diff options
| author | Ville Syrjälä <ville.syrjala@linux.intel.com> | 2022-02-18 17:47:06 +0200 |
|---|---|---|
| committer | Ville Syrjälä <ville.syrjala@linux.intel.com> | 2022-02-18 17:47:06 +0200 |
| commit | e1ab74623721da21eee91cb22b5824aa7d0c0d5b (patch) | |
| tree | 11573a3f5aa25700f12db4b1542651ee0980fd85 | |
| parent | eb89ef53badbc3c3bf25649b6c33dd045591e0c3 (diff) | |
| parent | b9c7babe2c2e37a50aa42401b38d597ea78f506e (diff) | |
Merge remote-tracking branch 'drm/drm-next' into drm-tip
865 files changed, 25036 insertions, 19233 deletions
diff --git a/Documentation/devicetree/bindings/display/bridge/lvds-codec.yaml b/Documentation/devicetree/bindings/display/bridge/lvds-codec.yaml index 5079c1cc337b..56b4c55f6f1b 100644 --- a/Documentation/devicetree/bindings/display/bridge/lvds-codec.yaml +++ b/Documentation/devicetree/bindings/display/bridge/lvds-codec.yaml @@ -39,6 +39,7 @@ properties: - const: lvds-encoder # Generic LVDS encoder compatible fallback - items: - enum: + - ti,ds90cf364a # For the DS90CF364A FPD-Link LVDS Receiver - ti,ds90cf384a # For the DS90CF384A FPD-Link LVDS Receiver - const: lvds-decoder # Generic LVDS decoders compatible fallback - enum: diff --git a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml index b446d0f0f1b4..48a97bb3e2e0 100644 --- a/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml +++ b/Documentation/devicetree/bindings/display/bridge/ti,sn65dsi83.yaml @@ -32,6 +32,9 @@ properties: maxItems: 1 description: GPIO specifier for bridge_en pin (active high). + vcc-supply: + description: A 1.8V power supply (see regulator/regulator.yaml). + ports: $ref: /schemas/graph.yaml#/properties/ports @@ -91,7 +94,6 @@ properties: required: - compatible - reg - - enable-gpios - ports allOf: @@ -133,6 +135,7 @@ examples: reg = <0x2d>; enable-gpios = <&gpio2 1 GPIO_ACTIVE_HIGH>; + vcc-supply = <®_sn65dsi83_1v8>; ports { #address-cells = <1>; diff --git a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml index 62f5f050c1bc..9cf5588a09d8 100644 --- a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml +++ b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml @@ -222,6 +222,8 @@ properties: - logictechno,lttd800480070-l6wh-rt # Mitsubishi "AA070MC01 7.0" WVGA TFT LCD panel - mitsubishi,aa070mc01-ca1 + # Multi-Inno Technology Co.,Ltd MI0700S4T-6 7" 800x480 TFT Resistive Touch Module + - multi-inno,mi0700s4t-6 # Multi-Inno Technology Co.,Ltd MI1010AIT-1CP 10.1" 1280x800 LVDS IPS Cap Touch Mod. - multi-inno,mi1010ait-1cp # NEC LCD Technologies, Ltd. 12.1" WXGA (1280x800) LVDS TFT LCD panel diff --git a/Documentation/devicetree/bindings/display/panel/sony,acx424akp.yaml b/Documentation/devicetree/bindings/display/panel/sony,acx424akp.yaml index 78d060097052..059cc6dbcfca 100644 --- a/Documentation/devicetree/bindings/display/panel/sony,acx424akp.yaml +++ b/Documentation/devicetree/bindings/display/panel/sony,acx424akp.yaml @@ -4,7 +4,12 @@ $id: http://devicetree.org/schemas/display/panel/sony,acx424akp.yaml# $schema: http://devicetree.org/meta-schemas/core.yaml# -title: Sony ACX424AKP 4" 480x864 AMOLED panel +title: Sony ACX424AKP/ACX424AKM 4" 480x864/480x854 AMOLED panel + +description: The Sony ACX424AKP and ACX424AKM are panels built around + the Novatek NT35560 display controller. The only difference is that + the AKM is configured to use 10 pixels less in the Y axis than the + AKP. maintainers: - Linus Walleij <linus.walleij@linaro.org> @@ -14,7 +19,9 @@ allOf: properties: compatible: - const: sony,acx424akp + enum: + - sony,acx424akp + - sony,acx424akm reg: true reset-gpios: true vddi-supply: diff --git a/Documentation/driver-api/device-io.rst b/Documentation/driver-api/device-io.rst index e9f04b1815d1..4d2baac0311c 100644 --- a/Documentation/driver-api/device-io.rst +++ b/Documentation/driver-api/device-io.rst @@ -502,6 +502,15 @@ pcim_iomap() Not using these wrappers may make drivers unusable on certain platforms with stricter rules for mapping I/O memory. +Generalizing Access to System and I/O Memory +============================================ + +.. kernel-doc:: include/linux/iosys-map.h + :doc: overview + +.. kernel-doc:: include/linux/iosys-map.h + :internal: + Public Functions Provided ========================= diff --git a/Documentation/driver-api/dma-buf.rst b/Documentation/driver-api/dma-buf.rst index 2cd7db82d9fe..55006678394a 100644 --- a/Documentation/driver-api/dma-buf.rst +++ b/Documentation/driver-api/dma-buf.rst @@ -128,15 +128,6 @@ Kernel Functions and Structures Reference .. kernel-doc:: include/linux/dma-buf.h :internal: -Buffer Mapping Helpers -~~~~~~~~~~~~~~~~~~~~~~ - -.. kernel-doc:: include/linux/dma-buf-map.h - :doc: overview - -.. kernel-doc:: include/linux/dma-buf-map.h - :internal: - Reservation Objects ------------------- diff --git a/Documentation/gpu/drm-internals.rst b/Documentation/gpu/drm-internals.rst index 607f78f0f189..38afed24a75c 100644 --- a/Documentation/gpu/drm-internals.rst +++ b/Documentation/gpu/drm-internals.rst @@ -75,6 +75,12 @@ update it, its value is mostly useless. The DRM core prints it to the kernel log at initialization time and passes it to userspace through the DRM_IOCTL_VERSION ioctl. +Module Initialization +--------------------- + +.. kernel-doc:: include/drm/drm_module.h + :doc: overview + Managing Ownership of the Framebuffer Aperture ---------------------------------------------- diff --git a/Documentation/gpu/drm-kms-helpers.rst b/Documentation/gpu/drm-kms-helpers.rst index 5bb55ec1b9b5..c3ce91eecbc1 100644 --- a/Documentation/gpu/drm-kms-helpers.rst +++ b/Documentation/gpu/drm-kms-helpers.rst @@ -232,34 +232,34 @@ HDCP Helper Functions Reference Display Port Helper Functions Reference ======================================= -.. kernel-doc:: drivers/gpu/drm/drm_dp_helper.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp.c :doc: dp helpers -.. kernel-doc:: include/drm/drm_dp_helper.h +.. kernel-doc:: include/drm/dp/drm_dp_helper.h :internal: -.. kernel-doc:: drivers/gpu/drm/drm_dp_helper.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp.c :export: Display Port CEC Helper Functions Reference =========================================== -.. kernel-doc:: drivers/gpu/drm/drm_dp_cec.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp_cec.c :doc: dp cec helpers -.. kernel-doc:: drivers/gpu/drm/drm_dp_cec.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp_cec.c :export: Display Port Dual Mode Adaptor Helper Functions Reference ========================================================= -.. kernel-doc:: drivers/gpu/drm/drm_dp_dual_mode_helper.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp_dual_mode_helper.c :doc: dp dual mode helpers -.. kernel-doc:: include/drm/drm_dp_dual_mode_helper.h +.. kernel-doc:: include/drm/dp/drm_dp_dual_mode_helper.h :internal: -.. kernel-doc:: drivers/gpu/drm/drm_dp_dual_mode_helper.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp_dual_mode_helper.c :export: Display Port MST Helpers @@ -268,19 +268,19 @@ Display Port MST Helpers Overview -------- -.. kernel-doc:: drivers/gpu/drm/drm_dp_mst_topology.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp_mst_topology.c :doc: dp mst helper -.. kernel-doc:: drivers/gpu/drm/drm_dp_mst_topology.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp_mst_topology.c :doc: Branch device and port refcounting Functions Reference ------------------- -.. kernel-doc:: include/drm/drm_dp_mst_helper.h +.. kernel-doc:: include/drm/dp/drm_dp_mst_helper.h :internal: -.. kernel-doc:: drivers/gpu/drm/drm_dp_mst_topology.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp_mst_topology.c :export: Topology Lifetime Internals @@ -289,7 +289,7 @@ Topology Lifetime Internals These functions aren't exported to drivers, but are documented here to help make the MST topology helpers easier to understand -.. kernel-doc:: drivers/gpu/drm/drm_dp_mst_topology.c +.. kernel-doc:: drivers/gpu/drm/dp/drm_dp_mst_topology.c :functions: drm_dp_mst_topology_try_get_mstb drm_dp_mst_topology_get_mstb drm_dp_mst_topology_put_mstb drm_dp_mst_topology_try_get_port drm_dp_mst_topology_get_port diff --git a/Documentation/gpu/drm-kms.rst b/Documentation/gpu/drm-kms.rst index d14bf1c35d7e..6f9c064fd323 100644 --- a/Documentation/gpu/drm-kms.rst +++ b/Documentation/gpu/drm-kms.rst @@ -423,12 +423,12 @@ Connector Functions Reference Writeback Connectors -------------------- -.. kernel-doc:: include/drm/drm_writeback.h - :internal: - .. kernel-doc:: drivers/gpu/drm/drm_writeback.c :doc: overview +.. kernel-doc:: include/drm/drm_writeback.h + :internal: + .. kernel-doc:: drivers/gpu/drm/drm_writeback.c :export: diff --git a/Documentation/gpu/drm-mm.rst b/Documentation/gpu/drm-mm.rst index e0538083a2c0..198bcc1affa1 100644 --- a/Documentation/gpu/drm-mm.rst +++ b/Documentation/gpu/drm-mm.rst @@ -8,7 +8,7 @@ the very dynamic nature of many of that data, managing graphics memory efficiently is thus crucial for the graphics stack and plays a central role in the DRM infrastructure. -The DRM core includes two memory managers, namely Translation Table Maps +The DRM core includes two memory managers, namely Translation Table Manager (TTM) and Graphics Execution Manager (GEM). TTM was the first DRM memory manager to be developed and tried to be a one-size-fits-them all solution. It provides a single userspace API to accommodate the need of diff --git a/Documentation/gpu/todo.rst b/Documentation/gpu/todo.rst index a1212b5b3026..7bf7f2111696 100644 --- a/Documentation/gpu/todo.rst +++ b/Documentation/gpu/todo.rst @@ -222,7 +222,7 @@ Convert drivers to use drm_fbdev_generic_setup() Most drivers can use drm_fbdev_generic_setup(). Driver have to implement atomic modesetting and GEM vmap support. Historically, generic fbdev emulation expected the framebuffer in system memory or system-like memory. By employing -struct dma_buf_map, drivers with frambuffers in I/O memory can be supported +struct iosys_map, drivers with frambuffers in I/O memory can be supported as well. Contact: Maintainer of the driver you plan to convert @@ -234,7 +234,7 @@ Reimplement functions in drm_fbdev_fb_ops without fbdev A number of callback functions in drm_fbdev_fb_ops could benefit from being rewritten without dependencies on the fbdev module. Some of the -helpers could further benefit from using struct dma_buf_map instead of +helpers could further benefit from using struct iosys_map instead of raw pointers. Contact: Thomas Zimmermann <tzimmermann@suse.de>, Daniel Vetter @@ -410,19 +410,19 @@ Contact: Emil Velikov, respective driver maintainers Level: Intermediate -Use struct dma_buf_map throughout codebase ------------------------------------------- +Use struct iosys_map throughout codebase +---------------------------------------- -Pointers to shared device memory are stored in struct dma_buf_map. Each +Pointers to shared device memory are stored in struct iosys_map. Each instance knows whether it refers to system or I/O memory. Most of the DRM-wide -interface have been converted to use struct dma_buf_map, but implementations +interface have been converted to use struct iosys_map, but implementations often still use raw pointers. -The task is to use struct dma_buf_map where it makes sense. +The task is to use struct iosys_map where it makes sense. -* Memory managers should use struct dma_buf_map for dma-buf-imported buffers. -* TTM might benefit from using struct dma_buf_map internally. -* Framebuffer copying and blitting helpers should operate on struct dma_buf_map. +* Memory managers should use struct iosys_map for dma-buf-imported buffers. +* TTM might benefit from using struct iosys_map internally. +* Framebuffer copying and blitting helpers should operate on struct iosys_map. Contact: Thomas Zimmermann <tzimmermann@suse.de>, Christian König, Daniel Vetter @@ -443,6 +443,21 @@ Contact: Thomas Zimmermann <tzimmermann@suse.de> Level: Intermediate +Request memory regions in all drivers +------------------------------------- + +Go through all drivers and add code to request the memory regions that the +driver uses. This requires adding calls to request_mem_region(), +pci_request_region() or similar functions. Use helpers for managed cleanup +where possible. + +Drivers are pretty bad at doing this and there used to be conflicts among +DRM and fbdev drivers. Still, it's the correct thing to do. + +Contact: Thomas Zimmermann <tzimmermann@suse.de> + +Level: Starter + Core refactorings ================= diff --git a/Documentation/gpu/vkms.rst b/Documentation/gpu/vkms.rst index 941f0e7e5eef..9c873c3912cc 100644 --- a/Documentation/gpu/vkms.rst +++ b/Documentation/gpu/vkms.rst @@ -124,8 +124,6 @@ Add Plane Features There's lots of plane features we could add support for: -- Multiple overlay planes. [Good to get started] - - Clearing primary plane: clear primary plane before plane composition (at the start) for correctness of pixel blend ops. It also guarantees alpha channel is cleared in the target buffer for stable crc. [Good to get started] diff --git a/MAINTAINERS b/MAINTAINERS index fca970a46e77..abfb59afd06d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5739,7 +5739,7 @@ T: git git://anongit.freedesktop.org/drm/drm-misc F: Documentation/driver-api/dma-buf.rst F: drivers/dma-buf/ F: include/linux/*fence.h -F: include/linux/dma-buf* +F: include/linux/dma-buf.h F: include/linux/dma-resv.h K: \bdma_(?:buf|fence|resv)\b @@ -10061,6 +10061,13 @@ F: include/linux/iova.h F: include/linux/of_iommu.h F: include/uapi/linux/iommu.h +IOSYS-MAP HELPERS +M: Thomas Zimmermann <tzimmermann@suse.de> +L: dri-devel@lists.freedesktop.org +S: Maintained +T: git git://anongit.freedesktop.org/drm/drm-misc +F: include/linux/iosys-map.h + IO_URING M: Jens Axboe <axboe@kernel.dk> R: Pavel Begunkov <asml.silence@gmail.com> diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c index dc7da08bc700..bd6dad83c65b 100644 --- a/arch/x86/kernel/early-quirks.c +++ b/arch/x86/kernel/early-quirks.c @@ -555,6 +555,7 @@ static const struct pci_device_id intel_early_ids[] __initconst = { INTEL_RKL_IDS(&gen11_early_ops), INTEL_ADLS_IDS(&gen11_early_ops), INTEL_ADLP_IDS(&gen11_early_ops), + INTEL_ADLN_IDS(&gen11_early_ops), INTEL_RPLS_IDS(&gen11_early_ops), }; diff --git a/drivers/char/agp/ati-agp.c b/drivers/char/agp/ati-agp.c index 20bf5f78a362..6f5530482d83 100644 --- a/drivers/char/agp/ati-agp.c +++ b/drivers/char/agp/ati-agp.c @@ -55,7 +55,7 @@ static struct _ati_generic_private { static int ati_create_page_map(struct ati_page_map *page_map) { - int i, err = 0; + int i, err; page_map->real = (unsigned long *) __get_free_page(GFP_KERNEL); if (page_map->real == NULL) @@ -63,6 +63,10 @@ static int ati_create_page_map(struct ati_page_map *page_map) set_memory_uc((unsigned long)page_map->real, 1); err = map_page_into_agp(virt_to_page(page_map->real)); + if (err) { + free_page((unsigned long)page_map->real); + return err; + } page_map->remapped = page_map->real; for (i = 0; i < PAGE_SIZE / sizeof(unsigned long); i++) { @@ -303,7 +307,7 @@ static int ati_insert_memory(struct agp_memory * mem, for (i = 0, j = pg_start; i < mem->page_count; i++, j++) { addr = (j * PAGE_SIZE) + agp_bridge->gart_bus_addr; cur_gatt = GET_GATT(addr); - writel(agp_bridge->driver->mask_memory(agp_bridge, + writel(agp_bridge->driver->mask_memory(agp_bridge, page_to_phys(mem->pages[i]), mem->type), cur_gatt+GET_GATT_OFF(addr)); diff --git a/drivers/char/agp/backend.c b/drivers/char/agp/backend.c index 004a3ce8ba72..0e19c600db53 100644 --- a/drivers/char/agp/backend.c +++ b/drivers/char/agp/backend.c @@ -62,6 +62,7 @@ EXPORT_SYMBOL(agp_find_bridge); /** * agp_backend_acquire - attempt to acquire an agp backend. + * @pdev: the PCI device * */ struct agp_bridge_data *agp_backend_acquire(struct pci_dev *pdev) @@ -83,6 +84,7 @@ EXPORT_SYMBOL(agp_backend_acquire); /** * agp_backend_release - release the lock on the agp backend. + * @bridge: the AGP backend to release * * The caller must insure that the graphics aperture translation table * is read for use by another entity. diff --git a/drivers/char/agp/frontend.c b/drivers/char/agp/frontend.c index 00ff5fcb808a..321118a9cfa5 100644 --- a/drivers/char/agp/frontend.c +++ b/drivers/char/agp/frontend.c @@ -39,7 +39,9 @@ #include <linux/fs.h> #include <linux/sched.h> #include <linux/uaccess.h> + #include "agp.h" +#include "compat_ioctl.h" struct agp_front_data agp_fe; @@ -1017,7 +1019,7 @@ static long agp_ioctl(struct file *file, case AGPIOC_UNBIND: ret_val = agpioc_unbind_wrap(curr_priv, (void __user *) arg); break; - + case AGPIOC_CHIPSET_FLUSH: break; } diff --git a/drivers/char/agp/nvidia-agp.c b/drivers/char/agp/nvidia-agp.c index f78e756157db..826dbd06f6bb 100644 --- a/drivers/char/agp/nvidia-agp.c +++ b/drivers/char/agp/nvidia-agp.c @@ -261,7 +261,8 @@ static int nvidia_remove_memory(struct agp_memory *mem, off_t pg_start, int type static void nvidia_tlbflush(struct agp_memory *mem) { unsigned long end; - u32 wbc_reg, temp; + u32 wbc_reg; + u32 __maybe_unused temp; int i; /* flush chipset */ diff --git a/drivers/char/agp/sworks-agp.c b/drivers/char/agp/sworks-agp.c index f875970bda65..b91da5998dd7 100644 --- a/drivers/char/agp/sworks-agp.c +++ b/drivers/char/agp/sworks-agp.c @@ -262,13 +262,10 @@ static void serverworks_tlbflush(struct agp_memory *temp) static int serverworks_configure(void) { - struct aper_size_info_lvl2 *current_size; u32 temp; u8 enable_reg; u16 cap_reg; - current_size = A_SIZE_LVL2(agp_bridge->current_size); - /* Get the memory mapped registers */ pci_read_config_dword(agp_bridge->dev, serverworks_private.mm_addr_ofs, &temp); temp = (temp & PCI_BASE_ADDRESS_MEM_MASK); @@ -350,7 +347,7 @@ static int serverworks_insert_memory(struct agp_memory *mem, for (i = 0, j = pg_start; i < mem->page_count; i++, j++) { addr = (j * PAGE_SIZE) + agp_bridge->gart_bus_addr; cur_gatt = SVRWRKS_GET_GATT(addr); - writel(agp_bridge->driver->mask_memory(agp_bridge, + writel(agp_bridge->driver->mask_memory(agp_bridge, page_to_phys(mem->pages[i]), mem->type), cur_gatt+GET_GATT_OFF(addr)); } diff --git a/drivers/char/agp/via-agp.c b/drivers/char/agp/via-agp.c index a460ae352772..b2f484f527fb 100644 --- a/drivers/char/agp/via-agp.c +++ b/drivers/char/agp/via-agp.c @@ -128,9 +128,6 @@ static int via_fetch_size_agp3(void) static int via_configure_agp3(void) { u32 temp; - struct aper_size_info_16 *current_size; - - current_size = A_SIZE_16(agp_bridge->current_size); /* address to map to */ agp_bridge->gart_bus_addr = pci_bus_address(agp_bridge->dev, diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c index 602b12d7470d..df23239b04fc 100644 --- a/drivers/dma-buf/dma-buf.c +++ b/drivers/dma-buf/dma-buf.c @@ -1047,8 +1047,8 @@ EXPORT_SYMBOL_NS_GPL(dma_buf_move_notify, DMA_BUF); * * Interfaces:: * - * void \*dma_buf_vmap(struct dma_buf \*dmabuf, struct dma_buf_map \*map) - * void dma_buf_vunmap(struct dma_buf \*dmabuf, struct dma_buf_map \*map) + * void \*dma_buf_vmap(struct dma_buf \*dmabuf, struct iosys_map \*map) + * void dma_buf_vunmap(struct dma_buf \*dmabuf, struct iosys_map \*map) * * The vmap call can fail if there is no vmap support in the exporter, or if * it runs out of vmalloc space. Note that the dma-buf layer keeps a reference @@ -1260,12 +1260,12 @@ EXPORT_SYMBOL_NS_GPL(dma_buf_mmap, DMA_BUF); * * Returns 0 on success, or a negative errno code otherwise. */ -int dma_buf_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map) +int dma_buf_vmap(struct dma_buf *dmabuf, struct iosys_map *map) { - struct dma_buf_map ptr; + struct iosys_map ptr; int ret = 0; - dma_buf_map_clear(map); + iosys_map_clear(map); if (WARN_ON(!dmabuf)) return -EINVAL; @@ -1276,12 +1276,12 @@ int dma_buf_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map) mutex_lock(&dmabuf->lock); if (dmabuf->vmapping_counter) { dmabuf->vmapping_counter++; - BUG_ON(dma_buf_map_is_null(&dmabuf->vmap_ptr)); + BUG_ON(iosys_map_is_null(&dmabuf->vmap_ptr)); *map = dmabuf->vmap_ptr; goto out_unlock; } - BUG_ON(dma_buf_map_is_set(&dmabuf->vmap_ptr)); + BUG_ON(iosys_map_is_set(&dmabuf->vmap_ptr)); ret = dmabuf->ops->vmap(dmabuf, &ptr); if (WARN_ON_ONCE(ret)) @@ -1303,20 +1303,20 @@ EXPORT_SYMBOL_NS_GPL(dma_buf_vmap, DMA_BUF); * @dmabuf: [in] buffer to vunmap * @map: [in] vmap pointer to vunmap */ -void dma_buf_vunmap(struct dma_buf *dmabuf, struct dma_buf_map *map) +void dma_buf_vunmap(struct dma_buf *dmabuf, struct iosys_map *map) { if (WARN_ON(!dmabuf)) return; - BUG_ON(dma_buf_map_is_null(&dmabuf->vmap_ptr)); + BUG_ON(iosys_map_is_null(&dmabuf->vmap_ptr)); BUG_ON(dmabuf->vmapping_counter == 0); - BUG_ON(!dma_buf_map_is_equal(&dmabuf->vmap_ptr, map)); + BUG_ON(!iosys_map_is_equal(&dmabuf->vmap_ptr, map)); mutex_lock(&dmabuf->lock); if (--dmabuf->vmapping_counter == 0) { if (dmabuf->ops->vunmap) dmabuf->ops->vunmap(dmabuf, map); - dma_buf_map_clear(&dmabuf->vmap_ptr); + iosys_map_clear(&dmabuf->vmap_ptr); } mutex_unlock(&dmabuf->lock); } diff --git a/drivers/dma-buf/dma-resv.c b/drivers/dma-buf/dma-resv.c index 4deea75c0b9c..6dd9a40b55d4 100644 --- a/drivers/dma-buf/dma-resv.c +++ b/drivers/dma-buf/dma-resv.c @@ -542,57 +542,45 @@ EXPORT_SYMBOL(dma_resv_copy_fences); * dma_resv_get_fences - Get an object's shared and exclusive * fences without update side lock held * @obj: the reservation object - * @fence_excl: the returned exclusive fence (or NULL) - * @shared_count: the number of shared fences returned - * @shared: the array of shared fence ptrs returned (array is krealloc'd to - * the required size, and must be freed by caller) + * @write: true if we should return all fences + * @num_fences: the number of fences returned + * @fences: the array of fence ptrs returned (array is krealloc'd to the + * required size, and must be freed by caller) * - * Retrieve all fences from the reservation object. If the pointer for the - * exclusive fence is not specified the fence is put into the array of the - * shared fences as well. Returns either zero or -ENOMEM. + * Retrieve all fences from the reservation object. + * Returns either zero or -ENOMEM. */ -int dma_resv_get_fences(struct dma_resv *obj, struct dma_fence **fence_excl, - unsigned int *shared_count, struct dma_fence ***shared) +int dma_resv_get_fences(struct dma_resv *obj, bool write, + unsigned int *num_fences, struct dma_fence ***fences) { struct dma_resv_iter cursor; struct dma_fence *fence; - *shared_count = 0; - *shared = NULL; + *num_fences = 0; + *fences = NULL; - if (fence_excl) - *fence_excl = NULL; - - dma_resv_iter_begin(&cursor, obj, true); + dma_resv_iter_begin(&cursor, obj, write); dma_resv_for_each_fence_unlocked(&cursor, fence) { if (dma_resv_iter_is_restarted(&cursor)) { unsigned int count; - while (*shared_count) - dma_fence_put((*shared)[--(*shared_count)]); - - if (fence_excl) - dma_fence_put(*fence_excl); + while (*num_fences) + dma_fence_put((*fences)[--(*num_fences)]); - count = cursor.shared_count; - count += fence_excl ? 0 : 1; + count = cursor.shared_count + 1; /* Eventually re-allocate the array */ - *shared = krealloc_array(*shared, count, + *fences = krealloc_array(*fences, count, sizeof(void *), GFP_KERNEL); - if (count && !*shared) { + if (count && !*fences) { dma_resv_iter_end(&cursor); return -ENOMEM; } } - dma_fence_get(fence); - if (dma_resv_iter_is_exclusive(&cursor) && fence_excl) - *fence_excl = fence; - else - (*shared)[(*shared_count)++] = fence; + (*fences)[(*num_fences)++] = dma_fence_get(fence); } dma_resv_iter_end(&cursor); diff --git a/drivers/dma-buf/heaps/cma_heap.c b/drivers/dma-buf/heaps/cma_heap.c index 83f02bd51dda..28fb04eccdd0 100644 --- a/drivers/dma-buf/heaps/cma_heap.c +++ b/drivers/dma-buf/heaps/cma_heap.c @@ -202,7 +202,7 @@ static void *cma_heap_do_vmap(struct cma_heap_buffer *buffer) return vaddr; } -static int cma_heap_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map) +static int cma_heap_vmap(struct dma_buf *dmabuf, struct iosys_map *map) { struct cma_heap_buffer *buffer = dmabuf->priv; void *vaddr; @@ -211,7 +211,7 @@ static int cma_heap_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map) mutex_lock(&buffer->lock); if (buffer->vmap_cnt) { buffer->vmap_cnt++; - dma_buf_map_set_vaddr(map, buffer->vaddr); + iosys_map_set_vaddr(map, buffer->vaddr); goto out; } @@ -222,14 +222,14 @@ static int cma_heap_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map) } buffer->vaddr = vaddr; buffer->vmap_cnt++; - dma_buf_map_set_vaddr(map, buffer->vaddr); + iosys_map_set_vaddr(map, buffer->vaddr); out: mutex_unlock(&buffer->lock); return ret; } -static void cma_heap_vunmap(struct dma_buf *dmabuf, struct dma_buf_map *map) +static void cma_heap_vunmap(struct dma_buf *dmabuf, struct iosys_map *map) { struct cma_heap_buffer *buffer = dmabuf->priv; @@ -239,7 +239,7 @@ static void cma_heap_vunmap(struct dma_buf *dmabuf, struct dma_buf_map *map) buffer->vaddr = NULL; } mutex_unlock(&buffer->lock); - dma_buf_map_clear(map); + iosys_map_clear(map); } static void cma_heap_dma_buf_release(struct dma_buf *dmabuf) diff --git a/drivers/dma-buf/heaps/system_heap.c b/drivers/dma-buf/heaps/system_heap.c index ab7fd896d2c4..fcf836ba9c1f 100644 --- a/drivers/dma-buf/heaps/system_heap.c +++ b/drivers/dma-buf/heaps/system_heap.c @@ -241,7 +241,7 @@ static void *system_heap_do_vmap(struct system_heap_buffer *buffer) return vaddr; } -static int system_heap_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map) +static int system_heap_vmap(struct dma_buf *dmabuf, struct iosys_map *map) { struct system_heap_buffer *buffer = dmabuf->priv; void *vaddr; @@ -250,7 +250,7 @@ static int system_heap_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map) mutex_lock(&buffer->lock); if (buffer->vmap_cnt) { buffer->vmap_cnt++; - dma_buf_map_set_vaddr(map, buffer->vaddr); + iosys_map_set_vaddr(map, buffer->vaddr); goto out; } @@ -262,14 +262,14 @@ static int system_heap_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map) buffer->vaddr = vaddr; buffer->vmap_cnt++; - dma_buf_map_set_vaddr(map, buffer->vaddr); + iosys_map_set_vaddr(map, buffer->vaddr); out: mutex_unlock(&buffer->lock); return ret; } -static void system_heap_vunmap(struct dma_buf *dmabuf, struct dma_buf_map *map) +static void system_heap_vunmap(struct dma_buf *dmabuf, struct iosys_map *map) { struct system_heap_buffer *buffer = dmabuf->priv; @@ -279,7 +279,7 @@ static void system_heap_vunmap(struct dma_buf *dmabuf, struct dma_buf_map *map) buffer->vaddr = NULL; } mutex_unlock(&buffer->lock); - dma_buf_map_clear(map); + iosys_map_clear(map); } static void system_heap_dma_buf_release(struct dma_buf *dmabuf) diff --git a/drivers/dma-buf/st-dma-resv.c b/drivers/dma-buf/st-dma-resv.c index bc32b3eedcb6..cbe999c6e7a6 100644 --- a/drivers/dma-buf/st-dma-resv.c +++ b/drivers/dma-buf/st-dma-resv.c @@ -275,7 +275,7 @@ static int test_shared_for_each_unlocked(void *arg) static int test_get_fences(void *arg, bool shared) { - struct dma_fence *f, *excl = NULL, **fences = NULL; + struct dma_fence *f, **fences = NULL; struct dma_resv resv; int r, i; @@ -304,35 +304,19 @@ static int test_get_fences(void *arg, bool shared) } dma_resv_unlock(&resv); - r = dma_resv_get_fences(&resv, &excl, &i, &fences); + r = dma_resv_get_fences(&resv, shared, &i, &fences); if (r) { pr_err("get_fences failed\n"); goto err_free; } - if (shared) { - if (excl != NULL) { - pr_err("get_fences returned unexpected excl fence\n"); - goto err_free; - } - if (i != 1 || fences[0] != f) { - pr_err("get_fences returned unexpected shared fence\n"); - goto err_free; - } - } else { - if (excl != f) { - pr_err("get_fences returned unexpected excl fence\n"); - goto err_free; - } - if (i != 0) { - pr_err("get_fences returned unexpected shared fence\n"); - goto err_free; - } + if (i != 1 || fences[0] != f) { + pr_err("get_fences returned unexpected fence\n"); + goto err_free; } dma_fence_signal(f); err_free: - dma_fence_put(excl); while (i--) dma_fence_put(fences[i]); kfree(fences); diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c index c57a609db75b..e7330684d3b8 100644 --- a/drivers/dma-buf/udmabuf.c +++ b/drivers/dma-buf/udmabuf.c @@ -190,6 +190,10 @@ static long udmabuf_create(struct miscdevice *device, if (ubuf->pagecount > pglimit) goto err; } + + if (!ubuf->pagecount) + goto err; + ubuf->pages = kmalloc_array(ubuf->pagecount, sizeof(*ubuf->pages), GFP_KERNEL); if (!ubuf->pages) { diff --git a/drivers/firmware/sysfb_simplefb.c b/drivers/firmware/sysfb_simplefb.c index 303a491e520d..76c4abc42a30 100644 --- a/drivers/firmware/sysfb_simplefb.c +++ b/drivers/firmware/sysfb_simplefb.c @@ -99,7 +99,7 @@ __init int sysfb_create_simplefb(const struct screen_info *si, /* setup IORESOURCE_MEM as framebuffer memory */ memset(&res, 0, sizeof(res)); - res.flags = IORESOURCE_MEM | IORESOURCE_BUSY; + res.flags = IORESOURCE_MEM; res.name = simplefb_resname; res.start = base; res.end = res.start + length - 1; diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index b1f22e457fd0..dfdd3ec5f793 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -68,6 +68,7 @@ config DRM_DEBUG_SELFTEST depends on DRM depends on DEBUG_KERNEL select PRIME_NUMBERS + select DRM_DP_HELPER select DRM_LIB_RANDOM select DRM_KMS_HELPER select DRM_EXPORT_FOR_TESTS if m @@ -80,6 +81,12 @@ config DRM_DEBUG_SELFTEST If in doubt, say "N". +config DRM_DP_HELPER + tristate + depends on DRM + help + DRM helpers for DisplayPort. + config DRM_KMS_HELPER tristate depends on DRM @@ -198,6 +205,12 @@ config DRM_TTM GPU memory types. Will be enabled automatically if a device driver uses it. +config DRM_BUDDY + tristate + depends on DRM + help + A page based buddy allocator + config DRM_VRAM_HELPER tristate depends on DRM @@ -236,6 +249,7 @@ config DRM_RADEON depends on DRM && PCI && MMU depends on AGP || !AGP select FW_LOADER + select DRM_DP_HELPER select DRM_KMS_HELPER select DRM_TTM select DRM_TTM_HELPER @@ -256,6 +270,7 @@ config DRM_AMDGPU tristate "AMD GPU" depends on DRM && PCI && MMU select FW_LOADER + select DRM_DP_HELPER select DRM_KMS_HELPER select DRM_SCHED select DRM_TTM diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index 301a44dc18e3..8675c2af7ae1 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -31,8 +31,6 @@ drm-$(CONFIG_DEBUG_FS) += drm_debugfs.o drm_debugfs_crc.o drm-$(CONFIG_DRM_LOAD_EDID_FIRMWARE) += drm_edid_load.o drm-$(CONFIG_DRM_PRIVACY_SCREEN) += drm_privacy_screen.o drm_privacy_screen_x86.o -obj-$(CONFIG_DRM_DP_AUX_BUS) += drm_dp_aux_bus.o - obj-$(CONFIG_DRM_NOMODESET) += drm_nomodeset.o drm_cma_helper-y := drm_gem_cma_helper.o @@ -42,27 +40,26 @@ obj-$(CONFIG_DRM_GEM_CMA_HELPER) += drm_cma_helper.o drm_shmem_helper-y := drm_gem_shmem_helper.o obj-$(CONFIG_DRM_GEM_SHMEM_HELPER) += drm_shmem_helper.o +obj-$(CONFIG_DRM_BUDDY) += drm_buddy.o + drm_vram_helper-y := drm_gem_vram_helper.o obj-$(CONFIG_DRM_VRAM_HELPER) += drm_vram_helper.o drm_ttm_helper-y := drm_gem_ttm_helper.o obj-$(CONFIG_DRM_TTM_HELPER) += drm_ttm_helper.o -drm_kms_helper-y := drm_bridge_connector.o drm_crtc_helper.o drm_dp_helper.o \ +drm_kms_helper-y := drm_bridge_connector.o drm_crtc_helper.o \ drm_dsc.o drm_encoder_slave.o drm_flip_work.o drm_hdcp.o \ drm_probe_helper.o \ - drm_plane_helper.o drm_dp_mst_topology.o drm_atomic_helper.o \ - drm_kms_helper_common.o drm_dp_dual_mode_helper.o \ + drm_plane_helper.o drm_atomic_helper.o \ + drm_kms_helper_common.o \ drm_simple_kms_helper.o drm_modeset_helper.o \ drm_scdc_helper.o drm_gem_atomic_helper.o \ drm_gem_framebuffer_helper.o \ drm_atomic_state_helper.o drm_damage_helper.o \ drm_format_helper.o drm_self_refresh_helper.o drm_rect.o - drm_kms_helper-$(CONFIG_DRM_PANEL_BRIDGE) += bridge/panel.o drm_kms_helper-$(CONFIG_DRM_FBDEV_EMULATION) += drm_fb_helper.o -drm_kms_helper-$(CONFIG_DRM_DP_AUX_CHARDEV) += drm_dp_aux_dev.o -drm_kms_helper-$(CONFIG_DRM_DP_CEC) += drm_dp_cec.o obj-$(CONFIG_DRM_KMS_HELPER) += drm_kms_helper.o obj-$(CONFIG_DRM_DEBUG_SELFTEST) += selftests/ @@ -72,6 +69,7 @@ obj-$(CONFIG_DRM_MIPI_DBI) += drm_mipi_dbi.o obj-$(CONFIG_DRM_MIPI_DSI) += drm_mipi_dsi.o obj-$(CONFIG_DRM_PANEL_ORIENTATION_QUIRKS) += drm_panel_orientation_quirks.o obj-y += arm/ +obj-y += dp/ obj-$(CONFIG_DRM_TTM) += ttm/ obj-$(CONFIG_DRM_SCHED) += scheduler/ obj-$(CONFIG_DRM_TDFX) += tdfx/ diff --git a/drivers/gpu/drm/amd/amdgpu/aldebaran.c b/drivers/gpu/drm/amd/amdgpu/aldebaran.c index bcfdb63b1d42..a545df4efce1 100644 --- a/drivers/gpu/drm/amd/amdgpu/aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/aldebaran.c @@ -260,7 +260,7 @@ static int aldebaran_mode2_restore_ip(struct amdgpu_device *adev) adev->gfx.rlc.funcs->resume(adev); /* Wait for FW reset event complete */ - r = smu_wait_for_event(adev, SMU_EVENT_RESET_COMPLETE, 0); + r = amdgpu_dpm_wait_for_event(adev, SMU_EVENT_RESET_COMPLETE, 0); if (r) { dev_err(adev->dev, "Failed to get response from firmware after reset\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 9a53a4de2bb7..2931c8ff4cc6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -99,7 +99,6 @@ #include "amdgpu_gem.h" #include "amdgpu_doorbell.h" #include "amdgpu_amdkfd.h" -#include "amdgpu_smu.h" #include "amdgpu_discovery.h" #include "amdgpu_mes.h" #include "amdgpu_umc.h" @@ -109,6 +108,7 @@ #include "amdgpu_smuio.h" #include "amdgpu_fdinfo.h" #include "amdgpu_mca.h" +#include "amdgpu_ras.h" #define MAX_GPU_INSTANCE 16 @@ -197,7 +197,6 @@ extern int amdgpu_emu_mode; extern uint amdgpu_smu_memory_pool_size; extern int amdgpu_smu_pptable_id; extern uint amdgpu_dc_feature_mask; -extern uint amdgpu_freesync_vid_mode; extern uint amdgpu_dc_debug_mask; extern uint amdgpu_dm_abm_level; extern int amdgpu_backlight; @@ -373,7 +372,8 @@ int amdgpu_device_ip_block_add(struct amdgpu_device *adev, */ bool amdgpu_get_bios(struct amdgpu_device *adev); bool amdgpu_read_bios(struct amdgpu_device *adev); - +bool amdgpu_soc15_read_bios_from_rom(struct amdgpu_device *adev, + u8 *bios, u32 length_bytes); /* * Clocks */ @@ -950,12 +950,6 @@ struct amdgpu_device { /* powerplay */ struct amd_powerplay powerplay; - bool pp_force_state_enabled; - - /* smu */ - struct smu_context smu; - - /* dpm */ struct amdgpu_pm pm; u32 cg_flags; u32 pg_flags; @@ -1100,6 +1094,8 @@ struct amdgpu_device { uint32_t ip_versions[MAX_HWIP][HWIP_MAX_INSTANCE]; bool ram_is_direct_mapped; + + struct list_head ras_list; }; static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev) @@ -1321,6 +1317,10 @@ void amdgpu_device_invalidate_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring); void amdgpu_device_halt(struct amdgpu_device *adev); +u32 amdgpu_device_pcie_port_rreg(struct amdgpu_device *adev, + u32 reg); +void amdgpu_device_pcie_port_wreg(struct amdgpu_device *adev, + u32 reg, u32 v); /* atpx handler */ #if defined(CONFIG_VGA_SWITCHEROO) @@ -1458,6 +1458,15 @@ int amdgpu_device_set_cg_state(struct amdgpu_device *adev, int amdgpu_device_set_pg_state(struct amdgpu_device *adev, enum amd_powergating_state state); +static inline bool amdgpu_device_has_timeouts_enabled(struct amdgpu_device *adev) +{ + return amdgpu_gpu_recovery != 0 && + adev->gfx_timeout != MAX_SCHEDULE_TIMEOUT && + adev->compute_timeout != MAX_SCHEDULE_TIMEOUT && + adev->sdma_timeout != MAX_SCHEDULE_TIMEOUT && + adev->video_timeout != MAX_SCHEDULE_TIMEOUT; +} + #include "amdgpu_object.h" static inline bool amdgpu_is_tmz(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index ac841ae8f5cc..4cb14c2fe53f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -131,6 +131,7 @@ struct amdkfd_process_info { atomic_t evicted_bos; struct delayed_work restore_userptr_work; struct pid *pid; + bool block_mmu_notifications; }; int amdgpu_amdkfd_init(void); @@ -268,7 +269,7 @@ uint64_t amdgpu_amdkfd_gpuvm_get_process_page_dir(void *drm_priv); int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( struct amdgpu_device *adev, uint64_t va, uint64_t size, void *drm_priv, struct kgd_mem **mem, - uint64_t *offset, uint32_t flags); + uint64_t *offset, uint32_t flags, bool criu_resume); int amdgpu_amdkfd_gpuvm_free_memory_of_gpu( struct amdgpu_device *adev, struct kgd_mem *mem, void *drm_priv, uint64_t *size); @@ -297,6 +298,10 @@ int amdgpu_amdkfd_get_tile_config(struct amdgpu_device *adev, struct tile_config *config); void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bool reset); +bool amdgpu_amdkfd_bo_mapped_to_dev(struct amdgpu_device *adev, struct kgd_mem *mem); +void amdgpu_amdkfd_block_mmu_notifications(void *p); +int amdgpu_amdkfd_criu_resume(void *p); + #if IS_ENABLED(CONFIG_HSA_AMD) void amdgpu_amdkfd_gpuvm_init_mem_limits(void); void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c index 46cd4ee6bafb..c8935d718207 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c @@ -37,10 +37,7 @@ const struct kfd2kgd_calls aldebaran_kfd2kgd = { .hqd_sdma_is_occupied = kgd_arcturus_hqd_sdma_is_occupied, .hqd_destroy = kgd_gfx_v9_hqd_destroy, .hqd_sdma_destroy = kgd_arcturus_hqd_sdma_destroy, - .address_watch_disable = kgd_gfx_v9_address_watch_disable, - .address_watch_execute = kgd_gfx_v9_address_watch_execute, .wave_control_execute = kgd_gfx_v9_wave_control_execute, - .address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, .set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c index abe93b3ff765..4191af5a3f13 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c @@ -289,10 +289,7 @@ const struct kfd2kgd_calls arcturus_kfd2kgd = { .hqd_sdma_is_occupied = kgd_arcturus_hqd_sdma_is_occupied, .hqd_destroy = kgd_gfx_v9_hqd_destroy, .hqd_sdma_destroy = kgd_arcturus_hqd_sdma_destroy, - .address_watch_disable = kgd_gfx_v9_address_watch_disable, - .address_watch_execute = kgd_gfx_v9_address_watch_execute, .wave_control_execute = kgd_gfx_v9_wave_control_execute, - .address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, .set_vm_context_page_table_base = diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c index 7b7f4b2764c1..9378fc79e9ea 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c @@ -671,20 +671,6 @@ static bool get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, return !!(value & ATC_VMID0_PASID_MAPPING__VALID_MASK); } -static int kgd_address_watch_disable(struct amdgpu_device *adev) -{ - return 0; -} - -static int kgd_address_watch_execute(struct amdgpu_device *adev, - unsigned int watch_point_id, - uint32_t cntl_val, - uint32_t addr_hi, - uint32_t addr_lo) -{ - return 0; -} - static int kgd_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, uint32_t sq_cmd) @@ -709,13 +695,6 @@ static int kgd_wave_control_execute(struct amdgpu_device *adev, return 0; } -static uint32_t kgd_address_watch_get_offset(struct amdgpu_device *adev, - unsigned int watch_point_id, - unsigned int reg_offset) -{ - return 0; -} - static void set_vm_context_page_table_base(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { @@ -767,10 +746,7 @@ const struct kfd2kgd_calls gfx_v10_kfd2kgd = { .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied, .hqd_destroy = kgd_hqd_destroy, .hqd_sdma_destroy = kgd_hqd_sdma_destroy, - .address_watch_disable = kgd_address_watch_disable, - .address_watch_execute = kgd_address_watch_execute, .wave_control_execute = kgd_wave_control_execute, - .address_watch_get_offset = kgd_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, .set_vm_context_page_table_base = set_vm_context_page_table_base, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c index 1f37d3574001..e9c80ce13f3e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c @@ -582,21 +582,6 @@ static int hqd_sdma_destroy_v10_3(struct amdgpu_device *adev, void *mqd, return 0; } - -static int address_watch_disable_v10_3(struct amdgpu_device *adev) -{ - return 0; -} - -static int address_watch_execute_v10_3(struct amdgpu_device *adev, - unsigned int watch_point_id, - uint32_t cntl_val, - uint32_t addr_hi, - uint32_t addr_lo) -{ - return 0; -} - static int wave_control_execute_v10_3(struct amdgpu_device *adev, uint32_t gfx_index_val, uint32_t sq_cmd) @@ -621,13 +606,6 @@ static int wave_control_execute_v10_3(struct amdgpu_device *adev, return 0; } -static uint32_t address_watch_get_offset_v10_3(struct amdgpu_device *adev, - unsigned int watch_point_id, - unsigned int reg_offset) -{ - return 0; -} - static void set_vm_context_page_table_base_v10_3(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { @@ -809,10 +787,7 @@ const struct kfd2kgd_calls gfx_v10_3_kfd2kgd = { .hqd_sdma_is_occupied = hqd_sdma_is_occupied_v10_3, .hqd_destroy = hqd_destroy_v10_3, .hqd_sdma_destroy = hqd_sdma_destroy_v10_3, - .address_watch_disable = address_watch_disable_v10_3, - .address_watch_execute = address_watch_execute_v10_3, .wave_control_execute = wave_control_execute_v10_3, - .address_watch_get_offset = address_watch_get_offset_v10_3, .get_atc_vmid_pasid_mapping_info = NULL, .set_vm_context_page_table_base = set_vm_context_page_table_base_v10_3, .program_trap_handler_settings = program_trap_handler_settings_v10_3, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c index 36528dad7684..65552bb7d2f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -45,43 +45,6 @@ enum { MAX_WATCH_ADDRESSES = 4 }; -enum { - ADDRESS_WATCH_REG_ADDR_HI = 0, - ADDRESS_WATCH_REG_ADDR_LO, - ADDRESS_WATCH_REG_CNTL, - ADDRESS_WATCH_REG_MAX -}; - -/* not defined in the CI/KV reg file */ -enum { - ADDRESS_WATCH_REG_CNTL_ATC_BIT = 0x10000000UL, - ADDRESS_WATCH_REG_CNTL_DEFAULT_MASK = 0x00FFFFFF, - ADDRESS_WATCH_REG_ADDLOW_MASK_EXTENSION = 0x03000000, - /* extend the mask to 26 bits to match the low address field */ - ADDRESS_WATCH_REG_ADDLOW_SHIFT = 6, - ADDRESS_WATCH_REG_ADDHIGH_MASK = 0xFFFF -}; - -static const uint32_t watchRegs[MAX_WATCH_ADDRESSES * ADDRESS_WATCH_REG_MAX] = { - mmTCP_WATCH0_ADDR_H, mmTCP_WATCH0_ADDR_L, mmTCP_WATCH0_CNTL, - mmTCP_WATCH1_ADDR_H, mmTCP_WATCH1_ADDR_L, mmTCP_WATCH1_CNTL, - mmTCP_WATCH2_ADDR_H, mmTCP_WATCH2_ADDR_L, mmTCP_WATCH2_CNTL, - mmTCP_WATCH3_ADDR_H, mmTCP_WATCH3_ADDR_L, mmTCP_WATCH3_CNTL -}; - -union TCP_WATCH_CNTL_BITS { - struct { - uint32_t mask:24; - uint32_t vmid:4; - uint32_t atc:1; - uint32_t mode:2; - uint32_t valid:1; - } bitfields, bits; - uint32_t u32All; - signed int i32All; - float f32All; -}; - static void lock_srbm(struct amdgpu_device *adev, uint32_t mec, uint32_t pipe, uint32_t queue, uint32_t vmid) { @@ -529,55 +492,6 @@ static int kgd_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd, return 0; } -static int kgd_address_watch_disable(struct amdgpu_device *adev) -{ - union TCP_WATCH_CNTL_BITS cntl; - unsigned int i; - - cntl.u32All = 0; - - cntl.bitfields.valid = 0; - cntl.bitfields.mask = ADDRESS_WATCH_REG_CNTL_DEFAULT_MASK; - cntl.bitfields.atc = 1; - - /* Turning off this address until we set all the registers */ - for (i = 0; i < MAX_WATCH_ADDRESSES; i++) - WREG32(watchRegs[i * ADDRESS_WATCH_REG_MAX + - ADDRESS_WATCH_REG_CNTL], cntl.u32All); - - return 0; -} - -static int kgd_address_watch_execute(struct amdgpu_device *adev, - unsigned int watch_point_id, - uint32_t cntl_val, - uint32_t addr_hi, - uint32_t addr_lo) -{ - union TCP_WATCH_CNTL_BITS cntl; - - cntl.u32All = cntl_val; - - /* Turning off this watch point until we set all the registers */ - cntl.bitfields.valid = 0; - WREG32(watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + - ADDRESS_WATCH_REG_CNTL], cntl.u32All); - - WREG32(watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + - ADDRESS_WATCH_REG_ADDR_HI], addr_hi); - - WREG32(watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + - ADDRESS_WATCH_REG_ADDR_LO], addr_lo); - - /* Enable the watch point */ - cntl.bitfields.valid = 1; - - WREG32(watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + - ADDRESS_WATCH_REG_CNTL], cntl.u32All); - - return 0; -} - static int kgd_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, uint32_t sq_cmd) @@ -602,13 +516,6 @@ static int kgd_wave_control_execute(struct amdgpu_device *adev, return 0; } -static uint32_t kgd_address_watch_get_offset(struct amdgpu_device *adev, - unsigned int watch_point_id, - unsigned int reg_offset) -{ - return watchRegs[watch_point_id * ADDRESS_WATCH_REG_MAX + reg_offset]; -} - static bool get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, uint8_t vmid, uint16_t *p_pasid) { @@ -665,10 +572,7 @@ const struct kfd2kgd_calls gfx_v7_kfd2kgd = { .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied, .hqd_destroy = kgd_hqd_destroy, .hqd_sdma_destroy = kgd_hqd_sdma_destroy, - .address_watch_disable = kgd_address_watch_disable, - .address_watch_execute = kgd_address_watch_execute, .wave_control_execute = kgd_wave_control_execute, - .address_watch_get_offset = kgd_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, .set_scratch_backing_va = set_scratch_backing_va, .set_vm_context_page_table_base = set_vm_context_page_table_base, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c index 52832cd69a93..9dc5f2a0cc07 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -538,20 +538,6 @@ static bool get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, return !!(value & ATC_VMID0_PASID_MAPPING__VALID_MASK); } -static int kgd_address_watch_disable(struct amdgpu_device *adev) -{ - return 0; -} - -static int kgd_address_watch_execute(struct amdgpu_device *adev, - unsigned int watch_point_id, - uint32_t cntl_val, - uint32_t addr_hi, - uint32_t addr_lo) -{ - return 0; -} - static int kgd_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, uint32_t sq_cmd) @@ -576,13 +562,6 @@ static int kgd_wave_control_execute(struct amdgpu_device *adev, return 0; } -static uint32_t kgd_address_watch_get_offset(struct amdgpu_device *adev, - unsigned int watch_point_id, - unsigned int reg_offset) -{ - return 0; -} - static void set_scratch_backing_va(struct amdgpu_device *adev, uint64_t va, uint32_t vmid) { @@ -614,10 +593,7 @@ const struct kfd2kgd_calls gfx_v8_kfd2kgd = { .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied, .hqd_destroy = kgd_hqd_destroy, .hqd_sdma_destroy = kgd_hqd_sdma_destroy, - .address_watch_disable = kgd_address_watch_disable, - .address_watch_execute = kgd_address_watch_execute, .wave_control_execute = kgd_wave_control_execute, - .address_watch_get_offset = kgd_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, .set_scratch_backing_va = set_scratch_backing_va, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index 1abf662a0e91..53895a41932e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -622,20 +622,6 @@ bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, return !!(value & ATC_VMID0_PASID_MAPPING__VALID_MASK); } -int kgd_gfx_v9_address_watch_disable(struct amdgpu_device *adev) -{ - return 0; -} - -int kgd_gfx_v9_address_watch_execute(struct amdgpu_device *adev, - unsigned int watch_point_id, - uint32_t cntl_val, - uint32_t addr_hi, - uint32_t addr_lo) -{ - return 0; -} - int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, uint32_t sq_cmd) @@ -660,13 +646,6 @@ int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev, return 0; } -uint32_t kgd_gfx_v9_address_watch_get_offset(struct amdgpu_device *adev, - unsigned int watch_point_id, - unsigned int reg_offset) -{ - return 0; -} - void kgd_gfx_v9_set_vm_context_page_table_base(struct amdgpu_device *adev, uint32_t vmid, uint64_t page_table_base) { @@ -888,10 +867,7 @@ const struct kfd2kgd_calls gfx_v9_kfd2kgd = { .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied, .hqd_destroy = kgd_gfx_v9_hqd_destroy, .hqd_sdma_destroy = kgd_hqd_sdma_destroy, - .address_watch_disable = kgd_gfx_v9_address_watch_disable, - .address_watch_execute = kgd_gfx_v9_address_watch_execute, .wave_control_execute = kgd_gfx_v9_wave_control_execute, - .address_watch_get_offset = kgd_gfx_v9_address_watch_get_offset, .get_atc_vmid_pasid_mapping_info = kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, .set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h index 24be49df26fd..c7ed3bc9053c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h @@ -46,19 +46,9 @@ int kgd_gfx_v9_hqd_destroy(struct amdgpu_device *adev, void *mqd, enum kfd_preempt_type reset_type, unsigned int utimeout, uint32_t pipe_id, uint32_t queue_id); -int kgd_gfx_v9_address_watch_disable(struct amdgpu_device *adev); -int kgd_gfx_v9_address_watch_execute(struct amdgpu_device *adev, - unsigned int watch_point_id, - uint32_t cntl_val, - uint32_t addr_hi, - uint32_t addr_lo); int kgd_gfx_v9_wave_control_execute(struct amdgpu_device *adev, uint32_t gfx_index_val, uint32_t sq_cmd); -uint32_t kgd_gfx_v9_address_watch_get_offset(struct amdgpu_device *adev, - unsigned int watch_point_id, - unsigned int reg_offset); - bool kgd_gfx_v9_get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, uint8_t vmid, uint16_t *p_pasid); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index f9bab963a948..2e00c3fb4bd3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -778,7 +778,7 @@ unwind: continue; if (attachment[i]->bo_va) { amdgpu_bo_reserve(bo[i], true); - amdgpu_vm_bo_rmv(adev, attachment[i]->bo_va); + amdgpu_vm_bo_del(adev, attachment[i]->bo_va); amdgpu_bo_unreserve(bo[i]); list_del(&attachment[i]->list); } @@ -795,7 +795,7 @@ static void kfd_mem_detach(struct kfd_mem_attachment *attachment) pr_debug("\t remove VA 0x%llx in entry %p\n", attachment->va, attachment); - amdgpu_vm_bo_rmv(attachment->adev, attachment->bo_va); + amdgpu_vm_bo_del(attachment->adev, attachment->bo_va); drm_gem_object_put(&bo->tbo.base); list_del(&attachment->list); kfree(attachment); @@ -842,7 +842,8 @@ static void remove_kgd_mem_from_kfd_bo_list(struct kgd_mem *mem, * * Returns 0 for success, negative errno for errors. */ -static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr) +static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr, + bool criu_resume) { struct amdkfd_process_info *process_info = mem->process_info; struct amdgpu_bo *bo = mem->bo; @@ -864,6 +865,18 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr) goto out; } + if (criu_resume) { + /* + * During a CRIU restore operation, the userptr buffer objects + * will be validated in the restore_userptr_work worker at a + * later stage when it is scheduled by another ioctl called by + * CRIU master process for the target pid for restore. + */ + atomic_inc(&mem->invalid); + mutex_unlock(&process_info->lock); + return 0; + } + ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); if (ret) { pr_err("%s: Failed to get user pages: %d\n", __func__, ret); @@ -1452,10 +1465,39 @@ uint64_t amdgpu_amdkfd_gpuvm_get_process_page_dir(void *drm_priv) return avm->pd_phys_addr; } +void amdgpu_amdkfd_block_mmu_notifications(void *p) +{ + struct amdkfd_process_info *pinfo = (struct amdkfd_process_info *)p; + + mutex_lock(&pinfo->lock); + WRITE_ONCE(pinfo->block_mmu_notifications, true); + mutex_unlock(&pinfo->lock); +} + +int amdgpu_amdkfd_criu_resume(void *p) +{ + int ret = 0; + struct amdkfd_process_info *pinfo = (struct amdkfd_process_info *)p; + + mutex_lock(&pinfo->lock); + pr_debug("scheduling work\n"); + atomic_inc(&pinfo->evicted_bos); + if (!READ_ONCE(pinfo->block_mmu_notifications)) { + ret = -EINVAL; + goto out_unlock; + } + WRITE_ONCE(pinfo->block_mmu_notifications, false); + schedule_delayed_work(&pinfo->restore_userptr_work, 0); + +out_unlock: + mutex_unlock(&pinfo->lock); + return ret; +} + int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( struct amdgpu_device *adev, uint64_t va, uint64_t size, void *drm_priv, struct kgd_mem **mem, - uint64_t *offset, uint32_t flags) + uint64_t *offset, uint32_t flags, bool criu_resume) { struct amdgpu_vm *avm = drm_priv_to_vm(drm_priv); enum ttm_bo_type bo_type = ttm_bo_type_device; @@ -1558,7 +1600,8 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, user_addr); if (user_addr) { - ret = init_user_pages(*mem, user_addr); + pr_debug("creating userptr BO for user_addr = %llu\n", user_addr); + ret = init_user_pages(*mem, user_addr, criu_resume); if (ret) goto allocate_init_user_pages_failed; } else if (flags & (KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL | @@ -1813,12 +1856,6 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu( true); ret = unreserve_bo_and_vms(&ctx, false, false); - /* Only apply no TLB flush on Aldebaran to - * workaround regressions on other Asics. - */ - if (table_freed && (adev->asic_type != CHIP_ALDEBARAN)) - *table_freed = true; - goto out; out_unreserve: @@ -2068,6 +2105,10 @@ int amdgpu_amdkfd_evict_userptr(struct kgd_mem *mem, int evicted_bos; int r = 0; + /* Do not process MMU notifications until stage-4 IOCTL is received */ + if (READ_ONCE(process_info->block_mmu_notifications)) + return 0; + atomic_inc(&mem->invalid); evicted_bos = atomic_inc_return(&process_info->evicted_bos); if (evicted_bos == 1) { @@ -2635,3 +2676,14 @@ int amdgpu_amdkfd_get_tile_config(struct amdgpu_device *adev, return 0; } + +bool amdgpu_amdkfd_bo_mapped_to_dev(struct amdgpu_device *adev, struct kgd_mem *mem) +{ + struct kfd_mem_attachment *entry; + + list_for_each_entry(entry, &mem->attachments, list) { + if (entry->is_mapped && entry->adev == adev) + return true; + } + return false; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c index 12a6b1c99c93..9ba4817a9148 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c @@ -1083,6 +1083,7 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev, return 0; } +#ifdef CONFIG_DRM_AMDGPU_SI int amdgpu_atombios_get_memory_pll_dividers(struct amdgpu_device *adev, u32 clock, bool strobe_mode, @@ -1503,6 +1504,7 @@ int amdgpu_atombios_init_mc_reg_table(struct amdgpu_device *adev, } return -EINVAL; } +#endif bool amdgpu_atombios_has_gpu_virtualization_table(struct amdgpu_device *adev) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h index 27e74b1fc260..4153d520e2a3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h @@ -160,6 +160,7 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev, bool strobe_mode, struct atom_clock_dividers *dividers); +#ifdef CONFIG_DRM_AMDGPU_SI int amdgpu_atombios_get_memory_pll_dividers(struct amdgpu_device *adev, u32 clock, bool strobe_mode, @@ -179,6 +180,17 @@ int amdgpu_atombios_get_voltage_table(struct amdgpu_device *adev, int amdgpu_atombios_init_mc_reg_table(struct amdgpu_device *adev, u8 module_index, struct atom_mc_reg_table *reg_table); +int amdgpu_atombios_get_max_vddc(struct amdgpu_device *adev, u8 voltage_type, + u16 voltage_id, u16 *voltage); +int amdgpu_atombios_get_leakage_vddc_based_on_leakage_idx(struct amdgpu_device *adev, + u16 *voltage, + u16 leakage_idx); +void amdgpu_atombios_get_default_voltages(struct amdgpu_device *adev, + u16 *vddc, u16 *vddci, u16 *mvdd); +int amdgpu_atombios_get_svi2_info(struct amdgpu_device *adev, + u8 voltage_type, + u8 *svd_gpio_id, u8 *svc_gpio_id); +#endif bool amdgpu_atombios_has_gpu_virtualization_table(struct amdgpu_device *adev); @@ -190,21 +202,11 @@ void amdgpu_atombios_scratch_regs_set_backlight_level(struct amdgpu_device *adev bool amdgpu_atombios_scratch_need_asic_init(struct amdgpu_device *adev); void amdgpu_atombios_copy_swap(u8 *dst, u8 *src, u8 num_bytes, bool to_le); -int amdgpu_atombios_get_max_vddc(struct amdgpu_device *adev, u8 voltage_type, - u16 voltage_id, u16 *voltage); -int amdgpu_atombios_get_leakage_vddc_based_on_leakage_idx(struct amdgpu_device *adev, - u16 *voltage, - u16 leakage_idx); -void amdgpu_atombios_get_default_voltages(struct amdgpu_device *adev, - u16 *vddc, u16 *vddci, u16 *mvdd); int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev, u8 clock_type, u32 clock, bool strobe_mode, struct atom_clock_dividers *dividers); -int amdgpu_atombios_get_svi2_info(struct amdgpu_device *adev, - u8 voltage_type, - u8 *svd_gpio_id, u8 *svc_gpio_id); int amdgpu_atombios_get_data_table(struct amdgpu_device *adev, uint32_t table, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c index 27b19503773b..0eddca795e96 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c @@ -464,3 +464,41 @@ success: adev->is_atom_fw = (adev->asic_type >= CHIP_VEGA10) ? true : false; return true; } + +/* helper function for soc15 and onwards to read bios from rom */ +bool amdgpu_soc15_read_bios_from_rom(struct amdgpu_device *adev, + u8 *bios, u32 length_bytes) +{ + u32 *dw_ptr; + u32 i, length_dw; + u32 rom_index_offset; + u32 rom_data_offset; + + if (bios == NULL) + return false; + if (length_bytes == 0) + return false; + /* APU vbios image is part of sbios image */ + if (adev->flags & AMD_IS_APU) + return false; + if (!adev->smuio.funcs || + !adev->smuio.funcs->get_rom_index_offset || + !adev->smuio.funcs->get_rom_data_offset) + return false; + + dw_ptr = (u32 *)bios; + length_dw = ALIGN(length_bytes, 4) / 4; + + rom_index_offset = + adev->smuio.funcs->get_rom_index_offset(adev); + rom_data_offset = + adev->smuio.funcs->get_rom_data_offset(adev); + + /* set rom index to 0 */ + WREG32(rom_index_offset, 0); + /* read out the rom data */ + for (i = 0; i < length_dw; i++) + dw_ptr[i] = RREG32(rom_data_offset); + + return true; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index c16a2704ced6..fa20261aa928 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -26,7 +26,7 @@ #include <drm/drm_edid.h> #include <drm/drm_fb_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_probe_helper.h> #include <drm/amdgpu_drm.h> #include "amdgpu.h" @@ -175,7 +175,7 @@ int amdgpu_connector_get_monitor_bpc(struct drm_connector *connector) /* Check if bpc is within clock limit. Try to degrade gracefully otherwise */ if ((bpc == 12) && (mode_clock * 3/2 > max_tmds_clock)) { - if ((connector->display_info.edid_hdmi_dc_modes & DRM_EDID_HDMI_DC_30) && + if ((connector->display_info.edid_hdmi_rgb444_dc_modes & DRM_EDID_HDMI_DC_30) && (mode_clock * 5/4 <= max_tmds_clock)) bpc = 10; else diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 06d07502a1f6..10b9e99c8941 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -127,8 +127,6 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, union drm_amdgpu_cs goto free_chunk; } - mutex_lock(&p->ctx->lock); - /* skip guilty context job */ if (atomic_read(&p->ctx->guilty) == 1) { ret = -ECANCELED; @@ -585,6 +583,16 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, } } + /* Move fence waiting after getting reservation lock of + * PD root. Then there is no need on a ctx mutex lock. + */ + r = amdgpu_ctx_wait_prev_fence(p->ctx, p->entity); + if (unlikely(r != 0)) { + if (r != -ERESTARTSYS) + DRM_ERROR("amdgpu_ctx_wait_prev_fence failed.\n"); + goto error_validate; + } + amdgpu_cs_get_threshold_for_moves(p->adev, &p->bytes_moved_threshold, &p->bytes_moved_vis_threshold); p->bytes_moved = 0; @@ -700,7 +708,6 @@ static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error, dma_fence_put(parser->fence); if (parser->ctx) { - mutex_unlock(&parser->ctx->lock); amdgpu_ctx_put(parser->ctx); } if (parser->bo_list) @@ -944,7 +951,7 @@ static int amdgpu_cs_ib_fill(struct amdgpu_device *adev, if (parser->job->uf_addr && ring->funcs->no_user_fence) return -EINVAL; - return amdgpu_ctx_wait_prev_fence(parser->ctx, parser->entity); + return 0; } static int amdgpu_cs_process_fence_dep(struct amdgpu_cs_parser *p, @@ -1274,14 +1281,11 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, /* * Work around dma_resv shortcommings by wrapping up the * submission in a dma_fence_chain and add it as exclusive - * fence, but first add the submission as shared fence to make - * sure that shared fences never signal before the exclusive - * one. + * fence. */ dma_fence_chain_init(chain, dma_resv_excl_fence(resv), dma_fence_get(p->fence), 1); - dma_resv_add_shared_fence(resv, p->fence); rcu_assign_pointer(resv->fence_excl, &chain->base); e->chain = NULL; } @@ -1363,7 +1367,6 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) goto out; r = amdgpu_cs_submit(&parser, cs); - out: amdgpu_cs_parser_fini(&parser, r, reserved_buffers); @@ -1509,6 +1512,7 @@ int amdgpu_cs_fence_to_handle_ioctl(struct drm_device *dev, void *data, return 0; default: + dma_fence_put(fence); return -EINVAL; } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c index da21e60bb827..c6d4d41c4393 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c @@ -98,7 +98,7 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm, if (r) { DRM_ERROR("failed to do bo_map on static CSA, err=%d\n", r); - amdgpu_vm_bo_rmv(adev, *bo_va); + amdgpu_vm_bo_del(adev, *bo_va); ttm_eu_backoff_reservation(&ticket, &list); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 468003583b2a..1c72f6095f08 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -230,13 +230,13 @@ static int amdgpu_ctx_init(struct amdgpu_device *adev, kref_init(&ctx->refcount); spin_lock_init(&ctx->ring_lock); - mutex_init(&ctx->lock); ctx->reset_counter = atomic_read(&adev->gpu_reset_counter); ctx->reset_counter_query = ctx->reset_counter; ctx->vram_lost_counter = atomic_read(&adev->vram_lost_counter); ctx->init_priority = priority; ctx->override_priority = AMDGPU_CTX_PRIORITY_UNSET; + ctx->stable_pstate = AMDGPU_CTX_STABLE_PSTATE_NONE; return 0; } @@ -255,6 +255,86 @@ static void amdgpu_ctx_fini_entity(struct amdgpu_ctx_entity *entity) kfree(entity); } +static int amdgpu_ctx_get_stable_pstate(struct amdgpu_ctx *ctx, + u32 *stable_pstate) +{ + struct amdgpu_device *adev = ctx->adev; + enum amd_dpm_forced_level current_level; + + if (!ctx) + return -EINVAL; + + current_level = amdgpu_dpm_get_performance_level(adev); + + switch (current_level) { + case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_STANDARD; + break; + case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_MIN_SCLK; + break; + case AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_MIN_MCLK; + break; + case AMD_DPM_FORCED_LEVEL_PROFILE_PEAK: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_PEAK; + break; + default: + *stable_pstate = AMDGPU_CTX_STABLE_PSTATE_NONE; + break; + } + return 0; +} + +static int amdgpu_ctx_set_stable_pstate(struct amdgpu_ctx *ctx, + u32 stable_pstate) +{ + struct amdgpu_device *adev = ctx->adev; + enum amd_dpm_forced_level level; + int r; + + if (!ctx) + return -EINVAL; + + mutex_lock(&adev->pm.stable_pstate_ctx_lock); + if (adev->pm.stable_pstate_ctx && adev->pm.stable_pstate_ctx != ctx) { + r = -EBUSY; + goto done; + } + + switch (stable_pstate) { + case AMDGPU_CTX_STABLE_PSTATE_NONE: + level = AMD_DPM_FORCED_LEVEL_AUTO; + break; + case AMDGPU_CTX_STABLE_PSTATE_STANDARD: + level = AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD; + break; + case AMDGPU_CTX_STABLE_PSTATE_MIN_SCLK: + level = AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK; + break; + case AMDGPU_CTX_STABLE_PSTATE_MIN_MCLK: + level = AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK; + break; + case AMDGPU_CTX_STABLE_PSTATE_PEAK: + level = AMD_DPM_FORCED_LEVEL_PROFILE_PEAK; + break; + default: + r = -EINVAL; + goto done; + } + + r = amdgpu_dpm_force_performance_level(adev, level); + + if (level == AMD_DPM_FORCED_LEVEL_AUTO) + adev->pm.stable_pstate_ctx = NULL; + else + adev->pm.stable_pstate_ctx = ctx; +done: + mutex_unlock(&adev->pm.stable_pstate_ctx_lock); + + return r; +} + static void amdgpu_ctx_fini(struct kref *ref) { struct amdgpu_ctx *ctx = container_of(ref, struct amdgpu_ctx, refcount); @@ -270,8 +350,7 @@ static void amdgpu_ctx_fini(struct kref *ref) ctx->entities[i][j] = NULL; } } - - mutex_destroy(&ctx->lock); + amdgpu_ctx_set_stable_pstate(ctx, AMDGPU_CTX_STABLE_PSTATE_NONE); kfree(ctx); } @@ -467,11 +546,41 @@ static int amdgpu_ctx_query2(struct amdgpu_device *adev, return 0; } + + +static int amdgpu_ctx_stable_pstate(struct amdgpu_device *adev, + struct amdgpu_fpriv *fpriv, uint32_t id, + bool set, u32 *stable_pstate) +{ + struct amdgpu_ctx *ctx; + struct amdgpu_ctx_mgr *mgr; + int r; + + if (!fpriv) + return -EINVAL; + + mgr = &fpriv->ctx_mgr; + mutex_lock(&mgr->lock); + ctx = idr_find(&mgr->ctx_handles, id); + if (!ctx) { + mutex_unlock(&mgr->lock); + return -EINVAL; + } + + if (set) + r = amdgpu_ctx_set_stable_pstate(ctx, *stable_pstate); + else + r = amdgpu_ctx_get_stable_pstate(ctx, stable_pstate); + + mutex_unlock(&mgr->lock); + return r; +} + int amdgpu_ctx_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) { int r; - uint32_t id; + uint32_t id, stable_pstate; int32_t priority; union drm_amdgpu_ctx *args = data; @@ -500,6 +609,20 @@ int amdgpu_ctx_ioctl(struct drm_device *dev, void *data, case AMDGPU_CTX_OP_QUERY_STATE2: r = amdgpu_ctx_query2(adev, fpriv, id, &args->out); break; + case AMDGPU_CTX_OP_GET_STABLE_PSTATE: + if (args->in.flags) + return -EINVAL; + r = amdgpu_ctx_stable_pstate(adev, fpriv, id, false, &stable_pstate); + args->out.pstate.flags = stable_pstate; + break; + case AMDGPU_CTX_OP_SET_STABLE_PSTATE: + if (args->in.flags & ~AMDGPU_CTX_STABLE_PSTATE_FLAGS_MASK) + return -EINVAL; + stable_pstate = args->in.flags & AMDGPU_CTX_STABLE_PSTATE_FLAGS_MASK; + if (stable_pstate > AMDGPU_CTX_STABLE_PSTATE_PEAK) + return -EINVAL; + r = amdgpu_ctx_stable_pstate(adev, fpriv, id, true, &stable_pstate); + break; default: return -EINVAL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h index a44b8b8ed39c..d0cbfcea90f7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h @@ -49,10 +49,10 @@ struct amdgpu_ctx { bool preamble_presented; int32_t init_priority; int32_t override_priority; - struct mutex lock; atomic_t guilty; unsigned long ras_counter_ce; unsigned long ras_counter_ue; + uint32_t stable_pstate; }; struct amdgpu_ctx_mgr { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index 25e2e5bf90eb..4b950de9bf66 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -1120,8 +1120,10 @@ static ssize_t amdgpu_debugfs_gfxoff_read(struct file *f, char __user *buf, return -EINVAL; r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) + if (r < 0) { + pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); return r; + } while (size) { uint32_t value; @@ -1585,22 +1587,25 @@ static int amdgpu_debugfs_sclk_set(void *data, u64 val) return ret; } - if (is_support_sw_smu(adev)) { - ret = smu_get_dpm_freq_range(&adev->smu, SMU_SCLK, &min_freq, &max_freq); - if (ret || val > max_freq || val < min_freq) - return -EINVAL; - ret = smu_set_soft_freq_range(&adev->smu, SMU_SCLK, (uint32_t)val, (uint32_t)val); - } else { - return 0; + ret = amdgpu_dpm_get_dpm_freq_range(adev, PP_SCLK, &min_freq, &max_freq); + if (ret == -EOPNOTSUPP) { + ret = 0; + goto out; } + if (ret || val > max_freq || val < min_freq) { + ret = -EINVAL; + goto out; + } + + ret = amdgpu_dpm_set_soft_freq_range(adev, PP_SCLK, (uint32_t)val, (uint32_t)val); + if (ret) + ret = -EINVAL; +out: pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - if (ret) - return -EINVAL; - - return 0; + return ret; } DEFINE_DEBUGFS_ATTRIBUTE(fops_ib_preempt, NULL, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index ed077de426d9..0b98d65056e3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -31,6 +31,7 @@ #include <linux/console.h> #include <linux/slab.h> #include <linux/iommu.h> +#include <linux/pci.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_probe_helper.h> @@ -55,7 +56,6 @@ #include "soc15.h" #include "nv.h" #include "bif/bif_4_1_d.h" -#include <linux/pci.h> #include <linux/firmware.h> #include "amdgpu_vf_error.h" @@ -566,7 +566,7 @@ void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, adev->gfx.rlc.funcs && adev->gfx.rlc.funcs->is_rlcg_access_range) { if (adev->gfx.rlc.funcs->is_rlcg_access_range(adev, reg)) - return adev->gfx.rlc.funcs->sriov_wreg(adev, reg, v, 0, 0); + return amdgpu_sriov_wreg(adev, reg, v, 0, 0); } else if ((reg * 4) >= adev->rmmio_size) { adev->pcie_wreg(adev, reg * 4, v); } else { @@ -2073,6 +2073,8 @@ out: */ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) { + struct drm_device *dev = adev_to_drm(adev); + struct pci_dev *parent; int i, r; amdgpu_device_enable_virtual_display(adev); @@ -2137,6 +2139,16 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) break; } + if (amdgpu_has_atpx() && + (amdgpu_is_atpx_hybrid() || + amdgpu_has_atpx_dgpu_power_cntl()) && + ((adev->flags & AMD_IS_APU) == 0) && + !pci_is_thunderbolt_attached(to_pci_dev(dev->dev))) + adev->flags |= AMD_IS_PX; + + parent = pci_upstream_bridge(adev->pdev); + adev->has_pr3 = parent ? pci_pr3_present(parent) : false; + amdgpu_amdkfd_device_probe(adev); adev->pm.pp_feature = amdgpu_pp_feature_mask; @@ -2624,7 +2636,7 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) /* For passthrough configuration on arcturus and aldebaran, enable special handling SBR */ if (amdgpu_passthrough(adev) && ((adev->asic_type == CHIP_ARCTURUS && adev->gmc.xgmi.num_physical_nodes > 1)|| adev->asic_type == CHIP_ALDEBARAN )) - smu_handle_passthrough_sbr(&adev->smu, true); + amdgpu_dpm_handle_passthrough_sbr(adev, true); if (adev->gmc.xgmi.num_physical_nodes > 1) { mutex_lock(&mgpu_info.mutex); @@ -2708,11 +2720,11 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) } } - amdgpu_amdkfd_suspend(adev, false); - amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); + amdgpu_amdkfd_suspend(adev, false); + /* Workaroud for ASICs need to disable SMC first */ amdgpu_device_smu_fini_early(adev); @@ -2881,7 +2893,7 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) int i, r; if (adev->in_s0ix) - amdgpu_gfx_state_change_set(adev, sGpuChangeState_D3Entry); + amdgpu_dpm_gfx_state_change(adev, sGpuChangeState_D3Entry); for (i = adev->num_ip_blocks - 1; i >= 0; i--) { if (!adev->ip_blocks[i].status.valid) @@ -3307,9 +3319,9 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) if (adev->asic_reset_res) goto fail; - if (adev->mmhub.ras_funcs && - adev->mmhub.ras_funcs->reset_ras_error_count) - adev->mmhub.ras_funcs->reset_ras_error_count(adev); + if (adev->mmhub.ras && adev->mmhub.ras->ras_block.hw_ops && + adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count) + adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count(adev); } else { task_barrier_full(&hive->tb); @@ -3497,8 +3509,9 @@ int amdgpu_device_init(struct amdgpu_device *adev, init_rwsem(&adev->reset_sem); mutex_init(&adev->psp.mutex); mutex_init(&adev->notifier_lock); + mutex_init(&adev->pm.stable_pstate_ctx_lock); - amdgpu_device_init_apu_flags(adev); + amdgpu_device_init_apu_flags(adev); r = amdgpu_device_check_arguments(adev); if (r) @@ -3519,6 +3532,8 @@ int amdgpu_device_init(struct amdgpu_device *adev, INIT_LIST_HEAD(&adev->reset_list); + INIT_LIST_HEAD(&adev->ras_list); + INIT_DELAYED_WORK(&adev->delayed_init_work, amdgpu_device_delayed_init_work_handler); INIT_DELAYED_WORK(&adev->gfx.gfx_off_delay_work, @@ -4044,7 +4059,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon) return 0; if (adev->in_s0ix) - amdgpu_gfx_state_change_set(adev, sGpuChangeState_D0Entry); + amdgpu_dpm_gfx_state_change(adev, sGpuChangeState_D0Entry); /* post card */ if (amdgpu_device_need_post(adev)) { @@ -4645,9 +4660,9 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle, if (!r && amdgpu_ras_intr_triggered()) { list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - if (tmp_adev->mmhub.ras_funcs && - tmp_adev->mmhub.ras_funcs->reset_ras_error_count) - tmp_adev->mmhub.ras_funcs->reset_ras_error_count(tmp_adev); + if (tmp_adev->mmhub.ras && tmp_adev->mmhub.ras->ras_block.hw_ops && + tmp_adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count) + tmp_adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count(tmp_adev); } amdgpu_ras_intr_cleared(); @@ -5726,3 +5741,36 @@ void amdgpu_device_halt(struct amdgpu_device *adev) pci_disable_device(pdev); pci_wait_for_pending_transaction(pdev); } + +u32 amdgpu_device_pcie_port_rreg(struct amdgpu_device *adev, + u32 reg) +{ + unsigned long flags, address, data; + u32 r; + + address = adev->nbio.funcs->get_pcie_port_index_offset(adev); + data = adev->nbio.funcs->get_pcie_port_data_offset(adev); + + spin_lock_irqsave(&adev->pcie_idx_lock, flags); + WREG32(address, reg * 4); + (void)RREG32(address); + r = RREG32(data); + spin_unlock_irqrestore(&adev->pcie_idx_lock, flags); + return r; +} + +void amdgpu_device_pcie_port_wreg(struct amdgpu_device *adev, + u32 reg, u32 v) +{ + unsigned long flags, address, data; + + address = adev->nbio.funcs->get_pcie_port_index_offset(adev); + data = adev->nbio.funcs->get_pcie_port_data_offset(adev); + + spin_lock_irqsave(&adev->pcie_idx_lock, flags); + WREG32(address, reg * 4); + (void)RREG32(address); + WREG32(data, v); + (void)RREG32(data); + spin_unlock_irqrestore(&adev->pcie_idx_lock, flags); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 81bfee978b74..cd7e8522c130 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -674,6 +674,7 @@ static int amdgpu_discovery_set_common_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): case IP_VERSION(10, 3, 0): case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 2): @@ -709,6 +710,7 @@ static int amdgpu_discovery_set_gmc_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): case IP_VERSION(10, 3, 0): case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 2): @@ -846,8 +848,14 @@ static int amdgpu_discovery_set_display_ip_blocks(struct amdgpu_device *adev) { if (adev->enable_virtual_display || amdgpu_sriov_vf(adev)) { amdgpu_device_ip_block_add(adev, &amdgpu_vkms_ip_block); + return 0; + } + + if (!amdgpu_device_has_dc_support(adev)) + return 0; + #if defined(CONFIG_DRM_AMD_DC) - } else if (adev->ip_versions[DCE_HWIP][0]) { + if (adev->ip_versions[DCE_HWIP][0]) { switch (adev->ip_versions[DCE_HWIP][0]) { case IP_VERSION(1, 0, 0): case IP_VERSION(1, 0, 1): @@ -882,8 +890,8 @@ static int amdgpu_discovery_set_display_ip_blocks(struct amdgpu_device *adev) adev->ip_versions[DCI_HWIP][0]); return -EINVAL; } -#endif } +#endif return 0; } @@ -904,6 +912,7 @@ static int amdgpu_discovery_set_gc_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): case IP_VERSION(10, 3, 0): case IP_VERSION(10, 3, 2): case IP_VERSION(10, 3, 1): @@ -1038,6 +1047,7 @@ static int amdgpu_discovery_set_mes_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): case IP_VERSION(10, 3, 0): case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 2): @@ -1217,11 +1227,6 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) return -EINVAL; amdgpu_discovery_harvest_ip(adev); - - if (!adev->mman.discovery_bin) { - DRM_ERROR("ip discovery uninitialized\n"); - return -EINVAL; - } break; } @@ -1242,6 +1247,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): case IP_VERSION(10, 3, 0): case IP_VERSION(10, 3, 2): case IP_VERSION(10, 3, 4): @@ -1258,6 +1264,20 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) return -EINVAL; } + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(9, 1, 0): + case IP_VERSION(9, 2, 2): + case IP_VERSION(9, 3, 0): + case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): + case IP_VERSION(10, 3, 1): + case IP_VERSION(10, 3, 3): + adev->flags |= AMD_IS_APU; + break; + default: + break; + } + if (adev->ip_versions[XGMI_HWIP][0] == IP_VERSION(4, 8, 0)) adev->gmc.xgmi.supported = true; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index 82011e75ed85..ec4c9ef5f795 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -200,8 +200,10 @@ int amdgpu_display_crtc_page_flip_target(struct drm_crtc *crtc, goto unpin; } - r = dma_resv_get_fences(new_abo->tbo.base.resv, NULL, - &work->shared_count, &work->shared); + /* TODO: Unify this with other drivers */ + r = dma_resv_get_fences(new_abo->tbo.base.resv, true, + &work->shared_count, + &work->shared); if (unlikely(r != 0)) { DRM_ERROR("failed to get fences for buffer\n"); goto unpin; @@ -510,19 +512,24 @@ uint32_t amdgpu_display_supported_domains(struct amdgpu_device *adev, case CHIP_STONEY: domain |= AMDGPU_GEM_DOMAIN_GTT; break; - case CHIP_RAVEN: - /* enable S/G on PCO and RV2 */ - if ((adev->apu_flags & AMD_APU_IS_RAVEN2) || - (adev->apu_flags & AMD_APU_IS_PICASSO)) - domain |= AMDGPU_GEM_DOMAIN_GTT; - break; - case CHIP_RENOIR: - case CHIP_VANGOGH: - case CHIP_YELLOW_CARP: - domain |= AMDGPU_GEM_DOMAIN_GTT; - break; - default: + switch (adev->ip_versions[DCE_HWIP][0]) { + case IP_VERSION(1, 0, 0): + case IP_VERSION(1, 0, 1): + /* enable S/G on PCO and RV2 */ + if ((adev->apu_flags & AMD_APU_IS_RAVEN2) || + (adev->apu_flags & AMD_APU_IS_PICASSO)) + domain |= AMDGPU_GEM_DOMAIN_GTT; + break; + case IP_VERSION(2, 1, 0): + case IP_VERSION(3, 0, 1): + case IP_VERSION(3, 1, 2): + case IP_VERSION(3, 1, 3): + domain |= AMDGPU_GEM_DOMAIN_GTT; + break; + default: + break; + } break; } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 63a089992645..5cdafdcfec59 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -99,9 +99,10 @@ * - 3.42.0 - Add 16bpc fixed point display support * - 3.43.0 - Add device hot plug/unplug support * - 3.44.0 - DCN3 supports DCC independent block settings: !64B && 128B, 64B && 128B + * - 3.45.0 - Add context ioctl stable pstate interface */ #define KMS_DRIVER_MAJOR 3 -#define KMS_DRIVER_MINOR 44 +#define KMS_DRIVER_MINOR 45 #define KMS_DRIVER_PATCHLEVEL 0 int amdgpu_vram_limit; @@ -174,7 +175,6 @@ int amdgpu_mes; int amdgpu_noretry = -1; int amdgpu_force_asic_type = -1; int amdgpu_tmz = -1; /* auto */ -uint amdgpu_freesync_vid_mode; int amdgpu_reset_method = -1; /* auto */ int amdgpu_num_kcq = -1; int amdgpu_smartshift_bias; @@ -844,32 +844,6 @@ MODULE_PARM_DESC(tmz, "Enable TMZ feature (-1 = auto (default), 0 = off, 1 = on) module_param_named(tmz, amdgpu_tmz, int, 0444); /** - * DOC: freesync_video (uint) - * Enable the optimization to adjust front porch timing to achieve seamless - * mode change experience when setting a freesync supported mode for which full - * modeset is not needed. - * - * The Display Core will add a set of modes derived from the base FreeSync - * video mode into the corresponding connector's mode list based on commonly - * used refresh rates and VRR range of the connected display, when users enable - * this feature. From the userspace perspective, they can see a seamless mode - * change experience when the change between different refresh rates under the - * same resolution. Additionally, userspace applications such as Video playback - * can read this modeset list and change the refresh rate based on the video - * frame rate. Finally, the userspace can also derive an appropriate mode for a - * particular refresh rate based on the FreeSync Mode and add it to the - * connector's mode list. - * - * Note: This is an experimental feature. - * - * The default value: 0 (off). - */ -MODULE_PARM_DESC( - freesync_video, - "Enable freesync modesetting optimization feature (0 = off (default), 1 = on)"); -module_param_named(freesync_video, amdgpu_freesync_vid_mode, uint, 0444); - -/** * DOC: reset_method (int) * GPU reset method (-1 = auto (default), 0 = legacy, 1 = mode0, 2 = mode1, 3 = mode2, 4 = baco, 5 = pci) */ @@ -1942,10 +1916,10 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x73FF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_DIMGREY_CAVEFISH}, /* Aldebaran */ - {0x1002, 0x7408, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ALDEBARAN|AMD_EXP_HW_SUPPORT}, - {0x1002, 0x740C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ALDEBARAN|AMD_EXP_HW_SUPPORT}, - {0x1002, 0x740F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ALDEBARAN|AMD_EXP_HW_SUPPORT}, - {0x1002, 0x7410, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ALDEBARAN|AMD_EXP_HW_SUPPORT}, + {0x1002, 0x7408, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ALDEBARAN}, + {0x1002, 0x740C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ALDEBARAN}, + {0x1002, 0x740F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ALDEBARAN}, + {0x1002, 0x7410, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ALDEBARAN}, /* CYAN_SKILLFISH */ {0x1002, 0x13FE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYAN_SKILLFISH|AMD_IS_APU}, @@ -1994,6 +1968,28 @@ static bool amdgpu_is_fw_framebuffer(resource_size_t base, return found; } +static void amdgpu_get_secondary_funcs(struct amdgpu_device *adev) +{ + struct pci_dev *p = NULL; + int i; + + /* 0 - GPU + * 1 - audio + * 2 - USB + * 3 - UCSI + */ + for (i = 1; i < 4; i++) { + p = pci_get_domain_bus_and_slot(pci_domain_nr(adev->pdev->bus), + adev->pdev->bus->number, i); + if (p) { + pm_runtime_get_sync(&p->dev); + pm_runtime_mark_last_busy(&p->dev); + pm_runtime_put_autosuspend(&p->dev); + pci_dev_put(p); + } + } +} + static int amdgpu_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { @@ -2126,6 +2122,48 @@ retry_init: if (ret) DRM_ERROR("Creating debugfs files failed (%d).\n", ret); + if (adev->runpm) { + /* only need to skip on ATPX */ + if (amdgpu_device_supports_px(ddev)) + dev_pm_set_driver_flags(ddev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); + /* we want direct complete for BOCO */ + if (amdgpu_device_supports_boco(ddev)) + dev_pm_set_driver_flags(ddev->dev, DPM_FLAG_SMART_PREPARE | + DPM_FLAG_SMART_SUSPEND | + DPM_FLAG_MAY_SKIP_RESUME); + pm_runtime_use_autosuspend(ddev->dev); + pm_runtime_set_autosuspend_delay(ddev->dev, 5000); + + pm_runtime_allow(ddev->dev); + + pm_runtime_mark_last_busy(ddev->dev); + pm_runtime_put_autosuspend(ddev->dev); + + /* + * For runpm implemented via BACO, PMFW will handle the + * timing for BACO in and out: + * - put ASIC into BACO state only when both video and + * audio functions are in D3 state. + * - pull ASIC out of BACO state when either video or + * audio function is in D0 state. + * Also, at startup, PMFW assumes both functions are in + * D0 state. + * + * So if snd driver was loaded prior to amdgpu driver + * and audio function was put into D3 state, there will + * be no PMFW-aware D-state transition(D0->D3) on runpm + * suspend. Thus the BACO will be not correctly kicked in. + * + * Via amdgpu_get_secondary_funcs(), the audio dev is put + * into D0 state. Then there will be a PMFW-aware D-state + * transition(D0->D3) on runpm suspend. + */ + if (amdgpu_device_supports_baco(ddev) && + !(adev->flags & AMD_IS_APU) && + (adev->asic_type >= CHIP_NAVI10)) + amdgpu_get_secondary_funcs(adev); + } + return 0; err_pci: @@ -2137,8 +2175,15 @@ static void amdgpu_pci_remove(struct pci_dev *pdev) { struct drm_device *dev = pci_get_drvdata(pdev); + struct amdgpu_device *adev = drm_to_adev(dev); drm_dev_unplug(dev); + + if (adev->runpm) { + pm_runtime_get_sync(dev->dev); + pm_runtime_forbid(dev->dev); + } + amdgpu_driver_unload_kms(dev); /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c index 2a786e788627..ecada5eadfe3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c @@ -30,7 +30,6 @@ #include "amdgpu_eeprom.h" #define FRU_EEPROM_MADDR 0x60000 -#define I2C_PRODUCT_INFO_OFFSET 0xC0 static bool is_fru_eeprom_supported(struct amdgpu_device *adev) { @@ -40,7 +39,13 @@ static bool is_fru_eeprom_supported(struct amdgpu_device *adev) */ struct atom_context *atom_ctx = adev->mode_info.atom_context; - /* VBIOS is of the format ###-DXXXYY-##. For SKU identification, + /* The i2c access is blocked on VF + * TODO: Need other way to get the info + */ + if (amdgpu_sriov_vf(adev)) + return false; + + /* VBIOS is of the format ###-DXXXYYYY-##. For SKU identification, * we can use just the "DXXX" portion. If there were more models, we * could convert the 3 characters to a hex integer and use a switch * for ease/speed/readability. For now, 2 string comparisons are @@ -59,17 +64,24 @@ static bool is_fru_eeprom_supported(struct amdgpu_device *adev) case CHIP_ALDEBARAN: /* All Aldebaran SKUs have the FRU */ return true; + case CHIP_SIENNA_CICHLID: + if (strnstr(atom_ctx->vbios_version, "D603", + sizeof(atom_ctx->vbios_version))) + return true; + else + return false; default: return false; } } static int amdgpu_fru_read_eeprom(struct amdgpu_device *adev, uint32_t addrptr, - unsigned char *buff) + unsigned char *buf, size_t buf_size) { - int ret, size; + int ret; + u8 size; - ret = amdgpu_eeprom_read(&adev->pm.smu_i2c, addrptr, buff, 1); + ret = amdgpu_eeprom_read(adev->pm.fru_eeprom_i2c_bus, addrptr, buf, 1); if (ret < 1) { DRM_WARN("FRU: Failed to get size field"); return ret; @@ -78,9 +90,11 @@ static int amdgpu_fru_read_eeprom(struct amdgpu_device *adev, uint32_t addrptr, /* The size returned by the i2c requires subtraction of 0xC0 since the * size apparently always reports as 0xC0+actual size. */ - size = buff[0] - I2C_PRODUCT_INFO_OFFSET; + size = buf[0] & 0x3F; + size = min_t(size_t, size, buf_size); - ret = amdgpu_eeprom_read(&adev->pm.smu_i2c, addrptr + 1, buff, size); + ret = amdgpu_eeprom_read(adev->pm.fru_eeprom_i2c_bus, addrptr + 1, + buf, size); if (ret < 1) { DRM_WARN("FRU: Failed to get data field"); return ret; @@ -91,19 +105,15 @@ static int amdgpu_fru_read_eeprom(struct amdgpu_device *adev, uint32_t addrptr, int amdgpu_fru_get_product_info(struct amdgpu_device *adev) { - unsigned char buff[AMDGPU_PRODUCT_NAME_LEN+2]; + unsigned char buf[AMDGPU_PRODUCT_NAME_LEN]; u32 addrptr; int size, len; - int offset = 2; if (!is_fru_eeprom_supported(adev)) return 0; - if (adev->asic_type == CHIP_ALDEBARAN) - offset = 0; - /* If algo exists, it means that the i2c_adapter's initialized */ - if (!adev->pm.smu_i2c.algo) { + if (!adev->pm.fru_eeprom_i2c_bus || !adev->pm.fru_eeprom_i2c_bus->algo) { DRM_WARN("Cannot access FRU, EEPROM accessor not initialized"); return -ENODEV; } @@ -121,7 +131,7 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) * and the language field, so just start from 0xb, manufacturer size */ addrptr = FRU_EEPROM_MADDR + 0xb; - size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + size = amdgpu_fru_read_eeprom(adev, addrptr, buf, sizeof(buf)); if (size < 1) { DRM_ERROR("Failed to read FRU Manufacturer, ret:%d", size); return -EINVAL; @@ -131,7 +141,7 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) * size field being 1 byte. This pattern continues below. */ addrptr += size + 1; - size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + size = amdgpu_fru_read_eeprom(adev, addrptr, buf, sizeof(buf)); if (size < 1) { DRM_ERROR("Failed to read FRU product name, ret:%d", size); return -EINVAL; @@ -143,12 +153,11 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) AMDGPU_PRODUCT_NAME_LEN); len = AMDGPU_PRODUCT_NAME_LEN - 1; } - /* Start at 2 due to buff using fields 0 and 1 for the address */ - memcpy(adev->product_name, &buff[offset], len); + memcpy(adev->product_name, buf, len); adev->product_name[len] = '\0'; addrptr += size + 1; - size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + size = amdgpu_fru_read_eeprom(adev, addrptr, buf, sizeof(buf)); if (size < 1) { DRM_ERROR("Failed to read FRU product number, ret:%d", size); return -EINVAL; @@ -162,11 +171,11 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) DRM_WARN("FRU Product Number is larger than 16 characters. This is likely a mistake"); len = sizeof(adev->product_number) - 1; } - memcpy(adev->product_number, &buff[offset], len); + memcpy(adev->product_number, buf, len); adev->product_number[len] = '\0'; addrptr += size + 1; - size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + size = amdgpu_fru_read_eeprom(adev, addrptr, buf, sizeof(buf)); if (size < 1) { DRM_ERROR("Failed to read FRU product version, ret:%d", size); @@ -174,7 +183,7 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) } addrptr += size + 1; - size = amdgpu_fru_read_eeprom(adev, addrptr, buff); + size = amdgpu_fru_read_eeprom(adev, addrptr, buf, sizeof(buf)); if (size < 1) { DRM_ERROR("Failed to read FRU serial number, ret:%d", size); @@ -189,7 +198,7 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) DRM_WARN("FRU Serial Number is larger than 16 characters. This is likely a mistake"); len = sizeof(adev->serial) - 1; } - memcpy(adev->serial, &buff[offset], len); + memcpy(adev->serial, buf, len); adev->serial[len] = '\0'; return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c index 645950a653a0..01cb89ffbd56 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c @@ -150,7 +150,7 @@ void amdgpu_gart_table_vram_free(struct amdgpu_device *adev) * replaces them with the dummy page (all asics). * Returns 0 for success, -EINVAL for failure. */ -int amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, +void amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, int pages) { unsigned t; @@ -161,13 +161,11 @@ int amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, uint64_t flags = 0; int idx; - if (!adev->gart.ready) { - WARN(1, "trying to unbind memory from uninitialized GART !\n"); - return -EINVAL; - } + if (!adev->gart.ptr) + return; if (!drm_dev_enter(adev_to_drm(adev), &idx)) - return 0; + return; t = offset / AMDGPU_GPU_PAGE_SIZE; p = t / AMDGPU_GPU_PAGES_IN_CPU_PAGE; @@ -188,7 +186,6 @@ int amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, amdgpu_gmc_flush_gpu_tlb(adev, 0, i, 0); drm_dev_exit(idx); - return 0; } /** @@ -204,7 +201,7 @@ int amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, * Map the dma_addresses into GART entries (all asics). * Returns 0 for success, -EINVAL for failure. */ -int amdgpu_gart_map(struct amdgpu_device *adev, uint64_t offset, +void amdgpu_gart_map(struct amdgpu_device *adev, uint64_t offset, int pages, dma_addr_t *dma_addr, uint64_t flags, void *dst) { @@ -212,13 +209,8 @@ int amdgpu_gart_map(struct amdgpu_device *adev, uint64_t offset, unsigned i, j, t; int idx; - if (!adev->gart.ready) { - WARN(1, "trying to bind memory to uninitialized GART !\n"); - return -EINVAL; - } - if (!drm_dev_enter(adev_to_drm(adev), &idx)) - return 0; + return; t = offset / AMDGPU_GPU_PAGE_SIZE; @@ -230,7 +222,6 @@ int amdgpu_gart_map(struct amdgpu_device *adev, uint64_t offset, } } drm_dev_exit(idx); - return 0; } /** @@ -246,20 +237,14 @@ int amdgpu_gart_map(struct amdgpu_device *adev, uint64_t offset, * (all asics). * Returns 0 for success, -EINVAL for failure. */ -int amdgpu_gart_bind(struct amdgpu_device *adev, uint64_t offset, +void amdgpu_gart_bind(struct amdgpu_device *adev, uint64_t offset, int pages, dma_addr_t *dma_addr, uint64_t flags) { - if (!adev->gart.ready) { - WARN(1, "trying to bind memory to uninitialized GART !\n"); - return -EINVAL; - } - if (!adev->gart.ptr) - return 0; + return; - return amdgpu_gart_map(adev, offset, pages, dma_addr, flags, - adev->gart.ptr); + amdgpu_gart_map(adev, offset, pages, dma_addr, flags, adev->gart.ptr); } /** @@ -274,6 +259,9 @@ void amdgpu_gart_invalidate_tlb(struct amdgpu_device *adev) { int i; + if (!adev->gart.ptr) + return; + mb(); amdgpu_device_flush_hdp(adev, NULL); for (i = 0; i < adev->num_vmhubs; i++) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h index 78895413cf9f..8fea3e04e411 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h @@ -46,7 +46,6 @@ struct amdgpu_gart { unsigned num_gpu_pages; unsigned num_cpu_pages; unsigned table_size; - bool ready; /* Asic default pte flags */ uint64_t gart_pte_flags; @@ -58,12 +57,12 @@ int amdgpu_gart_table_vram_pin(struct amdgpu_device *adev); void amdgpu_gart_table_vram_unpin(struct amdgpu_device *adev); int amdgpu_gart_init(struct amdgpu_device *adev); void amdgpu_gart_dummy_page_fini(struct amdgpu_device *adev); -int amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, - int pages); -int amdgpu_gart_map(struct amdgpu_device *adev, uint64_t offset, - int pages, dma_addr_t *dma_addr, uint64_t flags, - void *dst); -int amdgpu_gart_bind(struct amdgpu_device *adev, uint64_t offset, - int pages, dma_addr_t *dma_addr, uint64_t flags); +void amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset, + int pages); +void amdgpu_gart_map(struct amdgpu_device *adev, uint64_t offset, + int pages, dma_addr_t *dma_addr, uint64_t flags, + void *dst); +void amdgpu_gart_bind(struct amdgpu_device *adev, uint64_t offset, + int pages, dma_addr_t *dma_addr, uint64_t flags); void amdgpu_gart_invalidate_tlb(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index c0d8f40a5b45..57b74d35052f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -222,16 +222,10 @@ static void amdgpu_gem_object_close(struct drm_gem_object *obj, if (!bo_va || --bo_va->ref_count) goto out_unlock; - amdgpu_vm_bo_rmv(adev, bo_va); + amdgpu_vm_bo_del(adev, bo_va); if (!amdgpu_vm_ready(vm)) goto out_unlock; - fence = dma_resv_excl_fence(bo->tbo.base.resv); - if (fence) { - amdgpu_bo_fence(bo, fence, true); - fence = NULL; - } - r = amdgpu_vm_clear_freed(adev, vm, &fence); if (r || !fence) goto out_unlock; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 1916ec84dd71..43004822ec6f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -615,14 +615,14 @@ int amdgpu_get_gfx_off_status(struct amdgpu_device *adev, uint32_t *value) mutex_lock(&adev->gfx.gfx_off_mutex); - r = smu_get_status_gfxoff(adev, value); + r = amdgpu_dpm_get_status_gfxoff(adev, value); mutex_unlock(&adev->gfx.gfx_off_mutex); return r; } -int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev) +int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev, void *ras_info) { int r; struct ras_fs_if fs_info = { @@ -695,9 +695,9 @@ int amdgpu_gfx_process_ras_data_cb(struct amdgpu_device *adev, */ if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) { kgd2kfd_set_sram_ecc_flag(adev->kfd.dev); - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->query_ras_error_count) - adev->gfx.ras_funcs->query_ras_error_count(adev, err_data); + if (adev->gfx.ras && adev->gfx.ras->ras_block.hw_ops && + adev->gfx.ras->ras_block.hw_ops->query_ras_error_count) + adev->gfx.ras->ras_block.hw_ops->query_ras_error_count(adev, err_data); amdgpu_ras_reset_gpu(adev); } return AMDGPU_RAS_SUCCESS; @@ -852,19 +852,3 @@ int amdgpu_gfx_get_num_kcq(struct amdgpu_device *adev) } return amdgpu_num_kcq; } - -/* amdgpu_gfx_state_change_set - Handle gfx power state change set - * @adev: amdgpu_device pointer - * @state: gfx power state(1 -sGpuChangeState_D0Entry and 2 -sGpuChangeState_D3Entry) - * - */ - -void amdgpu_gfx_state_change_set(struct amdgpu_device *adev, enum gfx_change_state state) -{ - mutex_lock(&adev->pm.mutex); - if (adev->powerplay.pp_funcs && - adev->powerplay.pp_funcs->gfx_state_change_set) - ((adev)->powerplay.pp_funcs->gfx_state_change_set( - (adev)->powerplay.pp_handle, state)); - mutex_unlock(&adev->pm.mutex); -} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index f851196c83a5..f99eac544f6d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -31,6 +31,7 @@ #include "amdgpu_ring.h" #include "amdgpu_rlc.h" #include "soc15.h" +#include "amdgpu_ras.h" /* GFX current status */ #define AMDGPU_GFX_NORMAL_MODE 0x00000000L @@ -47,12 +48,6 @@ enum amdgpu_gfx_pipe_priority { AMDGPU_GFX_PIPE_PRIO_HIGH = AMDGPU_RING_PRIO_2 }; -/* Argument for PPSMC_MSG_GpuChangeState */ -enum gfx_change_state { - sGpuChangeState_D0Entry = 1, - sGpuChangeState_D3Entry, -}; - #define AMDGPU_GFX_QUEUE_PRIORITY_MINIMUM 0 #define AMDGPU_GFX_QUEUE_PRIORITY_MAXIMUM 15 @@ -204,16 +199,8 @@ struct amdgpu_cu_info { uint32_t bitmap[4][4]; }; -struct amdgpu_gfx_ras_funcs { - int (*ras_late_init)(struct amdgpu_device *adev); - void (*ras_fini)(struct amdgpu_device *adev); - int (*ras_error_inject)(struct amdgpu_device *adev, - void *inject_if); - int (*query_ras_error_count)(struct amdgpu_device *adev, - void *ras_error_status); - void (*reset_ras_error_count)(struct amdgpu_device *adev); - void (*query_ras_error_status)(struct amdgpu_device *adev); - void (*reset_ras_error_status)(struct amdgpu_device *adev); +struct amdgpu_gfx_ras { + struct amdgpu_ras_block_object ras_block; void (*enable_watchdog_timer)(struct amdgpu_device *adev); }; @@ -337,7 +324,7 @@ struct amdgpu_gfx { /*ras */ struct ras_common_if *ras_if; - const struct amdgpu_gfx_ras_funcs *ras_funcs; + struct amdgpu_gfx_ras *ras; }; #define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev)) @@ -399,7 +386,7 @@ bool amdgpu_gfx_is_me_queue_enabled(struct amdgpu_device *adev, int me, int pipe, int queue); void amdgpu_gfx_off_ctrl(struct amdgpu_device *adev, bool enable); int amdgpu_get_gfx_off_status(struct amdgpu_device *adev, uint32_t *value); -int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev); +int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev, void *ras_info); void amdgpu_gfx_ras_fini(struct amdgpu_device *adev); int amdgpu_gfx_process_ras_data_cb(struct amdgpu_device *adev, void *err_data, @@ -410,5 +397,4 @@ int amdgpu_gfx_cp_ecc_error_irq(struct amdgpu_device *adev, uint32_t amdgpu_kiq_rreg(struct amdgpu_device *adev, uint32_t reg); void amdgpu_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v); int amdgpu_gfx_get_num_kcq(struct amdgpu_device *adev); -void amdgpu_gfx_state_change_set(struct amdgpu_device *adev, enum gfx_change_state state); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 2430d6223c2d..ec1203d49799 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -436,58 +436,58 @@ void amdgpu_gmc_filter_faults_remove(struct amdgpu_device *adev, uint64_t addr, } while (fault->timestamp < tmp); } +int amdgpu_gmc_ras_early_init(struct amdgpu_device *adev) +{ + if (!adev->gmc.xgmi.connected_to_cpu) { + adev->gmc.xgmi.ras = &xgmi_ras; + amdgpu_ras_register_ras_block(adev, &adev->gmc.xgmi.ras->ras_block); + } + + return 0; +} + int amdgpu_gmc_ras_late_init(struct amdgpu_device *adev) { int r; - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->ras_late_init) { - r = adev->umc.ras_funcs->ras_late_init(adev); + if (adev->umc.ras && adev->umc.ras->ras_block.ras_late_init) { + r = adev->umc.ras->ras_block.ras_late_init(adev, NULL); if (r) return r; } - if (adev->mmhub.ras_funcs && - adev->mmhub.ras_funcs->ras_late_init) { - r = adev->mmhub.ras_funcs->ras_late_init(adev); + if (adev->mmhub.ras && adev->mmhub.ras->ras_block.ras_late_init) { + r = adev->mmhub.ras->ras_block.ras_late_init(adev, NULL); if (r) return r; } - if (!adev->gmc.xgmi.connected_to_cpu) - adev->gmc.xgmi.ras_funcs = &xgmi_ras_funcs; - - if (adev->gmc.xgmi.ras_funcs && - adev->gmc.xgmi.ras_funcs->ras_late_init) { - r = adev->gmc.xgmi.ras_funcs->ras_late_init(adev); + if (adev->gmc.xgmi.ras && adev->gmc.xgmi.ras->ras_block.ras_late_init) { + r = adev->gmc.xgmi.ras->ras_block.ras_late_init(adev, NULL); if (r) return r; } - if (adev->hdp.ras_funcs && - adev->hdp.ras_funcs->ras_late_init) { - r = adev->hdp.ras_funcs->ras_late_init(adev); + if (adev->hdp.ras && adev->hdp.ras->ras_block.ras_late_init) { + r = adev->hdp.ras->ras_block.ras_late_init(adev, NULL); if (r) return r; } - if (adev->mca.mp0.ras_funcs && - adev->mca.mp0.ras_funcs->ras_late_init) { - r = adev->mca.mp0.ras_funcs->ras_late_init(adev); + if (adev->mca.mp0.ras && adev->mca.mp0.ras->ras_block.ras_late_init) { + r = adev->mca.mp0.ras->ras_block.ras_late_init(adev, NULL); if (r) return r; } - if (adev->mca.mp1.ras_funcs && - adev->mca.mp1.ras_funcs->ras_late_init) { - r = adev->mca.mp1.ras_funcs->ras_late_init(adev); + if (adev->mca.mp1.ras && adev->mca.mp1.ras->ras_block.ras_late_init) { + r = adev->mca.mp1.ras->ras_block.ras_late_init(adev, NULL); if (r) return r; } - if (adev->mca.mpio.ras_funcs && - adev->mca.mpio.ras_funcs->ras_late_init) { - r = adev->mca.mpio.ras_funcs->ras_late_init(adev); + if (adev->mca.mpio.ras && adev->mca.mpio.ras->ras_block.ras_late_init) { + r = adev->mca.mpio.ras->ras_block.ras_late_init(adev, NULL); if (r) return r; } @@ -497,21 +497,17 @@ int amdgpu_gmc_ras_late_init(struct amdgpu_device *adev) void amdgpu_gmc_ras_fini(struct amdgpu_device *adev) { - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->ras_fini) - adev->umc.ras_funcs->ras_fini(adev); + if (adev->umc.ras && adev->umc.ras->ras_block.ras_fini) + adev->umc.ras->ras_block.ras_fini(adev); - if (adev->mmhub.ras_funcs && - adev->mmhub.ras_funcs->ras_fini) - adev->mmhub.ras_funcs->ras_fini(adev); + if (adev->mmhub.ras && adev->mmhub.ras->ras_block.ras_fini) + adev->mmhub.ras->ras_block.ras_fini(adev); - if (adev->gmc.xgmi.ras_funcs && - adev->gmc.xgmi.ras_funcs->ras_fini) - adev->gmc.xgmi.ras_funcs->ras_fini(adev); + if (adev->gmc.xgmi.ras && adev->gmc.xgmi.ras->ras_block.ras_fini) + adev->gmc.xgmi.ras->ras_block.ras_fini(adev); - if (adev->hdp.ras_funcs && - adev->hdp.ras_funcs->ras_fini) - adev->hdp.ras_funcs->ras_fini(adev); + if (adev->hdp.ras && adev->hdp.ras->ras_block.ras_fini) + adev->hdp.ras->ras_block.ras_fini(adev); } /* @@ -831,3 +827,49 @@ void amdgpu_gmc_get_reserved_allocation(struct amdgpu_device *adev) break; } } + +int amdgpu_gmc_vram_checking(struct amdgpu_device *adev) +{ + struct amdgpu_bo *vram_bo = NULL; + uint64_t vram_gpu = 0; + void *vram_ptr = NULL; + + int ret, size = 0x100000; + uint8_t cptr[10]; + + ret = amdgpu_bo_create_kernel(adev, size, PAGE_SIZE, + AMDGPU_GEM_DOMAIN_VRAM, + &vram_bo, + &vram_gpu, + &vram_ptr); + if (ret) + return ret; + + memset(vram_ptr, 0x86, size); + memset(cptr, 0x86, 10); + + /** + * Check the start, the mid, and the end of the memory if the content of + * each byte is the pattern "0x86". If yes, we suppose the vram bo is + * workable. + * + * Note: If check the each byte of whole 1M bo, it will cost too many + * seconds, so here, we just pick up three parts for emulation. + */ + ret = memcmp(vram_ptr, cptr, 10); + if (ret) + return ret; + + ret = memcmp(vram_ptr + (size / 2), cptr, 10); + if (ret) + return ret; + + ret = memcmp(vram_ptr + size - 10, cptr, 10); + if (ret) + return ret; + + amdgpu_bo_free_kernel(&vram_bo, &vram_gpu, + &vram_ptr); + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 8458cebc6d5b..93505bb0a36c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -29,6 +29,7 @@ #include <linux/types.h> #include "amdgpu_irq.h" +#include "amdgpu_ras.h" /* VA hole for 48bit addresses on Vega10 */ #define AMDGPU_GMC_HOLE_START 0x0000800000000000ULL @@ -135,12 +136,8 @@ struct amdgpu_gmc_funcs { unsigned int (*get_vbios_fb_size)(struct amdgpu_device *adev); }; -struct amdgpu_xgmi_ras_funcs { - int (*ras_late_init)(struct amdgpu_device *adev); - void (*ras_fini)(struct amdgpu_device *adev); - int (*query_ras_error_count)(struct amdgpu_device *adev, - void *ras_error_status); - void (*reset_ras_error_count)(struct amdgpu_device *adev); +struct amdgpu_xgmi_ras { + struct amdgpu_ras_block_object ras_block; }; struct amdgpu_xgmi { @@ -159,7 +156,7 @@ struct amdgpu_xgmi { struct ras_common_if *ras_if; bool connected_to_cpu; bool pending_reset; - const struct amdgpu_xgmi_ras_funcs *ras_funcs; + struct amdgpu_xgmi_ras *ras; }; struct amdgpu_gmc { @@ -321,6 +318,7 @@ bool amdgpu_gmc_filter_faults(struct amdgpu_device *adev, uint16_t pasid, uint64_t timestamp); void amdgpu_gmc_filter_faults_remove(struct amdgpu_device *adev, uint64_t addr, uint16_t pasid); +int amdgpu_gmc_ras_early_init(struct amdgpu_device *adev); int amdgpu_gmc_ras_late_init(struct amdgpu_device *adev); void amdgpu_gmc_ras_fini(struct amdgpu_device *adev); int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev); @@ -339,4 +337,5 @@ void amdgpu_gmc_init_pdb0(struct amdgpu_device *adev); uint64_t amdgpu_gmc_vram_mc2pa(struct amdgpu_device *adev, uint64_t mc_addr); uint64_t amdgpu_gmc_vram_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo); uint64_t amdgpu_gmc_vram_cpu_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo); +int amdgpu_gmc_vram_checking(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c index 72022df264f6..899a47011a67 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c @@ -167,6 +167,7 @@ static int amdgpu_gtt_mgr_new(struct ttm_resource_manager *man, return 0; err_free: + ttm_resource_fini(man, &node->base.base); kfree(node); err_out: @@ -198,6 +199,7 @@ static void amdgpu_gtt_mgr_del(struct ttm_resource_manager *man, if (!(res->placement & TTM_PL_FLAG_TEMPORARY)) atomic64_sub(res->num_pages, &mgr->used); + ttm_resource_fini(man, res); kfree(node); } @@ -220,26 +222,21 @@ uint64_t amdgpu_gtt_mgr_usage(struct amdgpu_gtt_mgr *mgr) * * Re-init the gart for each known BO in the GTT. */ -int amdgpu_gtt_mgr_recover(struct amdgpu_gtt_mgr *mgr) +void amdgpu_gtt_mgr_recover(struct amdgpu_gtt_mgr *mgr) { struct amdgpu_gtt_node *node; struct drm_mm_node *mm_node; struct amdgpu_device *adev; - int r = 0; adev = container_of(mgr, typeof(*adev), mman.gtt_mgr); spin_lock(&mgr->lock); drm_mm_for_each_node(mm_node, &mgr->mm) { node = container_of(mm_node, typeof(*node), base.mm_nodes[0]); - r = amdgpu_ttm_recover_gart(node->tbo); - if (r) - break; + amdgpu_ttm_recover_gart(node->tbo); } spin_unlock(&mgr->lock); amdgpu_gart_invalidate_tlb(adev); - - return r; } /** @@ -286,7 +283,8 @@ int amdgpu_gtt_mgr_init(struct amdgpu_device *adev, uint64_t gtt_size) man->use_tt = true; man->func = &amdgpu_gtt_mgr_func; - ttm_resource_manager_init(man, gtt_size >> PAGE_SHIFT); + ttm_resource_manager_init(man, &adev->mman.bdev, + gtt_size >> PAGE_SHIFT); start = AMDGPU_GTT_MAX_TRANSFER_SIZE * AMDGPU_GTT_NUM_TRANSFER_WINDOWS; size = (adev->gmc.gart_size >> PAGE_SHIFT) - start; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c index a766e1aad2b9..518966a26130 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.c @@ -24,7 +24,7 @@ #include "amdgpu.h" #include "amdgpu_ras.h" -int amdgpu_hdp_ras_late_init(struct amdgpu_device *adev) +int amdgpu_hdp_ras_late_init(struct amdgpu_device *adev, void *ras_info) { int r; struct ras_ih_if ih_info = { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h index 7ec99d591584..4af2c2a322e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hdp.h @@ -22,13 +22,10 @@ */ #ifndef __AMDGPU_HDP_H__ #define __AMDGPU_HDP_H__ +#include "amdgpu_ras.h" -struct amdgpu_hdp_ras_funcs { - int (*ras_late_init)(struct amdgpu_device *adev); - void (*ras_fini)(struct amdgpu_device *adev); - void (*query_ras_error_count)(struct amdgpu_device *adev, - void *ras_error_status); - void (*reset_ras_error_count)(struct amdgpu_device *adev); +struct amdgpu_hdp_ras { + struct amdgpu_ras_block_object ras_block; }; struct amdgpu_hdp_funcs { @@ -43,9 +40,9 @@ struct amdgpu_hdp_funcs { struct amdgpu_hdp { struct ras_common_if *ras_if; const struct amdgpu_hdp_funcs *funcs; - const struct amdgpu_hdp_ras_funcs *ras_funcs; + struct amdgpu_hdp_ras *ras; }; -int amdgpu_hdp_ras_late_init(struct amdgpu_device *adev); +int amdgpu_hdp_ras_late_init(struct amdgpu_device *adev, void *ras_info); void amdgpu_hdp_ras_fini(struct amdgpu_device *adev); #endif /* __AMDGPU_HDP_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c index b7fb72bff2c1..be48487e2ca7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c @@ -112,7 +112,7 @@ void amdgpu_pasid_free_delayed(struct dma_resv *resv, unsigned count; int r; - r = dma_resv_get_fences(resv, NULL, &count, &fences); + r = dma_resv_get_fences(resv, true, &count, &fences); if (r) goto fallback; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index f5cbc2747ac6..ea3e8c66211f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -199,13 +199,13 @@ static irqreturn_t amdgpu_irq_handler(int irq, void *arg) * ack the interrupt if it is there */ if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__PCIE_BIF)) { - if (adev->nbio.ras_funcs && - adev->nbio.ras_funcs->handle_ras_controller_intr_no_bifring) - adev->nbio.ras_funcs->handle_ras_controller_intr_no_bifring(adev); + if (adev->nbio.ras && + adev->nbio.ras->handle_ras_controller_intr_no_bifring) + adev->nbio.ras->handle_ras_controller_intr_no_bifring(adev); - if (adev->nbio.ras_funcs && - adev->nbio.ras_funcs->handle_ras_err_event_athub_intr_no_bifring) - adev->nbio.ras_funcs->handle_ras_err_event_athub_intr_no_bifring(adev); + if (adev->nbio.ras && + adev->nbio.ras->handle_ras_err_event_athub_intr_no_bifring) + adev->nbio.ras->handle_ras_err_event_athub_intr_no_bifring(adev); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 1ebb91db2274..9f985bd463be 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -87,11 +87,6 @@ void amdgpu_driver_unload_kms(struct drm_device *dev) if (adev->rmmio == NULL) return; - if (adev->runpm) { - pm_runtime_get_sync(dev->dev); - pm_runtime_forbid(dev->dev); - } - if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DRV_UNLOAD)) DRM_WARN("smart shift update failed\n"); @@ -124,22 +119,6 @@ void amdgpu_register_gpu_instance(struct amdgpu_device *adev) mutex_unlock(&mgpu_info.mutex); } -static void amdgpu_get_audio_func(struct amdgpu_device *adev) -{ - struct pci_dev *p = NULL; - - p = pci_get_domain_bus_and_slot(pci_domain_nr(adev->pdev->bus), - adev->pdev->bus->number, 1); - if (p) { - pm_runtime_get_sync(&p->dev); - - pm_runtime_mark_last_busy(&p->dev); - pm_runtime_put_autosuspend(&p->dev); - - pci_dev_put(p); - } -} - /** * amdgpu_driver_load_kms - Main load function for KMS. * @@ -152,21 +131,10 @@ static void amdgpu_get_audio_func(struct amdgpu_device *adev) int amdgpu_driver_load_kms(struct amdgpu_device *adev, unsigned long flags) { struct drm_device *dev; - struct pci_dev *parent; int r, acpi_status; dev = adev_to_drm(adev); - if (amdgpu_has_atpx() && - (amdgpu_is_atpx_hybrid() || - amdgpu_has_atpx_dgpu_power_cntl()) && - ((flags & AMD_IS_APU) == 0) && - !pci_is_thunderbolt_attached(to_pci_dev(dev->dev))) - flags |= AMD_IS_PX; - - parent = pci_upstream_bridge(adev->pdev); - adev->has_pr3 = parent ? pci_pr3_present(parent) : false; - /* amdgpu_device_init should report only fatal error * like memory allocation failure or iomapping failure, * or memory manager initialization failure, it must @@ -224,58 +192,12 @@ int amdgpu_driver_load_kms(struct amdgpu_device *adev, unsigned long flags) if (acpi_status) dev_dbg(dev->dev, "Error during ACPI methods call\n"); - if (adev->runpm) { - /* only need to skip on ATPX */ - if (amdgpu_device_supports_px(dev)) - dev_pm_set_driver_flags(dev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); - /* we want direct complete for BOCO */ - if (amdgpu_device_supports_boco(dev)) - dev_pm_set_driver_flags(dev->dev, DPM_FLAG_SMART_PREPARE | - DPM_FLAG_SMART_SUSPEND | - DPM_FLAG_MAY_SKIP_RESUME); - pm_runtime_use_autosuspend(dev->dev); - pm_runtime_set_autosuspend_delay(dev->dev, 5000); - - pm_runtime_allow(dev->dev); - - pm_runtime_mark_last_busy(dev->dev); - pm_runtime_put_autosuspend(dev->dev); - - /* - * For runpm implemented via BACO, PMFW will handle the - * timing for BACO in and out: - * - put ASIC into BACO state only when both video and - * audio functions are in D3 state. - * - pull ASIC out of BACO state when either video or - * audio function is in D0 state. - * Also, at startup, PMFW assumes both functions are in - * D0 state. - * - * So if snd driver was loaded prior to amdgpu driver - * and audio function was put into D3 state, there will - * be no PMFW-aware D-state transition(D0->D3) on runpm - * suspend. Thus the BACO will be not correctly kicked in. - * - * Via amdgpu_get_audio_func(), the audio dev is put - * into D0 state. Then there will be a PMFW-aware D-state - * transition(D0->D3) on runpm suspend. - */ - if (amdgpu_device_supports_baco(dev) && - !(adev->flags & AMD_IS_APU) && - (adev->asic_type >= CHIP_NAVI10)) - amdgpu_get_audio_func(adev); - } - if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DRV_LOAD)) DRM_WARN("smart shift update failed\n"); out: - if (r) { - /* balance pm_runtime_get_sync in amdgpu_driver_unload_kms */ - if (adev->rmmio && adev->runpm) - pm_runtime_put_noidle(dev->dev); + if (r) amdgpu_driver_unload_kms(dev); - } return r; } @@ -406,6 +328,10 @@ static int amdgpu_firmware_info(struct drm_amdgpu_info_firmware *fw_info, fw_info->ver = adev->psp.toc.fw_version; fw_info->feature = adev->psp.toc.feature_version; break; + case AMDGPU_INFO_FW_CAP: + fw_info->ver = adev->psp.cap_fw_version; + fw_info->feature = adev->psp.cap_feature_version; + break; default: return -EINVAL; } @@ -1268,18 +1194,20 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_VCE) != NULL) amdgpu_vce_free_handles(adev, file_priv); - amdgpu_vm_bo_rmv(adev, fpriv->prt_va); - if (amdgpu_mcbp || amdgpu_sriov_vf(adev)) { /* TODO: how to handle reserve failure */ BUG_ON(amdgpu_bo_reserve(adev->virt.csa_obj, true)); - amdgpu_vm_bo_rmv(adev, fpriv->csa_va); + amdgpu_vm_bo_del(adev, fpriv->csa_va); fpriv->csa_va = NULL; amdgpu_bo_unreserve(adev->virt.csa_obj); } pasid = fpriv->vm.pasid; pd = amdgpu_bo_ref(fpriv->vm.root.bo); + if (!WARN_ON(amdgpu_bo_reserve(pd, true))) { + amdgpu_vm_bo_del(adev, fpriv->prt_va); + amdgpu_bo_unreserve(pd); + } amdgpu_ctx_mgr_fini(&fpriv->ctx_mgr); amdgpu_vm_fini(adev, &fpriv->vm); @@ -1427,8 +1355,7 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused) struct drm_amdgpu_info_firmware fw_info; struct drm_amdgpu_query_fw query_fw; struct atom_context *ctx = adev->mode_info.atom_context; - uint8_t smu_minor, smu_debug; - uint16_t smu_major; + uint8_t smu_program, smu_major, smu_minor, smu_debug; int ret, i; static const char *ta_fw_name[TA_FW_TYPE_MAX_INDEX] = { @@ -1574,11 +1501,12 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused) ret = amdgpu_firmware_info(&fw_info, &query_fw, adev); if (ret) return ret; - smu_major = (fw_info.ver >> 16) & 0xffff; + smu_program = (fw_info.ver >> 24) & 0xff; + smu_major = (fw_info.ver >> 16) & 0xff; smu_minor = (fw_info.ver >> 8) & 0xff; smu_debug = (fw_info.ver >> 0) & 0xff; - seq_printf(m, "SMC feature version: %u, firmware version: 0x%08x (%d.%d.%d)\n", - fw_info.feature, fw_info.ver, smu_major, smu_minor, smu_debug); + seq_printf(m, "SMC feature version: %u, program: %d, firmware version: 0x%08x (%d.%d.%d)\n", + fw_info.feature, smu_program, fw_info.ver, smu_major, smu_minor, smu_debug); /* SDMA */ query_fw.fw_type = AMDGPU_INFO_FW_SDMA; @@ -1623,6 +1551,16 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused) seq_printf(m, "TOC feature version: %u, firmware version: 0x%08x\n", fw_info.feature, fw_info.ver); + /* CAP */ + if (adev->psp.cap_fw) { + query_fw.fw_type = AMDGPU_INFO_FW_CAP; + ret = amdgpu_firmware_info(&fw_info, &query_fw, adev); + if (ret) + return ret; + seq_printf(m, "CAP feature version: %u, firmware version: 0x%08x\n", + fw_info.feature, fw_info.ver); + } + seq_printf(m, "VBIOS version: %s\n", ctx->vbios_version); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c index ce538f4819f9..52a60c2316a2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c @@ -74,20 +74,23 @@ void amdgpu_mca_query_ras_error_count(struct amdgpu_device *adev, int amdgpu_mca_ras_late_init(struct amdgpu_device *adev, struct amdgpu_mca_ras *mca_dev) { + char sysfs_name[32] = {0}; int r; struct ras_ih_if ih_info = { .cb = NULL, }; - struct ras_fs_if fs_info = { - .sysfs_name = mca_dev->ras_funcs->sysfs_name, + struct ras_fs_if fs_info= { + .sysfs_name = sysfs_name, }; + snprintf(sysfs_name, sizeof(sysfs_name), "%s_err_count", mca_dev->ras->ras_block.name); + if (!mca_dev->ras_if) { mca_dev->ras_if = kmalloc(sizeof(struct ras_common_if), GFP_KERNEL); if (!mca_dev->ras_if) return -ENOMEM; - mca_dev->ras_if->block = mca_dev->ras_funcs->ras_block; - mca_dev->ras_if->sub_block_index = mca_dev->ras_funcs->ras_sub_block; + mca_dev->ras_if->block = mca_dev->ras->ras_block.block; + mca_dev->ras_if->sub_block_index = mca_dev->ras->ras_block.sub_block_index; mca_dev->ras_if->type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE; } ih_info.head = fs_info.head = *mca_dev->ras_if; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h index c74bc7177066..be030c4031d2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h @@ -21,21 +21,13 @@ #ifndef __AMDGPU_MCA_H__ #define __AMDGPU_MCA_H__ -struct amdgpu_mca_ras_funcs { - int (*ras_late_init)(struct amdgpu_device *adev); - void (*ras_fini)(struct amdgpu_device *adev); - void (*query_ras_error_count)(struct amdgpu_device *adev, - void *ras_error_status); - void (*query_ras_error_address)(struct amdgpu_device *adev, - void *ras_error_status); - uint32_t ras_block; - uint32_t ras_sub_block; - const char* sysfs_name; +struct amdgpu_mca_ras_block { + struct amdgpu_ras_block_object ras_block; }; struct amdgpu_mca_ras { struct ras_common_if *ras_if; - const struct amdgpu_mca_ras_funcs *ras_funcs; + struct amdgpu_mca_ras_block *ras; }; struct amdgpu_mca_funcs { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c index 24297dc51434..f9b5472a75d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c @@ -24,7 +24,7 @@ #include "amdgpu.h" #include "amdgpu_ras.h" -int amdgpu_mmhub_ras_late_init(struct amdgpu_device *adev) +int amdgpu_mmhub_ras_late_init(struct amdgpu_device *adev, void *ras_info) { int r; struct ras_ih_if ih_info = { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h index b27fcbccce2b..7deda9a3b81e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.h @@ -21,14 +21,8 @@ #ifndef __AMDGPU_MMHUB_H__ #define __AMDGPU_MMHUB_H__ -struct amdgpu_mmhub_ras_funcs { - int (*ras_late_init)(struct amdgpu_device *adev); - void (*ras_fini)(struct amdgpu_device *adev); - void (*query_ras_error_count)(struct amdgpu_device *adev, - void *ras_error_status); - void (*query_ras_error_status)(struct amdgpu_device *adev); - void (*reset_ras_error_count)(struct amdgpu_device *adev); - void (*reset_ras_error_status)(struct amdgpu_device *adev); +struct amdgpu_mmhub_ras { + struct amdgpu_ras_block_object ras_block; }; struct amdgpu_mmhub_funcs { @@ -50,10 +44,10 @@ struct amdgpu_mmhub_funcs { struct amdgpu_mmhub { struct ras_common_if *ras_if; const struct amdgpu_mmhub_funcs *funcs; - const struct amdgpu_mmhub_ras_funcs *ras_funcs; + struct amdgpu_mmhub_ras *ras; }; -int amdgpu_mmhub_ras_late_init(struct amdgpu_device *adev); +int amdgpu_mmhub_ras_late_init(struct amdgpu_device *adev, void *ras_info); void amdgpu_mmhub_ras_fini(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h index 6043bf6fd414..902235fae4cd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h @@ -33,7 +33,7 @@ #include <drm/drm_crtc.h> #include <drm/drm_edid.h> #include <drm/drm_encoder.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_fixed.h> #include <drm/drm_crtc_helper.h> #include <drm/drm_fb_helper.h> @@ -44,7 +44,7 @@ #include <linux/hrtimer.h> #include "amdgpu_irq.h" -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include "modules/inc/mod_freesync.h" #include "amdgpu_dm_irq_params.h" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c index 6afb02fef8cf..6ace2e390e77 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c @@ -22,7 +22,7 @@ #include "amdgpu.h" #include "amdgpu_ras.h" -int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev) +int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev, void *ras_info) { int r; struct ras_ih_if ih_info = { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h index 843052205bd5..4afb76d3cd97 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h @@ -47,15 +47,12 @@ struct nbio_hdp_flush_reg { u32 ref_and_mask_sdma7; }; -struct amdgpu_nbio_ras_funcs { +struct amdgpu_nbio_ras { + struct amdgpu_ras_block_object ras_block; void (*handle_ras_controller_intr_no_bifring)(struct amdgpu_device *adev); void (*handle_ras_err_event_athub_intr_no_bifring)(struct amdgpu_device *adev); int (*init_ras_controller_interrupt)(struct amdgpu_device *adev); int (*init_ras_err_event_athub_interrupt)(struct amdgpu_device *adev); - void (*query_ras_error_count)(struct amdgpu_device *adev, - void *ras_error_status); - int (*ras_late_init)(struct amdgpu_device *adev); - void (*ras_fini)(struct amdgpu_device *adev); }; struct amdgpu_nbio_funcs { @@ -104,9 +101,9 @@ struct amdgpu_nbio { struct amdgpu_irq_src ras_err_event_athub_irq; struct ras_common_if *ras_if; const struct amdgpu_nbio_funcs *funcs; - const struct amdgpu_nbio_ras_funcs *ras_funcs; + struct amdgpu_nbio_ras *ras; }; -int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev); +int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev, void *ras_info); void amdgpu_nbio_ras_fini(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 5661b82d84d4..23c9a60693ee 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -575,6 +575,9 @@ int amdgpu_bo_create(struct amdgpu_device *adev, if (!amdgpu_bo_support_uswc(bo->flags)) bo->flags &= ~AMDGPU_GEM_CREATE_CPU_GTT_USWC; + if (adev->ras_enabled) + bo->flags |= AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE; + bo->tbo.bdev = &adev->mman.bdev; if (bp->domain & (AMDGPU_GEM_DOMAIN_GWS | AMDGPU_GEM_DOMAIN_OA | AMDGPU_GEM_DOMAIN_GDS)) @@ -1303,7 +1306,8 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo) !(abo->flags & AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE)) return; - dma_resv_lock(bo->base.resv, NULL); + if (WARN_ON_ONCE(!dma_resv_trylock(bo->base.resv))) + return; r = amdgpu_fill_buffer(abo, AMDGPU_POISON, bo->base.resv, &fence); if (!WARN_ON(r)) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_preempt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_preempt_mgr.c index 786afe4f58f9..0d85c2096ab5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_preempt_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_preempt_mgr.c @@ -95,6 +95,7 @@ static void amdgpu_preempt_mgr_del(struct ttm_resource_manager *man, struct amdgpu_preempt_mgr *mgr = to_preempt_mgr(man); atomic64_sub(res->num_pages, &mgr->used); + ttm_resource_fini(man, res); kfree(res); } @@ -152,7 +153,7 @@ int amdgpu_preempt_mgr_init(struct amdgpu_device *adev) man->use_tt = true; man->func = &amdgpu_preempt_mgr_func; - ttm_resource_manager_init(man, (1 << 30)); + ttm_resource_manager_init(man, &adev->mman.bdev, (1 << 30)); atomic64_set(&mgr->used, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index dee17a0e1187..9bc9155cbf06 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -137,7 +137,8 @@ static int psp_early_init(void *handle) psp->autoload_supported = true; break; case IP_VERSION(11, 0, 8): - if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2) { + if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2 || + adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 1, 4)) { psp_v11_0_8_set_psp_funcs(psp); psp->autoload_supported = false; } @@ -259,6 +260,32 @@ static bool psp_get_runtime_db_entry(struct amdgpu_device *adev, return ret; } +static int psp_init_sriov_microcode(struct psp_context *psp) +{ + struct amdgpu_device *adev = psp->adev; + int ret = 0; + + switch (adev->ip_versions[MP0_HWIP][0]) { + case IP_VERSION(9, 0, 0): + ret = psp_init_cap_microcode(psp, "vega10"); + break; + case IP_VERSION(11, 0, 9): + ret = psp_init_cap_microcode(psp, "navi12"); + break; + case IP_VERSION(11, 0, 7): + ret = psp_init_cap_microcode(psp, "sienna_cichlid"); + break; + case IP_VERSION(13, 0, 2): + ret = psp_init_ta_microcode(psp, "aldebaran"); + break; + default: + BUG(); + break; + } + + return ret; +} + static int psp_sw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -273,19 +300,13 @@ static int psp_sw_init(void *handle) ret = -ENOMEM; } - if (!amdgpu_sriov_vf(adev)) { + if (amdgpu_sriov_vf(adev)) + ret = psp_init_sriov_microcode(psp); + else ret = psp_init_microcode(psp); - if (ret) { - DRM_ERROR("Failed to load psp firmware!\n"); - return ret; - } - } else if (amdgpu_sriov_vf(adev) && - adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 2)) { - ret = psp_init_ta_microcode(psp, "aldebaran"); - if (ret) { - DRM_ERROR("Failed to initialize ta microcode!\n"); - return ret; - } + if (ret) { + DRM_ERROR("Failed to load psp firmware!\n"); + return ret; } memset(&boot_cfg_entry, 0, sizeof(boot_cfg_entry)); @@ -353,6 +374,10 @@ static int psp_sw_fini(void *handle) release_firmware(psp->ta_fw); psp->ta_fw = NULL; } + if (adev->psp.cap_fw) { + release_firmware(psp->cap_fw); + psp->cap_fw = NULL; + } if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 0) || adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 7)) @@ -491,7 +516,10 @@ psp_cmd_submit_buf(struct psp_context *psp, DRM_WARN("psp gfx command %s(0x%X) failed and response status is (0x%X)\n", psp_gfx_cmd_name(psp->cmd_buf_mem->cmd_id), psp->cmd_buf_mem->cmd_id, psp->cmd_buf_mem->resp.status); - if (!timeout) { + /* If we load CAP FW, PSP must return 0 under SRIOV + * also return failure in case of timeout + */ + if ((ucode && (ucode->ucode_id == AMDGPU_UCODE_ID_CAP)) || !timeout) { ret = -EINVAL; goto exit; } @@ -914,19 +942,15 @@ static void psp_prep_ta_load_cmd_buf(struct psp_gfx_cmd_resp *cmd, static int psp_ta_init_shared_buf(struct psp_context *psp, struct ta_mem_context *mem_ctx) { - int ret; - /* * Allocate 16k memory aligned to 4k from Frame Buffer (local * physical) for ta to host memory */ - ret = amdgpu_bo_create_kernel(psp->adev, mem_ctx->shared_mem_size, + return amdgpu_bo_create_kernel(psp->adev, mem_ctx->shared_mem_size, PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM, &mem_ctx->shared_bo, &mem_ctx->shared_mc_addr, &mem_ctx->shared_buf); - - return ret; } static void psp_ta_free_shared_buf(struct ta_mem_context *mem_ctx) @@ -1308,6 +1332,11 @@ static void psp_ras_ta_check_status(struct psp_context *psp) break; case TA_RAS_STATUS__SUCCESS: break; + case TA_RAS_STATUS__TEE_ERROR_ACCESS_DENIED: + if (ras_cmd->cmd_id == TA_RAS_COMMAND__TRIGGER_ERROR) + dev_warn(psp->adev->dev, + "RAS WARNING: Inject error to critical region is not allowed\n"); + break; default: dev_warn(psp->adev->dev, "RAS WARNING: ras status = 0x%X\n", ras_cmd->ras_status); @@ -1520,7 +1549,9 @@ int psp_ras_trigger_error(struct psp_context *psp, if (amdgpu_ras_intr_triggered()) return 0; - if (ras_cmd->ras_status) + if (ras_cmd->ras_status == TA_RAS_STATUS__TEE_ERROR_ACCESS_DENIED) + return -EACCES; + else if (ras_cmd->ras_status) return -EINVAL; return 0; @@ -2051,6 +2082,9 @@ static int psp_get_fw_type(struct amdgpu_firmware_info *ucode, enum psp_gfx_fw_type *type) { switch (ucode->ucode_id) { + case AMDGPU_UCODE_ID_CAP: + *type = GFX_FW_TYPE_CAP; + break; case AMDGPU_UCODE_ID_SDMA0: *type = GFX_FW_TYPE_SDMA0; break; @@ -3217,6 +3251,58 @@ out: return err; } +int psp_init_cap_microcode(struct psp_context *psp, + const char *chip_name) +{ + struct amdgpu_device *adev = psp->adev; + char fw_name[PSP_FW_NAME_LEN]; + const struct psp_firmware_header_v1_0 *cap_hdr_v1_0; + struct amdgpu_firmware_info *info = NULL; + int err = 0; + + if (!chip_name) { + dev_err(adev->dev, "invalid chip name for cap microcode\n"); + return -EINVAL; + } + + if (!amdgpu_sriov_vf(adev)) { + dev_err(adev->dev, "cap microcode should only be loaded under SRIOV\n"); + return -EINVAL; + } + + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_cap.bin", chip_name); + err = request_firmware(&adev->psp.cap_fw, fw_name, adev->dev); + if (err) { + dev_warn(adev->dev, "cap microcode does not exist, skip\n"); + err = 0; + goto out; + } + + err = amdgpu_ucode_validate(adev->psp.cap_fw); + if (err) { + dev_err(adev->dev, "fail to initialize cap microcode\n"); + goto out; + } + + info = &adev->firmware.ucode[AMDGPU_UCODE_ID_CAP]; + info->ucode_id = AMDGPU_UCODE_ID_CAP; + info->fw = adev->psp.cap_fw; + cap_hdr_v1_0 = (const struct psp_firmware_header_v1_0 *) + adev->psp.cap_fw->data; + adev->firmware.fw_size += ALIGN( + le32_to_cpu(cap_hdr_v1_0->header.ucode_size_bytes), PAGE_SIZE); + adev->psp.cap_fw_version = le32_to_cpu(cap_hdr_v1_0->header.ucode_version); + adev->psp.cap_feature_version = le32_to_cpu(cap_hdr_v1_0->sos.fw_version); + adev->psp.cap_ucode_size = le32_to_cpu(cap_hdr_v1_0->header.ucode_size_bytes); + + return 0; + +out: + release_firmware(adev->psp.cap_fw); + adev->psp.cap_fw = NULL; + return err; +} + static int psp_set_clockgating_state(void *handle, enum amd_clockgating_state state) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index f29afabbff1f..ff7d533eb746 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -306,6 +306,9 @@ struct psp_context /* toc firmware */ const struct firmware *toc_fw; + /* cap firmware */ + const struct firmware *cap_fw; + /* fence buffer */ struct amdgpu_bo *fence_buf_bo; uint64_t fence_buf_mc_addr; @@ -327,6 +330,10 @@ struct psp_context const struct firmware *ta_fw; uint32_t ta_fw_version; + uint32_t cap_fw_version; + uint32_t cap_feature_version; + uint32_t cap_ucode_size; + struct ta_context asd_context; struct psp_xgmi_context xgmi_context; struct psp_ras_context ras_context; @@ -440,6 +447,8 @@ int psp_init_sos_microcode(struct psp_context *psp, const char *chip_name); int psp_init_ta_microcode(struct psp_context *psp, const char *chip_name); +int psp_init_cap_microcode(struct psp_context *psp, + const char *chip_name); int psp_get_fw_attestation_records_addr(struct psp_context *psp, uint64_t *output_ptr); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 8f47c14ecbc7..a9c133a09be5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -75,6 +75,13 @@ const char *ras_mca_block_string[] = { "mca_iohc", }; +struct amdgpu_ras_block_list { + /* ras block link */ + struct list_head node; + + struct amdgpu_ras_block_object *ras_obj; +}; + const char *get_ras_block_str(struct ras_common_if *ras_block) { if (!ras_block) @@ -89,6 +96,9 @@ const char *get_ras_block_str(struct ras_common_if *ras_block) return ras_block_string[ras_block->block]; } +#define ras_block_str(_BLOCK_) \ + (((_BLOCK_) < ARRAY_SIZE(ras_block_string)) ? ras_block_string[_BLOCK_] : "Out Of Range") + #define ras_err_str(i) (ras_error_string[ffs(i)]) #define RAS_DEFAULT_FLAGS (AMDGPU_RAS_FLAG_INIT_BY_VBIOS) @@ -155,14 +165,9 @@ static int amdgpu_reserve_page_direct(struct amdgpu_device *adev, uint64_t addre } memset(&err_rec, 0x0, sizeof(struct eeprom_table_record)); - - err_rec.address = address; - err_rec.retired_page = address >> AMDGPU_GPU_PAGE_SHIFT; - err_rec.ts = (uint64_t)ktime_get_real_seconds(); - err_rec.err_type = AMDGPU_RAS_EEPROM_ERR_NON_RECOVERABLE; - err_data.err_addr = &err_rec; - err_data.err_addr_cnt = 1; + amdgpu_umc_fill_error_record(&err_data, address, + (address >> AMDGPU_GPU_PAGE_SHIFT), 0, 0); if (amdgpu_bad_page_threshold != 0) { amdgpu_ras_add_bad_pages(adev, err_data.err_addr, @@ -452,7 +457,7 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, } if (ret) - return -EINVAL; + return ret; return size; } @@ -866,30 +871,47 @@ static int amdgpu_ras_enable_all_features(struct amdgpu_device *adev, } /* feature ctl end */ +static int amdgpu_ras_block_match_default(struct amdgpu_ras_block_object *block_obj, + enum amdgpu_ras_block block) +{ + if (!block_obj) + return -EINVAL; + + if (block_obj->block == block) + return 0; -static void amdgpu_ras_mca_query_error_status(struct amdgpu_device *adev, - struct ras_common_if *ras_block, - struct ras_err_data *err_data) + return -EINVAL; +} + +static struct amdgpu_ras_block_object *amdgpu_ras_get_ras_block(struct amdgpu_device *adev, + enum amdgpu_ras_block block, uint32_t sub_block_index) { - switch (ras_block->sub_block_index) { - case AMDGPU_RAS_MCA_BLOCK__MP0: - if (adev->mca.mp0.ras_funcs && - adev->mca.mp0.ras_funcs->query_ras_error_count) - adev->mca.mp0.ras_funcs->query_ras_error_count(adev, &err_data); - break; - case AMDGPU_RAS_MCA_BLOCK__MP1: - if (adev->mca.mp1.ras_funcs && - adev->mca.mp1.ras_funcs->query_ras_error_count) - adev->mca.mp1.ras_funcs->query_ras_error_count(adev, &err_data); - break; - case AMDGPU_RAS_MCA_BLOCK__MPIO: - if (adev->mca.mpio.ras_funcs && - adev->mca.mpio.ras_funcs->query_ras_error_count) - adev->mca.mpio.ras_funcs->query_ras_error_count(adev, &err_data); - break; - default: - break; + struct amdgpu_ras_block_list *node, *tmp; + struct amdgpu_ras_block_object *obj; + + if (block >= AMDGPU_RAS_BLOCK__LAST) + return NULL; + + if (!amdgpu_ras_is_supported(adev, block)) + return NULL; + + list_for_each_entry_safe(node, tmp, &adev->ras_list, node) { + if (!node->ras_obj) { + dev_warn(adev->dev, "Warning: abnormal ras list node.\n"); + continue; + } + + obj = node->ras_obj; + if (obj->ras_block_match) { + if (obj->ras_block_match(obj, block, sub_block_index) == 0) + return obj; + } else { + if (amdgpu_ras_block_match_default(obj, block) == 0) + return obj; + } } + + return NULL; } static void amdgpu_ras_get_ecc_info(struct amdgpu_device *adev, struct ras_err_data *err_data) @@ -901,26 +923,26 @@ static void amdgpu_ras_get_ecc_info(struct amdgpu_device *adev, struct ras_err_d * choosing right query method according to * whether smu support query error information */ - ret = smu_get_ecc_info(&adev->smu, (void *)&(ras->umc_ecc)); + ret = amdgpu_dpm_get_ecc_info(adev, (void *)&(ras->umc_ecc)); if (ret == -EOPNOTSUPP) { - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->query_ras_error_count) - adev->umc.ras_funcs->query_ras_error_count(adev, err_data); + if (adev->umc.ras && adev->umc.ras->ras_block.hw_ops && + adev->umc.ras->ras_block.hw_ops->query_ras_error_count) + adev->umc.ras->ras_block.hw_ops->query_ras_error_count(adev, err_data); /* umc query_ras_error_address is also responsible for clearing * error status */ - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->query_ras_error_address) - adev->umc.ras_funcs->query_ras_error_address(adev, err_data); + if (adev->umc.ras && adev->umc.ras->ras_block.hw_ops && + adev->umc.ras->ras_block.hw_ops->query_ras_error_address) + adev->umc.ras->ras_block.hw_ops->query_ras_error_address(adev, err_data); } else if (!ret) { - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->ecc_info_query_ras_error_count) - adev->umc.ras_funcs->ecc_info_query_ras_error_count(adev, err_data); + if (adev->umc.ras && + adev->umc.ras->ecc_info_query_ras_error_count) + adev->umc.ras->ecc_info_query_ras_error_count(adev, err_data); - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->ecc_info_query_ras_error_address) - adev->umc.ras_funcs->ecc_info_query_ras_error_address(adev, err_data); + if (adev->umc.ras && + adev->umc.ras->ecc_info_query_ras_error_address) + adev->umc.ras->ecc_info_query_ras_error_address(adev, err_data); } } @@ -928,62 +950,32 @@ static void amdgpu_ras_get_ecc_info(struct amdgpu_device *adev, struct ras_err_d int amdgpu_ras_query_error_status(struct amdgpu_device *adev, struct ras_query_if *info) { + struct amdgpu_ras_block_object *block_obj = NULL; struct ras_manager *obj = amdgpu_ras_find_obj(adev, &info->head); struct ras_err_data err_data = {0, 0, 0, NULL}; - int i; if (!obj) return -EINVAL; - switch (info->head.block) { - case AMDGPU_RAS_BLOCK__UMC: + if (info->head.block == AMDGPU_RAS_BLOCK__UMC) { amdgpu_ras_get_ecc_info(adev, &err_data); - break; - case AMDGPU_RAS_BLOCK__SDMA: - if (adev->sdma.funcs->query_ras_error_count) { - for (i = 0; i < adev->sdma.num_instances; i++) - adev->sdma.funcs->query_ras_error_count(adev, i, - &err_data); + } else { + block_obj = amdgpu_ras_get_ras_block(adev, info->head.block, 0); + if (!block_obj || !block_obj->hw_ops) { + dev_dbg_once(adev->dev, "%s doesn't config RAS function\n", + get_ras_block_str(&info->head)); + return -EINVAL; } - break; - case AMDGPU_RAS_BLOCK__GFX: - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->query_ras_error_count) - adev->gfx.ras_funcs->query_ras_error_count(adev, &err_data); - - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->query_ras_error_status) - adev->gfx.ras_funcs->query_ras_error_status(adev); - break; - case AMDGPU_RAS_BLOCK__MMHUB: - if (adev->mmhub.ras_funcs && - adev->mmhub.ras_funcs->query_ras_error_count) - adev->mmhub.ras_funcs->query_ras_error_count(adev, &err_data); - - if (adev->mmhub.ras_funcs && - adev->mmhub.ras_funcs->query_ras_error_status) - adev->mmhub.ras_funcs->query_ras_error_status(adev); - break; - case AMDGPU_RAS_BLOCK__PCIE_BIF: - if (adev->nbio.ras_funcs && - adev->nbio.ras_funcs->query_ras_error_count) - adev->nbio.ras_funcs->query_ras_error_count(adev, &err_data); - break; - case AMDGPU_RAS_BLOCK__XGMI_WAFL: - if (adev->gmc.xgmi.ras_funcs && - adev->gmc.xgmi.ras_funcs->query_ras_error_count) - adev->gmc.xgmi.ras_funcs->query_ras_error_count(adev, &err_data); - break; - case AMDGPU_RAS_BLOCK__HDP: - if (adev->hdp.ras_funcs && - adev->hdp.ras_funcs->query_ras_error_count) - adev->hdp.ras_funcs->query_ras_error_count(adev, &err_data); - break; - case AMDGPU_RAS_BLOCK__MCA: - amdgpu_ras_mca_query_error_status(adev, &info->head, &err_data); - break; - default: - break; + + if (block_obj->hw_ops->query_ras_error_count) + block_obj->hw_ops->query_ras_error_count(adev, &err_data); + + if ((info->head.block == AMDGPU_RAS_BLOCK__SDMA) || + (info->head.block == AMDGPU_RAS_BLOCK__GFX) || + (info->head.block == AMDGPU_RAS_BLOCK__MMHUB)) { + if (block_obj->hw_ops->query_ras_error_status) + block_obj->hw_ops->query_ras_error_status(adev); + } } obj->err_data.ue_count += err_data.ue_count; @@ -1040,68 +1032,27 @@ int amdgpu_ras_query_error_status(struct amdgpu_device *adev, int amdgpu_ras_reset_error_status(struct amdgpu_device *adev, enum amdgpu_ras_block block) { + struct amdgpu_ras_block_object *block_obj = amdgpu_ras_get_ras_block(adev, block, 0); + if (!amdgpu_ras_is_supported(adev, block)) return -EINVAL; - switch (block) { - case AMDGPU_RAS_BLOCK__GFX: - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->reset_ras_error_count) - adev->gfx.ras_funcs->reset_ras_error_count(adev); - - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->reset_ras_error_status) - adev->gfx.ras_funcs->reset_ras_error_status(adev); - break; - case AMDGPU_RAS_BLOCK__MMHUB: - if (adev->mmhub.ras_funcs && - adev->mmhub.ras_funcs->reset_ras_error_count) - adev->mmhub.ras_funcs->reset_ras_error_count(adev); - - if (adev->mmhub.ras_funcs && - adev->mmhub.ras_funcs->reset_ras_error_status) - adev->mmhub.ras_funcs->reset_ras_error_status(adev); - break; - case AMDGPU_RAS_BLOCK__SDMA: - if (adev->sdma.funcs->reset_ras_error_count) - adev->sdma.funcs->reset_ras_error_count(adev); - break; - case AMDGPU_RAS_BLOCK__HDP: - if (adev->hdp.ras_funcs && - adev->hdp.ras_funcs->reset_ras_error_count) - adev->hdp.ras_funcs->reset_ras_error_count(adev); - break; - default: - break; + if (!block_obj || !block_obj->hw_ops) { + dev_dbg_once(adev->dev, "%s doesn't config RAS function\n", + ras_block_str(block)); + return -EINVAL; } - return 0; -} - -/* Trigger XGMI/WAFL error */ -static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev, - struct ta_ras_trigger_error_input *block_info) -{ - int ret; + if (block_obj->hw_ops->reset_ras_error_count) + block_obj->hw_ops->reset_ras_error_count(adev); - if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) - dev_warn(adev->dev, "Failed to disallow df cstate"); - - if (amdgpu_dpm_allow_xgmi_power_down(adev, false)) - dev_warn(adev->dev, "Failed to disallow XGMI power down"); - - ret = psp_ras_trigger_error(&adev->psp, block_info); - - if (amdgpu_ras_intr_triggered()) - return ret; - - if (amdgpu_dpm_allow_xgmi_power_down(adev, true)) - dev_warn(adev->dev, "Failed to allow XGMI power down"); - - if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW)) - dev_warn(adev->dev, "Failed to allow df cstate"); + if ((block == AMDGPU_RAS_BLOCK__GFX) || + (block == AMDGPU_RAS_BLOCK__MMHUB)) { + if (block_obj->hw_ops->reset_ras_error_status) + block_obj->hw_ops->reset_ras_error_status(adev); + } - return ret; + return 0; } /* wrapper of psp_ras_trigger_error */ @@ -1116,11 +1067,20 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, .address = info->address, .value = info->value, }; - int ret = 0; + int ret = -EINVAL; + struct amdgpu_ras_block_object *block_obj = amdgpu_ras_get_ras_block(adev, + info->head.block, + info->head.sub_block_index); if (!obj) return -EINVAL; + if (!block_obj || !block_obj->hw_ops) { + dev_dbg_once(adev->dev, "%s doesn't config RAS function\n", + get_ras_block_str(&info->head)); + return -EINVAL; + } + /* Calculate XGMI relative offset */ if (adev->gmc.xgmi.num_physical_nodes > 1) { block_info.address = @@ -1128,28 +1088,15 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, block_info.address); } - switch (info->head.block) { - case AMDGPU_RAS_BLOCK__GFX: - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->ras_error_inject) - ret = adev->gfx.ras_funcs->ras_error_inject(adev, info); - else - ret = -EINVAL; - break; - case AMDGPU_RAS_BLOCK__UMC: - case AMDGPU_RAS_BLOCK__SDMA: - case AMDGPU_RAS_BLOCK__MMHUB: - case AMDGPU_RAS_BLOCK__PCIE_BIF: - case AMDGPU_RAS_BLOCK__MCA: - ret = psp_ras_trigger_error(&adev->psp, &block_info); - break; - case AMDGPU_RAS_BLOCK__XGMI_WAFL: - ret = amdgpu_ras_error_inject_xgmi(adev, &block_info); - break; - default: - dev_info(adev->dev, "%s error injection is not supported yet\n", - get_ras_block_str(&info->head)); - ret = -EINVAL; + if (info->head.block == AMDGPU_RAS_BLOCK__GFX) { + if (block_obj->hw_ops->ras_error_inject) + ret = block_obj->hw_ops->ras_error_inject(adev, info); + } else { + /* If defined special ras_error_inject(e.g: xgmi), implement special ras_error_inject */ + if (block_obj->hw_ops->ras_error_inject) + ret = block_obj->hw_ops->ras_error_inject(adev, &block_info); + else /*If not defined .ras_error_inject, use default ras_error_inject*/ + ret = psp_ras_trigger_error(&adev->psp, &block_info); } if (ret) @@ -1766,24 +1713,28 @@ static void amdgpu_ras_log_on_err_counter(struct amdgpu_device *adev) static void amdgpu_ras_error_status_query(struct amdgpu_device *adev, struct ras_query_if *info) { + struct amdgpu_ras_block_object *block_obj; /* * Only two block need to query read/write * RspStatus at current state */ - switch (info->head.block) { - case AMDGPU_RAS_BLOCK__GFX: - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->query_ras_error_status) - adev->gfx.ras_funcs->query_ras_error_status(adev); - break; - case AMDGPU_RAS_BLOCK__MMHUB: - if (adev->mmhub.ras_funcs && - adev->mmhub.ras_funcs->query_ras_error_status) - adev->mmhub.ras_funcs->query_ras_error_status(adev); - break; - default: - break; + if ((info->head.block != AMDGPU_RAS_BLOCK__GFX) && + (info->head.block != AMDGPU_RAS_BLOCK__MMHUB)) + return; + + block_obj = amdgpu_ras_get_ras_block(adev, + info->head.block, + info->head.sub_block_index); + + if (!block_obj || !block_obj->hw_ops) { + dev_dbg_once(adev->dev, "%s doesn't config RAS function\n", + get_ras_block_str(&info->head)); + return; } + + if (block_obj->hw_ops->query_ras_error_status) + block_obj->hw_ops->query_ras_error_status(adev); + } static void amdgpu_ras_query_err_status(struct amdgpu_device *adev) @@ -2141,8 +2092,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev) if (ret) goto free; - if (adev->smu.ppt_funcs && adev->smu.ppt_funcs->send_hbm_bad_pages_num) - adev->smu.ppt_funcs->send_hbm_bad_pages_num(&adev->smu, con->eeprom_control.ras_num_recs); + amdgpu_dpm_send_hbm_bad_pages_num(adev, con->eeprom_control.ras_num_recs); } #ifdef CONFIG_X86_MCE_AMD @@ -2348,24 +2298,26 @@ int amdgpu_ras_init(struct amdgpu_device *adev) case CHIP_VEGA20: case CHIP_ARCTURUS: case CHIP_ALDEBARAN: - if (!adev->gmc.xgmi.connected_to_cpu) - adev->nbio.ras_funcs = &nbio_v7_4_ras_funcs; + if (!adev->gmc.xgmi.connected_to_cpu) { + adev->nbio.ras = &nbio_v7_4_ras; + amdgpu_ras_register_ras_block(adev, &adev->nbio.ras->ras_block); + } break; default: /* nbio ras is not available */ break; } - if (adev->nbio.ras_funcs && - adev->nbio.ras_funcs->init_ras_controller_interrupt) { - r = adev->nbio.ras_funcs->init_ras_controller_interrupt(adev); + if (adev->nbio.ras && + adev->nbio.ras->init_ras_controller_interrupt) { + r = adev->nbio.ras->init_ras_controller_interrupt(adev); if (r) goto release_con; } - if (adev->nbio.ras_funcs && - adev->nbio.ras_funcs->init_ras_err_event_athub_interrupt) { - r = adev->nbio.ras_funcs->init_ras_err_event_athub_interrupt(adev); + if (adev->nbio.ras && + adev->nbio.ras->init_ras_err_event_athub_interrupt) { + r = adev->nbio.ras->init_ras_err_event_athub_interrupt(adev); if (r) goto release_con; } @@ -2377,12 +2329,12 @@ int amdgpu_ras_init(struct amdgpu_device *adev) } else if (adev->df.funcs && adev->df.funcs->query_ras_poison_mode && - adev->umc.ras_funcs && - adev->umc.ras_funcs->query_ras_poison_mode) { + adev->umc.ras && + adev->umc.ras->query_ras_poison_mode) { df_poison = adev->df.funcs->query_ras_poison_mode(adev); umc_poison = - adev->umc.ras_funcs->query_ras_poison_mode(adev); + adev->umc.ras->query_ras_poison_mode(adev); /* Only poison is set in both DF and UMC, we can support it */ if (df_poison && umc_poison) con->poison_supported = true; @@ -2585,6 +2537,7 @@ int amdgpu_ras_pre_fini(struct amdgpu_device *adev) int amdgpu_ras_fini(struct amdgpu_device *adev) { + struct amdgpu_ras_block_list *ras_node, *tmp; struct amdgpu_ras *con = amdgpu_ras_get_context(adev); if (!adev->ras_enabled || !con) @@ -2603,6 +2556,12 @@ int amdgpu_ras_fini(struct amdgpu_device *adev) amdgpu_ras_set_context(adev, NULL); kfree(con); + /* Clear ras blocks from ras_list and free ras block list node */ + list_for_each_entry_safe(ras_node, tmp, &adev->ras_list, node) { + list_del(&ras_node->node); + kfree(ras_node); + } + return 0; } @@ -2717,8 +2676,6 @@ static int amdgpu_bad_page_notifier(struct notifier_block *nb, dev_info(adev->dev, "Uncorrectable error detected in UMC inst: %d, chan_idx: %d", umc_inst, ch_inst); - memset(&err_rec, 0x0, sizeof(struct eeprom_table_record)); - /* * Translate UMC channel address to Physical address */ @@ -2730,16 +2687,10 @@ static int amdgpu_bad_page_notifier(struct notifier_block *nb, ADDR_OF_256B_BLOCK(channel_index) | OFFSET_IN_256B_BLOCK(m->addr); - err_rec.address = m->addr; - err_rec.retired_page = retired_page >> AMDGPU_GPU_PAGE_SHIFT; - err_rec.ts = (uint64_t)ktime_get_real_seconds(); - err_rec.err_type = AMDGPU_RAS_EEPROM_ERR_NON_RECOVERABLE; - err_rec.cu = 0; - err_rec.mem_channel = channel_index; - err_rec.mcumc_id = umc_inst; - + memset(&err_rec, 0x0, sizeof(struct eeprom_table_record)); err_data.err_addr = &err_rec; - err_data.err_addr_cnt = 1; + amdgpu_umc_fill_error_record(&err_data, m->addr, + retired_page, channel_index, umc_inst); if (amdgpu_bad_page_threshold != 0) { amdgpu_ras_add_bad_pages(adev, err_data.err_addr, @@ -2777,3 +2728,63 @@ static void amdgpu_register_bad_pages_mca_notifier(struct amdgpu_device *adev) } } #endif + +struct amdgpu_ras *amdgpu_ras_get_context(struct amdgpu_device *adev) +{ + if (!adev) + return NULL; + + return adev->psp.ras_context.ras; +} + +int amdgpu_ras_set_context(struct amdgpu_device *adev, struct amdgpu_ras *ras_con) +{ + if (!adev) + return -EINVAL; + + adev->psp.ras_context.ras = ras_con; + return 0; +} + +/* check if ras is supported on block, say, sdma, gfx */ +int amdgpu_ras_is_supported(struct amdgpu_device *adev, + unsigned int block) +{ + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + if (block >= AMDGPU_RAS_BLOCK_COUNT) + return 0; + return ras && (adev->ras_enabled & (1 << block)); +} + +int amdgpu_ras_reset_gpu(struct amdgpu_device *adev) +{ + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + if (atomic_cmpxchg(&ras->in_recovery, 0, 1) == 0) + schedule_work(&ras->recovery_work); + return 0; +} + + +/* Register each ip ras block into amdgpu ras */ +int amdgpu_ras_register_ras_block(struct amdgpu_device *adev, + struct amdgpu_ras_block_object *ras_block_obj) +{ + struct amdgpu_ras_block_list *ras_node; + if (!adev || !ras_block_obj) + return -EINVAL; + + if (!amdgpu_ras_asic_supported(adev)) + return 0; + + ras_node = kzalloc(sizeof(*ras_node), GFP_KERNEL); + if (!ras_node) + return -ENOMEM; + + INIT_LIST_HEAD(&ras_node->node); + ras_node->ras_obj = ras_block_obj; + list_add_tail(&ras_node->node, &adev->ras_list); + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index 1c708122d492..a55743b12d57 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -26,11 +26,11 @@ #include <linux/debugfs.h> #include <linux/list.h> -#include "amdgpu.h" -#include "amdgpu_psp.h" #include "ta_ras_if.h" #include "amdgpu_ras_eeprom.h" +struct amdgpu_iv_entry; + #define AMDGPU_RAS_FLAG_INIT_BY_VBIOS (0x1 << 0) enum amdgpu_ras_block { @@ -484,6 +484,31 @@ struct ras_debug_if { }; int op; }; + +struct amdgpu_ras_block_object { + /* block name */ + char name[32]; + + enum amdgpu_ras_block block; + + uint32_t sub_block_index; + + int (*ras_block_match)(struct amdgpu_ras_block_object *block_obj, + enum amdgpu_ras_block block, uint32_t sub_block_index); + int (*ras_late_init)(struct amdgpu_device *adev, void *ras_info); + void (*ras_fini)(struct amdgpu_device *adev); + const struct amdgpu_ras_block_hw_ops *hw_ops; +}; + +struct amdgpu_ras_block_hw_ops { + int (*ras_error_inject)(struct amdgpu_device *adev, void *inject_if); + void (*query_ras_error_count)(struct amdgpu_device *adev, void *ras_error_status); + void (*query_ras_error_status)(struct amdgpu_device *adev); + void (*query_ras_error_address)(struct amdgpu_device *adev, void *ras_error_status); + void (*reset_ras_error_count)(struct amdgpu_device *adev); + void (*reset_ras_error_status)(struct amdgpu_device *adev); +}; + /* work flow * vbios * 1: ras feature enable (enabled by default) @@ -498,19 +523,6 @@ struct ras_debug_if { * 8: feature disable */ -#define amdgpu_ras_get_context(adev) ((adev)->psp.ras_context.ras) -#define amdgpu_ras_set_context(adev, ras_con) ((adev)->psp.ras_context.ras = (ras_con)) - -/* check if ras is supported on block, say, sdma, gfx */ -static inline int amdgpu_ras_is_supported(struct amdgpu_device *adev, - unsigned int block) -{ - struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); - - if (block >= AMDGPU_RAS_BLOCK_COUNT) - return 0; - return ras && (adev->ras_enabled & (1 << block)); -} int amdgpu_ras_recovery_init(struct amdgpu_device *adev); @@ -527,15 +539,6 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev, int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev); -static inline int amdgpu_ras_reset_gpu(struct amdgpu_device *adev) -{ - struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); - - if (atomic_cmpxchg(&ras->in_recovery, 0, 1) == 0) - schedule_work(&ras->recovery_work); - return 0; -} - static inline enum ta_ras_block amdgpu_ras_block_to_ta(enum amdgpu_ras_block block) { switch (block) { @@ -667,4 +670,14 @@ const char *get_ras_block_str(struct ras_common_if *ras_block); bool amdgpu_ras_is_poison_mode_supported(struct amdgpu_device *adev); +int amdgpu_ras_is_supported(struct amdgpu_device *adev, unsigned int block); + +int amdgpu_ras_reset_gpu(struct amdgpu_device *adev); + +struct amdgpu_ras* amdgpu_ras_get_context(struct amdgpu_device *adev); + +int amdgpu_ras_set_context(struct amdgpu_device *adev, struct amdgpu_ras *ras_con); + +int amdgpu_ras_register_ras_block(struct amdgpu_device *adev, + struct amdgpu_ras_block_object *ras_block_obj); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index 05117eda105b..2b844a5aafdb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -194,7 +194,7 @@ static int __write_table_header(struct amdgpu_ras_eeprom_control *control) /* i2c may be unstable in gpu reset */ down_read(&adev->reset_sem); - res = amdgpu_eeprom_write(&adev->pm.smu_i2c, + res = amdgpu_eeprom_write(adev->pm.ras_eeprom_i2c_bus, control->i2c_address + control->ras_header_offset, buf, RAS_TABLE_HEADER_SIZE); @@ -263,6 +263,7 @@ static int amdgpu_ras_eeprom_correct_header_tag( */ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control) { + struct amdgpu_device *adev = to_amdgpu_device(control); struct amdgpu_ras_eeprom_table_header *hdr = &control->tbl_hdr; u8 csum; int res; @@ -282,6 +283,8 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control) control->ras_num_recs = 0; control->ras_fri = 0; + amdgpu_dpm_send_hbm_bad_pages_num(adev, control->ras_num_recs); + amdgpu_ras_debugfs_set_ret_size(control); mutex_unlock(&control->ras_tbl_mutex); @@ -389,7 +392,7 @@ static int __amdgpu_ras_eeprom_write(struct amdgpu_ras_eeprom_control *control, /* i2c may be unstable in gpu reset */ down_read(&adev->reset_sem); buf_size = num * RAS_TABLE_RECORD_SIZE; - res = amdgpu_eeprom_write(&adev->pm.smu_i2c, + res = amdgpu_eeprom_write(adev->pm.ras_eeprom_i2c_bus, control->i2c_address + RAS_INDEX_TO_OFFSET(control, fri), buf, buf_size); @@ -548,7 +551,7 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) } down_read(&adev->reset_sem); - res = amdgpu_eeprom_read(&adev->pm.smu_i2c, + res = amdgpu_eeprom_read(adev->pm.ras_eeprom_i2c_bus, control->i2c_address + control->ras_record_offset, buf, buf_size); @@ -644,7 +647,7 @@ static int __amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control, /* i2c may be unstable in gpu reset */ down_read(&adev->reset_sem); buf_size = num * RAS_TABLE_RECORD_SIZE; - res = amdgpu_eeprom_read(&adev->pm.smu_i2c, + res = amdgpu_eeprom_read(adev->pm.ras_eeprom_i2c_bus, control->i2c_address + RAS_INDEX_TO_OFFSET(control, fri), buf, buf_size); @@ -1009,7 +1012,7 @@ static int __verify_ras_table_checksum(struct amdgpu_ras_eeprom_control *control return -ENOMEM; } - res = amdgpu_eeprom_read(&adev->pm.smu_i2c, + res = amdgpu_eeprom_read(adev->pm.ras_eeprom_i2c_bus, control->i2c_address + control->ras_header_offset, buf, buf_size); @@ -1045,7 +1048,7 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control, return 0; /* Verify i2c adapter is initialized */ - if (!adev->pm.smu_i2c.algo) + if (!adev->pm.ras_eeprom_i2c_bus || !adev->pm.ras_eeprom_i2c_bus->algo) return -ENOENT; if (!__get_eeprom_i2c_addr(adev, control)) @@ -1057,7 +1060,7 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control, mutex_init(&control->ras_tbl_mutex); /* Read the table header from EEPROM address */ - res = amdgpu_eeprom_read(&adev->pm.smu_i2c, + res = amdgpu_eeprom_read(adev->pm.ras_eeprom_i2c_bus, control->i2c_address + control->ras_header_offset, buf, RAS_TABLE_HEADER_SIZE); if (res < RAS_TABLE_HEADER_SIZE) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h index 00afd0dcae86..3f671a62b009 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h @@ -127,11 +127,19 @@ struct amdgpu_rlc_funcs { void (*reset)(struct amdgpu_device *adev); void (*start)(struct amdgpu_device *adev); void (*update_spm_vmid)(struct amdgpu_device *adev, unsigned vmid); - void (*sriov_wreg)(struct amdgpu_device *adev, u32 offset, u32 v, u32 acc_flags, u32 hwip); - u32 (*sriov_rreg)(struct amdgpu_device *adev, u32 offset, u32 acc_flags, u32 hwip); bool (*is_rlcg_access_range)(struct amdgpu_device *adev, uint32_t reg); }; +struct amdgpu_rlcg_reg_access_ctrl { + uint32_t scratch_reg0; + uint32_t scratch_reg1; + uint32_t scratch_reg2; + uint32_t scratch_reg3; + uint32_t grbm_cntl; + uint32_t grbm_idx; + uint32_t spare_int; +}; + struct amdgpu_rlc { /* for power gating */ struct amdgpu_bo *save_restore_obj; @@ -191,6 +199,10 @@ struct amdgpu_rlc { struct amdgpu_bo *rlc_toc_bo; uint64_t rlc_toc_gpu_addr; void *rlc_toc_buf; + + bool rlcg_reg_access_supported; + /* registers for rlcg indirect reg access */ + struct amdgpu_rlcg_reg_access_ctrl reg_access_ctrl; }; void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h index f8fb755e3aa6..eaee12ab6518 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h @@ -23,6 +23,7 @@ #ifndef __AMDGPU_SDMA_H__ #define __AMDGPU_SDMA_H__ +#include "amdgpu_ras.h" /* max number of IP instances */ #define AMDGPU_MAX_SDMA_INSTANCES 8 @@ -50,13 +51,8 @@ struct amdgpu_sdma_instance { bool burst_nop; }; -struct amdgpu_sdma_ras_funcs { - int (*ras_late_init)(struct amdgpu_device *adev, - void *ras_ih_info); - void (*ras_fini)(struct amdgpu_device *adev); - int (*query_ras_error_count)(struct amdgpu_device *adev, - uint32_t instance, void *ras_error_status); - void (*reset_ras_error_count)(struct amdgpu_device *adev); +struct amdgpu_sdma_ras { + struct amdgpu_ras_block_object ras_block; }; struct amdgpu_sdma { @@ -73,7 +69,7 @@ struct amdgpu_sdma { uint32_t srbm_soft_reset; bool has_page_queue; struct ras_common_if *ras_if; - const struct amdgpu_sdma_ras_funcs *funcs; + struct amdgpu_sdma_ras *ras; }; /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h index d855cb53c7e0..a48b34d4ce31 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h @@ -358,11 +358,10 @@ TRACE_EVENT(amdgpu_vm_update_ptes, } ), TP_printk("pid:%u vm_ctx:0x%llx start:0x%010llx end:0x%010llx," - " flags:0x%llx, incr:%llu, dst:\n%s%s", __entry->pid, + " flags:0x%llx, incr:%llu, dst:\n%s", __entry->pid, __entry->vm_ctx, __entry->start, __entry->end, __entry->flags, __entry->incr, __print_array( - __get_dynamic_array(dst), min(__entry->nptes, 32u), 8), - __entry->nptes > 32 ? "..." : "") + __get_dynamic_array(dst), __entry->nptes, 8)) ); TRACE_EVENT(amdgpu_vm_set_ptes, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 4655702a5e00..414a22dddc78 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -50,6 +50,7 @@ #include <drm/ttm/ttm_range_manager.h> #include <drm/amdgpu_drm.h> +#include <drm/drm_drv.h> #include "amdgpu.h" #include "amdgpu_object.h" @@ -170,10 +171,10 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, * @bo: buffer object to map * @mem: memory object to map * @mm_cur: range to map - * @num_pages: number of pages to map * @window: which GART window to use * @ring: DMA ring to use for the copy * @tmz: if we should setup a TMZ enabled mapping + * @size: in number of bytes to map, out number of bytes mapped * @addr: resulting address inside the MC address space * * Setup one of the GART windows to access a specific piece of memory or return @@ -182,15 +183,14 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo, struct ttm_resource *mem, struct amdgpu_res_cursor *mm_cur, - unsigned num_pages, unsigned window, - struct amdgpu_ring *ring, bool tmz, - uint64_t *addr) + unsigned window, struct amdgpu_ring *ring, + bool tmz, uint64_t *size, uint64_t *addr) { struct amdgpu_device *adev = ring->adev; - struct amdgpu_job *job; - unsigned num_dw, num_bytes; - struct dma_fence *fence; + unsigned offset, num_pages, num_dw, num_bytes; uint64_t src_addr, dst_addr; + struct dma_fence *fence; + struct amdgpu_job *job; void *cpu_addr; uint64_t flags; unsigned int i; @@ -198,7 +198,9 @@ static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo, BUG_ON(adev->mman.buffer_funcs->copy_max_bytes < AMDGPU_GTT_MAX_TRANSFER_SIZE * 8); - BUG_ON(mem->mem_type == AMDGPU_PL_PREEMPT); + + if (WARN_ON(mem->mem_type == AMDGPU_PL_PREEMPT)) + return -EINVAL; /* Map only what can't be accessed directly */ if (!tmz && mem->start != AMDGPU_BO_INVALID_OFFSET) { @@ -207,10 +209,22 @@ static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo, return 0; } + + /* + * If start begins at an offset inside the page, then adjust the size + * and addr accordingly + */ + offset = mm_cur->start & ~PAGE_MASK; + + num_pages = PFN_UP(*size + offset); + num_pages = min_t(uint32_t, num_pages, AMDGPU_GTT_MAX_TRANSFER_SIZE); + + *size = min(*size, (uint64_t)num_pages * PAGE_SIZE - offset); + *addr = adev->gmc.gart_start; *addr += (u64)window * AMDGPU_GTT_MAX_TRANSFER_SIZE * AMDGPU_GPU_PAGE_SIZE; - *addr += mm_cur->start & ~PAGE_MASK; + *addr += offset; num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8); num_bytes = num_pages * 8 * AMDGPU_GPU_PAGES_IN_CPU_PAGE; @@ -241,10 +255,7 @@ static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo, dma_addr_t *dma_addr; dma_addr = &bo->ttm->dma_address[mm_cur->start >> PAGE_SHIFT]; - r = amdgpu_gart_map(adev, 0, num_pages, dma_addr, flags, - cpu_addr); - if (r) - goto error_free; + amdgpu_gart_map(adev, 0, num_pages, dma_addr, flags, cpu_addr); } else { dma_addr_t dma_address; @@ -252,11 +263,8 @@ static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo, dma_address += adev->vm_manager.vram_base_offset; for (i = 0; i < num_pages; ++i) { - r = amdgpu_gart_map(adev, i << PAGE_SHIFT, 1, - &dma_address, flags, cpu_addr); - if (r) - goto error_free; - + amdgpu_gart_map(adev, i << PAGE_SHIFT, 1, &dma_address, + flags, cpu_addr); dma_address += PAGE_SIZE; } } @@ -297,9 +305,6 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, struct dma_resv *resv, struct dma_fence **f) { - const uint32_t GTT_MAX_BYTES = (AMDGPU_GTT_MAX_TRANSFER_SIZE * - AMDGPU_GPU_PAGE_SIZE); - struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; struct amdgpu_res_cursor src_mm, dst_mm; struct dma_fence *fence = NULL; @@ -315,29 +320,20 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, mutex_lock(&adev->mman.gtt_window_lock); while (src_mm.remaining) { - uint32_t src_page_offset = src_mm.start & ~PAGE_MASK; - uint32_t dst_page_offset = dst_mm.start & ~PAGE_MASK; + uint64_t from, to, cur_size; struct dma_fence *next; - uint32_t cur_size; - uint64_t from, to; - /* Copy size cannot exceed GTT_MAX_BYTES. So if src or dst - * begins at an offset, then adjust the size accordingly - */ - cur_size = max(src_page_offset, dst_page_offset); - cur_size = min(min3(src_mm.size, dst_mm.size, size), - (uint64_t)(GTT_MAX_BYTES - cur_size)); + /* Never copy more than 256MiB at once to avoid a timeout */ + cur_size = min3(src_mm.size, dst_mm.size, 256ULL << 20); /* Map src to window 0 and dst to window 1. */ r = amdgpu_ttm_map_buffer(src->bo, src->mem, &src_mm, - PFN_UP(cur_size + src_page_offset), - 0, ring, tmz, &from); + 0, ring, tmz, &cur_size, &from); if (r) goto error; r = amdgpu_ttm_map_buffer(dst->bo, dst->mem, &dst_mm, - PFN_UP(cur_size + dst_page_offset), - 1, ring, tmz, &to); + 1, ring, tmz, &cur_size, &to); if (r) goto error; @@ -396,8 +392,7 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo, (abo->flags & AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE)) { struct dma_fence *wipe_fence = NULL; - r = amdgpu_fill_buffer(ttm_to_amdgpu_bo(bo), AMDGPU_POISON, - NULL, &wipe_fence); + r = amdgpu_fill_buffer(abo, AMDGPU_POISON, NULL, &wipe_fence); if (r) { goto error; } else if (wipe_fence) { @@ -821,14 +816,13 @@ static void amdgpu_ttm_tt_unpin_userptr(struct ttm_device *bdev, #endif } -static int amdgpu_ttm_gart_bind(struct amdgpu_device *adev, - struct ttm_buffer_object *tbo, - uint64_t flags) +static void amdgpu_ttm_gart_bind(struct amdgpu_device *adev, + struct ttm_buffer_object *tbo, + uint64_t flags) { struct amdgpu_bo *abo = ttm_to_amdgpu_bo(tbo); struct ttm_tt *ttm = tbo->ttm; struct amdgpu_ttm_tt *gtt = (void *)ttm; - int r; if (amdgpu_bo_encrypted(abo)) flags |= AMDGPU_PTE_TMZ; @@ -836,10 +830,8 @@ static int amdgpu_ttm_gart_bind(struct amdgpu_device *adev, if (abo->flags & AMDGPU_GEM_CREATE_CP_MQD_GFX9) { uint64_t page_idx = 1; - r = amdgpu_gart_bind(adev, gtt->offset, page_idx, - gtt->ttm.dma_address, flags); - if (r) - goto gart_bind_fail; + amdgpu_gart_bind(adev, gtt->offset, page_idx, + gtt->ttm.dma_address, flags); /* The memory type of the first page defaults to UC. Now * modify the memory type to NC from the second page of @@ -848,21 +840,13 @@ static int amdgpu_ttm_gart_bind(struct amdgpu_device *adev, flags &= ~AMDGPU_PTE_MTYPE_VG10_MASK; flags |= AMDGPU_PTE_MTYPE_VG10(AMDGPU_MTYPE_NC); - r = amdgpu_gart_bind(adev, - gtt->offset + (page_idx << PAGE_SHIFT), - ttm->num_pages - page_idx, - &(gtt->ttm.dma_address[page_idx]), flags); + amdgpu_gart_bind(adev, gtt->offset + (page_idx << PAGE_SHIFT), + ttm->num_pages - page_idx, + &(gtt->ttm.dma_address[page_idx]), flags); } else { - r = amdgpu_gart_bind(adev, gtt->offset, ttm->num_pages, - gtt->ttm.dma_address, flags); + amdgpu_gart_bind(adev, gtt->offset, ttm->num_pages, + gtt->ttm.dma_address, flags); } - -gart_bind_fail: - if (r) - DRM_ERROR("failed to bind %u pages at 0x%08llX\n", - ttm->num_pages, gtt->offset); - - return r; } /* @@ -878,7 +862,7 @@ static int amdgpu_ttm_backend_bind(struct ttm_device *bdev, struct amdgpu_device *adev = amdgpu_ttm_adev(bdev); struct amdgpu_ttm_tt *gtt = (void*)ttm; uint64_t flags; - int r = 0; + int r; if (!bo_mem) return -EINVAL; @@ -925,14 +909,10 @@ static int amdgpu_ttm_backend_bind(struct ttm_device *bdev, /* bind pages into GART page tables */ gtt->offset = (u64)bo_mem->start << PAGE_SHIFT; - r = amdgpu_gart_bind(adev, gtt->offset, ttm->num_pages, - gtt->ttm.dma_address, flags); - - if (r) - DRM_ERROR("failed to bind %u pages at 0x%08llX\n", - ttm->num_pages, gtt->offset); + amdgpu_gart_bind(adev, gtt->offset, ttm->num_pages, + gtt->ttm.dma_address, flags); gtt->bound = true; - return r; + return 0; } /* @@ -982,12 +962,7 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo) /* Bind pages */ gtt->offset = (u64)tmp->start << PAGE_SHIFT; - r = amdgpu_ttm_gart_bind(adev, bo, flags); - if (unlikely(r)) { - ttm_resource_free(bo, &tmp); - return r; - } - + amdgpu_ttm_gart_bind(adev, bo, flags); amdgpu_gart_invalidate_tlb(adev); ttm_resource_free(bo, &bo->resource); ttm_bo_assign_mem(bo, tmp); @@ -1001,19 +976,16 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo) * Called by amdgpu_gtt_mgr_recover() from amdgpu_device_reset() to * rebind GTT pages during a GPU reset. */ -int amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo) +void amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo) { struct amdgpu_device *adev = amdgpu_ttm_adev(tbo->bdev); uint64_t flags; - int r; if (!tbo->ttm) - return 0; + return; flags = amdgpu_ttm_tt_pte_flags(adev, tbo->ttm, tbo->resource); - r = amdgpu_ttm_gart_bind(adev, tbo, flags); - - return r; + amdgpu_ttm_gart_bind(adev, tbo, flags); } /* @@ -1027,7 +999,6 @@ static void amdgpu_ttm_backend_unbind(struct ttm_device *bdev, { struct amdgpu_device *adev = amdgpu_ttm_adev(bdev); struct amdgpu_ttm_tt *gtt = (void *)ttm; - int r; /* if the pages have userptr pinning then clear that first */ if (gtt->userptr) { @@ -1047,10 +1018,7 @@ static void amdgpu_ttm_backend_unbind(struct ttm_device *bdev, return; /* unbind shouldn't be done for GDS/GWS/OA in ttm_bo_clean_mm */ - r = amdgpu_gart_unbind(adev, gtt->offset, ttm->num_pages); - if (r) - DRM_ERROR("failed to unbind %u pages at 0x%08llX\n", - gtt->ttm.num_pages, gtt->offset); + amdgpu_gart_unbind(adev, gtt->offset, ttm->num_pages); gtt->bound = false; } @@ -1169,6 +1137,26 @@ static void amdgpu_ttm_tt_unpopulate(struct ttm_device *bdev, } /** + * amdgpu_ttm_tt_get_userptr - Return the userptr GTT ttm_tt for the current + * task + * + * @tbo: The ttm_buffer_object that contains the userptr + * @user_addr: The returned value + */ +int amdgpu_ttm_tt_get_userptr(const struct ttm_buffer_object *tbo, + uint64_t *user_addr) +{ + struct amdgpu_ttm_tt *gtt; + + if (!tbo->ttm) + return -EINVAL; + + gtt = (void *)tbo->ttm; + *user_addr = gtt->userptr; + return 0; +} + +/** * amdgpu_ttm_tt_set_userptr - Initialize userptr GTT ttm_tt for the current * task * @@ -1433,6 +1421,63 @@ static void amdgpu_ttm_vram_mm_access(struct amdgpu_device *adev, loff_t pos, } } +static int amdgpu_ttm_access_memory_sdma(struct ttm_buffer_object *bo, + unsigned long offset, void *buf, int len, int write) +{ + struct amdgpu_bo *abo = ttm_to_amdgpu_bo(bo); + struct amdgpu_device *adev = amdgpu_ttm_adev(abo->tbo.bdev); + struct amdgpu_res_cursor src_mm; + struct amdgpu_job *job; + struct dma_fence *fence; + uint64_t src_addr, dst_addr; + unsigned int num_dw; + int r, idx; + + if (len != PAGE_SIZE) + return -EINVAL; + + if (!adev->mman.sdma_access_ptr) + return -EACCES; + + if (!drm_dev_enter(adev_to_drm(adev), &idx)) + return -ENODEV; + + if (write) + memcpy(adev->mman.sdma_access_ptr, buf, len); + + num_dw = ALIGN(adev->mman.buffer_funcs->copy_num_dw, 8); + r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, AMDGPU_IB_POOL_DELAYED, &job); + if (r) + goto out; + + amdgpu_res_first(abo->tbo.resource, offset, len, &src_mm); + src_addr = amdgpu_ttm_domain_start(adev, bo->resource->mem_type) + src_mm.start; + dst_addr = amdgpu_bo_gpu_offset(adev->mman.sdma_access_bo); + if (write) + swap(src_addr, dst_addr); + + amdgpu_emit_copy_buffer(adev, &job->ibs[0], src_addr, dst_addr, PAGE_SIZE, false); + + amdgpu_ring_pad_ib(adev->mman.buffer_funcs_ring, &job->ibs[0]); + WARN_ON(job->ibs[0].length_dw > num_dw); + + r = amdgpu_job_submit(job, &adev->mman.entity, AMDGPU_FENCE_OWNER_UNDEFINED, &fence); + if (r) { + amdgpu_job_free(job); + goto out; + } + + if (!dma_fence_wait_timeout(fence, false, adev->sdma_timeout)) + r = -ETIMEDOUT; + dma_fence_put(fence); + + if (!(r || write)) + memcpy(buf, adev->mman.sdma_access_ptr, len); +out: + drm_dev_exit(idx); + return r; +} + /** * amdgpu_ttm_access_memory - Read or Write memory that backs a buffer object. * @@ -1457,6 +1502,10 @@ static int amdgpu_ttm_access_memory(struct ttm_buffer_object *bo, if (bo->resource->mem_type != TTM_PL_VRAM) return -EIO; + if (amdgpu_device_has_timeouts_enabled(adev) && + !amdgpu_ttm_access_memory_sdma(bo, offset, buf, len, write)) + return len; + amdgpu_res_first(bo->resource, offset, len, &cursor); while (cursor.remaining) { size_t count, size = cursor.size; @@ -1797,6 +1846,12 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) return r; } + if (amdgpu_bo_create_kernel(adev, PAGE_SIZE, PAGE_SIZE, + AMDGPU_GEM_DOMAIN_GTT, + &adev->mman.sdma_access_bo, NULL, + &adev->mman.sdma_access_ptr)) + DRM_WARN("Debug VRAM access will use slowpath MM access\n"); + return 0; } @@ -1818,6 +1873,8 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev) if (adev->mman.stolen_reserved_size) amdgpu_bo_free_kernel(&adev->mman.stolen_reserved_memory, NULL, NULL); + amdgpu_bo_free_kernel(&adev->mman.sdma_access_bo, NULL, + &adev->mman.sdma_access_ptr); amdgpu_ttm_fw_reserve_vram_fini(adev); if (drm_dev_enter(adev_to_drm(adev), &idx)) { @@ -1888,19 +1945,51 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) adev->mman.buffer_funcs_enabled = enable; } +static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev, + bool direct_submit, + unsigned int num_dw, + struct dma_resv *resv, + bool vm_needs_flush, + struct amdgpu_job **job) +{ + enum amdgpu_ib_pool_type pool = direct_submit ? + AMDGPU_IB_POOL_DIRECT : + AMDGPU_IB_POOL_DELAYED; + int r; + + r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, pool, job); + if (r) + return r; + + if (vm_needs_flush) { + (*job)->vm_pd_addr = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo ? + adev->gmc.pdb0_bo : + adev->gart.bo); + (*job)->vm_needs_flush = true; + } + if (resv) { + r = amdgpu_sync_resv(adev, &(*job)->sync, resv, + AMDGPU_SYNC_ALWAYS, + AMDGPU_FENCE_OWNER_UNDEFINED); + if (r) { + DRM_ERROR("sync failed (%d).\n", r); + amdgpu_job_free(*job); + return r; + } + } + return 0; +} + int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, uint64_t dst_offset, uint32_t byte_count, struct dma_resv *resv, struct dma_fence **fence, bool direct_submit, bool vm_needs_flush, bool tmz) { - enum amdgpu_ib_pool_type pool = direct_submit ? AMDGPU_IB_POOL_DIRECT : - AMDGPU_IB_POOL_DELAYED; struct amdgpu_device *adev = ring->adev; + unsigned num_loops, num_dw; struct amdgpu_job *job; - uint32_t max_bytes; - unsigned num_loops, num_dw; unsigned i; int r; @@ -1912,26 +2001,11 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, max_bytes = adev->mman.buffer_funcs->copy_max_bytes; num_loops = DIV_ROUND_UP(byte_count, max_bytes); num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->copy_num_dw, 8); - - r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, pool, &job); + r = amdgpu_ttm_prepare_job(adev, direct_submit, num_dw, + resv, vm_needs_flush, &job); if (r) return r; - if (vm_needs_flush) { - job->vm_pd_addr = amdgpu_gmc_pd_addr(adev->gmc.pdb0_bo ? - adev->gmc.pdb0_bo : adev->gart.bo); - job->vm_needs_flush = true; - } - if (resv) { - r = amdgpu_sync_resv(adev, &job->sync, resv, - AMDGPU_SYNC_ALWAYS, - AMDGPU_FENCE_OWNER_UNDEFINED); - if (r) { - DRM_ERROR("sync failed (%d).\n", r); - goto error_free; - } - } - for (i = 0; i < num_loops; i++) { uint32_t cur_size_in_bytes = min(byte_count, max_bytes); @@ -1961,77 +2035,35 @@ error_free: return r; } -int amdgpu_fill_buffer(struct amdgpu_bo *bo, - uint32_t src_data, - struct dma_resv *resv, - struct dma_fence **fence) +static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data, + uint64_t dst_addr, uint32_t byte_count, + struct dma_resv *resv, + struct dma_fence **fence, + bool vm_needs_flush) { - struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); - uint32_t max_bytes = adev->mman.buffer_funcs->fill_max_bytes; - struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; - - struct amdgpu_res_cursor cursor; + struct amdgpu_device *adev = ring->adev; unsigned int num_loops, num_dw; - uint64_t num_bytes; - struct amdgpu_job *job; + uint32_t max_bytes; + unsigned int i; int r; - if (!adev->mman.buffer_funcs_enabled) { - DRM_ERROR("Trying to clear memory with ring turned off.\n"); - return -EINVAL; - } - - if (bo->tbo.resource->mem_type == AMDGPU_PL_PREEMPT) { - DRM_ERROR("Trying to clear preemptible memory.\n"); - return -EINVAL; - } - - if (bo->tbo.resource->mem_type == TTM_PL_TT) { - r = amdgpu_ttm_alloc_gart(&bo->tbo); - if (r) - return r; - } - - num_bytes = bo->tbo.resource->num_pages << PAGE_SHIFT; - num_loops = 0; - - amdgpu_res_first(bo->tbo.resource, 0, num_bytes, &cursor); - while (cursor.remaining) { - num_loops += DIV_ROUND_UP_ULL(cursor.size, max_bytes); - amdgpu_res_next(&cursor, cursor.size); - } - num_dw = num_loops * adev->mman.buffer_funcs->fill_num_dw; - - /* for IB padding */ - num_dw += 64; - - r = amdgpu_job_alloc_with_ib(adev, num_dw * 4, AMDGPU_IB_POOL_DELAYED, - &job); + max_bytes = adev->mman.buffer_funcs->fill_max_bytes; + num_loops = DIV_ROUND_UP_ULL(byte_count, max_bytes); + num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->fill_num_dw, 8); + r = amdgpu_ttm_prepare_job(adev, false, num_dw, resv, vm_needs_flush, + &job); if (r) return r; - if (resv) { - r = amdgpu_sync_resv(adev, &job->sync, resv, - AMDGPU_SYNC_ALWAYS, - AMDGPU_FENCE_OWNER_UNDEFINED); - if (r) { - DRM_ERROR("sync failed (%d).\n", r); - goto error_free; - } - } - - amdgpu_res_first(bo->tbo.resource, 0, num_bytes, &cursor); - while (cursor.remaining) { - uint32_t cur_size = min_t(uint64_t, cursor.size, max_bytes); - uint64_t dst_addr = cursor.start; + for (i = 0; i < num_loops; i++) { + uint32_t cur_size = min(byte_count, max_bytes); - dst_addr += amdgpu_ttm_domain_start(adev, - bo->tbo.resource->mem_type); amdgpu_emit_fill_buffer(adev, &job->ibs[0], src_data, dst_addr, cur_size); - amdgpu_res_next(&cursor, cur_size); + dst_addr += cur_size; + byte_count -= cur_size; } amdgpu_ring_pad_ib(ring, &job->ibs[0]); @@ -2048,6 +2080,55 @@ error_free: return r; } +int amdgpu_fill_buffer(struct amdgpu_bo *bo, + uint32_t src_data, + struct dma_resv *resv, + struct dma_fence **f) +{ + struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); + struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring; + struct dma_fence *fence = NULL; + struct amdgpu_res_cursor dst; + int r; + + if (!adev->mman.buffer_funcs_enabled) { + DRM_ERROR("Trying to clear memory with ring turned off.\n"); + return -EINVAL; + } + + amdgpu_res_first(bo->tbo.resource, 0, amdgpu_bo_size(bo), &dst); + + mutex_lock(&adev->mman.gtt_window_lock); + while (dst.remaining) { + struct dma_fence *next; + uint64_t cur_size, to; + + /* Never fill more than 256MiB at once to avoid timeouts */ + cur_size = min(dst.size, 256ULL << 20); + + r = amdgpu_ttm_map_buffer(&bo->tbo, bo->tbo.resource, &dst, + 1, ring, false, &cur_size, &to); + if (r) + goto error; + + r = amdgpu_ttm_fill_mem(ring, src_data, to, cur_size, resv, + &next, true); + if (r) + goto error; + + dma_fence_put(fence); + fence = next; + + amdgpu_res_next(&dst, cur_size); + } +error: + mutex_unlock(&adev->mman.gtt_window_lock); + if (f) + *f = dma_fence_get(fence); + dma_fence_put(fence); + return r; +} + /** * amdgpu_ttm_evict_resources - evict memory buffers * @adev: amdgpu device object @@ -2087,7 +2168,7 @@ static int amdgpu_mm_vram_table_show(struct seq_file *m, void *unused) TTM_PL_VRAM); struct drm_printer p = drm_seq_file_printer(m); - man->func->debug(man, &p); + ttm_resource_manager_debug(man, &p); return 0; } @@ -2105,7 +2186,7 @@ static int amdgpu_mm_tt_table_show(struct seq_file *m, void *unused) TTM_PL_TT); struct drm_printer p = drm_seq_file_printer(m); - man->func->debug(man, &p); + ttm_resource_manager_debug(man, &p); return 0; } @@ -2116,7 +2197,7 @@ static int amdgpu_mm_gds_table_show(struct seq_file *m, void *unused) AMDGPU_PL_GDS); struct drm_printer p = drm_seq_file_printer(m); - man->func->debug(man, &p); + ttm_resource_manager_debug(man, &p); return 0; } @@ -2127,7 +2208,7 @@ static int amdgpu_mm_gws_table_show(struct seq_file *m, void *unused) AMDGPU_PL_GWS); struct drm_printer p = drm_seq_file_printer(m); - man->func->debug(man, &p); + ttm_resource_manager_debug(man, &p); return 0; } @@ -2138,7 +2219,7 @@ static int amdgpu_mm_oa_table_show(struct seq_file *m, void *unused) AMDGPU_PL_OA); struct drm_printer p = drm_seq_file_printer(m); - man->func->debug(man, &p); + ttm_resource_manager_debug(man, &p); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index f8f48be16d80..0e4ecc77db3f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -98,6 +98,10 @@ struct amdgpu_mman { u64 fw_vram_usage_size; struct amdgpu_bo *fw_vram_usage_reserved_bo; void *fw_vram_usage_va; + + /* PAGE_SIZE'd BO for process memory r/w over SDMA. */ + struct amdgpu_bo *sdma_access_bo; + void *sdma_access_ptr; }; struct amdgpu_copy_mem { @@ -115,7 +119,7 @@ void amdgpu_vram_mgr_fini(struct amdgpu_device *adev); bool amdgpu_gtt_mgr_has_gart_addr(struct ttm_resource *mem); uint64_t amdgpu_gtt_mgr_usage(struct amdgpu_gtt_mgr *mgr); -int amdgpu_gtt_mgr_recover(struct amdgpu_gtt_mgr *mgr); +void amdgpu_gtt_mgr_recover(struct amdgpu_gtt_mgr *mgr); uint64_t amdgpu_preempt_mgr_usage(struct ttm_resource_manager *man); @@ -158,7 +162,7 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo, struct dma_fence **fence); int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo); -int amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo); +void amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo); uint64_t amdgpu_ttm_domain_start(struct amdgpu_device *adev, uint32_t type); #if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR) @@ -177,6 +181,8 @@ static inline bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm) #endif void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages); +int amdgpu_ttm_tt_get_userptr(const struct ttm_buffer_object *tbo, + uint64_t *user_addr); int amdgpu_ttm_tt_set_userptr(struct ttm_buffer_object *bo, uint64_t addr, uint32_t flags); bool amdgpu_ttm_tt_has_userptr(struct ttm_tt *ttm); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h index 7c2538db3cd5..428f4df184d0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h @@ -378,6 +378,7 @@ enum AMDGPU_UCODE_ID { AMDGPU_UCODE_ID_VCN0_RAM, AMDGPU_UCODE_ID_VCN1_RAM, AMDGPU_UCODE_ID_DMCUB, + AMDGPU_UCODE_ID_CAP, AMDGPU_UCODE_ID_MAXIMUM, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index 46264a4002f7..ff7805beda38 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -21,7 +21,7 @@ * */ -#include "amdgpu_ras.h" +#include "amdgpu.h" static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, void *ras_error_status, @@ -33,14 +33,14 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, int ret = 0; kgd2kfd_set_sram_ecc_flag(adev->kfd.dev); - ret = smu_get_ecc_info(&adev->smu, (void *)&(con->umc_ecc)); + ret = amdgpu_dpm_get_ecc_info(adev, (void *)&(con->umc_ecc)); if (ret == -EOPNOTSUPP) { - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->query_ras_error_count) - adev->umc.ras_funcs->query_ras_error_count(adev, ras_error_status); + if (adev->umc.ras && adev->umc.ras->ras_block.hw_ops && + adev->umc.ras->ras_block.hw_ops->query_ras_error_count) + adev->umc.ras->ras_block.hw_ops->query_ras_error_count(adev, ras_error_status); - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->query_ras_error_address && + if (adev->umc.ras && adev->umc.ras->ras_block.hw_ops && + adev->umc.ras->ras_block.hw_ops->query_ras_error_address && adev->umc.max_ras_err_cnt_per_query) { err_data->err_addr = kcalloc(adev->umc.max_ras_err_cnt_per_query, @@ -56,15 +56,15 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, /* umc query_ras_error_address is also responsible for clearing * error status */ - adev->umc.ras_funcs->query_ras_error_address(adev, ras_error_status); + adev->umc.ras->ras_block.hw_ops->query_ras_error_address(adev, ras_error_status); } } else if (!ret) { - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->ecc_info_query_ras_error_count) - adev->umc.ras_funcs->ecc_info_query_ras_error_count(adev, ras_error_status); + if (adev->umc.ras && + adev->umc.ras->ecc_info_query_ras_error_count) + adev->umc.ras->ecc_info_query_ras_error_count(adev, ras_error_status); - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->ecc_info_query_ras_error_address && + if (adev->umc.ras && + adev->umc.ras->ecc_info_query_ras_error_address && adev->umc.max_ras_err_cnt_per_query) { err_data->err_addr = kcalloc(adev->umc.max_ras_err_cnt_per_query, @@ -80,7 +80,7 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, /* umc query_ras_error_address is also responsible for clearing * error status */ - adev->umc.ras_funcs->ecc_info_query_ras_error_address(adev, ras_error_status); + adev->umc.ras->ecc_info_query_ras_error_address(adev, ras_error_status); } } @@ -96,8 +96,7 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, err_data->err_addr_cnt); amdgpu_ras_save_bad_pages(adev); - if (adev->smu.ppt_funcs && adev->smu.ppt_funcs->send_hbm_bad_pages_num) - adev->smu.ppt_funcs->send_hbm_bad_pages_num(&adev->smu, con->eeprom_control.ras_num_recs); + amdgpu_dpm_send_hbm_bad_pages_num(adev, con->eeprom_control.ras_num_recs); } if (reset) @@ -137,7 +136,7 @@ static int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev, return amdgpu_umc_do_page_retirement(adev, ras_error_status, entry, true); } -int amdgpu_umc_ras_late_init(struct amdgpu_device *adev) +int amdgpu_umc_ras_late_init(struct amdgpu_device *adev, void *ras_info) { int r; struct ras_fs_if fs_info = { @@ -173,9 +172,9 @@ int amdgpu_umc_ras_late_init(struct amdgpu_device *adev) } /* ras init of specific umc version */ - if (adev->umc.ras_funcs && - adev->umc.ras_funcs->err_cnt_init) - adev->umc.ras_funcs->err_cnt_init(adev); + if (adev->umc.ras && + adev->umc.ras->err_cnt_init) + adev->umc.ras->err_cnt_init(adev); return 0; @@ -219,3 +218,24 @@ int amdgpu_umc_process_ecc_irq(struct amdgpu_device *adev, amdgpu_ras_interrupt_dispatch(adev, &ih_data); return 0; } + +void amdgpu_umc_fill_error_record(struct ras_err_data *err_data, + uint64_t err_addr, + uint64_t retired_page, + uint32_t channel_index, + uint32_t umc_inst) +{ + struct eeprom_table_record *err_rec = + &err_data->err_addr[err_data->err_addr_cnt]; + + err_rec->address = err_addr; + /* page frame address is saved */ + err_rec->retired_page = retired_page >> AMDGPU_GPU_PAGE_SHIFT; + err_rec->ts = (uint64_t)ktime_get_real_seconds(); + err_rec->err_type = AMDGPU_RAS_EEPROM_ERR_NON_RECOVERABLE; + err_rec->cu = 0; + err_rec->mem_channel = channel_index; + err_rec->mcumc_id = umc_inst; + + err_data->err_addr_cnt++; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h index b72194e8bfe5..4db0526d0be4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h @@ -20,6 +20,7 @@ */ #ifndef __AMDGPU_UMC_H__ #define __AMDGPU_UMC_H__ +#include "amdgpu_ras.h" /* * (addr / 256) * 4096, the higher 26 bits in ErrorAddr @@ -40,14 +41,9 @@ #define LOOP_UMC_CH_INST(ch_inst) for ((ch_inst) = 0; (ch_inst) < adev->umc.channel_inst_num; (ch_inst)++) #define LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) LOOP_UMC_INST((umc_inst)) LOOP_UMC_CH_INST((ch_inst)) -struct amdgpu_umc_ras_funcs { +struct amdgpu_umc_ras { + struct amdgpu_ras_block_object ras_block; void (*err_cnt_init)(struct amdgpu_device *adev); - int (*ras_late_init)(struct amdgpu_device *adev); - void (*ras_fini)(struct amdgpu_device *adev); - void (*query_ras_error_count)(struct amdgpu_device *adev, - void *ras_error_status); - void (*query_ras_error_address)(struct amdgpu_device *adev, - void *ras_error_status); bool (*query_ras_poison_mode)(struct amdgpu_device *adev); void (*ecc_info_query_ras_error_count)(struct amdgpu_device *adev, void *ras_error_status); @@ -73,10 +69,10 @@ struct amdgpu_umc { struct ras_common_if *ras_if; const struct amdgpu_umc_funcs *funcs; - const struct amdgpu_umc_ras_funcs *ras_funcs; + struct amdgpu_umc_ras *ras; }; -int amdgpu_umc_ras_late_init(struct amdgpu_device *adev); +int amdgpu_umc_ras_late_init(struct amdgpu_device *adev, void *ras_info); void amdgpu_umc_ras_fini(struct amdgpu_device *adev); int amdgpu_umc_poison_handler(struct amdgpu_device *adev, void *ras_error_status, @@ -84,4 +80,9 @@ int amdgpu_umc_poison_handler(struct amdgpu_device *adev, int amdgpu_umc_process_ecc_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry); +void amdgpu_umc_fill_error_record(struct ras_err_data *err_data, + uint64_t err_addr, + uint64_t retired_page, + uint32_t channel_index, + uint32_t umc_inst); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 6f8de11a17f1..9cc23b220537 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -834,6 +834,7 @@ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx, handle = msg[2]; if (handle == 0) { + amdgpu_bo_kunmap(bo); DRM_ERROR("Invalid UVD handle!\n"); return -EINVAL; } @@ -892,6 +893,7 @@ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx, DRM_ERROR("Illegal UVD message type (%d)!\n", msg_type); } + amdgpu_bo_kunmap(bo); return -EINVAL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 07bc0f504713..e1288901beb6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -820,3 +820,148 @@ void amdgpu_virt_update_sriov_video_codec(struct amdgpu_device *adev, } } } + +static bool amdgpu_virt_get_rlcg_reg_access_flag(struct amdgpu_device *adev, + u32 acc_flags, u32 hwip, + bool write, u32 *rlcg_flag) +{ + bool ret = false; + + switch (hwip) { + case GC_HWIP: + if (amdgpu_sriov_reg_indirect_gc(adev)) { + *rlcg_flag = + write ? AMDGPU_RLCG_GC_WRITE : AMDGPU_RLCG_GC_READ; + ret = true; + /* only in new version, AMDGPU_REGS_NO_KIQ and + * AMDGPU_REGS_RLC are enabled simultaneously */ + } else if ((acc_flags & AMDGPU_REGS_RLC) && + !(acc_flags & AMDGPU_REGS_NO_KIQ)) { + *rlcg_flag = AMDGPU_RLCG_GC_WRITE_LEGACY; + ret = true; + } + break; + case MMHUB_HWIP: + if (amdgpu_sriov_reg_indirect_mmhub(adev) && + (acc_flags & AMDGPU_REGS_RLC) && write) { + *rlcg_flag = AMDGPU_RLCG_MMHUB_WRITE; + ret = true; + } + break; + default: + break; + } + return ret; +} + +static u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v, u32 flag) +{ + struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl; + uint32_t timeout = 50000; + uint32_t i, tmp; + uint32_t ret = 0; + static void *scratch_reg0; + static void *scratch_reg1; + static void *scratch_reg2; + static void *scratch_reg3; + static void *spare_int; + + if (!adev->gfx.rlc.rlcg_reg_access_supported) { + dev_err(adev->dev, + "indirect registers access through rlcg is not available\n"); + return 0; + } + + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl; + scratch_reg0 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg0; + scratch_reg1 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg1; + scratch_reg2 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg2; + scratch_reg3 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg3; + if (reg_access_ctrl->spare_int) + spare_int = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->spare_int; + + if (offset == reg_access_ctrl->grbm_cntl) { + /* if the target reg offset is grbm_cntl, write to scratch_reg2 */ + writel(v, scratch_reg2); + writel(v, ((void __iomem *)adev->rmmio) + (offset * 4)); + } else if (offset == reg_access_ctrl->grbm_idx) { + /* if the target reg offset is grbm_idx, write to scratch_reg3 */ + writel(v, scratch_reg3); + writel(v, ((void __iomem *)adev->rmmio) + (offset * 4)); + } else { + /* + * SCRATCH_REG0 = read/write value + * SCRATCH_REG1[30:28] = command + * SCRATCH_REG1[19:0] = address in dword + * SCRATCH_REG1[26:24] = Error reporting + */ + writel(v, scratch_reg0); + writel((offset | flag), scratch_reg1); + if (reg_access_ctrl->spare_int) + writel(1, spare_int); + + for (i = 0; i < timeout; i++) { + tmp = readl(scratch_reg1); + if (!(tmp & flag)) + break; + udelay(10); + } + + if (i >= timeout) { + if (amdgpu_sriov_rlcg_error_report_enabled(adev)) { + if (tmp & AMDGPU_RLCG_VFGATE_DISABLED) { + dev_err(adev->dev, + "vfgate is disabled, rlcg failed to program reg: 0x%05x\n", offset); + } else if (tmp & AMDGPU_RLCG_WRONG_OPERATION_TYPE) { + dev_err(adev->dev, + "wrong operation type, rlcg failed to program reg: 0x%05x\n", offset); + } else if (tmp & AMDGPU_RLCG_REG_NOT_IN_RANGE) { + dev_err(adev->dev, + "regiser is not in range, rlcg failed to program reg: 0x%05x\n", offset); + } else { + dev_err(adev->dev, + "unknown error type, rlcg failed to program reg: 0x%05x\n", offset); + } + } else { + dev_err(adev->dev, + "timeout: rlcg faled to program reg: 0x%05x\n", offset); + } + } + } + + ret = readl(scratch_reg0); + return ret; +} + +void amdgpu_sriov_wreg(struct amdgpu_device *adev, + u32 offset, u32 value, + u32 acc_flags, u32 hwip) +{ + u32 rlcg_flag; + + if (!amdgpu_sriov_runtime(adev) && + amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags, hwip, true, &rlcg_flag)) { + amdgpu_virt_rlcg_reg_rw(adev, offset, value, rlcg_flag); + return; + } + + if (acc_flags & AMDGPU_REGS_NO_KIQ) + WREG32_NO_KIQ(offset, value); + else + WREG32(offset, value); +} + +u32 amdgpu_sriov_rreg(struct amdgpu_device *adev, + u32 offset, u32 acc_flags, u32 hwip) +{ + u32 rlcg_flag; + + if (!amdgpu_sriov_runtime(adev) && + amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags, hwip, false, &rlcg_flag)) + return amdgpu_virt_rlcg_reg_rw(adev, offset, 0, rlcg_flag); + + if (acc_flags & AMDGPU_REGS_NO_KIQ) + return RREG32_NO_KIQ(offset); + else + return RREG32(offset); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index 9adfb8d63280..645093610aa0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -32,6 +32,17 @@ #define AMDGPU_PASSTHROUGH_MODE (1 << 3) /* thw whole GPU is pass through for VM */ #define AMDGPU_SRIOV_CAPS_RUNTIME (1 << 4) /* is out of full access mode */ +/* flags for indirect register access path supported by rlcg for sriov */ +#define AMDGPU_RLCG_GC_WRITE_LEGACY (0x8 << 28) +#define AMDGPU_RLCG_GC_WRITE (0x0 << 28) +#define AMDGPU_RLCG_GC_READ (0x1 << 28) +#define AMDGPU_RLCG_MMHUB_WRITE (0x2 << 28) + +/* error code for indirect register access path supported by rlcg for sriov */ +#define AMDGPU_RLCG_VFGATE_DISABLED 0x4000000 +#define AMDGPU_RLCG_WRONG_OPERATION_TYPE 0x2000000 +#define AMDGPU_RLCG_REG_NOT_IN_RANGE 0x1000000 + /* all asic after AI use this offset */ #define mmRCC_IOV_FUNC_IDENTIFIER 0xDE5 /* tonga/fiji use this offset */ @@ -275,13 +286,18 @@ struct amdgpu_video_codec_info; (amdgpu_sriov_vf((adev)) && \ ((adev)->virt.reg_access & (AMDGIM_FEATURE_GC_REG_RLC_EN))) +#define amdgpu_sriov_rlcg_error_report_enabled(adev) \ + (amdgpu_sriov_reg_indirect_mmhub(adev) || amdgpu_sriov_reg_indirect_gc(adev)) + #define amdgpu_passthrough(adev) \ ((adev)->virt.caps & AMDGPU_PASSTHROUGH_MODE) static inline bool is_virtual_machine(void) { -#ifdef CONFIG_X86 +#if defined(CONFIG_X86) return boot_cpu_has(X86_FEATURE_HYPERVISOR); +#elif defined(CONFIG_ARM64) + return !is_kernel_in_hyp_mode(); #else return false; #endif @@ -293,7 +309,6 @@ static inline bool is_virtual_machine(void) ((!amdgpu_in_reset(adev)) && adev->virt.tdr_debug) #define amdgpu_sriov_is_normal(adev) \ ((!amdgpu_in_reset(adev)) && (!adev->virt.tdr_debug)) - bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev); void amdgpu_virt_init_setting(struct amdgpu_device *adev); void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev, @@ -321,4 +336,9 @@ enum amdgpu_sriov_vf_mode amdgpu_virt_get_sriov_vf_mode(struct amdgpu_device *ad void amdgpu_virt_update_sriov_video_codec(struct amdgpu_device *adev, struct amdgpu_video_codec_info *encode, uint32_t encode_array_size, struct amdgpu_video_codec_info *decode, uint32_t decode_array_size); +void amdgpu_sriov_wreg(struct amdgpu_device *adev, + u32 offset, u32 value, + u32 acc_flags, u32 hwip); +u32 amdgpu_sriov_rreg(struct amdgpu_device *adev, + u32 offset, u32 acc_flags, u32 hwip); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index b37fc7d7d2c7..37acd8911168 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -375,6 +375,8 @@ static void amdgpu_vm_bo_base_init(struct amdgpu_vm_bo_base *base, if (bo->tbo.base.resv != vm->root.bo->tbo.base.resv) return; + dma_resv_assert_held(vm->root.bo->tbo.base.resv); + vm->bulk_moveable = false; if (bo->tbo.type == ttm_bo_type_kernel && bo->parent) amdgpu_vm_bo_relocated(base); @@ -1634,7 +1636,7 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params, nptes = max(nptes, 1u); trace_amdgpu_vm_update_ptes(params, frag_start, upd_end, - nptes, dst, incr, upd_flags, + min(nptes, 32u), dst, incr, upd_flags, vm->task_info.pid, vm->immediate.fence_context); amdgpu_vm_update_flags(params, to_amdgpu_bo_vm(pt), @@ -2257,6 +2259,7 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev, if (!bo) return bo_va; + dma_resv_assert_held(bo->tbo.base.resv); if (amdgpu_dmabuf_is_xgmi_accessible(adev, bo)) { bo_va->is_xgmi = true; /* Power up XGMI if it can be potentially used */ @@ -2634,7 +2637,7 @@ void amdgpu_vm_bo_trace_cs(struct amdgpu_vm *vm, struct ww_acquire_ctx *ticket) } /** - * amdgpu_vm_bo_rmv - remove a bo to a specific vm + * amdgpu_vm_bo_del - remove a bo from a specific vm * * @adev: amdgpu_device pointer * @bo_va: requested bo_va @@ -2643,7 +2646,7 @@ void amdgpu_vm_bo_trace_cs(struct amdgpu_vm *vm, struct ww_acquire_ctx *ticket) * * Object have to be reserved! */ -void amdgpu_vm_bo_rmv(struct amdgpu_device *adev, +void amdgpu_vm_bo_del(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va) { struct amdgpu_bo_va_mapping *mapping, *next; @@ -2651,7 +2654,10 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev, struct amdgpu_vm *vm = bo_va->base.vm; struct amdgpu_vm_bo_base **base; + dma_resv_assert_held(vm->root.bo->tbo.base.resv); + if (bo) { + dma_resv_assert_held(bo->tbo.base.resv); if (bo->tbo.base.resv == vm->root.bo->tbo.base.resv) vm->bulk_moveable = false; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 85fcfb8c5efd..a40a6a993bb0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -435,7 +435,7 @@ int amdgpu_vm_bo_clear_mappings(struct amdgpu_device *adev, struct amdgpu_bo_va_mapping *amdgpu_vm_bo_lookup_mapping(struct amdgpu_vm *vm, uint64_t addr); void amdgpu_vm_bo_trace_cs(struct amdgpu_vm *vm, struct ww_acquire_ctx *ticket); -void amdgpu_vm_bo_rmv(struct amdgpu_device *adev, +void amdgpu_vm_bo_del(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va); void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint32_t min_vm_size, uint32_t fragment_size_default, unsigned max_level, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c index 7a2b487db57c..fce9a13a6ba1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c @@ -281,7 +281,7 @@ int amdgpu_vram_mgr_reserve_range(struct amdgpu_vram_mgr *mgr, rsv->mm_node.size = size >> PAGE_SHIFT; spin_lock(&mgr->lock); - list_add_tail(&mgr->reservations_pending, &rsv->node); + list_add_tail(&rsv->node, &mgr->reservations_pending); amdgpu_vram_mgr_do_reserve(&mgr->manager); spin_unlock(&mgr->lock); @@ -472,6 +472,7 @@ error_free: while (i--) drm_mm_remove_node(&node->mm_nodes[i]); spin_unlock(&mgr->lock); + ttm_resource_fini(man, &node->base); kvfree(node); error_sub: @@ -511,6 +512,7 @@ static void amdgpu_vram_mgr_del(struct ttm_resource_manager *man, atomic64_sub(usage, &mgr->usage); atomic64_sub(vis_usage, &mgr->vis_usage); + ttm_resource_fini(man, res); kvfree(node); } @@ -689,7 +691,8 @@ int amdgpu_vram_mgr_init(struct amdgpu_device *adev) struct amdgpu_vram_mgr *mgr = &adev->mman.vram_mgr; struct ttm_resource_manager *man = &mgr->manager; - ttm_resource_manager_init(man, adev->gmc.real_vram_size >> PAGE_SHIFT); + ttm_resource_manager_init(man, &adev->mman.bdev, + adev->gmc.real_vram_size >> PAGE_SHIFT); man->func = &amdgpu_vram_mgr_func; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index e8b8f28c2f72..5929d6f528c9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -732,7 +732,7 @@ int amdgpu_xgmi_remove_device(struct amdgpu_device *adev) return psp_xgmi_terminate(&adev->psp); } -static int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev) +static int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev, void *ras_info) { int r; struct ras_ih_if ih_info = { @@ -746,7 +746,7 @@ static int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev) adev->gmc.xgmi.num_physical_nodes == 0) return 0; - adev->gmc.xgmi.ras_funcs->reset_ras_error_count(adev); + adev->gmc.xgmi.ras->ras_block.hw_ops->reset_ras_error_count(adev); if (!adev->gmc.xgmi.ras_if) { adev->gmc.xgmi.ras_if = kmalloc(sizeof(struct ras_common_if), GFP_KERNEL); @@ -865,7 +865,7 @@ static int amdgpu_xgmi_query_pcs_error_status(struct amdgpu_device *adev, return 0; } -static int amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, +static void amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status) { struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; @@ -874,7 +874,7 @@ static int amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, uint32_t ue_cnt = 0, ce_cnt = 0; if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL)) - return -EINVAL; + return ; err_data->ue_count = 0; err_data->ce_count = 0; @@ -940,17 +940,51 @@ static int amdgpu_xgmi_query_ras_error_count(struct amdgpu_device *adev, break; } - adev->gmc.xgmi.ras_funcs->reset_ras_error_count(adev); + adev->gmc.xgmi.ras->ras_block.hw_ops->reset_ras_error_count(adev); err_data->ue_count += ue_cnt; err_data->ce_count += ce_cnt; +} - return 0; +/* Trigger XGMI/WAFL error */ +static int amdgpu_ras_error_inject_xgmi(struct amdgpu_device *adev, void *inject_if) +{ + int ret = 0; + struct ta_ras_trigger_error_input *block_info = + (struct ta_ras_trigger_error_input *)inject_if; + + if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_DISALLOW)) + dev_warn(adev->dev, "Failed to disallow df cstate"); + + if (amdgpu_dpm_allow_xgmi_power_down(adev, false)) + dev_warn(adev->dev, "Failed to disallow XGMI power down"); + + ret = psp_ras_trigger_error(&adev->psp, block_info); + + if (amdgpu_ras_intr_triggered()) + return ret; + + if (amdgpu_dpm_allow_xgmi_power_down(adev, true)) + dev_warn(adev->dev, "Failed to allow XGMI power down"); + + if (amdgpu_dpm_set_df_cstate(adev, DF_CSTATE_ALLOW)) + dev_warn(adev->dev, "Failed to allow df cstate"); + + return ret; } -const struct amdgpu_xgmi_ras_funcs xgmi_ras_funcs = { - .ras_late_init = amdgpu_xgmi_ras_late_init, - .ras_fini = amdgpu_xgmi_ras_fini, +struct amdgpu_ras_block_hw_ops xgmi_ras_hw_ops = { .query_ras_error_count = amdgpu_xgmi_query_ras_error_count, .reset_ras_error_count = amdgpu_xgmi_reset_ras_error_count, + .ras_error_inject = amdgpu_ras_error_inject_xgmi, +}; + +struct amdgpu_xgmi_ras xgmi_ras = { + .ras_block = { + .name = "xgmi", + .block = AMDGPU_RAS_BLOCK__XGMI_WAFL, + .hw_ops = &xgmi_ras_hw_ops, + .ras_late_init = amdgpu_xgmi_ras_late_init, + .ras_fini = amdgpu_xgmi_ras_fini, + }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h index d2189bf7d428..0afca51c3c0c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -24,7 +24,7 @@ #include <drm/task_barrier.h> #include "amdgpu_psp.h" - +#include "amdgpu_ras.h" struct amdgpu_hive_info { struct kobject kobj; @@ -50,7 +50,7 @@ struct amdgpu_pcs_ras_field { uint32_t pcs_err_shift; }; -extern const struct amdgpu_xgmi_ras_funcs xgmi_ras_funcs; +extern struct amdgpu_xgmi_ras xgmi_ras; struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev); void amdgpu_put_xgmi_hive(struct amdgpu_hive_info *hive); int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c index 3ea557864320..88642e7ecdf4 100644 --- a/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/athub_v1_0.c @@ -68,12 +68,13 @@ int athub_v1_0_set_clockgating(struct amdgpu_device *adev, if (amdgpu_sriov_vf(adev)) return 0; - switch (adev->asic_type) { - case CHIP_VEGA10: - case CHIP_VEGA12: - case CHIP_VEGA20: - case CHIP_RAVEN: - case CHIP_RENOIR: + switch (adev->ip_versions[ATHUB_HWIP][0]) { + case IP_VERSION(9, 0, 0): + case IP_VERSION(9, 1, 0): + case IP_VERSION(9, 2, 0): + case IP_VERSION(9, 3, 0): + case IP_VERSION(9, 4, 0): + case IP_VERSION(1, 5, 0): athub_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE); athub_update_medium_grain_light_sleep(adev, diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/athub_v2_0.c index ab6a07e5e8c4..a720436857b4 100644 --- a/drivers/gpu/drm/amd/amdgpu/athub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/athub_v2_0.c @@ -78,6 +78,7 @@ int athub_v2_0_set_clockgating(struct amdgpu_device *adev, return 0; switch (adev->ip_versions[ATHUB_HWIP][0]) { + case IP_VERSION(1, 3, 1): case IP_VERSION(2, 0, 0): case IP_VERSION(2, 0, 2): athub_v2_0_update_medium_grain_clock_gating(adev, diff --git a/drivers/gpu/drm/amd/amdgpu/athub_v2_1.c b/drivers/gpu/drm/amd/amdgpu/athub_v2_1.c index 2edefd10e56c..ad8e87d3d2cb 100644 --- a/drivers/gpu/drm/amd/amdgpu/athub_v2_1.c +++ b/drivers/gpu/drm/amd/amdgpu/athub_v2_1.c @@ -74,6 +74,7 @@ int athub_v2_1_set_clockgating(struct amdgpu_device *adev, case IP_VERSION(2, 1, 0): case IP_VERSION(2, 1, 1): case IP_VERSION(2, 1, 2): + case IP_VERSION(2, 4, 0): athub_v2_1_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE); athub_v2_1_update_medium_grain_light_sleep(adev, state == AMD_CG_STATE_GATE); break; diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c index f327becb022f..49a2f594fb2c 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c @@ -34,7 +34,7 @@ #include "atombios_dp.h" #include "amdgpu_connectors.h" #include "amdgpu_atombios.h" -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> /* move these to drm_dp_helper.c/h */ #define DP_LINK_CONFIGURATION_SIZE 9 diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index d1570a462a51..5d5205870861 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -2532,7 +2532,7 @@ static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode) break; } /* adjust pm to dpms */ - amdgpu_pm_compute_clocks(adev); + amdgpu_dpm_compute_clocks(adev); } static void dce_v10_0_crtc_prepare(struct drm_crtc *crtc) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index 18a7b3bd633b..4d812b22c54f 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -2608,7 +2608,7 @@ static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode) break; } /* adjust pm to dpms */ - amdgpu_pm_compute_clocks(adev); + amdgpu_dpm_compute_clocks(adev); } static void dce_v11_0_crtc_prepare(struct drm_crtc *crtc) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index c7803dc2b2d5..b90bc2adf778 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c @@ -2424,7 +2424,7 @@ static void dce_v6_0_crtc_dpms(struct drm_crtc *crtc, int mode) break; } /* adjust pm to dpms */ - amdgpu_pm_compute_clocks(adev); + amdgpu_dpm_compute_clocks(adev); } static void dce_v6_0_crtc_prepare(struct drm_crtc *crtc) diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index 8318ee8339f1..7c1379b02f94 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -2433,7 +2433,7 @@ static void dce_v8_0_crtc_dpms(struct drm_crtc *crtc, int mode) break; } /* adjust pm to dpms */ - amdgpu_pm_compute_clocks(adev); + amdgpu_dpm_compute_clocks(adev); } static void dce_v8_0_crtc_prepare(struct drm_crtc *crtc) diff --git a/drivers/gpu/drm/amd/amdgpu/df_v3_6.c b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c index 43c5e3ec9a39..f4dfca013ec5 100644 --- a/drivers/gpu/drm/amd/amdgpu/df_v3_6.c +++ b/drivers/gpu/drm/amd/amdgpu/df_v3_6.c @@ -458,7 +458,7 @@ static int df_v3_6_pmc_add_cntr(struct amdgpu_device *adev, #define DEFERRED_ARM_MASK (1 << 31) static int df_v3_6_pmc_set_deferred(struct amdgpu_device *adev, - int counter_idx, uint64_t config, + uint64_t config, int counter_idx, bool is_deferred) { @@ -476,8 +476,8 @@ static int df_v3_6_pmc_set_deferred(struct amdgpu_device *adev, } static bool df_v3_6_pmc_is_deferred(struct amdgpu_device *adev, - int counter_idx, - uint64_t config) + uint64_t config, + int counter_idx) { return (df_v3_6_pmc_has_counter(adev, config, counter_idx) && (adev->df_perfmon_config_assign_mask[counter_idx] diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index dbe7442fb25c..8fb4528c741f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -56,10 +56,6 @@ #define GFX10_NUM_GFX_RINGS_Sienna_Cichlid 1 #define GFX10_MEC_HPD_SIZE 2048 -#define RLCG_VFGATE_DISABLED 0x4000000 -#define RLCG_WRONG_OPERATION_TYPE 0x2000000 -#define RLCG_NOT_IN_RANGE 0x1000000 - #define F32_CE_PROGRAM_RAM_SIZE 65536 #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L @@ -180,14 +176,6 @@ #define mmRLC_SPARE_INT_0_Sienna_Cichlid 0x4ca5 #define mmRLC_SPARE_INT_0_Sienna_Cichlid_BASE_IDX 1 -#define GFX_RLCG_GC_WRITE_OLD (0x8 << 28) -#define GFX_RLCG_GC_WRITE (0x0 << 28) -#define GFX_RLCG_GC_READ (0x1 << 28) -#define GFX_RLCG_MMHUB_WRITE (0x2 << 28) - -#define RLCG_ERROR_REPORT_ENABLED(adev) \ - (amdgpu_sriov_reg_indirect_mmhub(adev) || amdgpu_sriov_reg_indirect_gc(adev)) - MODULE_FIRMWARE("amdgpu/navi10_ce.bin"); MODULE_FIRMWARE("amdgpu/navi10_pfp.bin"); MODULE_FIRMWARE("amdgpu/navi10_me.bin"); @@ -1463,143 +1451,6 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_2[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00c00000) }; -static bool gfx_v10_get_rlcg_flag(struct amdgpu_device *adev, u32 acc_flags, u32 hwip, - int write, u32 *rlcg_flag) -{ - switch (hwip) { - case GC_HWIP: - if (amdgpu_sriov_reg_indirect_gc(adev)) { - *rlcg_flag = write ? GFX_RLCG_GC_WRITE : GFX_RLCG_GC_READ; - - return true; - /* only in new version, AMDGPU_REGS_NO_KIQ and AMDGPU_REGS_RLC enabled simultaneously */ - } else if ((acc_flags & AMDGPU_REGS_RLC) && !(acc_flags & AMDGPU_REGS_NO_KIQ)) { - *rlcg_flag = GFX_RLCG_GC_WRITE_OLD; - - return true; - } - - break; - case MMHUB_HWIP: - if (amdgpu_sriov_reg_indirect_mmhub(adev) && - (acc_flags & AMDGPU_REGS_RLC) && write) { - *rlcg_flag = GFX_RLCG_MMHUB_WRITE; - return true; - } - - break; - default: - DRM_DEBUG("Not program register by RLCG\n"); - } - - return false; -} - -static u32 gfx_v10_rlcg_rw(struct amdgpu_device *adev, u32 offset, u32 v, uint32_t flag) -{ - static void *scratch_reg0; - static void *scratch_reg1; - static void *scratch_reg2; - static void *scratch_reg3; - static void *spare_int; - static uint32_t grbm_cntl; - static uint32_t grbm_idx; - uint32_t i = 0; - uint32_t retries = 50000; - u32 ret = 0; - u32 tmp; - - scratch_reg0 = adev->rmmio + - (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0) * 4; - scratch_reg1 = adev->rmmio + - (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG1) * 4; - scratch_reg2 = adev->rmmio + - (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG2) * 4; - scratch_reg3 = adev->rmmio + - (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG3) * 4; - - if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(10, 3, 0)) { - spare_int = adev->rmmio + - (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_0_Sienna_Cichlid_BASE_IDX] - + mmRLC_SPARE_INT_0_Sienna_Cichlid) * 4; - } else { - spare_int = adev->rmmio + - (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_BASE_IDX] + mmRLC_SPARE_INT) * 4; - } - - grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL; - grbm_idx = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_INDEX_BASE_IDX] + mmGRBM_GFX_INDEX; - - if (offset == grbm_cntl || offset == grbm_idx) { - if (offset == grbm_cntl) - writel(v, scratch_reg2); - else if (offset == grbm_idx) - writel(v, scratch_reg3); - - writel(v, ((void __iomem *)adev->rmmio) + (offset * 4)); - } else { - writel(v, scratch_reg0); - writel(offset | flag, scratch_reg1); - writel(1, spare_int); - - for (i = 0; i < retries; i++) { - tmp = readl(scratch_reg1); - if (!(tmp & flag)) - break; - - udelay(10); - } - - if (i >= retries) { - if (RLCG_ERROR_REPORT_ENABLED(adev)) { - if (tmp & RLCG_VFGATE_DISABLED) - pr_err("The vfgate is disabled, program reg:0x%05x failed!\n", offset); - else if (tmp & RLCG_WRONG_OPERATION_TYPE) - pr_err("Wrong operation type, program reg:0x%05x failed!\n", offset); - else if (tmp & RLCG_NOT_IN_RANGE) - pr_err("The register is not in range, program reg:0x%05x failed!\n", offset); - else - pr_err("Unknown error type, program reg:0x%05x failed!\n", offset); - } else - pr_err("timeout: rlcg program reg:0x%05x failed!\n", offset); - } - } - - ret = readl(scratch_reg0); - - return ret; -} - -static void gfx_v10_sriov_wreg(struct amdgpu_device *adev, u32 offset, u32 value, u32 acc_flags, u32 hwip) -{ - u32 rlcg_flag; - - if (!amdgpu_sriov_runtime(adev) && - gfx_v10_get_rlcg_flag(adev, acc_flags, hwip, 1, &rlcg_flag)) { - gfx_v10_rlcg_rw(adev, offset, value, rlcg_flag); - return; - } - - if (acc_flags & AMDGPU_REGS_NO_KIQ) - WREG32_NO_KIQ(offset, value); - else - WREG32(offset, value); -} - -static u32 gfx_v10_sriov_rreg(struct amdgpu_device *adev, u32 offset, u32 acc_flags, u32 hwip) -{ - u32 rlcg_flag; - - if (!amdgpu_sriov_runtime(adev) && - gfx_v10_get_rlcg_flag(adev, acc_flags, hwip, 0, &rlcg_flag)) - return gfx_v10_rlcg_rw(adev, offset, 0, rlcg_flag); - - if (acc_flags & AMDGPU_REGS_NO_KIQ) - return RREG32_NO_KIQ(offset); - else - return RREG32(offset); -} - static const struct soc15_reg_golden golden_settings_gc_10_1_nv14[] = { /* Pending on emulation bring up */ @@ -3790,6 +3641,7 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev) (const u32)ARRAY_SIZE(golden_settings_gc_10_3_5)); break; case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): soc15_program_register_sequence(adev, golden_settings_gc_10_0_cyan_skillfish, (const u32)ARRAY_SIZE(golden_settings_gc_10_0_cyan_skillfish)); @@ -3968,6 +3820,7 @@ static void gfx_v10_0_check_fw_write_wait(struct amdgpu_device *adev) case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): if ((adev->gfx.me_fw_version >= 0x00000046) && (adev->gfx.me_feature_version >= 27) && (adev->gfx.pfp_fw_version >= 0x00000068) && @@ -4108,6 +3961,9 @@ static int gfx_v10_0_init_microcode(struct amdgpu_device *adev) else chip_name = "cyan_skillfish"; break; + case IP_VERSION(10, 1, 4): + chip_name = "cyan_skillfish2"; + break; default: BUG(); } @@ -4448,6 +4304,30 @@ static void gfx_v10_0_rlc_fini(struct amdgpu_device *adev) (void **)&adev->gfx.rlc.cp_table_ptr); } +static void gfx_v10_0_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev) +{ + struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl; + + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl; + reg_access_ctrl->scratch_reg0 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG0); + reg_access_ctrl->scratch_reg1 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG1); + reg_access_ctrl->scratch_reg2 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG2); + reg_access_ctrl->scratch_reg3 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG3); + reg_access_ctrl->grbm_cntl = SOC15_REG_OFFSET(GC, 0, mmGRBM_GFX_CNTL); + reg_access_ctrl->grbm_idx = SOC15_REG_OFFSET(GC, 0, mmGRBM_GFX_INDEX); + switch (adev->ip_versions[GC_HWIP][0]) { + case IP_VERSION(10, 3, 0): + reg_access_ctrl->spare_int = + SOC15_REG_OFFSET(GC, 0, mmRLC_SPARE_INT_0_Sienna_Cichlid); + break; + default: + reg_access_ctrl->spare_int = + SOC15_REG_OFFSET(GC, 0, mmRLC_SPARE_INT); + break; + } + adev->gfx.rlc.rlcg_reg_access_supported = true; +} + static int gfx_v10_0_rlc_init(struct amdgpu_device *adev) { const struct cs_section_def *cs_data; @@ -4468,6 +4348,7 @@ static int gfx_v10_0_rlc_init(struct amdgpu_device *adev) if (adev->gfx.rlc.funcs->update_spm_vmid) adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + return 0; } @@ -4689,6 +4570,7 @@ static void gfx_v10_0_gpu_early_init(struct amdgpu_device *adev) 1 << REG_GET_FIELD(gb_addr_config, GB_ADDR_CONFIG, NUM_PKRS); break; case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): adev->gfx.config.max_hw_contexts = 8; adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; adev->gfx.config.sc_prim_fifo_size_backend = 0x100; @@ -4801,6 +4683,7 @@ static int gfx_v10_0_sw_init(void *handle) case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): adev->gfx.me.num_me = 1; adev->gfx.me.num_pipe_per_me = 1; adev->gfx.me.num_queue_per_pipe = 1; @@ -4865,10 +4748,14 @@ static int gfx_v10_0_sw_init(void *handle) if (r) return r; - r = gfx_v10_0_rlc_init(adev); - if (r) { - DRM_ERROR("Failed to init rlc BOs!\n"); - return r; + if (adev->gfx.rlc.funcs) { + if (adev->gfx.rlc.funcs->init) { + r = adev->gfx.rlc.funcs->init(adev); + if (r) { + dev_err(adev->dev, "Failed to init rlc BOs!\n"); + return r; + } + } } r = gfx_v10_0_mec_init(adev); @@ -7778,6 +7665,7 @@ static int gfx_v10_0_early_init(void *handle) case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): adev->gfx.num_gfx_rings = GFX10_NUM_GFX_RINGS_NV1X; break; case IP_VERSION(10, 3, 0): @@ -7801,6 +7689,9 @@ static int gfx_v10_0_early_init(void *handle) gfx_v10_0_set_gds_init(adev); gfx_v10_0_set_rlc_funcs(adev); + /* init rlcg reg access ctrl */ + gfx_v10_0_init_rlcg_reg_access_ctrl(adev); + return 0; } @@ -8377,8 +8268,6 @@ static const struct amdgpu_rlc_funcs gfx_v10_0_rlc_funcs_sriov = { .reset = gfx_v10_0_rlc_reset, .start = gfx_v10_0_rlc_start, .update_spm_vmid = gfx_v10_0_update_spm_vmid, - .sriov_wreg = gfx_v10_sriov_wreg, - .sriov_rreg = gfx_v10_sriov_rreg, .is_rlcg_access_range = gfx_v10_0_is_rlcg_access_range, }; @@ -9537,6 +9426,7 @@ static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev) case IP_VERSION(10, 1, 10): case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): case IP_VERSION(10, 3, 2): case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 4): diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 9189fb85a4dd..744253be5142 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -63,13 +63,6 @@ #define mmGCEA_PROBE_MAP 0x070c #define mmGCEA_PROBE_MAP_BASE_IDX 0 -#define GFX9_RLCG_GC_WRITE_OLD (0x8 << 28) -#define GFX9_RLCG_GC_WRITE (0x0 << 28) -#define GFX9_RLCG_GC_READ (0x1 << 28) -#define GFX9_RLCG_VFGATE_DISABLED 0x4000000 -#define GFX9_RLCG_WRONG_OPERATION_TYPE 0x2000000 -#define GFX9_RLCG_NOT_IN_RANGE 0x1000000 - MODULE_FIRMWARE("amdgpu/vega10_ce.bin"); MODULE_FIRMWARE("amdgpu/vega10_pfp.bin"); MODULE_FIRMWARE("amdgpu/vega10_me.bin"); @@ -746,128 +739,6 @@ static const u32 GFX_RLC_SRM_INDEX_CNTL_DATA_OFFSETS[] = mmRLC_SRM_INDEX_CNTL_DATA_7 - mmRLC_SRM_INDEX_CNTL_DATA_0, }; -static u32 gfx_v9_0_rlcg_rw(struct amdgpu_device *adev, u32 offset, u32 v, uint32_t flag) -{ - static void *scratch_reg0; - static void *scratch_reg1; - static void *scratch_reg2; - static void *scratch_reg3; - static void *spare_int; - static uint32_t grbm_cntl; - static uint32_t grbm_idx; - uint32_t i = 0; - uint32_t retries = 50000; - u32 ret = 0; - u32 tmp; - - scratch_reg0 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG0_BASE_IDX] + mmSCRATCH_REG0)*4; - scratch_reg1 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG1_BASE_IDX] + mmSCRATCH_REG1)*4; - scratch_reg2 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG2_BASE_IDX] + mmSCRATCH_REG2)*4; - scratch_reg3 = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmSCRATCH_REG3_BASE_IDX] + mmSCRATCH_REG3)*4; - spare_int = adev->rmmio + (adev->reg_offset[GC_HWIP][0][mmRLC_SPARE_INT_BASE_IDX] + mmRLC_SPARE_INT)*4; - - grbm_cntl = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_CNTL_BASE_IDX] + mmGRBM_GFX_CNTL; - grbm_idx = adev->reg_offset[GC_HWIP][0][mmGRBM_GFX_INDEX_BASE_IDX] + mmGRBM_GFX_INDEX; - - if (offset == grbm_cntl || offset == grbm_idx) { - if (offset == grbm_cntl) - writel(v, scratch_reg2); - else if (offset == grbm_idx) - writel(v, scratch_reg3); - - writel(v, ((void __iomem *)adev->rmmio) + (offset * 4)); - } else { - /* - * SCRATCH_REG0 = read/write value - * SCRATCH_REG1[30:28] = command - * SCRATCH_REG1[19:0] = address in dword - * SCRATCH_REG1[26:24] = Error reporting - */ - writel(v, scratch_reg0); - writel(offset | flag, scratch_reg1); - writel(1, spare_int); - - for (i = 0; i < retries; i++) { - tmp = readl(scratch_reg1); - if (!(tmp & flag)) - break; - - udelay(10); - } - - if (i >= retries) { - if (amdgpu_sriov_reg_indirect_gc(adev)) { - if (tmp & GFX9_RLCG_VFGATE_DISABLED) - pr_err("The vfgate is disabled, program reg:0x%05x failed!\n", offset); - else if (tmp & GFX9_RLCG_WRONG_OPERATION_TYPE) - pr_err("Wrong operation type, program reg:0x%05x failed!\n", offset); - else if (tmp & GFX9_RLCG_NOT_IN_RANGE) - pr_err("The register is not in range, program reg:0x%05x failed!\n", offset); - else - pr_err("Unknown error type, program reg:0x%05x failed!\n", offset); - } else - pr_err("timeout: rlcg program reg:0x%05x failed!\n", offset); - } - } - - ret = readl(scratch_reg0); - - return ret; -} - -static bool gfx_v9_0_get_rlcg_flag(struct amdgpu_device *adev, u32 acc_flags, u32 hwip, - int write, u32 *rlcg_flag) -{ - - switch (hwip) { - case GC_HWIP: - if (amdgpu_sriov_reg_indirect_gc(adev)) { - *rlcg_flag = write ? GFX9_RLCG_GC_WRITE : GFX9_RLCG_GC_READ; - - return true; - /* only in new version, AMDGPU_REGS_NO_KIQ and AMDGPU_REGS_RLC enabled simultaneously */ - } else if ((acc_flags & AMDGPU_REGS_RLC) && !(acc_flags & AMDGPU_REGS_NO_KIQ) && write) { - *rlcg_flag = GFX9_RLCG_GC_WRITE_OLD; - return true; - } - - break; - default: - return false; - } - - return false; -} - -static u32 gfx_v9_0_sriov_rreg(struct amdgpu_device *adev, u32 offset, u32 acc_flags, u32 hwip) -{ - u32 rlcg_flag; - - if (!amdgpu_sriov_runtime(adev) && gfx_v9_0_get_rlcg_flag(adev, acc_flags, hwip, 0, &rlcg_flag)) - return gfx_v9_0_rlcg_rw(adev, offset, 0, rlcg_flag); - - if (acc_flags & AMDGPU_REGS_NO_KIQ) - return RREG32_NO_KIQ(offset); - else - return RREG32(offset); -} - -static void gfx_v9_0_sriov_wreg(struct amdgpu_device *adev, u32 offset, - u32 value, u32 acc_flags, u32 hwip) -{ - u32 rlcg_flag; - - if (!amdgpu_sriov_runtime(adev) && gfx_v9_0_get_rlcg_flag(adev, acc_flags, hwip, 1, &rlcg_flag)) { - gfx_v9_0_rlcg_rw(adev, offset, value, rlcg_flag); - return; - } - - if (acc_flags & AMDGPU_REGS_NO_KIQ) - WREG32_NO_KIQ(offset, value); - else - WREG32(offset, value); -} - #define VEGA10_GB_ADDR_CONFIG_GOLDEN 0x2a114042 #define VEGA12_GB_ADDR_CONFIG_GOLDEN 0x24104041 #define RAVEN_GB_ADDR_CONFIG_GOLDEN 0x24000042 @@ -882,7 +753,7 @@ static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev, static uint64_t gfx_v9_0_get_gpu_clock_counter(struct amdgpu_device *adev); static void gfx_v9_0_ring_emit_de_meta(struct amdgpu_ring *ring); static u64 gfx_v9_0_ring_get_rptr_compute(struct amdgpu_ring *ring); -static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, +static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status); static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev, void *inject_if); @@ -2008,6 +1879,21 @@ static int gfx_v9_0_cp_jump_table_num(struct amdgpu_device *adev) return 4; } +static void gfx_v9_0_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev) +{ + struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl; + + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl; + reg_access_ctrl->scratch_reg0 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG0); + reg_access_ctrl->scratch_reg1 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG1); + reg_access_ctrl->scratch_reg2 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG2); + reg_access_ctrl->scratch_reg3 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG3); + reg_access_ctrl->grbm_cntl = SOC15_REG_OFFSET(GC, 0, mmGRBM_GFX_CNTL); + reg_access_ctrl->grbm_idx = SOC15_REG_OFFSET(GC, 0, mmGRBM_GFX_INDEX); + reg_access_ctrl->spare_int = SOC15_REG_OFFSET(GC, 0, mmRLC_SPARE_INT); + adev->gfx.rlc.rlcg_reg_access_supported = true; +} + static int gfx_v9_0_rlc_init(struct amdgpu_device *adev) { const struct cs_section_def *cs_data; @@ -2197,12 +2083,16 @@ static const struct amdgpu_gfx_funcs gfx_v9_0_gfx_funcs = { .select_me_pipe_q = &gfx_v9_0_select_me_pipe_q, }; -static const struct amdgpu_gfx_ras_funcs gfx_v9_0_ras_funcs = { - .ras_late_init = amdgpu_gfx_ras_late_init, - .ras_fini = amdgpu_gfx_ras_fini, - .ras_error_inject = &gfx_v9_0_ras_error_inject, - .query_ras_error_count = &gfx_v9_0_query_ras_error_count, - .reset_ras_error_count = &gfx_v9_0_reset_ras_error_count, +const struct amdgpu_ras_block_hw_ops gfx_v9_0_ras_ops = { + .ras_error_inject = &gfx_v9_0_ras_error_inject, + .query_ras_error_count = &gfx_v9_0_query_ras_error_count, + .reset_ras_error_count = &gfx_v9_0_reset_ras_error_count, +}; + +static struct amdgpu_gfx_ras gfx_v9_0_ras = { + .ras_block = { + .hw_ops = &gfx_v9_0_ras_ops, + }, }; static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) @@ -2231,7 +2121,7 @@ static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) DRM_INFO("fix gfx.config for vega12\n"); break; case IP_VERSION(9, 4, 0): - adev->gfx.ras_funcs = &gfx_v9_0_ras_funcs; + adev->gfx.ras = &gfx_v9_0_ras; adev->gfx.config.max_hw_contexts = 8; adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; adev->gfx.config.sc_prim_fifo_size_backend = 0x100; @@ -2258,7 +2148,7 @@ static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) gb_addr_config = RAVEN_GB_ADDR_CONFIG_GOLDEN; break; case IP_VERSION(9, 4, 1): - adev->gfx.ras_funcs = &gfx_v9_4_ras_funcs; + adev->gfx.ras = &gfx_v9_4_ras; adev->gfx.config.max_hw_contexts = 8; adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; adev->gfx.config.sc_prim_fifo_size_backend = 0x100; @@ -2279,7 +2169,7 @@ static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) gb_addr_config |= 0x22010042; break; case IP_VERSION(9, 4, 2): - adev->gfx.ras_funcs = &gfx_v9_4_2_ras_funcs; + adev->gfx.ras = &gfx_v9_4_2_ras; adev->gfx.config.max_hw_contexts = 8; adev->gfx.config.sc_prim_fifo_size_frontend = 0x20; adev->gfx.config.sc_prim_fifo_size_backend = 0x100; @@ -2298,6 +2188,25 @@ static int gfx_v9_0_gpu_early_init(struct amdgpu_device *adev) break; } + if (adev->gfx.ras) { + err = amdgpu_ras_register_ras_block(adev, &adev->gfx.ras->ras_block); + if (err) { + DRM_ERROR("Failed to register gfx ras block!\n"); + return err; + } + + strcpy(adev->gfx.ras->ras_block.name,"gfx"); + adev->gfx.ras->ras_block.block = AMDGPU_RAS_BLOCK__GFX; + + /* If not define special ras_late_init function, use gfx default ras_late_init */ + if (!adev->gfx.ras->ras_block.ras_late_init) + adev->gfx.ras->ras_block.ras_late_init = amdgpu_gfx_ras_late_init; + + /* If not define special ras_fini function, use gfx default ras_fini */ + if (!adev->gfx.ras->ras_block.ras_fini) + adev->gfx.ras->ras_block.ras_fini = amdgpu_gfx_ras_fini; + } + adev->gfx.config.gb_addr_config = gb_addr_config; adev->gfx.config.gb_addr_config_fields.num_pipes = 1 << @@ -2434,10 +2343,14 @@ static int gfx_v9_0_sw_init(void *handle) return r; } - r = adev->gfx.rlc.funcs->init(adev); - if (r) { - DRM_ERROR("Failed to init rlc BOs!\n"); - return r; + if (adev->gfx.rlc.funcs) { + if (adev->gfx.rlc.funcs->init) { + r = adev->gfx.rlc.funcs->init(adev); + if (r) { + dev_err(adev->dev, "Failed to init rlc BOs!\n"); + return r; + } + } } r = gfx_v9_0_mec_init(adev); @@ -2513,9 +2426,8 @@ static int gfx_v9_0_sw_fini(void *handle) int i; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->ras_fini) - adev->gfx.ras_funcs->ras_fini(adev); + if (adev->gfx.ras && adev->gfx.ras->ras_block.ras_fini) + adev->gfx.ras->ras_block.ras_fini(adev); for (i = 0; i < adev->gfx.num_gfx_rings; i++) amdgpu_ring_fini(&adev->gfx.gfx_ring[i]); @@ -4840,6 +4752,9 @@ static int gfx_v9_0_early_init(void *handle) gfx_v9_0_set_gds_init(adev); gfx_v9_0_set_rlc_funcs(adev); + /* init rlcg reg access ctrl */ + gfx_v9_0_init_rlcg_reg_access_ctrl(adev); + return 0; } @@ -4870,16 +4785,15 @@ static int gfx_v9_0_ecc_late_init(void *handle) if (r) return r; - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->ras_late_init) { - r = adev->gfx.ras_funcs->ras_late_init(adev); + if (adev->gfx.ras && adev->gfx.ras->ras_block.ras_late_init) { + r = adev->gfx.ras->ras_block.ras_late_init(adev, NULL); if (r) return r; } - if (adev->gfx.ras_funcs && - adev->gfx.ras_funcs->enable_watchdog_timer) - adev->gfx.ras_funcs->enable_watchdog_timer(adev); + if (adev->gfx.ras && + adev->gfx.ras->enable_watchdog_timer) + adev->gfx.ras->enable_watchdog_timer(adev); return 0; } @@ -5250,8 +5164,6 @@ static const struct amdgpu_rlc_funcs gfx_v9_0_rlc_funcs = { .reset = gfx_v9_0_rlc_reset, .start = gfx_v9_0_rlc_start, .update_spm_vmid = gfx_v9_0_update_spm_vmid, - .sriov_wreg = gfx_v9_0_sriov_wreg, - .sriov_rreg = gfx_v9_0_sriov_rreg, .is_rlcg_access_range = gfx_v9_0_is_rlcg_access_range, }; @@ -6819,7 +6731,7 @@ static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, mmATC_L2_CACHE_4K_EDC_INDEX, 255); } -static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, +static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status) { struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; @@ -6828,7 +6740,7 @@ static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, uint32_t reg_value; if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) - return -EINVAL; + return; err_data->ue_count = 0; err_data->ce_count = 0; @@ -6857,8 +6769,6 @@ static int gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, mutex_unlock(&adev->grbm_idx_mutex); gfx_v9_0_query_utc_edc_status(adev, err_data); - - return 0; } static void gfx_v9_0_emit_mem_sync(struct amdgpu_ring *ring) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c index b4789dfc2bb9..c67e387a97f5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c @@ -863,7 +863,7 @@ static int gfx_v9_4_ras_error_count(struct amdgpu_device *adev, return 0; } -static int gfx_v9_4_query_ras_error_count(struct amdgpu_device *adev, +static void gfx_v9_4_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status) { struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; @@ -872,7 +872,7 @@ static int gfx_v9_4_query_ras_error_count(struct amdgpu_device *adev, uint32_t reg_value; if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) - return -EINVAL; + return; err_data->ue_count = 0; err_data->ce_count = 0; @@ -903,7 +903,6 @@ static int gfx_v9_4_query_ras_error_count(struct amdgpu_device *adev, gfx_v9_4_query_utc_edc_status(adev, err_data); - return 0; } static void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev) @@ -1029,11 +1028,16 @@ static void gfx_v9_4_query_ras_error_status(struct amdgpu_device *adev) mutex_unlock(&adev->grbm_idx_mutex); } -const struct amdgpu_gfx_ras_funcs gfx_v9_4_ras_funcs = { - .ras_late_init = amdgpu_gfx_ras_late_init, - .ras_fini = amdgpu_gfx_ras_fini, - .ras_error_inject = &gfx_v9_4_ras_error_inject, - .query_ras_error_count = &gfx_v9_4_query_ras_error_count, - .reset_ras_error_count = &gfx_v9_4_reset_ras_error_count, - .query_ras_error_status = &gfx_v9_4_query_ras_error_status, + +const struct amdgpu_ras_block_hw_ops gfx_v9_4_ras_ops = { + .ras_error_inject = &gfx_v9_4_ras_error_inject, + .query_ras_error_count = &gfx_v9_4_query_ras_error_count, + .reset_ras_error_count = &gfx_v9_4_reset_ras_error_count, + .query_ras_error_status = &gfx_v9_4_query_ras_error_status, +}; + +struct amdgpu_gfx_ras gfx_v9_4_ras = { + .ras_block = { + .hw_ops = &gfx_v9_4_ras_ops, + }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h index bdd16b568021..ca520a767267 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.h @@ -24,6 +24,6 @@ #ifndef __GFX_V9_4_H__ #define __GFX_V9_4_H__ -extern const struct amdgpu_gfx_ras_funcs gfx_v9_4_ras_funcs; +extern struct amdgpu_gfx_ras gfx_v9_4_ras; #endif /* __GFX_V9_4_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c index c4f37a161875..7653ebd0e67b 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c @@ -1641,14 +1641,14 @@ static int gfx_v9_4_2_query_utc_edc_count(struct amdgpu_device *adev, return 0; } -static int gfx_v9_4_2_query_ras_error_count(struct amdgpu_device *adev, +static void gfx_v9_4_2_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status) { struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; uint32_t sec_count = 0, ded_count = 0; if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) - return -EINVAL; + return; err_data->ue_count = 0; err_data->ce_count = 0; @@ -1661,7 +1661,6 @@ static int gfx_v9_4_2_query_ras_error_count(struct amdgpu_device *adev, err_data->ce_count += sec_count; err_data->ue_count += ded_count; - return 0; } static void gfx_v9_4_2_reset_utc_err_status(struct amdgpu_device *adev) @@ -1931,13 +1930,17 @@ static void gfx_v9_4_2_reset_sq_timeout_status(struct amdgpu_device *adev) mutex_unlock(&adev->grbm_idx_mutex); } -const struct amdgpu_gfx_ras_funcs gfx_v9_4_2_ras_funcs = { - .ras_late_init = amdgpu_gfx_ras_late_init, - .ras_fini = amdgpu_gfx_ras_fini, - .ras_error_inject = &gfx_v9_4_2_ras_error_inject, - .query_ras_error_count = &gfx_v9_4_2_query_ras_error_count, - .reset_ras_error_count = &gfx_v9_4_2_reset_ras_error_count, - .query_ras_error_status = &gfx_v9_4_2_query_ras_error_status, - .reset_ras_error_status = &gfx_v9_4_2_reset_ras_error_status, +struct amdgpu_ras_block_hw_ops gfx_v9_4_2_ras_ops = { + .ras_error_inject = &gfx_v9_4_2_ras_error_inject, + .query_ras_error_count = &gfx_v9_4_2_query_ras_error_count, + .reset_ras_error_count = &gfx_v9_4_2_reset_ras_error_count, + .query_ras_error_status = &gfx_v9_4_2_query_ras_error_status, + .reset_ras_error_status = &gfx_v9_4_2_reset_ras_error_status, +}; + +struct amdgpu_gfx_ras gfx_v9_4_2_ras = { + .ras_block = { + .hw_ops = &gfx_v9_4_2_ras_ops, + }, .enable_watchdog_timer = &gfx_v9_4_2_enable_watchdog_timer, }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h index 6db1f88509af..7584624b641c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h @@ -31,6 +31,6 @@ void gfx_v9_4_2_init_golden_registers(struct amdgpu_device *adev, void gfx_v9_4_2_set_power_brake_sequence(struct amdgpu_device *adev); int gfx_v9_4_2_do_edc_gpr_workarounds(struct amdgpu_device *adev); -extern const struct amdgpu_gfx_ras_funcs gfx_v9_4_2_ras_funcs; +extern struct amdgpu_gfx_ras gfx_v9_4_2_ras; #endif /* __GFX_V9_4_2_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index a2f8ed0e6a64..c64e3a391c99 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -664,11 +664,25 @@ static void gmc_v10_0_set_umc_funcs(struct amdgpu_device *adev) adev->umc.umc_inst_num = UMC_V8_7_UMC_INSTANCE_NUM; adev->umc.channel_offs = UMC_V8_7_PER_CHANNEL_OFFSET_SIENNA; adev->umc.channel_idx_tbl = &umc_v8_7_channel_idx_tbl[0][0]; - adev->umc.ras_funcs = &umc_v8_7_ras_funcs; + adev->umc.ras = &umc_v8_7_ras; break; default: break; } + if (adev->umc.ras) { + amdgpu_ras_register_ras_block(adev, &adev->umc.ras->ras_block); + + strcpy(adev->umc.ras->ras_block.name, "umc"); + adev->umc.ras->ras_block.block = AMDGPU_RAS_BLOCK__UMC; + + /* If don't define special ras_late_init function, use default ras_late_init */ + if (!adev->umc.ras->ras_block.ras_late_init) + adev->umc.ras->ras_block.ras_late_init = amdgpu_umc_ras_late_init; + + /* If don't define special ras_fini function, use default ras_fini */ + if (!adev->umc.ras->ras_block.ras_fini) + adev->umc.ras->ras_block.ras_fini = amdgpu_umc_ras_fini; + } } @@ -705,6 +719,7 @@ static void gmc_v10_0_set_gfxhub_funcs(struct amdgpu_device *adev) static int gmc_v10_0_early_init(void *handle) { + int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; gmc_v10_0_set_mmhub_funcs(adev); @@ -720,6 +735,10 @@ static int gmc_v10_0_early_init(void *handle) adev->gmc.private_aperture_end = adev->gmc.private_aperture_start + (4ULL << 30) - 1; + r = amdgpu_gmc_ras_early_init(adev); + if (r) + return r; + return 0; } @@ -862,6 +881,7 @@ static int gmc_v10_0_sw_init(void *handle) case IP_VERSION(10, 1, 1): case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): case IP_VERSION(10, 3, 0): case IP_VERSION(10, 3, 2): case IP_VERSION(10, 3, 1): @@ -986,14 +1006,7 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev) return -EINVAL; } - if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) - goto skip_pin_bo; - - r = amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); - if (r) - return r; - -skip_pin_bo: + amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); r = adev->gfxhub.funcs->gart_enable(adev); if (r) return r; @@ -1019,8 +1032,6 @@ skip_pin_bo: (unsigned)(adev->gmc.gart_size >> 20), (unsigned long long)amdgpu_bo_gpu_offset(adev->gart.bo)); - adev->gart.ready = true; - return 0; } @@ -1043,6 +1054,12 @@ static int gmc_v10_0_hw_init(void *handle) if (r) return r; + if (amdgpu_emu_mode == 1) { + r = amdgpu_gmc_vram_checking(adev); + if (r) + return r; + } + if (adev->umc.funcs && adev->umc.funcs->init_registers) adev->umc.funcs->init_registers(adev); @@ -1140,7 +1157,8 @@ static void gmc_v10_0_get_clockgating_state(void *handle, u32 *flags) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 1, 3)) + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 1, 3) || + adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 1, 4)) return; adev->mmhub.funcs->get_clockgating(adev, flags); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index cd6c38e083d0..ec291d28edff 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -469,16 +469,14 @@ static void gmc_v6_0_set_prt(struct amdgpu_device *adev, bool enable) static int gmc_v6_0_gart_enable(struct amdgpu_device *adev) { uint64_t table_addr; - int r, i; u32 field; + int i; if (adev->gart.bo == NULL) { dev_err(adev->dev, "No VRAM object for PCIE GART.\n"); return -EINVAL; } - r = amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); - if (r) - return r; + amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); table_addr = amdgpu_bo_gpu_offset(adev->gart.bo); @@ -558,7 +556,6 @@ static int gmc_v6_0_gart_enable(struct amdgpu_device *adev) dev_info(adev->dev, "PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), (unsigned long long)table_addr); - adev->gart.ready = true; return 0; } @@ -922,7 +919,10 @@ static int gmc_v6_0_hw_init(void *handle) if (r) return r; - return r; + if (amdgpu_emu_mode == 1) + return amdgpu_gmc_vram_checking(adev); + else + return r; } static int gmc_v6_0_hw_fini(void *handle) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index ab8adbff9e2d..344d819b4c1b 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -613,17 +613,14 @@ static void gmc_v7_0_set_prt(struct amdgpu_device *adev, bool enable) static int gmc_v7_0_gart_enable(struct amdgpu_device *adev) { uint64_t table_addr; - int r, i; u32 tmp, field; + int i; if (adev->gart.bo == NULL) { dev_err(adev->dev, "No VRAM object for PCIE GART.\n"); return -EINVAL; } - r = amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); - if (r) - return r; - + amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); table_addr = amdgpu_bo_gpu_offset(adev->gart.bo); /* Setup TLB control */ @@ -712,7 +709,6 @@ static int gmc_v7_0_gart_enable(struct amdgpu_device *adev) DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), (unsigned long long)table_addr); - adev->gart.ready = true; return 0; } @@ -1111,7 +1107,10 @@ static int gmc_v7_0_hw_init(void *handle) if (r) return r; - return r; + if (amdgpu_emu_mode == 1) + return amdgpu_gmc_vram_checking(adev); + else + return r; } static int gmc_v7_0_hw_fini(void *handle) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 054733838292..ca9841d5669f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -837,17 +837,14 @@ static void gmc_v8_0_set_prt(struct amdgpu_device *adev, bool enable) static int gmc_v8_0_gart_enable(struct amdgpu_device *adev) { uint64_t table_addr; - int r, i; u32 tmp, field; + int i; if (adev->gart.bo == NULL) { dev_err(adev->dev, "No VRAM object for PCIE GART.\n"); return -EINVAL; } - r = amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); - if (r) - return r; - + amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); table_addr = amdgpu_bo_gpu_offset(adev->gart.bo); /* Setup TLB control */ @@ -953,7 +950,6 @@ static int gmc_v8_0_gart_enable(struct amdgpu_device *adev) DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", (unsigned)(adev->gmc.gart_size >> 20), (unsigned long long)table_addr); - adev->gart.ready = true; return 0; } @@ -1242,7 +1238,10 @@ static int gmc_v8_0_hw_init(void *handle) if (r) return r; - return r; + if (amdgpu_emu_mode == 1) + return amdgpu_gmc_vram_checking(adev); + else + return r; } static int gmc_v8_0_hw_fini(void *handle) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 88c1eb9ad068..4595027a8c63 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -1202,7 +1202,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev) adev->umc.umc_inst_num = UMC_V6_1_UMC_INSTANCE_NUM; adev->umc.channel_offs = UMC_V6_1_PER_CHANNEL_OFFSET_VG20; adev->umc.channel_idx_tbl = &umc_v6_1_channel_idx_tbl[0][0]; - adev->umc.ras_funcs = &umc_v6_1_ras_funcs; + adev->umc.ras = &umc_v6_1_ras; break; case IP_VERSION(6, 1, 2): adev->umc.max_ras_err_cnt_per_query = UMC_V6_1_TOTAL_CHANNEL_NUM; @@ -1210,15 +1210,16 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev) adev->umc.umc_inst_num = UMC_V6_1_UMC_INSTANCE_NUM; adev->umc.channel_offs = UMC_V6_1_PER_CHANNEL_OFFSET_ARCT; adev->umc.channel_idx_tbl = &umc_v6_1_channel_idx_tbl[0][0]; - adev->umc.ras_funcs = &umc_v6_1_ras_funcs; + adev->umc.ras = &umc_v6_1_ras; break; case IP_VERSION(6, 7, 0): - adev->umc.max_ras_err_cnt_per_query = UMC_V6_7_TOTAL_CHANNEL_NUM; + adev->umc.max_ras_err_cnt_per_query = + UMC_V6_7_TOTAL_CHANNEL_NUM * UMC_V6_7_BAD_PAGE_NUM_PER_CHANNEL; adev->umc.channel_inst_num = UMC_V6_7_CHANNEL_INSTANCE_NUM; adev->umc.umc_inst_num = UMC_V6_7_UMC_INSTANCE_NUM; adev->umc.channel_offs = UMC_V6_7_PER_CHANNEL_OFFSET; if (!adev->gmc.xgmi.connected_to_cpu) - adev->umc.ras_funcs = &umc_v6_7_ras_funcs; + adev->umc.ras = &umc_v6_7_ras; if (1 & adev->smuio.funcs->get_die_id(adev)) adev->umc.channel_idx_tbl = &umc_v6_7_channel_idx_tbl_first[0][0]; else @@ -1227,6 +1228,21 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev) default: break; } + + if (adev->umc.ras) { + amdgpu_ras_register_ras_block(adev, &adev->umc.ras->ras_block); + + strcpy(adev->umc.ras->ras_block.name, "umc"); + adev->umc.ras->ras_block.block = AMDGPU_RAS_BLOCK__UMC; + + /* If don't define special ras_late_init function, use default ras_late_init */ + if (!adev->umc.ras->ras_block.ras_late_init) + adev->umc.ras->ras_block.ras_late_init = amdgpu_umc_ras_late_init; + + /* If don't define special ras_fini function, use default ras_fini */ + if (!adev->umc.ras->ras_block.ras_fini) + adev->umc.ras->ras_block.ras_fini = amdgpu_umc_ras_fini; + } } static void gmc_v9_0_set_mmhub_funcs(struct amdgpu_device *adev) @@ -1248,18 +1264,33 @@ static void gmc_v9_0_set_mmhub_ras_funcs(struct amdgpu_device *adev) { switch (adev->ip_versions[MMHUB_HWIP][0]) { case IP_VERSION(9, 4, 0): - adev->mmhub.ras_funcs = &mmhub_v1_0_ras_funcs; + adev->mmhub.ras = &mmhub_v1_0_ras; break; case IP_VERSION(9, 4, 1): - adev->mmhub.ras_funcs = &mmhub_v9_4_ras_funcs; + adev->mmhub.ras = &mmhub_v9_4_ras; break; case IP_VERSION(9, 4, 2): - adev->mmhub.ras_funcs = &mmhub_v1_7_ras_funcs; + adev->mmhub.ras = &mmhub_v1_7_ras; break; default: /* mmhub ras is not available */ break; } + + if (adev->mmhub.ras) { + amdgpu_ras_register_ras_block(adev, &adev->mmhub.ras->ras_block); + + strcpy(adev->mmhub.ras->ras_block.name,"mmhub"); + adev->mmhub.ras->ras_block.block = AMDGPU_RAS_BLOCK__MMHUB; + + /* If don't define special ras_late_init function, use default ras_late_init */ + if (!adev->mmhub.ras->ras_block.ras_late_init) + adev->mmhub.ras->ras_block.ras_late_init = amdgpu_mmhub_ras_late_init; + + /* If don't define special ras_fini function, use default ras_fini */ + if (!adev->mmhub.ras->ras_block.ras_fini) + adev->mmhub.ras->ras_block.ras_fini = amdgpu_mmhub_ras_fini; + } } static void gmc_v9_0_set_gfxhub_funcs(struct amdgpu_device *adev) @@ -1269,7 +1300,8 @@ static void gmc_v9_0_set_gfxhub_funcs(struct amdgpu_device *adev) static void gmc_v9_0_set_hdp_ras_funcs(struct amdgpu_device *adev) { - adev->hdp.ras_funcs = &hdp_v4_0_ras_funcs; + adev->hdp.ras = &hdp_v4_0_ras; + amdgpu_ras_register_ras_block(adev, &adev->hdp.ras->ras_block); } static void gmc_v9_0_set_mca_funcs(struct amdgpu_device *adev) @@ -1287,6 +1319,7 @@ static void gmc_v9_0_set_mca_funcs(struct amdgpu_device *adev) static int gmc_v9_0_early_init(void *handle) { + int r; struct amdgpu_device *adev = (struct amdgpu_device *)handle; /* ARCT and VEGA20 don't have XGMI defined in their IP discovery tables */ @@ -1316,6 +1349,10 @@ static int gmc_v9_0_early_init(void *handle) adev->gmc.private_aperture_end = adev->gmc.private_aperture_start + (4ULL << 30) - 1; + r = amdgpu_gmc_ras_early_init(adev); + if (r) + return r; + return 0; } @@ -1342,13 +1379,13 @@ static int gmc_v9_0_late_init(void *handle) } if (!amdgpu_persistent_edc_harvesting_supported(adev)) { - if (adev->mmhub.ras_funcs && - adev->mmhub.ras_funcs->reset_ras_error_count) - adev->mmhub.ras_funcs->reset_ras_error_count(adev); + if (adev->mmhub.ras && adev->mmhub.ras->ras_block.hw_ops && + adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count) + adev->mmhub.ras->ras_block.hw_ops->reset_ras_error_count(adev); - if (adev->hdp.ras_funcs && - adev->hdp.ras_funcs->reset_ras_error_count) - adev->hdp.ras_funcs->reset_ras_error_count(adev); + if (adev->hdp.ras && adev->hdp.ras->ras_block.hw_ops && + adev->hdp.ras->ras_block.hw_ops->reset_ras_error_count) + adev->hdp.ras->ras_block.hw_ops->reset_ras_error_count(adev); } r = amdgpu_gmc_ras_late_init(adev); @@ -1752,14 +1789,7 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev) return -EINVAL; } - if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) - goto skip_pin_bo; - - r = amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); - if (r) - return r; - -skip_pin_bo: + amdgpu_gtt_mgr_recover(&adev->mman.gtt_mgr); r = adev->gfxhub.funcs->gart_enable(adev); if (r) return r; @@ -1776,7 +1806,6 @@ skip_pin_bo: DRM_INFO("PTB located at 0x%016llX\n", (unsigned long long)amdgpu_bo_gpu_offset(adev->gart.bo)); - adev->gart.ready = true; return 0; } @@ -1784,7 +1813,7 @@ static int gmc_v9_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; bool value; - int i; + int i, r; /* The sequence of these two function calls matters.*/ gmc_v9_0_init_golden_registers(adev); @@ -1819,7 +1848,14 @@ static int gmc_v9_0_hw_init(void *handle) if (adev->umc.funcs && adev->umc.funcs->init_registers) adev->umc.funcs->init_registers(adev); - return gmc_v9_0_gart_enable(adev); + r = gmc_v9_0_gart_enable(adev); + if (r) + return r; + + if (amdgpu_emu_mode == 1) + return amdgpu_gmc_vram_checking(adev); + else + return r; } /** diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c index eecfb1545c1e..6b41fcbf4875 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.c @@ -150,13 +150,21 @@ static void hdp_v4_0_init_registers(struct amdgpu_device *adev) WREG32_SOC15(HDP, 0, mmHDP_NONSURFACE_BASE_HI, (adev->gmc.vram_start >> 40)); } -const struct amdgpu_hdp_ras_funcs hdp_v4_0_ras_funcs = { - .ras_late_init = amdgpu_hdp_ras_late_init, - .ras_fini = amdgpu_hdp_ras_fini, +struct amdgpu_ras_block_hw_ops hdp_v4_0_ras_hw_ops = { .query_ras_error_count = hdp_v4_0_query_ras_error_count, .reset_ras_error_count = hdp_v4_0_reset_ras_error_count, }; +struct amdgpu_hdp_ras hdp_v4_0_ras = { + .ras_block = { + .name = "hdp", + .block = AMDGPU_RAS_BLOCK__HDP, + .hw_ops = &hdp_v4_0_ras_hw_ops, + .ras_late_init = amdgpu_hdp_ras_late_init, + .ras_fini = amdgpu_hdp_ras_fini, + }, +}; + const struct amdgpu_hdp_funcs hdp_v4_0_funcs = { .flush_hdp = hdp_v4_0_flush_hdp, .invalidate_hdp = hdp_v4_0_invalidate_hdp, diff --git a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.h b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.h index dc3a1b81dd62..c44eee9282ab 100644 --- a/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.h +++ b/drivers/gpu/drm/amd/amdgpu/hdp_v4_0.h @@ -27,6 +27,6 @@ #include "soc15_common.h" extern const struct amdgpu_hdp_funcs hdp_v4_0_funcs; -extern const struct amdgpu_hdp_ras_funcs hdp_v4_0_ras_funcs; +extern struct amdgpu_hdp_ras hdp_v4_0_ras; #endif diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c index 01c242c5abc3..41a00851b6c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c @@ -50,11 +50,16 @@ static int jpeg_v3_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->asic_type != CHIP_YELLOW_CARP) { - u32 harvest = RREG32_SOC15(JPEG, 0, mmCC_UVD_HARVESTING); + u32 harvest; + switch (adev->ip_versions[UVD_HWIP][0]) { + case IP_VERSION(3, 1, 1): + break; + default: + harvest = RREG32_SOC15(JPEG, 0, mmCC_UVD_HARVESTING); if (harvest & CC_UVD_HARVESTING__UVD_DISABLE_MASK) return -ENOENT; + break; } adev->jpeg.num_jpeg_inst = 1; diff --git a/drivers/gpu/drm/amd/amdgpu/mca_v3_0.c b/drivers/gpu/drm/amd/amdgpu/mca_v3_0.c index 8f7107d392af..68565262af9c 100644 --- a/drivers/gpu/drm/amd/amdgpu/mca_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mca_v3_0.c @@ -37,7 +37,7 @@ static void mca_v3_0_mp0_query_ras_error_count(struct amdgpu_device *adev, ras_error_status); } -static int mca_v3_0_mp0_ras_late_init(struct amdgpu_device *adev) +static int mca_v3_0_mp0_ras_late_init(struct amdgpu_device *adev, void *ras_info) { return amdgpu_mca_ras_late_init(adev, &adev->mca.mp0); } @@ -47,14 +47,35 @@ static void mca_v3_0_mp0_ras_fini(struct amdgpu_device *adev) amdgpu_mca_ras_fini(adev, &adev->mca.mp0); } -const struct amdgpu_mca_ras_funcs mca_v3_0_mp0_ras_funcs = { - .ras_late_init = mca_v3_0_mp0_ras_late_init, - .ras_fini = mca_v3_0_mp0_ras_fini, +static int mca_v3_0_ras_block_match(struct amdgpu_ras_block_object *block_obj, + enum amdgpu_ras_block block, uint32_t sub_block_index) +{ + if (!block_obj) + return -EINVAL; + + if ((block_obj->block == block) && + (block_obj->sub_block_index == sub_block_index)) { + return 0; + } + + return -EINVAL; +} + +const struct amdgpu_ras_block_hw_ops mca_v3_0_mp0_hw_ops = { .query_ras_error_count = mca_v3_0_mp0_query_ras_error_count, .query_ras_error_address = NULL, - .ras_block = AMDGPU_RAS_BLOCK__MCA, - .ras_sub_block = AMDGPU_RAS_MCA_BLOCK__MP0, - .sysfs_name = "mp0_err_count", +}; + +struct amdgpu_mca_ras_block mca_v3_0_mp0_ras = { + .ras_block = { + .block = AMDGPU_RAS_BLOCK__MCA, + .sub_block_index = AMDGPU_RAS_MCA_BLOCK__MP0, + .name = "mp0", + .hw_ops = &mca_v3_0_mp0_hw_ops, + .ras_block_match = mca_v3_0_ras_block_match, + .ras_late_init = mca_v3_0_mp0_ras_late_init, + .ras_fini = mca_v3_0_mp0_ras_fini, + }, }; static void mca_v3_0_mp1_query_ras_error_count(struct amdgpu_device *adev, @@ -65,7 +86,7 @@ static void mca_v3_0_mp1_query_ras_error_count(struct amdgpu_device *adev, ras_error_status); } -static int mca_v3_0_mp1_ras_late_init(struct amdgpu_device *adev) +static int mca_v3_0_mp1_ras_late_init(struct amdgpu_device *adev, void *ras_info) { return amdgpu_mca_ras_late_init(adev, &adev->mca.mp1); } @@ -75,14 +96,21 @@ static void mca_v3_0_mp1_ras_fini(struct amdgpu_device *adev) amdgpu_mca_ras_fini(adev, &adev->mca.mp1); } -const struct amdgpu_mca_ras_funcs mca_v3_0_mp1_ras_funcs = { - .ras_late_init = mca_v3_0_mp1_ras_late_init, - .ras_fini = mca_v3_0_mp1_ras_fini, +const struct amdgpu_ras_block_hw_ops mca_v3_0_mp1_hw_ops = { .query_ras_error_count = mca_v3_0_mp1_query_ras_error_count, .query_ras_error_address = NULL, - .ras_block = AMDGPU_RAS_BLOCK__MCA, - .ras_sub_block = AMDGPU_RAS_MCA_BLOCK__MP1, - .sysfs_name = "mp1_err_count", +}; + +struct amdgpu_mca_ras_block mca_v3_0_mp1_ras = { + .ras_block = { + .block = AMDGPU_RAS_BLOCK__MCA, + .sub_block_index = AMDGPU_RAS_MCA_BLOCK__MP1, + .name = "mp1", + .hw_ops = &mca_v3_0_mp1_hw_ops, + .ras_block_match = mca_v3_0_ras_block_match, + .ras_late_init = mca_v3_0_mp1_ras_late_init, + .ras_fini = mca_v3_0_mp1_ras_fini, + }, }; static void mca_v3_0_mpio_query_ras_error_count(struct amdgpu_device *adev, @@ -93,7 +121,7 @@ static void mca_v3_0_mpio_query_ras_error_count(struct amdgpu_device *adev, ras_error_status); } -static int mca_v3_0_mpio_ras_late_init(struct amdgpu_device *adev) +static int mca_v3_0_mpio_ras_late_init(struct amdgpu_device *adev, void *ras_info) { return amdgpu_mca_ras_late_init(adev, &adev->mca.mpio); } @@ -103,14 +131,21 @@ static void mca_v3_0_mpio_ras_fini(struct amdgpu_device *adev) amdgpu_mca_ras_fini(adev, &adev->mca.mpio); } -const struct amdgpu_mca_ras_funcs mca_v3_0_mpio_ras_funcs = { - .ras_late_init = mca_v3_0_mpio_ras_late_init, - .ras_fini = mca_v3_0_mpio_ras_fini, +const struct amdgpu_ras_block_hw_ops mca_v3_0_mpio_hw_ops = { .query_ras_error_count = mca_v3_0_mpio_query_ras_error_count, .query_ras_error_address = NULL, - .ras_block = AMDGPU_RAS_BLOCK__MCA, - .ras_sub_block = AMDGPU_RAS_MCA_BLOCK__MPIO, - .sysfs_name = "mpio_err_count", +}; + +struct amdgpu_mca_ras_block mca_v3_0_mpio_ras = { + .ras_block = { + .block = AMDGPU_RAS_BLOCK__MCA, + .sub_block_index = AMDGPU_RAS_MCA_BLOCK__MPIO, + .name = "mpio", + .hw_ops = &mca_v3_0_mpio_hw_ops, + .ras_block_match = mca_v3_0_ras_block_match, + .ras_late_init = mca_v3_0_mpio_ras_late_init, + .ras_fini = mca_v3_0_mpio_ras_fini, + }, }; @@ -118,9 +153,12 @@ static void mca_v3_0_init(struct amdgpu_device *adev) { struct amdgpu_mca *mca = &adev->mca; - mca->mp0.ras_funcs = &mca_v3_0_mp0_ras_funcs; - mca->mp1.ras_funcs = &mca_v3_0_mp1_ras_funcs; - mca->mpio.ras_funcs = &mca_v3_0_mpio_ras_funcs; + mca->mp0.ras = &mca_v3_0_mp0_ras; + mca->mp1.ras = &mca_v3_0_mp1_ras; + mca->mpio.ras = &mca_v3_0_mpio_ras; + amdgpu_ras_register_ras_block(adev, &mca->mp0.ras->ras_block); + amdgpu_ras_register_ras_block(adev, &mca->mp1.ras->ras_block); + amdgpu_ras_register_ras_block(adev, &mca->mpio.ras->ras_block); } const struct amdgpu_mca_funcs mca_v3_0_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 1da2ec692057..4c9f0c0f3116 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -774,13 +774,17 @@ static void mmhub_v1_0_reset_ras_error_count(struct amdgpu_device *adev) } } -const struct amdgpu_mmhub_ras_funcs mmhub_v1_0_ras_funcs = { - .ras_late_init = amdgpu_mmhub_ras_late_init, - .ras_fini = amdgpu_mmhub_ras_fini, +struct amdgpu_ras_block_hw_ops mmhub_v1_0_ras_hw_ops = { .query_ras_error_count = mmhub_v1_0_query_ras_error_count, .reset_ras_error_count = mmhub_v1_0_reset_ras_error_count, }; +struct amdgpu_mmhub_ras mmhub_v1_0_ras = { + .ras_block = { + .hw_ops = &mmhub_v1_0_ras_hw_ops, + }, +}; + const struct amdgpu_mmhub_funcs mmhub_v1_0_funcs = { .get_fb_location = mmhub_v1_0_get_fb_location, .init = mmhub_v1_0_init, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h index 4661b094e007..dae7ca48bd8b 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h @@ -24,6 +24,6 @@ #define __MMHUB_V1_0_H__ extern const struct amdgpu_mmhub_funcs mmhub_v1_0_funcs; -extern const struct amdgpu_mmhub_ras_funcs mmhub_v1_0_ras_funcs; +extern struct amdgpu_mmhub_ras mmhub_v1_0_ras; #endif diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c index f5f7181f9af5..3b901f941627 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c @@ -1321,15 +1321,19 @@ static void mmhub_v1_7_reset_ras_error_status(struct amdgpu_device *adev) } } -const struct amdgpu_mmhub_ras_funcs mmhub_v1_7_ras_funcs = { - .ras_late_init = amdgpu_mmhub_ras_late_init, - .ras_fini = amdgpu_mmhub_ras_fini, +struct amdgpu_ras_block_hw_ops mmhub_v1_7_ras_hw_ops = { .query_ras_error_count = mmhub_v1_7_query_ras_error_count, .reset_ras_error_count = mmhub_v1_7_reset_ras_error_count, .query_ras_error_status = mmhub_v1_7_query_ras_error_status, .reset_ras_error_status = mmhub_v1_7_reset_ras_error_status, }; +struct amdgpu_mmhub_ras mmhub_v1_7_ras = { + .ras_block = { + .hw_ops = &mmhub_v1_7_ras_hw_ops, + }, +}; + const struct amdgpu_mmhub_funcs mmhub_v1_7_funcs = { .get_fb_location = mmhub_v1_7_get_fb_location, .init = mmhub_v1_7_init, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.h index a7f9dfc24697..629f49052137 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.h +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.h @@ -24,6 +24,6 @@ #define __MMHUB_V1_7_H__ extern const struct amdgpu_mmhub_funcs mmhub_v1_7_funcs; -extern const struct amdgpu_mmhub_ras_funcs mmhub_v1_7_ras_funcs; +extern struct amdgpu_mmhub_ras mmhub_v1_7_ras; #endif diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c index ff49eeaf7882..619106f7d23d 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c @@ -1655,14 +1655,18 @@ static void mmhub_v9_4_query_ras_error_status(struct amdgpu_device *adev) } } -const struct amdgpu_mmhub_ras_funcs mmhub_v9_4_ras_funcs = { - .ras_late_init = amdgpu_mmhub_ras_late_init, - .ras_fini = amdgpu_mmhub_ras_fini, +const struct amdgpu_ras_block_hw_ops mmhub_v9_4_ras_hw_ops = { .query_ras_error_count = mmhub_v9_4_query_ras_error_count, .reset_ras_error_count = mmhub_v9_4_reset_ras_error_count, .query_ras_error_status = mmhub_v9_4_query_ras_error_status, }; +struct amdgpu_mmhub_ras mmhub_v9_4_ras = { + .ras_block = { + .hw_ops = &mmhub_v9_4_ras_hw_ops, + }, +}; + const struct amdgpu_mmhub_funcs mmhub_v9_4_funcs = { .get_fb_location = mmhub_v9_4_get_fb_location, .init = mmhub_v9_4_init, diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h index 90436efa92ef..a48329d95f71 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.h @@ -24,6 +24,6 @@ #define __MMHUB_V9_4_H__ extern const struct amdgpu_mmhub_funcs mmhub_v9_4_funcs; -extern const struct amdgpu_mmhub_ras_funcs mmhub_v9_4_ras_funcs; +extern struct amdgpu_mmhub_ras mmhub_v9_4_ras; #endif diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c index aef9d059ae52..a642c04cf17d 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c @@ -544,7 +544,7 @@ static int xgpu_vi_mailbox_rcv_irq(struct amdgpu_device *adev, { int r; - /* trigger gpu-reset by hypervisor only if TDR disbaled */ + /* trigger gpu-reset by hypervisor only if TDR disabled */ if (!amdgpu_gpu_recovery) { /* see what event we get */ r = xgpu_vi_mailbox_rcv_msg(adev, IDH_FLR_NOTIFICATION); diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c index 3444332ea110..44f17bbfeb6a 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_2.c @@ -59,10 +59,15 @@ static u32 nbio_v7_2_get_rev_id(struct amdgpu_device *adev) { u32 tmp; - if (adev->asic_type == CHIP_YELLOW_CARP) + switch (adev->ip_versions[NBIO_HWIP][0]) { + case IP_VERSION(7, 2, 1): + case IP_VERSION(7, 5, 0): tmp = RREG32_SOC15(NBIO, 0, regRCC_STRAP0_RCC_DEV0_EPF0_STRAP0_YC); - else + break; + default: tmp = RREG32_SOC15(NBIO, 0, regRCC_STRAP0_RCC_DEV0_EPF0_STRAP0); + break; + } tmp &= RCC_STRAP0_RCC_DEV0_EPF0_STRAP0__STRAP_ATI_REV_ID_DEV0_F0_MASK; tmp >>= RCC_STRAP0_RCC_DEV0_EPF0_STRAP0__STRAP_ATI_REV_ID_DEV0_F0__SHIFT; @@ -72,20 +77,25 @@ static u32 nbio_v7_2_get_rev_id(struct amdgpu_device *adev) static void nbio_v7_2_mc_access_enable(struct amdgpu_device *adev, bool enable) { - if (enable) - if (adev->asic_type == CHIP_YELLOW_CARP) + switch (adev->ip_versions[NBIO_HWIP][0]) { + case IP_VERSION(7, 2, 1): + case IP_VERSION(7, 5, 0): + if (enable) WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_FB_EN_YC, BIF_BX0_BIF_FB_EN__FB_READ_EN_MASK | BIF_BX0_BIF_FB_EN__FB_WRITE_EN_MASK); else + WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_FB_EN_YC, 0); + break; + default: + if (enable) WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_FB_EN, BIF_BX0_BIF_FB_EN__FB_READ_EN_MASK | BIF_BX0_BIF_FB_EN__FB_WRITE_EN_MASK); - else - if (adev->asic_type == CHIP_YELLOW_CARP) - WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_FB_EN_YC, 0); else WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_FB_EN, 0); + break; + } } static u32 nbio_v7_2_get_memsize(struct amdgpu_device *adev) @@ -250,7 +260,9 @@ static void nbio_v7_2_update_medium_grain_light_sleep(struct amdgpu_device *adev { uint32_t def, data; - if (adev->asic_type == CHIP_YELLOW_CARP) { + switch (adev->ip_versions[NBIO_HWIP][0]) { + case IP_VERSION(7, 2, 1): + case IP_VERSION(7, 5, 0): def = data = RREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regPCIE_CNTL2)); if (enable && (adev->cg_flags & AMD_CG_SUPPORT_BIF_LS)) data |= PCIE_CNTL2__SLV_MEM_LS_EN_MASK; @@ -260,8 +272,8 @@ static void nbio_v7_2_update_medium_grain_light_sleep(struct amdgpu_device *adev if (def != data) WREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regPCIE_CNTL2), data); - data = RREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regBIF1_PCIE_TX_POWER_CTRL_1)); - def = data; + def = data = RREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, + regBIF1_PCIE_TX_POWER_CTRL_1)); if (enable && (adev->cg_flags & AMD_CG_SUPPORT_BIF_LS)) data |= (BIF1_PCIE_TX_POWER_CTRL_1__MST_MEM_LS_EN_MASK | BIF1_PCIE_TX_POWER_CTRL_1__REPLAY_MEM_LS_EN_MASK); @@ -272,7 +284,8 @@ static void nbio_v7_2_update_medium_grain_light_sleep(struct amdgpu_device *adev if (def != data) WREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regBIF1_PCIE_TX_POWER_CTRL_1), data); - } else { + break; + default: def = data = RREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regPCIE_CNTL2)); if (enable && (adev->cg_flags & AMD_CG_SUPPORT_BIF_LS)) data |= (PCIE_CNTL2__SLV_MEM_LS_EN_MASK | @@ -285,6 +298,7 @@ static void nbio_v7_2_update_medium_grain_light_sleep(struct amdgpu_device *adev if (def != data) WREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regPCIE_CNTL2), data); + break; } } @@ -352,7 +366,9 @@ const struct nbio_hdp_flush_reg nbio_v7_2_hdp_flush_reg = { static void nbio_v7_2_init_registers(struct amdgpu_device *adev) { uint32_t def, data; - if (adev->asic_type == CHIP_YELLOW_CARP) { + switch (adev->ip_versions[NBIO_HWIP][0]) { + case IP_VERSION(7, 2, 1): + case IP_VERSION(7, 5, 0): def = data = RREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regBIF1_PCIE_MST_CTRL_3)); data = REG_SET_FIELD(data, BIF1_PCIE_MST_CTRL_3, CI_SWUS_MAX_READ_REQUEST_SIZE_MODE, 1); @@ -361,7 +377,8 @@ static void nbio_v7_2_init_registers(struct amdgpu_device *adev) if (def != data) WREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regBIF1_PCIE_MST_CTRL_3), data); - } else { + break; + default: def = data = RREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regPCIE_CONFIG_CNTL)); data = REG_SET_FIELD(data, PCIE_CONFIG_CNTL, CI_SWUS_MAX_READ_REQUEST_SIZE_MODE, 1); @@ -370,6 +387,7 @@ static void nbio_v7_2_init_registers(struct amdgpu_device *adev) if (def != data) WREG32_PCIE_PORT(SOC15_REG_OFFSET(NBIO, 0, regPCIE_CONFIG_CNTL), data); + break; } if (amdgpu_sriov_vf(adev)) diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c index dc5e93756fea..39974b449341 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c @@ -658,16 +658,25 @@ static void nbio_v7_4_enable_doorbell_interrupt(struct amdgpu_device *adev, DOORBELL_INTERRUPT_DISABLE, enable ? 0 : 1); } -const struct amdgpu_nbio_ras_funcs nbio_v7_4_ras_funcs = { +const struct amdgpu_ras_block_hw_ops nbio_v7_4_ras_hw_ops = { + .query_ras_error_count = nbio_v7_4_query_ras_error_count, +}; + +struct amdgpu_nbio_ras nbio_v7_4_ras = { + .ras_block = { + .name = "pcie_bif", + .block = AMDGPU_RAS_BLOCK__PCIE_BIF, + .hw_ops = &nbio_v7_4_ras_hw_ops, + .ras_late_init = amdgpu_nbio_ras_late_init, + .ras_fini = amdgpu_nbio_ras_fini, + }, .handle_ras_controller_intr_no_bifring = nbio_v7_4_handle_ras_controller_intr_no_bifring, .handle_ras_err_event_athub_intr_no_bifring = nbio_v7_4_handle_ras_err_event_athub_intr_no_bifring, .init_ras_controller_interrupt = nbio_v7_4_init_ras_controller_interrupt, .init_ras_err_event_athub_interrupt = nbio_v7_4_init_ras_err_event_athub_interrupt, - .query_ras_error_count = nbio_v7_4_query_ras_error_count, - .ras_late_init = amdgpu_nbio_ras_late_init, - .ras_fini = amdgpu_nbio_ras_fini, }; + static void nbio_v7_4_program_ltr(struct amdgpu_device *adev) { uint32_t def, data; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.h b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.h index cc5692db6f98..7490022d79d4 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.h +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.h @@ -29,6 +29,6 @@ extern const struct nbio_hdp_flush_reg nbio_v7_4_hdp_flush_reg; extern const struct nbio_hdp_flush_reg nbio_v7_4_hdp_flush_reg_ald; extern const struct amdgpu_nbio_funcs nbio_v7_4_funcs; -extern const struct amdgpu_nbio_ras_funcs nbio_v7_4_ras_funcs; +extern struct amdgpu_nbio_ras nbio_v7_4_ras; #endif diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index 2ec1ffb36b1f..5e9ab31fee6b 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -258,21 +258,6 @@ static u64 nv_pcie_rreg64(struct amdgpu_device *adev, u32 reg) return amdgpu_device_indirect_rreg64(adev, address, data, reg); } -static u32 nv_pcie_port_rreg(struct amdgpu_device *adev, u32 reg) -{ - unsigned long flags, address, data; - u32 r; - address = adev->nbio.funcs->get_pcie_port_index_offset(adev); - data = adev->nbio.funcs->get_pcie_port_data_offset(adev); - - spin_lock_irqsave(&adev->pcie_idx_lock, flags); - WREG32(address, reg * 4); - (void)RREG32(address); - r = RREG32(data); - spin_unlock_irqrestore(&adev->pcie_idx_lock, flags); - return r; -} - static void nv_pcie_wreg64(struct amdgpu_device *adev, u32 reg, u64 v) { unsigned long address, data; @@ -283,21 +268,6 @@ static void nv_pcie_wreg64(struct amdgpu_device *adev, u32 reg, u64 v) amdgpu_device_indirect_wreg64(adev, address, data, reg, v); } -static void nv_pcie_port_wreg(struct amdgpu_device *adev, u32 reg, u32 v) -{ - unsigned long flags, address, data; - - address = adev->nbio.funcs->get_pcie_port_index_offset(adev); - data = adev->nbio.funcs->get_pcie_port_data_offset(adev); - - spin_lock_irqsave(&adev->pcie_idx_lock, flags); - WREG32(address, reg * 4); - (void)RREG32(address); - WREG32(data, v); - (void)RREG32(data); - spin_unlock_irqrestore(&adev->pcie_idx_lock, flags); -} - static u32 nv_didt_rreg(struct amdgpu_device *adev, u32 reg) { unsigned long flags, address, data; @@ -360,38 +330,6 @@ static bool nv_read_disabled_bios(struct amdgpu_device *adev) return false; } -static bool nv_read_bios_from_rom(struct amdgpu_device *adev, - u8 *bios, u32 length_bytes) -{ - u32 *dw_ptr; - u32 i, length_dw; - u32 rom_index_offset, rom_data_offset; - - if (bios == NULL) - return false; - if (length_bytes == 0) - return false; - /* APU vbios image is part of sbios image */ - if (adev->flags & AMD_IS_APU) - return false; - - dw_ptr = (u32 *)bios; - length_dw = ALIGN(length_bytes, 4) / 4; - - rom_index_offset = - adev->smuio.funcs->get_rom_index_offset(adev); - rom_data_offset = - adev->smuio.funcs->get_rom_data_offset(adev); - - /* set rom index to 0 */ - WREG32(rom_index_offset, 0); - /* read out the rom data */ - for (i = 0; i < length_dw; i++) - dw_ptr[i] = RREG32(rom_data_offset); - - return true; -} - static struct soc15_allowed_register_entry nv_allowed_read_registers[] = { { SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS)}, { SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS2)}, @@ -708,7 +646,7 @@ static int nv_update_umd_stable_pstate(struct amdgpu_device *adev, static const struct amdgpu_asic_funcs nv_asic_funcs = { .read_disabled_bios = &nv_read_disabled_bios, - .read_bios_from_rom = &nv_read_bios_from_rom, + .read_bios_from_rom = &amdgpu_soc15_read_bios_from_rom, .read_register = &nv_read_register, .reset = &nv_asic_reset, .reset_method = &nv_asic_reset_method, @@ -742,8 +680,8 @@ static int nv_common_early_init(void *handle) adev->pcie_wreg = &nv_pcie_wreg; adev->pcie_rreg64 = &nv_pcie_rreg64; adev->pcie_wreg64 = &nv_pcie_wreg64; - adev->pciep_rreg = &nv_pcie_port_rreg; - adev->pciep_wreg = &nv_pcie_port_wreg; + adev->pciep_rreg = amdgpu_device_pcie_port_rreg; + adev->pciep_wreg = amdgpu_device_pcie_port_wreg; /* TODO: will add them during VCN v2 implementation */ adev->uvd_ctx_rreg = NULL; @@ -964,6 +902,7 @@ static int nv_common_early_init(void *handle) adev->external_rev_id = adev->rev_id + 0x01; break; case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): adev->cg_flags = 0; adev->pg_flags = 0; adev->external_rev_id = adev->rev_id + 0x82; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h index dd0dce254901..1f276ddd26e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h @@ -258,6 +258,7 @@ enum psp_gfx_fw_type { GFX_FW_TYPE_SDMA6 = 56, /* SDMA6 MI */ GFX_FW_TYPE_SDMA7 = 57, /* SDMA7 MI */ GFX_FW_TYPE_VCN1 = 58, /* VCN1 MI */ + GFX_FW_TYPE_CAP = 62, /* CAP_FW */ GFX_FW_TYPE_REG_LIST = 67, /* REG_LIST MI */ GFX_FW_TYPE_MAX }; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index d0e76b36d4ab..9518b4394a6e 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -53,11 +53,13 @@ MODULE_FIRMWARE("amdgpu/navi14_ta.bin"); MODULE_FIRMWARE("amdgpu/navi12_sos.bin"); MODULE_FIRMWARE("amdgpu/navi12_asd.bin"); MODULE_FIRMWARE("amdgpu/navi12_ta.bin"); +MODULE_FIRMWARE("amdgpu/navi12_cap.bin"); MODULE_FIRMWARE("amdgpu/arcturus_sos.bin"); MODULE_FIRMWARE("amdgpu/arcturus_asd.bin"); MODULE_FIRMWARE("amdgpu/arcturus_ta.bin"); MODULE_FIRMWARE("amdgpu/sienna_cichlid_sos.bin"); MODULE_FIRMWARE("amdgpu/sienna_cichlid_ta.bin"); +MODULE_FIRMWARE("amdgpu/sienna_cichlid_cap.bin"); MODULE_FIRMWARE("amdgpu/navy_flounder_sos.bin"); MODULE_FIRMWARE("amdgpu/navy_flounder_ta.bin"); MODULE_FIRMWARE("amdgpu/vangogh_asd.bin"); @@ -177,8 +179,6 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) err = psp_init_asd_microcode(psp, chip_name); if (err) return err; - if (amdgpu_sriov_vf(adev)) - break; snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ta.bin", chip_name); err = request_firmware(&adev->psp.ta_fw, fw_name, adev->dev); if (err) { diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c index 1ed357cb0f49..01f3bcc62a6c 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c @@ -44,6 +44,7 @@ MODULE_FIRMWARE("amdgpu/vega10_sos.bin"); MODULE_FIRMWARE("amdgpu/vega10_asd.bin"); +MODULE_FIRMWARE("amdgpu/vega10_cap.bin"); MODULE_FIRMWARE("amdgpu/vega12_sos.bin"); MODULE_FIRMWARE("amdgpu/vega12_asd.bin"); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index f0638db57111..02115d63b071 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -1892,13 +1892,13 @@ static int sdma_v4_0_late_init(void *handle) sdma_v4_0_setup_ulv(adev); if (!amdgpu_persistent_edc_harvesting_supported(adev)) { - if (adev->sdma.funcs && - adev->sdma.funcs->reset_ras_error_count) - adev->sdma.funcs->reset_ras_error_count(adev); + if (adev->sdma.ras && adev->sdma.ras->ras_block.hw_ops && + adev->sdma.ras->ras_block.hw_ops->reset_ras_error_count) + adev->sdma.ras->ras_block.hw_ops->reset_ras_error_count(adev); } - if (adev->sdma.funcs && adev->sdma.funcs->ras_late_init) - return adev->sdma.funcs->ras_late_init(adev, &ih_info); + if (adev->sdma.ras && adev->sdma.ras->ras_block.ras_late_init) + return adev->sdma.ras->ras_block.ras_late_init(adev, &ih_info); else return 0; } @@ -2001,8 +2001,9 @@ static int sdma_v4_0_sw_fini(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; int i; - if (adev->sdma.funcs && adev->sdma.funcs->ras_fini) - adev->sdma.funcs->ras_fini(adev); + if (adev->sdma.ras && adev->sdma.ras->ras_block.hw_ops && + adev->sdma.ras->ras_block.ras_fini) + adev->sdma.ras->ras_block.ras_fini(adev); for (i = 0; i < adev->sdma.num_instances; i++) { amdgpu_ring_fini(&adev->sdma.instance[i].ring); @@ -2748,7 +2749,7 @@ static void sdma_v4_0_get_ras_error_count(uint32_t value, } } -static int sdma_v4_0_query_ras_error_count(struct amdgpu_device *adev, +static int sdma_v4_0_query_ras_error_count_by_instance(struct amdgpu_device *adev, uint32_t instance, void *ras_error_status) { struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; @@ -2770,6 +2771,18 @@ static int sdma_v4_0_query_ras_error_count(struct amdgpu_device *adev, return 0; }; +static void sdma_v4_0_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status) +{ + int i = 0; + + for (i = 0; i < adev->sdma.num_instances; i++) { + if (sdma_v4_0_query_ras_error_count_by_instance(adev, i, ras_error_status)) { + dev_err(adev->dev, "Query ras error count failed in SDMA%d\n", i); + return; + } + } +} + static void sdma_v4_0_reset_ras_error_count(struct amdgpu_device *adev) { int i; @@ -2781,26 +2794,45 @@ static void sdma_v4_0_reset_ras_error_count(struct amdgpu_device *adev) } } -static const struct amdgpu_sdma_ras_funcs sdma_v4_0_ras_funcs = { - .ras_late_init = amdgpu_sdma_ras_late_init, - .ras_fini = amdgpu_sdma_ras_fini, +const struct amdgpu_ras_block_hw_ops sdma_v4_0_ras_hw_ops = { .query_ras_error_count = sdma_v4_0_query_ras_error_count, .reset_ras_error_count = sdma_v4_0_reset_ras_error_count, }; +static struct amdgpu_sdma_ras sdma_v4_0_ras = { + .ras_block = { + .hw_ops = &sdma_v4_0_ras_hw_ops, + }, +}; + static void sdma_v4_0_set_ras_funcs(struct amdgpu_device *adev) { switch (adev->ip_versions[SDMA0_HWIP][0]) { case IP_VERSION(4, 2, 0): case IP_VERSION(4, 2, 2): - adev->sdma.funcs = &sdma_v4_0_ras_funcs; + adev->sdma.ras = &sdma_v4_0_ras; break; case IP_VERSION(4, 4, 0): - adev->sdma.funcs = &sdma_v4_4_ras_funcs; + adev->sdma.ras = &sdma_v4_4_ras; break; default: break; } + + if (adev->sdma.ras) { + amdgpu_ras_register_ras_block(adev, &adev->sdma.ras->ras_block); + + strcpy(adev->sdma.ras->ras_block.name, "sdma"); + adev->sdma.ras->ras_block.block = AMDGPU_RAS_BLOCK__SDMA; + + /* If don't define special ras_late_init function, use default ras_late_init */ + if (!adev->sdma.ras->ras_block.ras_late_init) + adev->sdma.ras->ras_block.ras_late_init = amdgpu_sdma_ras_late_init; + + /* If don't define special ras_fini function, use default ras_fini */ + if (!adev->sdma.ras->ras_block.ras_fini) + adev->sdma.ras->ras_block.ras_fini = amdgpu_sdma_ras_fini; + } } const struct amdgpu_ip_block_version sdma_v4_0_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c index bf95007f0843..6f9895cdddb1 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.c @@ -188,7 +188,7 @@ static void sdma_v4_4_get_ras_error_count(struct amdgpu_device *adev, } } -static int sdma_v4_4_query_ras_error_count(struct amdgpu_device *adev, +static int sdma_v4_4_query_ras_error_count_by_instance(struct amdgpu_device *adev, uint32_t instance, void *ras_error_status) { @@ -245,9 +245,26 @@ static void sdma_v4_4_reset_ras_error_count(struct amdgpu_device *adev) } } -const struct amdgpu_sdma_ras_funcs sdma_v4_4_ras_funcs = { - .ras_late_init = amdgpu_sdma_ras_late_init, - .ras_fini = amdgpu_sdma_ras_fini, +static void sdma_v4_4_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status) +{ + int i = 0; + + for (i = 0; i < adev->sdma.num_instances; i++) { + if (sdma_v4_4_query_ras_error_count_by_instance(adev, i, ras_error_status)) { + dev_err(adev->dev, "Query ras error count failed in SDMA%d\n", i); + return; + } + } + +} + +const struct amdgpu_ras_block_hw_ops sdma_v4_4_ras_hw_ops = { .query_ras_error_count = sdma_v4_4_query_ras_error_count, .reset_ras_error_count = sdma_v4_4_reset_ras_error_count, }; + +struct amdgpu_sdma_ras sdma_v4_4_ras = { + .ras_block = { + .hw_ops = &sdma_v4_4_ras_hw_ops, + }, +}; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.h b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.h index 74a6e5b5e949..a9f0c68359e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.h +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4.h @@ -23,6 +23,6 @@ #ifndef __SDMA_V4_4_H__ #define __SDMA_V4_4_H__ -extern const struct amdgpu_sdma_ras_funcs sdma_v4_4_ras_funcs; +extern struct amdgpu_sdma_ras sdma_v4_4_ras; #endif diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 81e033549dda..45e10d5028c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -264,7 +264,8 @@ static int sdma_v5_0_init_microcode(struct amdgpu_device *adev) chip_name = "navi12"; break; case IP_VERSION(5, 0, 1): - if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2) + if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2 || + adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 1, 4)) chip_name = "cyan_skillfish2"; else chip_name = "cyan_skillfish"; diff --git a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c index 73ffa8fde3df..dd2d66090d23 100644 --- a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c +++ b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c @@ -26,6 +26,7 @@ #include "smu_v11_0_i2c.h" #include "amdgpu.h" +#include "amdgpu_dpm.h" #include "soc15_common.h" #include <drm/drm_fixed.h> #include <drm/drm_drv.h> @@ -43,11 +44,10 @@ #define I2C_X_RESTART BIT(31) -#define to_amdgpu_device(x) (container_of(x, struct amdgpu_device, pm.smu_i2c)) - static void smu_v11_0_i2c_set_clock_gating(struct i2c_adapter *control, bool en) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; uint32_t reg = RREG32_SOC15(SMUIO, 0, mmSMUIO_PWRMGT); reg = REG_SET_FIELD(reg, SMUIO_PWRMGT, i2c_clk_gate_en, en ? 1 : 0); @@ -75,7 +75,8 @@ static void smu_v11_0_i2c_set_clock_gating(struct i2c_adapter *control, bool en) static int smu_v11_0_i2c_enable(struct i2c_adapter *control, bool enable) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; WREG32_SOC15(SMUIO, 0, mmCKSVII2C_IC_ENABLE, enable ? 1 : 0); @@ -100,7 +101,8 @@ static int smu_v11_0_i2c_enable(struct i2c_adapter *control, bool enable) static void smu_v11_0_i2c_clear_status(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; /* do */ { RREG32_SOC15(SMUIO, 0, mmCKSVII2C_IC_CLR_INTR); @@ -110,7 +112,8 @@ static void smu_v11_0_i2c_clear_status(struct i2c_adapter *control) static void smu_v11_0_i2c_configure(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; uint32_t reg = 0; reg = REG_SET_FIELD(reg, CKSVII2C_IC_CON, IC_SLAVE_DISABLE, 1); @@ -131,7 +134,8 @@ static void smu_v11_0_i2c_configure(struct i2c_adapter *control) static void smu_v11_0_i2c_set_clock(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; /* * Standard mode speed, These values are taken from SMUIO MAS, @@ -154,7 +158,8 @@ static void smu_v11_0_i2c_set_clock(struct i2c_adapter *control) static void smu_v11_0_i2c_set_address(struct i2c_adapter *control, u16 address) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; /* The IC_TAR::IC_TAR field is 10-bits wide. * It takes a 7-bit or 10-bit addresses as an address, @@ -165,7 +170,8 @@ static void smu_v11_0_i2c_set_address(struct i2c_adapter *control, u16 address) static uint32_t smu_v11_0_i2c_poll_tx_status(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; uint32_t ret = I2C_OK; uint32_t reg, reg_c_tx_abrt_source; @@ -216,7 +222,8 @@ static uint32_t smu_v11_0_i2c_poll_tx_status(struct i2c_adapter *control) static uint32_t smu_v11_0_i2c_poll_rx_status(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; uint32_t ret = I2C_OK; uint32_t reg_ic_status, reg_c_tx_abrt_source; @@ -262,7 +269,8 @@ static uint32_t smu_v11_0_i2c_transmit(struct i2c_adapter *control, u16 address, u8 *data, u32 numbytes, u32 i2c_flag) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; u32 bytes_sent, reg, ret = I2C_OK; unsigned long timeout_counter; @@ -360,7 +368,8 @@ static uint32_t smu_v11_0_i2c_receive(struct i2c_adapter *control, u16 address, u8 *data, u32 numbytes, u32 i2c_flag) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; uint32_t bytes_received, ret = I2C_OK; bytes_received = 0; @@ -431,7 +440,8 @@ static uint32_t smu_v11_0_i2c_receive(struct i2c_adapter *control, static void smu_v11_0_i2c_abort(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; uint32_t reg = 0; /* Enable I2C engine; */ @@ -447,7 +457,8 @@ static void smu_v11_0_i2c_abort(struct i2c_adapter *control) static bool smu_v11_0_i2c_activity_done(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; const uint32_t IDLE_TIMEOUT = 1024; uint32_t timeout_count = 0; @@ -508,7 +519,8 @@ static void smu_v11_0_i2c_init(struct i2c_adapter *control) static void smu_v11_0_i2c_fini(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; u32 status, enable, en_stat; int res; @@ -543,7 +555,8 @@ static void smu_v11_0_i2c_fini(struct i2c_adapter *control) static bool smu_v11_0_i2c_bus_lock(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; /* Send PPSMC_MSG_RequestI2CBus */ if (!amdgpu_dpm_smu_i2c_bus_access(adev, true)) @@ -554,7 +567,8 @@ static bool smu_v11_0_i2c_bus_lock(struct i2c_adapter *control) static bool smu_v11_0_i2c_bus_unlock(struct i2c_adapter *control) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(control); + struct amdgpu_device *adev = smu_i2c->adev; /* Send PPSMC_MSG_ReleaseI2CBus */ if (!amdgpu_dpm_smu_i2c_bus_access(adev, false)) @@ -587,16 +601,17 @@ static uint32_t smu_v11_0_i2c_write_data(struct i2c_adapter *control, if (ret != I2C_OK) DRM_ERROR("WriteI2CData() - I2C error occurred :%x", ret); - + return ret; } static void lock_bus(struct i2c_adapter *i2c, unsigned int flags) { - struct amdgpu_device *adev = to_amdgpu_device(i2c); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(i2c); + struct amdgpu_device *adev = smu_i2c->adev; - mutex_lock(&adev->pm.smu_i2c_mutex); + mutex_lock(&smu_i2c->mutex); if (!smu_v11_0_i2c_bus_lock(i2c)) DRM_ERROR("Failed to lock the bus from SMU"); else @@ -611,13 +626,14 @@ static int trylock_bus(struct i2c_adapter *i2c, unsigned int flags) static void unlock_bus(struct i2c_adapter *i2c, unsigned int flags) { - struct amdgpu_device *adev = to_amdgpu_device(i2c); + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(i2c); + struct amdgpu_device *adev = smu_i2c->adev; if (!smu_v11_0_i2c_bus_unlock(i2c)) DRM_ERROR("Failed to unlock the bus from SMU"); else adev->pm.bus_locked = false; - mutex_unlock(&adev->pm.smu_i2c_mutex); + mutex_unlock(&smu_i2c->mutex); } static const struct i2c_lock_operations smu_v11_0_i2c_i2c_lock_ops = { @@ -706,19 +722,26 @@ static const struct i2c_adapter_quirks smu_v11_0_i2c_control_quirks = { .flags = I2C_AQ_NO_ZERO_LEN, }; -int smu_v11_0_i2c_control_init(struct i2c_adapter *control) +int smu_v11_0_i2c_control_init(struct amdgpu_device *adev) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[0]; + struct i2c_adapter *control = &smu_i2c->adapter; int res; - mutex_init(&adev->pm.smu_i2c_mutex); + smu_i2c->adev = adev; + smu_i2c->port = 0; + mutex_init(&smu_i2c->mutex); control->owner = THIS_MODULE; control->class = I2C_CLASS_HWMON; control->dev.parent = &adev->pdev->dev; control->algo = &smu_v11_0_i2c_algo; - snprintf(control->name, sizeof(control->name), "AMDGPU SMU"); + snprintf(control->name, sizeof(control->name), "AMDGPU SMU 0"); control->lock_ops = &smu_v11_0_i2c_i2c_lock_ops; control->quirks = &smu_v11_0_i2c_control_quirks; + i2c_set_adapdata(control, smu_i2c); + + adev->pm.ras_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter; + adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter; res = i2c_add_adapter(control); if (res) @@ -727,9 +750,13 @@ int smu_v11_0_i2c_control_init(struct i2c_adapter *control) return res; } -void smu_v11_0_i2c_control_fini(struct i2c_adapter *control) +void smu_v11_0_i2c_control_fini(struct amdgpu_device *adev) { + struct i2c_adapter *control = adev->pm.ras_eeprom_i2c_bus; + i2c_del_adapter(control); + adev->pm.ras_eeprom_i2c_bus = NULL; + adev->pm.fru_eeprom_i2c_bus = NULL; } /* diff --git a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.h b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.h index 44467c05f642..96ad14288a0c 100644 --- a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.h +++ b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.h @@ -26,9 +26,9 @@ #include <linux/types.h> -struct i2c_adapter; +struct amdgpu_device; -int smu_v11_0_i2c_control_init(struct i2c_adapter *control); -void smu_v11_0_i2c_control_fini(struct i2c_adapter *control); +int smu_v11_0_i2c_control_init(struct amdgpu_device *adev); +void smu_v11_0_i2c_control_fini(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 0fc1747e4a70..a0235f75dbcb 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -375,39 +375,6 @@ static bool soc15_read_disabled_bios(struct amdgpu_device *adev) return false; } -static bool soc15_read_bios_from_rom(struct amdgpu_device *adev, - u8 *bios, u32 length_bytes) -{ - u32 *dw_ptr; - u32 i, length_dw; - uint32_t rom_index_offset; - uint32_t rom_data_offset; - - if (bios == NULL) - return false; - if (length_bytes == 0) - return false; - /* APU vbios image is part of sbios image */ - if (adev->flags & AMD_IS_APU) - return false; - - dw_ptr = (u32 *)bios; - length_dw = ALIGN(length_bytes, 4) / 4; - - rom_index_offset = - adev->smuio.funcs->get_rom_index_offset(adev); - rom_data_offset = - adev->smuio.funcs->get_rom_data_offset(adev); - - /* set rom index to 0 */ - WREG32(rom_index_offset, 0); - /* read out the rom data */ - for (i = 0; i < length_dw; i++) - dw_ptr[i] = RREG32(rom_data_offset); - - return true; -} - static struct soc15_allowed_register_entry soc15_allowed_read_registers[] = { { SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS)}, { SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS2)}, @@ -925,7 +892,7 @@ static void soc15_pre_asic_init(struct amdgpu_device *adev) static const struct amdgpu_asic_funcs soc15_asic_funcs = { .read_disabled_bios = &soc15_read_disabled_bios, - .read_bios_from_rom = &soc15_read_bios_from_rom, + .read_bios_from_rom = &amdgpu_soc15_read_bios_from_rom, .read_register = &soc15_read_register, .reset = &soc15_asic_reset, .reset_method = &soc15_asic_reset_method, @@ -947,7 +914,7 @@ static const struct amdgpu_asic_funcs soc15_asic_funcs = static const struct amdgpu_asic_funcs vega20_asic_funcs = { .read_disabled_bios = &soc15_read_disabled_bios, - .read_bios_from_rom = &soc15_read_bios_from_rom, + .read_bios_from_rom = &amdgpu_soc15_read_bios_from_rom, .read_register = &soc15_read_register, .reset = &soc15_asic_reset, .reset_method = &soc15_asic_reset_method, @@ -1224,9 +1191,8 @@ static int soc15_common_late_init(void *handle) if (amdgpu_sriov_vf(adev)) xgpu_ai_mailbox_get_irq(adev); - if (adev->nbio.ras_funcs && - adev->nbio.ras_funcs->ras_late_init) - r = adev->nbio.ras_funcs->ras_late_init(adev); + if (adev->nbio.ras && adev->nbio.ras->ras_block.ras_late_init) + r = adev->nbio.ras->ras_block.ras_late_init(adev, NULL); return r; } @@ -1249,9 +1215,8 @@ static int soc15_common_sw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->nbio.ras_funcs && - adev->nbio.ras_funcs->ras_fini) - adev->nbio.ras_funcs->ras_fini(adev); + if (adev->nbio.ras && adev->nbio.ras->ras_block.ras_fini) + adev->nbio.ras->ras_block.ras_fini(adev); if (adev->df.funcs && adev->df.funcs->sw_fini) @@ -1318,11 +1283,11 @@ static int soc15_common_hw_fini(void *handle) if (adev->nbio.ras_if && amdgpu_ras_is_supported(adev, adev->nbio.ras_if->block)) { - if (adev->nbio.ras_funcs && - adev->nbio.ras_funcs->init_ras_controller_interrupt) + if (adev->nbio.ras && + adev->nbio.ras->init_ras_controller_interrupt) amdgpu_irq_put(adev, &adev->nbio.ras_controller_irq, 0); - if (adev->nbio.ras_funcs && - adev->nbio.ras_funcs->init_ras_err_event_athub_interrupt) + if (adev->nbio.ras && + adev->nbio.ras->init_ras_err_event_athub_interrupt) amdgpu_irq_put(adev, &adev->nbio.ras_err_event_athub_irq, 0); } diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h index 473767e03676..acce8c2e0328 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h @@ -28,13 +28,13 @@ #define SOC15_REG_OFFSET(ip, inst, reg) (adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) #define __WREG32_SOC15_RLC__(reg, value, flag, hwip) \ - ((amdgpu_sriov_vf(adev) && adev->gfx.rlc.funcs && adev->gfx.rlc.funcs->sriov_wreg) ? \ - adev->gfx.rlc.funcs->sriov_wreg(adev, reg, value, flag, hwip) : \ + ((amdgpu_sriov_vf(adev) && adev->gfx.rlc.funcs && adev->gfx.rlc.rlcg_reg_access_supported) ? \ + amdgpu_sriov_wreg(adev, reg, value, flag, hwip) : \ WREG32(reg, value)) #define __RREG32_SOC15_RLC__(reg, flag, hwip) \ - ((amdgpu_sriov_vf(adev) && adev->gfx.rlc.funcs && adev->gfx.rlc.funcs->sriov_rreg) ? \ - adev->gfx.rlc.funcs->sriov_rreg(adev, reg, flag, hwip) : \ + ((amdgpu_sriov_vf(adev) && adev->gfx.rlc.funcs && adev->gfx.rlc.rlcg_reg_access_supported) ? \ + amdgpu_sriov_rreg(adev, reg, flag, hwip) : \ RREG32(reg)) #define WREG32_FIELD15(ip, idx, reg, field, val) \ diff --git a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h index 5093826a43d1..509d8a1945eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h +++ b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h @@ -64,7 +64,8 @@ enum ta_ras_status { TA_RAS_STATUS__ERROR_PCS_STATE_ERROR = 0xA016, TA_RAS_STATUS__ERROR_PCS_STATE_HANG = 0xA017, TA_RAS_STATUS__ERROR_PCS_STATE_UNKNOWN = 0xA018, - TA_RAS_STATUS__ERROR_UNSUPPORTED_ERROR_INJ = 0xA019 + TA_RAS_STATUS__ERROR_UNSUPPORTED_ERROR_INJ = 0xA019, + TA_RAS_STATUS__TEE_ERROR_ACCESS_DENIED = 0xA01A }; enum ta_ras_block { diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c index 20b44983ac94..939cb203f7ad 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.c @@ -300,7 +300,6 @@ static void umc_v6_1_query_error_address(struct amdgpu_device *adev, { uint32_t lsb, mc_umc_status_addr; uint64_t mc_umc_status, err_addr, retired_page, mc_umc_addrt0; - struct eeprom_table_record *err_rec; uint32_t channel_index = adev->umc.channel_idx_tbl[umc_inst * adev->umc.channel_inst_num + ch_inst]; if (adev->asic_type == CHIP_ARCTURUS) { @@ -328,8 +327,6 @@ static void umc_v6_1_query_error_address(struct amdgpu_device *adev, return; } - err_rec = &err_data->err_addr[err_data->err_addr_cnt]; - /* calculate error address if ue/ce error is detected */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 || @@ -348,18 +345,9 @@ static void umc_v6_1_query_error_address(struct amdgpu_device *adev, /* we only save ue error information currently, ce is skipped */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) - == 1) { - err_rec->address = err_addr; - /* page frame address is saved */ - err_rec->retired_page = retired_page >> AMDGPU_GPU_PAGE_SHIFT; - err_rec->ts = (uint64_t)ktime_get_real_seconds(); - err_rec->err_type = AMDGPU_RAS_EEPROM_ERR_NON_RECOVERABLE; - err_rec->cu = 0; - err_rec->mem_channel = channel_index; - err_rec->mcumc_id = umc_inst; - - err_data->err_addr_cnt++; - } + == 1) + amdgpu_umc_fill_error_record(err_data, err_addr, + retired_page, channel_index, umc_inst); } /* clear umc status */ @@ -465,10 +453,14 @@ static void umc_v6_1_err_cnt_init(struct amdgpu_device *adev) umc_v6_1_enable_umc_index_mode(adev); } -const struct amdgpu_umc_ras_funcs umc_v6_1_ras_funcs = { - .err_cnt_init = umc_v6_1_err_cnt_init, - .ras_late_init = amdgpu_umc_ras_late_init, - .ras_fini = amdgpu_umc_ras_fini, +const struct amdgpu_ras_block_hw_ops umc_v6_1_ras_hw_ops = { .query_ras_error_count = umc_v6_1_query_ras_error_count, .query_ras_error_address = umc_v6_1_query_ras_error_address, }; + +struct amdgpu_umc_ras umc_v6_1_ras = { + .ras_block = { + .hw_ops = &umc_v6_1_ras_hw_ops, + }, + .err_cnt_init = umc_v6_1_err_cnt_init, +};
\ No newline at end of file diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.h b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.h index 5dc36c730bb2..50c632eb4cc6 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_1.h +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_1.h @@ -45,7 +45,7 @@ /* umc ce count initial value */ #define UMC_V6_1_CE_CNT_INIT (UMC_V6_1_CE_CNT_MAX - UMC_V6_1_CE_INT_THRESHOLD) -extern const struct amdgpu_umc_ras_funcs umc_v6_1_ras_funcs; +extern struct amdgpu_umc_ras umc_v6_1_ras; extern const uint32_t umc_v6_1_channel_idx_tbl[UMC_V6_1_UMC_INSTANCE_NUM][UMC_V6_1_CHANNEL_INSTANCE_NUM]; diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c b/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c index 6dd1e19e8d43..87e4ef18e151 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_7.c @@ -47,6 +47,13 @@ static inline uint32_t get_umc_v6_7_reg_offset(struct amdgpu_device *adev, uint32_t umc_inst, uint32_t ch_inst) { + uint32_t index = umc_inst * adev->umc.channel_inst_num + ch_inst; + + /* adjust umc and channel index offset, + * the register address is not linear on each umc instace */ + umc_inst = index / 4; + ch_inst = index % 4; + return adev->umc.channel_offs * ch_inst + UMC_V6_7_INST_DIST * umc_inst; } @@ -58,42 +65,33 @@ static inline uint32_t get_umc_v6_7_channel_index(struct amdgpu_device *adev, } static void umc_v6_7_ecc_info_query_correctable_error_count(struct amdgpu_device *adev, - uint32_t channel_index, + uint32_t umc_inst, uint32_t ch_inst, unsigned long *error_count) { - uint32_t ecc_err_cnt; uint64_t mc_umc_status; + uint32_t eccinfo_table_idx; struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); - /* - * select the lower chip and check the error count - * skip add error count, calc error counter only from mca_umc_status - */ - ecc_err_cnt = ras->umc_ecc.ecc[channel_index].ce_count_lo_chip; - - /* - * select the higher chip and check the err counter - * skip add error count, calc error counter only from mca_umc_status - */ - ecc_err_cnt = ras->umc_ecc.ecc[channel_index].ce_count_hi_chip; - + eccinfo_table_idx = umc_inst * adev->umc.channel_inst_num + ch_inst; /* check for SRAM correctable error MCUMC_STATUS is a 64 bit register */ - mc_umc_status = ras->umc_ecc.ecc[channel_index].mca_umc_status; + mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1) *error_count += 1; } static void umc_v6_7_ecc_info_querry_uncorrectable_error_count(struct amdgpu_device *adev, - uint32_t channel_index, + uint32_t umc_inst, uint32_t ch_inst, unsigned long *error_count) { uint64_t mc_umc_status; + uint32_t eccinfo_table_idx; struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + eccinfo_table_idx = umc_inst * adev->umc.channel_inst_num + ch_inst; /* check the MCUMC_STATUS */ - mc_umc_status = ras->umc_ecc.ecc[channel_index].mca_umc_status; + mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; if ((REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1 || REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 || @@ -110,42 +108,34 @@ static void umc_v6_7_ecc_info_query_ras_error_count(struct amdgpu_device *adev, uint32_t umc_inst = 0; uint32_t ch_inst = 0; - uint32_t umc_reg_offset = 0; - uint32_t channel_index = 0; /*TODO: driver needs to toggle DF Cstate to ensure * safe access of UMC registers. Will add the protection */ LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { - umc_reg_offset = get_umc_v6_7_reg_offset(adev, - umc_inst, - ch_inst); - channel_index = get_umc_v6_7_channel_index(adev, - umc_inst, - ch_inst); umc_v6_7_ecc_info_query_correctable_error_count(adev, - channel_index, + umc_inst, ch_inst, &(err_data->ce_count)); umc_v6_7_ecc_info_querry_uncorrectable_error_count(adev, - channel_index, + umc_inst, ch_inst, &(err_data->ue_count)); } } static void umc_v6_7_ecc_info_query_error_address(struct amdgpu_device *adev, struct ras_err_data *err_data, - uint32_t umc_reg_offset, uint32_t ch_inst, uint32_t umc_inst) { - uint64_t mc_umc_status, err_addr, retired_page; - struct eeprom_table_record *err_rec; + uint64_t mc_umc_status, err_addr, soc_pa, retired_page, column; uint32_t channel_index; + uint32_t eccinfo_table_idx; struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + eccinfo_table_idx = umc_inst * adev->umc.channel_inst_num + ch_inst; channel_index = adev->umc.channel_idx_tbl[umc_inst * adev->umc.channel_inst_num + ch_inst]; - mc_umc_status = ras->umc_ecc.ecc[channel_index].mca_umc_status; + mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; if (mc_umc_status == 0) return; @@ -153,34 +143,39 @@ static void umc_v6_7_ecc_info_query_error_address(struct amdgpu_device *adev, if (!err_data->err_addr) return; - err_rec = &err_data->err_addr[err_data->err_addr_cnt]; - /* calculate error address if ue/ce error is detected */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 || REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1)) { - err_addr = ras->umc_ecc.ecc[channel_index].mca_umc_addr; + err_addr = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_addr; err_addr = REG_GET_FIELD(err_addr, MCA_UMC_UMC0_MCUMC_ADDRT0, ErrorAddr); /* translate umc channel address to soc pa, 3 parts are included */ - retired_page = ADDR_OF_8KB_BLOCK(err_addr) | + soc_pa = ADDR_OF_8KB_BLOCK(err_addr) | ADDR_OF_256B_BLOCK(channel_index) | OFFSET_IN_256B_BLOCK(err_addr); + /* The umc channel bits are not original values, they are hashed */ + SET_CHANNEL_HASH(channel_index, soc_pa); + + /* clear [C4 C3 C2] in soc physical address */ + soc_pa &= ~(0x7ULL << UMC_V6_7_PA_C2_BIT); + /* we only save ue error information currently, ce is skipped */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1) { - err_rec->address = err_addr; - /* page frame address is saved */ - err_rec->retired_page = retired_page >> AMDGPU_GPU_PAGE_SHIFT; - err_rec->ts = (uint64_t)ktime_get_real_seconds(); - err_rec->err_type = AMDGPU_RAS_EEPROM_ERR_NON_RECOVERABLE; - err_rec->cu = 0; - err_rec->mem_channel = channel_index; - err_rec->mcumc_id = umc_inst; - - err_data->err_addr_cnt++; + /* loop for all possibilities of [C4 C3 C2] */ + for (column = 0; column < UMC_V6_7_NA_MAP_PA_NUM; column++) { + retired_page = soc_pa | (column << UMC_V6_7_PA_C2_BIT); + amdgpu_umc_fill_error_record(err_data, err_addr, + retired_page, channel_index, umc_inst); + + /* shift R14 bit */ + retired_page ^= (0x1ULL << UMC_V6_7_PA_R14_BIT); + amdgpu_umc_fill_error_record(err_data, err_addr, + retired_page, channel_index, umc_inst); + } } } } @@ -192,18 +187,13 @@ static void umc_v6_7_ecc_info_query_ras_error_address(struct amdgpu_device *adev uint32_t umc_inst = 0; uint32_t ch_inst = 0; - uint32_t umc_reg_offset = 0; /*TODO: driver needs to toggle DF Cstate to ensure * safe access of UMC resgisters. Will add the protection * when firmware interface is ready */ LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { - umc_reg_offset = get_umc_v6_7_reg_offset(adev, - umc_inst, - ch_inst); umc_v6_7_ecc_info_query_error_address(adev, err_data, - umc_reg_offset, ch_inst, umc_inst); } @@ -365,9 +355,9 @@ static void umc_v6_7_query_error_address(struct amdgpu_device *adev, uint32_t umc_inst) { uint32_t mc_umc_status_addr; - uint64_t mc_umc_status, err_addr, retired_page, mc_umc_addrt0; - struct eeprom_table_record *err_rec; uint32_t channel_index; + uint64_t mc_umc_status, mc_umc_addrt0; + uint64_t err_addr, soc_pa, retired_page, column; mc_umc_status_addr = SOC15_REG_OFFSET(UMC, 0, regMCA_UMC_UMC0_MCUMC_STATUST0); @@ -385,8 +375,6 @@ static void umc_v6_7_query_error_address(struct amdgpu_device *adev, return; } - err_rec = &err_data->err_addr[err_data->err_addr_cnt]; - channel_index = adev->umc.channel_idx_tbl[umc_inst * adev->umc.channel_inst_num + ch_inst]; @@ -399,23 +387,30 @@ static void umc_v6_7_query_error_address(struct amdgpu_device *adev, err_addr = REG_GET_FIELD(err_addr, MCA_UMC_UMC0_MCUMC_ADDRT0, ErrorAddr); /* translate umc channel address to soc pa, 3 parts are included */ - retired_page = ADDR_OF_8KB_BLOCK(err_addr) | + soc_pa = ADDR_OF_8KB_BLOCK(err_addr) | ADDR_OF_256B_BLOCK(channel_index) | OFFSET_IN_256B_BLOCK(err_addr); + /* The umc channel bits are not original values, they are hashed */ + SET_CHANNEL_HASH(channel_index, soc_pa); + + /* clear [C4 C3 C2] in soc physical address */ + soc_pa &= ~(0x7ULL << UMC_V6_7_PA_C2_BIT); + /* we only save ue error information currently, ce is skipped */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1) { - err_rec->address = err_addr; - /* page frame address is saved */ - err_rec->retired_page = retired_page >> AMDGPU_GPU_PAGE_SHIFT; - err_rec->ts = (uint64_t)ktime_get_real_seconds(); - err_rec->err_type = AMDGPU_RAS_EEPROM_ERR_NON_RECOVERABLE; - err_rec->cu = 0; - err_rec->mem_channel = channel_index; - err_rec->mcumc_id = umc_inst; - - err_data->err_addr_cnt++; + /* loop for all possibilities of [C4 C3 C2] */ + for (column = 0; column < UMC_V6_7_NA_MAP_PA_NUM; column++) { + retired_page = soc_pa | (column << UMC_V6_7_PA_C2_BIT); + amdgpu_umc_fill_error_record(err_data, err_addr, + retired_page, channel_index, umc_inst); + + /* shift R14 bit */ + retired_page ^= (0x1ULL << UMC_V6_7_PA_R14_BIT); + amdgpu_umc_fill_error_record(err_data, err_addr, + retired_page, channel_index, umc_inst); + } } } @@ -463,28 +458,24 @@ static uint32_t umc_v6_7_query_ras_poison_mode_per_channel( static bool umc_v6_7_query_ras_poison_mode(struct amdgpu_device *adev) { - uint32_t umc_inst = 0; - uint32_t ch_inst = 0; uint32_t umc_reg_offset = 0; - LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { - umc_reg_offset = get_umc_v6_7_reg_offset(adev, - umc_inst, - ch_inst); - /* Enabling fatal error in one channel will be considered - as fatal error mode */ - if (umc_v6_7_query_ras_poison_mode_per_channel(adev, umc_reg_offset)) - return false; - } - - return true; + /* Enabling fatal error in umc instance0 channel0 will be + * considered as fatal error mode + */ + umc_reg_offset = get_umc_v6_7_reg_offset(adev, 0, 0); + return !umc_v6_7_query_ras_poison_mode_per_channel(adev, umc_reg_offset); } -const struct amdgpu_umc_ras_funcs umc_v6_7_ras_funcs = { - .ras_late_init = amdgpu_umc_ras_late_init, - .ras_fini = amdgpu_umc_ras_fini, +const struct amdgpu_ras_block_hw_ops umc_v6_7_ras_hw_ops = { .query_ras_error_count = umc_v6_7_query_ras_error_count, .query_ras_error_address = umc_v6_7_query_ras_error_address, +}; + +struct amdgpu_umc_ras umc_v6_7_ras = { + .ras_block = { + .hw_ops = &umc_v6_7_ras_hw_ops, + }, .query_ras_poison_mode = umc_v6_7_query_ras_poison_mode, .ecc_info_query_ras_error_count = umc_v6_7_ecc_info_query_ras_error_count, .ecc_info_query_ras_error_address = umc_v6_7_ecc_info_query_ras_error_address, diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_7.h b/drivers/gpu/drm/amd/amdgpu/umc_v6_7.h index 57f2557e7aca..fe41ed2f5945 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_7.h +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_7.h @@ -41,9 +41,32 @@ #define UMC_V6_7_CHANNEL_INSTANCE_NUM 8 /* total channel instances in one umc block */ #define UMC_V6_7_TOTAL_CHANNEL_NUM (UMC_V6_7_CHANNEL_INSTANCE_NUM * UMC_V6_7_UMC_INSTANCE_NUM) +/* one piece of normalizing address is mapped to 8 pieces of physical address */ +#define UMC_V6_7_NA_MAP_PA_NUM 8 +/* R14 bit shift should be considered, double the number */ +#define UMC_V6_7_BAD_PAGE_NUM_PER_CHANNEL (UMC_V6_7_NA_MAP_PA_NUM * 2) +/* The CH4 bit in SOC physical address */ +#define UMC_V6_7_PA_CH4_BIT 12 +/* The C2 bit in SOC physical address */ +#define UMC_V6_7_PA_C2_BIT 17 +/* The R14 bit in SOC physical address */ +#define UMC_V6_7_PA_R14_BIT 34 /* UMC regiser per channel offset */ #define UMC_V6_7_PER_CHANNEL_OFFSET 0x400 -extern const struct amdgpu_umc_ras_funcs umc_v6_7_ras_funcs; + +/* XOR bit 20, 25, 34 of PA into CH4 bit (bit 12 of PA), + * hash bit is only effective when related setting is enabled + */ +#define CHANNEL_HASH(channel_idx, pa) (((channel_idx) >> 4) ^ \ + (((pa) >> 20) & 0x1ULL & adev->df.hash_status.hash_64k) ^ \ + (((pa) >> 25) & 0x1ULL & adev->df.hash_status.hash_2m) ^ \ + (((pa) >> 34) & 0x1ULL & adev->df.hash_status.hash_1g)) +#define SET_CHANNEL_HASH(channel_idx, pa) do { \ + (pa) &= ~(0x1ULL << UMC_V6_7_PA_CH4_BIT); \ + (pa) |= (CHANNEL_HASH(channel_idx, pa) << UMC_V6_7_PA_CH4_BIT); \ + } while (0) + +extern struct amdgpu_umc_ras umc_v6_7_ras; extern const uint32_t umc_v6_7_channel_idx_tbl_second[UMC_V6_7_UMC_INSTANCE_NUM][UMC_V6_7_CHANNEL_INSTANCE_NUM]; extern const uint32_t diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_7.c b/drivers/gpu/drm/amd/amdgpu/umc_v8_7.c index af59a35788e3..de85a998ef99 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v8_7.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_7.c @@ -40,13 +40,144 @@ const uint32_t {9, 0}, {15, 6} }; -static inline uint32_t get_umc_8_reg_offset(struct amdgpu_device *adev, +static inline uint32_t get_umc_v8_7_reg_offset(struct amdgpu_device *adev, uint32_t umc_inst, uint32_t ch_inst) { return adev->umc.channel_offs*ch_inst + UMC_8_INST_DIST*umc_inst; } +static inline uint32_t get_umc_v8_7_channel_index(struct amdgpu_device *adev, + uint32_t umc_inst, + uint32_t ch_inst) +{ + return adev->umc.channel_idx_tbl[umc_inst * adev->umc.channel_inst_num + ch_inst]; +} + +static void umc_v8_7_ecc_info_query_correctable_error_count(struct amdgpu_device *adev, + uint32_t umc_inst, uint32_t ch_inst, + unsigned long *error_count) +{ + uint64_t mc_umc_status; + uint32_t eccinfo_table_idx; + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + eccinfo_table_idx = umc_inst * adev->umc.channel_inst_num + ch_inst; + + /* check for SRAM correctable error + * MCUMC_STATUS is a 64 bit register + */ + mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; + if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1) + *error_count += 1; +} + +static void umc_v8_7_ecc_info_querry_uncorrectable_error_count(struct amdgpu_device *adev, + uint32_t umc_inst, uint32_t ch_inst, + unsigned long *error_count) +{ + uint64_t mc_umc_status; + uint32_t eccinfo_table_idx; + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + eccinfo_table_idx = umc_inst * adev->umc.channel_inst_num + ch_inst; + + /* check the MCUMC_STATUS */ + mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; + if ((REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1 || + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 || + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC) == 1 || + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UC) == 1 || + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, TCC) == 1)) + *error_count += 1; +} + +static void umc_v8_7_ecc_info_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + + uint32_t umc_inst = 0; + uint32_t ch_inst = 0; + + /* TODO: driver needs to toggle DF Cstate to ensure + * safe access of UMC registers. Will add the protection + */ + LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { + umc_v8_7_ecc_info_query_correctable_error_count(adev, + umc_inst, ch_inst, + &(err_data->ce_count)); + umc_v8_7_ecc_info_querry_uncorrectable_error_count(adev, + umc_inst, ch_inst, + &(err_data->ue_count)); + } +} + +static void umc_v8_7_ecc_info_query_error_address(struct amdgpu_device *adev, + struct ras_err_data *err_data, + uint32_t ch_inst, + uint32_t umc_inst) +{ + uint64_t mc_umc_status, err_addr, retired_page; + uint32_t channel_index; + uint32_t eccinfo_table_idx; + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + eccinfo_table_idx = umc_inst * adev->umc.channel_inst_num + ch_inst; + channel_index = + adev->umc.channel_idx_tbl[umc_inst * adev->umc.channel_inst_num + ch_inst]; + + mc_umc_status = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_status; + + if (mc_umc_status == 0) + return; + + if (!err_data->err_addr) + return; + + /* calculate error address if ue/ce error is detected */ + if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 || + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, CECC) == 1)) { + + err_addr = ras->umc_ecc.ecc[eccinfo_table_idx].mca_umc_addr; + err_addr = REG_GET_FIELD(err_addr, MCA_UMC_UMC0_MCUMC_ADDRT0, ErrorAddr); + + /* translate umc channel address to soc pa, 3 parts are included */ + retired_page = ADDR_OF_4KB_BLOCK(err_addr) | + ADDR_OF_256B_BLOCK(channel_index) | + OFFSET_IN_256B_BLOCK(err_addr); + + /* we only save ue error information currently, ce is skipped */ + if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) + == 1) + amdgpu_umc_fill_error_record(err_data, err_addr, + retired_page, channel_index, umc_inst); + } +} + +static void umc_v8_7_ecc_info_query_ras_error_address(struct amdgpu_device *adev, + void *ras_error_status) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + + uint32_t umc_inst = 0; + uint32_t ch_inst = 0; + + /* TODO: driver needs to toggle DF Cstate to ensure + * safe access of UMC resgisters. Will add the protection + * when firmware interface is ready + */ + LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { + umc_v8_7_ecc_info_query_error_address(adev, + err_data, + ch_inst, + umc_inst); + } +} + static void umc_v8_7_clear_error_count_per_channel(struct amdgpu_device *adev, uint32_t umc_reg_offset) { @@ -92,7 +223,7 @@ static void umc_v8_7_clear_error_count(struct amdgpu_device *adev) uint32_t umc_reg_offset = 0; LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { - umc_reg_offset = get_umc_8_reg_offset(adev, + umc_reg_offset = get_umc_v8_7_reg_offset(adev, umc_inst, ch_inst); @@ -178,7 +309,7 @@ static void umc_v8_7_query_ras_error_count(struct amdgpu_device *adev, uint32_t umc_reg_offset = 0; LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { - umc_reg_offset = get_umc_8_reg_offset(adev, + umc_reg_offset = get_umc_v8_7_reg_offset(adev, umc_inst, ch_inst); @@ -201,7 +332,6 @@ static void umc_v8_7_query_error_address(struct amdgpu_device *adev, { uint32_t lsb, mc_umc_status_addr; uint64_t mc_umc_status, err_addr, retired_page, mc_umc_addrt0; - struct eeprom_table_record *err_rec; uint32_t channel_index = adev->umc.channel_idx_tbl[umc_inst * adev->umc.channel_inst_num + ch_inst]; mc_umc_status_addr = @@ -220,8 +350,6 @@ static void umc_v8_7_query_error_address(struct amdgpu_device *adev, return; } - err_rec = &err_data->err_addr[err_data->err_addr_cnt]; - /* calculate error address if ue/ce error is detected */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) == 1 || @@ -240,18 +368,9 @@ static void umc_v8_7_query_error_address(struct amdgpu_device *adev, /* we only save ue error information currently, ce is skipped */ if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UECC) - == 1) { - err_rec->address = err_addr; - /* page frame address is saved */ - err_rec->retired_page = retired_page >> AMDGPU_GPU_PAGE_SHIFT; - err_rec->ts = (uint64_t)ktime_get_real_seconds(); - err_rec->err_type = AMDGPU_RAS_EEPROM_ERR_NON_RECOVERABLE; - err_rec->cu = 0; - err_rec->mem_channel = channel_index; - err_rec->mcumc_id = umc_inst; - - err_data->err_addr_cnt++; - } + == 1) + amdgpu_umc_fill_error_record(err_data, err_addr, + retired_page, channel_index, umc_inst); } /* clear umc status */ @@ -268,7 +387,7 @@ static void umc_v8_7_query_ras_error_address(struct amdgpu_device *adev, uint32_t umc_reg_offset = 0; LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { - umc_reg_offset = get_umc_8_reg_offset(adev, + umc_reg_offset = get_umc_v8_7_reg_offset(adev, umc_inst, ch_inst); @@ -316,7 +435,7 @@ static void umc_v8_7_err_cnt_init(struct amdgpu_device *adev) uint32_t umc_reg_offset = 0; LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) { - umc_reg_offset = get_umc_8_reg_offset(adev, + umc_reg_offset = get_umc_v8_7_reg_offset(adev, umc_inst, ch_inst); @@ -324,10 +443,16 @@ static void umc_v8_7_err_cnt_init(struct amdgpu_device *adev) } } -const struct amdgpu_umc_ras_funcs umc_v8_7_ras_funcs = { - .err_cnt_init = umc_v8_7_err_cnt_init, - .ras_late_init = amdgpu_umc_ras_late_init, - .ras_fini = amdgpu_umc_ras_fini, +const struct amdgpu_ras_block_hw_ops umc_v8_7_ras_hw_ops = { .query_ras_error_count = umc_v8_7_query_ras_error_count, .query_ras_error_address = umc_v8_7_query_ras_error_address, }; + +struct amdgpu_umc_ras umc_v8_7_ras = { + .ras_block = { + .hw_ops = &umc_v8_7_ras_hw_ops, + }, + .err_cnt_init = umc_v8_7_err_cnt_init, + .ecc_info_query_ras_error_count = umc_v8_7_ecc_info_query_ras_error_count, + .ecc_info_query_ras_error_address = umc_v8_7_ecc_info_query_ras_error_address, +}; diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v8_7.h b/drivers/gpu/drm/amd/amdgpu/umc_v8_7.h index 37e6dc7c28e0..dd4993f5f78f 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v8_7.h +++ b/drivers/gpu/drm/amd/amdgpu/umc_v8_7.h @@ -44,7 +44,7 @@ /* umc ce count initial value */ #define UMC_V8_7_CE_CNT_INIT (UMC_V8_7_CE_CNT_MAX - UMC_V8_7_CE_INT_THRESHOLD) -extern const struct amdgpu_umc_ras_funcs umc_v8_7_ras_funcs; +extern struct amdgpu_umc_ras umc_v8_7_ras; extern const uint32_t umc_v8_7_channel_idx_tbl[UMC_V8_7_UMC_INSTANCE_NUM][UMC_V8_7_CHANNEL_INSTANCE_NUM]; diff --git a/drivers/gpu/drm/amd/amdkfd/Makefile b/drivers/gpu/drm/amd/amdkfd/Makefile index c4f3aff11072..19cfbf9577b4 100644 --- a/drivers/gpu/drm/amd/amdkfd/Makefile +++ b/drivers/gpu/drm/amd/amdkfd/Makefile @@ -51,8 +51,6 @@ AMDKFD_FILES := $(AMDKFD_PATH)/kfd_module.o \ $(AMDKFD_PATH)/kfd_events.o \ $(AMDKFD_PATH)/cik_event_interrupt.o \ $(AMDKFD_PATH)/kfd_int_process_v9.o \ - $(AMDKFD_PATH)/kfd_dbgdev.o \ - $(AMDKFD_PATH)/kfd_dbgmgr.o \ $(AMDKFD_PATH)/kfd_smi_events.o \ $(AMDKFD_PATH)/kfd_crat.o diff --git a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c index d60576ce10cd..5c8023cba196 100644 --- a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c +++ b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c @@ -110,7 +110,7 @@ static void cik_event_interrupt_wq(struct kfd_dev *dev, struct kfd_vm_fault_info info; kfd_smi_event_update_vmfault(dev, pasid); - kfd_process_vm_fault(dev->dqm, pasid); + kfd_dqm_evict_pasid(dev->dqm, pasid); memset(&info, 0, sizeof(info)); amdgpu_amdkfd_gpuvm_get_vm_fault_info(dev->adev, &info); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 4bfc0c8ab764..54d997f304b5 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -33,14 +33,16 @@ #include <linux/time.h> #include <linux/mm.h> #include <linux/mman.h> +#include <linux/ptrace.h> #include <linux/dma-buf.h> +#include <linux/fdtable.h> #include <asm/processor.h> #include "kfd_priv.h" #include "kfd_device_queue_manager.h" -#include "kfd_dbgmgr.h" #include "kfd_svm.h" #include "amdgpu_amdkfd.h" #include "kfd_smi_events.h" +#include "amdgpu_dma_buf.h" static long kfd_ioctl(struct file *, unsigned int, unsigned long); static int kfd_open(struct inode *, struct file *); @@ -292,14 +294,17 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, return err; pr_debug("Looking for gpu id 0x%x\n", args->gpu_id); - dev = kfd_device_by_id(args->gpu_id); - if (!dev) { - pr_debug("Could not find gpu id 0x%x\n", args->gpu_id); - return -EINVAL; - } mutex_lock(&p->mutex); + pdd = kfd_process_device_data_by_id(p, args->gpu_id); + if (!pdd) { + pr_debug("Could not find gpu id 0x%x\n", args->gpu_id); + err = -EINVAL; + goto err_pdd; + } + dev = pdd->dev; + pdd = kfd_bind_process_to_device(dev, p); if (IS_ERR(pdd)) { err = -ESRCH; @@ -310,7 +315,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, p->pasid, dev->id); - err = pqm_create_queue(&p->pqm, dev, filep, &q_properties, &queue_id, + err = pqm_create_queue(&p->pqm, dev, filep, &q_properties, &queue_id, NULL, NULL, NULL, &doorbell_offset_in_process); if (err != 0) goto err_create_queue; @@ -344,6 +349,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, err_create_queue: err_bind_process: +err_pdd: mutex_unlock(&p->mutex); return err; } @@ -490,7 +496,6 @@ static int kfd_ioctl_set_memory_policy(struct file *filep, struct kfd_process *p, void *data) { struct kfd_ioctl_set_memory_policy_args *args = data; - struct kfd_dev *dev; int err = 0; struct kfd_process_device *pdd; enum cache_policy default_policy, alternate_policy; @@ -505,13 +510,15 @@ static int kfd_ioctl_set_memory_policy(struct file *filep, return -EINVAL; } - dev = kfd_device_by_id(args->gpu_id); - if (!dev) - return -EINVAL; - mutex_lock(&p->mutex); + pdd = kfd_process_device_data_by_id(p, args->gpu_id); + if (!pdd) { + pr_debug("Could not find gpu id 0x%x\n", args->gpu_id); + err = -EINVAL; + goto err_pdd; + } - pdd = kfd_bind_process_to_device(dev, p); + pdd = kfd_bind_process_to_device(pdd->dev, p); if (IS_ERR(pdd)) { err = -ESRCH; goto out; @@ -524,7 +531,7 @@ static int kfd_ioctl_set_memory_policy(struct file *filep, (args->alternate_policy == KFD_IOC_CACHE_POLICY_COHERENT) ? cache_policy_coherent : cache_policy_noncoherent; - if (!dev->dqm->ops.set_cache_memory_policy(dev->dqm, + if (!pdd->dev->dqm->ops.set_cache_memory_policy(pdd->dev->dqm, &pdd->qpd, default_policy, alternate_policy, @@ -533,6 +540,7 @@ static int kfd_ioctl_set_memory_policy(struct file *filep, err = -EINVAL; out: +err_pdd: mutex_unlock(&p->mutex); return err; @@ -542,17 +550,18 @@ static int kfd_ioctl_set_trap_handler(struct file *filep, struct kfd_process *p, void *data) { struct kfd_ioctl_set_trap_handler_args *args = data; - struct kfd_dev *dev; int err = 0; struct kfd_process_device *pdd; - dev = kfd_device_by_id(args->gpu_id); - if (!dev) - return -EINVAL; - mutex_lock(&p->mutex); - pdd = kfd_bind_process_to_device(dev, p); + pdd = kfd_process_device_data_by_id(p, args->gpu_id); + if (!pdd) { + err = -EINVAL; + goto err_pdd; + } + + pdd = kfd_bind_process_to_device(pdd->dev, p); if (IS_ERR(pdd)) { err = -ESRCH; goto out; @@ -561,6 +570,7 @@ static int kfd_ioctl_set_trap_handler(struct file *filep, kfd_process_set_trap_handler(&pdd->qpd, args->tba_addr, args->tma_addr); out: +err_pdd: mutex_unlock(&p->mutex); return err; @@ -569,289 +579,40 @@ out: static int kfd_ioctl_dbg_register(struct file *filep, struct kfd_process *p, void *data) { - struct kfd_ioctl_dbg_register_args *args = data; - struct kfd_dev *dev; - struct kfd_dbgmgr *dbgmgr_ptr; - struct kfd_process_device *pdd; - bool create_ok; - long status = 0; - - dev = kfd_device_by_id(args->gpu_id); - if (!dev) - return -EINVAL; - - if (dev->adev->asic_type == CHIP_CARRIZO) { - pr_debug("kfd_ioctl_dbg_register not supported on CZ\n"); - return -EINVAL; - } - - mutex_lock(&p->mutex); - mutex_lock(kfd_get_dbgmgr_mutex()); - - /* - * make sure that we have pdd, if this the first queue created for - * this process - */ - pdd = kfd_bind_process_to_device(dev, p); - if (IS_ERR(pdd)) { - status = PTR_ERR(pdd); - goto out; - } - - if (!dev->dbgmgr) { - /* In case of a legal call, we have no dbgmgr yet */ - create_ok = kfd_dbgmgr_create(&dbgmgr_ptr, dev); - if (create_ok) { - status = kfd_dbgmgr_register(dbgmgr_ptr, p); - if (status != 0) - kfd_dbgmgr_destroy(dbgmgr_ptr); - else - dev->dbgmgr = dbgmgr_ptr; - } - } else { - pr_debug("debugger already registered\n"); - status = -EINVAL; - } - -out: - mutex_unlock(kfd_get_dbgmgr_mutex()); - mutex_unlock(&p->mutex); - - return status; + return -EPERM; } static int kfd_ioctl_dbg_unregister(struct file *filep, struct kfd_process *p, void *data) { - struct kfd_ioctl_dbg_unregister_args *args = data; - struct kfd_dev *dev; - long status; - - dev = kfd_device_by_id(args->gpu_id); - if (!dev || !dev->dbgmgr) - return -EINVAL; - - if (dev->adev->asic_type == CHIP_CARRIZO) { - pr_debug("kfd_ioctl_dbg_unregister not supported on CZ\n"); - return -EINVAL; - } - - mutex_lock(kfd_get_dbgmgr_mutex()); - - status = kfd_dbgmgr_unregister(dev->dbgmgr, p); - if (!status) { - kfd_dbgmgr_destroy(dev->dbgmgr); - dev->dbgmgr = NULL; - } - - mutex_unlock(kfd_get_dbgmgr_mutex()); - - return status; + return -EPERM; } -/* - * Parse and generate variable size data structure for address watch. - * Total size of the buffer and # watch points is limited in order - * to prevent kernel abuse. (no bearing to the much smaller HW limitation - * which is enforced by dbgdev module) - * please also note that the watch address itself are not "copied from user", - * since it be set into the HW in user mode values. - * - */ static int kfd_ioctl_dbg_address_watch(struct file *filep, struct kfd_process *p, void *data) { - struct kfd_ioctl_dbg_address_watch_args *args = data; - struct kfd_dev *dev; - struct dbg_address_watch_info aw_info; - unsigned char *args_buff; - long status; - void __user *cmd_from_user; - uint64_t watch_mask_value = 0; - unsigned int args_idx = 0; - - memset((void *) &aw_info, 0, sizeof(struct dbg_address_watch_info)); - - dev = kfd_device_by_id(args->gpu_id); - if (!dev) - return -EINVAL; - - if (dev->adev->asic_type == CHIP_CARRIZO) { - pr_debug("kfd_ioctl_dbg_wave_control not supported on CZ\n"); - return -EINVAL; - } - - cmd_from_user = (void __user *) args->content_ptr; - - /* Validate arguments */ - - if ((args->buf_size_in_bytes > MAX_ALLOWED_AW_BUFF_SIZE) || - (args->buf_size_in_bytes <= sizeof(*args) + sizeof(int) * 2) || - (cmd_from_user == NULL)) - return -EINVAL; - - /* this is the actual buffer to work with */ - args_buff = memdup_user(cmd_from_user, - args->buf_size_in_bytes - sizeof(*args)); - if (IS_ERR(args_buff)) - return PTR_ERR(args_buff); - - aw_info.process = p; - - aw_info.num_watch_points = *((uint32_t *)(&args_buff[args_idx])); - args_idx += sizeof(aw_info.num_watch_points); - - aw_info.watch_mode = (enum HSA_DBG_WATCH_MODE *) &args_buff[args_idx]; - args_idx += sizeof(enum HSA_DBG_WATCH_MODE) * aw_info.num_watch_points; - - /* - * set watch address base pointer to point on the array base - * within args_buff - */ - aw_info.watch_address = (uint64_t *) &args_buff[args_idx]; - - /* skip over the addresses buffer */ - args_idx += sizeof(aw_info.watch_address) * aw_info.num_watch_points; - - if (args_idx >= args->buf_size_in_bytes - sizeof(*args)) { - status = -EINVAL; - goto out; - } - - watch_mask_value = (uint64_t) args_buff[args_idx]; - - if (watch_mask_value > 0) { - /* - * There is an array of masks. - * set watch mask base pointer to point on the array base - * within args_buff - */ - aw_info.watch_mask = (uint64_t *) &args_buff[args_idx]; - - /* skip over the masks buffer */ - args_idx += sizeof(aw_info.watch_mask) * - aw_info.num_watch_points; - } else { - /* just the NULL mask, set to NULL and skip over it */ - aw_info.watch_mask = NULL; - args_idx += sizeof(aw_info.watch_mask); - } - - if (args_idx >= args->buf_size_in_bytes - sizeof(args)) { - status = -EINVAL; - goto out; - } - - /* Currently HSA Event is not supported for DBG */ - aw_info.watch_event = NULL; - - mutex_lock(kfd_get_dbgmgr_mutex()); - - status = kfd_dbgmgr_address_watch(dev->dbgmgr, &aw_info); - - mutex_unlock(kfd_get_dbgmgr_mutex()); - -out: - kfree(args_buff); - - return status; + return -EPERM; } /* Parse and generate fixed size data structure for wave control */ static int kfd_ioctl_dbg_wave_control(struct file *filep, struct kfd_process *p, void *data) { - struct kfd_ioctl_dbg_wave_control_args *args = data; - struct kfd_dev *dev; - struct dbg_wave_control_info wac_info; - unsigned char *args_buff; - uint32_t computed_buff_size; - long status; - void __user *cmd_from_user; - unsigned int args_idx = 0; - - memset((void *) &wac_info, 0, sizeof(struct dbg_wave_control_info)); - - /* we use compact form, independent of the packing attribute value */ - computed_buff_size = sizeof(*args) + - sizeof(wac_info.mode) + - sizeof(wac_info.operand) + - sizeof(wac_info.dbgWave_msg.DbgWaveMsg) + - sizeof(wac_info.dbgWave_msg.MemoryVA) + - sizeof(wac_info.trapId); - - dev = kfd_device_by_id(args->gpu_id); - if (!dev) - return -EINVAL; - - if (dev->adev->asic_type == CHIP_CARRIZO) { - pr_debug("kfd_ioctl_dbg_wave_control not supported on CZ\n"); - return -EINVAL; - } - - /* input size must match the computed "compact" size */ - if (args->buf_size_in_bytes != computed_buff_size) { - pr_debug("size mismatch, computed : actual %u : %u\n", - args->buf_size_in_bytes, computed_buff_size); - return -EINVAL; - } - - cmd_from_user = (void __user *) args->content_ptr; - - if (cmd_from_user == NULL) - return -EINVAL; - - /* copy the entire buffer from user */ - - args_buff = memdup_user(cmd_from_user, - args->buf_size_in_bytes - sizeof(*args)); - if (IS_ERR(args_buff)) - return PTR_ERR(args_buff); - - /* move ptr to the start of the "pay-load" area */ - wac_info.process = p; - - wac_info.operand = *((enum HSA_DBG_WAVEOP *)(&args_buff[args_idx])); - args_idx += sizeof(wac_info.operand); - - wac_info.mode = *((enum HSA_DBG_WAVEMODE *)(&args_buff[args_idx])); - args_idx += sizeof(wac_info.mode); - - wac_info.trapId = *((uint32_t *)(&args_buff[args_idx])); - args_idx += sizeof(wac_info.trapId); - - wac_info.dbgWave_msg.DbgWaveMsg.WaveMsgInfoGen2.Value = - *((uint32_t *)(&args_buff[args_idx])); - wac_info.dbgWave_msg.MemoryVA = NULL; - - mutex_lock(kfd_get_dbgmgr_mutex()); - - pr_debug("Calling dbg manager process %p, operand %u, mode %u, trapId %u, message %u\n", - wac_info.process, wac_info.operand, - wac_info.mode, wac_info.trapId, - wac_info.dbgWave_msg.DbgWaveMsg.WaveMsgInfoGen2.Value); - - status = kfd_dbgmgr_wave_control(dev->dbgmgr, &wac_info); - - pr_debug("Returned status of dbg manager is %ld\n", status); - - mutex_unlock(kfd_get_dbgmgr_mutex()); - - kfree(args_buff); - - return status; + return -EPERM; } static int kfd_ioctl_get_clock_counters(struct file *filep, struct kfd_process *p, void *data) { struct kfd_ioctl_get_clock_counters_args *args = data; - struct kfd_dev *dev; + struct kfd_process_device *pdd; - dev = kfd_device_by_id(args->gpu_id); - if (dev) + mutex_lock(&p->mutex); + pdd = kfd_process_device_data_by_id(p, args->gpu_id); + mutex_unlock(&p->mutex); + if (pdd) /* Reading GPU clock counter from KGD */ - args->gpu_clock_counter = amdgpu_amdkfd_get_gpu_clock_counter(dev->adev); + args->gpu_clock_counter = amdgpu_amdkfd_get_gpu_clock_counter(pdd->dev->adev); else /* Node without GPU resource */ args->gpu_clock_counter = 0; @@ -1007,57 +768,11 @@ static int kfd_ioctl_create_event(struct file *filp, struct kfd_process *p, * through the event_page_offset field. */ if (args->event_page_offset) { - struct kfd_dev *kfd; - struct kfd_process_device *pdd; - void *mem, *kern_addr; - uint64_t size; - - kfd = kfd_device_by_id(GET_GPU_ID(args->event_page_offset)); - if (!kfd) { - pr_err("Getting device by id failed in %s\n", __func__); - return -EINVAL; - } - mutex_lock(&p->mutex); - - if (p->signal_page) { - pr_err("Event page is already set\n"); - err = -EINVAL; - goto out_unlock; - } - - pdd = kfd_bind_process_to_device(kfd, p); - if (IS_ERR(pdd)) { - err = PTR_ERR(pdd); - goto out_unlock; - } - - mem = kfd_process_device_translate_handle(pdd, - GET_IDR_HANDLE(args->event_page_offset)); - if (!mem) { - pr_err("Can't find BO, offset is 0x%llx\n", - args->event_page_offset); - err = -EINVAL; - goto out_unlock; - } - - err = amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel(kfd->adev, - mem, &kern_addr, &size); - if (err) { - pr_err("Failed to map event page to kernel\n"); - goto out_unlock; - } - - err = kfd_event_page_set(p, kern_addr, size); - if (err) { - pr_err("Failed to set event page\n"); - amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(kfd->adev, mem); - goto out_unlock; - } - - p->signal_handle = args->event_page_offset; - + err = kfd_kmap_event_page(p, args->event_page_offset); mutex_unlock(&p->mutex); + if (err) + return err; } err = kfd_event_create(filp, p, args->event_type, @@ -1066,10 +781,7 @@ static int kfd_ioctl_create_event(struct file *filp, struct kfd_process *p, &args->event_page_offset, &args->event_slot_index); - return err; - -out_unlock: - mutex_unlock(&p->mutex); + pr_debug("Created event (id:0x%08x) (%s)\n", args->event_id, __func__); return err; } @@ -1118,11 +830,13 @@ static int kfd_ioctl_set_scratch_backing_va(struct file *filep, struct kfd_dev *dev; long err; - dev = kfd_device_by_id(args->gpu_id); - if (!dev) - return -EINVAL; - mutex_lock(&p->mutex); + pdd = kfd_process_device_data_by_id(p, args->gpu_id); + if (!pdd) { + err = -EINVAL; + goto err_pdd; + } + dev = pdd->dev; pdd = kfd_bind_process_to_device(dev, p); if (IS_ERR(pdd)) { @@ -1142,6 +856,7 @@ static int kfd_ioctl_set_scratch_backing_va(struct file *filep, return 0; bind_process_to_device_fail: +err_pdd: mutex_unlock(&p->mutex); return err; } @@ -1150,15 +865,17 @@ static int kfd_ioctl_get_tile_config(struct file *filep, struct kfd_process *p, void *data) { struct kfd_ioctl_get_tile_config_args *args = data; - struct kfd_dev *dev; + struct kfd_process_device *pdd; struct tile_config config; int err = 0; - dev = kfd_device_by_id(args->gpu_id); - if (!dev) + mutex_lock(&p->mutex); + pdd = kfd_process_device_data_by_id(p, args->gpu_id); + mutex_unlock(&p->mutex); + if (!pdd) return -EINVAL; - amdgpu_amdkfd_get_tile_config(dev->adev, &config); + amdgpu_amdkfd_get_tile_config(pdd->dev->adev, &config); args->gb_addr_config = config.gb_addr_config; args->num_banks = config.num_banks; @@ -1193,40 +910,37 @@ static int kfd_ioctl_acquire_vm(struct file *filep, struct kfd_process *p, { struct kfd_ioctl_acquire_vm_args *args = data; struct kfd_process_device *pdd; - struct kfd_dev *dev; struct file *drm_file; int ret; - dev = kfd_device_by_id(args->gpu_id); - if (!dev) - return -EINVAL; - drm_file = fget(args->drm_fd); if (!drm_file) return -EINVAL; mutex_lock(&p->mutex); - - pdd = kfd_get_process_device_data(dev, p); + pdd = kfd_process_device_data_by_id(p, args->gpu_id); if (!pdd) { ret = -EINVAL; - goto err_unlock; + goto err_pdd; } if (pdd->drm_file) { ret = pdd->drm_file == drm_file ? 0 : -EBUSY; - goto err_unlock; + goto err_drm_file; } ret = kfd_process_device_init_vm(pdd, drm_file); if (ret) goto err_unlock; + /* On success, the PDD keeps the drm_file reference */ mutex_unlock(&p->mutex); return 0; err_unlock: +err_pdd: +err_drm_file: mutex_unlock(&p->mutex); fput(drm_file); return ret; @@ -1283,19 +997,23 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep, } mutex_unlock(&p->svms.lock); #endif - dev = kfd_device_by_id(args->gpu_id); - if (!dev) - return -EINVAL; + mutex_lock(&p->mutex); + pdd = kfd_process_device_data_by_id(p, args->gpu_id); + if (!pdd) { + err = -EINVAL; + goto err_pdd; + } + + dev = pdd->dev; if ((flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) && (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) && !kfd_dev_is_large_bar(dev)) { pr_err("Alloc host visible vram on small bar is not allowed\n"); - return -EINVAL; + err = -EINVAL; + goto err_large_bar; } - mutex_lock(&p->mutex); - pdd = kfd_bind_process_to_device(dev, p); if (IS_ERR(pdd)) { err = PTR_ERR(pdd); @@ -1323,7 +1041,7 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep, err = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( dev->adev, args->va_addr, args->size, pdd->drm_priv, (struct kgd_mem **) &mem, &offset, - flags); + flags, false); if (err) goto err_unlock; @@ -1356,6 +1074,8 @@ err_free: amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->adev, (struct kgd_mem *)mem, pdd->drm_priv, NULL); err_unlock: +err_pdd: +err_large_bar: mutex_unlock(&p->mutex); return err; } @@ -1366,14 +1086,9 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep, struct kfd_ioctl_free_memory_of_gpu_args *args = data; struct kfd_process_device *pdd; void *mem; - struct kfd_dev *dev; int ret; uint64_t size = 0; - dev = kfd_device_by_id(GET_GPU_ID(args->handle)); - if (!dev) - return -EINVAL; - mutex_lock(&p->mutex); /* * Safeguard to prevent user space from freeing signal BO. @@ -1385,11 +1100,11 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep, goto err_unlock; } - pdd = kfd_get_process_device_data(dev, p); + pdd = kfd_process_device_data_by_id(p, GET_GPU_ID(args->handle)); if (!pdd) { pr_err("Process device data doesn't exist\n"); ret = -EINVAL; - goto err_unlock; + goto err_pdd; } mem = kfd_process_device_translate_handle( @@ -1399,7 +1114,7 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep, goto err_unlock; } - ret = amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->adev, + ret = amdgpu_amdkfd_gpuvm_free_memory_of_gpu(pdd->dev->adev, (struct kgd_mem *)mem, pdd->drm_priv, &size); /* If freeing the buffer failed, leave the handle in place for @@ -1412,26 +1127,30 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep, WRITE_ONCE(pdd->vram_usage, pdd->vram_usage - size); err_unlock: +err_pdd: mutex_unlock(&p->mutex); return ret; } +static bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev) { + return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) || + (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) && + dev->adev->sdma.instance[0].fw_version >= 18) || + KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0); +} + static int kfd_ioctl_map_memory_to_gpu(struct file *filep, struct kfd_process *p, void *data) { struct kfd_ioctl_map_memory_to_gpu_args *args = data; struct kfd_process_device *pdd, *peer_pdd; void *mem; - struct kfd_dev *dev, *peer; + struct kfd_dev *dev; long err = 0; int i; uint32_t *devices_arr = NULL; bool table_freed = false; - dev = kfd_device_by_id(GET_GPU_ID(args->handle)); - if (!dev) - return -EINVAL; - if (!args->n_devices) { pr_debug("Device IDs array empty\n"); return -EINVAL; @@ -1455,6 +1174,12 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep, } mutex_lock(&p->mutex); + pdd = kfd_process_device_data_by_id(p, GET_GPU_ID(args->handle)); + if (!pdd) { + err = -EINVAL; + goto get_process_device_data_failed; + } + dev = pdd->dev; pdd = kfd_bind_process_to_device(dev, p); if (IS_ERR(pdd)) { @@ -1470,21 +1195,22 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep, } for (i = args->n_success; i < args->n_devices; i++) { - peer = kfd_device_by_id(devices_arr[i]); - if (!peer) { + peer_pdd = kfd_process_device_data_by_id(p, devices_arr[i]); + if (!peer_pdd) { pr_debug("Getting device by id failed for 0x%x\n", devices_arr[i]); err = -EINVAL; goto get_mem_obj_from_handle_failed; } - peer_pdd = kfd_bind_process_to_device(peer, p); + peer_pdd = kfd_bind_process_to_device(peer_pdd->dev, p); if (IS_ERR(peer_pdd)) { err = PTR_ERR(peer_pdd); goto get_mem_obj_from_handle_failed; } + err = amdgpu_amdkfd_gpuvm_map_memory_to_gpu( - peer->adev, (struct kgd_mem *)mem, + peer_pdd->dev->adev, (struct kgd_mem *)mem, peer_pdd->drm_priv, &table_freed); if (err) { pr_err("Failed to map to gpu %d/%d\n", @@ -1503,12 +1229,9 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep, } /* Flush TLBs after waiting for the page table updates to complete */ - if (table_freed) { + if (table_freed || !kfd_flush_tlb_after_unmap(dev)) { for (i = 0; i < args->n_devices; i++) { - peer = kfd_device_by_id(devices_arr[i]); - if (WARN_ON_ONCE(!peer)) - continue; - peer_pdd = kfd_get_process_device_data(peer, p); + peer_pdd = kfd_process_device_data_by_id(p, devices_arr[i]); if (WARN_ON_ONCE(!peer_pdd)) continue; kfd_flush_tlb(peer_pdd, TLB_FLUSH_LEGACY); @@ -1518,6 +1241,7 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep, return err; +get_process_device_data_failed: bind_process_to_device_failed: get_mem_obj_from_handle_failed: map_memory_to_gpu_failed: @@ -1535,14 +1259,9 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep, struct kfd_ioctl_unmap_memory_from_gpu_args *args = data; struct kfd_process_device *pdd, *peer_pdd; void *mem; - struct kfd_dev *dev, *peer; long err = 0; uint32_t *devices_arr = NULL, i; - dev = kfd_device_by_id(GET_GPU_ID(args->handle)); - if (!dev) - return -EINVAL; - if (!args->n_devices) { pr_debug("Device IDs array empty\n"); return -EINVAL; @@ -1566,8 +1285,7 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep, } mutex_lock(&p->mutex); - - pdd = kfd_get_process_device_data(dev, p); + pdd = kfd_process_device_data_by_id(p, GET_GPU_ID(args->handle)); if (!pdd) { err = -EINVAL; goto bind_process_to_device_failed; @@ -1581,19 +1299,13 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep, } for (i = args->n_success; i < args->n_devices; i++) { - peer = kfd_device_by_id(devices_arr[i]); - if (!peer) { - err = -EINVAL; - goto get_mem_obj_from_handle_failed; - } - - peer_pdd = kfd_get_process_device_data(peer, p); + peer_pdd = kfd_process_device_data_by_id(p, devices_arr[i]); if (!peer_pdd) { - err = -ENODEV; + err = -EINVAL; goto get_mem_obj_from_handle_failed; } err = amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu( - peer->adev, (struct kgd_mem *)mem, peer_pdd->drm_priv); + peer_pdd->dev->adev, (struct kgd_mem *)mem, peer_pdd->drm_priv); if (err) { pr_err("Failed to unmap from gpu %d/%d\n", i, args->n_devices); @@ -1603,8 +1315,8 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep, } mutex_unlock(&p->mutex); - if (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2)) { - err = amdgpu_amdkfd_gpuvm_sync_memory(dev->adev, + if (kfd_flush_tlb_after_unmap(pdd->dev)) { + err = amdgpu_amdkfd_gpuvm_sync_memory(pdd->dev->adev, (struct kgd_mem *) mem, true); if (err) { pr_debug("Sync memory failed, wait interrupted by user signal\n"); @@ -1613,10 +1325,7 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep, /* Flush TLBs after waiting for the page table updates to complete */ for (i = 0; i < args->n_devices; i++) { - peer = kfd_device_by_id(devices_arr[i]); - if (WARN_ON_ONCE(!peer)) - continue; - peer_pdd = kfd_get_process_device_data(peer, p); + peer_pdd = kfd_process_device_data_by_id(p, devices_arr[i]); if (WARN_ON_ONCE(!peer_pdd)) continue; kfd_flush_tlb(peer_pdd, TLB_FLUSH_HEAVYWEIGHT); @@ -1736,29 +1445,29 @@ static int kfd_ioctl_import_dmabuf(struct file *filep, struct kfd_ioctl_import_dmabuf_args *args = data; struct kfd_process_device *pdd; struct dma_buf *dmabuf; - struct kfd_dev *dev; int idr_handle; uint64_t size; void *mem; int r; - dev = kfd_device_by_id(args->gpu_id); - if (!dev) - return -EINVAL; - dmabuf = dma_buf_get(args->dmabuf_fd); if (IS_ERR(dmabuf)) return PTR_ERR(dmabuf); mutex_lock(&p->mutex); + pdd = kfd_process_device_data_by_id(p, args->gpu_id); + if (!pdd) { + r = -EINVAL; + goto err_unlock; + } - pdd = kfd_bind_process_to_device(dev, p); + pdd = kfd_bind_process_to_device(pdd->dev, p); if (IS_ERR(pdd)) { r = PTR_ERR(pdd); goto err_unlock; } - r = amdgpu_amdkfd_gpuvm_import_dmabuf(dev->adev, dmabuf, + r = amdgpu_amdkfd_gpuvm_import_dmabuf(pdd->dev->adev, dmabuf, args->va_addr, pdd->drm_priv, (struct kgd_mem **)&mem, &size, NULL); @@ -1779,7 +1488,7 @@ static int kfd_ioctl_import_dmabuf(struct file *filep, return 0; err_free: - amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->adev, (struct kgd_mem *)mem, + amdgpu_amdkfd_gpuvm_free_memory_of_gpu(pdd->dev->adev, (struct kgd_mem *)mem, pdd->drm_priv, NULL); err_unlock: mutex_unlock(&p->mutex); @@ -1792,13 +1501,16 @@ static int kfd_ioctl_smi_events(struct file *filep, struct kfd_process *p, void *data) { struct kfd_ioctl_smi_events_args *args = data; - struct kfd_dev *dev; + struct kfd_process_device *pdd; - dev = kfd_device_by_id(args->gpuid); - if (!dev) + mutex_lock(&p->mutex); + + pdd = kfd_process_device_data_by_id(p, args->gpuid); + mutex_unlock(&p->mutex); + if (!pdd) return -EINVAL; - return kfd_smi_event_open(dev, &args->anon_fd); + return kfd_smi_event_open(pdd->dev, &args->anon_fd); } static int kfd_ioctl_set_xnack_mode(struct file *filep, @@ -1840,13 +1552,9 @@ static int kfd_ioctl_svm(struct file *filep, struct kfd_process *p, void *data) if (!args->start_addr || !args->size) return -EINVAL; - mutex_lock(&p->mutex); - r = svm_ioctl(p, args->op, args->start_addr, args->size, args->nattr, args->attrs); - mutex_unlock(&p->mutex); - return r; } #else @@ -1856,6 +1564,1024 @@ static int kfd_ioctl_svm(struct file *filep, struct kfd_process *p, void *data) } #endif +static int criu_checkpoint_process(struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_offset) +{ + struct kfd_criu_process_priv_data process_priv; + int ret; + + memset(&process_priv, 0, sizeof(process_priv)); + + process_priv.version = KFD_CRIU_PRIV_VERSION; + /* For CR, we don't consider negative xnack mode which is used for + * querying without changing it, here 0 simply means disabled and 1 + * means enabled so retry for finding a valid PTE. + */ + process_priv.xnack_mode = p->xnack_enabled ? 1 : 0; + + ret = copy_to_user(user_priv_data + *priv_offset, + &process_priv, sizeof(process_priv)); + + if (ret) { + pr_err("Failed to copy process information to user\n"); + ret = -EFAULT; + } + + *priv_offset += sizeof(process_priv); + return ret; +} + +static int criu_checkpoint_devices(struct kfd_process *p, + uint32_t num_devices, + uint8_t __user *user_addr, + uint8_t __user *user_priv_data, + uint64_t *priv_offset) +{ + struct kfd_criu_device_priv_data *device_priv = NULL; + struct kfd_criu_device_bucket *device_buckets = NULL; + int ret = 0, i; + + device_buckets = kvzalloc(num_devices * sizeof(*device_buckets), GFP_KERNEL); + if (!device_buckets) { + ret = -ENOMEM; + goto exit; + } + + device_priv = kvzalloc(num_devices * sizeof(*device_priv), GFP_KERNEL); + if (!device_priv) { + ret = -ENOMEM; + goto exit; + } + + for (i = 0; i < num_devices; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + + device_buckets[i].user_gpu_id = pdd->user_gpu_id; + device_buckets[i].actual_gpu_id = pdd->dev->id; + + /* + * priv_data does not contain useful information for now and is reserved for + * future use, so we do not set its contents. + */ + } + + ret = copy_to_user(user_addr, device_buckets, num_devices * sizeof(*device_buckets)); + if (ret) { + pr_err("Failed to copy device information to user\n"); + ret = -EFAULT; + goto exit; + } + + ret = copy_to_user(user_priv_data + *priv_offset, + device_priv, + num_devices * sizeof(*device_priv)); + if (ret) { + pr_err("Failed to copy device information to user\n"); + ret = -EFAULT; + } + *priv_offset += num_devices * sizeof(*device_priv); + +exit: + kvfree(device_buckets); + kvfree(device_priv); + return ret; +} + +static uint32_t get_process_num_bos(struct kfd_process *p) +{ + uint32_t num_of_bos = 0; + int i; + + /* Run over all PDDs of the process */ + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + void *mem; + int id; + + idr_for_each_entry(&pdd->alloc_idr, mem, id) { + struct kgd_mem *kgd_mem = (struct kgd_mem *)mem; + + if ((uint64_t)kgd_mem->va > pdd->gpuvm_base) + num_of_bos++; + } + } + return num_of_bos; +} + +static int criu_get_prime_handle(struct drm_gem_object *gobj, int flags, + u32 *shared_fd) +{ + struct dma_buf *dmabuf; + int ret; + + dmabuf = amdgpu_gem_prime_export(gobj, flags); + if (IS_ERR(dmabuf)) { + ret = PTR_ERR(dmabuf); + pr_err("dmabuf export failed for the BO\n"); + return ret; + } + + ret = dma_buf_fd(dmabuf, flags); + if (ret < 0) { + pr_err("dmabuf create fd failed, ret:%d\n", ret); + goto out_free_dmabuf; + } + + *shared_fd = ret; + return 0; + +out_free_dmabuf: + dma_buf_put(dmabuf); + return ret; +} + +static int criu_checkpoint_bos(struct kfd_process *p, + uint32_t num_bos, + uint8_t __user *user_bos, + uint8_t __user *user_priv_data, + uint64_t *priv_offset) +{ + struct kfd_criu_bo_bucket *bo_buckets; + struct kfd_criu_bo_priv_data *bo_privs; + int ret = 0, pdd_index, bo_index = 0, id; + void *mem; + + bo_buckets = kvzalloc(num_bos * sizeof(*bo_buckets), GFP_KERNEL); + if (!bo_buckets) + return -ENOMEM; + + bo_privs = kvzalloc(num_bos * sizeof(*bo_privs), GFP_KERNEL); + if (!bo_privs) { + ret = -ENOMEM; + goto exit; + } + + for (pdd_index = 0; pdd_index < p->n_pdds; pdd_index++) { + struct kfd_process_device *pdd = p->pdds[pdd_index]; + struct amdgpu_bo *dumper_bo; + struct kgd_mem *kgd_mem; + + idr_for_each_entry(&pdd->alloc_idr, mem, id) { + struct kfd_criu_bo_bucket *bo_bucket; + struct kfd_criu_bo_priv_data *bo_priv; + int i, dev_idx = 0; + + if (!mem) { + ret = -ENOMEM; + goto exit; + } + + kgd_mem = (struct kgd_mem *)mem; + dumper_bo = kgd_mem->bo; + + if ((uint64_t)kgd_mem->va <= pdd->gpuvm_base) + continue; + + bo_bucket = &bo_buckets[bo_index]; + bo_priv = &bo_privs[bo_index]; + + bo_bucket->gpu_id = pdd->user_gpu_id; + bo_bucket->addr = (uint64_t)kgd_mem->va; + bo_bucket->size = amdgpu_bo_size(dumper_bo); + bo_bucket->alloc_flags = (uint32_t)kgd_mem->alloc_flags; + bo_priv->idr_handle = id; + + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { + ret = amdgpu_ttm_tt_get_userptr(&dumper_bo->tbo, + &bo_priv->user_addr); + if (ret) { + pr_err("Failed to obtain user address for user-pointer bo\n"); + goto exit; + } + } + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { + ret = criu_get_prime_handle(&dumper_bo->tbo.base, + bo_bucket->alloc_flags & + KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE ? DRM_RDWR : 0, + &bo_bucket->dmabuf_fd); + if (ret) + goto exit; + } + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) + bo_bucket->offset = KFD_MMAP_TYPE_DOORBELL | + KFD_MMAP_GPU_ID(pdd->dev->id); + else if (bo_bucket->alloc_flags & + KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) + bo_bucket->offset = KFD_MMAP_TYPE_MMIO | + KFD_MMAP_GPU_ID(pdd->dev->id); + else + bo_bucket->offset = amdgpu_bo_mmap_offset(dumper_bo); + + for (i = 0; i < p->n_pdds; i++) { + if (amdgpu_amdkfd_bo_mapped_to_dev(p->pdds[i]->dev->adev, kgd_mem)) + bo_priv->mapped_gpuids[dev_idx++] = p->pdds[i]->user_gpu_id; + } + + pr_debug("bo_size = 0x%llx, bo_addr = 0x%llx bo_offset = 0x%llx\n" + "gpu_id = 0x%x alloc_flags = 0x%x idr_handle = 0x%x", + bo_bucket->size, + bo_bucket->addr, + bo_bucket->offset, + bo_bucket->gpu_id, + bo_bucket->alloc_flags, + bo_priv->idr_handle); + bo_index++; + } + } + + ret = copy_to_user(user_bos, bo_buckets, num_bos * sizeof(*bo_buckets)); + if (ret) { + pr_err("Failed to copy BO information to user\n"); + ret = -EFAULT; + goto exit; + } + + ret = copy_to_user(user_priv_data + *priv_offset, bo_privs, num_bos * sizeof(*bo_privs)); + if (ret) { + pr_err("Failed to copy BO priv information to user\n"); + ret = -EFAULT; + goto exit; + } + + *priv_offset += num_bos * sizeof(*bo_privs); + +exit: + while (ret && bo_index--) { + if (bo_buckets[bo_index].alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) + close_fd(bo_buckets[bo_index].dmabuf_fd); + } + + kvfree(bo_buckets); + kvfree(bo_privs); + return ret; +} + +static int criu_get_process_object_info(struct kfd_process *p, + uint32_t *num_devices, + uint32_t *num_bos, + uint32_t *num_objects, + uint64_t *objs_priv_size) +{ + uint64_t queues_priv_data_size, svm_priv_data_size, priv_size; + uint32_t num_queues, num_events, num_svm_ranges; + int ret; + + *num_devices = p->n_pdds; + *num_bos = get_process_num_bos(p); + + ret = kfd_process_get_queue_info(p, &num_queues, &queues_priv_data_size); + if (ret) + return ret; + + num_events = kfd_get_num_events(p); + + ret = svm_range_get_info(p, &num_svm_ranges, &svm_priv_data_size); + if (ret) + return ret; + + *num_objects = num_queues + num_events + num_svm_ranges; + + if (objs_priv_size) { + priv_size = sizeof(struct kfd_criu_process_priv_data); + priv_size += *num_devices * sizeof(struct kfd_criu_device_priv_data); + priv_size += *num_bos * sizeof(struct kfd_criu_bo_priv_data); + priv_size += queues_priv_data_size; + priv_size += num_events * sizeof(struct kfd_criu_event_priv_data); + priv_size += svm_priv_data_size; + *objs_priv_size = priv_size; + } + return 0; +} + +static int criu_checkpoint(struct file *filep, + struct kfd_process *p, + struct kfd_ioctl_criu_args *args) +{ + int ret; + uint32_t num_devices, num_bos, num_objects; + uint64_t priv_size, priv_offset = 0; + + if (!args->devices || !args->bos || !args->priv_data) + return -EINVAL; + + mutex_lock(&p->mutex); + + if (!p->n_pdds) { + pr_err("No pdd for given process\n"); + ret = -ENODEV; + goto exit_unlock; + } + + /* Confirm all process queues are evicted */ + if (!p->queues_paused) { + pr_err("Cannot dump process when queues are not in evicted state\n"); + /* CRIU plugin did not call op PROCESS_INFO before checkpointing */ + ret = -EINVAL; + goto exit_unlock; + } + + ret = criu_get_process_object_info(p, &num_devices, &num_bos, &num_objects, &priv_size); + if (ret) + goto exit_unlock; + + if (num_devices != args->num_devices || + num_bos != args->num_bos || + num_objects != args->num_objects || + priv_size != args->priv_data_size) { + + ret = -EINVAL; + goto exit_unlock; + } + + /* each function will store private data inside priv_data and adjust priv_offset */ + ret = criu_checkpoint_process(p, (uint8_t __user *)args->priv_data, &priv_offset); + if (ret) + goto exit_unlock; + + ret = criu_checkpoint_devices(p, num_devices, (uint8_t __user *)args->devices, + (uint8_t __user *)args->priv_data, &priv_offset); + if (ret) + goto exit_unlock; + + ret = criu_checkpoint_bos(p, num_bos, (uint8_t __user *)args->bos, + (uint8_t __user *)args->priv_data, &priv_offset); + if (ret) + goto exit_unlock; + + if (num_objects) { + ret = kfd_criu_checkpoint_queues(p, (uint8_t __user *)args->priv_data, + &priv_offset); + if (ret) + goto close_bo_fds; + + ret = kfd_criu_checkpoint_events(p, (uint8_t __user *)args->priv_data, + &priv_offset); + if (ret) + goto close_bo_fds; + + ret = kfd_criu_checkpoint_svm(p, (uint8_t __user *)args->priv_data, &priv_offset); + if (ret) + goto close_bo_fds; + } + +close_bo_fds: + if (ret) { + /* If IOCTL returns err, user assumes all FDs opened in criu_dump_bos are closed */ + uint32_t i; + struct kfd_criu_bo_bucket *bo_buckets = (struct kfd_criu_bo_bucket *) args->bos; + + for (i = 0; i < num_bos; i++) { + if (bo_buckets[i].alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) + close_fd(bo_buckets[i].dmabuf_fd); + } + } + +exit_unlock: + mutex_unlock(&p->mutex); + if (ret) + pr_err("Failed to dump CRIU ret:%d\n", ret); + else + pr_debug("CRIU dump ret:%d\n", ret); + + return ret; +} + +static int criu_restore_process(struct kfd_process *p, + struct kfd_ioctl_criu_args *args, + uint64_t *priv_offset, + uint64_t max_priv_data_size) +{ + int ret = 0; + struct kfd_criu_process_priv_data process_priv; + + if (*priv_offset + sizeof(process_priv) > max_priv_data_size) + return -EINVAL; + + ret = copy_from_user(&process_priv, + (void __user *)(args->priv_data + *priv_offset), + sizeof(process_priv)); + if (ret) { + pr_err("Failed to copy process private information from user\n"); + ret = -EFAULT; + goto exit; + } + *priv_offset += sizeof(process_priv); + + if (process_priv.version != KFD_CRIU_PRIV_VERSION) { + pr_err("Invalid CRIU API version (checkpointed:%d current:%d)\n", + process_priv.version, KFD_CRIU_PRIV_VERSION); + return -EINVAL; + } + + pr_debug("Setting XNACK mode\n"); + if (process_priv.xnack_mode && !kfd_process_xnack_mode(p, true)) { + pr_err("xnack mode cannot be set\n"); + ret = -EPERM; + goto exit; + } else { + pr_debug("set xnack mode: %d\n", process_priv.xnack_mode); + p->xnack_enabled = process_priv.xnack_mode; + } + +exit: + return ret; +} + +static int criu_restore_devices(struct kfd_process *p, + struct kfd_ioctl_criu_args *args, + uint64_t *priv_offset, + uint64_t max_priv_data_size) +{ + struct kfd_criu_device_bucket *device_buckets; + struct kfd_criu_device_priv_data *device_privs; + int ret = 0; + uint32_t i; + + if (args->num_devices != p->n_pdds) + return -EINVAL; + + if (*priv_offset + (args->num_devices * sizeof(*device_privs)) > max_priv_data_size) + return -EINVAL; + + device_buckets = kmalloc_array(args->num_devices, sizeof(*device_buckets), GFP_KERNEL); + if (!device_buckets) + return -ENOMEM; + + ret = copy_from_user(device_buckets, (void __user *)args->devices, + args->num_devices * sizeof(*device_buckets)); + if (ret) { + pr_err("Failed to copy devices buckets from user\n"); + ret = -EFAULT; + goto exit; + } + + for (i = 0; i < args->num_devices; i++) { + struct kfd_dev *dev; + struct kfd_process_device *pdd; + struct file *drm_file; + + /* device private data is not currently used */ + + if (!device_buckets[i].user_gpu_id) { + pr_err("Invalid user gpu_id\n"); + ret = -EINVAL; + goto exit; + } + + dev = kfd_device_by_id(device_buckets[i].actual_gpu_id); + if (!dev) { + pr_err("Failed to find device with gpu_id = %x\n", + device_buckets[i].actual_gpu_id); + ret = -EINVAL; + goto exit; + } + + pdd = kfd_get_process_device_data(dev, p); + if (!pdd) { + pr_err("Failed to get pdd for gpu_id = %x\n", + device_buckets[i].actual_gpu_id); + ret = -EINVAL; + goto exit; + } + pdd->user_gpu_id = device_buckets[i].user_gpu_id; + + drm_file = fget(device_buckets[i].drm_fd); + if (!drm_file) { + pr_err("Invalid render node file descriptor sent from plugin (%d)\n", + device_buckets[i].drm_fd); + ret = -EINVAL; + goto exit; + } + + if (pdd->drm_file) { + ret = -EINVAL; + goto exit; + } + + /* create the vm using render nodes for kfd pdd */ + if (kfd_process_device_init_vm(pdd, drm_file)) { + pr_err("could not init vm for given pdd\n"); + /* On success, the PDD keeps the drm_file reference */ + fput(drm_file); + ret = -EINVAL; + goto exit; + } + /* + * pdd now already has the vm bound to render node so below api won't create a new + * exclusive kfd mapping but use existing one with renderDXXX but is still needed + * for iommu v2 binding and runtime pm. + */ + pdd = kfd_bind_process_to_device(dev, p); + if (IS_ERR(pdd)) { + ret = PTR_ERR(pdd); + goto exit; + } + } + + /* + * We are not copying device private data from user as we are not using the data for now, + * but we still adjust for its private data. + */ + *priv_offset += args->num_devices * sizeof(*device_privs); + +exit: + kfree(device_buckets); + return ret; +} + +static int criu_restore_bos(struct kfd_process *p, + struct kfd_ioctl_criu_args *args, + uint64_t *priv_offset, + uint64_t max_priv_data_size) +{ + struct kfd_criu_bo_bucket *bo_buckets; + struct kfd_criu_bo_priv_data *bo_privs; + const bool criu_resume = true; + bool flush_tlbs = false; + int ret = 0, j = 0; + uint32_t i = 0; + + if (*priv_offset + (args->num_bos * sizeof(*bo_privs)) > max_priv_data_size) + return -EINVAL; + + /* Prevent MMU notifications until stage-4 IOCTL (CRIU_RESUME) is received */ + amdgpu_amdkfd_block_mmu_notifications(p->kgd_process_info); + + bo_buckets = kvmalloc_array(args->num_bos, sizeof(*bo_buckets), GFP_KERNEL); + if (!bo_buckets) + return -ENOMEM; + + ret = copy_from_user(bo_buckets, (void __user *)args->bos, + args->num_bos * sizeof(*bo_buckets)); + if (ret) { + pr_err("Failed to copy BOs information from user\n"); + ret = -EFAULT; + goto exit; + } + + bo_privs = kvmalloc_array(args->num_bos, sizeof(*bo_privs), GFP_KERNEL); + if (!bo_privs) { + ret = -ENOMEM; + goto exit; + } + + ret = copy_from_user(bo_privs, (void __user *)args->priv_data + *priv_offset, + args->num_bos * sizeof(*bo_privs)); + if (ret) { + pr_err("Failed to copy BOs information from user\n"); + ret = -EFAULT; + goto exit; + } + *priv_offset += args->num_bos * sizeof(*bo_privs); + + /* Create and map new BOs */ + for (; i < args->num_bos; i++) { + struct kfd_criu_bo_bucket *bo_bucket; + struct kfd_criu_bo_priv_data *bo_priv; + struct kfd_dev *dev; + struct kfd_process_device *pdd; + struct kgd_mem *kgd_mem; + void *mem; + u64 offset; + int idr_handle; + + bo_bucket = &bo_buckets[i]; + bo_priv = &bo_privs[i]; + + pr_debug("kfd restore ioctl - bo_bucket[%d]:\n", i); + pr_debug("size = 0x%llx, bo_addr = 0x%llx bo_offset = 0x%llx\n" + "gpu_id = 0x%x alloc_flags = 0x%x\n" + "idr_handle = 0x%x\n", + bo_bucket->size, + bo_bucket->addr, + bo_bucket->offset, + bo_bucket->gpu_id, + bo_bucket->alloc_flags, + bo_priv->idr_handle); + + pdd = kfd_process_device_data_by_id(p, bo_bucket->gpu_id); + if (!pdd) { + pr_err("Failed to get pdd\n"); + ret = -ENODEV; + goto exit; + } + dev = pdd->dev; + + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) { + pr_debug("restore ioctl: KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL\n"); + if (bo_bucket->size != kfd_doorbell_process_slice(dev)) { + ret = -EINVAL; + goto exit; + } + offset = kfd_get_process_doorbells(pdd); + } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) { + /* MMIO BOs need remapped bus address */ + pr_debug("restore ioctl :KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP\n"); + if (bo_bucket->size != PAGE_SIZE) { + pr_err("Invalid page size\n"); + ret = -EINVAL; + goto exit; + } + offset = dev->adev->rmmio_remap.bus_addr; + if (!offset) { + pr_err("amdgpu_amdkfd_get_mmio_remap_phys_addr failed\n"); + ret = -ENOMEM; + goto exit; + } + } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { + offset = bo_priv->user_addr; + } + /* Create the BO */ + ret = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(dev->adev, + bo_bucket->addr, + bo_bucket->size, + pdd->drm_priv, + (struct kgd_mem **) &mem, + &offset, + bo_bucket->alloc_flags, + criu_resume); + if (ret) { + pr_err("Could not create the BO\n"); + ret = -ENOMEM; + goto exit; + } + pr_debug("New BO created: size = 0x%llx, bo_addr = 0x%llx bo_offset = 0x%llx\n", + bo_bucket->size, bo_bucket->addr, offset); + + /* Restore previuos IDR handle */ + pr_debug("Restoring old IDR handle for the BO"); + idr_handle = idr_alloc(&pdd->alloc_idr, mem, + bo_priv->idr_handle, + bo_priv->idr_handle + 1, GFP_KERNEL); + + if (idr_handle < 0) { + pr_err("Could not allocate idr\n"); + amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->adev, + (struct kgd_mem *)mem, + pdd->drm_priv, NULL); + ret = -ENOMEM; + goto exit; + } + + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) + bo_bucket->restored_offset = KFD_MMAP_TYPE_DOORBELL | + KFD_MMAP_GPU_ID(pdd->dev->id); + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP) { + bo_bucket->restored_offset = KFD_MMAP_TYPE_MMIO | + KFD_MMAP_GPU_ID(pdd->dev->id); + } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) { + bo_bucket->restored_offset = offset; + pr_debug("updating offset for GTT\n"); + } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { + bo_bucket->restored_offset = offset; + /* Update the VRAM usage count */ + WRITE_ONCE(pdd->vram_usage, pdd->vram_usage + bo_bucket->size); + pr_debug("updating offset for VRAM\n"); + } + + /* now map these BOs to GPU/s */ + for (j = 0; j < p->n_pdds; j++) { + struct kfd_dev *peer; + struct kfd_process_device *peer_pdd; + bool table_freed = false; + + if (!bo_priv->mapped_gpuids[j]) + break; + + peer_pdd = kfd_process_device_data_by_id(p, bo_priv->mapped_gpuids[j]); + if (!peer_pdd) { + ret = -EINVAL; + goto exit; + } + peer = peer_pdd->dev; + + peer_pdd = kfd_bind_process_to_device(peer, p); + if (IS_ERR(peer_pdd)) { + ret = PTR_ERR(peer_pdd); + goto exit; + } + pr_debug("map mem in restore ioctl -> 0x%llx\n", + ((struct kgd_mem *)mem)->va); + ret = amdgpu_amdkfd_gpuvm_map_memory_to_gpu(peer->adev, + (struct kgd_mem *)mem, peer_pdd->drm_priv, &table_freed); + if (ret) { + pr_err("Failed to map to gpu %d/%d\n", j, p->n_pdds); + goto exit; + } + if (table_freed) + flush_tlbs = true; + } + + ret = amdgpu_amdkfd_gpuvm_sync_memory(dev->adev, + (struct kgd_mem *) mem, true); + if (ret) { + pr_debug("Sync memory failed, wait interrupted by user signal\n"); + goto exit; + } + + pr_debug("map memory was successful for the BO\n"); + /* create the dmabuf object and export the bo */ + kgd_mem = (struct kgd_mem *)mem; + if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { + ret = criu_get_prime_handle(&kgd_mem->bo->tbo.base, + DRM_RDWR, + &bo_bucket->dmabuf_fd); + if (ret) + goto exit; + } + } /* done */ + + if (flush_tlbs) { + /* Flush TLBs after waiting for the page table updates to complete */ + for (j = 0; j < p->n_pdds; j++) { + struct kfd_dev *peer; + struct kfd_process_device *pdd = p->pdds[j]; + struct kfd_process_device *peer_pdd; + + peer = kfd_device_by_id(pdd->dev->id); + if (WARN_ON_ONCE(!peer)) + continue; + peer_pdd = kfd_get_process_device_data(peer, p); + if (WARN_ON_ONCE(!peer_pdd)) + continue; + kfd_flush_tlb(peer_pdd, TLB_FLUSH_LEGACY); + } + } + + /* Copy only the buckets back so user can read bo_buckets[N].restored_offset */ + ret = copy_to_user((void __user *)args->bos, + bo_buckets, + (args->num_bos * sizeof(*bo_buckets))); + if (ret) + ret = -EFAULT; + +exit: + while (ret && i--) { + if (bo_buckets[i].alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) + close_fd(bo_buckets[i].dmabuf_fd); + } + kvfree(bo_buckets); + kvfree(bo_privs); + return ret; +} + +static int criu_restore_objects(struct file *filep, + struct kfd_process *p, + struct kfd_ioctl_criu_args *args, + uint64_t *priv_offset, + uint64_t max_priv_data_size) +{ + int ret = 0; + uint32_t i; + + BUILD_BUG_ON(offsetof(struct kfd_criu_queue_priv_data, object_type)); + BUILD_BUG_ON(offsetof(struct kfd_criu_event_priv_data, object_type)); + BUILD_BUG_ON(offsetof(struct kfd_criu_svm_range_priv_data, object_type)); + + for (i = 0; i < args->num_objects; i++) { + uint32_t object_type; + + if (*priv_offset + sizeof(object_type) > max_priv_data_size) { + pr_err("Invalid private data size\n"); + return -EINVAL; + } + + ret = get_user(object_type, (uint32_t __user *)(args->priv_data + *priv_offset)); + if (ret) { + pr_err("Failed to copy private information from user\n"); + goto exit; + } + + switch (object_type) { + case KFD_CRIU_OBJECT_TYPE_QUEUE: + ret = kfd_criu_restore_queue(p, (uint8_t __user *)args->priv_data, + priv_offset, max_priv_data_size); + if (ret) + goto exit; + break; + case KFD_CRIU_OBJECT_TYPE_EVENT: + ret = kfd_criu_restore_event(filep, p, (uint8_t __user *)args->priv_data, + priv_offset, max_priv_data_size); + if (ret) + goto exit; + break; + case KFD_CRIU_OBJECT_TYPE_SVM_RANGE: + ret = kfd_criu_restore_svm(p, (uint8_t __user *)args->priv_data, + priv_offset, max_priv_data_size); + if (ret) + goto exit; + break; + default: + pr_err("Invalid object type:%u at index:%d\n", object_type, i); + ret = -EINVAL; + goto exit; + } + } +exit: + return ret; +} + +static int criu_restore(struct file *filep, + struct kfd_process *p, + struct kfd_ioctl_criu_args *args) +{ + uint64_t priv_offset = 0; + int ret = 0; + + pr_debug("CRIU restore (num_devices:%u num_bos:%u num_objects:%u priv_data_size:%llu)\n", + args->num_devices, args->num_bos, args->num_objects, args->priv_data_size); + + if (!args->bos || !args->devices || !args->priv_data || !args->priv_data_size || + !args->num_devices || !args->num_bos) + return -EINVAL; + + mutex_lock(&p->mutex); + + /* + * Set the process to evicted state to avoid running any new queues before all the memory + * mappings are ready. + */ + ret = kfd_process_evict_queues(p); + if (ret) + goto exit_unlock; + + /* Each function will adjust priv_offset based on how many bytes they consumed */ + ret = criu_restore_process(p, args, &priv_offset, args->priv_data_size); + if (ret) + goto exit_unlock; + + ret = criu_restore_devices(p, args, &priv_offset, args->priv_data_size); + if (ret) + goto exit_unlock; + + ret = criu_restore_bos(p, args, &priv_offset, args->priv_data_size); + if (ret) + goto exit_unlock; + + ret = criu_restore_objects(filep, p, args, &priv_offset, args->priv_data_size); + if (ret) + goto exit_unlock; + + if (priv_offset != args->priv_data_size) { + pr_err("Invalid private data size\n"); + ret = -EINVAL; + } + +exit_unlock: + mutex_unlock(&p->mutex); + if (ret) + pr_err("Failed to restore CRIU ret:%d\n", ret); + else + pr_debug("CRIU restore successful\n"); + + return ret; +} + +static int criu_unpause(struct file *filep, + struct kfd_process *p, + struct kfd_ioctl_criu_args *args) +{ + int ret; + + mutex_lock(&p->mutex); + + if (!p->queues_paused) { + mutex_unlock(&p->mutex); + return -EINVAL; + } + + ret = kfd_process_restore_queues(p); + if (ret) + pr_err("Failed to unpause queues ret:%d\n", ret); + else + p->queues_paused = false; + + mutex_unlock(&p->mutex); + + return ret; +} + +static int criu_resume(struct file *filep, + struct kfd_process *p, + struct kfd_ioctl_criu_args *args) +{ + struct kfd_process *target = NULL; + struct pid *pid = NULL; + int ret = 0; + + pr_debug("Inside %s, target pid for criu restore: %d\n", __func__, + args->pid); + + pid = find_get_pid(args->pid); + if (!pid) { + pr_err("Cannot find pid info for %i\n", args->pid); + return -ESRCH; + } + + pr_debug("calling kfd_lookup_process_by_pid\n"); + target = kfd_lookup_process_by_pid(pid); + + put_pid(pid); + + if (!target) { + pr_debug("Cannot find process info for %i\n", args->pid); + return -ESRCH; + } + + mutex_lock(&target->mutex); + ret = kfd_criu_resume_svm(target); + if (ret) { + pr_err("kfd_criu_resume_svm failed for %i\n", args->pid); + goto exit; + } + + ret = amdgpu_amdkfd_criu_resume(target->kgd_process_info); + if (ret) + pr_err("amdgpu_amdkfd_criu_resume failed for %i\n", args->pid); + +exit: + mutex_unlock(&target->mutex); + + kfd_unref_process(target); + return ret; +} + +static int criu_process_info(struct file *filep, + struct kfd_process *p, + struct kfd_ioctl_criu_args *args) +{ + int ret = 0; + + mutex_lock(&p->mutex); + + if (!p->n_pdds) { + pr_err("No pdd for given process\n"); + ret = -ENODEV; + goto err_unlock; + } + + ret = kfd_process_evict_queues(p); + if (ret) + goto err_unlock; + + p->queues_paused = true; + + args->pid = task_pid_nr_ns(p->lead_thread, + task_active_pid_ns(p->lead_thread)); + + ret = criu_get_process_object_info(p, &args->num_devices, &args->num_bos, + &args->num_objects, &args->priv_data_size); + if (ret) + goto err_unlock; + + dev_dbg(kfd_device, "Num of devices:%u bos:%u objects:%u priv_data_size:%lld\n", + args->num_devices, args->num_bos, args->num_objects, + args->priv_data_size); + +err_unlock: + if (ret) { + kfd_process_restore_queues(p); + p->queues_paused = false; + } + mutex_unlock(&p->mutex); + return ret; +} + +static int kfd_ioctl_criu(struct file *filep, struct kfd_process *p, void *data) +{ + struct kfd_ioctl_criu_args *args = data; + int ret; + + dev_dbg(kfd_device, "CRIU operation: %d\n", args->op); + switch (args->op) { + case KFD_CRIU_OP_PROCESS_INFO: + ret = criu_process_info(filep, p, args); + break; + case KFD_CRIU_OP_CHECKPOINT: + ret = criu_checkpoint(filep, p, args); + break; + case KFD_CRIU_OP_UNPAUSE: + ret = criu_unpause(filep, p, args); + break; + case KFD_CRIU_OP_RESTORE: + ret = criu_restore(filep, p, args); + break; + case KFD_CRIU_OP_RESUME: + ret = criu_resume(filep, p, args); + break; + default: + dev_dbg(kfd_device, "Unsupported CRIU operation:%d\n", args->op); + ret = -EINVAL; + break; + } + + if (ret) + dev_dbg(kfd_device, "CRIU operation:%d err:%d\n", args->op, ret); + + return ret; +} + #define AMDKFD_IOCTL_DEF(ioctl, _func, _flags) \ [_IOC_NR(ioctl)] = {.cmd = ioctl, .func = _func, .flags = _flags, \ .cmd_drv = 0, .name = #ioctl} @@ -1898,16 +2624,16 @@ static const struct amdkfd_ioctl_desc amdkfd_ioctls[] = { AMDKFD_IOCTL_DEF(AMDKFD_IOC_WAIT_EVENTS, kfd_ioctl_wait_events, 0), - AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_REGISTER, + AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_REGISTER_DEPRECATED, kfd_ioctl_dbg_register, 0), - AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_UNREGISTER, + AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_UNREGISTER_DEPRECATED, kfd_ioctl_dbg_unregister, 0), - AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_ADDRESS_WATCH, + AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_ADDRESS_WATCH_DEPRECATED, kfd_ioctl_dbg_address_watch, 0), - AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_WAVE_CONTROL, + AMDKFD_IOCTL_DEF(AMDKFD_IOC_DBG_WAVE_CONTROL_DEPRECATED, kfd_ioctl_dbg_wave_control, 0), AMDKFD_IOCTL_DEF(AMDKFD_IOC_SET_SCRATCH_BACKING_VA, @@ -1959,6 +2685,10 @@ static const struct amdkfd_ioctl_desc amdkfd_ioctls[] = { AMDKFD_IOCTL_DEF(AMDKFD_IOC_SET_XNACK_MODE, kfd_ioctl_set_xnack_mode, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_CRIU_OP, + kfd_ioctl_criu, KFD_IOC_FLAG_CHECKPOINT_RESTORE), + }; #define AMDKFD_CORE_IOCTL_COUNT ARRAY_SIZE(amdkfd_ioctls) @@ -1973,6 +2703,7 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg) char *kdata = NULL; unsigned int usize, asize; int retcode = -EINVAL; + bool ptrace_attached = false; if (nr >= AMDKFD_CORE_IOCTL_COUNT) goto err_i1; @@ -1998,7 +2729,15 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg) * processes need to create their own KFD device context. */ process = filep->private_data; - if (process->lead_thread != current->group_leader) { + + rcu_read_lock(); + if ((ioctl->flags & KFD_IOC_FLAG_CHECKPOINT_RESTORE) && + ptrace_parent(process->lead_thread) == current) + ptrace_attached = true; + rcu_read_unlock(); + + if (process->lead_thread != current->group_leader + && !ptrace_attached) { dev_dbg(kfd_device, "Using KFD FD in wrong process\n"); retcode = -EBADF; goto err_i1; @@ -2013,6 +2752,19 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg) goto err_i1; } + /* + * Versions of docker shipped in Ubuntu 18.xx and 20.xx do not support + * CAP_CHECKPOINT_RESTORE, so we also allow access if CAP_SYS_ADMIN as CAP_SYS_ADMIN is a + * more priviledged access. + */ + if (unlikely(ioctl->flags & KFD_IOC_FLAG_CHECKPOINT_RESTORE)) { + if (!capable(CAP_CHECKPOINT_RESTORE) && + !capable(CAP_SYS_ADMIN)) { + retcode = -EACCES; + goto err_i1; + } + } + if (cmd & (IOC_IN | IOC_OUT)) { if (asize <= sizeof(stack_kdata)) { kdata = stack_kdata; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index 9624bbe8b501..bb6e49661d13 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -1411,6 +1411,7 @@ static int kfd_fill_gpu_cache_info(struct kfd_dev *kdev, case IP_VERSION(10, 1, 10): case IP_VERSION(10, 1, 2): case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): pcache_info = navi10_cache_info; num_of_cache_types = ARRAY_SIZE(navi10_cache_info); break; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c deleted file mode 100644 index 1e30717b5253..000000000000 --- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c +++ /dev/null @@ -1,845 +0,0 @@ -/* - * Copyright 2014 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - */ - -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/log2.h> -#include <linux/sched.h> -#include <linux/slab.h> -#include <linux/mutex.h> -#include <linux/device.h> - -#include "kfd_pm4_headers.h" -#include "kfd_pm4_headers_diq.h" -#include "kfd_kernel_queue.h" -#include "kfd_priv.h" -#include "kfd_pm4_opcodes.h" -#include "cik_regs.h" -#include "kfd_dbgmgr.h" -#include "kfd_dbgdev.h" -#include "kfd_device_queue_manager.h" - -static void dbgdev_address_watch_disable_nodiq(struct kfd_dev *dev) -{ - dev->kfd2kgd->address_watch_disable(dev->adev); -} - -static int dbgdev_diq_submit_ib(struct kfd_dbgdev *dbgdev, - u32 pasid, uint64_t vmid0_address, - uint32_t *packet_buff, size_t size_in_bytes) -{ - struct pm4__release_mem *rm_packet; - struct pm4__indirect_buffer_pasid *ib_packet; - struct kfd_mem_obj *mem_obj; - size_t pq_packets_size_in_bytes; - union ULARGE_INTEGER *largep; - union ULARGE_INTEGER addr; - struct kernel_queue *kq; - uint64_t *rm_state; - unsigned int *ib_packet_buff; - int status; - - if (WARN_ON(!size_in_bytes)) - return -EINVAL; - - kq = dbgdev->kq; - - pq_packets_size_in_bytes = sizeof(struct pm4__release_mem) + - sizeof(struct pm4__indirect_buffer_pasid); - - /* - * We acquire a buffer from DIQ - * The receive packet buff will be sitting on the Indirect Buffer - * and in the PQ we put the IB packet + sync packet(s). - */ - status = kq_acquire_packet_buffer(kq, - pq_packets_size_in_bytes / sizeof(uint32_t), - &ib_packet_buff); - if (status) { - pr_err("kq_acquire_packet_buffer failed\n"); - return status; - } - - memset(ib_packet_buff, 0, pq_packets_size_in_bytes); - - ib_packet = (struct pm4__indirect_buffer_pasid *) (ib_packet_buff); - - ib_packet->header.count = 3; - ib_packet->header.opcode = IT_INDIRECT_BUFFER_PASID; - ib_packet->header.type = PM4_TYPE_3; - - largep = (union ULARGE_INTEGER *) &vmid0_address; - - ib_packet->bitfields2.ib_base_lo = largep->u.low_part >> 2; - ib_packet->bitfields3.ib_base_hi = largep->u.high_part; - - ib_packet->control = (1 << 23) | (1 << 31) | - ((size_in_bytes / 4) & 0xfffff); - - ib_packet->bitfields5.pasid = pasid; - - /* - * for now we use release mem for GPU-CPU synchronization - * Consider WaitRegMem + WriteData as a better alternative - * we get a GART allocations ( gpu/cpu mapping), - * for the sync variable, and wait until: - * (a) Sync with HW - * (b) Sync var is written by CP to mem. - */ - rm_packet = (struct pm4__release_mem *) (ib_packet_buff + - (sizeof(struct pm4__indirect_buffer_pasid) / - sizeof(unsigned int))); - - status = kfd_gtt_sa_allocate(dbgdev->dev, sizeof(uint64_t), - &mem_obj); - - if (status) { - pr_err("Failed to allocate GART memory\n"); - kq_rollback_packet(kq); - return status; - } - - rm_state = (uint64_t *) mem_obj->cpu_ptr; - - *rm_state = QUEUESTATE__ACTIVE_COMPLETION_PENDING; - - rm_packet->header.opcode = IT_RELEASE_MEM; - rm_packet->header.type = PM4_TYPE_3; - rm_packet->header.count = sizeof(struct pm4__release_mem) / 4 - 2; - - rm_packet->bitfields2.event_type = CACHE_FLUSH_AND_INV_TS_EVENT; - rm_packet->bitfields2.event_index = - event_index___release_mem__end_of_pipe; - - rm_packet->bitfields2.cache_policy = cache_policy___release_mem__lru; - rm_packet->bitfields2.atc = 0; - rm_packet->bitfields2.tc_wb_action_ena = 1; - - addr.quad_part = mem_obj->gpu_addr; - - rm_packet->bitfields4.address_lo_32b = addr.u.low_part >> 2; - rm_packet->address_hi = addr.u.high_part; - - rm_packet->bitfields3.data_sel = - data_sel___release_mem__send_64_bit_data; - - rm_packet->bitfields3.int_sel = - int_sel___release_mem__send_data_after_write_confirm; - - rm_packet->bitfields3.dst_sel = - dst_sel___release_mem__memory_controller; - - rm_packet->data_lo = QUEUESTATE__ACTIVE; - - kq_submit_packet(kq); - - /* Wait till CP writes sync code: */ - status = amdkfd_fence_wait_timeout( - rm_state, - QUEUESTATE__ACTIVE, 1500); - - kfd_gtt_sa_free(dbgdev->dev, mem_obj); - - return status; -} - -static int dbgdev_register_nodiq(struct kfd_dbgdev *dbgdev) -{ - /* - * no action is needed in this case, - * just make sure diq will not be used - */ - - dbgdev->kq = NULL; - - return 0; -} - -static int dbgdev_register_diq(struct kfd_dbgdev *dbgdev) -{ - struct queue_properties properties; - unsigned int qid; - struct kernel_queue *kq = NULL; - int status; - - properties.type = KFD_QUEUE_TYPE_DIQ; - - status = pqm_create_queue(dbgdev->pqm, dbgdev->dev, NULL, - &properties, &qid, NULL); - - if (status) { - pr_err("Failed to create DIQ\n"); - return status; - } - - pr_debug("DIQ Created with queue id: %d\n", qid); - - kq = pqm_get_kernel_queue(dbgdev->pqm, qid); - - if (!kq) { - pr_err("Error getting DIQ\n"); - pqm_destroy_queue(dbgdev->pqm, qid); - return -EFAULT; - } - - dbgdev->kq = kq; - - return status; -} - -static int dbgdev_unregister_nodiq(struct kfd_dbgdev *dbgdev) -{ - /* disable watch address */ - dbgdev_address_watch_disable_nodiq(dbgdev->dev); - return 0; -} - -static int dbgdev_unregister_diq(struct kfd_dbgdev *dbgdev) -{ - /* todo - disable address watch */ - int status; - - status = pqm_destroy_queue(dbgdev->pqm, - dbgdev->kq->queue->properties.queue_id); - dbgdev->kq = NULL; - - return status; -} - -static void dbgdev_address_watch_set_registers( - const struct dbg_address_watch_info *adw_info, - union TCP_WATCH_ADDR_H_BITS *addrHi, - union TCP_WATCH_ADDR_L_BITS *addrLo, - union TCP_WATCH_CNTL_BITS *cntl, - unsigned int index, unsigned int vmid) -{ - union ULARGE_INTEGER addr; - - addr.quad_part = 0; - addrHi->u32All = 0; - addrLo->u32All = 0; - cntl->u32All = 0; - - if (adw_info->watch_mask) - cntl->bitfields.mask = - (uint32_t) (adw_info->watch_mask[index] & - ADDRESS_WATCH_REG_CNTL_DEFAULT_MASK); - else - cntl->bitfields.mask = ADDRESS_WATCH_REG_CNTL_DEFAULT_MASK; - - addr.quad_part = (unsigned long long) adw_info->watch_address[index]; - - addrHi->bitfields.addr = addr.u.high_part & - ADDRESS_WATCH_REG_ADDHIGH_MASK; - addrLo->bitfields.addr = - (addr.u.low_part >> ADDRESS_WATCH_REG_ADDLOW_SHIFT); - - cntl->bitfields.mode = adw_info->watch_mode[index]; - cntl->bitfields.vmid = (uint32_t) vmid; - /* for now assume it is an ATC address */ - cntl->u32All |= ADDRESS_WATCH_REG_CNTL_ATC_BIT; - - pr_debug("\t\t%20s %08x\n", "set reg mask :", cntl->bitfields.mask); - pr_debug("\t\t%20s %08x\n", "set reg add high :", - addrHi->bitfields.addr); - pr_debug("\t\t%20s %08x\n", "set reg add low :", - addrLo->bitfields.addr); -} - -static int dbgdev_address_watch_nodiq(struct kfd_dbgdev *dbgdev, - struct dbg_address_watch_info *adw_info) -{ - union TCP_WATCH_ADDR_H_BITS addrHi; - union TCP_WATCH_ADDR_L_BITS addrLo; - union TCP_WATCH_CNTL_BITS cntl; - struct kfd_process_device *pdd; - unsigned int i; - - /* taking the vmid for that process on the safe way using pdd */ - pdd = kfd_get_process_device_data(dbgdev->dev, - adw_info->process); - if (!pdd) { - pr_err("Failed to get pdd for wave control no DIQ\n"); - return -EFAULT; - } - - addrHi.u32All = 0; - addrLo.u32All = 0; - cntl.u32All = 0; - - if ((adw_info->num_watch_points > MAX_WATCH_ADDRESSES) || - (adw_info->num_watch_points == 0)) { - pr_err("num_watch_points is invalid\n"); - return -EINVAL; - } - - if (!adw_info->watch_mode || !adw_info->watch_address) { - pr_err("adw_info fields are not valid\n"); - return -EINVAL; - } - - for (i = 0; i < adw_info->num_watch_points; i++) { - dbgdev_address_watch_set_registers(adw_info, &addrHi, &addrLo, - &cntl, i, pdd->qpd.vmid); - - pr_debug("\t\t%30s\n", "* * * * * * * * * * * * * * * * * *"); - pr_debug("\t\t%20s %08x\n", "register index :", i); - pr_debug("\t\t%20s %08x\n", "vmid is :", pdd->qpd.vmid); - pr_debug("\t\t%20s %08x\n", "Address Low is :", - addrLo.bitfields.addr); - pr_debug("\t\t%20s %08x\n", "Address high is :", - addrHi.bitfields.addr); - pr_debug("\t\t%20s %08x\n", "Address high is :", - addrHi.bitfields.addr); - pr_debug("\t\t%20s %08x\n", "Control Mask is :", - cntl.bitfields.mask); - pr_debug("\t\t%20s %08x\n", "Control Mode is :", - cntl.bitfields.mode); - pr_debug("\t\t%20s %08x\n", "Control Vmid is :", - cntl.bitfields.vmid); - pr_debug("\t\t%20s %08x\n", "Control atc is :", - cntl.bitfields.atc); - pr_debug("\t\t%30s\n", "* * * * * * * * * * * * * * * * * *"); - - pdd->dev->kfd2kgd->address_watch_execute( - dbgdev->dev->adev, - i, - cntl.u32All, - addrHi.u32All, - addrLo.u32All); - } - - return 0; -} - -static int dbgdev_address_watch_diq(struct kfd_dbgdev *dbgdev, - struct dbg_address_watch_info *adw_info) -{ - struct pm4__set_config_reg *packets_vec; - union TCP_WATCH_ADDR_H_BITS addrHi; - union TCP_WATCH_ADDR_L_BITS addrLo; - union TCP_WATCH_CNTL_BITS cntl; - struct kfd_mem_obj *mem_obj; - unsigned int aw_reg_add_dword; - uint32_t *packet_buff_uint; - unsigned int i; - int status; - size_t ib_size = sizeof(struct pm4__set_config_reg) * 4; - /* we do not control the vmid in DIQ mode, just a place holder */ - unsigned int vmid = 0; - - addrHi.u32All = 0; - addrLo.u32All = 0; - cntl.u32All = 0; - - if ((adw_info->num_watch_points > MAX_WATCH_ADDRESSES) || - (adw_info->num_watch_points == 0)) { - pr_err("num_watch_points is invalid\n"); - return -EINVAL; - } - - if (!adw_info->watch_mode || !adw_info->watch_address) { - pr_err("adw_info fields are not valid\n"); - return -EINVAL; - } - - status = kfd_gtt_sa_allocate(dbgdev->dev, ib_size, &mem_obj); - - if (status) { - pr_err("Failed to allocate GART memory\n"); - return status; - } - - packet_buff_uint = mem_obj->cpu_ptr; - - memset(packet_buff_uint, 0, ib_size); - - packets_vec = (struct pm4__set_config_reg *) (packet_buff_uint); - - packets_vec[0].header.count = 1; - packets_vec[0].header.opcode = IT_SET_CONFIG_REG; - packets_vec[0].header.type = PM4_TYPE_3; - packets_vec[0].bitfields2.vmid_shift = ADDRESS_WATCH_CNTL_OFFSET; - packets_vec[0].bitfields2.insert_vmid = 1; - packets_vec[1].ordinal1 = packets_vec[0].ordinal1; - packets_vec[1].bitfields2.insert_vmid = 0; - packets_vec[2].ordinal1 = packets_vec[0].ordinal1; - packets_vec[2].bitfields2.insert_vmid = 0; - packets_vec[3].ordinal1 = packets_vec[0].ordinal1; - packets_vec[3].bitfields2.vmid_shift = ADDRESS_WATCH_CNTL_OFFSET; - packets_vec[3].bitfields2.insert_vmid = 1; - - for (i = 0; i < adw_info->num_watch_points; i++) { - dbgdev_address_watch_set_registers(adw_info, - &addrHi, - &addrLo, - &cntl, - i, - vmid); - - pr_debug("\t\t%30s\n", "* * * * * * * * * * * * * * * * * *"); - pr_debug("\t\t%20s %08x\n", "register index :", i); - pr_debug("\t\t%20s %08x\n", "vmid is :", vmid); - pr_debug("\t\t%20s %p\n", "Add ptr is :", - adw_info->watch_address); - pr_debug("\t\t%20s %08llx\n", "Add is :", - adw_info->watch_address[i]); - pr_debug("\t\t%20s %08x\n", "Address Low is :", - addrLo.bitfields.addr); - pr_debug("\t\t%20s %08x\n", "Address high is :", - addrHi.bitfields.addr); - pr_debug("\t\t%20s %08x\n", "Control Mask is :", - cntl.bitfields.mask); - pr_debug("\t\t%20s %08x\n", "Control Mode is :", - cntl.bitfields.mode); - pr_debug("\t\t%20s %08x\n", "Control Vmid is :", - cntl.bitfields.vmid); - pr_debug("\t\t%20s %08x\n", "Control atc is :", - cntl.bitfields.atc); - pr_debug("\t\t%30s\n", "* * * * * * * * * * * * * * * * * *"); - - aw_reg_add_dword = - dbgdev->dev->kfd2kgd->address_watch_get_offset( - dbgdev->dev->adev, - i, - ADDRESS_WATCH_REG_CNTL); - - packets_vec[0].bitfields2.reg_offset = - aw_reg_add_dword - AMD_CONFIG_REG_BASE; - - packets_vec[0].reg_data[0] = cntl.u32All; - - aw_reg_add_dword = - dbgdev->dev->kfd2kgd->address_watch_get_offset( - dbgdev->dev->adev, - i, - ADDRESS_WATCH_REG_ADDR_HI); - - packets_vec[1].bitfields2.reg_offset = - aw_reg_add_dword - AMD_CONFIG_REG_BASE; - packets_vec[1].reg_data[0] = addrHi.u32All; - - aw_reg_add_dword = - dbgdev->dev->kfd2kgd->address_watch_get_offset( - dbgdev->dev->adev, - i, - ADDRESS_WATCH_REG_ADDR_LO); - - packets_vec[2].bitfields2.reg_offset = - aw_reg_add_dword - AMD_CONFIG_REG_BASE; - packets_vec[2].reg_data[0] = addrLo.u32All; - - /* enable watch flag if address is not zero*/ - if (adw_info->watch_address[i] > 0) - cntl.bitfields.valid = 1; - else - cntl.bitfields.valid = 0; - - aw_reg_add_dword = - dbgdev->dev->kfd2kgd->address_watch_get_offset( - dbgdev->dev->adev, - i, - ADDRESS_WATCH_REG_CNTL); - - packets_vec[3].bitfields2.reg_offset = - aw_reg_add_dword - AMD_CONFIG_REG_BASE; - packets_vec[3].reg_data[0] = cntl.u32All; - - status = dbgdev_diq_submit_ib( - dbgdev, - adw_info->process->pasid, - mem_obj->gpu_addr, - packet_buff_uint, - ib_size); - - if (status) { - pr_err("Failed to submit IB to DIQ\n"); - break; - } - } - - kfd_gtt_sa_free(dbgdev->dev, mem_obj); - return status; -} - -static int dbgdev_wave_control_set_registers( - struct dbg_wave_control_info *wac_info, - union SQ_CMD_BITS *in_reg_sq_cmd, - union GRBM_GFX_INDEX_BITS *in_reg_gfx_index) -{ - int status = 0; - union SQ_CMD_BITS reg_sq_cmd; - union GRBM_GFX_INDEX_BITS reg_gfx_index; - struct HsaDbgWaveMsgAMDGen2 *pMsg; - - reg_sq_cmd.u32All = 0; - reg_gfx_index.u32All = 0; - pMsg = &wac_info->dbgWave_msg.DbgWaveMsg.WaveMsgInfoGen2; - - switch (wac_info->mode) { - /* Send command to single wave */ - case HSA_DBG_WAVEMODE_SINGLE: - /* - * Limit access to the process waves only, - * by setting vmid check - */ - reg_sq_cmd.bits.check_vmid = 1; - reg_sq_cmd.bits.simd_id = pMsg->ui32.SIMD; - reg_sq_cmd.bits.wave_id = pMsg->ui32.WaveId; - reg_sq_cmd.bits.mode = SQ_IND_CMD_MODE_SINGLE; - - reg_gfx_index.bits.sh_index = pMsg->ui32.ShaderArray; - reg_gfx_index.bits.se_index = pMsg->ui32.ShaderEngine; - reg_gfx_index.bits.instance_index = pMsg->ui32.HSACU; - - break; - - /* Send command to all waves with matching VMID */ - case HSA_DBG_WAVEMODE_BROADCAST_PROCESS: - - reg_gfx_index.bits.sh_broadcast_writes = 1; - reg_gfx_index.bits.se_broadcast_writes = 1; - reg_gfx_index.bits.instance_broadcast_writes = 1; - - reg_sq_cmd.bits.mode = SQ_IND_CMD_MODE_BROADCAST; - - break; - - /* Send command to all CU waves with matching VMID */ - case HSA_DBG_WAVEMODE_BROADCAST_PROCESS_CU: - - reg_sq_cmd.bits.check_vmid = 1; - reg_sq_cmd.bits.mode = SQ_IND_CMD_MODE_BROADCAST; - - reg_gfx_index.bits.sh_index = pMsg->ui32.ShaderArray; - reg_gfx_index.bits.se_index = pMsg->ui32.ShaderEngine; - reg_gfx_index.bits.instance_index = pMsg->ui32.HSACU; - - break; - - default: - return -EINVAL; - } - - switch (wac_info->operand) { - case HSA_DBG_WAVEOP_HALT: - reg_sq_cmd.bits.cmd = SQ_IND_CMD_CMD_HALT; - break; - - case HSA_DBG_WAVEOP_RESUME: - reg_sq_cmd.bits.cmd = SQ_IND_CMD_CMD_RESUME; - break; - - case HSA_DBG_WAVEOP_KILL: - reg_sq_cmd.bits.cmd = SQ_IND_CMD_CMD_KILL; - break; - - case HSA_DBG_WAVEOP_DEBUG: - reg_sq_cmd.bits.cmd = SQ_IND_CMD_CMD_DEBUG; - break; - - case HSA_DBG_WAVEOP_TRAP: - if (wac_info->trapId < MAX_TRAPID) { - reg_sq_cmd.bits.cmd = SQ_IND_CMD_CMD_TRAP; - reg_sq_cmd.bits.trap_id = wac_info->trapId; - } else { - status = -EINVAL; - } - break; - - default: - status = -EINVAL; - break; - } - - if (status == 0) { - *in_reg_sq_cmd = reg_sq_cmd; - *in_reg_gfx_index = reg_gfx_index; - } - - return status; -} - -static int dbgdev_wave_control_diq(struct kfd_dbgdev *dbgdev, - struct dbg_wave_control_info *wac_info) -{ - - int status; - union SQ_CMD_BITS reg_sq_cmd; - union GRBM_GFX_INDEX_BITS reg_gfx_index; - struct kfd_mem_obj *mem_obj; - uint32_t *packet_buff_uint; - struct pm4__set_config_reg *packets_vec; - size_t ib_size = sizeof(struct pm4__set_config_reg) * 3; - - reg_sq_cmd.u32All = 0; - - status = dbgdev_wave_control_set_registers(wac_info, ®_sq_cmd, - ®_gfx_index); - if (status) { - pr_err("Failed to set wave control registers\n"); - return status; - } - - /* we do not control the VMID in DIQ, so reset it to a known value */ - reg_sq_cmd.bits.vm_id = 0; - - pr_debug("\t\t %30s\n", "* * * * * * * * * * * * * * * * * *"); - - pr_debug("\t\t mode is: %u\n", wac_info->mode); - pr_debug("\t\t operand is: %u\n", wac_info->operand); - pr_debug("\t\t trap id is: %u\n", wac_info->trapId); - pr_debug("\t\t msg value is: %u\n", - wac_info->dbgWave_msg.DbgWaveMsg.WaveMsgInfoGen2.Value); - pr_debug("\t\t vmid is: N/A\n"); - - pr_debug("\t\t chk_vmid is : %u\n", reg_sq_cmd.bitfields.check_vmid); - pr_debug("\t\t command is : %u\n", reg_sq_cmd.bitfields.cmd); - pr_debug("\t\t queue id is : %u\n", reg_sq_cmd.bitfields.queue_id); - pr_debug("\t\t simd id is : %u\n", reg_sq_cmd.bitfields.simd_id); - pr_debug("\t\t mode is : %u\n", reg_sq_cmd.bitfields.mode); - pr_debug("\t\t vm_id is : %u\n", reg_sq_cmd.bitfields.vm_id); - pr_debug("\t\t wave_id is : %u\n", reg_sq_cmd.bitfields.wave_id); - - pr_debug("\t\t ibw is : %u\n", - reg_gfx_index.bitfields.instance_broadcast_writes); - pr_debug("\t\t ii is : %u\n", - reg_gfx_index.bitfields.instance_index); - pr_debug("\t\t sebw is : %u\n", - reg_gfx_index.bitfields.se_broadcast_writes); - pr_debug("\t\t se_ind is : %u\n", reg_gfx_index.bitfields.se_index); - pr_debug("\t\t sh_ind is : %u\n", reg_gfx_index.bitfields.sh_index); - pr_debug("\t\t sbw is : %u\n", - reg_gfx_index.bitfields.sh_broadcast_writes); - - pr_debug("\t\t %30s\n", "* * * * * * * * * * * * * * * * * *"); - - status = kfd_gtt_sa_allocate(dbgdev->dev, ib_size, &mem_obj); - - if (status != 0) { - pr_err("Failed to allocate GART memory\n"); - return status; - } - - packet_buff_uint = mem_obj->cpu_ptr; - - memset(packet_buff_uint, 0, ib_size); - - packets_vec = (struct pm4__set_config_reg *) packet_buff_uint; - packets_vec[0].header.count = 1; - packets_vec[0].header.opcode = IT_SET_UCONFIG_REG; - packets_vec[0].header.type = PM4_TYPE_3; - packets_vec[0].bitfields2.reg_offset = - GRBM_GFX_INDEX / 4 - USERCONFIG_REG_BASE; - - packets_vec[0].bitfields2.insert_vmid = 0; - packets_vec[0].reg_data[0] = reg_gfx_index.u32All; - - packets_vec[1].header.count = 1; - packets_vec[1].header.opcode = IT_SET_CONFIG_REG; - packets_vec[1].header.type = PM4_TYPE_3; - packets_vec[1].bitfields2.reg_offset = SQ_CMD / 4 - AMD_CONFIG_REG_BASE; - - packets_vec[1].bitfields2.vmid_shift = SQ_CMD_VMID_OFFSET; - packets_vec[1].bitfields2.insert_vmid = 1; - packets_vec[1].reg_data[0] = reg_sq_cmd.u32All; - - /* Restore the GRBM_GFX_INDEX register */ - - reg_gfx_index.u32All = 0; - reg_gfx_index.bits.sh_broadcast_writes = 1; - reg_gfx_index.bits.instance_broadcast_writes = 1; - reg_gfx_index.bits.se_broadcast_writes = 1; - - - packets_vec[2].ordinal1 = packets_vec[0].ordinal1; - packets_vec[2].bitfields2.reg_offset = - GRBM_GFX_INDEX / 4 - USERCONFIG_REG_BASE; - - packets_vec[2].bitfields2.insert_vmid = 0; - packets_vec[2].reg_data[0] = reg_gfx_index.u32All; - - status = dbgdev_diq_submit_ib( - dbgdev, - wac_info->process->pasid, - mem_obj->gpu_addr, - packet_buff_uint, - ib_size); - - if (status) - pr_err("Failed to submit IB to DIQ\n"); - - kfd_gtt_sa_free(dbgdev->dev, mem_obj); - - return status; -} - -static int dbgdev_wave_control_nodiq(struct kfd_dbgdev *dbgdev, - struct dbg_wave_control_info *wac_info) -{ - int status; - union SQ_CMD_BITS reg_sq_cmd; - union GRBM_GFX_INDEX_BITS reg_gfx_index; - struct kfd_process_device *pdd; - - reg_sq_cmd.u32All = 0; - - /* taking the VMID for that process on the safe way using PDD */ - pdd = kfd_get_process_device_data(dbgdev->dev, wac_info->process); - - if (!pdd) { - pr_err("Failed to get pdd for wave control no DIQ\n"); - return -EFAULT; - } - status = dbgdev_wave_control_set_registers(wac_info, ®_sq_cmd, - ®_gfx_index); - if (status) { - pr_err("Failed to set wave control registers\n"); - return status; - } - - /* for non DIQ we need to patch the VMID: */ - - reg_sq_cmd.bits.vm_id = pdd->qpd.vmid; - - pr_debug("\t\t %30s\n", "* * * * * * * * * * * * * * * * * *"); - - pr_debug("\t\t mode is: %u\n", wac_info->mode); - pr_debug("\t\t operand is: %u\n", wac_info->operand); - pr_debug("\t\t trap id is: %u\n", wac_info->trapId); - pr_debug("\t\t msg value is: %u\n", - wac_info->dbgWave_msg.DbgWaveMsg.WaveMsgInfoGen2.Value); - pr_debug("\t\t vmid is: %u\n", pdd->qpd.vmid); - - pr_debug("\t\t chk_vmid is : %u\n", reg_sq_cmd.bitfields.check_vmid); - pr_debug("\t\t command is : %u\n", reg_sq_cmd.bitfields.cmd); - pr_debug("\t\t queue id is : %u\n", reg_sq_cmd.bitfields.queue_id); - pr_debug("\t\t simd id is : %u\n", reg_sq_cmd.bitfields.simd_id); - pr_debug("\t\t mode is : %u\n", reg_sq_cmd.bitfields.mode); - pr_debug("\t\t vm_id is : %u\n", reg_sq_cmd.bitfields.vm_id); - pr_debug("\t\t wave_id is : %u\n", reg_sq_cmd.bitfields.wave_id); - - pr_debug("\t\t ibw is : %u\n", - reg_gfx_index.bitfields.instance_broadcast_writes); - pr_debug("\t\t ii is : %u\n", - reg_gfx_index.bitfields.instance_index); - pr_debug("\t\t sebw is : %u\n", - reg_gfx_index.bitfields.se_broadcast_writes); - pr_debug("\t\t se_ind is : %u\n", reg_gfx_index.bitfields.se_index); - pr_debug("\t\t sh_ind is : %u\n", reg_gfx_index.bitfields.sh_index); - pr_debug("\t\t sbw is : %u\n", - reg_gfx_index.bitfields.sh_broadcast_writes); - - pr_debug("\t\t %30s\n", "* * * * * * * * * * * * * * * * * *"); - - return dbgdev->dev->kfd2kgd->wave_control_execute(dbgdev->dev->adev, - reg_gfx_index.u32All, - reg_sq_cmd.u32All); -} - -int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p) -{ - int status = 0; - unsigned int vmid; - uint16_t queried_pasid; - union SQ_CMD_BITS reg_sq_cmd; - union GRBM_GFX_INDEX_BITS reg_gfx_index; - struct kfd_process_device *pdd; - struct dbg_wave_control_info wac_info; - int first_vmid_to_scan = dev->vm_info.first_vmid_kfd; - int last_vmid_to_scan = dev->vm_info.last_vmid_kfd; - - reg_sq_cmd.u32All = 0; - status = 0; - - wac_info.mode = HSA_DBG_WAVEMODE_BROADCAST_PROCESS; - wac_info.operand = HSA_DBG_WAVEOP_KILL; - - pr_debug("Killing all process wavefronts\n"); - - /* Scan all registers in the range ATC_VMID8_PASID_MAPPING .. - * ATC_VMID15_PASID_MAPPING - * to check which VMID the current process is mapped to. - */ - - for (vmid = first_vmid_to_scan; vmid <= last_vmid_to_scan; vmid++) { - status = dev->kfd2kgd->get_atc_vmid_pasid_mapping_info - (dev->adev, vmid, &queried_pasid); - - if (status && queried_pasid == p->pasid) { - pr_debug("Killing wave fronts of vmid %d and pasid 0x%x\n", - vmid, p->pasid); - break; - } - } - - if (vmid > last_vmid_to_scan) { - pr_err("Didn't find vmid for pasid 0x%x\n", p->pasid); - return -EFAULT; - } - - /* taking the VMID for that process on the safe way using PDD */ - pdd = kfd_get_process_device_data(dev, p); - if (!pdd) - return -EFAULT; - - status = dbgdev_wave_control_set_registers(&wac_info, ®_sq_cmd, - ®_gfx_index); - if (status != 0) - return -EINVAL; - - /* for non DIQ we need to patch the VMID: */ - reg_sq_cmd.bits.vm_id = vmid; - - dev->kfd2kgd->wave_control_execute(dev->adev, - reg_gfx_index.u32All, - reg_sq_cmd.u32All); - - return 0; -} - -void kfd_dbgdev_init(struct kfd_dbgdev *pdbgdev, struct kfd_dev *pdev, - enum DBGDEV_TYPE type) -{ - pdbgdev->dev = pdev; - pdbgdev->kq = NULL; - pdbgdev->type = type; - pdbgdev->pqm = NULL; - - switch (type) { - case DBGDEV_TYPE_NODIQ: - pdbgdev->dbgdev_register = dbgdev_register_nodiq; - pdbgdev->dbgdev_unregister = dbgdev_unregister_nodiq; - pdbgdev->dbgdev_wave_control = dbgdev_wave_control_nodiq; - pdbgdev->dbgdev_address_watch = dbgdev_address_watch_nodiq; - break; - case DBGDEV_TYPE_DIQ: - default: - pdbgdev->dbgdev_register = dbgdev_register_diq; - pdbgdev->dbgdev_unregister = dbgdev_unregister_diq; - pdbgdev->dbgdev_wave_control = dbgdev_wave_control_diq; - pdbgdev->dbgdev_address_watch = dbgdev_address_watch_diq; - break; - } - -} diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.h b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.h deleted file mode 100644 index 0619c777b47e..000000000000 --- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.h +++ /dev/null @@ -1,230 +0,0 @@ -/* - * Copyright 2014 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - */ - -#ifndef KFD_DBGDEV_H_ -#define KFD_DBGDEV_H_ - -enum { - SQ_CMD_VMID_OFFSET = 28, - ADDRESS_WATCH_CNTL_OFFSET = 24 -}; - -enum { - PRIV_QUEUE_SYNC_TIME_MS = 200 -}; - -/* CONTEXT reg space definition */ -enum { - CONTEXT_REG_BASE = 0xA000, - CONTEXT_REG_END = 0xA400, - CONTEXT_REG_SIZE = CONTEXT_REG_END - CONTEXT_REG_BASE -}; - -/* USER CONFIG reg space definition */ -enum { - USERCONFIG_REG_BASE = 0xC000, - USERCONFIG_REG_END = 0x10000, - USERCONFIG_REG_SIZE = USERCONFIG_REG_END - USERCONFIG_REG_BASE -}; - -/* CONFIG reg space definition */ -enum { - AMD_CONFIG_REG_BASE = 0x2000, /* in dwords */ - AMD_CONFIG_REG_END = 0x2B00, - AMD_CONFIG_REG_SIZE = AMD_CONFIG_REG_END - AMD_CONFIG_REG_BASE -}; - -/* SH reg space definition */ -enum { - SH_REG_BASE = 0x2C00, - SH_REG_END = 0x3000, - SH_REG_SIZE = SH_REG_END - SH_REG_BASE -}; - -/* SQ_CMD definitions */ -#define SQ_CMD 0x8DEC - -enum SQ_IND_CMD_CMD { - SQ_IND_CMD_CMD_NULL = 0x00000000, - SQ_IND_CMD_CMD_HALT = 0x00000001, - SQ_IND_CMD_CMD_RESUME = 0x00000002, - SQ_IND_CMD_CMD_KILL = 0x00000003, - SQ_IND_CMD_CMD_DEBUG = 0x00000004, - SQ_IND_CMD_CMD_TRAP = 0x00000005, -}; - -enum SQ_IND_CMD_MODE { - SQ_IND_CMD_MODE_SINGLE = 0x00000000, - SQ_IND_CMD_MODE_BROADCAST = 0x00000001, - SQ_IND_CMD_MODE_BROADCAST_QUEUE = 0x00000002, - SQ_IND_CMD_MODE_BROADCAST_PIPE = 0x00000003, - SQ_IND_CMD_MODE_BROADCAST_ME = 0x00000004, -}; - -union SQ_IND_INDEX_BITS { - struct { - uint32_t wave_id:4; - uint32_t simd_id:2; - uint32_t thread_id:6; - uint32_t:1; - uint32_t force_read:1; - uint32_t read_timeout:1; - uint32_t unindexed:1; - uint32_t index:16; - - } bitfields, bits; - uint32_t u32All; - signed int i32All; - float f32All; -}; - -union SQ_IND_CMD_BITS { - struct { - uint32_t data:32; - } bitfields, bits; - uint32_t u32All; - signed int i32All; - float f32All; -}; - -union SQ_CMD_BITS { - struct { - uint32_t cmd:3; - uint32_t:1; - uint32_t mode:3; - uint32_t check_vmid:1; - uint32_t trap_id:3; - uint32_t:5; - uint32_t wave_id:4; - uint32_t simd_id:2; - uint32_t:2; - uint32_t queue_id:3; - uint32_t:1; - uint32_t vm_id:4; - } bitfields, bits; - uint32_t u32All; - signed int i32All; - float f32All; -}; - -union SQ_IND_DATA_BITS { - struct { - uint32_t data:32; - } bitfields, bits; - uint32_t u32All; - signed int i32All; - float f32All; -}; - -union GRBM_GFX_INDEX_BITS { - struct { - uint32_t instance_index:8; - uint32_t sh_index:8; - uint32_t se_index:8; - uint32_t:5; - uint32_t sh_broadcast_writes:1; - uint32_t instance_broadcast_writes:1; - uint32_t se_broadcast_writes:1; - } bitfields, bits; - uint32_t u32All; - signed int i32All; - float f32All; -}; - -union TCP_WATCH_ADDR_H_BITS { - struct { - uint32_t addr:16; - uint32_t:16; - - } bitfields, bits; - uint32_t u32All; - signed int i32All; - float f32All; -}; - -union TCP_WATCH_ADDR_L_BITS { - struct { - uint32_t:6; - uint32_t addr:26; - } bitfields, bits; - uint32_t u32All; - signed int i32All; - float f32All; -}; - -enum { - QUEUESTATE__INVALID = 0, /* so by default we'll get invalid state */ - QUEUESTATE__ACTIVE_COMPLETION_PENDING, - QUEUESTATE__ACTIVE -}; - -union ULARGE_INTEGER { - struct { - uint32_t low_part; - uint32_t high_part; - } u; - unsigned long long quad_part; -}; - - -#define KFD_CIK_VMID_START_OFFSET (8) -#define KFD_CIK_VMID_END_OFFSET (KFD_CIK_VMID_START_OFFSET + (8)) - - -void kfd_dbgdev_init(struct kfd_dbgdev *pdbgdev, struct kfd_dev *pdev, - enum DBGDEV_TYPE type); - -union TCP_WATCH_CNTL_BITS { - struct { - uint32_t mask:24; - uint32_t vmid:4; - uint32_t atc:1; - uint32_t mode:2; - uint32_t valid:1; - } bitfields, bits; - uint32_t u32All; - signed int i32All; - float f32All; -}; - -enum { - ADDRESS_WATCH_REG_CNTL_ATC_BIT = 0x10000000UL, - ADDRESS_WATCH_REG_CNTL_DEFAULT_MASK = 0x00FFFFFF, - ADDRESS_WATCH_REG_ADDLOW_MASK_EXTENSION = 0x03000000, - /* extend the mask to 26 bits in order to match the low address field */ - ADDRESS_WATCH_REG_ADDLOW_SHIFT = 6, - ADDRESS_WATCH_REG_ADDHIGH_MASK = 0xFFFF -}; - -enum { - MAX_TRAPID = 8, /* 3 bits in the bitfield. */ - MAX_WATCH_ADDRESSES = 4 -}; - -enum { - ADDRESS_WATCH_REG_ADDR_HI = 0, - ADDRESS_WATCH_REG_ADDR_LO, - ADDRESS_WATCH_REG_CNTL, - ADDRESS_WATCH_REG_MAX -}; - -#endif /* KFD_DBGDEV_H_ */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c b/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c deleted file mode 100644 index 9bfa50633654..000000000000 --- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.c +++ /dev/null @@ -1,158 +0,0 @@ -/* - * Copyright 2014 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - */ -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/log2.h> -#include <linux/sched.h> -#include <linux/slab.h> -#include <linux/device.h> - -#include "kfd_priv.h" -#include "cik_regs.h" -#include "kfd_pm4_headers.h" -#include "kfd_pm4_headers_diq.h" -#include "kfd_dbgmgr.h" -#include "kfd_dbgdev.h" -#include "kfd_device_queue_manager.h" - -static DEFINE_MUTEX(kfd_dbgmgr_mutex); - -struct mutex *kfd_get_dbgmgr_mutex(void) -{ - return &kfd_dbgmgr_mutex; -} - - -static void kfd_dbgmgr_uninitialize(struct kfd_dbgmgr *pmgr) -{ - kfree(pmgr->dbgdev); - - pmgr->dbgdev = NULL; - pmgr->pasid = 0; - pmgr->dev = NULL; -} - -void kfd_dbgmgr_destroy(struct kfd_dbgmgr *pmgr) -{ - if (pmgr) { - kfd_dbgmgr_uninitialize(pmgr); - kfree(pmgr); - } -} - -bool kfd_dbgmgr_create(struct kfd_dbgmgr **ppmgr, struct kfd_dev *pdev) -{ - enum DBGDEV_TYPE type = DBGDEV_TYPE_DIQ; - struct kfd_dbgmgr *new_buff; - - if (WARN_ON(!pdev->init_complete)) - return false; - - new_buff = kfd_alloc_struct(new_buff); - if (!new_buff) { - pr_err("Failed to allocate dbgmgr instance\n"); - return false; - } - - new_buff->pasid = 0; - new_buff->dev = pdev; - new_buff->dbgdev = kfd_alloc_struct(new_buff->dbgdev); - if (!new_buff->dbgdev) { - pr_err("Failed to allocate dbgdev instance\n"); - kfree(new_buff); - return false; - } - - /* get actual type of DBGDevice cpsch or not */ - if (pdev->dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) - type = DBGDEV_TYPE_NODIQ; - - kfd_dbgdev_init(new_buff->dbgdev, pdev, type); - *ppmgr = new_buff; - - return true; -} - -long kfd_dbgmgr_register(struct kfd_dbgmgr *pmgr, struct kfd_process *p) -{ - if (pmgr->pasid != 0) { - pr_debug("H/W debugger is already active using pasid 0x%x\n", - pmgr->pasid); - return -EBUSY; - } - - /* remember pasid */ - pmgr->pasid = p->pasid; - - /* provide the pqm for diq generation */ - pmgr->dbgdev->pqm = &p->pqm; - - /* activate the actual registering */ - pmgr->dbgdev->dbgdev_register(pmgr->dbgdev); - - return 0; -} - -long kfd_dbgmgr_unregister(struct kfd_dbgmgr *pmgr, struct kfd_process *p) -{ - /* Is the requests coming from the already registered process? */ - if (pmgr->pasid != p->pasid) { - pr_debug("H/W debugger is not registered by calling pasid 0x%x\n", - p->pasid); - return -EINVAL; - } - - pmgr->dbgdev->dbgdev_unregister(pmgr->dbgdev); - - pmgr->pasid = 0; - - return 0; -} - -long kfd_dbgmgr_wave_control(struct kfd_dbgmgr *pmgr, - struct dbg_wave_control_info *wac_info) -{ - /* Is the requests coming from the already registered process? */ - if (pmgr->pasid != wac_info->process->pasid) { - pr_debug("H/W debugger support was not registered for requester pasid 0x%x\n", - wac_info->process->pasid); - return -EINVAL; - } - - return (long) pmgr->dbgdev->dbgdev_wave_control(pmgr->dbgdev, wac_info); -} - -long kfd_dbgmgr_address_watch(struct kfd_dbgmgr *pmgr, - struct dbg_address_watch_info *adw_info) -{ - /* Is the requests coming from the already registered process? */ - if (pmgr->pasid != adw_info->process->pasid) { - pr_debug("H/W debugger support was not registered for requester pasid 0x%x\n", - adw_info->process->pasid); - return -EINVAL; - } - - return (long) pmgr->dbgdev->dbgdev_address_watch(pmgr->dbgdev, - adw_info); -} - diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.h b/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.h deleted file mode 100644 index f9c6df1fdc5c..000000000000 --- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgmgr.h +++ /dev/null @@ -1,293 +0,0 @@ -/* - * Copyright 2014 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - */ - -#ifndef KFD_DBGMGR_H_ -#define KFD_DBGMGR_H_ - -#include "kfd_priv.h" - -/* must align with hsakmttypes definition */ -#pragma pack(push, 4) - -enum HSA_DBG_WAVEOP { - HSA_DBG_WAVEOP_HALT = 1, /* Halts a wavefront */ - HSA_DBG_WAVEOP_RESUME = 2, /* Resumes a wavefront */ - HSA_DBG_WAVEOP_KILL = 3, /* Kills a wavefront */ - HSA_DBG_WAVEOP_DEBUG = 4, /* Causes wavefront to enter dbg mode */ - HSA_DBG_WAVEOP_TRAP = 5, /* Causes wavefront to take a trap */ - HSA_DBG_NUM_WAVEOP = 5, - HSA_DBG_MAX_WAVEOP = 0xFFFFFFFF -}; - -enum HSA_DBG_WAVEMODE { - /* send command to a single wave */ - HSA_DBG_WAVEMODE_SINGLE = 0, - /* - * Broadcast to all wavefronts of all processes is not - * supported for HSA user mode - */ - - /* send to waves within current process */ - HSA_DBG_WAVEMODE_BROADCAST_PROCESS = 2, - /* send to waves within current process on CU */ - HSA_DBG_WAVEMODE_BROADCAST_PROCESS_CU = 3, - HSA_DBG_NUM_WAVEMODE = 3, - HSA_DBG_MAX_WAVEMODE = 0xFFFFFFFF -}; - -enum HSA_DBG_WAVEMSG_TYPE { - HSA_DBG_WAVEMSG_AUTO = 0, - HSA_DBG_WAVEMSG_USER = 1, - HSA_DBG_WAVEMSG_ERROR = 2, - HSA_DBG_NUM_WAVEMSG, - HSA_DBG_MAX_WAVEMSG = 0xFFFFFFFF -}; - -enum HSA_DBG_WATCH_MODE { - HSA_DBG_WATCH_READ = 0, /* Read operations only */ - HSA_DBG_WATCH_NONREAD = 1, /* Write or Atomic operations only */ - HSA_DBG_WATCH_ATOMIC = 2, /* Atomic Operations only */ - HSA_DBG_WATCH_ALL = 3, /* Read, Write or Atomic operations */ - HSA_DBG_WATCH_NUM, - HSA_DBG_WATCH_SIZE = 0xFFFFFFFF -}; - -/* This structure is hardware specific and may change in the future */ -struct HsaDbgWaveMsgAMDGen2 { - union { - struct ui32 { - uint32_t UserData:8; /* user data */ - uint32_t ShaderArray:1; /* Shader array */ - uint32_t Priv:1; /* Privileged */ - uint32_t Reserved0:4; /* Reserved, should be 0 */ - uint32_t WaveId:4; /* wave id */ - uint32_t SIMD:2; /* SIMD id */ - uint32_t HSACU:4; /* Compute unit */ - uint32_t ShaderEngine:2;/* Shader engine */ - uint32_t MessageType:2; /* see HSA_DBG_WAVEMSG_TYPE */ - uint32_t Reserved1:4; /* Reserved, should be 0 */ - } ui32; - uint32_t Value; - }; - uint32_t Reserved2; -}; - -union HsaDbgWaveMessageAMD { - struct HsaDbgWaveMsgAMDGen2 WaveMsgInfoGen2; - /* for future HsaDbgWaveMsgAMDGen3; */ -}; - -struct HsaDbgWaveMessage { - void *MemoryVA; /* ptr to associated host-accessible data */ - union HsaDbgWaveMessageAMD DbgWaveMsg; -}; - -/* - * TODO: This definitions to be MOVED to kfd_event, once it is implemented. - * - * HSA sync primitive, Event and HW Exception notification API definitions. - * The API functions allow the runtime to define a so-called sync-primitive, - * a SW object combining a user-mode provided "syncvar" and a scheduler event - * that can be signaled through a defined GPU interrupt. A syncvar is - * a process virtual memory location of a certain size that can be accessed - * by CPU and GPU shader code within the process to set and query the content - * within that memory. The definition of the content is determined by the HSA - * runtime and potentially GPU shader code interfacing with the HSA runtime. - * The syncvar values may be commonly written through an PM4 WRITE_DATA packet - * in the user mode instruction stream. The OS scheduler event is typically - * associated and signaled by an interrupt issued by the GPU, but other HSA - * system interrupt conditions from other HW (e.g. IOMMUv2) may be surfaced - * by the KFD by this mechanism, too. - */ - -/* these are the new definitions for events */ -enum HSA_EVENTTYPE { - HSA_EVENTTYPE_SIGNAL = 0, /* user-mode generated GPU signal */ - HSA_EVENTTYPE_NODECHANGE = 1, /* HSA node change (attach/detach) */ - HSA_EVENTTYPE_DEVICESTATECHANGE = 2, /* HSA device state change - * (start/stop) - */ - HSA_EVENTTYPE_HW_EXCEPTION = 3, /* GPU shader exception event */ - HSA_EVENTTYPE_SYSTEM_EVENT = 4, /* GPU SYSCALL with parameter info */ - HSA_EVENTTYPE_DEBUG_EVENT = 5, /* GPU signal for debugging */ - HSA_EVENTTYPE_PROFILE_EVENT = 6,/* GPU signal for profiling */ - HSA_EVENTTYPE_QUEUE_EVENT = 7, /* GPU signal queue idle state - * (EOP pm4) - */ - /* ... */ - HSA_EVENTTYPE_MAXID, - HSA_EVENTTYPE_TYPE_SIZE = 0xFFFFFFFF -}; - -/* Sub-definitions for various event types: Syncvar */ -struct HsaSyncVar { - union SyncVar { - void *UserData; /* pointer to user mode data */ - uint64_t UserDataPtrValue; /* 64bit compatibility of value */ - } SyncVar; - uint64_t SyncVarSize; -}; - -/* Sub-definitions for various event types: NodeChange */ - -enum HSA_EVENTTYPE_NODECHANGE_FLAGS { - HSA_EVENTTYPE_NODECHANGE_ADD = 0, - HSA_EVENTTYPE_NODECHANGE_REMOVE = 1, - HSA_EVENTTYPE_NODECHANGE_SIZE = 0xFFFFFFFF -}; - -struct HsaNodeChange { - /* HSA node added/removed on the platform */ - enum HSA_EVENTTYPE_NODECHANGE_FLAGS Flags; -}; - -/* Sub-definitions for various event types: DeviceStateChange */ -enum HSA_EVENTTYPE_DEVICESTATECHANGE_FLAGS { - /* device started (and available) */ - HSA_EVENTTYPE_DEVICESTATUSCHANGE_START = 0, - /* device stopped (i.e. unavailable) */ - HSA_EVENTTYPE_DEVICESTATUSCHANGE_STOP = 1, - HSA_EVENTTYPE_DEVICESTATUSCHANGE_SIZE = 0xFFFFFFFF -}; - -enum HSA_DEVICE { - HSA_DEVICE_CPU = 0, - HSA_DEVICE_GPU = 1, - MAX_HSA_DEVICE = 2 -}; - -struct HsaDeviceStateChange { - uint32_t NodeId; /* F-NUMA node that contains the device */ - enum HSA_DEVICE Device; /* device type: GPU or CPU */ - enum HSA_EVENTTYPE_DEVICESTATECHANGE_FLAGS Flags; /* event flags */ -}; - -struct HsaEventData { - enum HSA_EVENTTYPE EventType; /* event type */ - union EventData { - /* - * return data associated with HSA_EVENTTYPE_SIGNAL - * and other events - */ - struct HsaSyncVar SyncVar; - - /* data associated with HSA_EVENTTYPE_NODE_CHANGE */ - struct HsaNodeChange NodeChangeState; - - /* data associated with HSA_EVENTTYPE_DEVICE_STATE_CHANGE */ - struct HsaDeviceStateChange DeviceState; - } EventData; - - /* the following data entries are internal to the KFD & thunk itself */ - - /* internal thunk store for Event data (OsEventHandle) */ - uint64_t HWData1; - /* internal thunk store for Event data (HWAddress) */ - uint64_t HWData2; - /* internal thunk store for Event data (HWData) */ - uint32_t HWData3; -}; - -struct HsaEventDescriptor { - /* event type to allocate */ - enum HSA_EVENTTYPE EventType; - /* H-NUMA node containing GPU device that is event source */ - uint32_t NodeId; - /* pointer to user mode syncvar data, syncvar->UserDataPtrValue - * may be NULL - */ - struct HsaSyncVar SyncVar; -}; - -struct HsaEvent { - uint32_t EventId; - struct HsaEventData EventData; -}; - -#pragma pack(pop) - -enum DBGDEV_TYPE { - DBGDEV_TYPE_ILLEGAL = 0, - DBGDEV_TYPE_NODIQ = 1, - DBGDEV_TYPE_DIQ = 2, - DBGDEV_TYPE_TEST = 3 -}; - -struct dbg_address_watch_info { - struct kfd_process *process; - enum HSA_DBG_WATCH_MODE *watch_mode; - uint64_t *watch_address; - uint64_t *watch_mask; - struct HsaEvent *watch_event; - uint32_t num_watch_points; -}; - -struct dbg_wave_control_info { - struct kfd_process *process; - uint32_t trapId; - enum HSA_DBG_WAVEOP operand; - enum HSA_DBG_WAVEMODE mode; - struct HsaDbgWaveMessage dbgWave_msg; -}; - -struct kfd_dbgdev { - - /* The device that owns this data. */ - struct kfd_dev *dev; - - /* kernel queue for DIQ */ - struct kernel_queue *kq; - - /* a pointer to the pqm of the calling process */ - struct process_queue_manager *pqm; - - /* type of debug device ( DIQ, non DIQ, etc. ) */ - enum DBGDEV_TYPE type; - - /* virtualized function pointers to device dbg */ - int (*dbgdev_register)(struct kfd_dbgdev *dbgdev); - int (*dbgdev_unregister)(struct kfd_dbgdev *dbgdev); - int (*dbgdev_address_watch)(struct kfd_dbgdev *dbgdev, - struct dbg_address_watch_info *adw_info); - int (*dbgdev_wave_control)(struct kfd_dbgdev *dbgdev, - struct dbg_wave_control_info *wac_info); - -}; - -struct kfd_dbgmgr { - u32 pasid; - struct kfd_dev *dev; - struct kfd_dbgdev *dbgdev; -}; - -/* prototypes for debug manager functions */ -struct mutex *kfd_get_dbgmgr_mutex(void); -void kfd_dbgmgr_destroy(struct kfd_dbgmgr *pmgr); -bool kfd_dbgmgr_create(struct kfd_dbgmgr **ppmgr, struct kfd_dev *pdev); -long kfd_dbgmgr_register(struct kfd_dbgmgr *pmgr, struct kfd_process *p); -long kfd_dbgmgr_unregister(struct kfd_dbgmgr *pmgr, struct kfd_process *p); -long kfd_dbgmgr_wave_control(struct kfd_dbgmgr *pmgr, - struct dbg_wave_control_info *wac_info); -long kfd_dbgmgr_address_watch(struct kfd_dbgmgr *pmgr, - struct dbg_address_watch_info *adw_info); -#endif /* KFD_DBGMGR_H_ */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index 2b65d0acae2c..7f1746289989 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -64,34 +64,33 @@ static void kfd_device_info_set_sdma_queue_num(struct kfd_dev *kfd) uint32_t sdma_version = kfd->adev->ip_versions[SDMA0_HWIP][0]; switch (sdma_version) { - case IP_VERSION(4, 0, 0):/* VEGA10 */ - case IP_VERSION(4, 0, 1):/* VEGA12 */ - case IP_VERSION(4, 1, 0):/* RAVEN */ - case IP_VERSION(4, 1, 1):/* RAVEN */ - case IP_VERSION(4, 1, 2):/* RENOIR */ - case IP_VERSION(5, 2, 1):/* VANGOGH */ - case IP_VERSION(5, 2, 3):/* YELLOW_CARP */ - kfd->device_info.num_sdma_queues_per_engine = 2; - break; - case IP_VERSION(4, 2, 0):/* VEGA20 */ - case IP_VERSION(4, 2, 2):/* ARCTURUS */ - case IP_VERSION(4, 4, 0):/* ALDEBARAN */ - case IP_VERSION(5, 0, 0):/* NAVI10 */ - case IP_VERSION(5, 0, 1):/* CYAN_SKILLFISH */ - case IP_VERSION(5, 0, 2):/* NAVI14 */ - case IP_VERSION(5, 0, 5):/* NAVI12 */ - case IP_VERSION(5, 2, 0):/* SIENNA_CICHLID */ - case IP_VERSION(5, 2, 2):/* NAVY_FLOUNDER */ - case IP_VERSION(5, 2, 4):/* DIMGREY_CAVEFISH */ - case IP_VERSION(5, 2, 5):/* BEIGE_GOBY */ - kfd->device_info.num_sdma_queues_per_engine = 8; - break; - default: - dev_warn(kfd_device, - "Default sdma queue per engine(8) is set due to " - "mismatch of sdma ip block(SDMA_HWIP:0x%x).\n", - sdma_version); - kfd->device_info.num_sdma_queues_per_engine = 8; + case IP_VERSION(4, 0, 0):/* VEGA10 */ + case IP_VERSION(4, 0, 1):/* VEGA12 */ + case IP_VERSION(4, 1, 0):/* RAVEN */ + case IP_VERSION(4, 1, 1):/* RAVEN */ + case IP_VERSION(4, 1, 2):/* RENOIR */ + case IP_VERSION(5, 2, 1):/* VANGOGH */ + case IP_VERSION(5, 2, 3):/* YELLOW_CARP */ + kfd->device_info.num_sdma_queues_per_engine = 2; + break; + case IP_VERSION(4, 2, 0):/* VEGA20 */ + case IP_VERSION(4, 2, 2):/* ARCTURUS */ + case IP_VERSION(4, 4, 0):/* ALDEBARAN */ + case IP_VERSION(5, 0, 0):/* NAVI10 */ + case IP_VERSION(5, 0, 1):/* CYAN_SKILLFISH */ + case IP_VERSION(5, 0, 2):/* NAVI14 */ + case IP_VERSION(5, 0, 5):/* NAVI12 */ + case IP_VERSION(5, 2, 0):/* SIENNA_CICHLID */ + case IP_VERSION(5, 2, 2):/* NAVY_FLOUNDER */ + case IP_VERSION(5, 2, 4):/* DIMGREY_CAVEFISH */ + case IP_VERSION(5, 2, 5):/* BEIGE_GOBY */ + kfd->device_info.num_sdma_queues_per_engine = 8; + break; + default: + dev_warn(kfd_device, + "Default sdma queue per engine(8) is set due to mismatch of sdma ip block(SDMA_HWIP:0x%x).\n", + sdma_version); + kfd->device_info.num_sdma_queues_per_engine = 8; } } @@ -111,6 +110,7 @@ static void kfd_device_info_set_event_interrupt_class(struct kfd_dev *kfd) case IP_VERSION(10, 3, 1): /* VANGOGH */ case IP_VERSION(10, 3, 3): /* YELLOW_CARP */ case IP_VERSION(10, 1, 3): /* CYAN_SKILLFISH */ + case IP_VERSION(10, 1, 4): case IP_VERSION(10, 1, 10): /* NAVI10 */ case IP_VERSION(10, 1, 2): /* NAVI12 */ case IP_VERSION(10, 1, 1): /* NAVI14 */ @@ -308,6 +308,7 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) break; /* Cyan Skillfish */ case IP_VERSION(10, 1, 3): + case IP_VERSION(10, 1, 4): gfx_target_version = 100103; if (!vf) f2g = &gfx_v10_kfd2kgd; @@ -576,8 +577,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, if (kfd_resume(kfd)) goto kfd_resume_error; - kfd->dbgmgr = NULL; - if (kfd_topology_add_device(kfd)) { dev_err(kfd_device, "Error adding device to topology\n"); goto kfd_topology_add_device_error; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 4b6814949aad..7f6f1a842b0b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -58,7 +58,7 @@ static inline void deallocate_hqd(struct device_queue_manager *dqm, struct queue *q); static int allocate_hqd(struct device_queue_manager *dqm, struct queue *q); static int allocate_sdma_queue(struct device_queue_manager *dqm, - struct queue *q); + struct queue *q, const uint32_t *restore_sdma_id); static void kfd_process_hw_exception(struct work_struct *work); static inline @@ -144,7 +144,13 @@ static void decrement_queue_count(struct device_queue_manager *dqm, dqm->active_cp_queue_count--; } -static int allocate_doorbell(struct qcm_process_device *qpd, struct queue *q) +/* + * Allocate a doorbell ID to this queue. + * If doorbell_id is passed in, make sure requested ID is valid then allocate it. + */ +static int allocate_doorbell(struct qcm_process_device *qpd, + struct queue *q, + uint32_t const *restore_id) { struct kfd_dev *dev = qpd->dqm->dev; @@ -152,6 +158,10 @@ static int allocate_doorbell(struct qcm_process_device *qpd, struct queue *q) /* On pre-SOC15 chips we need to use the queue ID to * preserve the user mode ABI. */ + + if (restore_id && *restore_id != q->properties.queue_id) + return -EINVAL; + q->doorbell_id = q->properties.queue_id; } else if (q->properties.type == KFD_QUEUE_TYPE_SDMA || q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { @@ -160,25 +170,37 @@ static int allocate_doorbell(struct qcm_process_device *qpd, struct queue *q) * The doobell index distance between RLC (2*i) and (2*i+1) * for a SDMA engine is 512. */ - uint32_t *idx_offset = - dev->shared_resources.sdma_doorbell_idx; - q->doorbell_id = idx_offset[q->properties.sdma_engine_id] - + (q->properties.sdma_queue_id & 1) - * KFD_QUEUE_DOORBELL_MIRROR_OFFSET - + (q->properties.sdma_queue_id >> 1); + uint32_t *idx_offset = dev->shared_resources.sdma_doorbell_idx; + uint32_t valid_id = idx_offset[q->properties.sdma_engine_id] + + (q->properties.sdma_queue_id & 1) + * KFD_QUEUE_DOORBELL_MIRROR_OFFSET + + (q->properties.sdma_queue_id >> 1); + + if (restore_id && *restore_id != valid_id) + return -EINVAL; + q->doorbell_id = valid_id; } else { - /* For CP queues on SOC15 reserve a free doorbell ID */ - unsigned int found; - - found = find_first_zero_bit(qpd->doorbell_bitmap, - KFD_MAX_NUM_OF_QUEUES_PER_PROCESS); - if (found >= KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) { - pr_debug("No doorbells available"); - return -EBUSY; + /* For CP queues on SOC15 */ + if (restore_id) { + /* make sure that ID is free */ + if (__test_and_set_bit(*restore_id, qpd->doorbell_bitmap)) + return -EINVAL; + + q->doorbell_id = *restore_id; + } else { + /* or reserve a free doorbell ID */ + unsigned int found; + + found = find_first_zero_bit(qpd->doorbell_bitmap, + KFD_MAX_NUM_OF_QUEUES_PER_PROCESS); + if (found >= KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) { + pr_debug("No doorbells available"); + return -EBUSY; + } + set_bit(found, qpd->doorbell_bitmap); + q->doorbell_id = found; } - set_bit(found, qpd->doorbell_bitmap); - q->doorbell_id = found; } q->properties.doorbell_off = @@ -299,7 +321,9 @@ static void deallocate_vmid(struct device_queue_manager *dqm, static int create_queue_nocpsch(struct device_queue_manager *dqm, struct queue *q, - struct qcm_process_device *qpd) + struct qcm_process_device *qpd, + const struct kfd_criu_queue_priv_data *qd, + const void *restore_mqd, const void *restore_ctl_stack) { struct mqd_manager *mqd_mgr; int retval; @@ -339,13 +363,13 @@ static int create_queue_nocpsch(struct device_queue_manager *dqm, q->pipe, q->queue); } else if (q->properties.type == KFD_QUEUE_TYPE_SDMA || q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { - retval = allocate_sdma_queue(dqm, q); + retval = allocate_sdma_queue(dqm, q, qd ? &qd->sdma_id : NULL); if (retval) goto deallocate_vmid; dqm->asic_ops.init_sdma_vm(dqm, q, qpd); } - retval = allocate_doorbell(qpd, q); + retval = allocate_doorbell(qpd, q, qd ? &qd->doorbell_id : NULL); if (retval) goto out_deallocate_hqd; @@ -358,8 +382,15 @@ static int create_queue_nocpsch(struct device_queue_manager *dqm, retval = -ENOMEM; goto out_deallocate_doorbell; } - mqd_mgr->init_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, - &q->gart_mqd_addr, &q->properties); + + if (qd) + mqd_mgr->restore_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, &q->gart_mqd_addr, + &q->properties, restore_mqd, restore_ctl_stack, + qd->ctl_stack_size); + else + mqd_mgr->init_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, + &q->gart_mqd_addr, &q->properties); + if (q->properties.is_active) { if (!dqm->sched_running) { WARN_ONCE(1, "Load non-HWS mqd while stopped\n"); @@ -449,6 +480,65 @@ static inline void deallocate_hqd(struct device_queue_manager *dqm, dqm->allocated_queues[q->pipe] |= (1 << q->queue); } +#define SQ_IND_CMD_CMD_KILL 0x00000003 +#define SQ_IND_CMD_MODE_BROADCAST 0x00000001 + +static int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p) +{ + int status = 0; + unsigned int vmid; + uint16_t queried_pasid; + union SQ_CMD_BITS reg_sq_cmd; + union GRBM_GFX_INDEX_BITS reg_gfx_index; + struct kfd_process_device *pdd; + int first_vmid_to_scan = dev->vm_info.first_vmid_kfd; + int last_vmid_to_scan = dev->vm_info.last_vmid_kfd; + + reg_sq_cmd.u32All = 0; + reg_gfx_index.u32All = 0; + + pr_debug("Killing all process wavefronts\n"); + + /* Scan all registers in the range ATC_VMID8_PASID_MAPPING .. + * ATC_VMID15_PASID_MAPPING + * to check which VMID the current process is mapped to. + */ + + for (vmid = first_vmid_to_scan; vmid <= last_vmid_to_scan; vmid++) { + status = dev->kfd2kgd->get_atc_vmid_pasid_mapping_info + (dev->adev, vmid, &queried_pasid); + + if (status && queried_pasid == p->pasid) { + pr_debug("Killing wave fronts of vmid %d and pasid 0x%x\n", + vmid, p->pasid); + break; + } + } + + if (vmid > last_vmid_to_scan) { + pr_err("Didn't find vmid for pasid 0x%x\n", p->pasid); + return -EFAULT; + } + + /* taking the VMID for that process on the safe way using PDD */ + pdd = kfd_get_process_device_data(dev, p); + if (!pdd) + return -EFAULT; + + reg_gfx_index.bits.sh_broadcast_writes = 1; + reg_gfx_index.bits.se_broadcast_writes = 1; + reg_gfx_index.bits.instance_broadcast_writes = 1; + reg_sq_cmd.bits.mode = SQ_IND_CMD_MODE_BROADCAST; + reg_sq_cmd.bits.cmd = SQ_IND_CMD_CMD_KILL; + reg_sq_cmd.bits.vm_id = vmid; + + dev->kfd2kgd->wave_control_execute(dev->adev, + reg_gfx_index.u32All, + reg_sq_cmd.u32All); + + return 0; +} + /* Access to DQM has to be locked before calling destroy_queue_nocpsch_locked * to avoid asynchronized access */ @@ -1034,7 +1124,7 @@ static void pre_reset(struct device_queue_manager *dqm) } static int allocate_sdma_queue(struct device_queue_manager *dqm, - struct queue *q) + struct queue *q, const uint32_t *restore_sdma_id) { int bit; @@ -1044,9 +1134,21 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm, return -ENOMEM; } - bit = __ffs64(dqm->sdma_bitmap); - dqm->sdma_bitmap &= ~(1ULL << bit); - q->sdma_id = bit; + if (restore_sdma_id) { + /* Re-use existing sdma_id */ + if (!(dqm->sdma_bitmap & (1ULL << *restore_sdma_id))) { + pr_err("SDMA queue already in use\n"); + return -EBUSY; + } + dqm->sdma_bitmap &= ~(1ULL << *restore_sdma_id); + q->sdma_id = *restore_sdma_id; + } else { + /* Find first available sdma_id */ + bit = __ffs64(dqm->sdma_bitmap); + dqm->sdma_bitmap &= ~(1ULL << bit); + q->sdma_id = bit; + } + q->properties.sdma_engine_id = q->sdma_id % kfd_get_num_sdma_engines(dqm->dev); q->properties.sdma_queue_id = q->sdma_id / @@ -1056,9 +1158,19 @@ static int allocate_sdma_queue(struct device_queue_manager *dqm, pr_err("No more XGMI SDMA queue to allocate\n"); return -ENOMEM; } - bit = __ffs64(dqm->xgmi_sdma_bitmap); - dqm->xgmi_sdma_bitmap &= ~(1ULL << bit); - q->sdma_id = bit; + if (restore_sdma_id) { + /* Re-use existing sdma_id */ + if (!(dqm->xgmi_sdma_bitmap & (1ULL << *restore_sdma_id))) { + pr_err("SDMA queue already in use\n"); + return -EBUSY; + } + dqm->xgmi_sdma_bitmap &= ~(1ULL << *restore_sdma_id); + q->sdma_id = *restore_sdma_id; + } else { + bit = __ffs64(dqm->xgmi_sdma_bitmap); + dqm->xgmi_sdma_bitmap &= ~(1ULL << bit); + q->sdma_id = bit; + } /* sdma_engine_id is sdma id including * both PCIe-optimized SDMAs and XGMI- * optimized SDMAs. The calculation below @@ -1288,7 +1400,9 @@ static void destroy_kernel_queue_cpsch(struct device_queue_manager *dqm, } static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q, - struct qcm_process_device *qpd) + struct qcm_process_device *qpd, + const struct kfd_criu_queue_priv_data *qd, + const void *restore_mqd, const void *restore_ctl_stack) { int retval; struct mqd_manager *mqd_mgr; @@ -1303,13 +1417,13 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q, if (q->properties.type == KFD_QUEUE_TYPE_SDMA || q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { dqm_lock(dqm); - retval = allocate_sdma_queue(dqm, q); + retval = allocate_sdma_queue(dqm, q, qd ? &qd->sdma_id : NULL); dqm_unlock(dqm); if (retval) goto out; } - retval = allocate_doorbell(qpd, q); + retval = allocate_doorbell(qpd, q, qd ? &qd->doorbell_id : NULL); if (retval) goto out_deallocate_sdma_queue; @@ -1334,8 +1448,14 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q, * updates the is_evicted flag but is a no-op otherwise. */ q->properties.is_evicted = !!qpd->evicted; - mqd_mgr->init_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, - &q->gart_mqd_addr, &q->properties); + + if (qd) + mqd_mgr->restore_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, &q->gart_mqd_addr, + &q->properties, restore_mqd, restore_ctl_stack, + qd->ctl_stack_size); + else + mqd_mgr->init_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, + &q->gart_mqd_addr, &q->properties); list_add(&q->list, &qpd->queues_list); qpd->queue_count++; @@ -1738,6 +1858,56 @@ static int get_wave_state(struct device_queue_manager *dqm, ctl_stack_used_size, save_area_used_size); } +static void get_queue_checkpoint_info(struct device_queue_manager *dqm, + const struct queue *q, + u32 *mqd_size, + u32 *ctl_stack_size) +{ + struct mqd_manager *mqd_mgr; + enum KFD_MQD_TYPE mqd_type = + get_mqd_type_from_queue_type(q->properties.type); + + dqm_lock(dqm); + mqd_mgr = dqm->mqd_mgrs[mqd_type]; + *mqd_size = mqd_mgr->mqd_size; + *ctl_stack_size = 0; + + if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE && mqd_mgr->get_checkpoint_info) + mqd_mgr->get_checkpoint_info(mqd_mgr, q->mqd, ctl_stack_size); + + dqm_unlock(dqm); +} + +static int checkpoint_mqd(struct device_queue_manager *dqm, + const struct queue *q, + void *mqd, + void *ctl_stack) +{ + struct mqd_manager *mqd_mgr; + int r = 0; + enum KFD_MQD_TYPE mqd_type = + get_mqd_type_from_queue_type(q->properties.type); + + dqm_lock(dqm); + + if (q->properties.is_active || !q->device->cwsr_enabled) { + r = -EINVAL; + goto dqm_unlock; + } + + mqd_mgr = dqm->mqd_mgrs[mqd_type]; + if (!mqd_mgr->checkpoint_mqd) { + r = -EOPNOTSUPP; + goto dqm_unlock; + } + + mqd_mgr->checkpoint_mqd(mqd_mgr, q->mqd, mqd, ctl_stack); + +dqm_unlock: + dqm_unlock(dqm); + return r; +} + static int process_termination_cpsch(struct device_queue_manager *dqm, struct qcm_process_device *qpd) { @@ -1915,6 +2085,8 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev) dqm->ops.restore_process_queues = restore_process_queues_cpsch; dqm->ops.get_wave_state = get_wave_state; dqm->ops.reset_queues = reset_queues_cpsch; + dqm->ops.get_queue_checkpoint_info = get_queue_checkpoint_info; + dqm->ops.checkpoint_mqd = checkpoint_mqd; break; case KFD_SCHED_POLICY_NO_HWS: /* initialize dqm for no cp scheduling */ @@ -1934,6 +2106,8 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev) dqm->ops.restore_process_queues = restore_process_queues_nocpsch; dqm->ops.get_wave_state = get_wave_state; + dqm->ops.get_queue_checkpoint_info = get_queue_checkpoint_info; + dqm->ops.checkpoint_mqd = checkpoint_mqd; break; default: pr_err("Invalid scheduling policy %d\n", dqm->sched_policy); @@ -2005,7 +2179,7 @@ void device_queue_manager_uninit(struct device_queue_manager *dqm) kfree(dqm); } -int kfd_process_vm_fault(struct device_queue_manager *dqm, u32 pasid) +int kfd_dqm_evict_pasid(struct device_queue_manager *dqm, u32 pasid) { struct kfd_process_device *pdd; struct kfd_process *p = kfd_lookup_process_by_pasid(pasid); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h index e145e4deb53a..05a9c17daa3e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h @@ -39,6 +39,41 @@ struct device_process_node { struct list_head list; }; +union SQ_CMD_BITS { + struct { + uint32_t cmd:3; + uint32_t:1; + uint32_t mode:3; + uint32_t check_vmid:1; + uint32_t trap_id:3; + uint32_t:5; + uint32_t wave_id:4; + uint32_t simd_id:2; + uint32_t:2; + uint32_t queue_id:3; + uint32_t:1; + uint32_t vm_id:4; + } bitfields, bits; + uint32_t u32All; + signed int i32All; + float f32All; +}; + +union GRBM_GFX_INDEX_BITS { + struct { + uint32_t instance_index:8; + uint32_t sh_index:8; + uint32_t se_index:8; + uint32_t:5; + uint32_t sh_broadcast_writes:1; + uint32_t instance_broadcast_writes:1; + uint32_t se_broadcast_writes:1; + } bitfields, bits; + uint32_t u32All; + signed int i32All; + float f32All; +}; + /** * struct device_queue_manager_ops * @@ -83,12 +118,18 @@ struct device_process_node { * control stack, if kept in the MQD, to the given userspace address. * * @reset_queues: reset queues which consume RAS poison + * @get_queue_checkpoint_info: Retrieves queue size information for CRIU checkpoint. + * + * @checkpoint_mqd: checkpoint queue MQD contents for CRIU. */ struct device_queue_manager_ops { int (*create_queue)(struct device_queue_manager *dqm, struct queue *q, - struct qcm_process_device *qpd); + struct qcm_process_device *qpd, + const struct kfd_criu_queue_priv_data *qd, + const void *restore_mqd, + const void *restore_ctl_stack); int (*destroy_queue)(struct device_queue_manager *dqm, struct qcm_process_device *qpd, @@ -139,6 +180,14 @@ struct device_queue_manager_ops { int (*reset_queues)(struct device_queue_manager *dqm, uint16_t pasid); + void (*get_queue_checkpoint_info)(struct device_queue_manager *dqm, + const struct queue *q, u32 *mqd_size, + u32 *ctl_stack_size); + + int (*checkpoint_mqd)(struct device_queue_manager *dqm, + const struct queue *q, + void *mqd, + void *ctl_stack); }; struct device_queue_manager_asic_ops { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index afe72dd11325..b5eda1e04f34 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -55,7 +55,6 @@ struct kfd_signal_page { bool need_to_free_pages; }; - static uint64_t *page_slots(struct kfd_signal_page *page) { return page->kernel_address; @@ -92,7 +91,8 @@ fail_alloc_signal_store: } static int allocate_event_notification_slot(struct kfd_process *p, - struct kfd_event *ev) + struct kfd_event *ev, + const int *restore_id) { int id; @@ -104,14 +104,19 @@ static int allocate_event_notification_slot(struct kfd_process *p, p->signal_mapped_size = 256*8; } - /* - * Compatibility with old user mode: Only use signal slots - * user mode has mapped, may be less than - * KFD_SIGNAL_EVENT_LIMIT. This also allows future increase - * of the event limit without breaking user mode. - */ - id = idr_alloc(&p->event_idr, ev, 0, p->signal_mapped_size / 8, - GFP_KERNEL); + if (restore_id) { + id = idr_alloc(&p->event_idr, ev, *restore_id, *restore_id + 1, + GFP_KERNEL); + } else { + /* + * Compatibility with old user mode: Only use signal slots + * user mode has mapped, may be less than + * KFD_SIGNAL_EVENT_LIMIT. This also allows future increase + * of the event limit without breaking user mode. + */ + id = idr_alloc(&p->event_idr, ev, 0, p->signal_mapped_size / 8, + GFP_KERNEL); + } if (id < 0) return id; @@ -178,9 +183,8 @@ static struct kfd_event *lookup_signaled_event_by_partial_id( return ev; } -static int create_signal_event(struct file *devkfd, - struct kfd_process *p, - struct kfd_event *ev) +static int create_signal_event(struct file *devkfd, struct kfd_process *p, + struct kfd_event *ev, const int *restore_id) { int ret; @@ -193,7 +197,7 @@ static int create_signal_event(struct file *devkfd, return -ENOSPC; } - ret = allocate_event_notification_slot(p, ev); + ret = allocate_event_notification_slot(p, ev, restore_id); if (ret) { pr_warn("Signal event wasn't created because out of kernel memory\n"); return ret; @@ -209,16 +213,22 @@ static int create_signal_event(struct file *devkfd, return 0; } -static int create_other_event(struct kfd_process *p, struct kfd_event *ev) +static int create_other_event(struct kfd_process *p, struct kfd_event *ev, const int *restore_id) { - /* Cast KFD_LAST_NONSIGNAL_EVENT to uint32_t. This allows an - * intentional integer overflow to -1 without a compiler - * warning. idr_alloc treats a negative value as "maximum - * signed integer". - */ - int id = idr_alloc(&p->event_idr, ev, KFD_FIRST_NONSIGNAL_EVENT_ID, - (uint32_t)KFD_LAST_NONSIGNAL_EVENT_ID + 1, - GFP_KERNEL); + int id; + + if (restore_id) + id = idr_alloc(&p->event_idr, ev, *restore_id, *restore_id + 1, + GFP_KERNEL); + else + /* Cast KFD_LAST_NONSIGNAL_EVENT to uint32_t. This allows an + * intentional integer overflow to -1 without a compiler + * warning. idr_alloc treats a negative value as "maximum + * signed integer". + */ + id = idr_alloc(&p->event_idr, ev, KFD_FIRST_NONSIGNAL_EVENT_ID, + (uint32_t)KFD_LAST_NONSIGNAL_EVENT_ID + 1, + GFP_KERNEL); if (id < 0) return id; @@ -295,8 +305,8 @@ static bool event_can_be_cpu_signaled(const struct kfd_event *ev) return ev->type == KFD_EVENT_TYPE_SIGNAL; } -int kfd_event_page_set(struct kfd_process *p, void *kernel_address, - uint64_t size) +static int kfd_event_page_set(struct kfd_process *p, void *kernel_address, + uint64_t size, uint64_t user_handle) { struct kfd_signal_page *page; @@ -315,10 +325,57 @@ int kfd_event_page_set(struct kfd_process *p, void *kernel_address, p->signal_page = page; p->signal_mapped_size = size; - + p->signal_handle = user_handle; return 0; } +int kfd_kmap_event_page(struct kfd_process *p, uint64_t event_page_offset) +{ + struct kfd_dev *kfd; + struct kfd_process_device *pdd; + void *mem, *kern_addr; + uint64_t size; + int err = 0; + + if (p->signal_page) { + pr_err("Event page is already set\n"); + return -EINVAL; + } + + pdd = kfd_process_device_data_by_id(p, GET_GPU_ID(event_page_offset)); + if (!pdd) { + pr_err("Getting device by id failed in %s\n", __func__); + return -EINVAL; + } + kfd = pdd->dev; + + pdd = kfd_bind_process_to_device(kfd, p); + if (IS_ERR(pdd)) + return PTR_ERR(pdd); + + mem = kfd_process_device_translate_handle(pdd, + GET_IDR_HANDLE(event_page_offset)); + if (!mem) { + pr_err("Can't find BO, offset is 0x%llx\n", event_page_offset); + return -EINVAL; + } + + err = amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel(kfd->adev, + mem, &kern_addr, &size); + if (err) { + pr_err("Failed to map event page to kernel\n"); + return err; + } + + err = kfd_event_page_set(p, kern_addr, size, event_page_offset); + if (err) { + pr_err("Failed to set event page\n"); + amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(kfd->adev, mem); + return err; + } + return err; +} + int kfd_event_create(struct file *devkfd, struct kfd_process *p, uint32_t event_type, bool auto_reset, uint32_t node_id, uint32_t *event_id, uint32_t *event_trigger_data, @@ -343,14 +400,14 @@ int kfd_event_create(struct file *devkfd, struct kfd_process *p, switch (event_type) { case KFD_EVENT_TYPE_SIGNAL: case KFD_EVENT_TYPE_DEBUG: - ret = create_signal_event(devkfd, p, ev); + ret = create_signal_event(devkfd, p, ev, NULL); if (!ret) { *event_page_offset = KFD_MMAP_TYPE_EVENTS; *event_slot_index = ev->event_id; } break; default: - ret = create_other_event(p, ev); + ret = create_other_event(p, ev, NULL); break; } @@ -366,6 +423,166 @@ int kfd_event_create(struct file *devkfd, struct kfd_process *p, return ret; } +int kfd_criu_restore_event(struct file *devkfd, + struct kfd_process *p, + uint8_t __user *user_priv_ptr, + uint64_t *priv_data_offset, + uint64_t max_priv_data_size) +{ + struct kfd_criu_event_priv_data *ev_priv; + struct kfd_event *ev = NULL; + int ret = 0; + + ev_priv = kmalloc(sizeof(*ev_priv), GFP_KERNEL); + if (!ev_priv) + return -ENOMEM; + + ev = kzalloc(sizeof(*ev), GFP_KERNEL); + if (!ev) { + ret = -ENOMEM; + goto exit; + } + + if (*priv_data_offset + sizeof(*ev_priv) > max_priv_data_size) { + ret = -EINVAL; + goto exit; + } + + ret = copy_from_user(ev_priv, user_priv_ptr + *priv_data_offset, sizeof(*ev_priv)); + if (ret) { + ret = -EFAULT; + goto exit; + } + *priv_data_offset += sizeof(*ev_priv); + + if (ev_priv->user_handle) { + ret = kfd_kmap_event_page(p, ev_priv->user_handle); + if (ret) + goto exit; + } + + ev->type = ev_priv->type; + ev->auto_reset = ev_priv->auto_reset; + ev->signaled = ev_priv->signaled; + + init_waitqueue_head(&ev->wq); + + mutex_lock(&p->event_mutex); + switch (ev->type) { + case KFD_EVENT_TYPE_SIGNAL: + case KFD_EVENT_TYPE_DEBUG: + ret = create_signal_event(devkfd, p, ev, &ev_priv->event_id); + break; + case KFD_EVENT_TYPE_MEMORY: + memcpy(&ev->memory_exception_data, + &ev_priv->memory_exception_data, + sizeof(struct kfd_hsa_memory_exception_data)); + + ret = create_other_event(p, ev, &ev_priv->event_id); + break; + case KFD_EVENT_TYPE_HW_EXCEPTION: + memcpy(&ev->hw_exception_data, + &ev_priv->hw_exception_data, + sizeof(struct kfd_hsa_hw_exception_data)); + + ret = create_other_event(p, ev, &ev_priv->event_id); + break; + } + +exit: + if (ret) + kfree(ev); + + kfree(ev_priv); + + mutex_unlock(&p->event_mutex); + + return ret; +} + +int kfd_criu_checkpoint_events(struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_data_offset) +{ + struct kfd_criu_event_priv_data *ev_privs; + int i = 0; + int ret = 0; + struct kfd_event *ev; + uint32_t ev_id; + + uint32_t num_events = kfd_get_num_events(p); + + if (!num_events) + return 0; + + ev_privs = kvzalloc(num_events * sizeof(*ev_privs), GFP_KERNEL); + if (!ev_privs) + return -ENOMEM; + + + idr_for_each_entry(&p->event_idr, ev, ev_id) { + struct kfd_criu_event_priv_data *ev_priv; + + /* + * Currently, all events have same size of private_data, but the current ioctl's + * and CRIU plugin supports private_data of variable sizes + */ + ev_priv = &ev_privs[i]; + + ev_priv->object_type = KFD_CRIU_OBJECT_TYPE_EVENT; + + /* We store the user_handle with the first event */ + if (i == 0 && p->signal_page) + ev_priv->user_handle = p->signal_handle; + + ev_priv->event_id = ev->event_id; + ev_priv->auto_reset = ev->auto_reset; + ev_priv->type = ev->type; + ev_priv->signaled = ev->signaled; + + if (ev_priv->type == KFD_EVENT_TYPE_MEMORY) + memcpy(&ev_priv->memory_exception_data, + &ev->memory_exception_data, + sizeof(struct kfd_hsa_memory_exception_data)); + else if (ev_priv->type == KFD_EVENT_TYPE_HW_EXCEPTION) + memcpy(&ev_priv->hw_exception_data, + &ev->hw_exception_data, + sizeof(struct kfd_hsa_hw_exception_data)); + + pr_debug("Checkpointed event[%d] id = 0x%08x auto_reset = %x type = %x signaled = %x\n", + i, + ev_priv->event_id, + ev_priv->auto_reset, + ev_priv->type, + ev_priv->signaled); + i++; + } + + ret = copy_to_user(user_priv_data + *priv_data_offset, + ev_privs, num_events * sizeof(*ev_privs)); + if (ret) { + pr_err("Failed to copy events priv to user\n"); + ret = -EFAULT; + } + + *priv_data_offset += num_events * sizeof(*ev_privs); + + kvfree(ev_privs); + return ret; +} + +int kfd_get_num_events(struct kfd_process *p) +{ + struct kfd_event *ev; + uint32_t id; + u32 num_events = 0; + + idr_for_each_entry(&p->event_idr, ev, id) + num_events++; + + return num_events; +} + /* Assumes that p is current. */ int kfd_event_destroy(struct kfd_process *p, uint32_t event_id) { @@ -878,6 +1095,7 @@ void kfd_signal_iommu_event(struct kfd_dev *dev, u32 pasid, { struct kfd_hsa_memory_exception_data memory_exception_data; struct vm_area_struct *vma; + int user_gpu_id; /* * Because we are called from arbitrary context (workqueue) as opposed @@ -899,12 +1117,17 @@ void kfd_signal_iommu_event(struct kfd_dev *dev, u32 pasid, return; /* Process is exiting */ } + user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id); + if (unlikely(user_gpu_id == -EINVAL)) { + WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id); + return; + } memset(&memory_exception_data, 0, sizeof(memory_exception_data)); mmap_read_lock(mm); vma = find_vma(mm, address); - memory_exception_data.gpu_id = dev->id; + memory_exception_data.gpu_id = user_gpu_id; memory_exception_data.va = address; /* Set failure reason */ memory_exception_data.failure.NotPresent = 1; @@ -980,11 +1203,19 @@ void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid, uint32_t id; struct kfd_process *p = kfd_lookup_process_by_pasid(pasid); struct kfd_hsa_memory_exception_data memory_exception_data; + int user_gpu_id; if (!p) return; /* Presumably process exited. */ + + user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id); + if (unlikely(user_gpu_id == -EINVAL)) { + WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id); + return; + } + memset(&memory_exception_data, 0, sizeof(memory_exception_data)); - memory_exception_data.gpu_id = dev->id; + memory_exception_data.gpu_id = user_gpu_id; memory_exception_data.failure.imprecise = true; /* Set failure reason */ if (info) { @@ -1024,27 +1255,34 @@ void kfd_signal_reset_event(struct kfd_dev *dev) /* Whole gpu reset caused by GPU hang and memory is lost */ memset(&hw_exception_data, 0, sizeof(hw_exception_data)); - hw_exception_data.gpu_id = dev->id; hw_exception_data.memory_lost = 1; hw_exception_data.reset_cause = reset_cause; memset(&memory_exception_data, 0, sizeof(memory_exception_data)); memory_exception_data.ErrorType = KFD_MEM_ERR_SRAM_ECC; - memory_exception_data.gpu_id = dev->id; memory_exception_data.failure.imprecise = true; idx = srcu_read_lock(&kfd_processes_srcu); hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) { + int user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id); + + if (unlikely(user_gpu_id == -EINVAL)) { + WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id); + continue; + } + mutex_lock(&p->event_mutex); id = KFD_FIRST_NONSIGNAL_EVENT_ID; idr_for_each_entry_continue(&p->event_idr, ev, id) { if (ev->type == KFD_EVENT_TYPE_HW_EXCEPTION) { ev->hw_exception_data = hw_exception_data; + ev->hw_exception_data.gpu_id = user_gpu_id; set_event(ev); } if (ev->type == KFD_EVENT_TYPE_MEMORY && reset_cause == KFD_HW_EXCEPTION_ECC) { ev->memory_exception_data = memory_exception_data; + ev->memory_exception_data.gpu_id = user_gpu_id; set_event(ev); } } @@ -1060,18 +1298,25 @@ void kfd_signal_poison_consumed_event(struct kfd_dev *dev, u32 pasid) struct kfd_hsa_hw_exception_data hw_exception_data; struct kfd_event *ev; uint32_t id = KFD_FIRST_NONSIGNAL_EVENT_ID; + int user_gpu_id; if (!p) return; /* Presumably process exited. */ + user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id); + if (unlikely(user_gpu_id == -EINVAL)) { + WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id); + return; + } + memset(&hw_exception_data, 0, sizeof(hw_exception_data)); - hw_exception_data.gpu_id = dev->id; + hw_exception_data.gpu_id = user_gpu_id; hw_exception_data.memory_lost = 1; hw_exception_data.reset_cause = KFD_HW_EXCEPTION_ECC; memset(&memory_exception_data, 0, sizeof(memory_exception_data)); memory_exception_data.ErrorType = KFD_MEM_ERR_POISON_CONSUMED; - memory_exception_data.gpu_id = dev->id; + memory_exception_data.gpu_id = user_gpu_id; memory_exception_data.failure.imprecise = true; mutex_lock(&p->event_mutex); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c index e8bc28009c22..68ee923a440b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c @@ -109,8 +109,7 @@ static void event_interrupt_poison_consumption(struct kfd_dev *dev, switch (source_id) { case SOC15_INTSRC_SQ_INTERRUPT_MSG: - if (dev->dqm->ops.reset_queues) - ret = dev->dqm->ops.reset_queues(dev->dqm, pasid); + kfd_dqm_evict_pasid(dev->dqm, pasid); break; case SOC15_INTSRC_SDMA_ECC: default: @@ -308,7 +307,7 @@ static void event_interrupt_wq_v9(struct kfd_dev *dev, info.prot_write = ring_id & 0x20; kfd_smi_event_update_vmfault(dev, pasid); - kfd_process_vm_fault(dev->dqm, pasid); + kfd_dqm_evict_pasid(dev->dqm, pasid); kfd_signal_vm_fault_event(dev, pasid, &info); } } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c index 66ad8d0b8f7f..fe62407dacb2 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c @@ -30,7 +30,6 @@ #include <linux/pci.h> #include <linux/amd-iommu.h> #include "kfd_priv.h" -#include "kfd_dbgmgr.h" #include "kfd_topology.h" #include "kfd_iommu.h" @@ -163,17 +162,6 @@ static void iommu_pasid_shutdown_callback(struct pci_dev *pdev, u32 pasid) pr_debug("Unbinding process 0x%x from IOMMU\n", pasid); - mutex_lock(kfd_get_dbgmgr_mutex()); - - if (dev->dbgmgr && dev->dbgmgr->pasid == p->pasid) { - if (!kfd_dbgmgr_unregister(dev->dbgmgr, p)) { - kfd_dbgmgr_destroy(dev->dbgmgr); - dev->dbgmgr = NULL; - } - } - - mutex_unlock(kfd_get_dbgmgr_mutex()); - mutex_lock(&p->mutex); pdd = kfd_get_process_device_data(dev, p); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c index ed5385137f48..7dcafd337bb2 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -36,7 +36,7 @@ #ifdef dev_fmt #undef dev_fmt #endif -#define dev_fmt(fmt) "kfd_migrate: %s: " fmt, __func__ +#define dev_fmt(fmt) "kfd_migrate: " fmt static uint64_t svm_migrate_direct_mapping_addr(struct amdgpu_device *adev, uint64_t addr) @@ -86,10 +86,7 @@ svm_migrate_gart_map(struct amdgpu_ring *ring, uint64_t npages, cpu_addr = &job->ibs[0].ptr[num_dw]; - r = amdgpu_gart_map(adev, 0, npages, addr, pte_flags, cpu_addr); - if (r) - goto error_free; - + amdgpu_gart_map(adev, 0, npages, addr, pte_flags, cpu_addr); r = amdgpu_job_submit(job, &adev->mman.entity, AMDGPU_FENCE_OWNER_UNDEFINED, &fence); if (r) @@ -315,7 +312,7 @@ svm_migrate_copy_to_vram(struct amdgpu_device *adev, struct svm_range *prange, r = svm_range_vram_node_new(adev, prange, true); if (r) { - dev_err(adev->dev, "fail %d to alloc vram\n", r); + dev_dbg(adev->dev, "fail %d to alloc vram\n", r); goto out; } @@ -334,7 +331,8 @@ svm_migrate_copy_to_vram(struct amdgpu_device *adev, struct svm_range *prange, DMA_TO_DEVICE); r = dma_mapping_error(dev, src[i]); if (r) { - dev_err(adev->dev, "fail %d dma_map_page\n", r); + dev_err(adev->dev, "%s: fail %d dma_map_page\n", + __func__, r); goto out_free_vram_pages; } } else { @@ -435,8 +433,8 @@ svm_migrate_vma_to_vram(struct amdgpu_device *adev, struct svm_range *prange, r = migrate_vma_setup(&migrate); if (r) { - dev_err(adev->dev, "vma setup fail %d range [0x%lx 0x%lx]\n", r, - prange->start, prange->last); + dev_err(adev->dev, "%s: vma setup fail %d range [0x%lx 0x%lx]\n", + __func__, r, prange->start, prange->last); goto out_free; } @@ -614,7 +612,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange, dst[i] = dma_map_page(dev, dpage, 0, PAGE_SIZE, DMA_FROM_DEVICE); r = dma_mapping_error(dev, dst[i]); if (r) { - dev_err(adev->dev, "fail %d dma_map_page\n", r); + dev_err(adev->dev, "%s: fail %d dma_map_page\n", __func__, r); goto out_oom; } @@ -674,8 +672,8 @@ svm_migrate_vma_to_ram(struct amdgpu_device *adev, struct svm_range *prange, r = migrate_vma_setup(&migrate); if (r) { - dev_err(adev->dev, "vma setup fail %d range [0x%lx 0x%lx]\n", r, - prange->start, prange->last); + dev_err(adev->dev, "%s: vma setup fail %d range [0x%lx 0x%lx]\n", + __func__, r, prange->start, prange->last); goto out_free; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c index e2825ad4d699..9b7544ddced7 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c @@ -173,3 +173,66 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, } } } + +int kfd_hiq_load_mqd_kiq(struct mqd_manager *mm, void *mqd, + uint32_t pipe_id, uint32_t queue_id, + struct queue_properties *p, struct mm_struct *mms) +{ + return mm->dev->kfd2kgd->hiq_mqd_load(mm->dev->adev, mqd, pipe_id, + queue_id, p->doorbell_off); +} + +int kfd_destroy_mqd_cp(struct mqd_manager *mm, void *mqd, + enum kfd_preempt_type type, unsigned int timeout, + uint32_t pipe_id, uint32_t queue_id) +{ + return mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, mqd, type, timeout, + pipe_id, queue_id); +} + +void kfd_free_mqd_cp(struct mqd_manager *mm, void *mqd, + struct kfd_mem_obj *mqd_mem_obj) +{ + if (mqd_mem_obj->gtt_mem) { + amdgpu_amdkfd_free_gtt_mem(mm->dev->adev, mqd_mem_obj->gtt_mem); + kfree(mqd_mem_obj); + } else { + kfd_gtt_sa_free(mm->dev, mqd_mem_obj); + } +} + +bool kfd_is_occupied_cp(struct mqd_manager *mm, void *mqd, + uint64_t queue_address, uint32_t pipe_id, + uint32_t queue_id) +{ + return mm->dev->kfd2kgd->hqd_is_occupied(mm->dev->adev, queue_address, + pipe_id, queue_id); +} + +int kfd_load_mqd_sdma(struct mqd_manager *mm, void *mqd, + uint32_t pipe_id, uint32_t queue_id, + struct queue_properties *p, struct mm_struct *mms) +{ + return mm->dev->kfd2kgd->hqd_sdma_load(mm->dev->adev, mqd, + (uint32_t __user *)p->write_ptr, + mms); +} + +/* + * preempt type here is ignored because there is only one way + * to preempt sdma queue + */ +int kfd_destroy_mqd_sdma(struct mqd_manager *mm, void *mqd, + enum kfd_preempt_type type, + unsigned int timeout, uint32_t pipe_id, + uint32_t queue_id) +{ + return mm->dev->kfd2kgd->hqd_sdma_destroy(mm->dev->adev, mqd, timeout); +} + +bool kfd_is_occupied_sdma(struct mqd_manager *mm, void *mqd, + uint64_t queue_address, uint32_t pipe_id, + uint32_t queue_id) +{ + return mm->dev->kfd2kgd->hqd_sdma_is_occupied(mm->dev->adev, mqd); +} diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h index 965e17c5dbb4..d81ae0699300 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h @@ -100,6 +100,20 @@ struct mqd_manager { u32 *ctl_stack_used_size, u32 *save_area_used_size); + void (*get_checkpoint_info)(struct mqd_manager *mm, void *mqd, uint32_t *ctl_stack_size); + + void (*checkpoint_mqd)(struct mqd_manager *mm, + void *mqd, + void *mqd_dst, + void *ctl_stack_dst); + + void (*restore_mqd)(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *p, + const void *mqd_src, + const void *ctl_stack_src, + const u32 ctl_stack_size); + #if defined(CONFIG_DEBUG_FS) int (*debugfs_show_mqd)(struct seq_file *m, void *data); #endif @@ -122,4 +136,31 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, const uint32_t *cu_mask, uint32_t cu_mask_count, uint32_t *se_mask); +int kfd_hiq_load_mqd_kiq(struct mqd_manager *mm, void *mqd, + uint32_t pipe_id, uint32_t queue_id, + struct queue_properties *p, struct mm_struct *mms); + +int kfd_destroy_mqd_cp(struct mqd_manager *mm, void *mqd, + enum kfd_preempt_type type, unsigned int timeout, + uint32_t pipe_id, uint32_t queue_id); + +void kfd_free_mqd_cp(struct mqd_manager *mm, void *mqd, + struct kfd_mem_obj *mqd_mem_obj); + +bool kfd_is_occupied_cp(struct mqd_manager *mm, void *mqd, + uint64_t queue_address, uint32_t pipe_id, + uint32_t queue_id); + +int kfd_load_mqd_sdma(struct mqd_manager *mm, void *mqd, + uint32_t pipe_id, uint32_t queue_id, + struct queue_properties *p, struct mm_struct *mms); + +int kfd_destroy_mqd_sdma(struct mqd_manager *mm, void *mqd, + enum kfd_preempt_type type, unsigned int timeout, + uint32_t pipe_id, uint32_t queue_id); + +bool kfd_is_occupied_sdma(struct mqd_manager *mm, void *mqd, + uint64_t queue_address, uint32_t pipe_id, + uint32_t queue_id); + #endif /* KFD_MQD_MANAGER_H_ */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c index e9a8e21e144e..06aae5df6a37 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c @@ -156,13 +156,6 @@ static void init_mqd_sdma(struct mqd_manager *mm, void **mqd, mm->update_mqd(mm, m, q, NULL); } -static void free_mqd(struct mqd_manager *mm, void *mqd, - struct kfd_mem_obj *mqd_mem_obj) -{ - kfd_gtt_sa_free(mm->dev, mqd_mem_obj); -} - - static int load_mqd(struct mqd_manager *mm, void *mqd, uint32_t pipe_id, uint32_t queue_id, struct queue_properties *p, struct mm_struct *mms) @@ -176,15 +169,6 @@ static int load_mqd(struct mqd_manager *mm, void *mqd, uint32_t pipe_id, wptr_shift, wptr_mask, mms); } -static int load_mqd_sdma(struct mqd_manager *mm, void *mqd, - uint32_t pipe_id, uint32_t queue_id, - struct queue_properties *p, struct mm_struct *mms) -{ - return mm->dev->kfd2kgd->hqd_sdma_load(mm->dev->adev, mqd, - (uint32_t __user *)p->write_ptr, - mms); -} - static void __update_mqd(struct mqd_manager *mm, void *mqd, struct queue_properties *q, struct mqd_update_info *minfo, unsigned int atc_bit) @@ -271,42 +255,75 @@ static void update_mqd_sdma(struct mqd_manager *mm, void *mqd, q->is_active = QUEUE_IS_ACTIVE(*q); } -static int destroy_mqd(struct mqd_manager *mm, void *mqd, - enum kfd_preempt_type type, - unsigned int timeout, uint32_t pipe_id, - uint32_t queue_id) +static void checkpoint_mqd(struct mqd_manager *mm, void *mqd, void *mqd_dst, void *ctl_stack_dst) { - return mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, mqd, type, timeout, - pipe_id, queue_id); + struct cik_mqd *m; + + m = get_mqd(mqd); + + memcpy(mqd_dst, m, sizeof(struct cik_mqd)); } -/* - * preempt type here is ignored because there is only one way - * to preempt sdma queue - */ -static int destroy_mqd_sdma(struct mqd_manager *mm, void *mqd, - enum kfd_preempt_type type, - unsigned int timeout, uint32_t pipe_id, - uint32_t queue_id) +static void restore_mqd(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, const u32 ctl_stack_size) { - return mm->dev->kfd2kgd->hqd_sdma_destroy(mm->dev->adev, mqd, timeout); + uint64_t addr; + struct cik_mqd *m; + + m = (struct cik_mqd *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + m->cp_hqd_pq_doorbell_control = DOORBELL_OFFSET(qp->doorbell_off); + + pr_debug("cp_hqd_pq_doorbell_control 0x%x\n", + m->cp_hqd_pq_doorbell_control); + + qp->is_active = 0; } -static bool is_occupied(struct mqd_manager *mm, void *mqd, - uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) +static void checkpoint_mqd_sdma(struct mqd_manager *mm, + void *mqd, + void *mqd_dst, + void *ctl_stack_dst) { + struct cik_sdma_rlc_registers *m; - return mm->dev->kfd2kgd->hqd_is_occupied(mm->dev->adev, queue_address, - pipe_id, queue_id); + m = get_sdma_mqd(mqd); + memcpy(mqd_dst, m, sizeof(struct cik_sdma_rlc_registers)); } -static bool is_occupied_sdma(struct mqd_manager *mm, void *mqd, - uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) +static void restore_mqd_sdma(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, const u32 ctl_stack_size) { - return mm->dev->kfd2kgd->hqd_sdma_is_occupied(mm->dev->adev, mqd); + uint64_t addr; + struct cik_sdma_rlc_registers *m; + + m = (struct cik_sdma_rlc_registers *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + m->sdma_rlc_doorbell = + qp->doorbell_off << SDMA0_RLC0_DOORBELL__OFFSET__SHIFT; + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + qp->is_active = 0; } /* @@ -389,11 +406,13 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, case KFD_MQD_TYPE_CP: mqd->allocate_mqd = allocate_mqd; mqd->init_mqd = init_mqd; - mqd->free_mqd = free_mqd; + mqd->free_mqd = kfd_free_mqd_cp; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; + mqd->checkpoint_mqd = checkpoint_mqd; + mqd->restore_mqd = restore_mqd; mqd->mqd_size = sizeof(struct cik_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -405,8 +424,8 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, mqd->free_mqd = free_mqd_hiq_sdma; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd_hiq; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct cik_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -416,11 +435,11 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, case KFD_MQD_TYPE_DIQ: mqd->allocate_mqd = allocate_mqd; mqd->init_mqd = init_mqd_hiq; - mqd->free_mqd = free_mqd; + mqd->free_mqd = kfd_free_mqd_cp; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd_hiq; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct cik_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -430,10 +449,12 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, mqd->allocate_mqd = allocate_sdma_mqd; mqd->init_mqd = init_mqd_sdma; mqd->free_mqd = free_mqd_hiq_sdma; - mqd->load_mqd = load_mqd_sdma; + mqd->load_mqd = kfd_load_mqd_sdma; mqd->update_mqd = update_mqd_sdma; - mqd->destroy_mqd = destroy_mqd_sdma; - mqd->is_occupied = is_occupied_sdma; + mqd->destroy_mqd = kfd_destroy_mqd_sdma; + mqd->is_occupied = kfd_is_occupied_sdma; + mqd->checkpoint_mqd = checkpoint_mqd_sdma; + mqd->restore_mqd = restore_mqd_sdma; mqd->mqd_size = sizeof(struct cik_sdma_rlc_registers); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c index d74d8a6ac27a..88228009ded4 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c @@ -154,14 +154,6 @@ static int load_mqd(struct mqd_manager *mm, void *mqd, return r; } -static int hiq_load_mqd_kiq(struct mqd_manager *mm, void *mqd, - uint32_t pipe_id, uint32_t queue_id, - struct queue_properties *p, struct mm_struct *mms) -{ - return mm->dev->kfd2kgd->hiq_mqd_load(mm->dev->adev, mqd, pipe_id, - queue_id, p->doorbell_off); -} - static void update_mqd(struct mqd_manager *mm, void *mqd, struct queue_properties *q, struct mqd_update_info *minfo) @@ -233,31 +225,6 @@ static uint32_t read_doorbell_id(void *mqd) return m->queue_doorbell_id0; } -static int destroy_mqd(struct mqd_manager *mm, void *mqd, - enum kfd_preempt_type type, - unsigned int timeout, uint32_t pipe_id, - uint32_t queue_id) -{ - return mm->dev->kfd2kgd->hqd_destroy - (mm->dev->adev, mqd, type, timeout, - pipe_id, queue_id); -} - -static void free_mqd(struct mqd_manager *mm, void *mqd, - struct kfd_mem_obj *mqd_mem_obj) -{ - kfd_gtt_sa_free(mm->dev, mqd_mem_obj); -} - -static bool is_occupied(struct mqd_manager *mm, void *mqd, - uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) -{ - return mm->dev->kfd2kgd->hqd_is_occupied( - mm->dev->adev, queue_address, - pipe_id, queue_id); -} - static int get_wave_state(struct mqd_manager *mm, void *mqd, void __user *ctl_stack, u32 *ctl_stack_used_size, @@ -285,6 +252,42 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd, return 0; } +static void checkpoint_mqd(struct mqd_manager *mm, void *mqd, void *mqd_dst, void *ctl_stack_dst) +{ + struct v10_compute_mqd *m; + + m = get_mqd(mqd); + + memcpy(mqd_dst, m, sizeof(struct v10_compute_mqd)); +} + +static void restore_mqd(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, const u32 ctl_stack_size) +{ + uint64_t addr; + struct v10_compute_mqd *m; + + m = (struct v10_compute_mqd *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + m->cp_hqd_pq_doorbell_control = + qp->doorbell_off << + CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; + pr_debug("cp_hqd_pq_doorbell_control 0x%x\n", + m->cp_hqd_pq_doorbell_control); + + qp->is_active = 0; +} + static void init_mqd_hiq(struct mqd_manager *mm, void **mqd, struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, struct queue_properties *q) @@ -316,15 +319,6 @@ static void init_mqd_sdma(struct mqd_manager *mm, void **mqd, mm->update_mqd(mm, m, q, NULL); } -static int load_mqd_sdma(struct mqd_manager *mm, void *mqd, - uint32_t pipe_id, uint32_t queue_id, - struct queue_properties *p, struct mm_struct *mms) -{ - return mm->dev->kfd2kgd->hqd_sdma_load(mm->dev->adev, mqd, - (uint32_t __user *)p->write_ptr, - mms); -} - #define SDMA_RLC_DUMMY_DEFAULT 0xf static void update_mqd_sdma(struct mqd_manager *mm, void *mqd, @@ -354,23 +348,41 @@ static void update_mqd_sdma(struct mqd_manager *mm, void *mqd, q->is_active = QUEUE_IS_ACTIVE(*q); } -/* - * * preempt type here is ignored because there is only one way - * * to preempt sdma queue - */ -static int destroy_mqd_sdma(struct mqd_manager *mm, void *mqd, - enum kfd_preempt_type type, - unsigned int timeout, uint32_t pipe_id, - uint32_t queue_id) +static void checkpoint_mqd_sdma(struct mqd_manager *mm, + void *mqd, + void *mqd_dst, + void *ctl_stack_dst) { - return mm->dev->kfd2kgd->hqd_sdma_destroy(mm->dev->adev, mqd, timeout); + struct v10_sdma_mqd *m; + + m = get_sdma_mqd(mqd); + + memcpy(mqd_dst, m, sizeof(struct v10_sdma_mqd)); } -static bool is_occupied_sdma(struct mqd_manager *mm, void *mqd, - uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) +static void restore_mqd_sdma(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, + const u32 ctl_stack_size) { - return mm->dev->kfd2kgd->hqd_sdma_is_occupied(mm->dev->adev, mqd); + uint64_t addr; + struct v10_sdma_mqd *m; + + m = (struct v10_sdma_mqd *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + m->sdmax_rlcx_doorbell_offset = + qp->doorbell_off << SDMA0_RLC0_DOORBELL_OFFSET__OFFSET__SHIFT; + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + qp->is_active = 0; } #if defined(CONFIG_DEBUG_FS) @@ -410,13 +422,15 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, pr_debug("%s@%i\n", __func__, __LINE__); mqd->allocate_mqd = allocate_mqd; mqd->init_mqd = init_mqd; - mqd->free_mqd = free_mqd; + mqd->free_mqd = kfd_free_mqd_cp; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v10_compute_mqd); mqd->get_wave_state = get_wave_state; + mqd->checkpoint_mqd = checkpoint_mqd; + mqd->restore_mqd = restore_mqd; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif @@ -427,10 +441,10 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, mqd->allocate_mqd = allocate_hiq_mqd; mqd->init_mqd = init_mqd_hiq; mqd->free_mqd = free_mqd_hiq_sdma; - mqd->load_mqd = hiq_load_mqd_kiq; + mqd->load_mqd = kfd_hiq_load_mqd_kiq; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v10_compute_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -441,11 +455,11 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, case KFD_MQD_TYPE_DIQ: mqd->allocate_mqd = allocate_mqd; mqd->init_mqd = init_mqd_hiq; - mqd->free_mqd = free_mqd; + mqd->free_mqd = kfd_free_mqd_cp; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v10_compute_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -456,10 +470,12 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, mqd->allocate_mqd = allocate_sdma_mqd; mqd->init_mqd = init_mqd_sdma; mqd->free_mqd = free_mqd_hiq_sdma; - mqd->load_mqd = load_mqd_sdma; + mqd->load_mqd = kfd_load_mqd_sdma; mqd->update_mqd = update_mqd_sdma; - mqd->destroy_mqd = destroy_mqd_sdma; - mqd->is_occupied = is_occupied_sdma; + mqd->destroy_mqd = kfd_destroy_mqd_sdma; + mqd->is_occupied = kfd_is_occupied_sdma; + mqd->checkpoint_mqd = checkpoint_mqd_sdma; + mqd->restore_mqd = restore_mqd_sdma; mqd->mqd_size = sizeof(struct v10_sdma_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c index 326eb2285029..dff58ee53557 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c @@ -204,14 +204,6 @@ static int load_mqd(struct mqd_manager *mm, void *mqd, wptr_shift, 0, mms); } -static int hiq_load_mqd_kiq(struct mqd_manager *mm, void *mqd, - uint32_t pipe_id, uint32_t queue_id, - struct queue_properties *p, struct mm_struct *mms) -{ - return mm->dev->kfd2kgd->hiq_mqd_load(mm->dev->adev, mqd, pipe_id, - queue_id, p->doorbell_off); -} - static void update_mqd(struct mqd_manager *mm, void *mqd, struct queue_properties *q, struct mqd_update_info *minfo) @@ -285,38 +277,6 @@ static uint32_t read_doorbell_id(void *mqd) return m->queue_doorbell_id0; } -static int destroy_mqd(struct mqd_manager *mm, void *mqd, - enum kfd_preempt_type type, - unsigned int timeout, uint32_t pipe_id, - uint32_t queue_id) -{ - return mm->dev->kfd2kgd->hqd_destroy - (mm->dev->adev, mqd, type, timeout, - pipe_id, queue_id); -} - -static void free_mqd(struct mqd_manager *mm, void *mqd, - struct kfd_mem_obj *mqd_mem_obj) -{ - struct kfd_dev *kfd = mm->dev; - - if (mqd_mem_obj->gtt_mem) { - amdgpu_amdkfd_free_gtt_mem(kfd->adev, mqd_mem_obj->gtt_mem); - kfree(mqd_mem_obj); - } else { - kfd_gtt_sa_free(mm->dev, mqd_mem_obj); - } -} - -static bool is_occupied(struct mqd_manager *mm, void *mqd, - uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) -{ - return mm->dev->kfd2kgd->hqd_is_occupied( - mm->dev->adev, queue_address, - pipe_id, queue_id); -} - static int get_wave_state(struct mqd_manager *mm, void *mqd, void __user *ctl_stack, u32 *ctl_stack_used_size, @@ -340,6 +300,57 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd, return 0; } +static void get_checkpoint_info(struct mqd_manager *mm, void *mqd, u32 *ctl_stack_size) +{ + struct v9_mqd *m = get_mqd(mqd); + + *ctl_stack_size = m->cp_hqd_cntl_stack_size; +} + +static void checkpoint_mqd(struct mqd_manager *mm, void *mqd, void *mqd_dst, void *ctl_stack_dst) +{ + struct v9_mqd *m; + /* Control stack is located one page after MQD. */ + void *ctl_stack = (void *)((uintptr_t)mqd + PAGE_SIZE); + + m = get_mqd(mqd); + + memcpy(mqd_dst, m, sizeof(struct v9_mqd)); + memcpy(ctl_stack_dst, ctl_stack, m->cp_hqd_cntl_stack_size); +} + +static void restore_mqd(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, u32 ctl_stack_size) +{ + uint64_t addr; + struct v9_mqd *m; + void *ctl_stack; + + m = (struct v9_mqd *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + /* Control stack is located one page after MQD. */ + ctl_stack = (void *)((uintptr_t)*mqd + PAGE_SIZE); + memcpy(ctl_stack, ctl_stack_src, ctl_stack_size); + + m->cp_hqd_pq_doorbell_control = + qp->doorbell_off << + CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; + pr_debug("cp_hqd_pq_doorbell_control 0x%x\n", + m->cp_hqd_pq_doorbell_control); + + qp->is_active = 0; +} + static void init_mqd_hiq(struct mqd_manager *mm, void **mqd, struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, struct queue_properties *q) @@ -371,15 +382,6 @@ static void init_mqd_sdma(struct mqd_manager *mm, void **mqd, mm->update_mqd(mm, m, q, NULL); } -static int load_mqd_sdma(struct mqd_manager *mm, void *mqd, - uint32_t pipe_id, uint32_t queue_id, - struct queue_properties *p, struct mm_struct *mms) -{ - return mm->dev->kfd2kgd->hqd_sdma_load(mm->dev->adev, mqd, - (uint32_t __user *)p->write_ptr, - mms); -} - #define SDMA_RLC_DUMMY_DEFAULT 0xf static void update_mqd_sdma(struct mqd_manager *mm, void *mqd, @@ -409,23 +411,40 @@ static void update_mqd_sdma(struct mqd_manager *mm, void *mqd, q->is_active = QUEUE_IS_ACTIVE(*q); } -/* - * * preempt type here is ignored because there is only one way - * * to preempt sdma queue - */ -static int destroy_mqd_sdma(struct mqd_manager *mm, void *mqd, - enum kfd_preempt_type type, - unsigned int timeout, uint32_t pipe_id, - uint32_t queue_id) +static void checkpoint_mqd_sdma(struct mqd_manager *mm, + void *mqd, + void *mqd_dst, + void *ctl_stack_dst) { - return mm->dev->kfd2kgd->hqd_sdma_destroy(mm->dev->adev, mqd, timeout); + struct v9_sdma_mqd *m; + + m = get_sdma_mqd(mqd); + + memcpy(mqd_dst, m, sizeof(struct v9_sdma_mqd)); } -static bool is_occupied_sdma(struct mqd_manager *mm, void *mqd, - uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) +static void restore_mqd_sdma(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, const u32 ctl_stack_size) { - return mm->dev->kfd2kgd->hqd_sdma_is_occupied(mm->dev->adev, mqd); + uint64_t addr; + struct v9_sdma_mqd *m; + + m = (struct v9_sdma_mqd *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + m->sdmax_rlcx_doorbell_offset = + qp->doorbell_off << SDMA0_RLC0_DOORBELL_OFFSET__OFFSET__SHIFT; + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + qp->is_active = 0; } #if defined(CONFIG_DEBUG_FS) @@ -464,12 +483,15 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, case KFD_MQD_TYPE_CP: mqd->allocate_mqd = allocate_mqd; mqd->init_mqd = init_mqd; - mqd->free_mqd = free_mqd; + mqd->free_mqd = kfd_free_mqd_cp; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->get_wave_state = get_wave_state; + mqd->get_checkpoint_info = get_checkpoint_info; + mqd->checkpoint_mqd = checkpoint_mqd; + mqd->restore_mqd = restore_mqd; mqd->mqd_size = sizeof(struct v9_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -479,10 +501,10 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, mqd->allocate_mqd = allocate_hiq_mqd; mqd->init_mqd = init_mqd_hiq; mqd->free_mqd = free_mqd_hiq_sdma; - mqd->load_mqd = hiq_load_mqd_kiq; + mqd->load_mqd = kfd_hiq_load_mqd_kiq; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v9_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -492,11 +514,11 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, case KFD_MQD_TYPE_DIQ: mqd->allocate_mqd = allocate_mqd; mqd->init_mqd = init_mqd_hiq; - mqd->free_mqd = free_mqd; + mqd->free_mqd = kfd_free_mqd_cp; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v9_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -506,10 +528,12 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, mqd->allocate_mqd = allocate_sdma_mqd; mqd->init_mqd = init_mqd_sdma; mqd->free_mqd = free_mqd_hiq_sdma; - mqd->load_mqd = load_mqd_sdma; + mqd->load_mqd = kfd_load_mqd_sdma; mqd->update_mqd = update_mqd_sdma; - mqd->destroy_mqd = destroy_mqd_sdma; - mqd->is_occupied = is_occupied_sdma; + mqd->destroy_mqd = kfd_destroy_mqd_sdma; + mqd->is_occupied = kfd_is_occupied_sdma; + mqd->checkpoint_mqd = checkpoint_mqd_sdma; + mqd->restore_mqd = restore_mqd_sdma; mqd->mqd_size = sizeof(struct v9_sdma_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c index d456e950ce1d..86fc5decacf0 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c @@ -259,31 +259,6 @@ static void update_mqd_tonga(struct mqd_manager *mm, void *mqd, __update_mqd(mm, mqd, q, minfo, MTYPE_UC, 0); } -static int destroy_mqd(struct mqd_manager *mm, void *mqd, - enum kfd_preempt_type type, - unsigned int timeout, uint32_t pipe_id, - uint32_t queue_id) -{ - return mm->dev->kfd2kgd->hqd_destroy - (mm->dev->adev, mqd, type, timeout, - pipe_id, queue_id); -} - -static void free_mqd(struct mqd_manager *mm, void *mqd, - struct kfd_mem_obj *mqd_mem_obj) -{ - kfd_gtt_sa_free(mm->dev, mqd_mem_obj); -} - -static bool is_occupied(struct mqd_manager *mm, void *mqd, - uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) -{ - return mm->dev->kfd2kgd->hqd_is_occupied( - mm->dev->adev, queue_address, - pipe_id, queue_id); -} - static int get_wave_state(struct mqd_manager *mm, void *mqd, void __user *ctl_stack, u32 *ctl_stack_used_size, @@ -306,6 +281,48 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd, return 0; } +static void get_checkpoint_info(struct mqd_manager *mm, void *mqd, u32 *ctl_stack_size) +{ + /* Control stack is stored in user mode */ + *ctl_stack_size = 0; +} + +static void checkpoint_mqd(struct mqd_manager *mm, void *mqd, void *mqd_dst, void *ctl_stack_dst) +{ + struct vi_mqd *m; + + m = get_mqd(mqd); + + memcpy(mqd_dst, m, sizeof(struct vi_mqd)); +} + +static void restore_mqd(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, const u32 ctl_stack_size) +{ + uint64_t addr; + struct vi_mqd *m; + + m = (struct vi_mqd *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + m->cp_hqd_pq_doorbell_control = + qp->doorbell_off << + CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; + pr_debug("cp_hqd_pq_doorbell_control 0x%x\n", + m->cp_hqd_pq_doorbell_control); + + qp->is_active = 0; +} + static void init_mqd_hiq(struct mqd_manager *mm, void **mqd, struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, struct queue_properties *q) @@ -343,15 +360,6 @@ static void init_mqd_sdma(struct mqd_manager *mm, void **mqd, mm->update_mqd(mm, m, q, NULL); } -static int load_mqd_sdma(struct mqd_manager *mm, void *mqd, - uint32_t pipe_id, uint32_t queue_id, - struct queue_properties *p, struct mm_struct *mms) -{ - return mm->dev->kfd2kgd->hqd_sdma_load(mm->dev->adev, mqd, - (uint32_t __user *)p->write_ptr, - mms); -} - static void update_mqd_sdma(struct mqd_manager *mm, void *mqd, struct queue_properties *q, struct mqd_update_info *minfo) @@ -380,27 +388,45 @@ static void update_mqd_sdma(struct mqd_manager *mm, void *mqd, q->is_active = QUEUE_IS_ACTIVE(*q); } -/* - * * preempt type here is ignored because there is only one way - * * to preempt sdma queue - */ -static int destroy_mqd_sdma(struct mqd_manager *mm, void *mqd, - enum kfd_preempt_type type, - unsigned int timeout, uint32_t pipe_id, - uint32_t queue_id) +static void checkpoint_mqd_sdma(struct mqd_manager *mm, + void *mqd, + void *mqd_dst, + void *ctl_stack_dst) { - return mm->dev->kfd2kgd->hqd_sdma_destroy(mm->dev->adev, mqd, timeout); + struct vi_sdma_mqd *m; + + m = get_sdma_mqd(mqd); + + memcpy(mqd_dst, m, sizeof(struct vi_sdma_mqd)); } -static bool is_occupied_sdma(struct mqd_manager *mm, void *mqd, - uint64_t queue_address, uint32_t pipe_id, - uint32_t queue_id) +static void restore_mqd_sdma(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, const u32 ctl_stack_size) { - return mm->dev->kfd2kgd->hqd_sdma_is_occupied(mm->dev->adev, mqd); + uint64_t addr; + struct vi_sdma_mqd *m; + + m = (struct vi_sdma_mqd *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + m->sdmax_rlcx_doorbell = + qp->doorbell_off << SDMA0_RLC0_DOORBELL__OFFSET__SHIFT; + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + qp->is_active = 0; } #if defined(CONFIG_DEBUG_FS) + static int debugfs_show_mqd(struct seq_file *m, void *data) { seq_hex_dump(m, " ", DUMP_PREFIX_OFFSET, 32, 4, @@ -435,12 +461,15 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, case KFD_MQD_TYPE_CP: mqd->allocate_mqd = allocate_mqd; mqd->init_mqd = init_mqd; - mqd->free_mqd = free_mqd; + mqd->free_mqd = kfd_free_mqd_cp; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->get_wave_state = get_wave_state; + mqd->get_checkpoint_info = get_checkpoint_info; + mqd->checkpoint_mqd = checkpoint_mqd; + mqd->restore_mqd = restore_mqd; mqd->mqd_size = sizeof(struct vi_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -452,8 +481,8 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, mqd->free_mqd = free_mqd_hiq_sdma; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd_hiq; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct vi_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -463,11 +492,11 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, case KFD_MQD_TYPE_DIQ: mqd->allocate_mqd = allocate_mqd; mqd->init_mqd = init_mqd_hiq; - mqd->free_mqd = free_mqd; + mqd->free_mqd = kfd_free_mqd_cp; mqd->load_mqd = load_mqd; mqd->update_mqd = update_mqd_hiq; - mqd->destroy_mqd = destroy_mqd; - mqd->is_occupied = is_occupied; + mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct vi_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; @@ -477,10 +506,12 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, mqd->allocate_mqd = allocate_sdma_mqd; mqd->init_mqd = init_mqd_sdma; mqd->free_mqd = free_mqd_hiq_sdma; - mqd->load_mqd = load_mqd_sdma; + mqd->load_mqd = kfd_load_mqd_sdma; mqd->update_mqd = update_mqd_sdma; - mqd->destroy_mqd = destroy_mqd_sdma; - mqd->is_occupied = is_occupied_sdma; + mqd->destroy_mqd = kfd_destroy_mqd_sdma; + mqd->is_occupied = kfd_is_occupied_sdma; + mqd->checkpoint_mqd = checkpoint_mqd_sdma; + mqd->restore_mqd = restore_mqd_sdma; mqd->mqd_size = sizeof(struct vi_sdma_mqd); #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index ea68f3b3a4e9..b6790a637f5c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -121,7 +121,26 @@ */ #define KFD_QUEUE_DOORBELL_MIRROR_OFFSET 512 - +/** + * enum kfd_ioctl_flags - KFD ioctl flags + * Various flags that can be set in &amdkfd_ioctl_desc.flags to control how + * userspace can use a given ioctl. + */ +enum kfd_ioctl_flags { + /* + * @KFD_IOC_FLAG_CHECKPOINT_RESTORE: + * Certain KFD ioctls such as AMDKFD_IOC_CRIU_OP can potentially + * perform privileged operations and load arbitrary data into MQDs and + * eventually HQD registers when the queue is mapped by HWS. In order to + * prevent this we should perform additional security checks. + * + * This is equivalent to callers with the CHECKPOINT_RESTORE capability. + * + * Note: Since earlier versions of docker do not support CHECKPOINT_RESTORE, + * we also allow ioctls with SYS_ADMIN capability. + */ + KFD_IOC_FLAG_CHECKPOINT_RESTORE = BIT(0), +}; /* * Kernel module parameter to specify maximum number of supported queues per * device @@ -281,9 +300,6 @@ struct kfd_dev { */ bool interrupts_active; - /* Debug manager */ - struct kfd_dbgmgr *dbgmgr; - /* Firmware versions */ uint16_t mec_fw_version; uint16_t mec2_fw_version; @@ -442,6 +458,7 @@ enum KFD_QUEUE_PRIORITY { * it's user mode or kernel mode queue. * */ + struct queue_properties { enum kfd_queue_type type; enum kfd_queue_format format; @@ -754,6 +771,12 @@ struct kfd_process_device { uint64_t faults; uint64_t page_in; uint64_t page_out; + /* + * If this process has been checkpointed before, then the user + * application will use the original gpu_id on the + * checkpointed node to refer to this device. + */ + uint32_t user_gpu_id; }; #define qpd_to_pdd(x) container_of(x, struct kfd_process_device, qpd) @@ -764,6 +787,7 @@ struct svm_range_list { struct list_head list; struct work_struct deferred_list_work; struct list_head deferred_range_list; + struct list_head criu_svm_metadata_list; spinlock_t deferred_list_lock; atomic_t evicted_ranges; atomic_t drain_pagefaults; @@ -858,6 +882,8 @@ struct kfd_process { bool xnack_enabled; atomic_t poison; + /* Queues are in paused stated because we are in the process of doing a CRIU checkpoint */ + bool queues_paused; }; #define KFD_PROCESS_TABLE_SIZE 5 /* bits: 32 entries */ @@ -911,6 +937,11 @@ int kfd_process_restore_queues(struct kfd_process *p); void kfd_suspend_all_processes(void); int kfd_resume_all_processes(void); +struct kfd_process_device *kfd_process_device_data_by_id(struct kfd_process *process, + uint32_t gpu_id); + +int kfd_process_get_user_gpu_id(struct kfd_process *p, uint32_t actual_gpu_id); + int kfd_process_device_init_vm(struct kfd_process_device *pdd, struct file *drm_file); struct kfd_process_device *kfd_bind_process_to_device(struct kfd_dev *dev, @@ -932,6 +963,7 @@ void *kfd_process_device_translate_handle(struct kfd_process_device *p, int handle); void kfd_process_device_remove_obj_handle(struct kfd_process_device *pdd, int handle); +struct kfd_process *kfd_lookup_process_by_pid(struct pid *pid); /* PASIDs */ int kfd_pasid_init(void); @@ -1006,6 +1038,116 @@ void kfd_process_set_trap_handler(struct qcm_process_device *qpd, uint64_t tba_addr, uint64_t tma_addr); +/* CRIU */ +/* + * Need to increment KFD_CRIU_PRIV_VERSION each time a change is made to any of the CRIU private + * structures: + * kfd_criu_process_priv_data + * kfd_criu_device_priv_data + * kfd_criu_bo_priv_data + * kfd_criu_queue_priv_data + * kfd_criu_event_priv_data + * kfd_criu_svm_range_priv_data + */ + +#define KFD_CRIU_PRIV_VERSION 1 + +struct kfd_criu_process_priv_data { + uint32_t version; + uint32_t xnack_mode; +}; + +struct kfd_criu_device_priv_data { + /* For future use */ + uint64_t reserved; +}; + +struct kfd_criu_bo_priv_data { + uint64_t user_addr; + uint32_t idr_handle; + uint32_t mapped_gpuids[MAX_GPU_INSTANCE]; +}; + +/* + * The first 4 bytes of kfd_criu_queue_priv_data, kfd_criu_event_priv_data, + * kfd_criu_svm_range_priv_data is the object type + */ +enum kfd_criu_object_type { + KFD_CRIU_OBJECT_TYPE_QUEUE, + KFD_CRIU_OBJECT_TYPE_EVENT, + KFD_CRIU_OBJECT_TYPE_SVM_RANGE, +}; + +struct kfd_criu_svm_range_priv_data { + uint32_t object_type; + uint64_t start_addr; + uint64_t size; + /* Variable length array of attributes */ + struct kfd_ioctl_svm_attribute attrs[0]; +}; + +struct kfd_criu_queue_priv_data { + uint32_t object_type; + uint64_t q_address; + uint64_t q_size; + uint64_t read_ptr_addr; + uint64_t write_ptr_addr; + uint64_t doorbell_off; + uint64_t eop_ring_buffer_address; + uint64_t ctx_save_restore_area_address; + uint32_t gpu_id; + uint32_t type; + uint32_t format; + uint32_t q_id; + uint32_t priority; + uint32_t q_percent; + uint32_t doorbell_id; + uint32_t is_gws; + uint32_t sdma_id; + uint32_t eop_ring_buffer_size; + uint32_t ctx_save_restore_area_size; + uint32_t ctl_stack_size; + uint32_t mqd_size; +}; + +struct kfd_criu_event_priv_data { + uint32_t object_type; + uint64_t user_handle; + uint32_t event_id; + uint32_t auto_reset; + uint32_t type; + uint32_t signaled; + + union { + struct kfd_hsa_memory_exception_data memory_exception_data; + struct kfd_hsa_hw_exception_data hw_exception_data; + }; +}; + +int kfd_process_get_queue_info(struct kfd_process *p, + uint32_t *num_queues, + uint64_t *priv_data_sizes); + +int kfd_criu_checkpoint_queues(struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_data_offset); + +int kfd_criu_restore_queue(struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_data_offset, + uint64_t max_priv_data_size); + +int kfd_criu_checkpoint_events(struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_data_offset); + +int kfd_criu_restore_event(struct file *devkfd, + struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_data_offset, + uint64_t max_priv_data_size); +/* CRIU - End */ + /* Queue Context Management */ int init_queue(struct queue **q, const struct queue_properties *properties); void uninit_queue(struct queue *q); @@ -1029,7 +1171,7 @@ void device_queue_manager_uninit(struct device_queue_manager *dqm); struct kernel_queue *kernel_queue_init(struct kfd_dev *dev, enum kfd_queue_type type); void kernel_queue_uninit(struct kernel_queue *kq, bool hanging); -int kfd_process_vm_fault(struct device_queue_manager *dqm, u32 pasid); +int kfd_dqm_evict_pasid(struct device_queue_manager *dqm, u32 pasid); /* Process Queue Manager */ struct process_queue_node { @@ -1047,6 +1189,9 @@ int pqm_create_queue(struct process_queue_manager *pqm, struct file *f, struct queue_properties *properties, unsigned int *qid, + const struct kfd_criu_queue_priv_data *q_data, + const void *restore_mqd, + const void *restore_ctl_stack, uint32_t *p_doorbell_offset_in_process); int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid); int pqm_update_queue_properties(struct process_queue_manager *pqm, unsigned int qid, @@ -1069,6 +1214,10 @@ int amdkfd_fence_wait_timeout(uint64_t *fence_addr, uint64_t fence_value, unsigned int timeout_ms); +int pqm_get_queue_checkpoint_info(struct process_queue_manager *pqm, + unsigned int qid, + u32 *mqd_size, + u32 *ctl_stack_size); /* Packet Manager */ #define KFD_FENCE_COMPLETED (100) @@ -1160,12 +1309,14 @@ void kfd_signal_iommu_event(struct kfd_dev *dev, void kfd_signal_hw_exception_event(u32 pasid); int kfd_set_event(struct kfd_process *p, uint32_t event_id); int kfd_reset_event(struct kfd_process *p, uint32_t event_id); -int kfd_event_page_set(struct kfd_process *p, void *kernel_address, - uint64_t size); +int kfd_kmap_event_page(struct kfd_process *p, uint64_t event_page_offset); + int kfd_event_create(struct file *devkfd, struct kfd_process *p, uint32_t event_type, bool auto_reset, uint32_t node_id, uint32_t *event_id, uint32_t *event_trigger_data, uint64_t *event_page_offset, uint32_t *event_slot_index); + +int kfd_get_num_events(struct kfd_process *p); int kfd_event_destroy(struct kfd_process *p, uint32_t event_id); void kfd_signal_vm_fault_event(struct kfd_dev *dev, u32 pasid, @@ -1177,8 +1328,6 @@ void kfd_signal_poison_consumed_event(struct kfd_dev *dev, u32 pasid); void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type); -int dbgdev_wave_reset_wavefronts(struct kfd_dev *dev, struct kfd_process *p); - bool kfd_is_locked(void); /* Compute profile */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index d1145da5348f..8c6a48add76e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -40,7 +40,6 @@ struct mm_struct; #include "kfd_priv.h" #include "kfd_device_queue_manager.h" -#include "kfd_dbgmgr.h" #include "kfd_iommu.h" #include "kfd_svm.h" @@ -64,7 +63,8 @@ static struct workqueue_struct *kfd_process_wq; */ static struct workqueue_struct *kfd_restore_wq; -static struct kfd_process *find_process(const struct task_struct *thread); +static struct kfd_process *find_process(const struct task_struct *thread, + bool ref); static void kfd_process_ref_release(struct kref *ref); static struct kfd_process *create_process(const struct task_struct *thread); static int kfd_process_init_cwsr_apu(struct kfd_process *p, struct file *filep); @@ -715,7 +715,8 @@ static int kfd_process_alloc_gpuvm(struct kfd_process_device *pdd, int err; err = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(kdev->adev, gpu_va, size, - pdd->drm_priv, mem, NULL, flags); + pdd->drm_priv, mem, NULL, + flags, false); if (err) goto err_alloc_mem; @@ -816,7 +817,7 @@ struct kfd_process *kfd_create_process(struct file *filep) mutex_lock(&kfd_processes_mutex); /* A prior open of /dev/kfd could have already created the process. */ - process = find_process(thread); + process = find_process(thread, false); if (process) { pr_debug("Process already found\n"); } else { @@ -884,7 +885,7 @@ struct kfd_process *kfd_get_process(const struct task_struct *thread) if (thread->group_leader->mm != thread->mm) return ERR_PTR(-EINVAL); - process = find_process(thread); + process = find_process(thread, false); if (!process) return ERR_PTR(-EINVAL); @@ -903,13 +904,16 @@ static struct kfd_process *find_process_by_mm(const struct mm_struct *mm) return NULL; } -static struct kfd_process *find_process(const struct task_struct *thread) +static struct kfd_process *find_process(const struct task_struct *thread, + bool ref) { struct kfd_process *p; int idx; idx = srcu_read_lock(&kfd_processes_srcu); p = find_process_by_mm(thread->mm); + if (p && ref) + kref_get(&p->ref); srcu_read_unlock(&kfd_processes_srcu, idx); return p; @@ -920,6 +924,26 @@ void kfd_unref_process(struct kfd_process *p) kref_put(&p->ref, kfd_process_ref_release); } +/* This increments the process->ref counter. */ +struct kfd_process *kfd_lookup_process_by_pid(struct pid *pid) +{ + struct task_struct *task = NULL; + struct kfd_process *p = NULL; + + if (!pid) { + task = current; + get_task_struct(task); + } else { + task = get_pid_task(pid, PIDTYPE_PID); + } + + if (task) { + p = find_process(task, true); + put_task_struct(task); + } + + return p; +} static void kfd_process_device_free_bos(struct kfd_process_device *pdd) { @@ -1133,7 +1157,6 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn, struct mm_struct *mm) { struct kfd_process *p; - int i; /* * The kfd_process structure can not be free because the @@ -1150,27 +1173,9 @@ static void kfd_process_notifier_release(struct mmu_notifier *mn, cancel_delayed_work_sync(&p->eviction_work); cancel_delayed_work_sync(&p->restore_work); - cancel_delayed_work_sync(&p->svms.restore_work); mutex_lock(&p->mutex); - /* Iterate over all process device data structures and if the - * pdd is in debug mode, we should first force unregistration, - * then we will be able to destroy the queues - */ - for (i = 0; i < p->n_pdds; i++) { - struct kfd_dev *dev = p->pdds[i]->dev; - - mutex_lock(kfd_get_dbgmgr_mutex()); - if (dev && dev->dbgmgr && dev->dbgmgr->pasid == p->pasid) { - if (!kfd_dbgmgr_unregister(dev->dbgmgr, p)) { - kfd_dbgmgr_destroy(dev->dbgmgr); - dev->dbgmgr = NULL; - } - } - mutex_unlock(kfd_get_dbgmgr_mutex()); - } - kfd_process_dequeue_from_all_devices(p); pqm_uninit(&p->pqm); @@ -1360,6 +1365,7 @@ static struct kfd_process *create_process(const struct task_struct *thread) process->mm = thread->mm; process->lead_thread = thread->group_leader; process->n_pdds = 0; + process->queues_paused = false; INIT_DELAYED_WORK(&process->eviction_work, evict_process_worker); INIT_DELAYED_WORK(&process->restore_work, restore_process_worker); process->last_restore_timestamp = get_jiffies_64(); @@ -1501,6 +1507,7 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_dev *dev, pdd->runtime_inuse = false; pdd->vram_usage = 0; pdd->sdma_past_activity_counter = 0; + pdd->user_gpu_id = dev->id; atomic64_set(&pdd->evict_duration_counter, 0); p->pdds[p->n_pdds++] = pdd; @@ -1771,7 +1778,7 @@ int kfd_process_gpuidx_from_gpuid(struct kfd_process *p, uint32_t gpu_id) int i; for (i = 0; i < p->n_pdds; i++) - if (p->pdds[i] && gpu_id == p->pdds[i]->dev->id) + if (p->pdds[i] && gpu_id == p->pdds[i]->user_gpu_id) return i; return -EINVAL; } @@ -1784,7 +1791,7 @@ kfd_process_gpuid_from_adev(struct kfd_process *p, struct amdgpu_device *adev, for (i = 0; i < p->n_pdds; i++) if (p->pdds[i] && p->pdds[i]->dev->adev == adev) { - *gpuid = p->pdds[i]->dev->id; + *gpuid = p->pdds[i]->user_gpu_id; *gpuidx = i; return 0; } @@ -1956,6 +1963,37 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type) } } +struct kfd_process_device *kfd_process_device_data_by_id(struct kfd_process *p, uint32_t gpu_id) +{ + int i; + + if (gpu_id) { + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + + if (pdd->user_gpu_id == gpu_id) + return pdd; + } + } + return NULL; +} + +int kfd_process_get_user_gpu_id(struct kfd_process *p, uint32_t actual_gpu_id) +{ + int i; + + if (!actual_gpu_id) + return 0; + + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + + if (pdd->dev->id == actual_gpu_id) + return pdd->user_gpu_id; + } + return -EINVAL; +} + #if defined(CONFIG_DEBUG_FS) int kfd_debugfs_mqds_by_process(struct seq_file *m, void *data) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index 5e5c84a8e1ef..70ea4849e6a2 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -42,6 +42,20 @@ static inline struct process_queue_node *get_queue_by_qid( return NULL; } +static int assign_queue_slot_by_qid(struct process_queue_manager *pqm, + unsigned int qid) +{ + if (qid >= KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) + return -EINVAL; + + if (__test_and_set_bit(qid, pqm->queue_slot_bitmap)) { + pr_err("Cannot create new queue because requested qid(%u) is in use\n", qid); + return -ENOSPC; + } + + return 0; +} + static int find_available_queue_slot(struct process_queue_manager *pqm, unsigned int *qid) { @@ -193,6 +207,9 @@ int pqm_create_queue(struct process_queue_manager *pqm, struct file *f, struct queue_properties *properties, unsigned int *qid, + const struct kfd_criu_queue_priv_data *q_data, + const void *restore_mqd, + const void *restore_ctl_stack, uint32_t *p_doorbell_offset_in_process) { int retval; @@ -224,7 +241,12 @@ int pqm_create_queue(struct process_queue_manager *pqm, if (pdd->qpd.queue_count >= max_queues) return -ENOSPC; - retval = find_available_queue_slot(pqm, qid); + if (q_data) { + retval = assign_queue_slot_by_qid(pqm, q_data->q_id); + *qid = q_data->q_id; + } else + retval = find_available_queue_slot(pqm, qid); + if (retval != 0) return retval; @@ -252,7 +274,8 @@ int pqm_create_queue(struct process_queue_manager *pqm, goto err_create_queue; pqn->q = q; pqn->kq = NULL; - retval = dev->dqm->ops.create_queue(dev->dqm, q, &pdd->qpd); + retval = dev->dqm->ops.create_queue(dev->dqm, q, &pdd->qpd, q_data, + restore_mqd, restore_ctl_stack); print_queue(q); break; @@ -272,7 +295,8 @@ int pqm_create_queue(struct process_queue_manager *pqm, goto err_create_queue; pqn->q = q; pqn->kq = NULL; - retval = dev->dqm->ops.create_queue(dev->dqm, q, &pdd->qpd); + retval = dev->dqm->ops.create_queue(dev->dqm, q, &pdd->qpd, q_data, + restore_mqd, restore_ctl_stack); print_queue(q); break; case KFD_QUEUE_TYPE_DIQ: @@ -497,6 +521,348 @@ int pqm_get_wave_state(struct process_queue_manager *pqm, save_area_used_size); } +static int get_queue_data_sizes(struct kfd_process_device *pdd, + struct queue *q, + uint32_t *mqd_size, + uint32_t *ctl_stack_size) +{ + int ret; + + ret = pqm_get_queue_checkpoint_info(&pdd->process->pqm, + q->properties.queue_id, + mqd_size, + ctl_stack_size); + if (ret) + pr_err("Failed to get queue dump info (%d)\n", ret); + + return ret; +} + +int kfd_process_get_queue_info(struct kfd_process *p, + uint32_t *num_queues, + uint64_t *priv_data_sizes) +{ + uint32_t extra_data_sizes = 0; + struct queue *q; + int i; + int ret; + + *num_queues = 0; + + /* Run over all PDDs of the process */ + for (i = 0; i < p->n_pdds; i++) { + struct kfd_process_device *pdd = p->pdds[i]; + + list_for_each_entry(q, &pdd->qpd.queues_list, list) { + if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE || + q->properties.type == KFD_QUEUE_TYPE_SDMA || + q->properties.type == KFD_QUEUE_TYPE_SDMA_XGMI) { + uint32_t mqd_size, ctl_stack_size; + + *num_queues = *num_queues + 1; + + ret = get_queue_data_sizes(pdd, q, &mqd_size, &ctl_stack_size); + if (ret) + return ret; + + extra_data_sizes += mqd_size + ctl_stack_size; + } else { + pr_err("Unsupported queue type (%d)\n", q->properties.type); + return -EOPNOTSUPP; + } + } + } + *priv_data_sizes = extra_data_sizes + + (*num_queues * sizeof(struct kfd_criu_queue_priv_data)); + + return 0; +} + +static int pqm_checkpoint_mqd(struct process_queue_manager *pqm, + unsigned int qid, + void *mqd, + void *ctl_stack) +{ + struct process_queue_node *pqn; + + pqn = get_queue_by_qid(pqm, qid); + if (!pqn) { + pr_debug("amdkfd: No queue %d exists for operation\n", qid); + return -EFAULT; + } + + if (!pqn->q->device->dqm->ops.checkpoint_mqd) { + pr_err("amdkfd: queue dumping not supported on this device\n"); + return -EOPNOTSUPP; + } + + return pqn->q->device->dqm->ops.checkpoint_mqd(pqn->q->device->dqm, + pqn->q, mqd, ctl_stack); +} + +static int criu_checkpoint_queue(struct kfd_process_device *pdd, + struct queue *q, + struct kfd_criu_queue_priv_data *q_data) +{ + uint8_t *mqd, *ctl_stack; + int ret; + + mqd = (void *)(q_data + 1); + ctl_stack = mqd + q_data->mqd_size; + + q_data->gpu_id = pdd->user_gpu_id; + q_data->type = q->properties.type; + q_data->format = q->properties.format; + q_data->q_id = q->properties.queue_id; + q_data->q_address = q->properties.queue_address; + q_data->q_size = q->properties.queue_size; + q_data->priority = q->properties.priority; + q_data->q_percent = q->properties.queue_percent; + q_data->read_ptr_addr = (uint64_t)q->properties.read_ptr; + q_data->write_ptr_addr = (uint64_t)q->properties.write_ptr; + q_data->doorbell_id = q->doorbell_id; + + q_data->sdma_id = q->sdma_id; + + q_data->eop_ring_buffer_address = + q->properties.eop_ring_buffer_address; + + q_data->eop_ring_buffer_size = q->properties.eop_ring_buffer_size; + + q_data->ctx_save_restore_area_address = + q->properties.ctx_save_restore_area_address; + + q_data->ctx_save_restore_area_size = + q->properties.ctx_save_restore_area_size; + + ret = pqm_checkpoint_mqd(&pdd->process->pqm, q->properties.queue_id, mqd, ctl_stack); + if (ret) { + pr_err("Failed checkpoint queue_mqd (%d)\n", ret); + return ret; + } + + pr_debug("Dumping Queue: gpu_id:%x queue_id:%u\n", q_data->gpu_id, q_data->q_id); + return ret; +} + +static int criu_checkpoint_queues_device(struct kfd_process_device *pdd, + uint8_t __user *user_priv, + unsigned int *q_index, + uint64_t *queues_priv_data_offset) +{ + unsigned int q_private_data_size = 0; + uint8_t *q_private_data = NULL; /* Local buffer to store individual queue private data */ + struct queue *q; + int ret = 0; + + list_for_each_entry(q, &pdd->qpd.queues_list, list) { + struct kfd_criu_queue_priv_data *q_data; + uint64_t q_data_size; + uint32_t mqd_size; + uint32_t ctl_stack_size; + + if (q->properties.type != KFD_QUEUE_TYPE_COMPUTE && + q->properties.type != KFD_QUEUE_TYPE_SDMA && + q->properties.type != KFD_QUEUE_TYPE_SDMA_XGMI) { + + pr_err("Unsupported queue type (%d)\n", q->properties.type); + ret = -EOPNOTSUPP; + break; + } + + ret = get_queue_data_sizes(pdd, q, &mqd_size, &ctl_stack_size); + if (ret) + break; + + q_data_size = sizeof(*q_data) + mqd_size + ctl_stack_size; + + /* Increase local buffer space if needed */ + if (q_private_data_size < q_data_size) { + kfree(q_private_data); + + q_private_data = kzalloc(q_data_size, GFP_KERNEL); + if (!q_private_data) { + ret = -ENOMEM; + break; + } + q_private_data_size = q_data_size; + } + + q_data = (struct kfd_criu_queue_priv_data *)q_private_data; + + /* data stored in this order: priv_data, mqd, ctl_stack */ + q_data->mqd_size = mqd_size; + q_data->ctl_stack_size = ctl_stack_size; + + ret = criu_checkpoint_queue(pdd, q, q_data); + if (ret) + break; + + q_data->object_type = KFD_CRIU_OBJECT_TYPE_QUEUE; + + ret = copy_to_user(user_priv + *queues_priv_data_offset, + q_data, q_data_size); + if (ret) { + ret = -EFAULT; + break; + } + *queues_priv_data_offset += q_data_size; + *q_index = *q_index + 1; + } + + kfree(q_private_data); + + return ret; +} + +int kfd_criu_checkpoint_queues(struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_data_offset) +{ + int ret = 0, pdd_index, q_index = 0; + + for (pdd_index = 0; pdd_index < p->n_pdds; pdd_index++) { + struct kfd_process_device *pdd = p->pdds[pdd_index]; + + /* + * criu_checkpoint_queues_device will copy data to user and update q_index and + * queues_priv_data_offset + */ + ret = criu_checkpoint_queues_device(pdd, user_priv_data, &q_index, + priv_data_offset); + + if (ret) + break; + } + + return ret; +} + +static void set_queue_properties_from_criu(struct queue_properties *qp, + struct kfd_criu_queue_priv_data *q_data) +{ + qp->is_interop = false; + qp->is_gws = q_data->is_gws; + qp->queue_percent = q_data->q_percent; + qp->priority = q_data->priority; + qp->queue_address = q_data->q_address; + qp->queue_size = q_data->q_size; + qp->read_ptr = (uint32_t *) q_data->read_ptr_addr; + qp->write_ptr = (uint32_t *) q_data->write_ptr_addr; + qp->eop_ring_buffer_address = q_data->eop_ring_buffer_address; + qp->eop_ring_buffer_size = q_data->eop_ring_buffer_size; + qp->ctx_save_restore_area_address = q_data->ctx_save_restore_area_address; + qp->ctx_save_restore_area_size = q_data->ctx_save_restore_area_size; + qp->ctl_stack_size = q_data->ctl_stack_size; + qp->type = q_data->type; + qp->format = q_data->format; +} + +int kfd_criu_restore_queue(struct kfd_process *p, + uint8_t __user *user_priv_ptr, + uint64_t *priv_data_offset, + uint64_t max_priv_data_size) +{ + uint8_t *mqd, *ctl_stack, *q_extra_data = NULL; + struct kfd_criu_queue_priv_data *q_data; + struct kfd_process_device *pdd; + uint64_t q_extra_data_size; + struct queue_properties qp; + unsigned int queue_id; + int ret = 0; + + if (*priv_data_offset + sizeof(*q_data) > max_priv_data_size) + return -EINVAL; + + q_data = kmalloc(sizeof(*q_data), GFP_KERNEL); + if (!q_data) + return -ENOMEM; + + ret = copy_from_user(q_data, user_priv_ptr + *priv_data_offset, sizeof(*q_data)); + if (ret) { + ret = -EFAULT; + goto exit; + } + + *priv_data_offset += sizeof(*q_data); + q_extra_data_size = q_data->ctl_stack_size + q_data->mqd_size; + + if (*priv_data_offset + q_extra_data_size > max_priv_data_size) { + ret = -EINVAL; + goto exit; + } + + q_extra_data = kmalloc(q_extra_data_size, GFP_KERNEL); + if (!q_extra_data) { + ret = -ENOMEM; + goto exit; + } + + ret = copy_from_user(q_extra_data, user_priv_ptr + *priv_data_offset, q_extra_data_size); + if (ret) { + ret = -EFAULT; + goto exit; + } + + *priv_data_offset += q_extra_data_size; + + pdd = kfd_process_device_data_by_id(p, q_data->gpu_id); + if (!pdd) { + pr_err("Failed to get pdd\n"); + ret = -EINVAL; + goto exit; + } + /* data stored in this order: mqd, ctl_stack */ + mqd = q_extra_data; + ctl_stack = mqd + q_data->mqd_size; + + memset(&qp, 0, sizeof(qp)); + set_queue_properties_from_criu(&qp, q_data); + + print_queue_properties(&qp); + + ret = pqm_create_queue(&p->pqm, pdd->dev, NULL, &qp, &queue_id, q_data, mqd, ctl_stack, + NULL); + if (ret) { + pr_err("Failed to create new queue err:%d\n", ret); + ret = -EINVAL; + } + +exit: + if (ret) + pr_err("Failed to create queue (%d)\n", ret); + else + pr_debug("Queue id %d was restored successfully\n", queue_id); + + kfree(q_data); + + return ret; +} + +int pqm_get_queue_checkpoint_info(struct process_queue_manager *pqm, + unsigned int qid, + uint32_t *mqd_size, + uint32_t *ctl_stack_size) +{ + struct process_queue_node *pqn; + + pqn = get_queue_by_qid(pqm, qid); + if (!pqn) { + pr_debug("amdkfd: No queue %d exists for operation\n", qid); + return -EFAULT; + } + + if (!pqn->q->device->dqm->ops.get_queue_checkpoint_info) { + pr_err("amdkfd: queue dumping not supported on this device\n"); + return -EOPNOTSUPP; + } + + pqn->q->device->dqm->ops.get_queue_checkpoint_info(pqn->q->device->dqm, + pqn->q, mqd_size, + ctl_stack_size); + return 0; +} + #if defined(CONFIG_DEBUG_FS) int pqm_debugfs_mqds(struct seq_file *m, void *data) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c index deae12dc777d..329a4c89f1e6 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c @@ -222,7 +222,7 @@ void kfd_smi_event_update_thermal_throttling(struct kfd_dev *dev, len = snprintf(fifo_in, sizeof(fifo_in), "%x %llx:%llx\n", KFD_SMI_EVENT_THERMAL_THROTTLE, throttle_bitmask, - atomic64_read(&dev->adev->smu.throttle_int_counter)); + amdgpu_dpm_get_thermal_throttling_counter(dev->adev)); add_event_to_kfifo(dev, KFD_SMI_EVENT_THERMAL_THROTTLE, fifo_in, len); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index f2805ba74c80..b71d47afd243 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -45,6 +45,11 @@ */ #define AMDGPU_SVM_RANGE_RETRY_FAULT_PENDING 2000 +struct criu_svm_metadata { + struct list_head list; + struct kfd_criu_svm_range_priv_data data; +}; + static void svm_range_evict_svm_bo_worker(struct work_struct *work); static bool svm_range_cpu_invalidate_pagetables(struct mmu_interval_notifier *mni, @@ -1224,19 +1229,20 @@ svm_range_unmap_from_gpus(struct svm_range *prange, unsigned long start, if (r) break; } - amdgpu_amdkfd_flush_gpu_tlb_pasid(pdd->dev->adev, - p->pasid, TLB_FLUSH_HEAVYWEIGHT); + kfd_flush_tlb(pdd, TLB_FLUSH_HEAVYWEIGHT); } return r; } static int -svm_range_map_to_gpu(struct amdgpu_device *adev, struct amdgpu_vm *vm, - struct svm_range *prange, unsigned long offset, - unsigned long npages, bool readonly, dma_addr_t *dma_addr, - struct amdgpu_device *bo_adev, struct dma_fence **fence) +svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange, + unsigned long offset, unsigned long npages, bool readonly, + dma_addr_t *dma_addr, struct amdgpu_device *bo_adev, + struct dma_fence **fence) { + struct amdgpu_device *adev = pdd->dev->adev; + struct amdgpu_vm *vm = drm_priv_to_vm(pdd->drm_priv); bool table_freed = false; uint64_t pte_flags; unsigned long last_start; @@ -1300,12 +1306,8 @@ svm_range_map_to_gpu(struct amdgpu_device *adev, struct amdgpu_vm *vm, if (fence) *fence = dma_fence_get(vm->last_update); - if (table_freed) { - struct kfd_process *p; - - p = container_of(prange->svms, struct kfd_process, svms); - amdgpu_amdkfd_flush_gpu_tlb_pasid(adev, p->pasid, TLB_FLUSH_LEGACY); - } + if (table_freed) + kfd_flush_tlb(pdd, TLB_FLUSH_LEGACY); out: return r; } @@ -1346,8 +1348,7 @@ svm_range_map_to_gpus(struct svm_range *prange, unsigned long offset, continue; } - r = svm_range_map_to_gpu(pdd->dev->adev, drm_priv_to_vm(pdd->drm_priv), - prange, offset, npages, readonly, + r = svm_range_map_to_gpu(pdd, prange, offset, npages, readonly, prange->dma_addr[gpuidx], bo_adev, wait ? &fence : NULL); if (r) @@ -1643,13 +1644,14 @@ static void svm_range_restore_work(struct work_struct *work) pr_debug("restore svm ranges\n"); - /* kfd_process_notifier_release destroys this worker thread. So during - * the lifetime of this thread, kfd_process and mm will be valid. - */ p = container_of(svms, struct kfd_process, svms); - mm = p->mm; - if (!mm) + + /* Keep mm reference when svm_range_validate_and_map ranges */ + mm = get_task_mm(p->lead_thread); + if (!mm) { + pr_debug("svms 0x%p process mm gone\n", svms); return; + } svm_range_list_lock_and_flush_work(svms, mm); mutex_lock(&svms->lock); @@ -1703,6 +1705,7 @@ static void svm_range_restore_work(struct work_struct *work) out_reschedule: mutex_unlock(&svms->lock); mmap_write_unlock(mm); + mmput(mm); /* If validation failed, reschedule another attempt */ if (evicted_ranges) { @@ -1985,10 +1988,9 @@ svm_range_update_notifier_and_interval_tree(struct mm_struct *mm, } static void -svm_range_handle_list_op(struct svm_range_list *svms, struct svm_range *prange) +svm_range_handle_list_op(struct svm_range_list *svms, struct svm_range *prange, + struct mm_struct *mm) { - struct mm_struct *mm = prange->work_item.mm; - switch (prange->work_item.op) { case SVM_OP_NULL: pr_debug("NULL OP 0x%p prange 0x%p [0x%lx 0x%lx]\n", @@ -2065,40 +2067,44 @@ static void svm_range_deferred_list_work(struct work_struct *work) struct svm_range_list *svms; struct svm_range *prange; struct mm_struct *mm; - struct kfd_process *p; svms = container_of(work, struct svm_range_list, deferred_list_work); pr_debug("enter svms 0x%p\n", svms); - p = container_of(svms, struct kfd_process, svms); - /* Avoid mm is gone when inserting mmu notifier */ - mm = get_task_mm(p->lead_thread); - if (!mm) { - pr_debug("svms 0x%p process mm gone\n", svms); - return; - } -retry: - mmap_write_lock(mm); - - /* Checking for the need to drain retry faults must be inside - * mmap write lock to serialize with munmap notifiers. - */ - if (unlikely(atomic_read(&svms->drain_pagefaults))) { - mmap_write_unlock(mm); - svm_range_drain_retry_fault(svms); - goto retry; - } - spin_lock(&svms->deferred_list_lock); while (!list_empty(&svms->deferred_range_list)) { prange = list_first_entry(&svms->deferred_range_list, struct svm_range, deferred_list); - list_del_init(&prange->deferred_list); spin_unlock(&svms->deferred_list_lock); pr_debug("prange 0x%p [0x%lx 0x%lx] op %d\n", prange, prange->start, prange->last, prange->work_item.op); + mm = prange->work_item.mm; +retry: + mmap_write_lock(mm); + + /* Checking for the need to drain retry faults must be inside + * mmap write lock to serialize with munmap notifiers. + */ + if (unlikely(atomic_read(&svms->drain_pagefaults))) { + mmap_write_unlock(mm); + svm_range_drain_retry_fault(svms); + goto retry; + } + + /* Remove from deferred_list must be inside mmap write lock, for + * two race cases: + * 1. unmap_from_cpu may change work_item.op and add the range + * to deferred_list again, cause use after free bug. + * 2. svm_range_list_lock_and_flush_work may hold mmap write + * lock and continue because deferred_list is empty, but + * deferred_list work is actually waiting for mmap lock. + */ + spin_lock(&svms->deferred_list_lock); + list_del_init(&prange->deferred_list); + spin_unlock(&svms->deferred_list_lock); + mutex_lock(&svms->lock); mutex_lock(&prange->migrate_mutex); while (!list_empty(&prange->child_list)) { @@ -2109,19 +2115,20 @@ retry: pr_debug("child prange 0x%p op %d\n", pchild, pchild->work_item.op); list_del_init(&pchild->child_list); - svm_range_handle_list_op(svms, pchild); + svm_range_handle_list_op(svms, pchild, mm); } mutex_unlock(&prange->migrate_mutex); - svm_range_handle_list_op(svms, prange); + svm_range_handle_list_op(svms, prange, mm); mutex_unlock(&svms->lock); + mmap_write_unlock(mm); + + /* Pairs with mmget in svm_range_add_list_work */ + mmput(mm); spin_lock(&svms->deferred_list_lock); } spin_unlock(&svms->deferred_list_lock); - - mmap_write_unlock(mm); - mmput(mm); pr_debug("exit svms 0x%p\n", svms); } @@ -2139,6 +2146,9 @@ svm_range_add_list_work(struct svm_range_list *svms, struct svm_range *prange, prange->work_item.op = op; } else { prange->work_item.op = op; + + /* Pairs with mmput in deferred_list_work */ + mmget(mm); prange->work_item.mm = mm; list_add_tail(&prange->deferred_list, &prange->svms->deferred_range_list); @@ -2830,6 +2840,8 @@ void svm_range_list_fini(struct kfd_process *p) pr_debug("pasid 0x%x svms 0x%p\n", p->pasid, &p->svms); + cancel_delayed_work_sync(&p->svms.restore_work); + /* Ensure list work is finished before process is destroyed */ flush_work(&p->svms.deferred_list_work); @@ -2840,7 +2852,6 @@ void svm_range_list_fini(struct kfd_process *p) atomic_inc(&p->svms.drain_pagefaults); svm_range_drain_retry_fault(&p->svms); - list_for_each_entry_safe(prange, next, &p->svms.list, list) { svm_range_unlink(prange); svm_range_remove_notifier(prange); @@ -2865,6 +2876,7 @@ int svm_range_list_init(struct kfd_process *p) INIT_DELAYED_WORK(&svms->restore_work, svm_range_restore_work); INIT_WORK(&svms->deferred_list_work, svm_range_deferred_list_work); INIT_LIST_HEAD(&svms->deferred_range_list); + INIT_LIST_HEAD(&svms->criu_svm_metadata_list); spin_lock_init(&svms->deferred_list_lock); for (i = 0; i < p->n_pdds; i++) @@ -3193,10 +3205,10 @@ static void svm_range_evict_svm_bo_worker(struct work_struct *work) } static int -svm_range_set_attr(struct kfd_process *p, uint64_t start, uint64_t size, - uint32_t nattr, struct kfd_ioctl_svm_attribute *attrs) +svm_range_set_attr(struct kfd_process *p, struct mm_struct *mm, + uint64_t start, uint64_t size, uint32_t nattr, + struct kfd_ioctl_svm_attribute *attrs) { - struct mm_struct *mm = current->mm; struct list_head update_list; struct list_head insert_list; struct list_head remove_list; @@ -3295,8 +3307,9 @@ out: } static int -svm_range_get_attr(struct kfd_process *p, uint64_t start, uint64_t size, - uint32_t nattr, struct kfd_ioctl_svm_attribute *attrs) +svm_range_get_attr(struct kfd_process *p, struct mm_struct *mm, + uint64_t start, uint64_t size, uint32_t nattr, + struct kfd_ioctl_svm_attribute *attrs) { DECLARE_BITMAP(bitmap_access, MAX_GPU_INSTANCE); DECLARE_BITMAP(bitmap_aip, MAX_GPU_INSTANCE); @@ -3306,7 +3319,6 @@ svm_range_get_attr(struct kfd_process *p, uint64_t start, uint64_t size, bool get_accessible = false; bool get_flags = false; uint64_t last = start + size - 1UL; - struct mm_struct *mm = current->mm; uint8_t granularity = 0xff; struct interval_tree_node *node; struct svm_range_list *svms; @@ -3471,10 +3483,321 @@ fill_values: return 0; } +int kfd_criu_resume_svm(struct kfd_process *p) +{ + struct kfd_ioctl_svm_attribute *set_attr_new, *set_attr = NULL; + int nattr_common = 4, nattr_accessibility = 1; + struct criu_svm_metadata *criu_svm_md = NULL; + struct svm_range_list *svms = &p->svms; + struct criu_svm_metadata *next = NULL; + uint32_t set_flags = 0xffffffff; + int i, j, num_attrs, ret = 0; + uint64_t set_attr_size; + struct mm_struct *mm; + + if (list_empty(&svms->criu_svm_metadata_list)) { + pr_debug("No SVM data from CRIU restore stage 2\n"); + return ret; + } + + mm = get_task_mm(p->lead_thread); + if (!mm) { + pr_err("failed to get mm for the target process\n"); + return -ESRCH; + } + + num_attrs = nattr_common + (nattr_accessibility * p->n_pdds); + + i = j = 0; + list_for_each_entry(criu_svm_md, &svms->criu_svm_metadata_list, list) { + pr_debug("criu_svm_md[%d]\n\tstart: 0x%llx size: 0x%llx (npages)\n", + i, criu_svm_md->data.start_addr, criu_svm_md->data.size); + + for (j = 0; j < num_attrs; j++) { + pr_debug("\ncriu_svm_md[%d]->attrs[%d].type : 0x%x\ncriu_svm_md[%d]->attrs[%d].value : 0x%x\n", + i, j, criu_svm_md->data.attrs[j].type, + i, j, criu_svm_md->data.attrs[j].value); + switch (criu_svm_md->data.attrs[j].type) { + /* During Checkpoint operation, the query for + * KFD_IOCTL_SVM_ATTR_PREFETCH_LOC attribute might + * return KFD_IOCTL_SVM_LOCATION_UNDEFINED if they were + * not used by the range which was checkpointed. Care + * must be taken to not restore with an invalid value + * otherwise the gpuidx value will be invalid and + * set_attr would eventually fail so just replace those + * with another dummy attribute such as + * KFD_IOCTL_SVM_ATTR_SET_FLAGS. + */ + case KFD_IOCTL_SVM_ATTR_PREFETCH_LOC: + if (criu_svm_md->data.attrs[j].value == + KFD_IOCTL_SVM_LOCATION_UNDEFINED) { + criu_svm_md->data.attrs[j].type = + KFD_IOCTL_SVM_ATTR_SET_FLAGS; + criu_svm_md->data.attrs[j].value = 0; + } + break; + case KFD_IOCTL_SVM_ATTR_SET_FLAGS: + set_flags = criu_svm_md->data.attrs[j].value; + break; + default: + break; + } + } + + /* CLR_FLAGS is not available via get_attr during checkpoint but + * it needs to be inserted before restoring the ranges so + * allocate extra space for it before calling set_attr + */ + set_attr_size = sizeof(struct kfd_ioctl_svm_attribute) * + (num_attrs + 1); + set_attr_new = krealloc(set_attr, set_attr_size, + GFP_KERNEL); + if (!set_attr_new) { + ret = -ENOMEM; + goto exit; + } + set_attr = set_attr_new; + + memcpy(set_attr, criu_svm_md->data.attrs, num_attrs * + sizeof(struct kfd_ioctl_svm_attribute)); + set_attr[num_attrs].type = KFD_IOCTL_SVM_ATTR_CLR_FLAGS; + set_attr[num_attrs].value = ~set_flags; + + ret = svm_range_set_attr(p, mm, criu_svm_md->data.start_addr, + criu_svm_md->data.size, num_attrs + 1, + set_attr); + if (ret) { + pr_err("CRIU: failed to set range attributes\n"); + goto exit; + } + + i++; + } +exit: + kfree(set_attr); + list_for_each_entry_safe(criu_svm_md, next, &svms->criu_svm_metadata_list, list) { + pr_debug("freeing criu_svm_md[]\n\tstart: 0x%llx\n", + criu_svm_md->data.start_addr); + kfree(criu_svm_md); + } + + mmput(mm); + return ret; + +} + +int kfd_criu_restore_svm(struct kfd_process *p, + uint8_t __user *user_priv_ptr, + uint64_t *priv_data_offset, + uint64_t max_priv_data_size) +{ + uint64_t svm_priv_data_size, svm_object_md_size, svm_attrs_size; + int nattr_common = 4, nattr_accessibility = 1; + struct criu_svm_metadata *criu_svm_md = NULL; + struct svm_range_list *svms = &p->svms; + uint32_t num_devices; + int ret = 0; + + num_devices = p->n_pdds; + /* Handle one SVM range object at a time, also the number of gpus are + * assumed to be same on the restore node, checking must be done while + * evaluating the topology earlier + */ + + svm_attrs_size = sizeof(struct kfd_ioctl_svm_attribute) * + (nattr_common + nattr_accessibility * num_devices); + svm_object_md_size = sizeof(struct criu_svm_metadata) + svm_attrs_size; + + svm_priv_data_size = sizeof(struct kfd_criu_svm_range_priv_data) + + svm_attrs_size; + + criu_svm_md = kzalloc(svm_object_md_size, GFP_KERNEL); + if (!criu_svm_md) { + pr_err("failed to allocate memory to store svm metadata\n"); + return -ENOMEM; + } + if (*priv_data_offset + svm_priv_data_size > max_priv_data_size) { + ret = -EINVAL; + goto exit; + } + + ret = copy_from_user(&criu_svm_md->data, user_priv_ptr + *priv_data_offset, + svm_priv_data_size); + if (ret) { + ret = -EFAULT; + goto exit; + } + *priv_data_offset += svm_priv_data_size; + + list_add_tail(&criu_svm_md->list, &svms->criu_svm_metadata_list); + + return 0; + + +exit: + kfree(criu_svm_md); + return ret; +} + +int svm_range_get_info(struct kfd_process *p, uint32_t *num_svm_ranges, + uint64_t *svm_priv_data_size) +{ + uint64_t total_size, accessibility_size, common_attr_size; + int nattr_common = 4, nattr_accessibility = 1; + int num_devices = p->n_pdds; + struct svm_range_list *svms; + struct svm_range *prange; + uint32_t count = 0; + + *svm_priv_data_size = 0; + + svms = &p->svms; + if (!svms) + return -EINVAL; + + mutex_lock(&svms->lock); + list_for_each_entry(prange, &svms->list, list) { + pr_debug("prange: 0x%p start: 0x%lx\t npages: 0x%llx\t end: 0x%llx\n", + prange, prange->start, prange->npages, + prange->start + prange->npages - 1); + count++; + } + mutex_unlock(&svms->lock); + + *num_svm_ranges = count; + /* Only the accessbility attributes need to be queried for all the gpus + * individually, remaining ones are spanned across the entire process + * regardless of the various gpu nodes. Of the remaining attributes, + * KFD_IOCTL_SVM_ATTR_CLR_FLAGS need not be saved. + * + * KFD_IOCTL_SVM_ATTR_PREFERRED_LOC + * KFD_IOCTL_SVM_ATTR_PREFETCH_LOC + * KFD_IOCTL_SVM_ATTR_SET_FLAGS + * KFD_IOCTL_SVM_ATTR_GRANULARITY + * + * ** ACCESSBILITY ATTRIBUTES ** + * (Considered as one, type is altered during query, value is gpuid) + * KFD_IOCTL_SVM_ATTR_ACCESS + * KFD_IOCTL_SVM_ATTR_ACCESS_IN_PLACE + * KFD_IOCTL_SVM_ATTR_NO_ACCESS + */ + if (*num_svm_ranges > 0) { + common_attr_size = sizeof(struct kfd_ioctl_svm_attribute) * + nattr_common; + accessibility_size = sizeof(struct kfd_ioctl_svm_attribute) * + nattr_accessibility * num_devices; + + total_size = sizeof(struct kfd_criu_svm_range_priv_data) + + common_attr_size + accessibility_size; + + *svm_priv_data_size = *num_svm_ranges * total_size; + } + + pr_debug("num_svm_ranges %u total_priv_size %llu\n", *num_svm_ranges, + *svm_priv_data_size); + return 0; +} + +int kfd_criu_checkpoint_svm(struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_data_offset) +{ + struct kfd_criu_svm_range_priv_data *svm_priv = NULL; + struct kfd_ioctl_svm_attribute *query_attr = NULL; + uint64_t svm_priv_data_size, query_attr_size = 0; + int index, nattr_common = 4, ret = 0; + struct svm_range_list *svms; + int num_devices = p->n_pdds; + struct svm_range *prange; + struct mm_struct *mm; + + svms = &p->svms; + if (!svms) + return -EINVAL; + + mm = get_task_mm(p->lead_thread); + if (!mm) { + pr_err("failed to get mm for the target process\n"); + return -ESRCH; + } + + query_attr_size = sizeof(struct kfd_ioctl_svm_attribute) * + (nattr_common + num_devices); + + query_attr = kzalloc(query_attr_size, GFP_KERNEL); + if (!query_attr) { + ret = -ENOMEM; + goto exit; + } + + query_attr[0].type = KFD_IOCTL_SVM_ATTR_PREFERRED_LOC; + query_attr[1].type = KFD_IOCTL_SVM_ATTR_PREFETCH_LOC; + query_attr[2].type = KFD_IOCTL_SVM_ATTR_SET_FLAGS; + query_attr[3].type = KFD_IOCTL_SVM_ATTR_GRANULARITY; + + for (index = 0; index < num_devices; index++) { + struct kfd_process_device *pdd = p->pdds[index]; + + query_attr[index + nattr_common].type = + KFD_IOCTL_SVM_ATTR_ACCESS; + query_attr[index + nattr_common].value = pdd->user_gpu_id; + } + + svm_priv_data_size = sizeof(*svm_priv) + query_attr_size; + + svm_priv = kzalloc(svm_priv_data_size, GFP_KERNEL); + if (!svm_priv) { + ret = -ENOMEM; + goto exit_query; + } + + index = 0; + list_for_each_entry(prange, &svms->list, list) { + + svm_priv->object_type = KFD_CRIU_OBJECT_TYPE_SVM_RANGE; + svm_priv->start_addr = prange->start; + svm_priv->size = prange->npages; + memcpy(&svm_priv->attrs, query_attr, query_attr_size); + pr_debug("CRIU: prange: 0x%p start: 0x%lx\t npages: 0x%llx end: 0x%llx\t size: 0x%llx\n", + prange, prange->start, prange->npages, + prange->start + prange->npages - 1, + prange->npages * PAGE_SIZE); + + ret = svm_range_get_attr(p, mm, svm_priv->start_addr, + svm_priv->size, + (nattr_common + num_devices), + svm_priv->attrs); + if (ret) { + pr_err("CRIU: failed to obtain range attributes\n"); + goto exit_priv; + } + + if (copy_to_user(user_priv_data + *priv_data_offset, svm_priv, + svm_priv_data_size)) { + pr_err("Failed to copy svm priv to user\n"); + ret = -EFAULT; + goto exit_priv; + } + + *priv_data_offset += svm_priv_data_size; + + } + + +exit_priv: + kfree(svm_priv); +exit_query: + kfree(query_attr); +exit: + mmput(mm); + return ret; +} + int svm_ioctl(struct kfd_process *p, enum kfd_ioctl_svm_op op, uint64_t start, uint64_t size, uint32_t nattrs, struct kfd_ioctl_svm_attribute *attrs) { + struct mm_struct *mm = current->mm; int r; start >>= PAGE_SHIFT; @@ -3482,10 +3805,10 @@ svm_ioctl(struct kfd_process *p, enum kfd_ioctl_svm_op op, uint64_t start, switch (op) { case KFD_IOCTL_SVM_OP_SET_ATTR: - r = svm_range_set_attr(p, start, size, nattrs, attrs); + r = svm_range_set_attr(p, mm, start, size, nattrs, attrs); break; case KFD_IOCTL_SVM_OP_GET_ATTR: - r = svm_range_get_attr(p, start, size, nattrs, attrs); + r = svm_range_get_attr(p, mm, start, size, nattrs, attrs); break; default: r = EINVAL; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h index 949b477e2f4c..66c77f00ac3e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h @@ -183,6 +183,16 @@ void svm_range_dma_unmap(struct device *dev, dma_addr_t *dma_addr, void svm_range_free_dma_mappings(struct svm_range *prange); void svm_range_prefault(struct svm_range *prange, struct mm_struct *mm, void *owner); +int svm_range_get_info(struct kfd_process *p, uint32_t *num_svm_ranges, + uint64_t *svm_priv_data_size); +int kfd_criu_checkpoint_svm(struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_offset); +int kfd_criu_restore_svm(struct kfd_process *p, + uint8_t __user *user_priv_ptr, + uint64_t *priv_data_offset, + uint64_t max_priv_data_size); +int kfd_criu_resume_svm(struct kfd_process *p); struct kfd_process_device * svm_range_get_pdd_by_adev(struct svm_range *prange, struct amdgpu_device *adev); void svm_range_list_lock_and_flush_work(struct svm_range_list *svms, struct mm_struct *mm); @@ -220,6 +230,35 @@ static inline int svm_range_schedule_evict_svm_bo( return -EINVAL; } +static inline int svm_range_get_info(struct kfd_process *p, + uint32_t *num_svm_ranges, + uint64_t *svm_priv_data_size) +{ + *num_svm_ranges = 0; + *svm_priv_data_size = 0; + return 0; +} + +static inline int kfd_criu_checkpoint_svm(struct kfd_process *p, + uint8_t __user *user_priv_data, + uint64_t *priv_offset) +{ + return 0; +} + +static inline int kfd_criu_restore_svm(struct kfd_process *p, + uint8_t __user *user_priv_ptr, + uint64_t *priv_data_offset, + uint64_t max_priv_data_size) +{ + return -EINVAL; +} + +static inline int kfd_criu_resume_svm(struct kfd_process *p) +{ + return 0; +} + #define KFD_IS_SVM_API_SUPPORTED(dev) false #endif /* IS_ENABLED(CONFIG_HSA_AMD_SVM) */ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 7c1c623ba799..be4852757818 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -76,7 +76,7 @@ #include <drm/drm_atomic.h> #include <drm/drm_atomic_uapi.h> #include <drm/drm_atomic_helper.h> -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_fb_helper.h> #include <drm/drm_fourcc.h> #include <drm/drm_edid.h> @@ -1027,7 +1027,6 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev) const unsigned char *fw_inst_const, *fw_bss_data; uint32_t i, fw_inst_const_size, fw_bss_data_size; bool has_hw_support; - struct dc *dc = adev->dm.dc; if (!dmub_srv) /* DMUB isn't supported on the ASIC. */ @@ -1119,14 +1118,12 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev) for (i = 0; i < fb_info->num_fb; ++i) hw_params.fb[i] = &fb_info->fb[i]; - switch (adev->asic_type) { - case CHIP_YELLOW_CARP: - if (dc->ctx->asic_id.hw_internal_rev != YELLOW_CARP_A0) { - hw_params.dpia_supported = true; + switch (adev->ip_versions[DCE_HWIP][0]) { + case IP_VERSION(3, 1, 3): /* Only for this asic hw internal rev B0 */ + hw_params.dpia_supported = true; #if defined(CONFIG_DRM_AMD_DC_DCN) - hw_params.disable_dpia = dc->debug.dpia_debug.bits.disable_dpia; + hw_params.disable_dpia = adev->dm.dc->debug.dpia_debug.bits.disable_dpia; #endif - } break; default: break; @@ -1496,10 +1493,10 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) init_data.flags.allow_lttpr_non_transparent_mode.bits.DP2_0 = true; #endif - init_data.flags.power_down_display_on_boot = true; + init_data.flags.seamless_boot_edp_requested = false; if (check_seamless_boot_capability(adev)) { - init_data.flags.power_down_display_on_boot = false; + init_data.flags.seamless_boot_edp_requested = true; init_data.flags.allow_seamless_boot_optimization = true; DRM_INFO("Seamless boot condition check passed\n"); } @@ -2179,12 +2176,8 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend) static int amdgpu_dm_smu_write_watermarks_table(struct amdgpu_device *adev) { - struct smu_context *smu = &adev->smu; int ret = 0; - if (!is_support_sw_smu(adev)) - return 0; - /* This interface is for dGPU Navi1x.Linux dc-pplib interface depends * on window driver dc implementation. * For Navi1x, clock settings of dcn watermarks are fixed. the settings @@ -2223,7 +2216,7 @@ static int amdgpu_dm_smu_write_watermarks_table(struct amdgpu_device *adev) return 0; } - ret = smu_write_watermarks_table(smu); + ret = amdgpu_dpm_write_watermarks_table(adev); if (ret) { DRM_ERROR("Failed to update WMTABLE!\n"); return ret; @@ -5856,7 +5849,7 @@ static void fill_stream_properties_from_drm_display_mode( else if (drm_mode_is_420_also(info, mode_in) && aconnector->force_yuv420_output) timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR420; - else if ((connector->display_info.color_formats & DRM_COLOR_FORMAT_YCRCB444) + else if ((connector->display_info.color_formats & DRM_COLOR_FORMAT_YCBCR444) && stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR444; else @@ -6435,8 +6428,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, */ DRM_DEBUG_DRIVER("No preferred mode found\n"); } else { - recalculate_timing = amdgpu_freesync_vid_mode && - is_freesync_video_mode(&mode, aconnector); + recalculate_timing = is_freesync_video_mode(&mode, aconnector); if (recalculate_timing) { freesync_mode = get_highest_refresh_rate_mode(aconnector, false); saved_mode = mode; @@ -6499,7 +6491,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, if (stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED) stream->use_vsc_sdp_for_colorimetry = true; } - mod_build_vsc_infopacket(stream, &stream->vsc_infopacket); + mod_build_vsc_infopacket(stream, &stream->vsc_infopacket, stream->output_color_space); aconnector->psr_skip_count = AMDGPU_DM_PSR_ENTRY_DELAY; } @@ -8143,6 +8135,9 @@ static void amdgpu_dm_connector_add_common_modes(struct drm_encoder *encoder, mode = amdgpu_dm_create_common_mode(encoder, common_modes[i].name, common_modes[i].w, common_modes[i].h); + if (!mode) + continue; + drm_mode_probed_add(connector, mode); amdgpu_dm_connector->num_modes++; } @@ -8304,7 +8299,7 @@ static void amdgpu_dm_connector_add_freesync_modes(struct drm_connector *connect struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); - if (!(amdgpu_freesync_vid_mode && edid)) + if (!edid) return; if (amdgpu_dm_connector->max_vfreq - amdgpu_dm_connector->min_vfreq > 10) @@ -8371,7 +8366,7 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, break; case DRM_MODE_CONNECTOR_DisplayPort: aconnector->base.polled = DRM_CONNECTOR_POLL_HPD; - link->link_enc = dp_get_link_enc(link); + link->link_enc = link_enc_cfg_get_link_enc(link); ASSERT(link->link_enc); if (link->link_enc) aconnector->base.ycbcr_420_allowed = @@ -10271,8 +10266,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, * TODO: Refactor this function to allow this check to work * in all conditions. */ - if (amdgpu_freesync_vid_mode && - dm_new_crtc_state->stream && + if (dm_new_crtc_state->stream && is_timing_unchanged_for_freesync(new_crtc_state, old_crtc_state)) goto skip_modeset; @@ -10307,7 +10301,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, if (!dm_old_crtc_state->stream) goto skip_modeset; - if (amdgpu_freesync_vid_mode && dm_new_crtc_state->stream && + if (dm_new_crtc_state->stream && is_timing_unchanged_for_freesync(new_crtc_state, old_crtc_state)) { new_crtc_state->mode_changed = false; @@ -10319,7 +10313,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, set_freesync_fixed_config(dm_new_crtc_state); goto skip_modeset; - } else if (amdgpu_freesync_vid_mode && aconnector && + } else if (aconnector && is_freesync_video_mode(&new_crtc_state->mode, aconnector)) { struct drm_display_mode *high_mode; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index b9a69b0cef23..28a2b9d476b4 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -29,7 +29,7 @@ #include <drm/drm_atomic.h> #include <drm/drm_connector.h> #include <drm/drm_crtc.h> -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_plane.h> /* @@ -604,6 +604,7 @@ struct amdgpu_dm_connector { #endif bool force_yuv420_output; struct dsc_preferred_settings dsc_settings; + union dp_downstream_port_present mst_downstream_port_present; /* Cached display modes */ struct drm_display_mode freesync_vid_base; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 26719efa5396..bdea177fae55 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -227,8 +227,10 @@ static ssize_t dp_link_settings_read(struct file *f, char __user *buf, break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -389,8 +391,10 @@ static ssize_t dp_phy_settings_read(struct file *f, char __user *buf, break; r = put_user((*(rd_buf + result)), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -1359,8 +1363,10 @@ static ssize_t dp_dsc_clock_en_read(struct file *f, char __user *buf, break; } - if (!pipe_ctx) + if (!pipe_ctx) { + kfree(rd_buf); return -ENXIO; + } dsc = pipe_ctx->stream_res.dsc; if (dsc) @@ -1376,8 +1382,10 @@ static ssize_t dp_dsc_clock_en_read(struct file *f, char __user *buf, break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -1546,8 +1554,10 @@ static ssize_t dp_dsc_slice_width_read(struct file *f, char __user *buf, break; } - if (!pipe_ctx) + if (!pipe_ctx) { + kfree(rd_buf); return -ENXIO; + } dsc = pipe_ctx->stream_res.dsc; if (dsc) @@ -1563,8 +1573,10 @@ static ssize_t dp_dsc_slice_width_read(struct file *f, char __user *buf, break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -1731,8 +1743,10 @@ static ssize_t dp_dsc_slice_height_read(struct file *f, char __user *buf, break; } - if (!pipe_ctx) + if (!pipe_ctx) { + kfree(rd_buf); return -ENXIO; + } dsc = pipe_ctx->stream_res.dsc; if (dsc) @@ -1748,8 +1762,10 @@ static ssize_t dp_dsc_slice_height_read(struct file *f, char __user *buf, break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -1912,8 +1928,10 @@ static ssize_t dp_dsc_bits_per_pixel_read(struct file *f, char __user *buf, break; } - if (!pipe_ctx) + if (!pipe_ctx) { + kfree(rd_buf); return -ENXIO; + } dsc = pipe_ctx->stream_res.dsc; if (dsc) @@ -1929,8 +1947,10 @@ static ssize_t dp_dsc_bits_per_pixel_read(struct file *f, char __user *buf, break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -2088,8 +2108,10 @@ static ssize_t dp_dsc_pic_width_read(struct file *f, char __user *buf, break; } - if (!pipe_ctx) + if (!pipe_ctx) { + kfree(rd_buf); return -ENXIO; + } dsc = pipe_ctx->stream_res.dsc; if (dsc) @@ -2105,8 +2127,10 @@ static ssize_t dp_dsc_pic_width_read(struct file *f, char __user *buf, break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -2145,8 +2169,10 @@ static ssize_t dp_dsc_pic_height_read(struct file *f, char __user *buf, break; } - if (!pipe_ctx) + if (!pipe_ctx) { + kfree(rd_buf); return -ENXIO; + } dsc = pipe_ctx->stream_res.dsc; if (dsc) @@ -2162,8 +2188,10 @@ static ssize_t dp_dsc_pic_height_read(struct file *f, char __user *buf, break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -2217,8 +2245,10 @@ static ssize_t dp_dsc_chunk_size_read(struct file *f, char __user *buf, break; } - if (!pipe_ctx) + if (!pipe_ctx) { + kfree(rd_buf); return -ENXIO; + } dsc = pipe_ctx->stream_res.dsc; if (dsc) @@ -2234,8 +2264,10 @@ static ssize_t dp_dsc_chunk_size_read(struct file *f, char __user *buf, break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -2289,8 +2321,10 @@ static ssize_t dp_dsc_slice_bpg_offset_read(struct file *f, char __user *buf, break; } - if (!pipe_ctx) + if (!pipe_ctx) { + kfree(rd_buf); return -ENXIO; + } dsc = pipe_ctx->stream_res.dsc; if (dsc) @@ -2306,8 +2340,10 @@ static ssize_t dp_dsc_slice_bpg_offset_read(struct file *f, char __user *buf, break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; @@ -2851,7 +2887,7 @@ static ssize_t edp_ilr_write(struct file *f, const char __user *buf, kfree(wr_buf); DRM_DEBUG_DRIVER("Invalid Input value. No HW will be programmed\n"); prefer_link_settings.use_link_rate_set = false; - dc_link_set_preferred_training_settings(dc, NULL, NULL, link, true); + dc_link_set_preferred_training_settings(dc, NULL, NULL, link, false); return size; } @@ -3395,6 +3431,30 @@ static int dp_force_sst_get(void *data, u64 *val) } DEFINE_DEBUGFS_ATTRIBUTE(dp_set_mst_en_for_sst_ops, dp_force_sst_get, dp_force_sst_set, "%llu\n"); + +/* + * Force DP2 sequence without VESA certified cable. + * Example usage: echo 1 > /sys/kernel/debug/dri/0/amdgpu_dm_dp_ignore_cable_id + */ +static int dp_ignore_cable_id_set(void *data, u64 val) +{ + struct amdgpu_device *adev = data; + + adev->dm.dc->debug.ignore_cable_id = val; + + return 0; +} + +static int dp_ignore_cable_id_get(void *data, u64 *val) +{ + struct amdgpu_device *adev = data; + + *val = adev->dm.dc->debug.ignore_cable_id; + + return 0; +} +DEFINE_DEBUGFS_ATTRIBUTE(dp_ignore_cable_id_ops, dp_ignore_cable_id_get, + dp_ignore_cable_id_set, "%llu\n"); #endif /* @@ -3459,8 +3519,10 @@ static ssize_t dcc_en_bits_read( dc->hwss.get_dcc_en_bits(dc, dcc_en_bits); rd_buf = kcalloc(rd_buf_size, sizeof(char), GFP_KERNEL); - if (!rd_buf) + if (!rd_buf) { + kfree(dcc_en_bits); return -ENOMEM; + } for (i = 0; i < num_pipes; i++) offset += snprintf(rd_buf + offset, rd_buf_size - offset, @@ -3473,8 +3535,10 @@ static ssize_t dcc_en_bits_read( if (*pos >= rd_buf_size) break; r = put_user(*(rd_buf + result), buf); - if (r) + if (r) { + kfree(rd_buf); return r; /* r = -EFAULT */ + } buf += 1; size -= 1; *pos += 1; @@ -3509,6 +3573,8 @@ void dtn_debugfs_init(struct amdgpu_device *adev) #if defined(CONFIG_DRM_AMD_DC_DCN) debugfs_create_file("amdgpu_dm_dp_set_mst_en_for_sst", 0644, root, adev, &dp_set_mst_en_for_sst_ops); + debugfs_create_file("amdgpu_dm_dp_ignore_cable_id", 0644, root, adev, + &dp_ignore_cable_id_ops); #endif debugfs_create_file_unsafe("amdgpu_dm_visual_confirm", 0644, root, adev, diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index 29f07c26d080..db4ab01267e4 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -39,6 +39,7 @@ #include "amdgpu_dm_mst_types.h" #include "dm_helpers.h" +#include "ddc_service_types.h" struct monitor_patch_info { unsigned int manufacturer_id; @@ -445,40 +446,24 @@ bool dm_helpers_dp_mst_start_top_mgr( return (drm_dp_mst_topology_mgr_set_mst(&aconnector->mst_mgr, true) == 0); } -void dm_helpers_dp_mst_stop_top_mgr( +bool dm_helpers_dp_mst_stop_top_mgr( struct dc_context *ctx, struct dc_link *link) { struct amdgpu_dm_connector *aconnector = link->priv; - uint8_t i; if (!aconnector) { DRM_ERROR("Failed to find connector for link!"); - return; + return false; } DRM_INFO("DM_MST: stopping TM on aconnector: %p [id: %d]\n", aconnector, aconnector->base.base.id); - if (aconnector->mst_mgr.mst_state == true) { + if (aconnector->mst_mgr.mst_state == true) drm_dp_mst_topology_mgr_set_mst(&aconnector->mst_mgr, false); - for (i = 0; i < MAX_SINKS_PER_LINK; i++) { - if (link->remote_sinks[i] == NULL) - continue; - - if (link->remote_sinks[i]->sink_signal == - SIGNAL_TYPE_DISPLAY_PORT_MST) { - dc_link_remove_remote_sink(link, link->remote_sinks[i]); - - if (aconnector->dc_sink) { - dc_sink_release(aconnector->dc_sink); - aconnector->dc_sink = NULL; - aconnector->dc_link->cur_link_settings.lane_count = 0; - } - } - } - } + return false; } bool dm_helpers_dp_read_dpcd( @@ -552,6 +537,177 @@ bool dm_helpers_submit_i2c( return result; } + +#if defined(CONFIG_DRM_AMD_DC_DCN) +static bool execute_synaptics_rc_command(struct drm_dp_aux *aux, + bool is_write_cmd, + unsigned char cmd, + unsigned int length, + unsigned int offset, + unsigned char *data) +{ + bool success = false; + unsigned char rc_data[16] = {0}; + unsigned char rc_offset[4] = {0}; + unsigned char rc_length[2] = {0}; + unsigned char rc_cmd = 0; + unsigned char rc_result = 0xFF; + unsigned char i = 0; + uint8_t ret = 0; + + if (is_write_cmd) { + // write rc data + memmove(rc_data, data, length); + ret = drm_dp_dpcd_write(aux, SYNAPTICS_RC_DATA, rc_data, sizeof(rc_data)); + } + + // write rc offset + rc_offset[0] = (unsigned char) offset & 0xFF; + rc_offset[1] = (unsigned char) (offset >> 8) & 0xFF; + rc_offset[2] = (unsigned char) (offset >> 16) & 0xFF; + rc_offset[3] = (unsigned char) (offset >> 24) & 0xFF; + ret = drm_dp_dpcd_write(aux, SYNAPTICS_RC_OFFSET, rc_offset, sizeof(rc_offset)); + + // write rc length + rc_length[0] = (unsigned char) length & 0xFF; + rc_length[1] = (unsigned char) (length >> 8) & 0xFF; + ret = drm_dp_dpcd_write(aux, SYNAPTICS_RC_LENGTH, rc_length, sizeof(rc_length)); + + // write rc cmd + rc_cmd = cmd | 0x80; + ret = drm_dp_dpcd_write(aux, SYNAPTICS_RC_COMMAND, &rc_cmd, sizeof(rc_cmd)); + + if (ret < 0) { + DRM_ERROR(" execute_synaptics_rc_command - write cmd ..., err = %d\n", ret); + return false; + } + + // poll until active is 0 + for (i = 0; i < 10; i++) { + drm_dp_dpcd_read(aux, SYNAPTICS_RC_COMMAND, &rc_cmd, sizeof(rc_cmd)); + if (rc_cmd == cmd) + // active is 0 + break; + msleep(10); + } + + // read rc result + drm_dp_dpcd_read(aux, SYNAPTICS_RC_RESULT, &rc_result, sizeof(rc_result)); + success = (rc_result == 0); + + if (success && !is_write_cmd) { + // read rc data + drm_dp_dpcd_read(aux, SYNAPTICS_RC_DATA, data, length); + } + + DC_LOG_DC(" execute_synaptics_rc_command - success = %d\n", success); + + return success; +} + +static void apply_synaptics_fifo_reset_wa(struct drm_dp_aux *aux) +{ + unsigned char data[16] = {0}; + + DC_LOG_DC("Start apply_synaptics_fifo_reset_wa\n"); + + // Step 2 + data[0] = 'P'; + data[1] = 'R'; + data[2] = 'I'; + data[3] = 'U'; + data[4] = 'S'; + + if (!execute_synaptics_rc_command(aux, true, 0x01, 5, 0, data)) + return; + + // Step 3 and 4 + if (!execute_synaptics_rc_command(aux, false, 0x31, 4, 0x220998, data)) + return; + + data[0] &= (~(1 << 1)); // set bit 1 to 0 + if (!execute_synaptics_rc_command(aux, true, 0x21, 4, 0x220998, data)) + return; + + if (!execute_synaptics_rc_command(aux, false, 0x31, 4, 0x220D98, data)) + return; + + data[0] &= (~(1 << 1)); // set bit 1 to 0 + if (!execute_synaptics_rc_command(aux, true, 0x21, 4, 0x220D98, data)) + return; + + if (!execute_synaptics_rc_command(aux, false, 0x31, 4, 0x221198, data)) + return; + + data[0] &= (~(1 << 1)); // set bit 1 to 0 + if (!execute_synaptics_rc_command(aux, true, 0x21, 4, 0x221198, data)) + return; + + // Step 3 and 5 + if (!execute_synaptics_rc_command(aux, false, 0x31, 4, 0x220998, data)) + return; + + data[0] |= (1 << 1); // set bit 1 to 1 + if (!execute_synaptics_rc_command(aux, true, 0x21, 4, 0x220998, data)) + return; + + if (!execute_synaptics_rc_command(aux, false, 0x31, 4, 0x220D98, data)) + return; + + data[0] |= (1 << 1); // set bit 1 to 1 + return; + + if (!execute_synaptics_rc_command(aux, false, 0x31, 4, 0x221198, data)) + return; + + data[0] |= (1 << 1); // set bit 1 to 1 + if (!execute_synaptics_rc_command(aux, true, 0x21, 4, 0x221198, data)) + return; + + // Step 6 + if (!execute_synaptics_rc_command(aux, true, 0x02, 0, 0, NULL)) + return; + + DC_LOG_DC("Done apply_synaptics_fifo_reset_wa\n"); +} + +static uint8_t write_dsc_enable_synaptics_non_virtual_dpcd_mst( + struct drm_dp_aux *aux, + const struct dc_stream_state *stream, + bool enable) +{ + uint8_t ret = 0; + + DC_LOG_DC("Configure DSC to non-virtual dpcd synaptics\n"); + + if (enable) { + /* When DSC is enabled on previous boot and reboot with the hub, + * there is a chance that Synaptics hub gets stuck during reboot sequence. + * Applying a workaround to reset Synaptics SDP fifo before enabling the first stream + */ + if (!stream->link->link_status.link_active && + memcmp(stream->link->dpcd_caps.branch_dev_name, + (int8_t *)SYNAPTICS_DEVICE_ID, 4) == 0) + apply_synaptics_fifo_reset_wa(aux); + + ret = drm_dp_dpcd_write(aux, DP_DSC_ENABLE, &enable, 1); + DRM_INFO("Send DSC enable to synaptics\n"); + + } else { + /* Synaptics hub not support virtual dpcd, + * external monitor occur garbage while disable DSC, + * Disable DSC only when entire link status turn to false, + */ + if (!stream->link->link_status.link_active) { + ret = drm_dp_dpcd_write(aux, DP_DSC_ENABLE, &enable, 1); + DRM_INFO("Send DSC disable to synaptics\n"); + } + } + + return ret; +} +#endif + bool dm_helpers_dp_write_dsc_enable( struct dc_context *ctx, const struct dc_stream_state *stream, @@ -570,7 +726,16 @@ bool dm_helpers_dp_write_dsc_enable( if (!aconnector->dsc_aux) return false; +#if defined(CONFIG_DRM_AMD_DC_DCN) + // apply w/a to synaptics + if (needs_dsc_aux_workaround(aconnector->dc_link) && + (aconnector->mst_downstream_port_present.byte & 0x7) != 0x3) + return write_dsc_enable_synaptics_non_virtual_dpcd_mst( + aconnector->dsc_aux, stream, enable_dsc); +#endif + ret = drm_dp_dpcd_write(aconnector->dsc_aux, DP_DSC_ENABLE, &enable_dsc, 1); + DC_LOG_DC("Send DSC %s to MST RX\n", enable_dsc ? "enable" : "disable"); } if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT || stream->signal == SIGNAL_TYPE_EDP) { @@ -648,14 +813,6 @@ enum dc_edid_status dm_helpers_read_local_edid( /* We don't need the original edid anymore */ kfree(edid); - /* connector->display_info is parsed from EDID and saved - * into drm_connector->display_info - * - * drm_connector->display_info will be used by amdgpu_dm funcs, - * like fill_stream_properties_from_drm_display_mode - */ - amdgpu_dm_update_connector_after_detect(aconnector); - edid_status = dm_helpers_parse_edid_caps( link, &sink->dc_edid, @@ -797,16 +954,12 @@ void dm_helpers_mst_enable_stream_features(const struct dc_stream_state *stream) sizeof(new_downspread)); } -#if defined(CONFIG_DRM_AMD_DC_DCN) void dm_set_phyd32clk(struct dc_context *ctx, int freq_khz) { - // FPGA programming for this clock in diags framework that - // needs to go through dm layer, therefore leave dummy interace here + // TODO } - void dm_helpers_enable_periodic_detection(struct dc_context *ctx, bool enable) { /* TODO: add peridic detection implementation */ } -#endif diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index cc34a35d0bcb..d24be9fb5845 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -25,8 +25,8 @@ #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> -#include <drm/drm_dp_mst_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_helper.h> #include "dm_services.h" #include "amdgpu.h" #include "amdgpu_dm.h" @@ -159,7 +159,7 @@ static const struct drm_connector_funcs dm_dp_mst_connector_funcs = { }; #if defined(CONFIG_DRM_AMD_DC_DCN) -static bool needs_dsc_aux_workaround(struct dc_link *link) +bool needs_dsc_aux_workaround(struct dc_link *link) { if (link->dpcd_caps.branch_dev_id == DP_BRANCH_DEVICE_ID_90CC24 && (link->dpcd_caps.dpcd_rev.raw == DPCD_REV_14 || link->dpcd_caps.dpcd_rev.raw == DPCD_REV_12) && @@ -209,6 +209,25 @@ static bool validate_dsc_caps_on_connector(struct amdgpu_dm_connector *aconnecto return true; } + +bool retrieve_downstream_port_device(struct amdgpu_dm_connector *aconnector) +{ + union dp_downstream_port_present ds_port_present; + + if (!aconnector->dsc_aux) + return false; + + if (drm_dp_dpcd_read(aconnector->dsc_aux, DP_DOWNSTREAMPORT_PRESENT, &ds_port_present, 1) < 0) { + DRM_INFO("Failed to read downstream_port_present 0x05 from DFP of branch device\n"); + return false; + } + + aconnector->mst_downstream_port_present = ds_port_present; + DRM_INFO("Downstream port present %d, type %d\n", + ds_port_present.fields.PORT_PRESENT, ds_port_present.fields.PORT_TYPE); + + return true; +} #endif static int dm_dp_mst_get_modes(struct drm_connector *connector) @@ -289,6 +308,10 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) if (!validate_dsc_caps_on_connector(aconnector)) memset(&aconnector->dc_sink->dsc_caps, 0, sizeof(aconnector->dc_sink->dsc_caps)); + + if (!retrieve_downstream_port_device(aconnector)) + memset(&aconnector->mst_downstream_port_present, + 0, sizeof(aconnector->mst_downstream_port_present)); #endif } } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h index 900d3f7a8498..5da28ca03372 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h @@ -26,6 +26,14 @@ #ifndef __DAL_AMDGPU_DM_MST_TYPES_H__ #define __DAL_AMDGPU_DM_MST_TYPES_H__ +#define DP_BRANCH_DEVICE_ID_90CC24 0x90CC24 + +#define SYNAPTICS_RC_COMMAND 0x4B2 +#define SYNAPTICS_RC_RESULT 0x4B3 +#define SYNAPTICS_RC_LENGTH 0x4B8 +#define SYNAPTICS_RC_OFFSET 0x4BC +#define SYNAPTICS_RC_DATA 0x4C0 + struct amdgpu_display_manager; struct amdgpu_dm_connector; @@ -50,6 +58,8 @@ struct dsc_mst_fairness_vars { bool compute_mst_dsc_configs_for_state(struct drm_atomic_state *state, struct dc_state *dc_state, struct dsc_mst_fairness_vars *vars); + +bool needs_dsc_aux_workaround(struct dc_link *link); #endif #endif diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c index eba270121698..75284e2cec74 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c @@ -99,12 +99,9 @@ bool dm_pp_apply_display_requirements( adev->pm.pm_display_cfg.displays[i].controller_id = dc_cfg->pipe_idx + 1; } - if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->display_configuration_change) - adev->powerplay.pp_funcs->display_configuration_change( - adev->powerplay.pp_handle, - &adev->pm.pm_display_cfg); + amdgpu_dpm_display_configuration_change(adev, &adev->pm.pm_display_cfg); - amdgpu_pm_compute_clocks(adev); + amdgpu_dpm_compute_clocks(adev); } return true; @@ -298,31 +295,25 @@ bool dm_pp_get_clock_levels_by_type( struct dm_pp_clock_levels *dc_clks) { struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; struct amd_pp_clocks pp_clks = { 0 }; struct amd_pp_simple_clock_info validation_clks = { 0 }; uint32_t i; - if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->get_clock_by_type) { - if (adev->powerplay.pp_funcs->get_clock_by_type(pp_handle, - dc_to_pp_clock_type(clk_type), &pp_clks)) { - /* Error in pplib. Provide default values. */ - get_default_clock_levels(clk_type, dc_clks); - return true; - } + if (amdgpu_dpm_get_clock_by_type(adev, + dc_to_pp_clock_type(clk_type), &pp_clks)) { + /* Error in pplib. Provide default values. */ + get_default_clock_levels(clk_type, dc_clks); + return true; } pp_to_dc_clock_levels(&pp_clks, dc_clks, clk_type); - if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->get_display_mode_validation_clocks) { - if (adev->powerplay.pp_funcs->get_display_mode_validation_clocks( - pp_handle, &validation_clks)) { - /* Error in pplib. Provide default values. */ - DRM_INFO("DM_PPLIB: Warning: using default validation clocks!\n"); - validation_clks.engine_max_clock = 72000; - validation_clks.memory_max_clock = 80000; - validation_clks.level = 0; - } + if (amdgpu_dpm_get_display_mode_validation_clks(adev, &validation_clks)) { + /* Error in pplib. Provide default values. */ + DRM_INFO("DM_PPLIB: Warning: using default validation clocks!\n"); + validation_clks.engine_max_clock = 72000; + validation_clks.memory_max_clock = 80000; + validation_clks.level = 0; } DRM_INFO("DM_PPLIB: Validation clocks:\n"); @@ -370,18 +361,14 @@ bool dm_pp_get_clock_levels_by_type_with_latency( struct dm_pp_clock_levels_with_latency *clk_level_info) { struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; struct pp_clock_levels_with_latency pp_clks = { 0 }; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; int ret; - if (pp_funcs && pp_funcs->get_clock_by_type_with_latency) { - ret = pp_funcs->get_clock_by_type_with_latency(pp_handle, - dc_to_pp_clock_type(clk_type), - &pp_clks); - if (ret) - return false; - } + ret = amdgpu_dpm_get_clock_by_type_with_latency(adev, + dc_to_pp_clock_type(clk_type), + &pp_clks); + if (ret) + return false; pp_to_dc_clock_levels_with_latency(&pp_clks, clk_level_info, clk_type); @@ -394,18 +381,14 @@ bool dm_pp_get_clock_levels_by_type_with_voltage( struct dm_pp_clock_levels_with_voltage *clk_level_info) { struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; struct pp_clock_levels_with_voltage pp_clk_info = {0}; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; int ret; - if (pp_funcs && pp_funcs->get_clock_by_type_with_voltage) { - ret = pp_funcs->get_clock_by_type_with_voltage(pp_handle, - dc_to_pp_clock_type(clk_type), - &pp_clk_info); - if (ret) - return false; - } + ret = amdgpu_dpm_get_clock_by_type_with_voltage(adev, + dc_to_pp_clock_type(clk_type), + &pp_clk_info); + if (ret) + return false; pp_to_dc_clock_levels_with_voltage(&pp_clk_info, clk_level_info, clk_type); @@ -417,19 +400,16 @@ bool dm_pp_notify_wm_clock_changes( struct dm_pp_wm_sets_with_clock_ranges *wm_with_clock_ranges) { struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; /* * Limit this watermark setting for Polaris for now * TODO: expand this to other ASICs */ - if ((adev->asic_type >= CHIP_POLARIS10) && (adev->asic_type <= CHIP_VEGAM) - && pp_funcs && pp_funcs->set_watermarks_for_clocks_ranges) { - if (!pp_funcs->set_watermarks_for_clocks_ranges(pp_handle, + if ((adev->asic_type >= CHIP_POLARIS10) && + (adev->asic_type <= CHIP_VEGAM) && + !amdgpu_dpm_set_watermarks_for_clocks_ranges(adev, (void *)wm_with_clock_ranges)) return true; - } return false; } @@ -456,12 +436,10 @@ bool dm_pp_apply_clock_for_voltage_request( if (!pp_clock_request.clock_type) return false; - if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->display_clock_voltage_request) - ret = adev->powerplay.pp_funcs->display_clock_voltage_request( - adev->powerplay.pp_handle, - &pp_clock_request); - if (ret) + ret = amdgpu_dpm_display_clock_voltage_request(adev, &pp_clock_request); + if (ret && (ret != -EOPNOTSUPP)) return false; + return true; } @@ -471,15 +449,8 @@ bool dm_pp_get_static_clocks( { struct amdgpu_device *adev = ctx->driver_context; struct amd_pp_clock_info pp_clk_info = {0}; - int ret = 0; - if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->get_current_clocks) - ret = adev->powerplay.pp_funcs->get_current_clocks( - adev->powerplay.pp_handle, - &pp_clk_info); - else - return false; - if (ret) + if (amdgpu_dpm_get_current_clocks(adev, &pp_clk_info)) return false; static_clk_info->max_clocks_state = pp_to_dc_powerlevel_state(pp_clk_info.max_clocks_state); @@ -494,8 +465,6 @@ static void pp_rv_set_wm_ranges(struct pp_smu *pp, { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; struct dm_pp_wm_sets_with_clock_ranges_soc15 wm_with_clock_ranges; struct dm_pp_clock_range_for_dmif_wm_set_soc15 *wm_dce_clocks = wm_with_clock_ranges.wm_dmif_clocks_ranges; struct dm_pp_clock_range_for_mcif_wm_set_soc15 *wm_soc_clocks = wm_with_clock_ranges.wm_mcif_clocks_ranges; @@ -536,72 +505,48 @@ static void pp_rv_set_wm_ranges(struct pp_smu *pp, ranges->writer_wm_sets[i].min_drain_clk_mhz * 1000; } - if (pp_funcs && pp_funcs->set_watermarks_for_clocks_ranges) - pp_funcs->set_watermarks_for_clocks_ranges(pp_handle, - &wm_with_clock_ranges); + amdgpu_dpm_set_watermarks_for_clocks_ranges(adev, + &wm_with_clock_ranges); } static void pp_rv_set_pme_wa_enable(struct pp_smu *pp) { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - if (pp_funcs && pp_funcs->notify_smu_enable_pwe) - pp_funcs->notify_smu_enable_pwe(pp_handle); + amdgpu_dpm_notify_smu_enable_pwe(adev); } static void pp_rv_set_active_display_count(struct pp_smu *pp, int count) { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - if (!pp_funcs || !pp_funcs->set_active_display_count) - return; - - pp_funcs->set_active_display_count(pp_handle, count); + amdgpu_dpm_set_active_display_count(adev, count); } static void pp_rv_set_min_deep_sleep_dcfclk(struct pp_smu *pp, int clock) { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - - if (!pp_funcs || !pp_funcs->set_min_deep_sleep_dcefclk) - return; - pp_funcs->set_min_deep_sleep_dcefclk(pp_handle, clock); + amdgpu_dpm_set_min_deep_sleep_dcefclk(adev, clock); } static void pp_rv_set_hard_min_dcefclk_by_freq(struct pp_smu *pp, int clock) { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - if (!pp_funcs || !pp_funcs->set_hard_min_dcefclk_by_freq) - return; - - pp_funcs->set_hard_min_dcefclk_by_freq(pp_handle, clock); + amdgpu_dpm_set_hard_min_dcefclk_by_freq(adev, clock); } static void pp_rv_set_hard_min_fclk_by_freq(struct pp_smu *pp, int mhz) { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - - if (!pp_funcs || !pp_funcs->set_hard_min_fclk_by_freq) - return; - pp_funcs->set_hard_min_fclk_by_freq(pp_handle, mhz); + amdgpu_dpm_set_hard_min_fclk_by_freq(adev, mhz); } static enum pp_smu_status pp_nv_set_wm_ranges(struct pp_smu *pp, @@ -609,11 +554,8 @@ static enum pp_smu_status pp_nv_set_wm_ranges(struct pp_smu *pp, { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - if (pp_funcs && pp_funcs->set_watermarks_for_clocks_ranges) - pp_funcs->set_watermarks_for_clocks_ranges(pp_handle, ranges); + amdgpu_dpm_set_watermarks_for_clocks_ranges(adev, ranges); return PP_SMU_RESULT_OK; } @@ -622,14 +564,13 @@ static enum pp_smu_status pp_nv_set_display_count(struct pp_smu *pp, int count) { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; - if (!pp_funcs || !pp_funcs->set_active_display_count) + ret = amdgpu_dpm_set_active_display_count(adev, count); + if (ret == -EOPNOTSUPP) return PP_SMU_RESULT_UNSUPPORTED; - - /* 0: successful or smu.ppt_funcs->set_display_count = NULL; 1: fail */ - if (pp_funcs->set_active_display_count(pp_handle, count)) + else if (ret) + /* 0: successful or smu.ppt_funcs->set_display_count = NULL; 1: fail */ return PP_SMU_RESULT_FAIL; return PP_SMU_RESULT_OK; @@ -640,14 +581,13 @@ pp_nv_set_min_deep_sleep_dcfclk(struct pp_smu *pp, int mhz) { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - - if (!pp_funcs || !pp_funcs->set_min_deep_sleep_dcefclk) - return PP_SMU_RESULT_UNSUPPORTED; + int ret = 0; /* 0: successful or smu.ppt_funcs->set_deep_sleep_dcefclk = NULL;1: fail */ - if (pp_funcs->set_min_deep_sleep_dcefclk(pp_handle, mhz)) + ret = amdgpu_dpm_set_min_deep_sleep_dcefclk(adev, mhz); + if (ret == -EOPNOTSUPP) + return PP_SMU_RESULT_UNSUPPORTED; + else if (ret) return PP_SMU_RESULT_FAIL; return PP_SMU_RESULT_OK; @@ -658,12 +598,8 @@ static enum pp_smu_status pp_nv_set_hard_min_dcefclk_by_freq( { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; struct pp_display_clock_request clock_req; - - if (!pp_funcs || !pp_funcs->display_clock_voltage_request) - return PP_SMU_RESULT_UNSUPPORTED; + int ret = 0; clock_req.clock_type = amd_pp_dcef_clock; clock_req.clock_freq_in_khz = mhz * 1000; @@ -671,7 +607,10 @@ static enum pp_smu_status pp_nv_set_hard_min_dcefclk_by_freq( /* 0: successful or smu.ppt_funcs->display_clock_voltage_request = NULL * 1: fail */ - if (pp_funcs->display_clock_voltage_request(pp_handle, &clock_req)) + ret = amdgpu_dpm_display_clock_voltage_request(adev, &clock_req); + if (ret == -EOPNOTSUPP) + return PP_SMU_RESULT_UNSUPPORTED; + else if (ret) return PP_SMU_RESULT_FAIL; return PP_SMU_RESULT_OK; @@ -682,12 +621,8 @@ pp_nv_set_hard_min_uclk_by_freq(struct pp_smu *pp, int mhz) { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; struct pp_display_clock_request clock_req; - - if (!pp_funcs || !pp_funcs->display_clock_voltage_request) - return PP_SMU_RESULT_UNSUPPORTED; + int ret = 0; clock_req.clock_type = amd_pp_mem_clock; clock_req.clock_freq_in_khz = mhz * 1000; @@ -695,7 +630,10 @@ pp_nv_set_hard_min_uclk_by_freq(struct pp_smu *pp, int mhz) /* 0: successful or smu.ppt_funcs->display_clock_voltage_request = NULL * 1: fail */ - if (pp_funcs->display_clock_voltage_request(pp_handle, &clock_req)) + ret = amdgpu_dpm_display_clock_voltage_request(adev, &clock_req); + if (ret == -EOPNOTSUPP) + return PP_SMU_RESULT_UNSUPPORTED; + else if (ret) return PP_SMU_RESULT_FAIL; return PP_SMU_RESULT_OK; @@ -706,14 +644,10 @@ static enum pp_smu_status pp_nv_set_pstate_handshake_support( { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - if (pp_funcs && pp_funcs->display_disable_memory_clock_switch) { - if (pp_funcs->display_disable_memory_clock_switch(pp_handle, - !pstate_handshake_supported)) - return PP_SMU_RESULT_FAIL; - } + if (amdgpu_dpm_display_disable_memory_clock_switch(adev, + !pstate_handshake_supported)) + return PP_SMU_RESULT_FAIL; return PP_SMU_RESULT_OK; } @@ -723,12 +657,8 @@ static enum pp_smu_status pp_nv_set_voltage_by_freq(struct pp_smu *pp, { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; struct pp_display_clock_request clock_req; - - if (!pp_funcs || !pp_funcs->display_clock_voltage_request) - return PP_SMU_RESULT_UNSUPPORTED; + int ret = 0; switch (clock_id) { case PP_SMU_NV_DISPCLK: @@ -748,7 +678,10 @@ static enum pp_smu_status pp_nv_set_voltage_by_freq(struct pp_smu *pp, /* 0: successful or smu.ppt_funcs->display_clock_voltage_request = NULL * 1: fail */ - if (pp_funcs->display_clock_voltage_request(pp_handle, &clock_req)) + ret = amdgpu_dpm_display_clock_voltage_request(adev, &clock_req); + if (ret == -EOPNOTSUPP) + return PP_SMU_RESULT_UNSUPPORTED; + else if (ret) return PP_SMU_RESULT_FAIL; return PP_SMU_RESULT_OK; @@ -759,16 +692,16 @@ static enum pp_smu_status pp_nv_get_maximum_sustainable_clocks( { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; - if (!pp_funcs || !pp_funcs->get_max_sustainable_clocks_by_dc) + ret = amdgpu_dpm_get_max_sustainable_clocks_by_dc(adev, + max_clocks); + if (ret == -EOPNOTSUPP) return PP_SMU_RESULT_UNSUPPORTED; + else if (ret) + return PP_SMU_RESULT_FAIL; - if (!pp_funcs->get_max_sustainable_clocks_by_dc(pp_handle, max_clocks)) - return PP_SMU_RESULT_OK; - - return PP_SMU_RESULT_FAIL; + return PP_SMU_RESULT_OK; } static enum pp_smu_status pp_nv_get_uclk_dpm_states(struct pp_smu *pp, @@ -776,18 +709,17 @@ static enum pp_smu_status pp_nv_get_uclk_dpm_states(struct pp_smu *pp, { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; - if (!pp_funcs || !pp_funcs->get_uclk_dpm_states) + ret = amdgpu_dpm_get_uclk_dpm_states(adev, + clock_values_in_khz, + num_states); + if (ret == -EOPNOTSUPP) return PP_SMU_RESULT_UNSUPPORTED; + else if (ret) + return PP_SMU_RESULT_FAIL; - if (!pp_funcs->get_uclk_dpm_states(pp_handle, - clock_values_in_khz, - num_states)) - return PP_SMU_RESULT_OK; - - return PP_SMU_RESULT_FAIL; + return PP_SMU_RESULT_OK; } static enum pp_smu_status pp_rn_get_dpm_clock_table( @@ -795,16 +727,15 @@ static enum pp_smu_status pp_rn_get_dpm_clock_table( { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; - if (!pp_funcs || !pp_funcs->get_dpm_clock_table) + ret = amdgpu_dpm_get_dpm_clock_table(adev, clock_table); + if (ret == -EOPNOTSUPP) return PP_SMU_RESULT_UNSUPPORTED; + else if (ret) + return PP_SMU_RESULT_FAIL; - if (!pp_funcs->get_dpm_clock_table(pp_handle, clock_table)) - return PP_SMU_RESULT_OK; - - return PP_SMU_RESULT_FAIL; + return PP_SMU_RESULT_OK; } static enum pp_smu_status pp_rn_set_wm_ranges(struct pp_smu *pp, @@ -812,11 +743,8 @@ static enum pp_smu_status pp_rn_set_wm_ranges(struct pp_smu *pp, { const struct dc_context *ctx = pp->dm; struct amdgpu_device *adev = ctx->driver_context; - void *pp_handle = adev->powerplay.pp_handle; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - if (pp_funcs && pp_funcs->set_watermarks_for_clocks_ranges) - pp_funcs->set_watermarks_for_clocks_ranges(pp_handle, ranges); + amdgpu_dpm_set_watermarks_for_clocks_ranges(adev, ranges); return PP_SMU_RESULT_OK; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c index c510638b4f99..a009fc654ac9 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c @@ -149,10 +149,8 @@ bool amdgpu_dm_link_setup_psr(struct dc_stream_state *stream) link = stream->link; - psr_config.psr_version = link->dpcd_caps.psr_caps.psr_version; - - if (psr_config.psr_version > 0) { - psr_config.psr_exit_link_training_required = 0x1; + if (link->psr_settings.psr_version != DC_PSR_VERSION_UNSUPPORTED) { + psr_config.psr_version = link->psr_settings.psr_version; psr_config.psr_frame_capture_indication_req = 0; psr_config.psr_rfb_setup_time = 0x37; psr_config.psr_sdp_transmit_line_num_deadline = 0x20; diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile index b1f0d6260226..0aaf394b73ff 100644 --- a/drivers/gpu/drm/amd/display/dc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/Makefile @@ -23,12 +23,12 @@ # Makefile for Display Core (dc) component. # -DC_LIBS = basics bios calcs clk_mgr dce gpio irq virtual +DC_LIBS = basics bios dml clk_mgr dce gpio irq link virtual ifdef CONFIG_DRM_AMD_DC_DCN DC_LIBS += dcn20 DC_LIBS += dsc -DC_LIBS += dcn10 dml +DC_LIBS += dcn10 DC_LIBS += dcn21 DC_LIBS += dcn201 DC_LIBS += dcn30 @@ -58,7 +58,7 @@ AMD_DC = $(addsuffix /Makefile, $(addprefix $(FULL_AMD_DISPLAY_PATH)/dc/,$(DC_LI include $(AMD_DC) DISPLAY_CORE = dc.o dc_stat.o dc_link.o dc_resource.o dc_hw_sequencer.o dc_sink.o \ -dc_surface.o dc_link_hwss.o dc_link_dp.o dc_link_ddc.o dc_debug.o dc_stream.o \ +dc_surface.o dc_link_dp.o dc_link_ddc.o dc_debug.o dc_stream.o \ dc_link_enc_cfg.o dc_link_dpia.o dc_link_dpcd.o ifdef CONFIG_DRM_AMD_DC_DCN diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index 1e385d55e7fb..23a3b640f0ee 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -1692,7 +1692,6 @@ static enum bp_result bios_parser_get_encoder_cap_info( ATOM_ENCODER_CAP_RECORD_HBR3_EN) ? 1 : 0; info->HDMI_6GB_EN = (record->encodercaps & ATOM_ENCODER_CAP_RECORD_HDMI6Gbps_EN) ? 1 : 0; -#if defined(CONFIG_DRM_AMD_DC_DCN) info->IS_DP2_CAPABLE = (record->encodercaps & ATOM_ENCODER_CAP_RECORD_DP2) ? 1 : 0; info->DP_UHBR10_EN = (record->encodercaps & @@ -1701,7 +1700,6 @@ static enum bp_result bios_parser_get_encoder_cap_info( ATOM_ENCODER_CAP_RECORD_UHBR13_5_EN) ? 1 : 0; info->DP_UHBR20_EN = (record->encodercaps & ATOM_ENCODER_CAP_RECORD_UHBR20_EN) ? 1 : 0; -#endif info->DP_IS_USB_C = (record->encodercaps & ATOM_ENCODER_CAP_RECORD_USB_C_TYPE) ? 1 : 0; diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index 9afa5eb2e6d3..f52f7ff7ead4 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -338,12 +338,10 @@ static enum bp_result transmitter_control_v1_7( const struct command_table_helper *cmd = bp->cmd_helper; struct dmub_dig_transmitter_control_data_v1_7 dig_v1_7 = {0}; -#if defined(CONFIG_DRM_AMD_DC_DCN) uint8_t hpo_instance = (uint8_t)cntl->hpo_engine_id - ENGINE_ID_HPO_0; if (dc_is_dp_signal(cntl->signal)) hpo_instance = (uint8_t)cntl->hpo_engine_id - ENGINE_ID_HPO_DP_0; -#endif dig_v1_7.phyid = cmd->phy_id_to_atom(cntl->transmitter); dig_v1_7.action = (uint8_t)cntl->action; @@ -358,9 +356,7 @@ static enum bp_result transmitter_control_v1_7( dig_v1_7.hpdsel = cmd->hpd_sel_to_atom(cntl->hpd_sel); dig_v1_7.digfe_sel = cmd->dig_encoder_sel_to_atom(cntl->engine_id); dig_v1_7.connobj_id = (uint8_t)cntl->connector_obj_id.id; -#if defined(CONFIG_DRM_AMD_DC_DCN) dig_v1_7.HPO_instance = hpo_instance; -#endif dig_v1_7.symclk_units.symclk_10khz = cntl->pixel_clock/10; if (cntl->action == TRANSMITTER_CONTROL_ENABLE || diff --git a/drivers/gpu/drm/amd/display/dc/calcs/Makefile b/drivers/gpu/drm/amd/display/dc/calcs/Makefile deleted file mode 100644 index f3c00f479e1c..000000000000 --- a/drivers/gpu/drm/amd/display/dc/calcs/Makefile +++ /dev/null @@ -1,68 +0,0 @@ -# -# Copyright 2017 Advanced Micro Devices, Inc. -# Copyright 2019 Raptor Engineering, LLC -# -# Permission is hereby granted, free of charge, to any person obtaining a -# copy of this software and associated documentation files (the "Software"), -# to deal in the Software without restriction, including without limitation -# the rights to use, copy, modify, merge, publish, distribute, sublicense, -# and/or sell copies of the Software, and to permit persons to whom the -# Software is furnished to do so, subject to the following conditions: -# -# The above copyright notice and this permission notice shall be included in -# all copies or substantial portions of the Software. -# -# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR -# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, -# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL -# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR -# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, -# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR -# OTHER DEALINGS IN THE SOFTWARE. -# -# -# Makefile for the 'calcs' sub-component of DAL. -# It calculates Bandwidth and Watermarks values for HW programming -# - -ifdef CONFIG_X86 -calcs_ccflags := -mhard-float -msse -endif - -ifdef CONFIG_PPC64 -calcs_ccflags := -mhard-float -maltivec -endif - -ifdef CONFIG_CC_IS_GCC -ifeq ($(call cc-ifversion, -lt, 0701, y), y) -IS_OLD_GCC = 1 -endif -endif - -ifdef CONFIG_X86 -ifdef IS_OLD_GCC -# Stack alignment mismatch, proceed with caution. -# GCC < 7.1 cannot compile code using `double` and -mpreferred-stack-boundary=3 -# (8B stack alignment). -calcs_ccflags += -mpreferred-stack-boundary=4 -else -calcs_ccflags += -msse2 -endif -endif - -CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calcs.o := $(calcs_ccflags) -CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_auto.o := $(calcs_ccflags) -CFLAGS_$(AMDDALPATH)/dc/calcs/dcn_calc_math.o := $(calcs_ccflags) -Wno-tautological-compare -CFLAGS_REMOVE_$(AMDDALPATH)/dc/calcs/dcn_calcs.o := $(calcs_rcflags) -CFLAGS_REMOVE_$(AMDDALPATH)/dc/calcs/dcn_calc_auto.o := $(calcs_rcflags) -CFLAGS_REMOVE_$(AMDDALPATH)/dc/calcs/dcn_calc_math.o := $(calcs_rcflags) - -BW_CALCS = dce_calcs.o bw_fixed.o custom_float.o - -ifdef CONFIG_DRM_AMD_DC_DCN -BW_CALCS += dcn_calcs.o dcn_calc_math.o dcn_calc_auto.o -endif - -AMD_DAL_BW_CALCS = $(addprefix $(AMDDALPATH)/dc/calcs/,$(BW_CALCS)) - -AMD_DISPLAY_FILES += $(AMD_DAL_BW_CALCS) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c index f977f29907df..589131d415fd 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c @@ -42,8 +42,8 @@ #include "nbio/nbio_7_4_offset.h" -#include "dcn/dpcs_3_0_0_offset.h" -#include "dcn/dpcs_3_0_0_sh_mask.h" +#include "dpcs/dpcs_3_0_0_offset.h" +#include "dpcs/dpcs_3_0_0_sh_mask.h" #include "mmhub/mmhub_2_0_0_offset.h" #include "mmhub/mmhub_2_0_0_sh_mask.h" @@ -184,6 +184,7 @@ void dcn3_init_clocks(struct clk_mgr *clk_mgr_base) dcn3_init_single_clock(clk_mgr, PPCLK_DCEFCLK, &clk_mgr_base->bw_params->clk_table.entries[0].dcfclk_mhz, &num_levels); + dcn30_smu_set_min_deep_sleep_dcef_clk(clk_mgr, 0); /* DTBCLK */ dcn3_init_single_clock(clk_mgr, PPCLK_DTBCLK, diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c index 9d17c5a5ae01..e17c9938cee5 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c @@ -139,9 +139,9 @@ void dcn31_update_clocks(struct clk_mgr *clk_mgr_base, * also if safe to lower is false, we just go in the higher state */ if (safe_to_lower) { - if (new_clocks->zstate_support == DCN_ZSTATE_SUPPORT_ALLOW && + if (new_clocks->zstate_support != DCN_ZSTATE_SUPPORT_DISALLOW && new_clocks->zstate_support != clk_mgr_base->clks.zstate_support) { - dcn31_smu_set_Z9_support(clk_mgr, true); + dcn31_smu_set_zstate_support(clk_mgr, new_clocks->zstate_support); dm_helpers_enable_periodic_detection(clk_mgr_base->ctx, true); clk_mgr_base->clks.zstate_support = new_clocks->zstate_support; } @@ -167,7 +167,7 @@ void dcn31_update_clocks(struct clk_mgr *clk_mgr_base, } else { if (new_clocks->zstate_support == DCN_ZSTATE_SUPPORT_DISALLOW && new_clocks->zstate_support != clk_mgr_base->clks.zstate_support) { - dcn31_smu_set_Z9_support(clk_mgr, false); + dcn31_smu_set_zstate_support(clk_mgr, DCN_ZSTATE_SUPPORT_DISALLOW); dm_helpers_enable_periodic_detection(clk_mgr_base->ctx, false); clk_mgr_base->clks.zstate_support = new_clocks->zstate_support; } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c index de3f4643eeef..c5d7d075026f 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c @@ -310,23 +310,32 @@ void dcn31_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr) VBIOSSMC_MSG_TransferTableDram2Smu, TABLE_WATERMARKS); } -void dcn31_smu_set_Z9_support(struct clk_mgr_internal *clk_mgr, bool support) +void dcn31_smu_set_zstate_support(struct clk_mgr_internal *clk_mgr, enum dcn_zstate_support_state support) { - //TODO: Work with smu team to define optimization options. - unsigned int msg_id; + unsigned int msg_id, param; if (!clk_mgr->smu_present) return; - if (support) - msg_id = VBIOSSMC_MSG_AllowZstatesEntry; + if (!clk_mgr->base.ctx->dc->debug.enable_z9_disable_interface && + (support == DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY)) + support = DCN_ZSTATE_SUPPORT_DISALLOW; + + + if (support == DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY) + param = 1; else + param = 0; + + if (support == DCN_ZSTATE_SUPPORT_DISALLOW) msg_id = VBIOSSMC_MSG_DisallowZstatesEntry; + else + msg_id = VBIOSSMC_MSG_AllowZstatesEntry; dcn31_smu_send_msg_with_param( clk_mgr, msg_id, - 0); + param); } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.h index cd0b7e1e685f..dfa25a76a6d1 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.h @@ -265,7 +265,7 @@ void dcn31_smu_set_dram_addr_low(struct clk_mgr_internal *clk_mgr, uint32_t addr void dcn31_smu_transfer_dpm_table_smu_2_dram(struct clk_mgr_internal *clk_mgr); void dcn31_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr); -void dcn31_smu_set_Z9_support(struct clk_mgr_internal *clk_mgr, bool support); +void dcn31_smu_set_zstate_support(struct clk_mgr_internal *clk_mgr, enum dcn_zstate_support_state support); void dcn31_smu_set_dtbclk(struct clk_mgr_internal *clk_mgr, bool enable); #endif /* DAL_DC_31_SMU_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index d18e9f3ea998..467f606ba2c7 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -1469,7 +1469,7 @@ static bool context_changed( return false; } -bool dc_validate_seamless_boot_timing(const struct dc *dc, +bool dc_validate_boot_timing(const struct dc *dc, const struct dc_sink *sink, struct dc_crtc_timing *crtc_timing) { @@ -2379,10 +2379,8 @@ static enum surface_update_type check_update_surfaces_for_stream( if (stream_update->dsc_config) su_flags->bits.dsc_changed = 1; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (stream_update->mst_bw_update) su_flags->bits.mst_bw = 1; -#endif if (su_flags->raw != 0) overall_type = UPDATE_TYPE_FULL; @@ -2724,6 +2722,9 @@ static void commit_planes_do_stream_update(struct dc *dc, stream_update->vsp_infopacket) { resource_build_info_frame(pipe_ctx); dc->hwss.update_info_frame(pipe_ctx); + + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + dp_source_sequence_trace(pipe_ctx->stream->link, DPCD_SOURCE_SEQ_AFTER_UPDATE_INFO_FRAME); } if (stream_update->hdr_static_metadata && @@ -2761,14 +2762,12 @@ static void commit_planes_do_stream_update(struct dc *dc, if (stream_update->dsc_config) dp_update_dsc_config(pipe_ctx); -#if defined(CONFIG_DRM_AMD_DC_DCN) if (stream_update->mst_bw_update) { if (stream_update->mst_bw_update->is_increase) dc_link_increase_mst_payload(pipe_ctx, stream_update->mst_bw_update->mst_stream_bw); else dc_link_reduce_mst_payload(pipe_ctx, stream_update->mst_bw_update->mst_stream_bw); } -#endif if (stream_update->pending_test_pattern) { dc_link_dp_set_test_pattern(stream->link, diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index b5e570d33ca9..b1718600fa02 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -720,35 +720,8 @@ static bool detect_dp(struct dc_link *link, sink_caps->signal = SIGNAL_TYPE_DISPLAY_PORT; if (!detect_dp_sink_caps(link)) return false; - if (is_mst_supported(link)) { - sink_caps->signal = SIGNAL_TYPE_DISPLAY_PORT_MST; - link->type = dc_connection_mst_branch; - dal_ddc_service_set_transaction_type(link->ddc, - sink_caps->transaction_type); - -#if defined(CONFIG_DRM_AMD_DC_DCN) - /* Apply work around for tunneled MST on certain USB4 docks. Always use DSC if dock - * reports DSC support. - */ - if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && - link->type == dc_connection_mst_branch && - link->dpcd_caps.branch_dev_id == DP_BRANCH_DEVICE_ID_90CC24 && - link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_SUPPORT && - !link->dc->debug.dpia_debug.bits.disable_mst_dsc_work_around) - link->wa_flags.dpia_mst_dsc_always_on = true; -#endif - -#if defined(CONFIG_DRM_AMD_DC_HDCP) - /* In case of fallback to SST when topology discovery below fails - * HDCP caps will be querried again later by the upper layer (caller - * of this function). */ - query_hdcp_capability(SIGNAL_TYPE_DISPLAY_PORT_MST, link); -#endif - } - - if (link->type != dc_connection_mst_branch && - is_dp_branch_device(link)) + if (is_dp_branch_device(link)) /* DP SST branch */ link->type = dc_connection_sst_branch; } else { @@ -824,15 +797,213 @@ static bool wait_for_entering_dp_alt_mode(struct dc_link *link) return false; } -/* - * dc_link_detect() - Detect if a sink is attached to a given link +static void apply_dpia_mst_dsc_always_on_wa(struct dc_link *link) +{ +#if defined(CONFIG_DRM_AMD_DC_DCN) + /* Apply work around for tunneled MST on certain USB4 docks. Always use DSC if dock + * reports DSC support. + */ + if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && + link->type == dc_connection_mst_branch && + link->dpcd_caps.branch_dev_id == DP_BRANCH_DEVICE_ID_90CC24 && + link->dpcd_caps.branch_hw_revision == DP_BRANCH_HW_REV_20 && + link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_SUPPORT && + !link->dc->debug.dpia_debug.bits.disable_mst_dsc_work_around) + link->wa_flags.dpia_mst_dsc_always_on = true; +#endif +} + +static void revert_dpia_mst_dsc_always_on_wa(struct dc_link *link) +{ + /* Disable work around which keeps DSC on for tunneled MST on certain USB4 docks. */ + if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) + link->wa_flags.dpia_mst_dsc_always_on = false; +} + +static bool discover_dp_mst_topology(struct dc_link *link, enum dc_detect_reason reason) +{ + DC_LOGGER_INIT(link->ctx->logger); + + LINK_INFO("link=%d, mst branch is now Connected\n", + link->link_index); + + apply_dpia_mst_dsc_always_on_wa(link); + link->type = dc_connection_mst_branch; + dm_helpers_dp_update_branch_info(link->ctx, link); + if (dm_helpers_dp_mst_start_top_mgr(link->ctx, + link, (reason == DETECT_REASON_BOOT || reason == DETECT_REASON_RESUMEFROMS3S4))) { + link_disconnect_sink(link); + } else { + link->type = dc_connection_sst_branch; + } + + return link->type == dc_connection_mst_branch; +} + +static bool reset_cur_dp_mst_topology(struct dc_link *link) +{ + bool result = false; + DC_LOGGER_INIT(link->ctx->logger); + + LINK_INFO("link=%d, mst branch is now Disconnected\n", + link->link_index); + + revert_dpia_mst_dsc_always_on_wa(link); + result = dm_helpers_dp_mst_stop_top_mgr(link->ctx, link); + + link->mst_stream_alloc_table.stream_count = 0; + memset(link->mst_stream_alloc_table.stream_allocations, + 0, + sizeof(link->mst_stream_alloc_table.stream_allocations)); + return result; +} + +static bool should_prepare_phy_clocks_for_link_verification(const struct dc *dc, + enum dc_detect_reason reason) +{ + int i; + bool can_apply_seamless_boot = false; + + for (i = 0; i < dc->current_state->stream_count; i++) { + if (dc->current_state->streams[i]->apply_seamless_boot_optimization) { + can_apply_seamless_boot = true; + break; + } + } + + return !can_apply_seamless_boot && reason != DETECT_REASON_BOOT; +} + +static void prepare_phy_clocks_for_destructive_link_verification(const struct dc *dc) +{ +#if defined(CONFIG_DRM_AMD_DC_DCN) + dc_z10_restore(dc); +#endif + clk_mgr_exit_optimized_pwr_state(dc, dc->clk_mgr); +} + +static void restore_phy_clocks_for_destructive_link_verification(const struct dc *dc) +{ + clk_mgr_optimize_pwr_state(dc, dc->clk_mgr); +} + +static void set_all_streams_dpms_off_for_link(struct dc_link *link) +{ + int i; + struct pipe_ctx *pipe_ctx; + struct dc_stream_update stream_update; + bool dpms_off = true; + struct link_resource link_res = {0}; + + memset(&stream_update, 0, sizeof(stream_update)); + stream_update.dpms_off = &dpms_off; + + for (i = 0; i < MAX_PIPES; i++) { + pipe_ctx = &link->dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx && pipe_ctx->stream && !pipe_ctx->stream->dpms_off && + pipe_ctx->stream->link == link && !pipe_ctx->prev_odm_pipe) { + stream_update.stream = pipe_ctx->stream; + dc_commit_updates_for_stream(link->ctx->dc, NULL, 0, + pipe_ctx->stream, &stream_update, + link->ctx->dc->current_state); + } + } + + /* link can be also enabled by vbios. In this case it is not recorded + * in pipe_ctx. Disable link phy here to make sure it is completely off + */ + dp_disable_link_phy(link, &link_res, link->connector_signal); +} + +static void verify_link_capability_destructive(struct dc_link *link, + struct dc_sink *sink, + enum dc_detect_reason reason) +{ + bool should_prepare_phy_clocks = + should_prepare_phy_clocks_for_link_verification(link->dc, reason); + + if (should_prepare_phy_clocks) + prepare_phy_clocks_for_destructive_link_verification(link->dc); + + if (dc_is_dp_signal(link->local_sink->sink_signal)) { + struct dc_link_settings known_limit_link_setting = + dp_get_max_link_cap(link); + set_all_streams_dpms_off_for_link(link); + dp_verify_link_cap_with_retries( + link, &known_limit_link_setting, + LINK_TRAINING_MAX_VERIFY_RETRY); + } else { + ASSERT(0); + } + + if (should_prepare_phy_clocks) + restore_phy_clocks_for_destructive_link_verification(link->dc); +} + +static void verify_link_capability_non_destructive(struct dc_link *link) +{ + if (dc_is_dp_signal(link->local_sink->sink_signal)) { + if (dc_is_embedded_signal(link->local_sink->sink_signal) || + link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) + /* TODO - should we check link encoder's max link caps here? + * How do we know which link encoder to check from? + */ + link->verified_link_cap = link->reported_link_cap; + else + link->verified_link_cap = dp_get_max_link_cap(link); + } +} + +static bool should_verify_link_capability_destructively(struct dc_link *link, + enum dc_detect_reason reason) +{ + bool destrictive = false; + struct dc_link_settings max_link_cap; + bool is_link_enc_unavailable = link->link_enc && + link->dc->res_pool->funcs->link_encs_assign && + !link_enc_cfg_is_link_enc_avail( + link->ctx->dc, + link->link_enc->preferred_engine, + link); + + if (dc_is_dp_signal(link->local_sink->sink_signal)) { + max_link_cap = dp_get_max_link_cap(link); + destrictive = true; + + if (link->dc->debug.skip_detection_link_training || + dc_is_embedded_signal(link->local_sink->sink_signal) || + link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) { + destrictive = false; + } else if (dp_get_link_encoding_format(&max_link_cap) == + DP_8b_10b_ENCODING) { + if (link->dpcd_caps.is_mst_capable || + is_link_enc_unavailable) { + destrictive = false; + } + } + } + + return destrictive; +} + +static void verify_link_capability(struct dc_link *link, struct dc_sink *sink, + enum dc_detect_reason reason) +{ + if (should_verify_link_capability_destructively(link, reason)) + verify_link_capability_destructive(link, sink, reason); + else + verify_link_capability_non_destructive(link); +} + + +/** + * detect_link_and_local_sink() - Detect if a sink is attached to a given link * * link->local_sink is created or destroyed as needed. * - * This does not create remote sinks but will trigger DM - * to start MST detection if a branch is detected. + * This does not create remote sinks. */ -static bool dc_link_detect_helper(struct dc_link *link, +static bool detect_link_and_local_sink(struct dc_link *link, enum dc_detect_reason reason) { struct dc_sink_init_data sink_init_data = { 0 }; @@ -848,9 +1019,7 @@ static bool dc_link_detect_helper(struct dc_link *link, struct dpcd_caps prev_dpcd_caps; enum dc_connection_type new_connection_type = dc_connection_none; enum dc_connection_type pre_connection_type = dc_connection_none; - bool perform_dp_seamless_boot = false; const uint32_t post_oui_delay = 30; // 30ms - struct link_resource link_res = { 0 }; DC_LOGGER_INIT(link->ctx->logger); @@ -862,7 +1031,8 @@ static bool dc_link_detect_helper(struct dc_link *link, (!link->dc->config.allow_edp_hotplug_detection)) && link->local_sink) { // need to re-write OUI and brightness in resume case - if (link->connector_signal == SIGNAL_TYPE_EDP) { + if (link->connector_signal == SIGNAL_TYPE_EDP && + (link->dpcd_sink_ext_caps.bits.oled == 1)) { dpcd_set_source_specific_data(link); msleep(post_oui_delay); dc_link_set_default_brightness_aux(link); @@ -943,61 +1113,6 @@ static bool dc_link_detect_helper(struct dc_link *link, return false; } -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dp_get_link_encoding_format(&link->reported_link_cap) == DP_128b_132b_ENCODING) - link_res.hpo_dp_link_enc = resource_get_hpo_dp_link_enc_for_det_lt( - &link->dc->current_state->res_ctx, - link->dc->res_pool, - link); -#endif - - if (link->type == dc_connection_mst_branch) { - LINK_INFO("link=%d, mst branch is now Connected\n", - link->link_index); - /* Need to setup mst link_cap struct here - * otherwise dc_link_detect() will leave mst link_cap - * empty which leads to allocate_mst_payload() has "0" - * pbn_per_slot value leading to exception on dc_fixpt_div() - */ - dp_verify_mst_link_cap(link, &link_res); - - /* - * This call will initiate MST topology discovery. Which - * will detect MST ports and add new DRM connector DRM - * framework. Then read EDID via remote i2c over aux. In - * the end, will notify DRM detect result and save EDID - * into DRM framework. - * - * .detect is called by .fill_modes. - * .fill_modes is called by user mode ioctl - * DRM_IOCTL_MODE_GETCONNECTOR. - * - * .get_modes is called by .fill_modes. - * - * call .get_modes, AMDGPU DM implementation will create - * new dc_sink and add to dc_link. For long HPD plug - * in/out, MST has its own handle. - * - * Therefore, just after dc_create, link->sink is not - * created for MST until user mode app calls - * DRM_IOCTL_MODE_GETCONNECTOR. - * - * Need check ->sink usages in case ->sink = NULL - * TODO: s3 resume check - */ - - dm_helpers_dp_update_branch_info(link->ctx, link); - if (dm_helpers_dp_mst_start_top_mgr(link->ctx, - link, reason == DETECT_REASON_BOOT)) { - if (prev_sink) - dc_sink_release(prev_sink); - return false; - } else { - link->type = dc_connection_sst_branch; - sink_caps.signal = SIGNAL_TYPE_DISPLAY_PORT; - } - } - /* Active SST downstream branch device unplug*/ if (link->type == dc_connection_sst_branch && link->dpcd_caps.sink_count.bits.SINK_COUNT == 0) { @@ -1018,19 +1133,6 @@ static bool dc_link_detect_helper(struct dc_link *link, if (pre_connection_type == dc_connection_mst_branch && link->type != dc_connection_mst_branch) dm_helpers_dp_mst_stop_top_mgr(link->ctx, link); - - - // For seamless boot, to skip verify link cap, we read UEFI settings and set them as verified. - if (reason == DETECT_REASON_BOOT && - !dc_ctx->dc->config.power_down_display_on_boot && - link->link_status.link_active) - perform_dp_seamless_boot = true; - - if (perform_dp_seamless_boot) { - read_current_link_settings_on_detect(link); - link->verified_link_cap = link->reported_link_cap; - } - break; } @@ -1119,13 +1221,6 @@ static bool dc_link_detect_helper(struct dc_link *link, #if defined(CONFIG_DRM_AMD_DC_HDCP) query_hdcp_capability(sink->sink_signal, link); #endif - - // verify link cap for SST non-seamless boot - if (!perform_dp_seamless_boot) - dp_verify_link_cap_with_retries(link, - &link_res, - &link->reported_link_cap, - LINK_TRAINING_MAX_VERIFY_RETRY); } else { // If edid is the same, then discard new sink and revert back to original sink if (same_edid) { @@ -1185,27 +1280,6 @@ static bool dc_link_detect_helper(struct dc_link *link, } } else { /* From Connected-to-Disconnected. */ - if (link->type == dc_connection_mst_branch) { - LINK_INFO("link=%d, mst branch is now Disconnected\n", - link->link_index); - - /* Disable work around which keeps DSC on for tunneled MST on certain USB4 docks. */ - if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) - link->wa_flags.dpia_mst_dsc_always_on = false; - - dm_helpers_dp_mst_stop_top_mgr(link->ctx, link); - - link->mst_stream_alloc_table.stream_count = 0; - memset(link->mst_stream_alloc_table.stream_allocations, - 0, - sizeof(link->mst_stream_alloc_table.stream_allocations)); - } - -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dp_get_link_encoding_format(&link->cur_link_settings) == DP_128b_132b_ENCODING) - reset_dp_hpo_stream_encoders_for_link(link); -#endif - link->type = dc_connection_none; sink_caps.signal = SIGNAL_TYPE_NONE; /* When we unplug a passive DP-HDMI dongle connection, dongle_max_pix_clk @@ -1214,6 +1288,8 @@ static bool dc_link_detect_helper(struct dc_link *link, * Clear dongle_max_pix_clk on disconnect to fix this */ link->dongle_max_pix_clk = 0; + + dc_link_dp_clear_rx_status(link); } LINK_INFO("link=%d, dc_sink_in=%p is now %s prev_sink=%p edid same=%d\n", @@ -1230,33 +1306,26 @@ static bool dc_link_detect_helper(struct dc_link *link, bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) { - const struct dc *dc = link->dc; - bool ret; - bool can_apply_seamless_boot = false; - int i; + bool is_local_sink_detect_success; + bool is_delegated_to_mst_top_mgr = false; + enum dc_connection_type pre_link_type = link->type; - for (i = 0; i < dc->current_state->stream_count; i++) { - if (dc->current_state->streams[i]->apply_seamless_boot_optimization) { - can_apply_seamless_boot = true; - break; - } - } + is_local_sink_detect_success = detect_link_and_local_sink(link, reason); -#if defined(CONFIG_DRM_AMD_DC_DCN) - dc_z10_restore(dc); -#endif - - /* get out of low power state */ - if (!can_apply_seamless_boot && reason != DETECT_REASON_BOOT) - clk_mgr_exit_optimized_pwr_state(dc, dc->clk_mgr); + if (is_local_sink_detect_success && link->local_sink) + verify_link_capability(link, link->local_sink, reason); - ret = dc_link_detect_helper(link, reason); + if (is_local_sink_detect_success && link->local_sink && + dc_is_dp_signal(link->local_sink->sink_signal) && + link->dpcd_caps.is_mst_capable) + is_delegated_to_mst_top_mgr = discover_dp_mst_topology(link, reason); - /* Go back to power optimized state */ - if (!can_apply_seamless_boot && reason != DETECT_REASON_BOOT) - clk_mgr_optimize_pwr_state(dc, dc->clk_mgr); + if (is_local_sink_detect_success && + pre_link_type == dc_connection_mst_branch && + link->type != dc_connection_mst_branch) + is_delegated_to_mst_top_mgr = reset_cur_dp_mst_topology(link); - return ret; + return is_local_sink_detect_success && !is_delegated_to_mst_top_mgr; } bool dc_link_get_hpd_state(struct dc_link *dc_link) @@ -1586,9 +1655,7 @@ static bool dc_link_construct_legacy(struct dc_link *link, } DC_LOG_DC("BIOS object table - DP_IS_USB_C: %d", link->link_enc->features.flags.bits.DP_IS_USB_C); -#if defined(CONFIG_DRM_AMD_DC_DCN) DC_LOG_DC("BIOS object table - IS_DP2_CAPABLE: %d", link->link_enc->features.flags.bits.IS_DP2_CAPABLE); -#endif /* Update link encoder tracking variables. These are used for the dynamic * assignment of link encoders to streams. @@ -1871,7 +1938,6 @@ static enum dc_status enable_link_dp(struct dc_state *state, if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) do_fallback = true; -#if defined(CONFIG_DRM_AMD_DC_DCN) /* * Temporary w/a to get DP2.0 link rates to work with SST. * TODO DP2.0 - Workaround: Remove w/a if and when the issue is resolved. @@ -1881,7 +1947,6 @@ static enum dc_status enable_link_dp(struct dc_state *state, link->dc->debug.set_mst_en_for_sst) { dp_enable_mst_on_sink(link, true); } -#endif if (pipe_ctx->stream->signal == SIGNAL_TYPE_EDP) { /*in case it is not on*/ @@ -1889,7 +1954,6 @@ static enum dc_status enable_link_dp(struct dc_state *state, link->dc->hwss.edp_wait_for_hpd_ready(link, true); } -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(&link_settings) == DP_128b_132b_ENCODING) { /* TODO - DP2.0 HW: calculate 32 symbol clock for HPO encoder */ } else { @@ -1899,19 +1963,15 @@ static enum dc_status enable_link_dp(struct dc_state *state, state->clk_mgr->funcs->update_clocks(state->clk_mgr, state, false); } -#else - pipe_ctx->stream_res.pix_clk_params.requested_sym_clk = - link_settings.link_rate * LINK_RATE_REF_FREQ_IN_KHZ; - if (state->clk_mgr && !apply_seamless_boot_optimization) - state->clk_mgr->funcs->update_clocks(state->clk_mgr, - state, false); -#endif // during mode switch we do DP_SET_POWER off then on, and OUI is lost dpcd_set_source_specific_data(link); if (link->dpcd_sink_ext_caps.raw != 0) msleep(post_oui_delay); + // similarly, mode switch can cause loss of cable ID + dpcd_update_cable_id(link); + skip_video_pattern = true; if (link_settings.link_rate == LINK_RATE_LOW) @@ -1934,12 +1994,8 @@ static enum dc_status enable_link_dp(struct dc_state *state, else fec_enable = true; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(&link_settings) == DP_8b_10b_ENCODING) dp_set_fec_enable(link, fec_enable); -#else - dp_set_fec_enable(link, fec_enable); -#endif // during mode set we do DP_SET_POWER off then on, aux writes are lost if (link->dpcd_sink_ext_caps.bits.oled == 1 || @@ -2495,9 +2551,7 @@ static void disable_link(struct dc_link *link, const struct link_resource *link_ if (dc_is_dp_signal(signal)) { /* SST DP, eDP */ -#if defined(CONFIG_DRM_AMD_DC_DCN) struct dc_link_settings link_settings = link->cur_link_settings; -#endif if (dc_is_dp_sst_signal(signal)) dp_disable_link_phy(link, link_res, signal); else @@ -2505,15 +2559,10 @@ static void disable_link(struct dc_link *link, const struct link_resource *link_ if (dc_is_dp_sst_signal(signal) || link->mst_stream_alloc_table.stream_count == 0) { -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(&link_settings) == DP_8b_10b_ENCODING) { dp_set_fec_enable(link, false); dp_set_fec_ready(link, link_res, false); } -#else - dp_set_fec_enable(link, false); - dp_set_fec_ready(link, link_res, false); -#endif } } else { if (signal != SIGNAL_TYPE_VIRTUAL) @@ -2696,72 +2745,63 @@ static bool dp_active_dongle_validate_timing( break; } -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dpcd_caps->dongle_type == DISPLAY_DONGLE_DP_HDMI_CONVERTER && dongle_caps->extendedCapValid == true) { -#else - if (dpcd_caps->dongle_type != DISPLAY_DONGLE_DP_HDMI_CONVERTER || - dongle_caps->extendedCapValid == false) - return true; -#endif - - /* Check Pixel Encoding */ - switch (timing->pixel_encoding) { - case PIXEL_ENCODING_RGB: - case PIXEL_ENCODING_YCBCR444: - break; - case PIXEL_ENCODING_YCBCR422: - if (!dongle_caps->is_dp_hdmi_ycbcr422_pass_through) - return false; - break; - case PIXEL_ENCODING_YCBCR420: - if (!dongle_caps->is_dp_hdmi_ycbcr420_pass_through) + /* Check Pixel Encoding */ + switch (timing->pixel_encoding) { + case PIXEL_ENCODING_RGB: + case PIXEL_ENCODING_YCBCR444: + break; + case PIXEL_ENCODING_YCBCR422: + if (!dongle_caps->is_dp_hdmi_ycbcr422_pass_through) + return false; + break; + case PIXEL_ENCODING_YCBCR420: + if (!dongle_caps->is_dp_hdmi_ycbcr420_pass_through) + return false; + break; + default: + /* Invalid Pixel Encoding*/ return false; - break; - default: - /* Invalid Pixel Encoding*/ - return false; - } + } - switch (timing->display_color_depth) { - case COLOR_DEPTH_666: - case COLOR_DEPTH_888: - /*888 and 666 should always be supported*/ - break; - case COLOR_DEPTH_101010: - if (dongle_caps->dp_hdmi_max_bpc < 10) - return false; - break; - case COLOR_DEPTH_121212: - if (dongle_caps->dp_hdmi_max_bpc < 12) + switch (timing->display_color_depth) { + case COLOR_DEPTH_666: + case COLOR_DEPTH_888: + /*888 and 666 should always be supported*/ + break; + case COLOR_DEPTH_101010: + if (dongle_caps->dp_hdmi_max_bpc < 10) + return false; + break; + case COLOR_DEPTH_121212: + if (dongle_caps->dp_hdmi_max_bpc < 12) + return false; + break; + case COLOR_DEPTH_141414: + case COLOR_DEPTH_161616: + default: + /* These color depths are currently not supported */ return false; - break; - case COLOR_DEPTH_141414: - case COLOR_DEPTH_161616: - default: - /* These color depths are currently not supported */ - return false; - } + } #if defined(CONFIG_DRM_AMD_DC_DCN) - if (dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps > 0) { // DP to HDMI FRL converter - struct dc_crtc_timing outputTiming = *timing; + if (dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps > 0) { // DP to HDMI FRL converter + struct dc_crtc_timing outputTiming = *timing; - if (timing->flags.DSC && !timing->dsc_cfg.is_frl) - /* DP input has DSC, HDMI FRL output doesn't have DSC, remove DSC from output timing */ - outputTiming.flags.DSC = 0; - if (dc_bandwidth_in_kbps_from_timing(&outputTiming) > dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps) - return false; - } else { // DP to HDMI TMDS converter + if (timing->flags.DSC && !timing->dsc_cfg.is_frl) + /* DP input has DSC, HDMI FRL output doesn't have DSC, remove DSC from output timing */ + outputTiming.flags.DSC = 0; + if (dc_bandwidth_in_kbps_from_timing(&outputTiming) > dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps) + return false; + } else { // DP to HDMI TMDS converter + if (get_timing_pixel_clock_100hz(timing) > (dongle_caps->dp_hdmi_max_pixel_clk_in_khz * 10)) + return false; + } +#else if (get_timing_pixel_clock_100hz(timing) > (dongle_caps->dp_hdmi_max_pixel_clk_in_khz * 10)) return false; - } -#else - if (get_timing_pixel_clock_100hz(timing) > (dongle_caps->dp_hdmi_max_pixel_clk_in_khz * 10)) - return false; #endif - -#if defined(CONFIG_DRM_AMD_DC_DCN) } if (dpcd_caps->channel_coding_cap.bits.DP_128b_132b_SUPPORTED == 0 && @@ -2842,7 +2882,6 @@ static bool dp_active_dongle_validate_timing( return false; } } -#endif return true; } @@ -3315,9 +3354,7 @@ static struct fixed31_32 get_pbn_from_timing(struct pipe_ctx *pipe_ctx) static void update_mst_stream_alloc_table( struct dc_link *link, struct stream_encoder *stream_enc, -#if defined(CONFIG_DRM_AMD_DC_DCN) struct hpo_dp_stream_encoder *hpo_dp_stream_enc, // TODO: Rename stream_enc to dio_stream_enc? -#endif const struct dp_mst_stream_allocation_table *proposed_table) { struct link_mst_stream_allocation work_table[MAX_CONTROLLER_NUM] = { 0 }; @@ -3353,9 +3390,7 @@ static void update_mst_stream_alloc_table( work_table[i].slot_count = proposed_table->stream_allocations[i].slot_count; work_table[i].stream_enc = stream_enc; -#if defined(CONFIG_DRM_AMD_DC_DCN) work_table[i].hpo_dp_stream_enc = hpo_dp_stream_enc; -#endif } } @@ -3366,7 +3401,7 @@ static void update_mst_stream_alloc_table( link->mst_stream_alloc_table.stream_allocations[i] = work_table[i]; } -#if defined(CONFIG_DRM_AMD_DC_DCN) + static void dc_log_vcp_x_y(const struct dc_link *link, struct fixed31_32 avg_time_slots_per_mtp) { const uint32_t VCP_Y_PRECISION = 1000; @@ -3402,6 +3437,8 @@ static enum dc_status dc_link_update_sst_payload(struct pipe_ctx *pipe_ctx, struct hpo_dp_stream_encoder *hpo_dp_stream_encoder = pipe_ctx->stream_res.hpo_dp_stream_enc; struct link_mst_stream_allocation_table proposed_table = {0}; struct fixed31_32 avg_time_slots_per_mtp; + const struct dc_link_settings empty_link_settings = {0}; + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); DC_LOGGER_INIT(link->ctx->logger); /* slot X.Y for SST payload deallocate */ @@ -3410,10 +3447,13 @@ static enum dc_status dc_link_update_sst_payload(struct pipe_ctx *pipe_ctx, dc_log_vcp_x_y(link, avg_time_slots_per_mtp); - hpo_dp_link_encoder->funcs->set_throttled_vcp_size( - hpo_dp_link_encoder, - hpo_dp_stream_encoder->inst, - avg_time_slots_per_mtp); + if (link_hwss->ext.set_throttled_vcp_size) + link_hwss->ext.set_throttled_vcp_size(pipe_ctx, + avg_time_slots_per_mtp); + if (link_hwss->ext.set_hblank_min_symbol_width) + link_hwss->ext.set_hblank_min_symbol_width(pipe_ctx, + &empty_link_settings, + avg_time_slots_per_mtp); } /* calculate VC payload and update branch with new payload allocation table*/ @@ -3457,10 +3497,13 @@ static enum dc_status dc_link_update_sst_payload(struct pipe_ctx *pipe_ctx, dc_log_vcp_x_y(link, avg_time_slots_per_mtp); - hpo_dp_link_encoder->funcs->set_throttled_vcp_size( - hpo_dp_link_encoder, - hpo_dp_stream_encoder->inst, - avg_time_slots_per_mtp); + if (link_hwss->ext.set_throttled_vcp_size) + link_hwss->ext.set_throttled_vcp_size(pipe_ctx, + avg_time_slots_per_mtp); + if (link_hwss->ext.set_hblank_min_symbol_width) + link_hwss->ext.set_hblank_min_symbol_width(pipe_ctx, + &link->cur_link_settings, + avg_time_slots_per_mtp); } /* Always return DC_OK. @@ -3468,7 +3511,6 @@ static enum dc_status dc_link_update_sst_payload(struct pipe_ctx *pipe_ctx, */ return DC_OK; } -#endif /* convert link_mst_stream_alloc_table to dm dp_mst_stream_alloc_table * because stream_encoder is not exposed to dm @@ -3478,24 +3520,17 @@ enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx) struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; struct link_encoder *link_encoder = NULL; - struct stream_encoder *stream_encoder = pipe_ctx->stream_res.stream_enc; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct hpo_dp_link_encoder *hpo_dp_link_encoder = pipe_ctx->link_res.hpo_dp_link_enc; - struct hpo_dp_stream_encoder *hpo_dp_stream_encoder = pipe_ctx->stream_res.hpo_dp_stream_enc; -#endif struct dp_mst_stream_allocation_table proposed_table = {0}; struct fixed31_32 avg_time_slots_per_mtp; struct fixed31_32 pbn; struct fixed31_32 pbn_per_slot; int i; enum act_return_status ret; + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); DC_LOGGER_INIT(link->ctx->logger); - /* Link encoder may have been dynamically assigned to non-physical display endpoint. */ - if (link->ep_type == DISPLAY_ENDPOINT_PHY) - link_encoder = link->link_enc; - else if (link->dc->res_pool->funcs->link_encs_assign) - link_encoder = link_enc_cfg_get_link_enc_used_by_stream(pipe_ctx->stream->ctx->dc, stream); + link_encoder = link_enc_cfg_get_link_enc(link); ASSERT(link_encoder); /* enable_link_dp_mst already check link->enabled_stream_count @@ -3508,17 +3543,12 @@ enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx) stream->ctx, stream, &proposed_table, - true)) { + true)) update_mst_stream_alloc_table( -#if defined(CONFIG_DRM_AMD_DC_DCN) link, pipe_ctx->stream_res.stream_enc, pipe_ctx->stream_res.hpo_dp_stream_enc, &proposed_table); -#else - link, pipe_ctx->stream_res.stream_enc, &proposed_table); -#endif - } else DC_LOG_WARNING("Failed to update" "MST allocation table for" @@ -3531,7 +3561,6 @@ enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx) link->mst_stream_alloc_table.stream_count); for (i = 0; i < MAX_CONTROLLER_NUM; i++) { -#if defined(CONFIG_DRM_AMD_DC_DCN) DC_LOG_MST("stream_enc[%d]: %p " "stream[%d].hpo_dp_stream_enc: %p " "stream[%d].vcp_id: %d " @@ -3544,17 +3573,6 @@ enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx) link->mst_stream_alloc_table.stream_allocations[i].vcp_id, i, link->mst_stream_alloc_table.stream_allocations[i].slot_count); -#else - DC_LOG_MST("stream_enc[%d]: %p " - "stream[%d].vcp_id: %d " - "stream[%d].slot_count: %d\n", - i, - (void *) link->mst_stream_alloc_table.stream_allocations[i].stream_enc, - i, - link->mst_stream_alloc_table.stream_allocations[i].vcp_id, - i, - link->mst_stream_alloc_table.stream_allocations[i].slot_count); -#endif } ASSERT(proposed_table.stream_count > 0); @@ -3574,7 +3592,6 @@ enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx) } /* program DP source TX for payload */ -#if defined(CONFIG_DRM_AMD_DC_DCN) switch (dp_get_link_encoding_format(&link->cur_link_settings)) { case DP_8b_10b_ENCODING: link_encoder->funcs->update_mst_stream_allocation_table( @@ -3590,11 +3607,6 @@ enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx) DC_LOG_ERROR("Failure: unknown encoding format\n"); return DC_ERROR_UNEXPECTED; } -#else - link_encoder->funcs->update_mst_stream_allocation_table( - link_encoder, - &link->mst_stream_alloc_table); -#endif /* send down message */ ret = dm_helpers_dp_mst_poll_for_allocation_change_trigger( @@ -3617,34 +3629,19 @@ enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx) pbn = get_pbn_from_timing(pipe_ctx); avg_time_slots_per_mtp = dc_fixpt_div(pbn, pbn_per_slot); -#if defined(CONFIG_DRM_AMD_DC_DCN) - switch (dp_get_link_encoding_format(&link->cur_link_settings)) { - case DP_8b_10b_ENCODING: - stream_encoder->funcs->set_throttled_vcp_size( - stream_encoder, - avg_time_slots_per_mtp); - break; - case DP_128b_132b_ENCODING: - hpo_dp_link_encoder->funcs->set_throttled_vcp_size( - hpo_dp_link_encoder, - hpo_dp_stream_encoder->inst, + dc_log_vcp_x_y(link, avg_time_slots_per_mtp); + + if (link_hwss->ext.set_throttled_vcp_size) + link_hwss->ext.set_throttled_vcp_size(pipe_ctx, avg_time_slots_per_mtp); + if (link_hwss->ext.set_hblank_min_symbol_width) + link_hwss->ext.set_hblank_min_symbol_width(pipe_ctx, + &link->cur_link_settings, avg_time_slots_per_mtp); - break; - case DP_UNKNOWN_ENCODING: - DC_LOG_ERROR("Failure: unknown encoding format\n"); - return DC_ERROR_UNEXPECTED; - } -#else - stream_encoder->funcs->set_throttled_vcp_size( - stream_encoder, - avg_time_slots_per_mtp); -#endif return DC_OK; } -#if defined(CONFIG_DRM_AMD_DC_DCN) enum dc_status dc_link_reduce_mst_payload(struct pipe_ctx *pipe_ctx, uint32_t bw_in_kbps) { struct dc_stream_state *stream = pipe_ctx->stream; @@ -3653,10 +3650,10 @@ enum dc_status dc_link_reduce_mst_payload(struct pipe_ctx *pipe_ctx, uint32_t bw struct fixed31_32 pbn; struct fixed31_32 pbn_per_slot; struct link_encoder *link_encoder = link->link_enc; - struct stream_encoder *stream_encoder = pipe_ctx->stream_res.stream_enc; struct dp_mst_stream_allocation_table proposed_table = {0}; uint8_t i; enum act_return_status ret; + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); DC_LOGGER_INIT(link->ctx->logger); /* decrease throttled vcp size */ @@ -3664,8 +3661,11 @@ enum dc_status dc_link_reduce_mst_payload(struct pipe_ctx *pipe_ctx, uint32_t bw pbn = get_pbn_from_bw_in_kbps(bw_in_kbps); avg_time_slots_per_mtp = dc_fixpt_div(pbn, pbn_per_slot); - stream_encoder->funcs->set_throttled_vcp_size( - stream_encoder, + if (link_hwss->ext.set_throttled_vcp_size) + link_hwss->ext.set_throttled_vcp_size(pipe_ctx, avg_time_slots_per_mtp); + if (link_hwss->ext.set_hblank_min_symbol_width) + link_hwss->ext.set_hblank_min_symbol_width(pipe_ctx, + &link->cur_link_settings, avg_time_slots_per_mtp); /* send ALLOCATE_PAYLOAD sideband message with updated pbn */ @@ -3733,10 +3733,10 @@ enum dc_status dc_link_increase_mst_payload(struct pipe_ctx *pipe_ctx, uint32_t struct fixed31_32 pbn; struct fixed31_32 pbn_per_slot; struct link_encoder *link_encoder = link->link_enc; - struct stream_encoder *stream_encoder = pipe_ctx->stream_res.stream_enc; struct dp_mst_stream_allocation_table proposed_table = {0}; uint8_t i; enum act_return_status ret; + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); DC_LOGGER_INIT(link->ctx->logger); /* notify immediate branch device table update */ @@ -3795,35 +3795,31 @@ enum dc_status dc_link_increase_mst_payload(struct pipe_ctx *pipe_ctx, uint32_t pbn_per_slot = get_pbn_per_slot(stream); avg_time_slots_per_mtp = dc_fixpt_div(pbn, pbn_per_slot); - stream_encoder->funcs->set_throttled_vcp_size( - stream_encoder, + if (link_hwss->ext.set_throttled_vcp_size) + link_hwss->ext.set_throttled_vcp_size(pipe_ctx, avg_time_slots_per_mtp); + if (link_hwss->ext.set_hblank_min_symbol_width) + link_hwss->ext.set_hblank_min_symbol_width(pipe_ctx, + &link->cur_link_settings, avg_time_slots_per_mtp); return DC_OK; } -#endif static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx) { struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; struct link_encoder *link_encoder = NULL; - struct stream_encoder *stream_encoder = pipe_ctx->stream_res.stream_enc; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct hpo_dp_link_encoder *hpo_dp_link_encoder = pipe_ctx->link_res.hpo_dp_link_enc; - struct hpo_dp_stream_encoder *hpo_dp_stream_encoder = pipe_ctx->stream_res.hpo_dp_stream_enc; -#endif struct dp_mst_stream_allocation_table proposed_table = {0}; struct fixed31_32 avg_time_slots_per_mtp = dc_fixpt_from_int(0); int i; bool mst_mode = (link->type == dc_connection_mst_branch); + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); + const struct dc_link_settings empty_link_settings = {0}; DC_LOGGER_INIT(link->ctx->logger); - /* Link encoder may have been dynamically assigned to non-physical display endpoint. */ - if (link->ep_type == DISPLAY_ENDPOINT_PHY) - link_encoder = link->link_enc; - else if (link->dc->res_pool->funcs->link_encs_assign) - link_encoder = link_enc_cfg_get_link_enc_used_by_stream(pipe_ctx->stream->ctx->dc, stream); + link_encoder = link_enc_cfg_get_link_enc(link); ASSERT(link_encoder); /* deallocate_mst_payload is called before disable link. When mode or @@ -3834,28 +3830,12 @@ static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx) */ /* slot X.Y */ -#if defined(CONFIG_DRM_AMD_DC_DCN) - switch (dp_get_link_encoding_format(&link->cur_link_settings)) { - case DP_8b_10b_ENCODING: - stream_encoder->funcs->set_throttled_vcp_size( - stream_encoder, - avg_time_slots_per_mtp); - break; - case DP_128b_132b_ENCODING: - hpo_dp_link_encoder->funcs->set_throttled_vcp_size( - hpo_dp_link_encoder, - hpo_dp_stream_encoder->inst, + if (link_hwss->ext.set_throttled_vcp_size) + link_hwss->ext.set_throttled_vcp_size(pipe_ctx, avg_time_slots_per_mtp); + if (link_hwss->ext.set_hblank_min_symbol_width) + link_hwss->ext.set_hblank_min_symbol_width(pipe_ctx, + &empty_link_settings, avg_time_slots_per_mtp); - break; - case DP_UNKNOWN_ENCODING: - DC_LOG_ERROR("Failure: unknown encoding format\n"); - return DC_ERROR_UNEXPECTED; - } -#else - stream_encoder->funcs->set_throttled_vcp_size( - stream_encoder, - avg_time_slots_per_mtp); -#endif /* TODO: which component is responsible for remove payload table? */ if (mst_mode) { @@ -3865,16 +3845,11 @@ static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx) &proposed_table, false)) { -#if defined(CONFIG_DRM_AMD_DC_DCN) update_mst_stream_alloc_table( link, pipe_ctx->stream_res.stream_enc, pipe_ctx->stream_res.hpo_dp_stream_enc, &proposed_table); -#else - update_mst_stream_alloc_table( - link, pipe_ctx->stream_res.stream_enc, &proposed_table); -#endif } else { DC_LOG_WARNING("Failed to update" @@ -3890,7 +3865,6 @@ static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx) link->mst_stream_alloc_table.stream_count); for (i = 0; i < MAX_CONTROLLER_NUM; i++) { -#if defined(CONFIG_DRM_AMD_DC_DCN) DC_LOG_MST("stream_enc[%d]: %p " "stream[%d].hpo_dp_stream_enc: %p " "stream[%d].vcp_id: %d " @@ -3903,17 +3877,6 @@ static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx) link->mst_stream_alloc_table.stream_allocations[i].vcp_id, i, link->mst_stream_alloc_table.stream_allocations[i].slot_count); -#else - DC_LOG_MST("stream_enc[%d]: %p " - "stream[%d].vcp_id: %d " - "stream[%d].slot_count: %d\n", - i, - (void *) link->mst_stream_alloc_table.stream_allocations[i].stream_enc, - i, - link->mst_stream_alloc_table.stream_allocations[i].vcp_id, - i, - link->mst_stream_alloc_table.stream_allocations[i].slot_count); -#endif } if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) { @@ -3930,7 +3893,6 @@ static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx) status, mst_alloc_slots, prev_mst_slots_in_use); } -#if defined(CONFIG_DRM_AMD_DC_DCN) switch (dp_get_link_encoding_format(&link->cur_link_settings)) { case DP_8b_10b_ENCODING: link_encoder->funcs->update_mst_stream_allocation_table( @@ -3946,11 +3908,6 @@ static enum dc_status deallocate_mst_payload(struct pipe_ctx *pipe_ctx) DC_LOG_ERROR("Failure: unknown encoding format\n"); return DC_ERROR_UNEXPECTED; } -#else - link_encoder->funcs->update_mst_stream_allocation_table( - link_encoder, - &link->mst_stream_alloc_table); -#endif if (mst_mode) { dm_helpers_dp_mst_poll_for_allocation_change_trigger( @@ -3979,13 +3936,7 @@ static void update_psp_stream_config(struct pipe_ctx *pipe_ctx, bool dpms_off) if (cp_psp == NULL || cp_psp->funcs.update_stream_config == NULL) return; - if (pipe_ctx->stream->link->ep_type == DISPLAY_ENDPOINT_PHY) - link_enc = pipe_ctx->stream->link->link_enc; - else if (pipe_ctx->stream->link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && - pipe_ctx->stream->link->dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_stream( - pipe_ctx->stream->ctx->dc, - pipe_ctx->stream); + link_enc = link_enc_cfg_get_link_enc(pipe_ctx->stream->link); ASSERT(link_enc); if (link_enc == NULL) return; @@ -3998,21 +3949,18 @@ static void update_psp_stream_config(struct pipe_ctx *pipe_ctx, bool dpms_off) /* stream encoder index */ config.stream_enc_idx = pipe_ctx->stream_res.stream_enc->id - ENGINE_ID_DIGA; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) config.stream_enc_idx = pipe_ctx->stream_res.hpo_dp_stream_enc->id - ENGINE_ID_HPO_DP_0; -#endif /* dig back end */ config.dig_be = pipe_ctx->stream->link->link_enc_hw_inst; /* link encoder index */ config.link_enc_idx = link_enc->transmitter - TRANSMITTER_UNIPHY_A; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) config.link_enc_idx = pipe_ctx->link_res.hpo_dp_link_enc->inst; -#endif + /* dio output index */ config.dio_output_idx = link_enc->transmitter - TRANSMITTER_UNIPHY_A; @@ -4027,9 +3975,7 @@ static void update_psp_stream_config(struct pipe_ctx *pipe_ctx, bool dpms_off) config.assr_enabled = (panel_mode == DP_PANEL_MODE_EDP) ? 1 : 0; config.mst_enabled = (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) ? 1 : 0; -#if defined(CONFIG_DRM_AMD_DC_DCN) config.dp2_enabled = is_dp_128b_132b_signal(pipe_ctx) ? 1 : 0; -#endif config.usb4_enabled = (pipe_ctx->stream->link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) ? 1 : 0; config.dpms_off = dpms_off; @@ -4041,7 +3987,6 @@ static void update_psp_stream_config(struct pipe_ctx *pipe_ctx, bool dpms_off) } #endif -#if defined(CONFIG_DRM_AMD_DC_DCN) static void fpga_dp_hpo_enable_link_and_stream(struct dc_state *state, struct pipe_ctx *pipe_ctx) { struct dc *dc = pipe_ctx->stream->ctx->dc; @@ -4050,15 +3995,17 @@ static void fpga_dp_hpo_enable_link_and_stream(struct dc_state *state, struct pi struct fixed31_32 avg_time_slots_per_mtp; uint8_t req_slot_count = 0; uint8_t vc_id = 1; /// VC ID always 1 for SST - struct dc_link_settings link_settings = {0}; + const struct link_hwss *link_hwss = get_link_hwss(stream->link, &pipe_ctx->link_res); DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger); decide_link_settings(stream, &link_settings); stream->link->cur_link_settings = link_settings; - /* Enable clock, Configure lane count, and Enable Link Encoder*/ - enable_dp_hpo_output(stream->link, &pipe_ctx->link_res, &stream->link->cur_link_settings); + if (link_hwss->ext.enable_dp_link_output) + link_hwss->ext.enable_dp_link_output(stream->link, &pipe_ctx->link_res, + stream->signal, pipe_ctx->clock_source->id, + &link_settings); #ifdef DIAGS_BUILD /* Workaround for FPGA HPO capture DP link data: @@ -4112,16 +4059,11 @@ static void fpga_dp_hpo_enable_link_and_stream(struct dc_state *state, struct pi pipe_ctx->link_res.hpo_dp_link_enc, &proposed_table); - pipe_ctx->link_res.hpo_dp_link_enc->funcs->set_throttled_vcp_size( - pipe_ctx->link_res.hpo_dp_link_enc, - pipe_ctx->stream_res.hpo_dp_stream_enc->inst, - avg_time_slots_per_mtp); - - + if (link_hwss->ext.set_throttled_vcp_size) + link_hwss->ext.set_throttled_vcp_size(pipe_ctx, avg_time_slots_per_mtp); dc->hwss.unblank_stream(pipe_ctx, &stream->link->cur_link_settings); } -#endif void core_link_enable_stream( struct dc_state *state, @@ -4132,31 +4074,23 @@ void core_link_enable_stream( struct dc_link *link = stream->sink->link; enum dc_status status; struct link_encoder *link_enc; -#if defined(CONFIG_DRM_AMD_DC_DCN) enum otg_out_mux_dest otg_out_dest = OUT_MUX_DIO; struct vpg *vpg = pipe_ctx->stream_res.stream_enc->vpg; if (is_dp_128b_132b_signal(pipe_ctx)) vpg = pipe_ctx->stream_res.hpo_dp_stream_enc->vpg; -#endif + DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger); if (!IS_DIAG_DC(dc->ctx->dce_environment) && dc_is_virtual_signal(pipe_ctx->stream->signal)) return; - if (dc->res_pool->funcs->link_encs_assign && stream->link->ep_type != DISPLAY_ENDPOINT_PHY) - link_enc = link_enc_cfg_get_link_enc_used_by_stream(dc, stream); - else - link_enc = stream->link->link_enc; + link_enc = link_enc_cfg_get_link_enc(link); ASSERT(link_enc); -#if defined(CONFIG_DRM_AMD_DC_DCN) if (!dc_is_virtual_signal(pipe_ctx->stream->signal) && !is_dp_128b_132b_signal(pipe_ctx)) { -#else - if (!dc_is_virtual_signal(pipe_ctx->stream->signal)) { -#endif if (link_enc) link_enc->funcs->setup( link_enc, @@ -4167,7 +4101,6 @@ void core_link_enable_stream( stream->timing.timing_3d_format != TIMING_3D_FORMAT_NONE); } -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) { pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->set_stream_attribute( pipe_ctx->stream_res.hpo_dp_stream_enc, @@ -4185,14 +4118,6 @@ void core_link_enable_stream( stream->use_vsc_sdp_for_colorimetry, stream->link->dpcd_caps.dprx_feature.bits.SST_SPLIT_SDP_CAP); } -#else - pipe_ctx->stream_res.stream_enc->funcs->dp_set_stream_attribute( - pipe_ctx->stream_res.stream_enc, - &stream->timing, - stream->output_color_space, - stream->use_vsc_sdp_for_colorimetry, - stream->link->dpcd_caps.dprx_feature.bits.SST_SPLIT_SDP_CAP); -#endif if (dc_is_dp_signal(pipe_ctx->stream->signal)) dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_DP_STREAM_ATTR); @@ -4206,10 +4131,8 @@ void core_link_enable_stream( pipe_ctx->stream->link->link_state_valid = true; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (pipe_ctx->stream_res.tg->funcs->set_out_mux) pipe_ctx->stream_res.tg->funcs->set_out_mux(pipe_ctx->stream_res.tg, otg_out_dest); -#endif if (dc_is_dvi_signal(pipe_ctx->stream->signal)) pipe_ctx->stream_res.stream_enc->funcs->dvi_set_stream_attribute( @@ -4229,11 +4152,9 @@ void core_link_enable_stream( pipe_ctx->stream->apply_edp_fast_boot_optimization = false; -#if defined(CONFIG_DRM_AMD_DC_DCN) // Enable VPG before building infoframe if (vpg && vpg->funcs->vpg_poweron) vpg->funcs->vpg_poweron(vpg); -#endif resource_build_info_frame(pipe_ctx); dc->hwss.update_info_frame(pipe_ctx); @@ -4319,12 +4240,8 @@ void core_link_enable_stream( * as a workaround for the incorrect value being applied * from transmitter control. */ -#if defined(CONFIG_DRM_AMD_DC_DCN) if (!(dc_is_virtual_signal(pipe_ctx->stream->signal) || is_dp_128b_132b_signal(pipe_ctx))) -#else - if (!dc_is_virtual_signal(pipe_ctx->stream->signal)) -#endif if (link_enc) link_enc->funcs->setup( link_enc, @@ -4343,11 +4260,9 @@ void core_link_enable_stream( if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) dc_link_allocate_mst_payload(pipe_ctx); -#if defined(CONFIG_DRM_AMD_DC_DCN) else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT && is_dp_128b_132b_signal(pipe_ctx)) dc_link_update_sst_payload(pipe_ctx, true); -#endif dc->hwss.unblank_stream(pipe_ctx, &pipe_ctx->stream->link->cur_link_settings); @@ -4364,11 +4279,9 @@ void core_link_enable_stream( dc->hwss.enable_audio_stream(pipe_ctx); } else { // if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) { fpga_dp_hpo_enable_link_and_stream(state, pipe_ctx); } -#endif if (dc_is_dp_signal(pipe_ctx->stream->signal) || dc_is_virtual_signal(pipe_ctx->stream->signal)) dp_set_dsc_enable(pipe_ctx, true); @@ -4385,12 +4298,10 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx) struct dc *dc = pipe_ctx->stream->ctx->dc; struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->sink->link; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct vpg *vpg = pipe_ctx->stream_res.stream_enc->vpg; if (is_dp_128b_132b_signal(pipe_ctx)) vpg = pipe_ctx->stream_res.hpo_dp_stream_enc->vpg; -#endif if (!IS_DIAG_DC(dc->ctx->dce_environment) && dc_is_virtual_signal(pipe_ctx->stream->signal)) @@ -4410,11 +4321,9 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx) if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) deallocate_mst_payload(pipe_ctx); -#if defined(CONFIG_DRM_AMD_DC_DCN) else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT && is_dp_128b_132b_signal(pipe_ctx)) dc_link_update_sst_payload(pipe_ctx, false); -#endif if (dc_is_hdmi_signal(pipe_ctx->stream->signal)) { struct ext_hdmi_settings settings = {0}; @@ -4441,7 +4350,6 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx) } } -#if defined(CONFIG_DRM_AMD_DC_DCN) if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT && !is_dp_128b_132b_signal(pipe_ctx)) { @@ -4458,27 +4366,18 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx) dc->hwss.disable_stream(pipe_ctx); disable_link(pipe_ctx->stream->link, &pipe_ctx->link_res, pipe_ctx->stream->signal); } -#else - disable_link(pipe_ctx->stream->link, &pipe_ctx->link_res, pipe_ctx->stream->signal); - - dc->hwss.disable_stream(pipe_ctx); -#endif if (pipe_ctx->stream->timing.flags.DSC) { if (dc_is_dp_signal(pipe_ctx->stream->signal)) dp_set_dsc_enable(pipe_ctx, false); } -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) { if (pipe_ctx->stream_res.tg->funcs->set_out_mux) pipe_ctx->stream_res.tg->funcs->set_out_mux(pipe_ctx->stream_res.tg, OUT_MUX_DIO); } -#endif -#if defined(CONFIG_DRM_AMD_DC_DCN) if (vpg && vpg->funcs->vpg_powerdown) vpg->funcs->vpg_powerdown(vpg); -#endif } void core_link_set_avmute(struct pipe_ctx *pipe_ctx, bool enable) @@ -4542,22 +4441,17 @@ void dc_link_set_drive_settings(struct dc *dc, { int i; - struct pipe_ctx *pipe = NULL; - const struct link_resource *link_res; + struct link_resource link_res; - link_res = dc_link_get_cur_link_res(link); + for (i = 0; i < dc->link_count; i++) + if (dc->links[i] == link) + break; - for (i = 0; i < MAX_PIPES; i++) { - pipe = &dc->current_state->res_ctx.pipe_ctx[i]; - if (pipe->stream && pipe->stream->link) { - if (pipe->stream->link == link) - break; - } - } - if (pipe && link_res) - dc_link_dp_set_drive_settings(pipe->stream->link, link_res, lt_settings); - else + if (i >= dc->link_count) ASSERT_CRITICAL(false); + + dc_link_get_cur_link_res(link, &link_res); + dc_link_dp_set_drive_settings(dc->links[i], &link_res, lt_settings); } void dc_link_set_preferred_link_settings(struct dc *dc, @@ -4617,11 +4511,9 @@ void dc_link_set_preferred_training_settings(struct dc *dc, if (link_setting != NULL) { link->preferred_link_setting = *link_setting; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(link_setting) == DP_128b_132b_ENCODING) /* TODO: add dc update for acquiring link res */ skip_immediate_retrain = true; -#endif } else { link->preferred_link_setting.lane_count = LANE_COUNT_UNKNOWN; link->preferred_link_setting.link_rate = LINK_RATE_UNKNOWN; @@ -4663,7 +4555,6 @@ uint32_t dc_link_bandwidth_kbps( const struct dc_link *link, const struct dc_link_settings *link_setting) { -#if defined(CONFIG_DRM_AMD_DC_DCN) uint32_t total_data_bw_efficiency_x10000 = 0; uint32_t link_rate_per_lane_kbps = 0; @@ -4694,40 +4585,6 @@ uint32_t dc_link_bandwidth_kbps( /* overall effective link bandwidth = link rate per lane * lane count * total data bandwidth efficiency */ return link_rate_per_lane_kbps * link_setting->lane_count / 10000 * total_data_bw_efficiency_x10000; -#else - uint32_t link_bw_kbps = - link_setting->link_rate * LINK_RATE_REF_FREQ_IN_KHZ; /* bytes per sec */ - - link_bw_kbps *= 8; /* 8 bits per byte*/ - link_bw_kbps *= link_setting->lane_count; - - if (dc_link_should_enable_fec(link)) { - /* Account for FEC overhead. - * We have to do it based on caps, - * and not based on FEC being set ready, - * because FEC is set ready too late in - * the process to correctly be picked up - * by mode enumeration. - * - * There's enough zeros at the end of 'kbps' - * that make the below operation 100% precise - * for our purposes. - * 'long long' makes it work even for HDMI 2.1 - * max bandwidth (and much, much bigger bandwidths - * than that, actually). - * - * NOTE: Reducing link BW by 3% may not be precise - * because it may be a stream BT that increases by 3%, and so - * 1/1.03 = 0.970873 factor should have been used instead, - * but the difference is minimal and is in a safe direction, - * which all works well around potential ambiguity of DP 1.4a spec. - */ - long long fec_link_bw_kbps = link_bw_kbps * 970LL; - link_bw_kbps = (uint32_t)(div64_s64(fec_link_bw_kbps, 1000LL)); - } - return link_bw_kbps; - -#endif } const struct dc_link_settings *dc_link_get_link_cap( @@ -4752,16 +4609,7 @@ bool dc_link_is_fec_supported(const struct dc_link *link) */ struct link_encoder *link_enc = NULL; - /* Links supporting dynamically assigned link encoder will be assigned next - * available encoder if one not already assigned. - */ - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) { - link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - if (link_enc == NULL) - link_enc = link_enc_cfg_get_next_avail_link_enc(link->ctx->dc); - } else - link_enc = link->link_enc; + link_enc = link_enc_cfg_get_link_enc(link); ASSERT(link_enc); return (dc_is_dp_signal(link->connector_signal) && link_enc && @@ -4845,23 +4693,24 @@ uint32_t dc_bandwidth_in_kbps_from_timing( } -const struct link_resource *dc_link_get_cur_link_res(const struct dc_link *link) +void dc_link_get_cur_link_res(const struct dc_link *link, + struct link_resource *link_res) { int i; struct pipe_ctx *pipe = NULL; - const struct link_resource *link_res = NULL; + + memset(link_res, 0, sizeof(*link_res)); for (i = 0; i < MAX_PIPES; i++) { pipe = &link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe->stream && pipe->stream->link && pipe->top_pipe == NULL) { if (pipe->stream->link == link) { - link_res = &pipe->link_res; + *link_res = pipe->link_res; break; } } } - return link_res; } /** @@ -4883,9 +4732,8 @@ const struct link_resource *dc_link_get_cur_link_res(const struct dc_link *link) */ void dc_get_cur_link_res_map(const struct dc *dc, uint32_t *map) { -#if defined(CONFIG_DRM_AMD_DC_DCN) struct dc_link *link; - uint8_t i; + uint32_t i; uint32_t hpo_dp_recycle_map = 0; *map = 0; @@ -4903,7 +4751,6 @@ void dc_get_cur_link_res_map(const struct dc *dc, uint32_t *map) } *map |= (hpo_dp_recycle_map << LINK_RES_HPO_DP_REC_MAP__SHIFT); } -#endif } /** @@ -4926,9 +4773,8 @@ void dc_get_cur_link_res_map(const struct dc *dc, uint32_t *map) */ void dc_restore_link_res_map(const struct dc *dc, uint32_t *map) { -#if defined(CONFIG_DRM_AMD_DC_DCN) struct dc_link *link; - uint8_t i; + uint32_t i; unsigned int available_hpo_dp_count; uint32_t hpo_dp_recycle_map = (*map & LINK_RES_HPO_DP_REC_MAP__MASK) >> LINK_RES_HPO_DP_REC_MAP__SHIFT; @@ -4964,5 +4810,4 @@ void dc_restore_link_res_map(const struct dc *dc, uint32_t *map) } } } -#endif } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index 61b8f29a0c30..cd9c31b5e55d 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -27,6 +27,7 @@ #include "dm_helpers.h" #include "opp.h" #include "dsc.h" +#include "clk_mgr.h" #include "resource.h" #include "inc/core_types.h" @@ -62,7 +63,6 @@ enum { POST_LT_ADJ_REQ_TIMEOUT = 200 }; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct dp_lt_fallback_entry { enum dc_lane_count lane_count; enum dc_link_rate link_rate; @@ -97,16 +97,18 @@ static const struct dp_lt_fallback_entry dp_lt_fallbacks[] = { {LANE_COUNT_ONE, LINK_RATE_HIGH}, {LANE_COUNT_ONE, LINK_RATE_LOW}, }; -#endif + +static const struct dc_link_settings fail_safe_link_settings = { + .lane_count = LANE_COUNT_ONE, + .link_rate = LINK_RATE_LOW, + .link_spread = LINK_SPREAD_DISABLED, +}; static bool decide_fallback_link_setting( struct dc_link *link, struct dc_link_settings initial_link_settings, struct dc_link_settings *current_link_setting, enum link_training_result training_result); -static struct dc_link_settings get_common_supported_link_settings( - struct dc_link_settings link_setting_a, - struct dc_link_settings link_setting_b); static void maximize_lane_settings(const struct link_training_settings *lt_settings, struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]); static void override_lane_settings(const struct link_training_settings *lt_settings, @@ -117,7 +119,7 @@ static uint32_t get_cr_training_aux_rd_interval(struct dc_link *link, { union training_aux_rd_interval training_rd_interval; uint32_t wait_in_micro_secs = 100; -#if defined(CONFIG_DRM_AMD_DC_DCN) + memset(&training_rd_interval, 0, sizeof(training_rd_interval)); if (dp_get_link_encoding_format(link_settings) == DP_8b_10b_ENCODING && link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_12) { @@ -129,15 +131,7 @@ static uint32_t get_cr_training_aux_rd_interval(struct dc_link *link, if (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL) wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000; } -#else - core_link_read_dpcd( - link, - DP_TRAINING_AUX_RD_INTERVAL, - (uint8_t *)&training_rd_interval, - sizeof(training_rd_interval)); - if (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL) - wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000; -#endif + return wait_in_micro_secs; } @@ -145,7 +139,6 @@ static uint32_t get_eq_training_aux_rd_interval( struct dc_link *link, const struct dc_link_settings *link_settings) { -#if defined(CONFIG_DRM_AMD_DC_DCN) union training_aux_rd_interval training_rd_interval; memset(&training_rd_interval, 0, sizeof(training_rd_interval)); @@ -174,41 +167,16 @@ static uint32_t get_eq_training_aux_rd_interval( case 6: return 64000; default: return 400; } -#else - union training_aux_rd_interval training_rd_interval; - uint32_t wait_in_micro_secs = 400; - - memset(&training_rd_interval, 0, sizeof(training_rd_interval)); - /* overwrite the delay if rev > 1.1*/ - if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_12) { - /* DP 1.2 or later - retrieve delay through - * "DPCD_ADDR_TRAINING_AUX_RD_INTERVAL" register */ - core_link_read_dpcd( - link, - DP_TRAINING_AUX_RD_INTERVAL, - (uint8_t *)&training_rd_interval, - sizeof(training_rd_interval)); - - if (training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL) - wait_in_micro_secs = training_rd_interval.bits.TRAINIG_AUX_RD_INTERVAL * 4000; - } - - return wait_in_micro_secs; -#endif } void dp_wait_for_training_aux_rd_interval( struct dc_link *link, uint32_t wait_in_micro_secs) { -#if defined(CONFIG_DRM_AMD_DC_DCN) if (wait_in_micro_secs > 1000) msleep(wait_in_micro_secs/1000); else udelay(wait_in_micro_secs); -#else - udelay(wait_in_micro_secs); -#endif DC_LOG_HW_LINK_TRAINING("%s:\n wait = %d\n", __func__, @@ -236,7 +204,6 @@ enum dpcd_training_patterns case DP_TRAINING_PATTERN_SEQUENCE_4: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_4; break; -#if defined(CONFIG_DRM_AMD_DC_DCN) case DP_128b_132b_TPS1: dpcd_tr_pattern = DPCD_128b_132b_TPS1; break; @@ -246,7 +213,6 @@ enum dpcd_training_patterns case DP_128b_132b_TPS2_CDS: dpcd_tr_pattern = DPCD_128b_132b_TPS2_CDS; break; -#endif case DP_TRAINING_PATTERN_VIDEOIDLE: dpcd_tr_pattern = DPCD_TRAINING_PATTERN_VIDEOIDLE; break; @@ -289,10 +255,8 @@ static enum dc_dp_training_pattern decide_cr_training_pattern( case DP_8b_10b_ENCODING: default: return DP_TRAINING_PATTERN_SEQUENCE_1; -#if defined(CONFIG_DRM_AMD_DC_DCN) case DP_128b_132b_ENCODING: return DP_128b_132b_TPS1; -#endif } } @@ -300,19 +264,11 @@ static enum dc_dp_training_pattern decide_eq_training_pattern(struct dc_link *li const struct dc_link_settings *link_settings) { struct link_encoder *link_enc; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct encoder_feature_support *enc_caps; struct dpcd_caps *rx_caps = &link->dpcd_caps; enum dc_dp_training_pattern pattern = DP_TRAINING_PATTERN_SEQUENCE_2; - /* Access link encoder capability based on whether it is statically - * or dynamically assigned to a link. - */ - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - else - link_enc = link->link_enc; + link_enc = link_enc_cfg_get_link_enc(link); ASSERT(link_enc); enc_caps = &link_enc->features; @@ -335,41 +291,8 @@ static enum dc_dp_training_pattern decide_eq_training_pattern(struct dc_link *li break; } return pattern; -#else - enum dc_dp_training_pattern highest_tp = DP_TRAINING_PATTERN_SEQUENCE_2; - struct encoder_feature_support *features; - struct dpcd_caps *dpcd_caps = &link->dpcd_caps; - - /* Access link encoder capability based on whether it is statically - * or dynamically assigned to a link. - */ - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - else - link_enc = link->link_enc; - ASSERT(link_enc); - features = &link_enc->features; - - if (features->flags.bits.IS_TPS3_CAPABLE) - highest_tp = DP_TRAINING_PATTERN_SEQUENCE_3; - - if (features->flags.bits.IS_TPS4_CAPABLE) - highest_tp = DP_TRAINING_PATTERN_SEQUENCE_4; - - if (dpcd_caps->max_down_spread.bits.TPS4_SUPPORTED && - highest_tp >= DP_TRAINING_PATTERN_SEQUENCE_4) - return DP_TRAINING_PATTERN_SEQUENCE_4; - - if (dpcd_caps->max_ln_count.bits.TPS3_SUPPORTED && - highest_tp >= DP_TRAINING_PATTERN_SEQUENCE_3) - return DP_TRAINING_PATTERN_SEQUENCE_3; - - return DP_TRAINING_PATTERN_SEQUENCE_2; -#endif } -#if defined(CONFIG_DRM_AMD_DC_DCN) static uint8_t get_dpcd_link_rate(const struct dc_link_settings *link_settings) { uint8_t link_rate = 0; @@ -397,7 +320,6 @@ static uint8_t get_dpcd_link_rate(const struct dc_link_settings *link_settings) return link_rate; } -#endif static void vendor_specific_lttpr_wa_one_start(struct dc_link *link) { @@ -540,10 +462,8 @@ static void vendor_specific_lttpr_wa_four( const uint8_t offset = dp_convert_to_count( link->dpcd_caps.lttpr_caps.phy_repeater_cnt); uint32_t vendor_lttpr_write_address = 0xF004F; -#if defined(CONFIG_DRM_AMD_DC_DP2_0) uint8_t sink_status = 0; uint8_t i; -#endif if (offset != 0xFF) vendor_lttpr_write_address += @@ -569,7 +489,6 @@ static void vendor_specific_lttpr_wa_four( sizeof(vendor_lttpr_write_data_two)); } -#if defined(CONFIG_DRM_AMD_DC_DP2_0) /* poll for intra-hop disable */ for (i = 0; i < 10; i++) { if ((core_link_read_dpcd(link, DP_SINK_STATUS, &sink_status, 1) == DC_OK) && @@ -577,7 +496,6 @@ static void vendor_specific_lttpr_wa_four( break; udelay(1000); } -#endif } static void vendor_specific_lttpr_wa_five( @@ -665,11 +583,7 @@ enum dc_status dpcd_set_link_settings( status = core_link_write_dpcd(link, DP_LINK_RATE_SET, <_settings->link_settings.link_rate_set, 1); } else { -#if defined(CONFIG_DRM_AMD_DC_DCN) rate = get_dpcd_link_rate(<_settings->link_settings); -#else - rate = (uint8_t) (lt_settings->link_settings.link_rate); -#endif if (link->dc->debug.apply_vendor_specific_lttpr_wa && (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && link->lttpr_mode == LTTPR_MODE_TRANSPARENT) @@ -720,10 +634,8 @@ uint8_t dc_dp_initialize_scrambling_data_symbols( disable_scrabled_data_symbols = 1; break; case DP_TRAINING_PATTERN_SEQUENCE_4: -#if defined(CONFIG_DRM_AMD_DC_DCN) case DP_128b_132b_TPS1: case DP_128b_132b_TPS2: -#endif disable_scrabled_data_symbols = 0; break; default: @@ -794,7 +706,6 @@ static void dpcd_set_lt_pattern_and_lane_settings( size_in_bytes); if (is_repeater(link, offset)) { -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(<_settings->link_settings) == DP_128b_132b_ENCODING) DC_LOG_HW_LINK_TRAINING("%s:\n LTTPR Repeater ID: %d\n" @@ -805,7 +716,6 @@ static void dpcd_set_lt_pattern_and_lane_settings( lt_settings->dpcd_lane_settings[0].tx_ffe.PRESET_VALUE); else if (dp_get_link_encoding_format(<_settings->link_settings) == DP_8b_10b_ENCODING) -#endif DC_LOG_HW_LINK_TRAINING("%s:\n LTTPR Repeater ID: %d\n" " 0x%X VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", __func__, @@ -816,7 +726,6 @@ static void dpcd_set_lt_pattern_and_lane_settings( lt_settings->dpcd_lane_settings[0].bits.MAX_SWING_REACHED, lt_settings->dpcd_lane_settings[0].bits.MAX_PRE_EMPHASIS_REACHED); } else { -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(<_settings->link_settings) == DP_128b_132b_ENCODING) DC_LOG_HW_LINK_TRAINING("%s:\n 0x%X TX_FFE_PRESET_VALUE = %x\n", @@ -825,14 +734,13 @@ static void dpcd_set_lt_pattern_and_lane_settings( lt_settings->dpcd_lane_settings[0].tx_ffe.PRESET_VALUE); else if (dp_get_link_encoding_format(<_settings->link_settings) == DP_8b_10b_ENCODING) -#endif - DC_LOG_HW_LINK_TRAINING("%s:\n 0x%X VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", - __func__, - dpcd_base_lt_offset, - lt_settings->dpcd_lane_settings[0].bits.VOLTAGE_SWING_SET, - lt_settings->dpcd_lane_settings[0].bits.PRE_EMPHASIS_SET, - lt_settings->dpcd_lane_settings[0].bits.MAX_SWING_REACHED, - lt_settings->dpcd_lane_settings[0].bits.MAX_PRE_EMPHASIS_REACHED); + DC_LOG_HW_LINK_TRAINING("%s:\n 0x%X VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", + __func__, + dpcd_base_lt_offset, + lt_settings->dpcd_lane_settings[0].bits.VOLTAGE_SWING_SET, + lt_settings->dpcd_lane_settings[0].bits.PRE_EMPHASIS_SET, + lt_settings->dpcd_lane_settings[0].bits.MAX_SWING_REACHED, + lt_settings->dpcd_lane_settings[0].bits.MAX_PRE_EMPHASIS_REACHED); } if (edp_workaround) { /* for eDP write in 2 parts because the 5-byte burst is @@ -850,7 +758,6 @@ static void dpcd_set_lt_pattern_and_lane_settings( (uint8_t *)(lt_settings->dpcd_lane_settings), size_in_bytes); -#if defined(CONFIG_DRM_AMD_DC_DCN) } else if (dp_get_link_encoding_format(<_settings->link_settings) == DP_128b_132b_ENCODING) { core_link_write_dpcd( @@ -858,10 +765,9 @@ static void dpcd_set_lt_pattern_and_lane_settings( dpcd_base_lt_offset, dpcd_lt_buffer, sizeof(dpcd_lt_buffer)); -#endif - } else + } else /* write it all in (1 + number-of-lanes)-byte burst*/ - core_link_write_dpcd( + core_link_write_dpcd( link, dpcd_base_lt_offset, dpcd_lt_buffer, @@ -928,13 +834,11 @@ void dp_hw_to_dpcd_lane_settings( (hw_lane_settings[lane].PRE_EMPHASIS == PRE_EMPHASIS_MAX_LEVEL ? 1 : 0); } -#if defined(CONFIG_DRM_AMD_DC_DCN) else if (dp_get_link_encoding_format(<_settings->link_settings) == DP_128b_132b_ENCODING) { dpcd_lane_settings[lane].tx_ffe.PRESET_VALUE = hw_lane_settings[lane].FFE_PRESET.settings.level; } -#endif } } @@ -956,13 +860,11 @@ void dp_decide_lane_settings( (enum dc_pre_emphasis)(ln_adjust[lane].bits. PRE_EMPHASIS_LANE); } -#if defined(CONFIG_DRM_AMD_DC_DCN) else if (dp_get_link_encoding_format(<_settings->link_settings) == DP_128b_132b_ENCODING) { hw_lane_settings[lane].FFE_PRESET.raw = ln_adjust[lane].tx_ffe.PRESET_VALUE; } -#endif } dp_hw_to_dpcd_lane_settings(lt_settings, hw_lane_settings, dpcd_lane_settings); @@ -1013,9 +915,7 @@ static void maximize_lane_settings(const struct link_training_settings *lt_setti max_requested.VOLTAGE_SWING = lane_settings[0].VOLTAGE_SWING; max_requested.PRE_EMPHASIS = lane_settings[0].PRE_EMPHASIS; -#if defined(CONFIG_DRM_AMD_DC_DCN) max_requested.FFE_PRESET = lane_settings[0].FFE_PRESET; -#endif /* Determine what the maximum of the requested settings are*/ for (lane = 1; lane < lt_settings->link_settings.lane_count; lane++) { @@ -1024,12 +924,10 @@ static void maximize_lane_settings(const struct link_training_settings *lt_setti if (lane_settings[lane].PRE_EMPHASIS > max_requested.PRE_EMPHASIS) max_requested.PRE_EMPHASIS = lane_settings[lane].PRE_EMPHASIS; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (lane_settings[lane].FFE_PRESET.settings.level > max_requested.FFE_PRESET.settings.level) max_requested.FFE_PRESET.settings.level = lane_settings[lane].FFE_PRESET.settings.level; -#endif } /* make sure the requested settings are @@ -1039,10 +937,8 @@ static void maximize_lane_settings(const struct link_training_settings *lt_setti if (max_requested.PRE_EMPHASIS > PRE_EMPHASIS_MAX_LEVEL) max_requested.PRE_EMPHASIS = PRE_EMPHASIS_MAX_LEVEL; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (max_requested.FFE_PRESET.settings.level > DP_FFE_PRESET_MAX_LEVEL) max_requested.FFE_PRESET.settings.level = DP_FFE_PRESET_MAX_LEVEL; -#endif /* make sure the pre-emphasis matches the voltage swing*/ if (max_requested.PRE_EMPHASIS > @@ -1055,9 +951,7 @@ static void maximize_lane_settings(const struct link_training_settings *lt_setti for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { lane_settings[lane].VOLTAGE_SWING = max_requested.VOLTAGE_SWING; lane_settings[lane].PRE_EMPHASIS = max_requested.PRE_EMPHASIS; -#if defined(CONFIG_DRM_AMD_DC_DCN) lane_settings[lane].FFE_PRESET = max_requested.FFE_PRESET; -#endif } } @@ -1068,9 +962,7 @@ static void override_lane_settings(const struct link_training_settings *lt_setti if (lt_settings->voltage_swing == NULL && lt_settings->pre_emphasis == NULL && -#if defined(CONFIG_DRM_AMD_DC_DCN) lt_settings->ffe_preset == NULL && -#endif lt_settings->post_cursor2 == NULL) return; @@ -1082,10 +974,8 @@ static void override_lane_settings(const struct link_training_settings *lt_setti lane_settings[lane].PRE_EMPHASIS = *lt_settings->pre_emphasis; if (lt_settings->post_cursor2) lane_settings[lane].POST_CURSOR2 = *lt_settings->post_cursor2; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (lt_settings->ffe_preset) lane_settings[lane].FFE_PRESET = *lt_settings->ffe_preset; -#endif } } @@ -1189,7 +1079,6 @@ enum dc_status dpcd_set_lane_settings( link_training_setting->link_settings.lane_count); if (is_repeater(link, offset)) { -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(&link_training_setting->link_settings) == DP_128b_132b_ENCODING) DC_LOG_HW_LINK_TRAINING("%s:\n LTTPR Repeater ID: %d\n" @@ -1200,7 +1089,6 @@ enum dc_status dpcd_set_lane_settings( link_training_setting->dpcd_lane_settings[0].tx_ffe.PRESET_VALUE); else if (dp_get_link_encoding_format(&link_training_setting->link_settings) == DP_8b_10b_ENCODING) -#endif DC_LOG_HW_LINK_TRAINING("%s\n LTTPR Repeater ID: %d\n" " 0x%X VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", __func__, @@ -1212,7 +1100,6 @@ enum dc_status dpcd_set_lane_settings( link_training_setting->dpcd_lane_settings[0].bits.MAX_PRE_EMPHASIS_REACHED); } else { -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(&link_training_setting->link_settings) == DP_128b_132b_ENCODING) DC_LOG_HW_LINK_TRAINING("%s:\n 0x%X TX_FFE_PRESET_VALUE = %x\n", @@ -1221,7 +1108,6 @@ enum dc_status dpcd_set_lane_settings( link_training_setting->dpcd_lane_settings[0].tx_ffe.PRESET_VALUE); else if (dp_get_link_encoding_format(&link_training_setting->link_settings) == DP_8b_10b_ENCODING) -#endif DC_LOG_HW_LINK_TRAINING("%s\n 0x%X VS set = %x PE set = %x max VS Reached = %x max PE Reached = %x\n", __func__, lane0_set_address, @@ -1357,14 +1243,12 @@ uint32_t dp_translate_training_aux_read_interval(uint32_t dpcd_aux_read_interval case 0x04: aux_rd_interval_us = 16000; break; -#if defined(CONFIG_DRM_AMD_DC_DCN) case 0x05: aux_rd_interval_us = 32000; break; case 0x06: aux_rd_interval_us = 64000; break; -#endif default: break; } @@ -1405,13 +1289,8 @@ static enum link_training_result perform_channel_equalization_sequence( /* Note: also check that TPS4 is a supported feature*/ tr_pattern = lt_settings->pattern_for_eq; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_repeater(link, offset) && dp_get_link_encoding_format(<_settings->link_settings) == DP_8b_10b_ENCODING) tr_pattern = DP_TRAINING_PATTERN_SEQUENCE_4; -#else - if (is_repeater(link, offset)) - tr_pattern = DP_TRAINING_PATTERN_SEQUENCE_4; -#endif dp_set_hw_training_pattern(link, link_res, tr_pattern, offset); @@ -1582,15 +1461,10 @@ static enum link_training_result perform_clock_recovery_sequence( return LINK_TRAINING_SUCCESS; /* 6. max VS reached*/ -#if defined(CONFIG_DRM_AMD_DC_DCN) if ((dp_get_link_encoding_format(<_settings->link_settings) == DP_8b_10b_ENCODING) && dp_is_max_vs_reached(lt_settings)) break; -#else - if (dp_is_max_vs_reached(lt_settings)) - break; -#endif /* 7. same lane settings*/ /* Note: settings are the same for all lanes, @@ -1599,12 +1473,10 @@ static enum link_training_result perform_clock_recovery_sequence( lt_settings->dpcd_lane_settings[0].bits.VOLTAGE_SWING_SET == dpcd_lane_adjust[0].bits.VOLTAGE_SWING_LANE) retries_cr++; -#if defined(CONFIG_DRM_AMD_DC_DCN) else if ((dp_get_link_encoding_format(<_settings->link_settings) == DP_128b_132b_ENCODING) && lt_settings->dpcd_lane_settings[0].tx_ffe.PRESET_VALUE == dpcd_lane_adjust[0].tx_ffe.PRESET_VALUE) retries_cr++; -#endif else retries_cr = 0; @@ -1642,11 +1514,7 @@ static inline enum link_training_result dp_transition_to_video_idle( * TPS4 must be used instead of POST_LT_ADJ_REQ. */ if (link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED != 1 || -#if defined(CONFIG_DRM_AMD_DC_DCN) lt_settings->pattern_for_eq >= DP_TRAINING_PATTERN_SEQUENCE_4) { -#else - lt_settings->pattern_for_eq == DP_TRAINING_PATTERN_SEQUENCE_4) { -#endif /* delay 5ms after Main Link output idle pattern and then check * DPCD 0202h. */ @@ -1745,7 +1613,6 @@ static inline void decide_8b_10b_training_settings( dp_hw_to_dpcd_lane_settings(lt_settings, lt_settings->hw_lane_settings, lt_settings->dpcd_lane_settings); } -#if defined(CONFIG_DRM_AMD_DC_DCN) static inline void decide_128b_132b_training_settings(struct dc_link *link, const struct dc_link_settings *link_settings, struct link_training_settings *lt_settings) @@ -1772,7 +1639,6 @@ static inline void decide_128b_132b_training_settings(struct dc_link *link, dp_hw_to_dpcd_lane_settings(lt_settings, lt_settings->hw_lane_settings, lt_settings->dpcd_lane_settings); } -#endif void dp_decide_training_settings( struct dc_link *link, @@ -1781,10 +1647,8 @@ void dp_decide_training_settings( { if (dp_get_link_encoding_format(link_settings) == DP_8b_10b_ENCODING) decide_8b_10b_training_settings(link, link_settings, lt_settings); -#if defined(CONFIG_DRM_AMD_DC_DCN) else if (dp_get_link_encoding_format(link_settings) == DP_128b_132b_ENCODING) decide_128b_132b_training_settings(link, link_settings, lt_settings); -#endif } static void override_training_settings( @@ -1807,10 +1671,8 @@ static void override_training_settings( lt_settings->pre_emphasis = overrides->pre_emphasis; if (overrides->post_cursor2 != NULL) lt_settings->post_cursor2 = overrides->post_cursor2; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (overrides->ffe_preset != NULL) lt_settings->ffe_preset = overrides->ffe_preset; -#endif /* Override HW lane settings with BIOS forced values if present */ if (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN && link->lttpr_mode == LTTPR_MODE_TRANSPARENT) { @@ -2014,7 +1876,6 @@ static void print_status_message( case LINK_RATE_HIGH3: link_rate = "HBR3"; break; -#if defined(CONFIG_DRM_AMD_DC_DCN) case LINK_RATE_UHBR10: link_rate = "UHBR10"; break; @@ -2024,7 +1885,6 @@ static void print_status_message( case LINK_RATE_UHBR20: link_rate = "UHBR20"; break; -#endif default: break; } @@ -2054,7 +1914,6 @@ static void print_status_message( case LINK_TRAINING_LINK_LOSS: lt_result = "Link loss"; break; -#if defined(CONFIG_DRM_AMD_DC_DCN) case DP_128b_132b_LT_FAILED: lt_result = "LT_FAILED received"; break; @@ -2067,7 +1926,6 @@ static void print_status_message( case DP_128b_132b_CDS_DONE_TIMEOUT: lt_result = "CDS timeout"; break; -#endif default: break; } @@ -2087,9 +1945,9 @@ static void print_status_message( } /* Connectivity log: link training */ -#if defined(CONFIG_DRM_AMD_DC_DCN) + /* TODO - DP2.0 Log: add connectivity log for FFE PRESET */ -#endif + CONN_MSG_LT(link, "%sx%d %s VS=%d, PE=%d, DS=%s", link_rate, lt_settings->link_settings.lane_count, @@ -2177,15 +2035,12 @@ enum dc_status dpcd_configure_lttpr_mode(struct dc_link *link, struct link_train static void dpcd_exit_training_mode(struct dc_link *link) { -#if defined(CONFIG_DRM_AMD_DC_DCN) uint8_t sink_status = 0; uint8_t i; -#endif /* clear training pattern set */ dpcd_set_training_pattern(link, DP_TRAINING_PATTERN_VIDEOIDLE); -#if defined(CONFIG_DRM_AMD_DC_DCN) /* poll for intra-hop disable */ for (i = 0; i < 10; i++) { if ((core_link_read_dpcd(link, DP_SINK_STATUS, &sink_status, 1) == DC_OK) && @@ -2193,7 +2048,6 @@ static void dpcd_exit_training_mode(struct dc_link *link) break; udelay(1000); } -#endif } enum dc_status dpcd_configure_channel_coding(struct dc_link *link, @@ -2217,7 +2071,6 @@ enum dc_status dpcd_configure_channel_coding(struct dc_link *link, return status; } -#if defined(CONFIG_DRM_AMD_DC_DCN) static void dpcd_128b_132b_get_aux_rd_interval(struct dc_link *link, uint32_t *interval_in_us) { @@ -2348,7 +2201,6 @@ static enum link_training_result dp_perform_128b_132b_cds_done_sequence( return status; } -#endif static enum link_training_result dp_perform_8b_10b_link_training( struct dc_link *link, @@ -2398,18 +2250,17 @@ static enum link_training_result dp_perform_8b_10b_link_training( if (status == LINK_TRAINING_SUCCESS) { status = perform_clock_recovery_sequence(link, link_res, lt_settings, DPRX); - if (status == LINK_TRAINING_SUCCESS) { - status = perform_channel_equalization_sequence(link, - link_res, - lt_settings, - DPRX); + if (status == LINK_TRAINING_SUCCESS) { + status = perform_channel_equalization_sequence(link, + link_res, + lt_settings, + DPRX); } } return status; } -#if defined(CONFIG_DRM_AMD_DC_DCN) static enum link_training_result dp_perform_128b_132b_link_training( struct dc_link *link, const struct link_resource *link_res, @@ -2437,7 +2288,6 @@ static enum link_training_result dp_perform_128b_132b_link_training( return result; } -#endif static enum link_training_result dc_link_dp_perform_fixed_vs_pe_training_sequence( struct dc_link *link, @@ -2514,11 +2364,7 @@ static enum link_training_result dc_link_dp_perform_fixed_vs_pe_training_sequenc core_link_write_dpcd(link, DP_LANE_COUNT_SET, &lane_count_set.raw, 1); -#if defined(CONFIG_DRM_AMD_DC_DCN) rate = get_dpcd_link_rate(<_settings->link_settings); -#else - rate = (uint8_t) (lt_settings->link_settings.link_rate); -#endif /* Vendor specific: Toggle link rate */ toggle_rate = (rate == 0x6) ? 0xA : 0x6; @@ -2819,10 +2665,8 @@ enum link_training_result dc_link_dp_perform_link_training( status = dc_link_dp_perform_fixed_vs_pe_training_sequence(link, link_res, <_settings); else if (encoding == DP_8b_10b_ENCODING) status = dp_perform_8b_10b_link_training(link, link_res, <_settings); -#if defined(CONFIG_DRM_AMD_DC_DCN) else if (encoding == DP_128b_132b_ENCODING) status = dp_perform_128b_132b_link_training(link, link_res, <_settings); -#endif else ASSERT(0); @@ -2863,26 +2707,15 @@ bool perform_link_training_with_retries( struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; enum dp_panel_mode panel_mode = dp_get_panel_mode(link); - struct link_encoder *link_enc; enum link_training_result status = LINK_TRAINING_CR_FAIL_LANE0; struct dc_link_settings current_setting = *link_setting; + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); - /* Dynamically assigned link encoders associated with stream rather than - * link. - */ - if (link->is_dig_mapping_flexible && link->dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_stream(link->ctx->dc, pipe_ctx->stream); - else - link_enc = link->link_enc; - - /* We need to do this before the link training to ensure the idle pattern in SST - * mode will be sent right after the link training - */ - if (dp_get_link_encoding_format(¤t_setting) == DP_8b_10b_ENCODING) { - link_enc->funcs->connect_dig_be_to_fe(link_enc, - pipe_ctx->stream_res.stream_enc->id, true); - dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_CONNECT_DIG_FE_BE); - } + if (dp_get_link_encoding_format(¤t_setting) == DP_8b_10b_ENCODING) + /* We need to do this before the link training to ensure the idle + * pattern in SST mode will be sent right after the link training + */ + link_hwss->setup_stream_encoder(pipe_ctx); for (j = 0; j < attempts; ++j) { @@ -3074,14 +2907,10 @@ enum link_training_result dc_link_dp_sync_lt_attempt( dp_cs_id, link_settings); /* Set FEC enable */ -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(link_settings) == DP_8b_10b_ENCODING) { -#endif fec_enable = lt_overrides->fec_enable && *lt_overrides->fec_enable; dp_set_fec_ready(link, NULL, fec_enable); -#if defined(CONFIG_DRM_AMD_DC_DCN) } -#endif if (lt_overrides->alternate_scrambler_reset) { if (*lt_overrides->alternate_scrambler_reset) @@ -3124,13 +2953,9 @@ bool dc_link_dp_sync_lt_end(struct dc_link *link, bool link_down) * Still shouldn't turn off dp_receiver (DPCD:600h) */ if (link_down == true) { -#if defined(CONFIG_DRM_AMD_DC_DCN) struct dc_link_settings link_settings = link->cur_link_settings; -#endif dp_disable_link_phy(link, NULL, link->connector_signal); -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(&link_settings) == DP_8b_10b_ENCODING) -#endif dp_set_fec_ready(link, NULL, false); } @@ -3138,7 +2963,6 @@ bool dc_link_dp_sync_lt_end(struct dc_link *link, bool link_down) return true; } -#if defined(CONFIG_DRM_AMD_DC_DCN) static enum dc_link_rate get_lttpr_max_link_rate(struct dc_link *link) { enum dc_link_rate lttpr_max_link_rate = link->dpcd_caps.lttpr_caps.max_link_rate; @@ -3152,7 +2976,20 @@ static enum dc_link_rate get_lttpr_max_link_rate(struct dc_link *link) return lttpr_max_link_rate; } -#endif + +static enum dc_link_rate get_cable_max_link_rate(struct dc_link *link) +{ + enum dc_link_rate cable_max_link_rate = LINK_RATE_HIGH3; + + if (link->dpcd_caps.cable_attributes.bits.UHBR10_20_CAPABILITY & DP_UHBR20) + cable_max_link_rate = LINK_RATE_UHBR20; + else if (link->dpcd_caps.cable_attributes.bits.UHBR13_5_CAPABILITY) + cable_max_link_rate = LINK_RATE_UHBR13_5; + else if (link->dpcd_caps.cable_attributes.bits.UHBR10_20_CAPABILITY & DP_UHBR10) + cable_max_link_rate = LINK_RATE_UHBR10; + + return cable_max_link_rate; +} bool dc_link_dp_get_max_link_enc_cap(const struct dc_link *link, struct dc_link_settings *max_link_enc_cap) { @@ -3163,16 +3000,7 @@ bool dc_link_dp_get_max_link_enc_cap(const struct dc_link *link, struct dc_link_ return false; } - /* Links supporting dynamically assigned link encoder will be assigned next - * available encoder if one not already assigned. - */ - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) { - link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - if (link_enc == NULL) - link_enc = link_enc_cfg_get_next_avail_link_enc(link->ctx->dc); - } else - link_enc = link->link_enc; + link_enc = link_enc_cfg_get_link_enc(link); ASSERT(link_enc); if (link_enc && link_enc->funcs->get_max_link_cap) { @@ -3186,37 +3014,21 @@ bool dc_link_dp_get_max_link_enc_cap(const struct dc_link *link, struct dc_link_ return false; } -static struct dc_link_settings get_max_link_cap(struct dc_link *link, - const struct link_resource *link_res) + +struct dc_link_settings dp_get_max_link_cap(struct dc_link *link) { struct dc_link_settings max_link_cap = {0}; -#if defined(CONFIG_DRM_AMD_DC_DCN) enum dc_link_rate lttpr_max_link_rate; -#endif + enum dc_link_rate cable_max_link_rate; struct link_encoder *link_enc = NULL; - /* Links supporting dynamically assigned link encoder will be assigned next - * available encoder if one not already assigned. - */ - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) { - link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - if (link_enc == NULL) - link_enc = link_enc_cfg_get_next_avail_link_enc(link->ctx->dc); - } else - link_enc = link->link_enc; + + link_enc = link_enc_cfg_get_link_enc(link); ASSERT(link_enc); /* get max link encoder capability */ if (link_enc) link_enc->funcs->get_max_link_cap(link_enc, &max_link_cap); -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (max_link_cap.link_rate >= LINK_RATE_UHBR10) { - if (!link_res->hpo_dp_link_enc || - link->dc->debug.disable_uhbr) - max_link_cap.link_rate = LINK_RATE_HIGH3; - } -#endif /* Lower link settings based on sink's link cap */ if (link->reported_link_cap.lane_count < max_link_cap.lane_count) @@ -3229,6 +3041,14 @@ static struct dc_link_settings get_max_link_cap(struct dc_link *link, max_link_cap.link_spread) max_link_cap.link_spread = link->reported_link_cap.link_spread; + + /* Lower link settings based on cable attributes */ + cable_max_link_rate = get_cable_max_link_rate(link); + + if (!link->dc->debug.ignore_cable_id && + cable_max_link_rate < max_link_cap.link_rate) + max_link_cap.link_rate = cable_max_link_rate; + /* * account for lttpr repeaters cap * notes: repeaters do not snoop in the DPRX Capabilities addresses (3.6.3). @@ -3236,22 +3056,21 @@ static struct dc_link_settings get_max_link_cap(struct dc_link *link, if (link->lttpr_mode != LTTPR_MODE_NON_LTTPR) { if (link->dpcd_caps.lttpr_caps.max_lane_count < max_link_cap.lane_count) max_link_cap.lane_count = link->dpcd_caps.lttpr_caps.max_lane_count; - -#if defined(CONFIG_DRM_AMD_DC_DCN) lttpr_max_link_rate = get_lttpr_max_link_rate(link); if (lttpr_max_link_rate < max_link_cap.link_rate) max_link_cap.link_rate = lttpr_max_link_rate; -#else - if (link->dpcd_caps.lttpr_caps.max_link_rate < max_link_cap.link_rate) - max_link_cap.link_rate = link->dpcd_caps.lttpr_caps.max_link_rate; -#endif DC_LOG_HW_LINK_TRAINING("%s\n Training with LTTPR, max_lane count %d max_link rate %d \n", __func__, max_link_cap.lane_count, max_link_cap.link_rate); } + + if (dp_get_link_encoding_format(&max_link_cap) == DP_128b_132b_ENCODING && + link->dc->debug.disable_uhbr) + max_link_cap.link_rate = LINK_RATE_HIGH3; + return max_link_cap; } @@ -3370,48 +3189,22 @@ bool hpd_rx_irq_check_link_loss_status( return return_code; } -bool dp_verify_link_cap( +static bool dp_verify_link_cap( struct dc_link *link, - const struct link_resource *link_res, struct dc_link_settings *known_limit_link_setting, int *fail_count) { - struct dc_link_settings max_link_cap = {0}; - struct dc_link_settings cur_link_setting = {0}; - struct dc_link_settings *cur = &cur_link_setting; - struct dc_link_settings initial_link_settings = {0}; - bool success; - bool skip_link_training; + struct dc_link_settings cur_link_settings = {0}; + struct dc_link_settings initial_link_settings = *known_limit_link_setting; + bool success = false; bool skip_video_pattern; - enum clock_source_id dp_cs_id = CLOCK_SOURCE_ID_EXTERNAL; - enum link_training_result status; + enum clock_source_id dp_cs_id = get_clock_source_id(link); + enum link_training_result status = LINK_TRAINING_SUCCESS; union hpd_irq_data irq_data; - - /* link training starts with the maximum common settings - * supported by both sink and ASIC. - */ - max_link_cap = get_max_link_cap(link, link_res); - initial_link_settings = get_common_supported_link_settings( - *known_limit_link_setting, - max_link_cap); - - /* Accept reported capabilities if link supports flexible encoder mapping or encoder already in use. */ - if (link->dc->debug.skip_detection_link_training || - link->is_dig_mapping_flexible) { - /* TODO - should we check link encoder's max link caps here? - * How do we know which link encoder to check from? - */ - link->verified_link_cap = *known_limit_link_setting; - return true; - } else if (link->link_enc && link->dc->res_pool->funcs->link_encs_assign && - !link_enc_cfg_is_link_enc_avail(link->ctx->dc, link->link_enc->preferred_engine, link)) { - link->verified_link_cap = initial_link_settings; - return true; - } + struct link_resource link_res; memset(&irq_data, 0, sizeof(irq_data)); - success = false; - skip_link_training = false; + cur_link_settings = initial_link_settings; /* Grant extended timeout request */ if ((link->lttpr_mode == LTTPR_MODE_NON_TRANSPARENT) && (link->dpcd_caps.lttpr_caps.max_ext_timeout > 0)) { @@ -3420,104 +3213,72 @@ bool dp_verify_link_cap( core_link_write_dpcd(link, DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT, &grant, sizeof(grant)); } -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dp_get_link_encoding_format(&link->cur_link_settings) == DP_128b_132b_ENCODING) - reset_dp_hpo_stream_encoders_for_link(link); -#endif - /* TODO implement override and monitor patch later */ - - /* try to train the link from high to low to - * find the physical link capability - */ - /* disable PHY done possible by BIOS, will be done by driver itself */ - dp_disable_link_phy(link, link_res, link->connector_signal); - - dp_cs_id = get_clock_source_id(link); - - cur_link_setting = initial_link_settings; - - /* Temporary Renoir-specific workaround for SWDEV-215184; - * PHY will sometimes be in bad state on hotplugging display from certain USB-C dongle, - * so add extra cycle of enabling and disabling the PHY before first link training. - */ - if (link->link_enc && link->link_enc->features.flags.bits.DP_IS_USB_C && - link->dc->debug.usbc_combo_phy_reset_wa) { - dp_enable_link_phy(link, link_res, link->connector_signal, dp_cs_id, cur); - dp_disable_link_phy(link, link_res, link->connector_signal); - } - do { - skip_video_pattern = true; - - if (cur->link_rate == LINK_RATE_LOW) - skip_video_pattern = false; + if (!get_temp_dp_link_res(link, &link_res, &cur_link_settings)) + continue; + skip_video_pattern = cur_link_settings.link_rate != LINK_RATE_LOW; dp_enable_link_phy( link, - link_res, + &link_res, link->connector_signal, dp_cs_id, - cur); + &cur_link_settings); + status = dc_link_dp_perform_link_training( + link, + &link_res, + &cur_link_settings, + skip_video_pattern); - if (skip_link_training) + if (status == LINK_TRAINING_SUCCESS) { success = true; - else { - status = dc_link_dp_perform_link_training( + udelay(1000); + if (read_hpd_rx_irq_data(link, &irq_data) == DC_OK && + hpd_rx_irq_check_link_loss_status( link, - link_res, - cur, - skip_video_pattern); - if (status == LINK_TRAINING_SUCCESS) - success = true; - else + &irq_data)) (*fail_count)++; - } - if (success) { - link->verified_link_cap = *cur; - udelay(1000); - if (read_hpd_rx_irq_data(link, &irq_data) == DC_OK) - if (hpd_rx_irq_check_link_loss_status( - link, - &irq_data)) - (*fail_count)++; + } else { + (*fail_count)++; } - /* always disable the link before trying another - * setting or before returning we'll enable it later - * based on the actual mode we're driving - */ - dp_disable_link_phy(link, link_res, link->connector_signal); + dp_disable_link_phy(link, &link_res, link->connector_signal); } while (!success && decide_fallback_link_setting(link, - initial_link_settings, cur, status)); - - /* Link Training failed for all Link Settings - * (Lane Count is still unknown) - */ - if (!success) { - /* If all LT fails for all settings, - * set verified = failed safe (1 lane low) - */ - link->verified_link_cap.lane_count = LANE_COUNT_ONE; - link->verified_link_cap.link_rate = LINK_RATE_LOW; + initial_link_settings, &cur_link_settings, status)); - link->verified_link_cap.link_spread = - LINK_SPREAD_DISABLED; - } + link->verified_link_cap = success ? + cur_link_settings : fail_safe_link_settings; + return success; +} +static void apply_usbc_combo_phy_reset_wa(struct dc_link *link, + struct dc_link_settings *link_settings) +{ + /* Temporary Renoir-specific workaround PHY will sometimes be in bad + * state on hotplugging display from certain USB-C dongle, so add extra + * cycle of enabling and disabling the PHY before first link training. + */ + struct link_resource link_res = {0}; + enum clock_source_id dp_cs_id = get_clock_source_id(link); - return success; + dp_enable_link_phy(link, &link_res, link->connector_signal, + dp_cs_id, link_settings); + dp_disable_link_phy(link, &link_res, link->connector_signal); } bool dp_verify_link_cap_with_retries( struct dc_link *link, - const struct link_resource *link_res, struct dc_link_settings *known_limit_link_setting, int attempts) { int i = 0; bool success = false; + if (link->link_enc && link->link_enc->features.flags.bits.DP_IS_USB_C && + link->dc->debug.usbc_combo_phy_reset_wa) + apply_usbc_combo_phy_reset_wa(link, known_limit_link_setting); + for (i = 0; i < attempts; i++) { int fail_count = 0; enum dc_connection_type type = dc_connection_none; @@ -3525,12 +3286,9 @@ bool dp_verify_link_cap_with_retries( memset(&link->verified_link_cap, 0, sizeof(struct dc_link_settings)); if (!dc_link_detect_sink(link, &type) || type == dc_connection_none) { - link->verified_link_cap.lane_count = LANE_COUNT_ONE; - link->verified_link_cap.link_rate = LINK_RATE_LOW; - link->verified_link_cap.link_spread = LINK_SPREAD_DISABLED; + link->verified_link_cap = fail_safe_link_settings; break; - } else if (dp_verify_link_cap(link, link_res, - known_limit_link_setting, + } else if (dp_verify_link_cap(link, known_limit_link_setting, &fail_count) && fail_count == 0) { success = true; break; @@ -3540,79 +3298,32 @@ bool dp_verify_link_cap_with_retries( return success; } -bool dp_verify_mst_link_cap( - struct dc_link *link, const struct link_resource *link_res) -{ - struct dc_link_settings max_link_cap = {0}; - - if (dp_get_link_encoding_format(&link->reported_link_cap) == - DP_8b_10b_ENCODING) { - max_link_cap = get_max_link_cap(link, link_res); - link->verified_link_cap = get_common_supported_link_settings( - link->reported_link_cap, - max_link_cap); - } -#if defined(CONFIG_DRM_AMD_DC_DCN) - else if (dp_get_link_encoding_format(&link->reported_link_cap) == - DP_128b_132b_ENCODING) { - dp_verify_link_cap_with_retries(link, - link_res, - &link->reported_link_cap, - LINK_TRAINING_MAX_VERIFY_RETRY); - } -#endif - return true; -} - -static struct dc_link_settings get_common_supported_link_settings( - struct dc_link_settings link_setting_a, - struct dc_link_settings link_setting_b) +/* in DP compliance test, DPR-120 may have + * a random value in its MAX_LINK_BW dpcd field. + * We map it to the maximum supported link rate that + * is smaller than MAX_LINK_BW in this case. + */ +static enum dc_link_rate get_link_rate_from_max_link_bw( + uint8_t max_link_bw) { - struct dc_link_settings link_settings = {0}; + enum dc_link_rate link_rate; - link_settings.lane_count = - (link_setting_a.lane_count <= - link_setting_b.lane_count) ? - link_setting_a.lane_count : - link_setting_b.lane_count; - link_settings.link_rate = - (link_setting_a.link_rate <= - link_setting_b.link_rate) ? - link_setting_a.link_rate : - link_setting_b.link_rate; - link_settings.link_spread = LINK_SPREAD_DISABLED; - - /* in DP compliance test, DPR-120 may have - * a random value in its MAX_LINK_BW dpcd field. - * We map it to the maximum supported link rate that - * is smaller than MAX_LINK_BW in this case. - */ -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (link_settings.link_rate > LINK_RATE_UHBR20) { - link_settings.link_rate = LINK_RATE_UHBR20; - } else if (link_settings.link_rate < LINK_RATE_UHBR20 && - link_settings.link_rate > LINK_RATE_UHBR13_5) { - link_settings.link_rate = LINK_RATE_UHBR13_5; - } else if (link_settings.link_rate < LINK_RATE_UHBR10 && - link_settings.link_rate > LINK_RATE_HIGH3) { -#else - if (link_settings.link_rate > LINK_RATE_HIGH3) { -#endif - link_settings.link_rate = LINK_RATE_HIGH3; - } else if (link_settings.link_rate < LINK_RATE_HIGH3 - && link_settings.link_rate > LINK_RATE_HIGH2) { - link_settings.link_rate = LINK_RATE_HIGH2; - } else if (link_settings.link_rate < LINK_RATE_HIGH2 - && link_settings.link_rate > LINK_RATE_HIGH) { - link_settings.link_rate = LINK_RATE_HIGH; - } else if (link_settings.link_rate < LINK_RATE_HIGH - && link_settings.link_rate > LINK_RATE_LOW) { - link_settings.link_rate = LINK_RATE_LOW; - } else if (link_settings.link_rate < LINK_RATE_LOW) { - link_settings.link_rate = LINK_RATE_UNKNOWN; + if (max_link_bw >= LINK_RATE_HIGH3) { + link_rate = LINK_RATE_HIGH3; + } else if (max_link_bw < LINK_RATE_HIGH3 + && max_link_bw >= LINK_RATE_HIGH2) { + link_rate = LINK_RATE_HIGH2; + } else if (max_link_bw < LINK_RATE_HIGH2 + && max_link_bw >= LINK_RATE_HIGH) { + link_rate = LINK_RATE_HIGH; + } else if (max_link_bw < LINK_RATE_HIGH + && max_link_bw >= LINK_RATE_LOW) { + link_rate = LINK_RATE_LOW; + } else { + link_rate = LINK_RATE_UNKNOWN; } - return link_settings; + return link_rate; } static inline bool reached_minimum_lane_count(enum dc_lane_count lane_count) @@ -3642,14 +3353,12 @@ static enum dc_lane_count reduce_lane_count(enum dc_lane_count lane_count) static enum dc_link_rate reduce_link_rate(enum dc_link_rate link_rate) { switch (link_rate) { -#if defined(CONFIG_DRM_AMD_DC_DCN) case LINK_RATE_UHBR20: return LINK_RATE_UHBR13_5; case LINK_RATE_UHBR13_5: return LINK_RATE_UHBR10; case LINK_RATE_UHBR10: return LINK_RATE_HIGH3; -#endif case LINK_RATE_HIGH3: return LINK_RATE_HIGH2; case LINK_RATE_HIGH2: @@ -3684,20 +3393,17 @@ static enum dc_link_rate increase_link_rate(enum dc_link_rate link_rate) return LINK_RATE_HIGH2; case LINK_RATE_HIGH2: return LINK_RATE_HIGH3; -#if defined(CONFIG_DRM_AMD_DC_DCN) case LINK_RATE_HIGH3: return LINK_RATE_UHBR10; case LINK_RATE_UHBR10: return LINK_RATE_UHBR13_5; case LINK_RATE_UHBR13_5: return LINK_RATE_UHBR20; -#endif default: return LINK_RATE_UNKNOWN; } } -#if defined(CONFIG_DRM_AMD_DC_DCN) static bool decide_fallback_link_setting_max_bw_policy( const struct dc_link_settings *max, struct dc_link_settings *cur) @@ -3731,7 +3437,6 @@ static bool decide_fallback_link_setting_max_bw_policy( return found; } -#endif /* * function: set link rate and lane count fallback based @@ -3749,12 +3454,10 @@ static bool decide_fallback_link_setting( { if (!current_link_setting) return false; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (dp_get_link_encoding_format(&initial_link_settings) == DP_128b_132b_ENCODING || link->dc->debug.force_dp2_lt_fallback_method) return decide_fallback_link_setting_max_bw_policy(&initial_link_settings, current_link_setting); -#endif switch (training_result) { case LINK_TRAINING_CR_FAIL_LANE0: @@ -3801,6 +3504,7 @@ static bool decide_fallback_link_setting( current_link_setting->link_rate = reduce_link_rate( current_link_setting->link_rate); + current_link_setting->lane_count = initial_link_settings.lane_count; } else { return false; } @@ -3813,6 +3517,7 @@ static bool decide_fallback_link_setting( current_link_setting->link_rate = reduce_link_rate( current_link_setting->link_rate); + current_link_setting->lane_count = initial_link_settings.lane_count; } else { return false; } @@ -4292,15 +3997,9 @@ static void dp_test_send_phy_test_pattern(struct dc_link *link) union phy_test_pattern dpcd_test_pattern; union lane_adjust dpcd_lane_adjustment[2]; unsigned char dpcd_post_cursor_2_adjustment = 0; -#if defined(CONFIG_DRM_AMD_DC_DCN) unsigned char test_pattern_buffer[ (DP_TEST_264BIT_CUSTOM_PATTERN_263_256 - DP_TEST_264BIT_CUSTOM_PATTERN_7_0)+1] = {0}; -#else - unsigned char test_pattern_buffer[ - (DP_TEST_80BIT_CUSTOM_PATTERN_79_72 - - DP_TEST_80BIT_CUSTOM_PATTERN_7_0)+1] = {0}; -#endif unsigned int test_pattern_size = 0; enum dp_test_pattern test_pattern; union lane_adjust dpcd_lane_adjust; @@ -4371,7 +4070,6 @@ static void dp_test_send_phy_test_pattern(struct dc_link *link) case PHY_TEST_PATTERN_CP2520_3: test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN4; break; -#if defined(CONFIG_DRM_AMD_DC_DCN) case PHY_TEST_PATTERN_128b_132b_TPS1: test_pattern = DP_TEST_PATTERN_128b_132b_TPS1; break; @@ -4399,7 +4097,6 @@ static void dp_test_send_phy_test_pattern(struct dc_link *link) case PHY_TEST_PATTERN_SQUARE_PULSE: test_pattern = DP_TEST_PATTERN_SQUARE_PULSE; break; -#endif default: test_pattern = DP_TEST_PATTERN_VIDEO_MODE; break; @@ -4415,7 +4112,6 @@ static void dp_test_send_phy_test_pattern(struct dc_link *link) test_pattern_size); } -#if defined(CONFIG_DRM_AMD_DC_DCN) if (test_pattern == DP_TEST_PATTERN_SQUARE_PULSE) { test_pattern_size = 1; // Square pattern data is 1 byte (DP spec) core_link_read_dpcd( @@ -4434,7 +4130,6 @@ static void dp_test_send_phy_test_pattern(struct dc_link *link) test_pattern_buffer, test_pattern_size); } -#endif /* prepare link training settings */ link_training_settings.link_settings = link->cur_link_settings; @@ -4455,14 +4150,11 @@ static void dp_test_send_phy_test_pattern(struct dc_link *link) link_training_settings.hw_lane_settings[lane].POST_CURSOR2 = (enum dc_post_cursor2) ((dpcd_post_cursor_2_adjustment >> (lane * 2)) & 0x03); - } -#if defined(CONFIG_DRM_AMD_DC_DCN) - else if (dp_get_link_encoding_format(&link->cur_link_settings) == + } else if (dp_get_link_encoding_format(&link->cur_link_settings) == DP_128b_132b_ENCODING) { link_training_settings.hw_lane_settings[lane].FFE_PRESET.raw = dpcd_lane_adjust.tx_ffe.PRESET_VALUE; } -#endif } dp_hw_to_dpcd_lane_settings(&link_training_settings, @@ -4970,7 +4662,7 @@ uint32_t dc_link_bw_kbps_from_raw_frl_link_rate_data(uint8_t bw) return 0; } -/** +/* * Return PCON's post FRL link training supported BW if its non-zero, otherwise return max_supported_frl_bw. */ static uint32_t intersect_frl_link_bw_support( @@ -5159,7 +4851,6 @@ static void get_active_converter_info( dp_hw_fw_revision.ieee_fw_rev, sizeof(dp_hw_fw_revision.ieee_fw_rev)); } -#if defined(CONFIG_DRM_AMD_DC_DCN) if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_14 && link->dpcd_caps.dongle_type != DISPLAY_DONGLE_NONE) { union dp_dfp_cap_ext dfp_cap_ext; @@ -5195,7 +4886,6 @@ static void get_active_converter_info( DC_LOG_DP2("\tdfp_cap_ext.max_video_h_active_width = %d", link->dpcd_caps.dongle_caps.dfp_cap_ext.max_video_h_active_width); DC_LOG_DP2("\tdfp_cap_ext.max_video_v_active_height = %d", link->dpcd_caps.dongle_caps.dfp_cap_ext.max_video_v_active_height); } -#endif } static void dp_wa_power_up_0010FA(struct dc_link *link, uint8_t *dpcd_data, @@ -5255,12 +4945,8 @@ static bool dpcd_read_sink_ext_caps(struct dc_link *link) bool dp_retrieve_lttpr_cap(struct dc_link *link) { -#if defined(CONFIG_DRM_AMD_DC_DCN) uint8_t lttpr_dpcd_data[8]; bool allow_lttpr_non_transparent_mode = 0; -#else - uint8_t lttpr_dpcd_data[6]; -#endif bool vbios_lttpr_enable = link->dc->caps.vbios_lttpr_enable; bool vbios_lttpr_interop = link->dc->caps.vbios_lttpr_aware; enum dc_status status = DC_ERROR_UNEXPECTED; @@ -5268,7 +4954,6 @@ bool dp_retrieve_lttpr_cap(struct dc_link *link) memset(lttpr_dpcd_data, '\0', sizeof(lttpr_dpcd_data)); -#if defined(CONFIG_DRM_AMD_DC_DCN) if ((link->dc->config.allow_lttpr_non_transparent_mode.bits.DP2_0 && link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED)) { allow_lttpr_non_transparent_mode = 1; @@ -5276,7 +4961,6 @@ bool dp_retrieve_lttpr_cap(struct dc_link *link) !link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED) { allow_lttpr_non_transparent_mode = 1; } -#endif /* * Logic to determine LTTPR mode @@ -5285,21 +4969,12 @@ bool dp_retrieve_lttpr_cap(struct dc_link *link) if (vbios_lttpr_enable && vbios_lttpr_interop) link->lttpr_mode = LTTPR_MODE_NON_TRANSPARENT; else if (!vbios_lttpr_enable && vbios_lttpr_interop) { -#if defined(CONFIG_DRM_AMD_DC_DCN) if (allow_lttpr_non_transparent_mode) -#else - if (link->dc->config.allow_lttpr_non_transparent_mode) -#endif link->lttpr_mode = LTTPR_MODE_NON_TRANSPARENT; else link->lttpr_mode = LTTPR_MODE_TRANSPARENT; } else if (!vbios_lttpr_enable && !vbios_lttpr_interop) { -#if defined(CONFIG_DRM_AMD_DC_DCN) if (!allow_lttpr_non_transparent_mode || !link->dc->caps.extended_aux_timeout_support) -#else - if (!link->dc->config.allow_lttpr_non_transparent_mode - || !link->dc->caps.extended_aux_timeout_support) -#endif link->lttpr_mode = LTTPR_MODE_NON_LTTPR; else link->lttpr_mode = LTTPR_MODE_NON_TRANSPARENT; @@ -5349,7 +5024,6 @@ bool dp_retrieve_lttpr_cap(struct dc_link *link) lttpr_dpcd_data[DP_PHY_REPEATER_EXTENDED_WAIT_TIMEOUT - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; -#if defined(CONFIG_DRM_AMD_DC_DCN) link->dpcd_caps.lttpr_caps.main_link_channel_coding.raw = lttpr_dpcd_data[DP_MAIN_LINK_CHANNEL_CODING_PHY_REPEATER - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; @@ -5357,12 +5031,10 @@ bool dp_retrieve_lttpr_cap(struct dc_link *link) link->dpcd_caps.lttpr_caps.supported_128b_132b_rates.raw = lttpr_dpcd_data[DP_PHY_REPEATER_128b_132b_RATES - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; -#endif /* Attempt to train in LTTPR transparent mode if repeater count exceeds 8. */ - is_lttpr_present = (dp_convert_to_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt) != 0 && + is_lttpr_present = (link->dpcd_caps.lttpr_caps.max_lane_count > 0 && link->dpcd_caps.lttpr_caps.phy_repeater_cnt < 0xff && - link->dpcd_caps.lttpr_caps.max_lane_count > 0 && link->dpcd_caps.lttpr_caps.max_lane_count <= 4 && link->dpcd_caps.lttpr_caps.revision.raw >= 0x14); if (is_lttpr_present) { @@ -5374,6 +5046,13 @@ bool dp_retrieve_lttpr_cap(struct dc_link *link) return is_lttpr_present; } + +static bool is_usbc_connector(struct dc_link *link) +{ + return link->link_enc && + link->link_enc->features.flags.bits.DP_IS_USB_C; +} + static bool retrieve_link_cap(struct dc_link *link) { /* DP_ADAPTER_CAP - DP_DPCD_REV + 1 == 16 and also DP_DSC_BITS_PER_PIXEL_INC - DP_DSC_SUPPORT + 1 == 16, @@ -5430,6 +5109,9 @@ static bool retrieve_link_cap(struct dc_link *link) */ msleep(post_oui_delay); + /* Read cable ID and update receiver */ + dpcd_update_cable_id(link); + for (i = 0; i < read_dpcd_retry_cnt; i++) { status = core_link_read_dpcd( link, @@ -5516,6 +5198,9 @@ static bool retrieve_link_cap(struct dc_link *link) read_dp_device_vendor_id(link); + /* TODO - decouple raw mst capability from policy decision */ + link->dpcd_caps.is_mst_capable = is_mst_supported(link); + get_active_converter_info(ds_port.byte, link); dp_wa_power_up_0010FA(link, dpcd_data, sizeof(dpcd_data)); @@ -5534,8 +5219,8 @@ static bool retrieve_link_cap(struct dc_link *link) link->reported_link_cap.lane_count = link->dpcd_caps.max_ln_count.bits.MAX_LANE_COUNT; - link->reported_link_cap.link_rate = dpcd_data[ - DP_MAX_LINK_RATE - DP_DPCD_REV]; + link->reported_link_cap.link_rate = get_link_rate_from_max_link_bw( + dpcd_data[DP_MAX_LINK_RATE - DP_DPCD_REV]); link->reported_link_cap.link_spread = link->dpcd_caps.max_down_spread.bits.MAX_DOWN_SPREAD ? LINK_SPREAD_05_DOWNSPREAD_30KHZ : LINK_SPREAD_DISABLED; @@ -5632,7 +5317,6 @@ static bool retrieve_link_cap(struct dc_link *link) DP_DSC_SUPPORT, link->dpcd_caps.dsc_caps.dsc_basic_caps.raw, sizeof(link->dpcd_caps.dsc_caps.dsc_basic_caps.raw)); -#if defined(CONFIG_DRM_AMD_DC_DCN) if (link->dpcd_caps.dongle_type != DISPLAY_DONGLE_NONE) { status = core_link_read_dpcd( link, @@ -5647,19 +5331,33 @@ static bool retrieve_link_cap(struct dc_link *link) DC_LOG_DSC("\tBRANCH_MAX_LINE_WIDTH 0x%02x", link->dpcd_caps.dsc_caps.dsc_branch_decoder_caps.fields.BRANCH_MAX_LINE_WIDTH); } -#else - status = core_link_read_dpcd( - link, - DP_DSC_BRANCH_OVERALL_THROUGHPUT_0, - link->dpcd_caps.dsc_caps.dsc_branch_decoder_caps.raw, - sizeof(link->dpcd_caps.dsc_caps.dsc_branch_decoder_caps.raw)); + + /* Apply work around to disable FEC and DSC for USB4 tunneling in TBT3 compatibility mode + * only if required. + */ + if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && +#if defined(CONFIG_DRM_AMD_DC_DCN) + !link->dc->debug.dpia_debug.bits.disable_force_tbt3_work_around && #endif + link->dpcd_caps.is_branch_dev && + link->dpcd_caps.branch_dev_id == DP_BRANCH_DEVICE_ID_90CC24 && + link->dpcd_caps.branch_hw_revision == DP_BRANCH_HW_REV_10 && + (link->dpcd_caps.fec_cap.bits.FEC_CAPABLE || + link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_SUPPORT)) { + /* A TBT3 device is expected to report no support for FEC or DSC to a USB4 DPIA. + * Clear FEC and DSC capabilities as a work around if that is not the case. + */ + link->wa_flags.dpia_forced_tbt3_mode = true; + memset(&link->dpcd_caps.dsc_caps, '\0', sizeof(link->dpcd_caps.dsc_caps)); + memset(&link->dpcd_caps.fec_cap, '\0', sizeof(link->dpcd_caps.fec_cap)); + DC_LOG_DSC("Clear DSC SUPPORT for USB4 link(%d) in TBT3 compatibility mode", link->link_index); + } else + link->wa_flags.dpia_forced_tbt3_mode = false; } if (!dpcd_read_sink_ext_caps(link)) link->dpcd_sink_ext_caps.raw = 0; -#if defined(CONFIG_DRM_AMD_DC_DCN) link->dpcd_caps.channel_coding_cap.raw = dpcd_data[DP_MAIN_LINK_CHANNEL_CODING_CAP - DP_DPCD_REV]; if (link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED) { @@ -5706,7 +5404,6 @@ static bool retrieve_link_cap(struct dc_link *link) if (link->dpcd_caps.fec_cap1.bits.AGGREGATED_ERROR_COUNTERS_CAPABLE) DC_LOG_DP2("\tFEC aggregated error counters are supported"); } -#endif /* Connectivity log: detection */ CONN_DATA_DETECT(link, dpcd_data, sizeof(dpcd_data), "Rx Caps: "); @@ -5779,14 +5476,6 @@ bool dp_overwrite_extended_receiver_cap(struct dc_link *link) bool detect_dp_sink_caps(struct dc_link *link) { return retrieve_link_cap(link); - - /* dc init_hw has power encoder using default - * signal for connector. For native DP, no - * need to power up encoder again. If not native - * DP, hw_init may need check signal or power up - * encoder here. - */ - /* TODO save sink caps in link->sink */ } static enum dc_link_rate linkRateInKHzToLinkRateMultiplier(uint32_t link_rate_in_khz) @@ -5864,8 +5553,6 @@ void detect_edp_sink_caps(struct dc_link *link) } } } - link->verified_link_cap = link->reported_link_cap; - core_link_read_dpcd(link, DP_EDP_BACKLIGHT_ADJUSTMENT_CAP, &backlight_adj_cap, sizeof(backlight_adj_cap)); @@ -6031,7 +5718,7 @@ static void set_crtc_test_pattern(struct dc_link *link, else if (link->dc->hwss.set_disp_pattern_generator) { struct pipe_ctx *odm_pipe; int opp_cnt = 1; - int dpg_width = width; + int dpg_width; for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) opp_cnt++; @@ -6188,7 +5875,6 @@ bool dc_link_dp_set_test_pattern( case DP_TEST_PATTERN_CP2520_3: pattern = PHY_TEST_PATTERN_CP2520_3; break; -#if defined(CONFIG_DRM_AMD_DC_DCN) case DP_TEST_PATTERN_128b_132b_TPS1: pattern = PHY_TEST_PATTERN_128b_132b_TPS1; break; @@ -6216,7 +5902,6 @@ bool dc_link_dp_set_test_pattern( case DP_TEST_PATTERN_SQUARE_PULSE: pattern = PHY_TEST_PATTERN_SQUARE_PULSE; break; -#endif default: return false; } @@ -6482,14 +6167,7 @@ enum dc_status dp_set_fec_ready(struct dc_link *link, const struct link_resource enum dc_status status = DC_OK; uint8_t fec_config = 0; - /* Access link encoder based on whether it is statically - * or dynamically assigned to a link. - */ - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - else - link_enc = link->link_enc; + link_enc = link_enc_cfg_get_link_enc(link); ASSERT(link_enc); if (!dc_link_should_enable_fec(link)) @@ -6529,14 +6207,7 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) { struct link_encoder *link_enc = NULL; - /* Access link encoder based on whether it is statically - * or dynamically assigned to a link. - */ - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - else - link_enc = link->link_enc; + link_enc = link_enc_cfg_get_link_enc(link); ASSERT(link_enc); if (!dc_link_should_enable_fec(link)) @@ -6562,23 +6233,6 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) } } -struct link_encoder *dp_get_link_enc(struct dc_link *link) -{ - struct link_encoder *link_enc; - - link_enc = link->link_enc; - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) { - link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, - link); - if (!link->link_enc) - link_enc = link_enc_cfg_get_next_avail_link_enc( - link->ctx->dc); - } - - return link_enc; -} - void dpcd_set_source_specific_data(struct dc_link *link) { if (!link->dc->vendor_signature.is_valid) { @@ -6655,6 +6309,31 @@ void dpcd_set_source_specific_data(struct dc_link *link) } } +void dpcd_update_cable_id(struct dc_link *link) +{ + struct link_encoder *link_enc = NULL; + + link_enc = link_enc_cfg_get_link_enc(link); + + if (!link_enc || + !link_enc->features.flags.bits.IS_UHBR10_CAPABLE || + link->dprx_status.cable_id_updated) + return; + + /* Retrieve cable attributes */ + if (!is_usbc_connector(link)) + core_link_read_dpcd(link, DP_CABLE_ATTRIBUTES_UPDATED_BY_DPRX, + &link->dpcd_caps.cable_attributes.raw, + sizeof(uint8_t)); + + /* Update receiver with cable attributes */ + core_link_write_dpcd(link, DP_CABLE_ATTRIBUTES_UPDATED_BY_DPTX, + &link->dpcd_caps.cable_attributes.raw, + sizeof(link->dpcd_caps.cable_attributes.raw)); + + link->dprx_status.cable_id_updated = 1; +} + bool dc_link_set_backlight_level_nits(struct dc_link *link, bool isHDR, uint32_t backlight_millinits, @@ -6819,15 +6498,12 @@ enum dp_link_encoding dp_get_link_encoding_format(const struct dc_link_settings if ((link_settings->link_rate >= LINK_RATE_LOW) && (link_settings->link_rate <= LINK_RATE_HIGH3)) return DP_8b_10b_ENCODING; -#if defined(CONFIG_DRM_AMD_DC_DCN) else if ((link_settings->link_rate >= LINK_RATE_UHBR10) && (link_settings->link_rate <= LINK_RATE_UHBR20)) return DP_128b_132b_ENCODING; -#endif return DP_UNKNOWN_ENCODING; } -#if defined(CONFIG_DRM_AMD_DC_DCN) enum dp_link_encoding dc_link_dp_mst_decide_link_encoding_format(const struct dc_link *link) { struct dc_link_settings link_settings = {0}; @@ -7044,7 +6720,6 @@ bool is_dp_128b_132b_signal(struct pipe_ctx *pipe_ctx) pipe_ctx->link_res.hpo_dp_link_enc && dc_is_dp_signal(pipe_ctx->stream->signal)); } -#endif void edp_panel_backlight_power_on(struct dc_link *link) { @@ -7056,3 +6731,621 @@ void edp_panel_backlight_power_on(struct dc_link *link) if (link->dc->hwss.edp_backlight_control) link->dc->hwss.edp_backlight_control(link, true); } + +void dc_link_dp_clear_rx_status(struct dc_link *link) +{ + memset(&link->dprx_status, 0, sizeof(link->dprx_status)); +} + +void dp_receiver_power_ctrl(struct dc_link *link, bool on) +{ + uint8_t state; + + state = on ? DP_POWER_STATE_D0 : DP_POWER_STATE_D3; + + if (link->sync_lt_in_progress) + return; + + core_link_write_dpcd(link, DP_SET_POWER, &state, + sizeof(state)); + +} + +void dp_source_sequence_trace(struct dc_link *link, uint8_t dp_test_mode) +{ + if (link != NULL && link->dc->debug.enable_driver_sequence_debug) + core_link_write_dpcd(link, DP_SOURCE_SEQUENCE, + &dp_test_mode, sizeof(dp_test_mode)); +} + + +static uint8_t convert_to_count(uint8_t lttpr_repeater_count) +{ + switch (lttpr_repeater_count) { + case 0x80: // 1 lttpr repeater + return 1; + case 0x40: // 2 lttpr repeaters + return 2; + case 0x20: // 3 lttpr repeaters + return 3; + case 0x10: // 4 lttpr repeaters + return 4; + case 0x08: // 5 lttpr repeaters + return 5; + case 0x04: // 6 lttpr repeaters + return 6; + case 0x02: // 7 lttpr repeaters + return 7; + case 0x01: // 8 lttpr repeaters + return 8; + default: + break; + } + return 0; // invalid value +} + +static inline bool is_immediate_downstream(struct dc_link *link, uint32_t offset) +{ + return (convert_to_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt) == offset); +} + +void dp_enable_link_phy( + struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings) +{ + struct dc *dc = link->ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; + struct pipe_ctx *pipes = + link->dc->current_state->res_ctx.pipe_ctx; + struct clock_source *dp_cs = + link->dc->res_pool->dp_clock_source; + const struct link_hwss *link_hwss = get_link_hwss(link, link_res); + unsigned int i; + + if (link->connector_signal == SIGNAL_TYPE_EDP) { + link->dc->hwss.edp_power_control(link, true); + link->dc->hwss.edp_wait_for_hpd_ready(link, true); + } + + /* If the current pixel clock source is not DTO(happens after + * switching from HDMI passive dongle to DP on the same connector), + * switch the pixel clock source to DTO. + */ + for (i = 0; i < MAX_PIPES; i++) { + if (pipes[i].stream != NULL && + pipes[i].stream->link == link) { + if (pipes[i].clock_source != NULL && + pipes[i].clock_source->id != CLOCK_SOURCE_ID_DP_DTO) { + pipes[i].clock_source = dp_cs; + pipes[i].stream_res.pix_clk_params.requested_pix_clk_100hz = + pipes[i].stream->timing.pix_clk_100hz; + pipes[i].clock_source->funcs->program_pix_clk( + pipes[i].clock_source, + &pipes[i].stream_res.pix_clk_params, + &pipes[i].pll_settings); + } + } + } + + link->cur_link_settings = *link_settings; + + if (dp_get_link_encoding_format(link_settings) == DP_8b_10b_ENCODING) { + if (dc->clk_mgr->funcs->notify_link_rate_change) + dc->clk_mgr->funcs->notify_link_rate_change(dc->clk_mgr, link); + } + + if (dmcu != NULL && dmcu->funcs->lock_phy) + dmcu->funcs->lock_phy(dmcu); + + if (link_hwss->ext.enable_dp_link_output) + link_hwss->ext.enable_dp_link_output(link, link_res, signal, + clock_source, link_settings); + + if (dmcu != NULL && dmcu->funcs->unlock_phy) + dmcu->funcs->unlock_phy(dmcu); + + dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_ENABLE_LINK_PHY); + dp_receiver_power_ctrl(link, true); +} + +void edp_add_delay_for_T9(struct dc_link *link) +{ + if (link->local_sink && + link->local_sink->edid_caps.panel_patch.extra_delay_backlight_off > 0) + udelay(link->local_sink->edid_caps.panel_patch.extra_delay_backlight_off * 1000); +} + +bool edp_receiver_ready_T9(struct dc_link *link) +{ + unsigned int tries = 0; + unsigned char sinkstatus = 0; + unsigned char edpRev = 0; + enum dc_status result = DC_OK; + + result = core_link_read_dpcd(link, DP_EDP_DPCD_REV, &edpRev, sizeof(edpRev)); + + /* start from eDP version 1.2, SINK_STAUS indicate the sink is ready.*/ + if (result == DC_OK && edpRev >= DP_EDP_12) { + do { + sinkstatus = 1; + result = core_link_read_dpcd(link, DP_SINK_STATUS, &sinkstatus, sizeof(sinkstatus)); + if (sinkstatus == 0) + break; + if (result != DC_OK) + break; + udelay(100); //MAx T9 + } while (++tries < 50); + } + + return result; +} +bool edp_receiver_ready_T7(struct dc_link *link) +{ + unsigned char sinkstatus = 0; + unsigned char edpRev = 0; + enum dc_status result = DC_OK; + + /* use absolute time stamp to constrain max T7*/ + unsigned long long enter_timestamp = 0; + unsigned long long finish_timestamp = 0; + unsigned long long time_taken_in_ns = 0; + + result = core_link_read_dpcd(link, DP_EDP_DPCD_REV, &edpRev, sizeof(edpRev)); + + if (result == DC_OK && edpRev >= DP_EDP_12) { + /* start from eDP version 1.2, SINK_STAUS indicate the sink is ready.*/ + enter_timestamp = dm_get_timestamp(link->ctx); + do { + sinkstatus = 0; + result = core_link_read_dpcd(link, DP_SINK_STATUS, &sinkstatus, sizeof(sinkstatus)); + if (sinkstatus == 1) + break; + if (result != DC_OK) + break; + udelay(25); + finish_timestamp = dm_get_timestamp(link->ctx); + time_taken_in_ns = dm_get_elapse_time_in_ns(link->ctx, finish_timestamp, enter_timestamp); + } while (time_taken_in_ns < 50 * 1000000); //MAx T7 is 50ms + } + + if (link->local_sink && + link->local_sink->edid_caps.panel_patch.extra_t7_ms > 0) + udelay(link->local_sink->edid_caps.panel_patch.extra_t7_ms * 1000); + + return result; +} + +void dp_disable_link_phy(struct dc_link *link, const struct link_resource *link_res, + enum signal_type signal) +{ + struct dc *dc = link->ctx->dc; + struct dmcu *dmcu = dc->res_pool->dmcu; + const struct link_hwss *link_hwss = get_link_hwss(link, link_res); + + if (!link->wa_flags.dp_keep_receiver_powered) + dp_receiver_power_ctrl(link, false); + + if (signal == SIGNAL_TYPE_EDP) { + if (link->dc->hwss.edp_backlight_control) + link->dc->hwss.edp_backlight_control(link, false); + if (link_hwss->ext.disable_dp_link_output) + link_hwss->ext.disable_dp_link_output(link, link_res, signal); + link->dc->hwss.edp_power_control(link, false); + } else { + if (dmcu != NULL && dmcu->funcs->lock_phy) + dmcu->funcs->lock_phy(dmcu); + if (link_hwss->ext.disable_dp_link_output) + link_hwss->ext.disable_dp_link_output(link, link_res, signal); + if (dmcu != NULL && dmcu->funcs->unlock_phy) + dmcu->funcs->unlock_phy(dmcu); + } + + dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY); + + /* Clear current link setting.*/ + memset(&link->cur_link_settings, 0, + sizeof(link->cur_link_settings)); + + if (dc->clk_mgr->funcs->notify_link_rate_change) + dc->clk_mgr->funcs->notify_link_rate_change(dc->clk_mgr, link); +} + +void dp_disable_link_phy_mst(struct dc_link *link, const struct link_resource *link_res, + enum signal_type signal) +{ + /* MST disable link only when no stream use the link */ + if (link->mst_stream_alloc_table.stream_count > 0) + return; + + dp_disable_link_phy(link, link_res, signal); + + /* set the sink to SST mode after disabling the link */ + dp_enable_mst_on_sink(link, false); +} + +bool dp_set_hw_training_pattern( + struct dc_link *link, + const struct link_resource *link_res, + enum dc_dp_training_pattern pattern, + uint32_t offset) +{ + enum dp_test_pattern test_pattern = DP_TEST_PATTERN_UNSUPPORTED; + + switch (pattern) { + case DP_TRAINING_PATTERN_SEQUENCE_1: + test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN1; + break; + case DP_TRAINING_PATTERN_SEQUENCE_2: + test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN2; + break; + case DP_TRAINING_PATTERN_SEQUENCE_3: + test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN3; + break; + case DP_TRAINING_PATTERN_SEQUENCE_4: + test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN4; + break; + case DP_128b_132b_TPS1: + test_pattern = DP_TEST_PATTERN_128b_132b_TPS1_TRAINING_MODE; + break; + case DP_128b_132b_TPS2: + test_pattern = DP_TEST_PATTERN_128b_132b_TPS2_TRAINING_MODE; + break; + default: + break; + } + + dp_set_hw_test_pattern(link, link_res, test_pattern, NULL, 0); + + return true; +} + +void dp_set_hw_lane_settings( + struct dc_link *link, + const struct link_resource *link_res, + const struct link_training_settings *link_settings, + uint32_t offset) +{ + const struct link_hwss *link_hwss = get_link_hwss(link, link_res); + + if ((link->lttpr_mode == LTTPR_MODE_NON_TRANSPARENT) && !is_immediate_downstream(link, offset)) + return; + + if (link_hwss->ext.set_dp_lane_settings) + link_hwss->ext.set_dp_lane_settings(link, link_res, + &link_settings->link_settings, + link_settings->hw_lane_settings); + + memmove(link->cur_lane_setting, + link_settings->hw_lane_settings, + sizeof(link->cur_lane_setting)); +} + +void dp_set_hw_test_pattern( + struct dc_link *link, + const struct link_resource *link_res, + enum dp_test_pattern test_pattern, + uint8_t *custom_pattern, + uint32_t custom_pattern_size) +{ + const struct link_hwss *link_hwss = get_link_hwss(link, link_res); + struct encoder_set_dp_phy_pattern_param pattern_param = {0}; + + pattern_param.dp_phy_pattern = test_pattern; + pattern_param.custom_pattern = custom_pattern; + pattern_param.custom_pattern_size = custom_pattern_size; + pattern_param.dp_panel_mode = dp_get_panel_mode(link); + + if (link_hwss->ext.set_dp_link_test_pattern) + link_hwss->ext.set_dp_link_test_pattern(link, link_res, &pattern_param); +} + +void dp_retrain_link_dp_test(struct dc_link *link, + struct dc_link_settings *link_setting, + bool skip_video_pattern) +{ + struct pipe_ctx *pipes = + &link->dc->current_state->res_ctx.pipe_ctx[0]; + unsigned int i; + + + for (i = 0; i < MAX_PIPES; i++) { + if (pipes[i].stream != NULL && + !pipes[i].top_pipe && !pipes[i].prev_odm_pipe && + pipes[i].stream->link != NULL && + pipes[i].stream_res.stream_enc != NULL && + pipes[i].stream->link == link) { + udelay(100); + + pipes[i].stream_res.stream_enc->funcs->dp_blank(link, + pipes[i].stream_res.stream_enc); + + /* disable any test pattern that might be active */ + dp_set_hw_test_pattern(link, &pipes[i].link_res, + DP_TEST_PATTERN_VIDEO_MODE, NULL, 0); + + dp_receiver_power_ctrl(link, false); + + link->dc->hwss.disable_stream(&pipes[i]); + if ((&pipes[i])->stream_res.audio && !link->dc->debug.az_endpoint_mute_only) + (&pipes[i])->stream_res.audio->funcs->az_disable((&pipes[i])->stream_res.audio); + + if (link->link_enc) + link->link_enc->funcs->disable_output( + link->link_enc, + SIGNAL_TYPE_DISPLAY_PORT); + + /* Clear current link setting. */ + memset(&link->cur_link_settings, 0, + sizeof(link->cur_link_settings)); + + perform_link_training_with_retries( + link_setting, + skip_video_pattern, + LINK_TRAINING_ATTEMPTS, + &pipes[i], + SIGNAL_TYPE_DISPLAY_PORT, + false); + + link->dc->hwss.enable_stream(&pipes[i]); + + link->dc->hwss.unblank_stream(&pipes[i], + link_setting); + + if (pipes[i].stream_res.audio) { + /* notify audio driver for + * audio modes of monitor */ + pipes[i].stream_res.audio->funcs->az_enable( + pipes[i].stream_res.audio); + + /* un-mute audio */ + /* TODO: audio should be per stream rather than + * per link */ + pipes[i].stream_res.stream_enc->funcs-> + audio_mute_control( + pipes[i].stream_res.stream_enc, false); + } + } + } +} + +#undef DC_LOGGER +#define DC_LOGGER \ + dsc->ctx->logger +static void dsc_optc_config_log(struct display_stream_compressor *dsc, + struct dsc_optc_config *config) +{ + uint32_t precision = 1 << 28; + uint32_t bytes_per_pixel_int = config->bytes_per_pixel / precision; + uint32_t bytes_per_pixel_mod = config->bytes_per_pixel % precision; + uint64_t ll_bytes_per_pix_fraq = bytes_per_pixel_mod; + + /* 7 fractional digits decimal precision for bytes per pixel is enough because DSC + * bits per pixel precision is 1/16th of a pixel, which means bytes per pixel precision is + * 1/16/8 = 1/128 of a byte, or 0.0078125 decimal + */ + ll_bytes_per_pix_fraq *= 10000000; + ll_bytes_per_pix_fraq /= precision; + + DC_LOG_DSC("\tbytes_per_pixel 0x%08x (%d.%07d)", + config->bytes_per_pixel, bytes_per_pixel_int, (uint32_t)ll_bytes_per_pix_fraq); + DC_LOG_DSC("\tis_pixel_format_444 %d", config->is_pixel_format_444); + DC_LOG_DSC("\tslice_width %d", config->slice_width); +} + +bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable) +{ + struct dc *dc = pipe_ctx->stream->ctx->dc; + struct dc_stream_state *stream = pipe_ctx->stream; + bool result = false; + + if (dc_is_virtual_signal(stream->signal) || IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) + result = true; + else + result = dm_helpers_dp_write_dsc_enable(dc->ctx, stream, enable); + return result; +} + +/* The stream with these settings can be sent (unblanked) only after DSC was enabled on RX first, + * i.e. after dp_enable_dsc_on_rx() had been called + */ +void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) +{ + struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; + struct dc *dc = pipe_ctx->stream->ctx->dc; + struct dc_stream_state *stream = pipe_ctx->stream; + struct pipe_ctx *odm_pipe; + int opp_cnt = 1; + + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + opp_cnt++; + + if (enable) { + struct dsc_config dsc_cfg; + struct dsc_optc_config dsc_optc_cfg; + enum optc_dsc_mode optc_dsc_mode; + + /* Enable DSC hw block */ + dsc_cfg.pic_width = (stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right) / opp_cnt; + dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; + dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; + dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; + dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; + ASSERT(dsc_cfg.dc_dsc_cfg.num_slices_h % opp_cnt == 0); + dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; + + dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg); + dsc->funcs->dsc_enable(dsc, pipe_ctx->stream_res.opp->inst); + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + struct display_stream_compressor *odm_dsc = odm_pipe->stream_res.dsc; + + odm_dsc->funcs->dsc_set_config(odm_dsc, &dsc_cfg, &dsc_optc_cfg); + odm_dsc->funcs->dsc_enable(odm_dsc, odm_pipe->stream_res.opp->inst); + } + dsc_cfg.dc_dsc_cfg.num_slices_h *= opp_cnt; + dsc_cfg.pic_width *= opp_cnt; + + optc_dsc_mode = dsc_optc_cfg.is_pixel_format_444 ? OPTC_DSC_ENABLED_444 : OPTC_DSC_ENABLED_NATIVE_SUBSAMPLED; + + /* Enable DSC in encoder */ + if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) + && !is_dp_128b_132b_signal(pipe_ctx)) { + DC_LOG_DSC("Setting stream encoder DSC config for engine %d:", (int)pipe_ctx->stream_res.stream_enc->id); + dsc_optc_config_log(dsc, &dsc_optc_cfg); + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(pipe_ctx->stream_res.stream_enc, + optc_dsc_mode, + dsc_optc_cfg.bytes_per_pixel, + dsc_optc_cfg.slice_width); + + /* PPS SDP is set elsewhere because it has to be done after DIG FE is connected to DIG BE */ + } + + /* Enable DSC in OPTC */ + DC_LOG_DSC("Setting optc DSC config for tg instance %d:", pipe_ctx->stream_res.tg->inst); + dsc_optc_config_log(dsc, &dsc_optc_cfg); + pipe_ctx->stream_res.tg->funcs->set_dsc_config(pipe_ctx->stream_res.tg, + optc_dsc_mode, + dsc_optc_cfg.bytes_per_pixel, + dsc_optc_cfg.slice_width); + } else { + /* disable DSC in OPTC */ + pipe_ctx->stream_res.tg->funcs->set_dsc_config( + pipe_ctx->stream_res.tg, + OPTC_DSC_DISABLED, 0, 0); + + /* disable DSC in stream encoder */ + if (dc_is_dp_signal(stream->signal)) { + if (is_dp_128b_132b_signal(pipe_ctx)) + pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.hpo_dp_stream_enc, + false, + NULL, + true); + else if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config( + pipe_ctx->stream_res.stream_enc, + OPTC_DSC_DISABLED, 0, 0); + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.stream_enc, false, NULL, true); + } + } + + /* disable DSC block */ + pipe_ctx->stream_res.dsc->funcs->dsc_disable(pipe_ctx->stream_res.dsc); + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + odm_pipe->stream_res.dsc->funcs->dsc_disable(odm_pipe->stream_res.dsc); + } +} + +bool dp_set_dsc_enable(struct pipe_ctx *pipe_ctx, bool enable) +{ + struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; + bool result = false; + + if (!pipe_ctx->stream->timing.flags.DSC) + goto out; + if (!dsc) + goto out; + + if (enable) { + { + dp_set_dsc_on_stream(pipe_ctx, true); + result = true; + } + } else { + dp_set_dsc_on_rx(pipe_ctx, false); + dp_set_dsc_on_stream(pipe_ctx, false); + result = true; + } +out: + return result; +} + +/* + * For dynamic bpp change case, dsc is programmed with MASTER_UPDATE_LOCK enabled; + * hence PPS info packet update need to use frame update instead of immediate update. + * Added parameter immediate_update for this purpose. + * The decision to use frame update is hard-coded in function dp_update_dsc_config(), + * which is the only place where a "false" would be passed in for param immediate_update. + * + * immediate_update is only applicable when DSC is enabled. + */ +bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable, bool immediate_update) +{ + struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; + struct dc_stream_state *stream = pipe_ctx->stream; + + if (!pipe_ctx->stream->timing.flags.DSC || !dsc) + return false; + + if (enable) { + struct dsc_config dsc_cfg; + uint8_t dsc_packed_pps[128]; + + memset(&dsc_cfg, 0, sizeof(dsc_cfg)); + memset(dsc_packed_pps, 0, 128); + + /* Enable DSC hw block */ + dsc_cfg.pic_width = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; + dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; + dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; + dsc_cfg.color_depth = stream->timing.display_color_depth; + dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; + dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; + + DC_LOG_DSC(" "); + dsc->funcs->dsc_get_packed_pps(dsc, &dsc_cfg, &dsc_packed_pps[0]); + if (dc_is_dp_signal(stream->signal)) { + DC_LOG_DSC("Setting stream encoder DSC PPS SDP for engine %d\n", (int)pipe_ctx->stream_res.stream_enc->id); + if (is_dp_128b_132b_signal(pipe_ctx)) + pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.hpo_dp_stream_enc, + true, + &dsc_packed_pps[0], + immediate_update); + else + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.stream_enc, + true, + &dsc_packed_pps[0], + immediate_update); + } + } else { + /* disable DSC PPS in stream encoder */ + if (dc_is_dp_signal(stream->signal)) { + if (is_dp_128b_132b_signal(pipe_ctx)) + pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.hpo_dp_stream_enc, + false, + NULL, + true); + else + pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( + pipe_ctx->stream_res.stream_enc, false, NULL, true); + } + } + + return true; +} + + +bool dp_update_dsc_config(struct pipe_ctx *pipe_ctx) +{ + struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; + + if (!pipe_ctx->stream->timing.flags.DSC) + return false; + if (!dsc) + return false; + + dp_set_dsc_on_stream(pipe_ctx, true); + dp_set_dsc_pps_sdp(pipe_ctx, true, false); + return true; +} + +#undef DC_LOGGER +#define DC_LOGGER \ + link->ctx->logger diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dpcd.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dpcd.c index 7f25c11f4248..48a18766f002 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dpcd.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dpcd.c @@ -27,7 +27,7 @@ #include <dc_link.h> #include <inc/link_hwss.h> #include <inc/link_dpcd.h> -#include "drm/drm_dp_helper.h" +#include <drm/dp/drm_dp_helper.h> #include <dc_dp_types.h> #include "dm_helpers.h" diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c index a55944da8d53..047c626a4a34 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c @@ -122,6 +122,7 @@ static void remove_link_enc_assignment( stream->link_enc = NULL; state->res_ctx.link_enc_cfg_ctx.link_enc_assignments[i].eng_id = ENGINE_ID_UNKNOWN; state->res_ctx.link_enc_cfg_ctx.link_enc_assignments[i].stream = NULL; + dc_stream_release(stream); break; } } @@ -486,7 +487,8 @@ struct link_encoder *link_enc_cfg_get_next_avail_link_enc(struct dc *dc) } for (i = 0; i < dc->res_pool->res_cap->num_dig_link_enc; i++) { - if (encs_assigned[i] == ENGINE_ID_UNKNOWN) { + if (encs_assigned[i] == ENGINE_ID_UNKNOWN && + dc->res_pool->link_encoders[i] != NULL) { link_enc = dc->res_pool->link_encoders[i]; break; } @@ -506,6 +508,26 @@ struct link_encoder *link_enc_cfg_get_link_enc_used_by_stream( return link_enc; } +struct link_encoder *link_enc_cfg_get_link_enc( + const struct dc_link *link) +{ + struct link_encoder *link_enc = NULL; + + /* Links supporting dynamically assigned link encoder will be assigned next + * available encoder if one not already assigned. + */ + if (link->is_dig_mapping_flexible && + link->dc->res_pool->funcs->link_encs_assign) { + link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); + if (link_enc == NULL) + link_enc = link_enc_cfg_get_next_avail_link_enc( + link->ctx->dc); + } else + link_enc = link->link_enc; + + return link_enc; +} + bool link_enc_cfg_is_link_enc_avail(struct dc *dc, enum engine_id eng_id, struct dc_link *link) { bool is_avail = true; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c deleted file mode 100644 index 45d03d3a95c3..000000000000 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c +++ /dev/null @@ -1,917 +0,0 @@ -/* Copyright 2015 Advanced Micro Devices, Inc. */ - - -#include "dm_services.h" -#include "dc.h" -#include "inc/core_types.h" -#include "include/ddc_service_types.h" -#include "include/i2caux_interface.h" -#include "link_hwss.h" -#include "hw_sequencer.h" -#include "dc_link_dp.h" -#include "dc_link_ddc.h" -#include "dm_helpers.h" -#include "dpcd_defs.h" -#include "dsc.h" -#include "resource.h" -#include "link_enc_cfg.h" -#include "clk_mgr.h" -#include "inc/link_dpcd.h" -#include "dccg.h" - -static uint8_t convert_to_count(uint8_t lttpr_repeater_count) -{ - switch (lttpr_repeater_count) { - case 0x80: // 1 lttpr repeater - return 1; - case 0x40: // 2 lttpr repeaters - return 2; - case 0x20: // 3 lttpr repeaters - return 3; - case 0x10: // 4 lttpr repeaters - return 4; - case 0x08: // 5 lttpr repeaters - return 5; - case 0x04: // 6 lttpr repeaters - return 6; - case 0x02: // 7 lttpr repeaters - return 7; - case 0x01: // 8 lttpr repeaters - return 8; - default: - break; - } - return 0; // invalid value -} - -static inline bool is_immediate_downstream(struct dc_link *link, uint32_t offset) -{ - return (convert_to_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt) == offset); -} - -void dp_receiver_power_ctrl(struct dc_link *link, bool on) -{ - uint8_t state; - - state = on ? DP_POWER_STATE_D0 : DP_POWER_STATE_D3; - - if (link->sync_lt_in_progress) - return; - - core_link_write_dpcd(link, DP_SET_POWER, &state, - sizeof(state)); -} - -void dp_source_sequence_trace(struct dc_link *link, uint8_t dp_test_mode) -{ - if (link != NULL && link->dc->debug.enable_driver_sequence_debug) - core_link_write_dpcd(link, DP_SOURCE_SEQUENCE, - &dp_test_mode, sizeof(dp_test_mode)); -} - -void dp_enable_link_phy( - struct dc_link *link, - const struct link_resource *link_res, - enum signal_type signal, - enum clock_source_id clock_source, - const struct dc_link_settings *link_settings) -{ - struct link_encoder *link_enc; - struct dc *dc = link->ctx->dc; - struct dmcu *dmcu = dc->res_pool->dmcu; - - struct pipe_ctx *pipes = - link->dc->current_state->res_ctx.pipe_ctx; - struct clock_source *dp_cs = - link->dc->res_pool->dp_clock_source; - unsigned int i; - - /* Link should always be assigned encoder when en-/disabling. */ - if (link->is_dig_mapping_flexible && dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_link(dc, link); - else - link_enc = link->link_enc; - ASSERT(link_enc); - - if (link->connector_signal == SIGNAL_TYPE_EDP) { - link->dc->hwss.edp_power_control(link, true); - link->dc->hwss.edp_wait_for_hpd_ready(link, true); - } - - /* If the current pixel clock source is not DTO(happens after - * switching from HDMI passive dongle to DP on the same connector), - * switch the pixel clock source to DTO. - */ - for (i = 0; i < MAX_PIPES; i++) { - if (pipes[i].stream != NULL && - pipes[i].stream->link == link) { - if (pipes[i].clock_source != NULL && - pipes[i].clock_source->id != CLOCK_SOURCE_ID_DP_DTO) { - pipes[i].clock_source = dp_cs; - pipes[i].stream_res.pix_clk_params.requested_pix_clk_100hz = - pipes[i].stream->timing.pix_clk_100hz; - pipes[i].clock_source->funcs->program_pix_clk( - pipes[i].clock_source, - &pipes[i].stream_res.pix_clk_params, - &pipes[i].pll_settings); - } - } - } - - link->cur_link_settings = *link_settings; - -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dp_get_link_encoding_format(link_settings) == DP_128b_132b_ENCODING) { - /* TODO - DP2.0 HW: notify link rate change here */ - } else if (dp_get_link_encoding_format(link_settings) == DP_8b_10b_ENCODING) { - if (dc->clk_mgr->funcs->notify_link_rate_change) - dc->clk_mgr->funcs->notify_link_rate_change(dc->clk_mgr, link); - } -#else - if (dc->clk_mgr->funcs->notify_link_rate_change) - dc->clk_mgr->funcs->notify_link_rate_change(dc->clk_mgr, link); -#endif - if (dmcu != NULL && dmcu->funcs->lock_phy) - dmcu->funcs->lock_phy(dmcu); - -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dp_get_link_encoding_format(link_settings) == DP_128b_132b_ENCODING) { - enable_dp_hpo_output(link, link_res, link_settings); - } else if (dp_get_link_encoding_format(link_settings) == DP_8b_10b_ENCODING) { - if (dc_is_dp_sst_signal(signal)) { - link_enc->funcs->enable_dp_output( - link_enc, - link_settings, - clock_source); - } else { - link_enc->funcs->enable_dp_mst_output( - link_enc, - link_settings, - clock_source); - } - } -#else - if (dc_is_dp_sst_signal(signal)) { - link_enc->funcs->enable_dp_output( - link_enc, - link_settings, - clock_source); - } else { - link_enc->funcs->enable_dp_mst_output( - link_enc, - link_settings, - clock_source); - } -#endif - if (dmcu != NULL && dmcu->funcs->unlock_phy) - dmcu->funcs->unlock_phy(dmcu); - - dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_ENABLE_LINK_PHY); - dp_receiver_power_ctrl(link, true); -} - -void edp_add_delay_for_T9(struct dc_link *link) -{ - if (link->local_sink && - link->local_sink->edid_caps.panel_patch.extra_delay_backlight_off > 0) - udelay(link->local_sink->edid_caps.panel_patch.extra_delay_backlight_off * 1000); -} - -bool edp_receiver_ready_T9(struct dc_link *link) -{ - unsigned int tries = 0; - unsigned char sinkstatus = 0; - unsigned char edpRev = 0; - enum dc_status result; - - result = core_link_read_dpcd(link, DP_EDP_DPCD_REV, &edpRev, sizeof(edpRev)); - - /* start from eDP version 1.2, SINK_STAUS indicate the sink is ready.*/ - if (result == DC_OK && edpRev >= DP_EDP_12) { - do { - sinkstatus = 1; - result = core_link_read_dpcd(link, DP_SINK_STATUS, &sinkstatus, sizeof(sinkstatus)); - if (sinkstatus == 0) - break; - if (result != DC_OK) - break; - udelay(100); //MAx T9 - } while (++tries < 50); - } - - return result; -} -bool edp_receiver_ready_T7(struct dc_link *link) -{ - unsigned char sinkstatus = 0; - unsigned char edpRev = 0; - enum dc_status result; - - /* use absolute time stamp to constrain max T7*/ - unsigned long long enter_timestamp = 0; - unsigned long long finish_timestamp = 0; - unsigned long long time_taken_in_ns = 0; - - result = core_link_read_dpcd(link, DP_EDP_DPCD_REV, &edpRev, sizeof(edpRev)); - - if (result == DC_OK && edpRev >= DP_EDP_12) { - /* start from eDP version 1.2, SINK_STAUS indicate the sink is ready.*/ - enter_timestamp = dm_get_timestamp(link->ctx); - do { - sinkstatus = 0; - result = core_link_read_dpcd(link, DP_SINK_STATUS, &sinkstatus, sizeof(sinkstatus)); - if (sinkstatus == 1) - break; - if (result != DC_OK) - break; - udelay(25); - finish_timestamp = dm_get_timestamp(link->ctx); - time_taken_in_ns = dm_get_elapse_time_in_ns(link->ctx, finish_timestamp, enter_timestamp); - } while (time_taken_in_ns < 50 * 1000000); //MAx T7 is 50ms - } - - if (link->local_sink && - link->local_sink->edid_caps.panel_patch.extra_t7_ms > 0) - udelay(link->local_sink->edid_caps.panel_patch.extra_t7_ms * 1000); - - return result; -} - -void dp_disable_link_phy(struct dc_link *link, const struct link_resource *link_res, - enum signal_type signal) -{ - struct dc *dc = link->ctx->dc; - struct dmcu *dmcu = dc->res_pool->dmcu; -#if defined(CONFIG_DRM_AMD_DC_DCN) - struct hpo_dp_link_encoder *hpo_link_enc = link_res->hpo_dp_link_enc; -#endif - struct link_encoder *link_enc; - - /* Link should always be assigned encoder when en-/disabling. */ - if (link->is_dig_mapping_flexible && dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_link(dc, link); - else - link_enc = link->link_enc; - ASSERT(link_enc); - - if (!link->wa_flags.dp_keep_receiver_powered) - dp_receiver_power_ctrl(link, false); - - if (signal == SIGNAL_TYPE_EDP) { - if (link->dc->hwss.edp_backlight_control) - link->dc->hwss.edp_backlight_control(link, false); -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dp_get_link_encoding_format(&link->cur_link_settings) == DP_128b_132b_ENCODING) - disable_dp_hpo_output(link, link_res, signal); - else - link_enc->funcs->disable_output(link_enc, signal); -#else - link_enc->funcs->disable_output(link_enc, signal); -#endif - link->dc->hwss.edp_power_control(link, false); - } else { - if (dmcu != NULL && dmcu->funcs->lock_phy) - dmcu->funcs->lock_phy(dmcu); - -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dp_get_link_encoding_format(&link->cur_link_settings) == DP_128b_132b_ENCODING && - hpo_link_enc) - disable_dp_hpo_output(link, link_res, signal); - else - link_enc->funcs->disable_output(link_enc, signal); -#else - link_enc->funcs->disable_output(link_enc, signal); -#endif - if (dmcu != NULL && dmcu->funcs->unlock_phy) - dmcu->funcs->unlock_phy(dmcu); - } - - dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY); - - /* Clear current link setting.*/ - memset(&link->cur_link_settings, 0, - sizeof(link->cur_link_settings)); - - if (dc->clk_mgr->funcs->notify_link_rate_change) - dc->clk_mgr->funcs->notify_link_rate_change(dc->clk_mgr, link); -} - -void dp_disable_link_phy_mst(struct dc_link *link, const struct link_resource *link_res, - enum signal_type signal) -{ - /* MST disable link only when no stream use the link */ - if (link->mst_stream_alloc_table.stream_count > 0) - return; - - dp_disable_link_phy(link, link_res, signal); - - /* set the sink to SST mode after disabling the link */ - dp_enable_mst_on_sink(link, false); -} - -bool dp_set_hw_training_pattern( - struct dc_link *link, - const struct link_resource *link_res, - enum dc_dp_training_pattern pattern, - uint32_t offset) -{ - enum dp_test_pattern test_pattern = DP_TEST_PATTERN_UNSUPPORTED; - - switch (pattern) { - case DP_TRAINING_PATTERN_SEQUENCE_1: - test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN1; - break; - case DP_TRAINING_PATTERN_SEQUENCE_2: - test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN2; - break; - case DP_TRAINING_PATTERN_SEQUENCE_3: - test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN3; - break; - case DP_TRAINING_PATTERN_SEQUENCE_4: - test_pattern = DP_TEST_PATTERN_TRAINING_PATTERN4; - break; -#if defined(CONFIG_DRM_AMD_DC_DCN) - case DP_128b_132b_TPS1: - test_pattern = DP_TEST_PATTERN_128b_132b_TPS1_TRAINING_MODE; - break; - case DP_128b_132b_TPS2: - test_pattern = DP_TEST_PATTERN_128b_132b_TPS2_TRAINING_MODE; - break; -#endif - default: - break; - } - - dp_set_hw_test_pattern(link, link_res, test_pattern, NULL, 0); - - return true; -} - -#if defined(CONFIG_DRM_AMD_DC_DCN) -#define DC_LOGGER \ - link->ctx->logger -#endif -void dp_set_hw_lane_settings( - struct dc_link *link, - const struct link_resource *link_res, - const struct link_training_settings *link_settings, - uint32_t offset) -{ - struct link_encoder *encoder = link->link_enc; - - if ((link->lttpr_mode == LTTPR_MODE_NON_TRANSPARENT) && !is_immediate_downstream(link, offset)) - return; - - /* call Encoder to set lane settings */ -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dp_get_link_encoding_format(&link_settings->link_settings) == - DP_128b_132b_ENCODING) { - link_res->hpo_dp_link_enc->funcs->set_ffe( - link_res->hpo_dp_link_enc, - &link_settings->link_settings, - link_settings->lane_settings[0].FFE_PRESET.raw); - } else if (dp_get_link_encoding_format(&link_settings->link_settings) - == DP_8b_10b_ENCODING) { - encoder->funcs->dp_set_lane_settings(encoder, link_settings); - } -#else - encoder->funcs->dp_set_lane_settings(encoder, link_settings); -#endif - memmove(link->cur_lane_setting, - link_settings->lane_settings, - sizeof(link->cur_lane_setting)); -} - -void dp_set_hw_test_pattern( - struct dc_link *link, - const struct link_resource *link_res, - enum dp_test_pattern test_pattern, - uint8_t *custom_pattern, - uint32_t custom_pattern_size) -{ - struct encoder_set_dp_phy_pattern_param pattern_param = {0}; - struct link_encoder *encoder; -#if defined(CONFIG_DRM_AMD_DC_DCN) - enum dp_link_encoding link_encoding_format = dp_get_link_encoding_format(&link->cur_link_settings); -#endif - - /* Access link encoder based on whether it is statically - * or dynamically assigned to a link. - */ - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) - encoder = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - else - encoder = link->link_enc; - - pattern_param.dp_phy_pattern = test_pattern; - pattern_param.custom_pattern = custom_pattern; - pattern_param.custom_pattern_size = custom_pattern_size; - pattern_param.dp_panel_mode = dp_get_panel_mode(link); - -#if defined(CONFIG_DRM_AMD_DC_DCN) - switch (link_encoding_format) { - case DP_128b_132b_ENCODING: - link_res->hpo_dp_link_enc->funcs->set_link_test_pattern( - link_res->hpo_dp_link_enc, &pattern_param); - break; - case DP_8b_10b_ENCODING: - ASSERT(encoder); - encoder->funcs->dp_set_phy_pattern(encoder, &pattern_param); - break; - default: - DC_LOG_ERROR("%s: Unknown link encoding format.", __func__); - break; - } -#else - encoder->funcs->dp_set_phy_pattern(encoder, &pattern_param); -#endif - dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_SET_SOURCE_PATTERN); -} -#if defined(CONFIG_DRM_AMD_DC_DCN) -#undef DC_LOGGER -#endif - -void dp_retrain_link_dp_test(struct dc_link *link, - struct dc_link_settings *link_setting, - bool skip_video_pattern) -{ - struct pipe_ctx *pipes = - &link->dc->current_state->res_ctx.pipe_ctx[0]; - unsigned int i; - - for (i = 0; i < MAX_PIPES; i++) { - if (pipes[i].stream != NULL && - !pipes[i].top_pipe && !pipes[i].prev_odm_pipe && - pipes[i].stream->link != NULL && - pipes[i].stream_res.stream_enc != NULL && - pipes[i].stream->link == link) { - udelay(100); - - pipes[i].stream_res.stream_enc->funcs->dp_blank(link, - pipes[i].stream_res.stream_enc); - - /* disable any test pattern that might be active */ - dp_set_hw_test_pattern(link, &pipes[i].link_res, - DP_TEST_PATTERN_VIDEO_MODE, NULL, 0); - - dp_receiver_power_ctrl(link, false); - - link->dc->hwss.disable_stream(&pipes[i]); - if ((&pipes[i])->stream_res.audio && !link->dc->debug.az_endpoint_mute_only) - (&pipes[i])->stream_res.audio->funcs->az_disable((&pipes[i])->stream_res.audio); - - if (link->link_enc) - link->link_enc->funcs->disable_output( - link->link_enc, - SIGNAL_TYPE_DISPLAY_PORT); - - /* Clear current link setting. */ - memset(&link->cur_link_settings, 0, - sizeof(link->cur_link_settings)); - - perform_link_training_with_retries( - link_setting, - skip_video_pattern, - LINK_TRAINING_ATTEMPTS, - &pipes[i], - SIGNAL_TYPE_DISPLAY_PORT, - false); - - link->dc->hwss.enable_stream(&pipes[i]); - - link->dc->hwss.unblank_stream(&pipes[i], - link_setting); - - if (pipes[i].stream_res.audio) { - /* notify audio driver for - * audio modes of monitor */ - pipes[i].stream_res.audio->funcs->az_enable( - pipes[i].stream_res.audio); - - /* un-mute audio */ - /* TODO: audio should be per stream rather than - * per link */ - pipes[i].stream_res.stream_enc->funcs-> - audio_mute_control( - pipes[i].stream_res.stream_enc, false); - } - } - } -} - -#define DC_LOGGER \ - dsc->ctx->logger -static void dsc_optc_config_log(struct display_stream_compressor *dsc, - struct dsc_optc_config *config) -{ - uint32_t precision = 1 << 28; - uint32_t bytes_per_pixel_int = config->bytes_per_pixel / precision; - uint32_t bytes_per_pixel_mod = config->bytes_per_pixel % precision; - uint64_t ll_bytes_per_pix_fraq = bytes_per_pixel_mod; - - /* 7 fractional digits decimal precision for bytes per pixel is enough because DSC - * bits per pixel precision is 1/16th of a pixel, which means bytes per pixel precision is - * 1/16/8 = 1/128 of a byte, or 0.0078125 decimal - */ - ll_bytes_per_pix_fraq *= 10000000; - ll_bytes_per_pix_fraq /= precision; - - DC_LOG_DSC("\tbytes_per_pixel 0x%08x (%d.%07d)", - config->bytes_per_pixel, bytes_per_pixel_int, (uint32_t)ll_bytes_per_pix_fraq); - DC_LOG_DSC("\tis_pixel_format_444 %d", config->is_pixel_format_444); - DC_LOG_DSC("\tslice_width %d", config->slice_width); -} - -bool dp_set_dsc_on_rx(struct pipe_ctx *pipe_ctx, bool enable) -{ - struct dc *dc = pipe_ctx->stream->ctx->dc; - struct dc_stream_state *stream = pipe_ctx->stream; - bool result = false; - - if (dc_is_virtual_signal(stream->signal) || IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) - result = true; - else - result = dm_helpers_dp_write_dsc_enable(dc->ctx, stream, enable); - return result; -} - -/* The stream with these settings can be sent (unblanked) only after DSC was enabled on RX first, - * i.e. after dp_enable_dsc_on_rx() had been called - */ -void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) -{ - struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; - struct dc *dc = pipe_ctx->stream->ctx->dc; - struct dc_stream_state *stream = pipe_ctx->stream; - struct pipe_ctx *odm_pipe; - int opp_cnt = 1; - - for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) - opp_cnt++; - - if (enable) { - struct dsc_config dsc_cfg; - struct dsc_optc_config dsc_optc_cfg; - enum optc_dsc_mode optc_dsc_mode; - - /* Enable DSC hw block */ - dsc_cfg.pic_width = (stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right) / opp_cnt; - dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; - dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; - dsc_cfg.color_depth = stream->timing.display_color_depth; - dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; - dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; - ASSERT(dsc_cfg.dc_dsc_cfg.num_slices_h % opp_cnt == 0); - dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; - - dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg); - dsc->funcs->dsc_enable(dsc, pipe_ctx->stream_res.opp->inst); - for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { - struct display_stream_compressor *odm_dsc = odm_pipe->stream_res.dsc; - - odm_dsc->funcs->dsc_set_config(odm_dsc, &dsc_cfg, &dsc_optc_cfg); - odm_dsc->funcs->dsc_enable(odm_dsc, odm_pipe->stream_res.opp->inst); - } - dsc_cfg.dc_dsc_cfg.num_slices_h *= opp_cnt; - dsc_cfg.pic_width *= opp_cnt; - - optc_dsc_mode = dsc_optc_cfg.is_pixel_format_444 ? OPTC_DSC_ENABLED_444 : OPTC_DSC_ENABLED_NATIVE_SUBSAMPLED; - - /* Enable DSC in encoder */ -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment) - && !is_dp_128b_132b_signal(pipe_ctx)) { -#else - if (dc_is_dp_signal(stream->signal) && !IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { -#endif - DC_LOG_DSC("Setting stream encoder DSC config for engine %d:", (int)pipe_ctx->stream_res.stream_enc->id); - dsc_optc_config_log(dsc, &dsc_optc_cfg); - pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config(pipe_ctx->stream_res.stream_enc, - optc_dsc_mode, - dsc_optc_cfg.bytes_per_pixel, - dsc_optc_cfg.slice_width); - - /* PPS SDP is set elsewhere because it has to be done after DIG FE is connected to DIG BE */ - } - - /* Enable DSC in OPTC */ - DC_LOG_DSC("Setting optc DSC config for tg instance %d:", pipe_ctx->stream_res.tg->inst); - dsc_optc_config_log(dsc, &dsc_optc_cfg); - pipe_ctx->stream_res.tg->funcs->set_dsc_config(pipe_ctx->stream_res.tg, - optc_dsc_mode, - dsc_optc_cfg.bytes_per_pixel, - dsc_optc_cfg.slice_width); - } else { - /* disable DSC in OPTC */ - pipe_ctx->stream_res.tg->funcs->set_dsc_config( - pipe_ctx->stream_res.tg, - OPTC_DSC_DISABLED, 0, 0); - - /* disable DSC in stream encoder */ - if (dc_is_dp_signal(stream->signal)) { - -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (is_dp_128b_132b_signal(pipe_ctx)) - pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_set_dsc_pps_info_packet( - pipe_ctx->stream_res.hpo_dp_stream_enc, - false, - NULL, - true); - else -#endif - if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_config( - pipe_ctx->stream_res.stream_enc, - OPTC_DSC_DISABLED, 0, 0); - pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( - pipe_ctx->stream_res.stream_enc, false, NULL, true); - } - } - - /* disable DSC block */ - pipe_ctx->stream_res.dsc->funcs->dsc_disable(pipe_ctx->stream_res.dsc); - for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) - odm_pipe->stream_res.dsc->funcs->dsc_disable(odm_pipe->stream_res.dsc); - } -} - -bool dp_set_dsc_enable(struct pipe_ctx *pipe_ctx, bool enable) -{ - struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; - bool result = false; - - if (!pipe_ctx->stream->timing.flags.DSC) - goto out; - if (!dsc) - goto out; - - if (enable) { - { - dp_set_dsc_on_stream(pipe_ctx, true); - result = true; - } - } else { - dp_set_dsc_on_rx(pipe_ctx, false); - dp_set_dsc_on_stream(pipe_ctx, false); - result = true; - } -out: - return result; -} - -/* - * For dynamic bpp change case, dsc is programmed with MASTER_UPDATE_LOCK enabled; - * hence PPS info packet update need to use frame update instead of immediate update. - * Added parameter immediate_update for this purpose. - * The decision to use frame update is hard-coded in function dp_update_dsc_config(), - * which is the only place where a "false" would be passed in for param immediate_update. - * - * immediate_update is only applicable when DSC is enabled. - */ -bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable, bool immediate_update) -{ - struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; - struct dc_stream_state *stream = pipe_ctx->stream; - - if (!pipe_ctx->stream->timing.flags.DSC || !dsc) - return false; - - if (enable) { - struct dsc_config dsc_cfg; - uint8_t dsc_packed_pps[128]; - - memset(&dsc_cfg, 0, sizeof(dsc_cfg)); - memset(dsc_packed_pps, 0, 128); - - /* Enable DSC hw block */ - dsc_cfg.pic_width = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; - dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom; - dsc_cfg.pixel_encoding = stream->timing.pixel_encoding; - dsc_cfg.color_depth = stream->timing.display_color_depth; - dsc_cfg.is_odm = pipe_ctx->next_odm_pipe ? true : false; - dsc_cfg.dc_dsc_cfg = stream->timing.dsc_cfg; - - DC_LOG_DSC(" "); - dsc->funcs->dsc_get_packed_pps(dsc, &dsc_cfg, &dsc_packed_pps[0]); - if (dc_is_dp_signal(stream->signal)) { - DC_LOG_DSC("Setting stream encoder DSC PPS SDP for engine %d\n", (int)pipe_ctx->stream_res.stream_enc->id); -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (is_dp_128b_132b_signal(pipe_ctx)) - pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_set_dsc_pps_info_packet( - pipe_ctx->stream_res.hpo_dp_stream_enc, - true, - &dsc_packed_pps[0], - immediate_update); - else -#endif - pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( - pipe_ctx->stream_res.stream_enc, - true, - &dsc_packed_pps[0], - immediate_update); - } - } else { - /* disable DSC PPS in stream encoder */ - if (dc_is_dp_signal(stream->signal)) { -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (is_dp_128b_132b_signal(pipe_ctx)) - pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_set_dsc_pps_info_packet( - pipe_ctx->stream_res.hpo_dp_stream_enc, - false, - NULL, - true); - else -#endif - pipe_ctx->stream_res.stream_enc->funcs->dp_set_dsc_pps_info_packet( - pipe_ctx->stream_res.stream_enc, false, NULL, true); - } - } - - return true; -} - - -bool dp_update_dsc_config(struct pipe_ctx *pipe_ctx) -{ - struct display_stream_compressor *dsc = pipe_ctx->stream_res.dsc; - - if (!pipe_ctx->stream->timing.flags.DSC) - return false; - if (!dsc) - return false; - - dp_set_dsc_on_stream(pipe_ctx, true); - dp_set_dsc_pps_sdp(pipe_ctx, true, false); - return true; -} - -#if defined(CONFIG_DRM_AMD_DC_DCN) -#undef DC_LOGGER -#define DC_LOGGER \ - link->ctx->logger - -static enum phyd32clk_clock_source get_phyd32clk_src(struct dc_link *link) -{ - switch (link->link_enc->transmitter) { - case TRANSMITTER_UNIPHY_A: - return PHYD32CLKA; - case TRANSMITTER_UNIPHY_B: - return PHYD32CLKB; - case TRANSMITTER_UNIPHY_C: - return PHYD32CLKC; - case TRANSMITTER_UNIPHY_D: - return PHYD32CLKD; - case TRANSMITTER_UNIPHY_E: - return PHYD32CLKE; - default: - return PHYD32CLKA; - } -} - -void enable_dp_hpo_output(struct dc_link *link, - const struct link_resource *link_res, - const struct dc_link_settings *link_settings) -{ - const struct dc *dc = link->dc; - enum phyd32clk_clock_source phyd32clk; - - /* Enable PHY PLL at target bit rate - * UHBR10 = 10Gbps (SYMCLK32 = 312.5MHz) - * UBR13.5 = 13.5Gbps (SYMCLK32 = 421.875MHz) - * UHBR20 = 20Gbps (SYMCLK32 = 625MHz) - */ - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - switch (link_settings->link_rate) { - case LINK_RATE_UHBR10: - dm_set_phyd32clk(dc->ctx, 312500); - break; - case LINK_RATE_UHBR13_5: - dm_set_phyd32clk(dc->ctx, 412875); - break; - case LINK_RATE_UHBR20: - dm_set_phyd32clk(dc->ctx, 625000); - break; - default: - return; - } - } else { - /* DP2.0 HW: call transmitter control to enable PHY */ - link_res->hpo_dp_link_enc->funcs->enable_link_phy( - link_res->hpo_dp_link_enc, - link_settings, - link->link_enc->transmitter, - link->link_enc->hpd_source); - } - - /* DCCG muxing and DTBCLK DTO */ - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - dc->res_pool->dccg->funcs->set_physymclk( - dc->res_pool->dccg, - link->link_enc_hw_inst, - PHYSYMCLK_FORCE_SRC_PHYD32CLK, - true); - - phyd32clk = get_phyd32clk_src(link); - dc->res_pool->dccg->funcs->enable_symclk32_le( - dc->res_pool->dccg, - link_res->hpo_dp_link_enc->inst, - phyd32clk); - link_res->hpo_dp_link_enc->funcs->link_enable( - link_res->hpo_dp_link_enc, - link_settings->lane_count); - } -} - -void disable_dp_hpo_output(struct dc_link *link, - const struct link_resource *link_res, - enum signal_type signal) -{ - const struct dc *dc = link->dc; - - link_res->hpo_dp_link_enc->funcs->link_disable(link_res->hpo_dp_link_enc); - - if (IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) { - dc->res_pool->dccg->funcs->disable_symclk32_le( - dc->res_pool->dccg, - link_res->hpo_dp_link_enc->inst); - - dc->res_pool->dccg->funcs->set_physymclk( - dc->res_pool->dccg, - link->link_enc_hw_inst, - PHYSYMCLK_FORCE_SRC_SYMCLK, - false); - - dm_set_phyd32clk(dc->ctx, 0); - } else { - /* DP2.0 HW: call transmitter control to disable PHY */ - link_res->hpo_dp_link_enc->funcs->disable_link_phy( - link_res->hpo_dp_link_enc, - signal); - } -} - -void setup_dp_hpo_stream(struct pipe_ctx *pipe_ctx, bool enable) -{ - struct dc_stream_state *stream = pipe_ctx->stream; - struct dc *dc = pipe_ctx->stream->ctx->dc; - struct pipe_ctx *odm_pipe; - int odm_combine_num_segments = 1; - enum phyd32clk_clock_source phyd32clk; - - if (enable) { - for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) - odm_combine_num_segments++; - - dc->res_pool->dccg->funcs->set_dpstreamclk( - dc->res_pool->dccg, - DTBCLK0, - pipe_ctx->stream_res.tg->inst); - - phyd32clk = get_phyd32clk_src(stream->link); - dc->res_pool->dccg->funcs->enable_symclk32_se( - dc->res_pool->dccg, - pipe_ctx->stream_res.hpo_dp_stream_enc->inst, - phyd32clk); - - dc->res_pool->dccg->funcs->set_dtbclk_dto( - dc->res_pool->dccg, - pipe_ctx->stream_res.tg->inst, - stream->phy_pix_clk, - odm_combine_num_segments, - &stream->timing); - } else { - dc->res_pool->dccg->funcs->set_dtbclk_dto( - dc->res_pool->dccg, - pipe_ctx->stream_res.tg->inst, - 0, - 0, - &stream->timing); - dc->res_pool->dccg->funcs->disable_symclk32_se( - dc->res_pool->dccg, - pipe_ctx->stream_res.hpo_dp_stream_enc->inst); - dc->res_pool->dccg->funcs->set_dpstreamclk( - dc->res_pool->dccg, - REFCLK, - pipe_ctx->stream_res.tg->inst); - } -} - -void reset_dp_hpo_stream_encoders_for_link(struct dc_link *link) -{ - const struct dc *dc = link->dc; - struct dc_state *state = dc->current_state; - uint8_t i; - - for (i = 0; i < MAX_PIPES; i++) { - if (state->res_ctx.pipe_ctx[i].stream_res.hpo_dp_stream_enc && - state->res_ctx.pipe_ctx[i].stream && - state->res_ctx.pipe_ctx[i].stream->link == link && - !state->res_ctx.pipe_ctx[i].stream->dpms_off) { - setup_dp_hpo_stream(&state->res_ctx.pipe_ctx[i], false); - } - } -} - -#undef DC_LOGGER -#endif diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index b3912ff9dc91..71b393194c55 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -43,6 +43,10 @@ #include "dpcd_defs.h" #include "link_enc_cfg.h" #include "dc_link_dp.h" +#include "virtual/virtual_link_hwss.h" +#include "link/link_hwss_dio.h" +#include "link/link_hwss_dpia.h" +#include "link/link_hwss_hpo_dp.h" #if defined(CONFIG_DRM_AMD_DC_SI) #include "dce60/dce60_resource.h" @@ -131,7 +135,7 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id) case FAMILY_NV: dc_version = DCN_VERSION_2_0; - if (asic_id.chip_id == DEVICE_ID_NV_13FE) { + if (asic_id.chip_id == DEVICE_ID_NV_13FE || asic_id.chip_id == DEVICE_ID_NV_143F) { dc_version = DCN_VERSION_2_01; break; } @@ -356,7 +360,6 @@ bool resource_construct( } } -#if defined(CONFIG_DRM_AMD_DC_DCN) pool->hpo_dp_stream_enc_count = 0; if (create_funcs->create_hpo_dp_stream_encoder) { for (i = 0; i < caps->num_hpo_dp_stream_encoder; i++) { @@ -377,7 +380,6 @@ bool resource_construct( pool->hpo_dp_link_enc_count++; } } -#endif #if defined(CONFIG_DRM_AMD_DC_DCN) for (i = 0; i < caps->num_mpc_3dlut; i++) { @@ -1640,6 +1642,9 @@ static bool are_stream_backends_same( if (is_timing_changed(stream_a, stream_b)) return false; + if (stream_a->signal != stream_b->signal) + return false; + if (stream_a->dpms_off != stream_b->dpms_off) return false; @@ -1710,7 +1715,6 @@ static void update_stream_engine_usage( } } -#if defined(CONFIG_DRM_AMD_DC_DCN) static void update_hpo_dp_stream_engine_usage( struct resource_context *res_ctx, const struct resource_pool *pool, @@ -1812,7 +1816,6 @@ static void remove_hpo_dp_link_enc_from_ctx(struct resource_context *res_ctx, pipe_ctx->link_res.hpo_dp_link_enc = NULL; } } -#endif /* TODO: release audio object */ void update_audio_usage( @@ -1858,7 +1861,6 @@ static int acquire_first_free_pipe( return -1; } -#if defined(CONFIG_DRM_AMD_DC_DCN) static struct hpo_dp_stream_encoder *find_first_free_match_hpo_dp_stream_enc_for_link( struct resource_context *res_ctx, const struct resource_pool *pool, @@ -1876,7 +1878,6 @@ static struct hpo_dp_stream_encoder *find_first_free_match_hpo_dp_stream_enc_for return NULL; } -#endif static struct audio *find_first_free_audio( struct resource_context *res_ctx, @@ -1964,11 +1965,6 @@ enum dc_status dc_remove_stream_from_ctx( dc->res_pool, del_pipe->stream_res.stream_enc, false); - /* Release link encoder from stream in new dc_state. */ - if (dc->res_pool->funcs->link_enc_unassign) - dc->res_pool->funcs->link_enc_unassign(new_ctx, del_pipe->stream); - -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(del_pipe)) { update_hpo_dp_stream_engine_usage( &new_ctx->res_ctx, dc->res_pool, @@ -1976,7 +1972,6 @@ enum dc_status dc_remove_stream_from_ctx( false); remove_hpo_dp_link_enc_from_ctx(&new_ctx->res_ctx, del_pipe, del_pipe->stream); } -#endif if (del_pipe->stream_res.audio) update_audio_usage( @@ -2173,7 +2168,7 @@ static void mark_seamless_boot_stream( if (dc->config.allow_seamless_boot_optimization && !dcb->funcs->is_accelerated_mode(dcb)) { - if (dc_validate_seamless_boot_timing(dc, stream->sink, &stream->timing)) + if (dc_validate_boot_timing(dc, stream->sink, &stream->timing)) stream->apply_seamless_boot_optimization = true; } } @@ -2229,7 +2224,6 @@ enum dc_status resource_map_pool_resources( pipe_ctx->stream_res.stream_enc, true); -#if defined(CONFIG_DRM_AMD_DC_DCN) /* Allocate DP HPO Stream Encoder based on signal, hw capabilities * and link settings */ @@ -2254,7 +2248,6 @@ enum dc_status resource_map_pool_resources( return DC_NO_LINK_ENC_RESOURCE; } } -#endif /* TODO: Add check if ASIC support and EDID audio */ if (!stream->converter_disable_audio && @@ -2924,12 +2917,10 @@ bool pipe_need_reprogram( if (pipe_ctx_old->stream_res.dsc != pipe_ctx->stream_res.dsc) return true; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (pipe_ctx_old->stream_res.hpo_dp_stream_enc != pipe_ctx->stream_res.hpo_dp_stream_enc) return true; if (pipe_ctx_old->link_res.hpo_dp_link_enc != pipe_ctx->link_res.hpo_dp_link_enc) return true; -#endif /* DIG link encoder resource assignment for stream changed. */ if (pipe_ctx_old->stream->ctx->dc->res_pool->funcs->link_encs_assign) { @@ -3196,10 +3187,9 @@ void get_audio_check(struct audio_info *aud_modes, } } -#if defined(CONFIG_DRM_AMD_DC_DCN) -struct hpo_dp_link_encoder *resource_get_hpo_dp_link_enc_for_det_lt( +static struct hpo_dp_link_encoder *get_temp_hpo_dp_link_enc( const struct resource_context *res_ctx, - const struct resource_pool *pool, + const struct resource_pool *const pool, const struct dc_link *link) { struct hpo_dp_link_encoder *hpo_dp_link_enc = NULL; @@ -3215,7 +3205,24 @@ struct hpo_dp_link_encoder *resource_get_hpo_dp_link_enc_for_det_lt( return hpo_dp_link_enc; } -#endif + +bool get_temp_dp_link_res(struct dc_link *link, + struct link_resource *link_res, + struct dc_link_settings *link_settings) +{ + const struct dc *dc = link->dc; + const struct resource_context *res_ctx = &dc->current_state->res_ctx; + + memset(link_res, 0, sizeof(*link_res)); + + if (dp_get_link_encoding_format(link_settings) == DP_128b_132b_ENCODING) { + link_res->hpo_dp_link_enc = get_temp_hpo_dp_link_enc(res_ctx, + dc->res_pool, link); + if (!link_res->hpo_dp_link_enc) + return false; + } + return true; +} void reset_syncd_pipes_from_disabled_pipes(struct dc *dc, struct dc_state *context) @@ -3303,3 +3310,36 @@ uint8_t resource_transmitter_to_phy_idx(const struct dc *dc, enum transmitter tr #endif return phy_idx; } + +const struct link_hwss *get_link_hwss(const struct dc_link *link, + const struct link_resource *link_res) +{ + /* Link_hwss is only accessible by getter function instead of accessing + * by pointers in dc with the intent to protect against breaking polymorphism. + */ + if (can_use_hpo_dp_link_hwss(link, link_res)) + /* TODO: some assumes that if decided link settings is 128b/132b + * channel coding format hpo_dp_link_enc should be used. + * Others believe that if hpo_dp_link_enc is available in link + * resource then hpo_dp_link_enc must be used. This bound between + * hpo_dp_link_enc != NULL and decided link settings is loosely coupled + * with a premise that both hpo_dp_link_enc pointer and decided link + * settings are determined based on single policy function like + * "decide_link_settings" from upper layer. This "convention" + * cannot be maintained and enforced at current level. + * Therefore a refactor is due so we can enforce a strong bound + * between those two parameters at this level. + * + * To put it simple, we want to make enforcement at low level so that + * we will not return link hwss if caller plans to do 8b/10b + * with an hpo encoder. Or we can return a very dummy one that doesn't + * do work for all functions + */ + return get_hpo_dp_link_hwss(); + else if (can_use_dpia_link_hwss(link, link_res)) + return get_dpia_link_hwss(); + else if (can_use_dio_link_hwss(link, link_res)) + return get_dio_link_hwss(); + else + return get_virtual_link_hwss(); +} diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index b51864890621..fcb021bf0c96 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -47,7 +47,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.167" +#define DC_VER "3.2.172" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -195,10 +195,8 @@ struct dc_caps { unsigned int cursor_cache_size; struct dc_plane_cap planes[MAX_PLANES]; struct dc_color_caps color; -#if defined(CONFIG_DRM_AMD_DC_DCN) bool dp_hpo; bool hdmi_frl_pcon_support; -#endif bool edp_dsc_support; bool vbios_lttpr_aware; bool vbios_lttpr_enable; @@ -307,7 +305,6 @@ struct dc_cap_funcs { struct link_training_settings; -#if defined(CONFIG_DRM_AMD_DC_DCN) union allow_lttpr_non_transparent_mode { struct { bool DP1_4A : 1; @@ -315,7 +312,7 @@ union allow_lttpr_non_transparent_mode { } bits; unsigned char raw; }; -#endif + /* Structure to hold configuration flags set by dm at dc creation. */ struct dc_config { bool gpu_vm_support; @@ -323,16 +320,12 @@ struct dc_config { bool fbc_support; bool disable_fractional_pwm; bool allow_seamless_boot_optimization; - bool power_down_display_on_boot; + bool seamless_boot_edp_requested; bool edp_not_connected; bool edp_no_power_sequencing; bool force_enum_edp; bool forced_clocks; -#if defined(CONFIG_DRM_AMD_DC_DCN) union allow_lttpr_non_transparent_mode allow_lttpr_non_transparent_mode; -#else - bool allow_lttpr_non_transparent_mode; -#endif bool multi_mon_pp_mclk_switch; bool disable_dmcu; bool enable_4to1MPC; @@ -397,6 +390,7 @@ enum dcn_pwr_state { enum dcn_zstate_support_state { DCN_ZSTATE_SUPPORT_UNKNOWN, DCN_ZSTATE_SUPPORT_ALLOW, + DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY, DCN_ZSTATE_SUPPORT_DISALLOW, }; #endif @@ -519,12 +513,13 @@ union root_clock_optimization_options { union dpia_debug_options { struct { - uint32_t disable_dpia:1; - uint32_t force_non_lttpr:1; - uint32_t extend_aux_rd_interval:1; - uint32_t disable_mst_dsc_work_around:1; - uint32_t hpd_delay_in_ms:12; - uint32_t reserved:16; + uint32_t disable_dpia:1; /* bit 0 */ + uint32_t force_non_lttpr:1; /* bit 1 */ + uint32_t extend_aux_rd_interval:1; /* bit 2 */ + uint32_t disable_mst_dsc_work_around:1; /* bit 3 */ + uint32_t hpd_delay_in_ms:12; /* bits 4-15 */ + uint32_t disable_force_tbt3_work_around:1; /* bit 16 */ + uint32_t reserved:15; } bits; uint32_t raw; }; @@ -688,13 +683,12 @@ struct dc_debug_options { bool disable_dsc_edp; unsigned int force_dsc_edp_policy; bool enable_dram_clock_change_one_display_vactive; -#if defined(CONFIG_DRM_AMD_DC_DCN) /* TODO - remove once tested */ bool legacy_dp2_lt; bool set_mst_en_for_sst; bool disable_uhbr; bool force_dp2_lt_fallback_method; -#endif + bool ignore_cable_id; union mem_low_power_enable_options enable_mem_low_power; union root_clock_optimization_options root_clock_optimization; bool hpo_optimization; @@ -710,6 +704,7 @@ struct dc_debug_options { int crb_alloc_policy_min_disp_count; #if defined(CONFIG_DRM_AMD_DC_DCN) bool disable_z10; + bool enable_z9_disable_interface; bool enable_sw_cntl_psr; union dpia_debug_options dpia_debug; #endif @@ -1131,7 +1126,7 @@ struct dc_validation_set { uint8_t plane_count; }; -bool dc_validate_seamless_boot_timing(const struct dc *dc, +bool dc_validate_boot_timing(const struct dc *dc, const struct dc_sink *sink, struct dc_crtc_timing *crtc_timing); @@ -1212,6 +1207,7 @@ struct dpcd_caps { bool is_branch_dev; /* Dongle's downstream count. */ union sink_count sink_count; + bool is_mst_capable; /* If dongle_type == DISPLAY_DONGLE_DP_HDMI_CONVERTER, indicates 'Frame Sequential-to-lllFrame Pack' conversion capability.*/ struct dc_dongle_caps dongle_caps; @@ -1237,12 +1233,11 @@ struct dpcd_caps { struct psr_caps psr_caps; struct dpcd_usb4_dp_tunneling_info usb4_dp_tun_info; -#if defined(CONFIG_DRM_AMD_DC_DCN) union dp_128b_132b_supported_link_rates dp_128b_132b_supported_link_rates; union dp_main_line_channel_coding_cap channel_coding_cap; union dp_sink_video_fallback_formats fallback_formats; union dp_fec_capability1 fec_cap1; -#endif + union dp_cable_attributes cable_attributes; }; union dpcd_sink_ext_caps { diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index 353dac420f34..772084406795 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -53,7 +53,6 @@ enum dc_link_rate { LINK_RATE_RBR2 = 0x0C, // Rate_5 (RBR2)- 3.24 Gbps/Lane LINK_RATE_RATE_6 = 0x10, // Rate_6 - 4.32 Gbps/Lane LINK_RATE_HIGH2 = 0x14, // Rate_7 (HBR2)- 5.40 Gbps/Lane -#if defined(CONFIG_DRM_AMD_DC_DCN) LINK_RATE_HIGH3 = 0x1E, // Rate_8 (HBR3)- 8.10 Gbps/Lane /* Starting from DP2.0 link rate enum directly represents actual * link rate value in unit of 10 mbps @@ -61,9 +60,6 @@ enum dc_link_rate { LINK_RATE_UHBR10 = 1000, // UHBR10 - 10.0 Gbps/Lane LINK_RATE_UHBR13_5 = 1350, // UHBR13.5 - 13.5 Gbps/Lane LINK_RATE_UHBR20 = 2000, // UHBR10 - 20.0 Gbps/Lane -#else - LINK_RATE_HIGH3 = 0x1E // Rate_8 (HBR3)- 8.10 Gbps/Lane -#endif }; enum dc_link_spread { @@ -100,7 +96,6 @@ enum dc_post_cursor2 { POST_CURSOR2_MAX_LEVEL = POST_CURSOR2_LEVEL3, }; -#if defined(CONFIG_DRM_AMD_DC_DCN) enum dc_dp_ffe_preset_level { DP_FFE_PRESET_LEVEL0 = 0, DP_FFE_PRESET_LEVEL1, @@ -120,7 +115,6 @@ enum dc_dp_ffe_preset_level { DP_FFE_PRESET_LEVEL15, DP_FFE_PRESET_MAX_LEVEL = DP_FFE_PRESET_LEVEL15, }; -#endif enum dc_dp_training_pattern { DP_TRAINING_PATTERN_SEQUENCE_1 = 0, @@ -128,19 +122,15 @@ enum dc_dp_training_pattern { DP_TRAINING_PATTERN_SEQUENCE_3, DP_TRAINING_PATTERN_SEQUENCE_4, DP_TRAINING_PATTERN_VIDEOIDLE, -#if defined(CONFIG_DRM_AMD_DC_DCN) DP_128b_132b_TPS1, DP_128b_132b_TPS2, DP_128b_132b_TPS2_CDS, -#endif }; enum dp_link_encoding { DP_UNKNOWN_ENCODING = 0, DP_8b_10b_ENCODING = 1, -#if defined(CONFIG_DRM_AMD_DC_DCN) DP_128b_132b_ENCODING = 2, -#endif }; struct dc_link_settings { @@ -152,7 +142,6 @@ struct dc_link_settings { bool dpcd_source_device_specific_field_support; }; -#if defined(CONFIG_DRM_AMD_DC_DCN) union dc_dp_ffe_preset { struct { uint8_t level : 4; @@ -163,24 +152,19 @@ union dc_dp_ffe_preset { } settings; uint8_t raw; }; -#endif struct dc_lane_settings { enum dc_voltage_swing VOLTAGE_SWING; enum dc_pre_emphasis PRE_EMPHASIS; enum dc_post_cursor2 POST_CURSOR2; -#if defined(CONFIG_DRM_AMD_DC_DCN) union dc_dp_ffe_preset FFE_PRESET; -#endif }; struct dc_link_training_overrides { enum dc_voltage_swing *voltage_swing; enum dc_pre_emphasis *pre_emphasis; enum dc_post_cursor2 *post_cursor2; -#if defined(CONFIG_DRM_AMD_DC_DCN) union dc_dp_ffe_preset *ffe_preset; -#endif uint16_t *cr_pattern_time; uint16_t *eq_pattern_time; @@ -194,7 +178,6 @@ struct dc_link_training_overrides { bool *fec_enable; }; -#if defined(CONFIG_DRM_AMD_DC_DCN) union payload_table_update_status { struct { uint8_t VC_PAYLOAD_TABLE_UPDATED:1; @@ -202,7 +185,6 @@ union payload_table_update_status { } bits; uint8_t raw; }; -#endif union dpcd_rev { struct { @@ -291,14 +273,10 @@ union lane_align_status_updated { struct { uint8_t INTERLANE_ALIGN_DONE:1; uint8_t POST_LT_ADJ_REQ_IN_PROGRESS:1; -#if defined(CONFIG_DRM_AMD_DC_DCN) uint8_t EQ_INTERLANE_ALIGN_DONE_128b_132b:1; uint8_t CDS_INTERLANE_ALIGN_DONE_128b_132b:1; uint8_t LT_FAILED_128b_132b:1; uint8_t RESERVED:1; -#else - uint8_t RESERVED:4; -#endif uint8_t DOWNSTREAM_PORT_STATUS_CHANGED:1; uint8_t LINK_STATUS_UPDATED:1; } bits; @@ -311,12 +289,10 @@ union lane_adjust { uint8_t PRE_EMPHASIS_LANE:2; uint8_t RESERVED:4; } bits; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct { uint8_t PRESET_VALUE :4; uint8_t RESERVED :4; } tx_ffe; -#endif uint8_t raw; }; @@ -346,12 +322,10 @@ union dpcd_training_lane { uint8_t MAX_PRE_EMPHASIS_REACHED:1; uint8_t RESERVED:2; } bits; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct { uint8_t PRESET_VALUE :4; uint8_t RESERVED :4; } tx_ffe; -#endif uint8_t raw; }; @@ -665,18 +639,9 @@ union test_response { union phy_test_pattern { struct { -#if defined(CONFIG_DRM_AMD_DC_DCN) /* This field is 7 bits for DP2.0 */ uint8_t PATTERN :7; uint8_t RESERVED :1; -#else - /* DpcdPhyTestPatterns. This field is 2 bits for DP1.1 - * and 3 bits for DP1.2. - */ - uint8_t PATTERN :3; - /* BY speci, bit7:2 is 0 for DP1.1. */ - uint8_t RESERVED :5; -#endif } bits; uint8_t raw; }; @@ -754,14 +719,10 @@ union dpcd_fec_capability { uint8_t UNCORRECTED_BLOCK_ERROR_COUNT_CAPABLE:1; uint8_t CORRECTED_BLOCK_ERROR_COUNT_CAPABLE:1; uint8_t BIT_ERROR_COUNT_CAPABLE:1; -#if defined(CONFIG_DRM_AMD_DC_DCN) uint8_t PARITY_BLOCK_ERROR_COUNT_CAPABLE:1; uint8_t ARITY_BIT_ERROR_COUNT_CAPABLE:1; uint8_t FEC_RUNNING_INDICATOR_SUPPORTED:1; uint8_t FEC_ERROR_REPORTING_POLICY_SUPPORTED:1; -#else - uint8_t RESERVED:4; -#endif } bits; uint8_t raw; }; @@ -925,7 +886,6 @@ struct dpcd_usb4_dp_tunneling_info { uint8_t usb4_topology_id[DPCD_USB4_TOPOLOGY_ID_LEN]; }; -#if defined(CONFIG_DRM_AMD_DC_DCN) #ifndef DP_MAIN_LINK_CHANNEL_CODING_CAP #define DP_MAIN_LINK_CHANNEL_CODING_CAP 0x006 #endif @@ -941,6 +901,9 @@ struct dpcd_usb4_dp_tunneling_info { #ifndef DP_LINK_SQUARE_PATTERN #define DP_LINK_SQUARE_PATTERN 0x10F #endif +#ifndef DP_CABLE_ATTRIBUTES_UPDATED_BY_DPTX +#define DP_CABLE_ATTRIBUTES_UPDATED_BY_DPTX 0x110 +#endif #ifndef DP_DSC_CONFIGURATION #define DP_DSC_CONFIGURATION 0x161 #endif @@ -953,6 +916,9 @@ struct dpcd_usb4_dp_tunneling_info { #ifndef DP_128b_132b_TRAINING_AUX_RD_INTERVAL #define DP_128b_132b_TRAINING_AUX_RD_INTERVAL 0x2216 #endif +#ifndef DP_CABLE_ATTRIBUTES_UPDATED_BY_DPRX +#define DP_CABLE_ATTRIBUTES_UPDATED_BY_DPRX 0x2217 +#endif #ifndef DP_TEST_264BIT_CUSTOM_PATTERN_7_0 #define DP_TEST_264BIT_CUSTOM_PATTERN_7_0 0X2230 #endif @@ -994,8 +960,8 @@ struct dpcd_usb4_dp_tunneling_info { #endif #ifndef DP_INTRA_HOP_AUX_REPLY_INDICATION #define DP_INTRA_HOP_AUX_REPLY_INDICATION (1 << 3) -#endif /* TODO - Use DRM header to replace above once available */ +#endif // DP_INTRA_HOP_AUX_REPLY_INDICATION union dp_main_line_channel_coding_cap { struct { @@ -1052,6 +1018,16 @@ union dp_fec_capability1 { uint8_t raw; }; +union dp_cable_attributes { + struct { + uint8_t UHBR10_20_CAPABILITY :2; + uint8_t UHBR13_5_CAPABILITY :1; + uint8_t CABLE_TYPE :3; + uint8_t RESERVED :2; + } bits; + uint8_t raw; +}; + struct dp_color_depth_caps { uint8_t support_6bpc :1; uint8_t support_8bpc :1; @@ -1091,6 +1067,5 @@ union dp_128b_132b_training_aux_rd_interval { } bits; uint8_t raw; }; -#endif #endif /* DC_DP_TYPES_H */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c index ab6bc5d79012..f43cce16bb6c 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c @@ -588,6 +588,66 @@ uint32_t generic_indirect_reg_update_ex(const struct dc_context *ctx, return reg_val; } + +uint32_t generic_indirect_reg_update_ex_sync(const struct dc_context *ctx, + uint32_t index, uint32_t reg_val, int n, + uint8_t shift1, uint32_t mask1, uint32_t field_value1, + ...) +{ + uint32_t shift, mask, field_value; + int i = 1; + + va_list ap; + + va_start(ap, field_value1); + + reg_val = set_reg_field_value_ex(reg_val, field_value1, mask1, shift1); + + while (i < n) { + shift = va_arg(ap, uint32_t); + mask = va_arg(ap, uint32_t); + field_value = va_arg(ap, uint32_t); + + reg_val = set_reg_field_value_ex(reg_val, field_value, mask, shift); + i++; + } + + dm_write_index_reg(ctx, CGS_IND_REG__PCIE, index, reg_val); + va_end(ap); + + return reg_val; +} + +uint32_t generic_indirect_reg_get_sync(const struct dc_context *ctx, + uint32_t index, int n, + uint8_t shift1, uint32_t mask1, uint32_t *field_value1, + ...) +{ + uint32_t shift, mask, *field_value; + uint32_t value = 0; + int i = 1; + + va_list ap; + + va_start(ap, field_value1); + + value = dm_read_index_reg(ctx, CGS_IND_REG__PCIE, index); + *field_value1 = get_reg_field_value_ex(value, mask1, shift1); + + while (i < n) { + shift = va_arg(ap, uint32_t); + mask = va_arg(ap, uint32_t); + field_value = va_arg(ap, uint32_t *); + + *field_value = get_reg_field_value_ex(value, mask, shift); + i++; + } + + va_end(ap); + + return value; +} + void reg_sequence_start_gather(const struct dc_context *ctx) { /* if reg sequence is supported and enabled, set flag to diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index c0e37ad0e26c..9ad3ee4079c3 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -43,14 +43,16 @@ struct dc_link_status { struct dpcd_caps *dpcd_caps; }; +struct dp_receiver_status { + bool cable_id_updated; +}; + /* DP MST stream allocation (payload bandwidth number) */ struct link_mst_stream_allocation { /* DIG front */ const struct stream_encoder *stream_enc; -#if defined(CONFIG_DRM_AMD_DC_DCN) /* HPO DP Stream Encoder */ const struct hpo_dp_stream_encoder *hpo_dp_stream_enc; -#endif /* associate DRM payload table with DC stream encoder */ uint8_t vcp_id; /* number of slots required for the DP stream in transport packet */ @@ -197,10 +199,13 @@ struct dc_link { bool dp_mot_reset_segment; /* Some USB4 docks do not handle turning off MST DSC once it has been enabled. */ bool dpia_mst_dsc_always_on; + /* Forced DPIA into TBT3 compatibility mode. */ + bool dpia_forced_tbt3_mode; } wa_flags; struct link_mst_stream_allocation_table mst_stream_alloc_table; struct dc_link_status link_status; + struct dp_receiver_status dprx_status; struct link_trace link_trace; struct gpio *hpd_gpio; @@ -307,6 +312,7 @@ void dc_link_blank_dp_stream(struct dc_link *link, bool hw_init); */ enum dc_detect_reason { DETECT_REASON_BOOT, + DETECT_REASON_RESUMEFROMS3S4, DETECT_REASON_HPD, DETECT_REASON_HPDRX, DETECT_REASON_FALLBACK, @@ -316,10 +322,8 @@ enum dc_detect_reason { bool dc_link_detect(struct dc_link *dc_link, enum dc_detect_reason reason); bool dc_link_get_hpd_state(struct dc_link *dc_link); enum dc_status dc_link_allocate_mst_payload(struct pipe_ctx *pipe_ctx); -#if defined(CONFIG_DRM_AMD_DC_DCN) enum dc_status dc_link_reduce_mst_payload(struct pipe_ctx *pipe_ctx, uint32_t req_pbn); enum dc_status dc_link_increase_mst_payload(struct pipe_ctx *pipe_ctx, uint32_t req_pbn); -#endif /* Notify DC about DP RX Interrupt (aka Short Pulse Interrupt). * Return: @@ -453,14 +457,17 @@ uint32_t dc_bandwidth_in_kbps_from_timing( bool dc_link_is_fec_supported(const struct dc_link *link); bool dc_link_should_enable_fec(const struct dc_link *link); -#if defined(CONFIG_DRM_AMD_DC_DCN) uint32_t dc_link_bw_kbps_from_raw_frl_link_rate_data(uint8_t bw); enum dp_link_encoding dc_link_dp_mst_decide_link_encoding_format(const struct dc_link *link); -#endif -const struct link_resource *dc_link_get_cur_link_res(const struct dc_link *link); +void dc_link_get_cur_link_res(const struct dc_link *link, + struct link_resource *link_res); /* take a snapshot of current link resource allocation state */ void dc_get_cur_link_res_map(const struct dc *dc, uint32_t *map); /* restore link resource allocation state from a snapshot */ void dc_restore_link_res_map(const struct dc *dc, uint32_t *map); +void dc_link_dp_clear_rx_status(struct dc_link *link); +struct gpio *get_hpd_gpio(struct dc_bios *dcb, + struct graphics_object_id link_id, + struct gpio_service *gpio_service); #endif /* DC_LINK_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index e37c4a10bfd5..f631b61abedd 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -115,12 +115,10 @@ struct periodic_interrupt_config { int lines_offset; }; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct dc_mst_stream_bw_update { bool is_increase; // is bandwidth reduced or increased uint32_t mst_stream_bw; // new mst bandwidth in kbps }; -#endif union stream_update_flags { struct { @@ -132,9 +130,7 @@ union stream_update_flags { uint32_t gamut_remap:1; uint32_t wb_update:1; uint32_t dsc_changed : 1; -#if defined(CONFIG_DRM_AMD_DC_DCN) uint32_t mst_bw : 1; -#endif } bits; uint32_t raw; @@ -288,9 +284,7 @@ struct dc_stream_update { struct dc_writeback_update *wb_update; struct dc_dsc_config *dsc_config; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct dc_mst_stream_bw_update *mst_bw_update; -#endif struct dc_transfer_func *func_shaper; struct dc_3dlut *lut3d_func; diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 0285a4b38d05..48859d5fc172 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -395,14 +395,11 @@ struct dc_lttpr_caps { uint8_t max_link_rate; uint8_t phy_repeater_cnt; uint8_t max_ext_timeout; -#if defined(CONFIG_DRM_AMD_DC_DCN) union dp_main_link_channel_coding_lttpr_cap main_link_channel_coding; union dp_128b_132b_supported_lttpr_link_rates supported_128b_132b_rates; -#endif uint8_t aux_rd_interval[MAX_REPEATER_CNT - 1]; }; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct dc_dongle_dfp_cap_ext { bool supported; uint16_t max_pixel_rate_in_mps; @@ -414,7 +411,6 @@ struct dc_dongle_dfp_cap_ext { struct dp_color_depth_caps ycbcr422_color_depth_caps; struct dp_color_depth_caps ycbcr420_color_depth_caps; }; -#endif struct dc_dongle_caps { /* dongle type (DP converter, CV smart dongle) */ @@ -429,10 +425,8 @@ struct dc_dongle_caps { bool is_dp_hdmi_ycbcr420_converter; uint32_t dp_hdmi_max_bpc; uint32_t dp_hdmi_max_pixel_clk_in_khz; -#if defined(CONFIG_DRM_AMD_DC_DCN) uint32_t dp_hdmi_frl_max_link_bw_in_kbps; struct dc_dongle_dfp_cap_ext dfp_cap_ext; -#endif }; /* Scaling format */ enum scaling_transformation { diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c index 6d42a9cc9916..74b05b3aef08 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c @@ -878,7 +878,7 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc, default: DC_TRACE_LEVEL_MESSAGE(DAL_TRACE_LEVEL_ERROR, LOG_FLAG_Error_I2cAux, - "dce_aux_transfer_with_retries: AUX_RET_SUCCESS: FAILURE: AUX_TRANSACTION_REPLY_* unknown, default case."); + "dce_aux_transfer_with_retries: AUX_RET_SUCCESS: FAILURE: AUX_TRANSACTION_REPLY_* unknown, default case. Reply: %d", *payload->reply); goto fail; } break; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c index f1c61d5aee6c..0bc41414481e 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c @@ -1325,7 +1325,8 @@ void dce110_link_encoder_disable_output( void dce110_link_encoder_dp_set_lane_settings( struct link_encoder *enc, - const struct link_training_settings *link_settings) + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]) { struct dce110_link_encoder *enc110 = TO_DCE110_LINK_ENC(enc); union dpcd_training_lane_set training_lane_set = { { 0 } }; @@ -1340,26 +1341,26 @@ void dce110_link_encoder_dp_set_lane_settings( cntl.action = TRANSMITTER_CONTROL_SET_VOLTAGE_AND_PREEMPASIS; cntl.transmitter = enc110->base.transmitter; cntl.connector_obj_id = enc110->base.connector; - cntl.lanes_number = link_settings->link_settings.lane_count; + cntl.lanes_number = link_settings->lane_count; cntl.hpd_sel = enc110->base.hpd_source; - cntl.pixel_clock = link_settings->link_settings.link_rate * + cntl.pixel_clock = link_settings->link_rate * LINK_RATE_REF_FREQ_IN_KHZ; - for (lane = 0; lane < link_settings->link_settings.lane_count; lane++) { + for (lane = 0; lane < link_settings->lane_count; lane++) { /* translate lane settings */ training_lane_set.bits.VOLTAGE_SWING_SET = - link_settings->lane_settings[lane].VOLTAGE_SWING; + lane_settings[lane].VOLTAGE_SWING; training_lane_set.bits.PRE_EMPHASIS_SET = - link_settings->lane_settings[lane].PRE_EMPHASIS; + lane_settings[lane].PRE_EMPHASIS; /* post cursor 2 setting only applies to HBR2 link rate */ - if (link_settings->link_settings.link_rate == LINK_RATE_HIGH2) { + if (link_settings->link_rate == LINK_RATE_HIGH2) { /* this is passed to VBIOS * to program post cursor 2 level */ training_lane_set.bits.POST_CURSOR2_SET = - link_settings->lane_settings[lane].POST_CURSOR2; + lane_settings[lane].POST_CURSOR2; } cntl.lane_select = lane; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h index fc6ade824c23..261c70e01e33 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.h @@ -279,7 +279,8 @@ void dce110_link_encoder_disable_output( /* set DP lane settings */ void dce110_link_encoder_dp_set_lane_settings( struct link_encoder *enc, - const struct link_training_settings *link_settings); + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]); void dce110_link_encoder_dp_set_phy_pattern( struct link_encoder *enc, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c index 87ed48d5530d..8bd265b40847 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c @@ -138,6 +138,10 @@ static bool dmub_psr_set_version(struct dmub_psr *dmub, struct dc_stream_state * cmd.psr_set_version.psr_set_version_data.version = PSR_VERSION_UNSUPPORTED; break; } + + if (cmd.psr_set_version.psr_set_version_data.version == PSR_VERSION_UNSUPPORTED) + return false; + cmd.psr_set_version.psr_set_version_data.cmd_version = DMUB_CMD_PSR_CONTROL_VERSION_1; cmd.psr_set_version.psr_set_version_data.panel_inst = panel_inst; cmd.psr_set_version.header.payload_bytes = sizeof(struct dmub_cmd_psr_set_version_data); diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index eb2755bdb30e..ace04e2ed34e 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -49,9 +49,7 @@ #include "link_enc_cfg.h" #include "link_hwss.h" #include "dc_link_dp.h" -#if defined(CONFIG_DRM_AMD_DC_DCN) #include "dccg.h" -#endif #include "clock_source.h" #include "clk_mgr.h" #include "abm.h" @@ -669,17 +667,12 @@ void dce110_enable_stream(struct pipe_ctx *pipe_ctx) struct dc_crtc_timing *timing = &pipe_ctx->stream->timing; struct dc_link *link = pipe_ctx->stream->link; const struct dc *dc = link->dc; - + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); uint32_t active_total_with_borders; uint32_t early_control = 0; struct timing_generator *tg = pipe_ctx->stream_res.tg; - /* For MST, there are multiply stream go to only one link. - * connect DIG back_end to front_end while enable_stream and - * disconnect them during disable_stream - * BY this, it is logic clean to separate stream and link */ - link->link_enc->funcs->connect_dig_be_to_fe(link->link_enc, - pipe_ctx->stream_res.stream_enc->id, true); + link_hwss->setup_stream_encoder(pipe_ctx); dc->hwss.update_info_frame(pipe_ctx); @@ -795,7 +788,7 @@ void dce110_edp_wait_for_hpd_ready( dal_gpio_destroy_irq(&hpd); if (false == edp_hpd_high) { - DC_LOG_ERROR( + DC_LOG_WARNING( "%s: wait timed out!\n", __func__); } } @@ -1112,17 +1105,12 @@ void dce110_enable_audio_stream(struct pipe_ctx *pipe_ctx) clk_mgr->funcs->enable_pme_wa(clk_mgr); /* un-mute audio */ /* TODO: audio should be per stream rather than per link */ -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->audio_mute_control( pipe_ctx->stream_res.hpo_dp_stream_enc, false); else pipe_ctx->stream_res.stream_enc->funcs->audio_mute_control( pipe_ctx->stream_res.stream_enc, false); -#else - pipe_ctx->stream_res.stream_enc->funcs->audio_mute_control( - pipe_ctx->stream_res.stream_enc, false); -#endif if (pipe_ctx->stream_res.audio) pipe_ctx->stream_res.audio->enabled = true; } @@ -1145,32 +1133,22 @@ void dce110_disable_audio_stream(struct pipe_ctx *pipe_ctx) if (pipe_ctx->stream_res.audio && pipe_ctx->stream_res.audio->enabled == false) return; -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->audio_mute_control( pipe_ctx->stream_res.hpo_dp_stream_enc, true); else pipe_ctx->stream_res.stream_enc->funcs->audio_mute_control( pipe_ctx->stream_res.stream_enc, true); -#else - pipe_ctx->stream_res.stream_enc->funcs->audio_mute_control( - pipe_ctx->stream_res.stream_enc, true); -#endif if (pipe_ctx->stream_res.audio) { pipe_ctx->stream_res.audio->enabled = false; if (dc_is_dp_signal(pipe_ctx->stream->signal)) -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_audio_disable( pipe_ctx->stream_res.hpo_dp_stream_enc); else pipe_ctx->stream_res.stream_enc->funcs->dp_audio_disable( pipe_ctx->stream_res.stream_enc); -#else - pipe_ctx->stream_res.stream_enc->funcs->dp_audio_disable( - pipe_ctx->stream_res.stream_enc); -#endif else pipe_ctx->stream_res.stream_enc->funcs->hdmi_audio_disable( pipe_ctx->stream_res.stream_enc); @@ -1195,7 +1173,7 @@ void dce110_disable_stream(struct pipe_ctx *pipe_ctx) struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; struct dc *dc = pipe_ctx->stream->ctx->dc; - struct link_encoder *link_enc = NULL; + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) { pipe_ctx->stream_res.stream_enc->funcs->stop_hdmi_info_packets( @@ -1204,54 +1182,26 @@ void dce110_disable_stream(struct pipe_ctx *pipe_ctx) pipe_ctx->stream_res.stream_enc); } -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) { pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->stop_dp_info_packets( pipe_ctx->stream_res.hpo_dp_stream_enc); } else if (dc_is_dp_signal(pipe_ctx->stream->signal)) -#else - if (dc_is_dp_signal(pipe_ctx->stream->signal)) -#endif pipe_ctx->stream_res.stream_enc->funcs->stop_dp_info_packets( pipe_ctx->stream_res.stream_enc); dc->hwss.disable_audio_stream(pipe_ctx); - /* Link encoder may have been dynamically assigned to non-physical display endpoint. */ - if (link->ep_type == DISPLAY_ENDPOINT_PHY) - link_enc = link->link_enc; - else if (dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - ASSERT(link_enc); + link_hwss->reset_stream_encoder(pipe_ctx); -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) { - pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->disable( - pipe_ctx->stream_res.hpo_dp_stream_enc); - setup_dp_hpo_stream(pipe_ctx, false); - /* TODO - DP2.0 HW: unmap stream from link encoder here */ - } else { - if (link_enc) - link_enc->funcs->connect_dig_be_to_fe( - link_enc, - pipe_ctx->stream_res.stream_enc->id, - false); - } -#else - if (link_enc) - link_enc->funcs->connect_dig_be_to_fe( - link->link_enc, - pipe_ctx->stream_res.stream_enc->id, - false); -#endif - if (dc_is_dp_signal(pipe_ctx->stream->signal)) - dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_DISCONNECT_DIG_FE_BE); - -#if defined(CONFIG_DRM_AMD_DC_DCN) - if (dc->hwseq->funcs.setup_hpo_hw_control && is_dp_128b_132b_signal(pipe_ctx)) - dc->hwseq->funcs.setup_hpo_hw_control(dc->hwseq, false); -#endif - + /* TODO: This looks like a bug to me as we are disabling HPO IO when + * we are just disabling a single HPO stream. Shouldn't we disable HPO + * HW control only when HPOs for all streams are disabled? + */ + if (pipe_ctx->stream->ctx->dc->hwseq->funcs.setup_hpo_hw_control) + pipe_ctx->stream->ctx->dc->hwseq->funcs.setup_hpo_hw_control( + pipe_ctx->stream->ctx->dc->hwseq, false); + } } void dce110_unblank_stream(struct pipe_ctx *pipe_ctx, @@ -1285,15 +1235,11 @@ void dce110_blank_stream(struct pipe_ctx *pipe_ctx) link->dc->hwss.set_abm_immediate_disable(pipe_ctx); } -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) { /* TODO - DP2.0 HW: Set ODM mode in dp hpo encoder here */ pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_blank( pipe_ctx->stream_res.hpo_dp_stream_enc); } else if (dc_is_dp_signal(pipe_ctx->stream->signal)) { -#else - if (dc_is_dp_signal(pipe_ctx->stream->signal)) { -#endif pipe_ctx->stream_res.stream_enc->funcs->dp_blank(link, pipe_ctx->stream_res.stream_enc); if (!dc_is_embedded_signal(pipe_ctx->stream->signal)) { @@ -1535,7 +1481,6 @@ static enum dc_status apply_single_controller_ctx_to_hw( build_audio_output(context, pipe_ctx, &audio_output); if (dc_is_dp_signal(pipe_ctx->stream->signal)) -#if defined(CONFIG_DRM_AMD_DC_DCN) if (is_dp_128b_132b_signal(pipe_ctx)) pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_audio_setup( pipe_ctx->stream_res.hpo_dp_stream_enc, @@ -1546,12 +1491,6 @@ static enum dc_status apply_single_controller_ctx_to_hw( pipe_ctx->stream_res.stream_enc, pipe_ctx->stream_res.audio->inst, &pipe_ctx->stream->audio_info); -#else - pipe_ctx->stream_res.stream_enc->funcs->dp_audio_setup( - pipe_ctx->stream_res.stream_enc, - pipe_ctx->stream_res.audio->inst, - &pipe_ctx->stream->audio_info); -#endif else pipe_ctx->stream_res.stream_enc->funcs->hdmi_audio_setup( pipe_ctx->stream_res.stream_enc, @@ -1570,14 +1509,12 @@ static enum dc_status apply_single_controller_ctx_to_hw( if (!pipe_ctx->stream->apply_seamless_boot_optimization && dc->config.use_pipe_ctx_sync_logic) check_syncd_pipes_for_disabled_master_pipe(dc, context, pipe_ctx->pipe_idx); -#if defined(CONFIG_DRM_AMD_DC_DCN) /* DCN3.1 FPGA Workaround * Need to enable HPO DP Stream Encoder before setting OTG master enable. * To do so, move calling function enable_stream_timing to only be done AFTER calling * function core_link_enable_stream */ if (!(hws->wa.dp_hpo_and_otg_sequence && is_dp_128b_132b_signal(pipe_ctx))) -#endif /* */ /* Do not touch stream timing on seamless boot optimization. */ if (!pipe_ctx->stream->apply_seamless_boot_optimization) @@ -1638,7 +1575,6 @@ static enum dc_status apply_single_controller_ctx_to_hw( if (!stream->dpms_off) core_link_enable_stream(context, pipe_ctx); -#if defined(CONFIG_DRM_AMD_DC_DCN) /* DCN3.1 FPGA Workaround * Need to enable HPO DP Stream Encoder before setting OTG master enable. * To do so, move calling function enable_stream_timing to only be done AFTER calling @@ -1648,7 +1584,6 @@ static enum dc_status apply_single_controller_ctx_to_hw( if (!pipe_ctx->stream->apply_seamless_boot_optimization) hws->funcs.enable_stream_timing(pipe_ctx, context, dc); } -#endif pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->bottom_pipe != 0; @@ -1826,7 +1761,8 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context) edp_link->link_status.link_active) { struct dc_stream_state *edp_stream = edp_streams[0]; - can_apply_edp_fast_boot = !is_edp_ilr_optimization_required(edp_stream->link, &edp_stream->timing); + can_apply_edp_fast_boot = dc_validate_boot_timing(dc, + edp_stream->sink, &edp_stream->timing); edp_stream->apply_edp_fast_boot_optimization = can_apply_edp_fast_boot; if (can_apply_edp_fast_boot) DC_LOG_EVENT_LINK_TRAINING("eDP fast boot disabled to optimize link rate\n"); @@ -2248,8 +2184,6 @@ static void dce110_setup_audio_dto( build_audio_output(context, pipe_ctx, &audio_output); -#if defined(CONFIG_DRM_AMD_DC_DCN) - /* For DCN3.1, audio to HPO FRL encoder is using audio DTBCLK DTO */ if (dc->res_pool->dccg && dc->res_pool->dccg->funcs->set_audio_dtbclk_dto) { /* disable audio DTBCLK DTO */ dc->res_pool->dccg->funcs->set_audio_dtbclk_dto( @@ -2266,13 +2200,6 @@ static void dce110_setup_audio_dto( pipe_ctx->stream->signal, &audio_output.crtc_info, &audio_output.pll_info); -#else - pipe_ctx->stream_res.audio->funcs->wall_dto_setup( - pipe_ctx->stream_res.audio, - pipe_ctx->stream->signal, - &audio_output.crtc_info, - &audio_output.pll_info); -#endif break; } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index 530a72e3eefe..915eecb40788 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -1505,8 +1505,7 @@ void dcn10_init_hw(struct dc *dc) dmub_enable_outbox_notification(dc); /* we want to turn off all dp displays before doing detection */ - if (dc->config.power_down_display_on_boot) - dc_link_blank_all_dp_displays(dc); + dc_link_blank_all_dp_displays(dc); /* If taking control over from VBIOS, we may want to optimize our first * mode set, so we need to skip powering down pipes until we know which @@ -1514,7 +1513,7 @@ void dcn10_init_hw(struct dc *dc) * Otherwise, if taking control is not possible, we need to power * everything down. */ - if (dcb->funcs->is_accelerated_mode(dcb) || dc->config.power_down_display_on_boot) { + if (dcb->funcs->is_accelerated_mode(dcb) || !dc->config.seamless_boot_edp_requested) { if (!is_optimized_init_done) { hws->funcs.init_pipes(dc, dc->current_state); if (dc->res_pool->hubbub->funcs->allow_self_refresh_control) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c index f4b34c110eae..ca39361f71c8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c @@ -1101,7 +1101,8 @@ void dcn10_link_encoder_disable_output( void dcn10_link_encoder_dp_set_lane_settings( struct link_encoder *enc, - const struct link_training_settings *link_settings) + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]) { struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); union dpcd_training_lane_set training_lane_set = { { 0 } }; @@ -1116,26 +1117,25 @@ void dcn10_link_encoder_dp_set_lane_settings( cntl.action = TRANSMITTER_CONTROL_SET_VOLTAGE_AND_PREEMPASIS; cntl.transmitter = enc10->base.transmitter; cntl.connector_obj_id = enc10->base.connector; - cntl.lanes_number = link_settings->link_settings.lane_count; + cntl.lanes_number = link_settings->lane_count; cntl.hpd_sel = enc10->base.hpd_source; - cntl.pixel_clock = link_settings->link_settings.link_rate * - LINK_RATE_REF_FREQ_IN_KHZ; + cntl.pixel_clock = link_settings->link_rate * LINK_RATE_REF_FREQ_IN_KHZ; - for (lane = 0; lane < link_settings->link_settings.lane_count; lane++) { + for (lane = 0; lane < link_settings->lane_count; lane++) { /* translate lane settings */ training_lane_set.bits.VOLTAGE_SWING_SET = - link_settings->lane_settings[lane].VOLTAGE_SWING; + lane_settings[lane].VOLTAGE_SWING; training_lane_set.bits.PRE_EMPHASIS_SET = - link_settings->lane_settings[lane].PRE_EMPHASIS; + lane_settings[lane].PRE_EMPHASIS; /* post cursor 2 setting only applies to HBR2 link rate */ - if (link_settings->link_settings.link_rate == LINK_RATE_HIGH2) { + if (link_settings->link_rate == LINK_RATE_HIGH2) { /* this is passed to VBIOS * to program post cursor 2 level */ training_lane_set.bits.POST_CURSOR2_SET = - link_settings->lane_settings[lane].POST_CURSOR2; + lane_settings[lane].POST_CURSOR2; } cntl.lane_select = lane; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h index c337588231ff..663aac0a164a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.h @@ -581,7 +581,8 @@ void dcn10_link_encoder_disable_output( /* set DP lane settings */ void dcn10_link_encoder_dp_set_lane_settings( struct link_encoder *enc, - const struct link_training_settings *link_settings); + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]); void dcn10_link_encoder_dp_set_phy_pattern( struct link_encoder *enc, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h index f98aba308028..b3c9a9724efd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h @@ -183,15 +183,25 @@ type SYMCLK32_ROOT_SE1_GATE_DISABLE;\ type SYMCLK32_ROOT_SE2_GATE_DISABLE;\ type SYMCLK32_ROOT_SE3_GATE_DISABLE;\ + type SYMCLK32_SE0_GATE_DISABLE;\ + type SYMCLK32_SE1_GATE_DISABLE;\ + type SYMCLK32_SE2_GATE_DISABLE;\ + type SYMCLK32_SE3_GATE_DISABLE;\ type SYMCLK32_ROOT_LE0_GATE_DISABLE;\ type SYMCLK32_ROOT_LE1_GATE_DISABLE;\ + type SYMCLK32_LE0_GATE_DISABLE;\ + type SYMCLK32_LE1_GATE_DISABLE;\ type DPSTREAMCLK_ROOT_GATE_DISABLE;\ type DPSTREAMCLK_GATE_DISABLE;\ type HDMISTREAMCLK0_DTO_PHASE;\ type HDMISTREAMCLK0_DTO_MODULO;\ type HDMICHARCLK0_GATE_DISABLE;\ - type HDMICHARCLK0_ROOT_GATE_DISABLE; - + type HDMICHARCLK0_ROOT_GATE_DISABLE; \ + type PHYASYMCLK_GATE_DISABLE; \ + type PHYBSYMCLK_GATE_DISABLE; \ + type PHYCSYMCLK_GATE_DISABLE; \ + type PHYDSYMCLK_GATE_DISABLE; \ + type PHYESYMCLK_GATE_DISABLE; struct dccg_shift { DCCG_REG_FIELD_LIST(uint8_t) @@ -233,6 +243,7 @@ struct dccg_registers { uint32_t DSCCLK2_DTO_PARAM; uint32_t DPSTREAMCLK_ROOT_GATE_DISABLE; uint32_t DPSTREAMCLK_GATE_DISABLE; + uint32_t DCCG_GATE_DISABLE_CNTL2; uint32_t DCCG_GATE_DISABLE_CNTL3; uint32_t HDMISTREAMCLK0_DTO_PARAM; uint32_t DCCG_GATE_DISABLE_CNTL4; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index 4991e93e5308..20a9cbb7c0a8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -55,6 +55,7 @@ #include "inc/link_dpcd.h" #include "dpcd_defs.h" #include "inc/link_enc_cfg.h" +#include "link_hwss.h" #define DC_LOGGER_INIT(logger) @@ -2390,46 +2391,22 @@ void dcn20_enable_stream(struct pipe_ctx *pipe_ctx) uint32_t active_total_with_borders; uint32_t early_control = 0; struct timing_generator *tg = pipe_ctx->stream_res.tg; - struct link_encoder *link_enc; + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); + struct dc *dc = pipe_ctx->stream->ctx->dc; - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) - link_enc = link_enc_cfg_get_link_enc_used_by_stream(link->ctx->dc, pipe_ctx->stream); - else - link_enc = link->link_enc; - ASSERT(link_enc); - - /* For MST, there are multiply stream go to only one link. - * connect DIG back_end to front_end while enable_stream and - * disconnect them during disable_stream - * BY this, it is logic clean to separate stream and link - */ if (is_dp_128b_132b_signal(pipe_ctx)) { - if (pipe_ctx->stream->ctx->dc->hwseq->funcs.setup_hpo_hw_control) - pipe_ctx->stream->ctx->dc->hwseq->funcs.setup_hpo_hw_control( - pipe_ctx->stream->ctx->dc->hwseq, true); - setup_dp_hpo_stream(pipe_ctx, true); - pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->enable_stream( - pipe_ctx->stream_res.hpo_dp_stream_enc); - pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->map_stream_to_link( - pipe_ctx->stream_res.hpo_dp_stream_enc, - pipe_ctx->stream_res.hpo_dp_stream_enc->inst, - pipe_ctx->link_res.hpo_dp_link_enc->inst); + if (dc->hwseq->funcs.setup_hpo_hw_control) + dc->hwseq->funcs.setup_hpo_hw_control(dc->hwseq, true); } - if (!is_dp_128b_132b_signal(pipe_ctx) && link_enc) - link_enc->funcs->connect_dig_be_to_fe( - link_enc, pipe_ctx->stream_res.stream_enc->id, true); - - if (dc_is_dp_signal(pipe_ctx->stream->signal)) - dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_CONNECT_DIG_FE_BE); + link_hwss->setup_stream_encoder(pipe_ctx); if (pipe_ctx->plane_state && pipe_ctx->plane_state->flip_immediate != 1) { - if (link->dc->hwss.program_dmdata_engine) - link->dc->hwss.program_dmdata_engine(pipe_ctx); + if (dc->hwss.program_dmdata_engine) + dc->hwss.program_dmdata_engine(pipe_ctx); } - link->dc->hwss.update_info_frame(pipe_ctx); + dc->hwss.update_info_frame(pipe_ctx); if (dc_is_dp_signal(pipe_ctx->stream->signal)) dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_UPDATE_INFO_FRAME); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index 2a72517e2b28..dfe2e1c25a26 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -1605,16 +1605,7 @@ static void get_pixel_clock_parameters( pixel_clk_params->requested_pix_clk_100hz = stream->timing.pix_clk_100hz; - /* Links supporting dynamically assigned link encoder will be assigned next - * available encoder if one not already assigned. - */ - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) { - link_enc = link_enc_cfg_get_link_enc_used_by_stream(stream->ctx->dc, stream); - if (link_enc == NULL) - link_enc = link_enc_cfg_get_next_avail_link_enc(stream->ctx->dc); - } else - link_enc = stream->link->link_enc; + link_enc = link_enc_cfg_get_link_enc(link); ASSERT(link_enc); if (link_enc) @@ -3093,8 +3084,14 @@ static enum dcn_zstate_support_state decide_zstate_support(struct dc *dc, struc else if (context->stream_count == 1 && context->streams[0]->signal == SIGNAL_TYPE_EDP) { struct dc_link *link = context->streams[0]->sink->link; - if (link->link_index == 0 && context->bw_ctx.dml.vba.StutterPeriod > 5000.0) + /* zstate only supported on PWRSEQ0 */ + if (link->link_index != 0) + return DCN_ZSTATE_SUPPORT_DISALLOW; + + if (context->bw_ctx.dml.vba.StutterPeriod > 5000.0) return DCN_ZSTATE_SUPPORT_ALLOW; + else if (link->psr_settings.psr_version == DC_PSR_VERSION_1 && !dc->debug.disable_psr) + return DCN_ZSTATE_SUPPORT_ALLOW_Z10_ONLY; else return DCN_ZSTATE_SUPPORT_DISALLOW; } else diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c index 1db1ca19411d..ed0a0e5fd805 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c @@ -545,8 +545,7 @@ void dcn30_init_hw(struct dc *dc) hws->funcs.dsc_pg_control(hws, res_pool->dscs[i]->inst, false); /* we want to turn off all dp displays before doing detection */ - if (dc->config.power_down_display_on_boot) - dc_link_blank_all_dp_displays(dc); + dc_link_blank_all_dp_displays(dc); /* If taking control over from VBIOS, we may want to optimize our first * mode set, so we need to skip powering down pipes until we know which @@ -554,7 +553,7 @@ void dcn30_init_hw(struct dc *dc) * Otherwise, if taking control is not possible, we need to power * everything down. */ - if (dcb->funcs->is_accelerated_mode(dcb) || dc->config.power_down_display_on_boot) { + if (dcb->funcs->is_accelerated_mode(dcb) || !dc->config.seamless_boot_edp_requested) { hws->funcs.init_pipes(dc, dc->current_state); if (dc->res_pool->hubbub->funcs->allow_self_refresh_control) dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, @@ -566,7 +565,7 @@ void dcn30_init_hw(struct dc *dc) * To avoid this, power down hardware on boot * if DIG is turned on and seamless boot not enabled */ - if (dc->config.power_down_display_on_boot) { + if (!dc->config.seamless_boot_edp_requested) { struct dc_link *edp_links[MAX_NUM_EDP]; struct dc_link *edp_link = NULL; diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c index 8ca26383b568..f10f7a0ca02a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c @@ -72,8 +72,8 @@ #include "nbio/nbio_7_4_offset.h" -#include "dcn/dpcs_3_0_0_offset.h" -#include "dcn/dpcs_3_0_0_sh_mask.h" +#include "dpcs/dpcs_3_0_0_offset.h" +#include "dpcs/dpcs_3_0_0_sh_mask.h" #include "mmhub/mmhub_2_0_0_offset.h" #include "mmhub/mmhub_2_0_0_sh_mask.h" diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c index 5d9637b07429..4daf8931aa7c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c @@ -73,8 +73,8 @@ #include "nbio/nbio_7_2_0_offset.h" -#include "dcn/dpcs_3_0_0_offset.h" -#include "dcn/dpcs_3_0_0_sh_mask.h" +#include "dpcs/dpcs_3_0_0_offset.h" +#include "dpcs/dpcs_3_0_0_sh_mask.h" #include "reg_helper.h" #include "dce/dmub_abm.h" diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/Makefile b/drivers/gpu/drm/amd/display/dc/dcn302/Makefile index 101620a8867a..f9561d7f97a1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn302/Makefile @@ -1,11 +1,6 @@ # # (c) Copyright 2020 Advanced Micro Devices, Inc. All the rights reserved # -# All rights reserved. This notice is intended as a precaution against -# inadvertent publication and does not imply publication or any waiver -# of confidentiality. The year included in the foregoing notice is the -# year of creation of the work. -# # Authors: AMD # # Makefile for dcn302. @@ -20,13 +15,6 @@ ifdef CONFIG_PPC64 CFLAGS_$(AMDDALPATH)/dc/dcn302/dcn302_resource.o := -mhard-float -maltivec endif -ifdef CONFIG_CC_IS_GCC -ifeq ($(call cc-ifversion, -lt, 0701, y), y) -IS_OLD_GCC = 1 -endif -CFLAGS_$(AMDDALPATH)/dc/dcn302/dcn302_resource.o += -mhard-float -endif - ifdef CONFIG_X86 ifdef IS_OLD_GCC # Stack alignment mismatch, proceed with caution. diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c index 2e9cbfa7663b..88318e8ffca8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c @@ -61,150 +61,18 @@ #include "resource.h" #include "vm_helper.h" +#include "dml/dcn302/dcn302_fpu.h" + #include "dimgrey_cavefish_ip_offset.h" #include "dcn/dcn_3_0_2_offset.h" #include "dcn/dcn_3_0_2_sh_mask.h" -#include "dcn/dpcs_3_0_0_offset.h" -#include "dcn/dpcs_3_0_0_sh_mask.h" +#include "dpcs/dpcs_3_0_0_offset.h" +#include "dpcs/dpcs_3_0_0_sh_mask.h" #include "nbio/nbio_7_4_offset.h" #include "amdgpu_socbb.h" #define DC_LOGGER_INIT(logger) -struct _vcs_dpi_ip_params_st dcn3_02_ip = { - .use_min_dcfclk = 0, - .clamp_min_dcfclk = 0, - .odm_capable = 1, - .gpuvm_enable = 1, - .hostvm_enable = 0, - .gpuvm_max_page_table_levels = 4, - .hostvm_max_page_table_levels = 4, - .hostvm_cached_page_table_levels = 0, - .pte_group_size_bytes = 2048, - .num_dsc = 5, - .rob_buffer_size_kbytes = 184, - .det_buffer_size_kbytes = 184, - .dpte_buffer_size_in_pte_reqs_luma = 64, - .dpte_buffer_size_in_pte_reqs_chroma = 34, - .pde_proc_buffer_size_64k_reqs = 48, - .dpp_output_buffer_pixels = 2560, - .opp_output_buffer_lines = 1, - .pixel_chunk_size_kbytes = 8, - .pte_enable = 1, - .max_page_table_levels = 2, - .pte_chunk_size_kbytes = 2, // ? - .meta_chunk_size_kbytes = 2, - .writeback_chunk_size_kbytes = 8, - .line_buffer_size_bits = 789504, - .is_line_buffer_bpp_fixed = 0, // ? - .line_buffer_fixed_bpp = 0, // ? - .dcc_supported = true, - .writeback_interface_buffer_size_kbytes = 90, - .writeback_line_buffer_buffer_size = 0, - .max_line_buffer_lines = 12, - .writeback_luma_buffer_size_kbytes = 12, // writeback_line_buffer_buffer_size = 656640 - .writeback_chroma_buffer_size_kbytes = 8, - .writeback_chroma_line_buffer_width_pixels = 4, - .writeback_max_hscl_ratio = 1, - .writeback_max_vscl_ratio = 1, - .writeback_min_hscl_ratio = 1, - .writeback_min_vscl_ratio = 1, - .writeback_max_hscl_taps = 1, - .writeback_max_vscl_taps = 1, - .writeback_line_buffer_luma_buffer_size = 0, - .writeback_line_buffer_chroma_buffer_size = 14643, - .cursor_buffer_size = 8, - .cursor_chunk_size = 2, - .max_num_otg = 5, - .max_num_dpp = 5, - .max_num_wb = 1, - .max_dchub_pscl_bw_pix_per_clk = 4, - .max_pscl_lb_bw_pix_per_clk = 2, - .max_lb_vscl_bw_pix_per_clk = 4, - .max_vscl_hscl_bw_pix_per_clk = 4, - .max_hscl_ratio = 6, - .max_vscl_ratio = 6, - .hscl_mults = 4, - .vscl_mults = 4, - .max_hscl_taps = 8, - .max_vscl_taps = 8, - .dispclk_ramp_margin_percent = 1, - .underscan_factor = 1.11, - .min_vblank_lines = 32, - .dppclk_delay_subtotal = 46, - .dynamic_metadata_vm_enabled = true, - .dppclk_delay_scl_lb_only = 16, - .dppclk_delay_scl = 50, - .dppclk_delay_cnvc_formatter = 27, - .dppclk_delay_cnvc_cursor = 6, - .dispclk_delay_subtotal = 119, - .dcfclk_cstate_latency = 5.2, // SRExitTime - .max_inter_dcn_tile_repeaters = 8, - .max_num_hdmi_frl_outputs = 1, - .odm_combine_4to1_supported = true, - - .xfc_supported = false, - .xfc_fill_bw_overhead_percent = 10.0, - .xfc_fill_constant_bytes = 0, - .gfx7_compat_tiling_supported = 0, - .number_of_cursors = 1, -}; - -struct _vcs_dpi_soc_bounding_box_st dcn3_02_soc = { - .clock_limits = { - { - .state = 0, - .dispclk_mhz = 562.0, - .dppclk_mhz = 300.0, - .phyclk_mhz = 300.0, - .phyclk_d18_mhz = 667.0, - .dscclk_mhz = 405.6, - }, - }, - - .min_dcfclk = 500.0, /* TODO: set this to actual min DCFCLK */ - .num_states = 1, - .sr_exit_time_us = 26.5, - .sr_enter_plus_exit_time_us = 31, - .urgent_latency_us = 4.0, - .urgent_latency_pixel_data_only_us = 4.0, - .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, - .urgent_latency_vm_data_only_us = 4.0, - .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, - .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, - .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, - .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, - .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, - .max_avg_sdp_bw_use_normal_percent = 60.0, - .max_avg_dram_bw_use_normal_percent = 40.0, - .writeback_latency_us = 12.0, - .max_request_size_bytes = 256, - .fabric_datapath_to_dcn_data_return_bytes = 64, - .dcn_downspread_percent = 0.5, - .downspread_percent = 0.38, - .dram_page_open_time_ns = 50.0, - .dram_rw_turnaround_time_ns = 17.5, - .dram_return_buffer_per_channel_bytes = 8192, - .round_trip_ping_latency_dcfclk_cycles = 156, - .urgent_out_of_order_return_per_channel_bytes = 4096, - .channel_interleave_bytes = 256, - .num_banks = 8, - .gpuvm_min_page_size_bytes = 4096, - .hostvm_min_page_size_bytes = 4096, - .dram_clock_change_latency_us = 404, - .dummy_pstate_latency_us = 5, - .writeback_dram_clock_change_latency_us = 23.0, - .return_bus_width_bytes = 64, - .dispclk_dppclk_vco_speed_mhz = 3650, - .xfc_bus_transport_time_us = 20, // ? - .xfc_xbuf_latency_tolerance_us = 4, // ? - .use_urgent_burst_bw = 1, // ? - .do_urgent_latency_adjustment = true, - .urgent_latency_adjustment_fabric_clock_component_us = 1.0, - .urgent_latency_adjustment_fabric_clock_reference_mhz = 1000, -}; - static const struct dc_debug_options debug_defaults_drv = { .disable_dmcu = true, .force_abm_enable = false, @@ -1105,24 +973,19 @@ static bool init_soc_bounding_box(struct dc *dc, struct resource_pool *pool) loaded_ip->max_num_otg = pool->pipe_count; loaded_ip->max_num_dpp = pool->pipe_count; loaded_ip->clamp_min_dcfclk = dc->config.clamp_min_dcfclk; + DC_FP_START(); dcn20_patch_bounding_box(dc, loaded_bb); + DC_FP_END(); if (dc->ctx->dc_bios->funcs->get_soc_bb_info) { struct bp_soc_bb_info bb_info = { 0 }; if (dc->ctx->dc_bios->funcs->get_soc_bb_info( dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) { - if (bb_info.dram_clock_change_latency_100ns > 0) - dcn3_02_soc.dram_clock_change_latency_us = - bb_info.dram_clock_change_latency_100ns * 10; - if (bb_info.dram_sr_enter_exit_latency_100ns > 0) - dcn3_02_soc.sr_enter_plus_exit_time_us = - bb_info.dram_sr_enter_exit_latency_100ns * 10; - - if (bb_info.dram_sr_exit_latency_100ns > 0) - dcn3_02_soc.sr_exit_time_us = - bb_info.dram_sr_exit_latency_100ns * 10; + DC_FP_START(); + dcn302_fpu_init_soc_bounding_box(bb_info); + DC_FP_END(); } } @@ -1257,170 +1120,11 @@ static void dcn302_destroy_resource_pool(struct resource_pool **pool) *pool = NULL; } -static void dcn302_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts, - unsigned int *optimal_dcfclk, - unsigned int *optimal_fclk) -{ - double bw_from_dram, bw_from_dram1, bw_from_dram2; - - bw_from_dram1 = uclk_mts * dcn3_02_soc.num_chans * - dcn3_02_soc.dram_channel_width_bytes * (dcn3_02_soc.max_avg_dram_bw_use_normal_percent / 100); - bw_from_dram2 = uclk_mts * dcn3_02_soc.num_chans * - dcn3_02_soc.dram_channel_width_bytes * (dcn3_02_soc.max_avg_sdp_bw_use_normal_percent / 100); - - bw_from_dram = (bw_from_dram1 < bw_from_dram2) ? bw_from_dram1 : bw_from_dram2; - - if (optimal_fclk) - *optimal_fclk = bw_from_dram / - (dcn3_02_soc.fabric_datapath_to_dcn_data_return_bytes * (dcn3_02_soc.max_avg_sdp_bw_use_normal_percent / 100)); - - if (optimal_dcfclk) - *optimal_dcfclk = bw_from_dram / - (dcn3_02_soc.return_bus_width_bytes * (dcn3_02_soc.max_avg_sdp_bw_use_normal_percent / 100)); -} - void dcn302_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { - unsigned int i, j; - unsigned int num_states = 0; - - unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0}; - unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0}; - unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0}; - unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0}; - - unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {694, 875, 1000, 1200}; - unsigned int num_dcfclk_sta_targets = 4; - unsigned int num_uclk_states; - - - if (dc->ctx->dc_bios->vram_info.num_chans) - dcn3_02_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans; - - if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) - dcn3_02_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; - - dcn3_02_soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; - dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; - - if (bw_params->clk_table.entries[0].memclk_mhz) { - int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, max_phyclk_mhz = 0; - - for (i = 0; i < MAX_NUM_DPM_LVL; i++) { - if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz) - max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; - if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz) - max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; - if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz) - max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; - if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz) - max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; - } - if (!max_dcfclk_mhz) - max_dcfclk_mhz = dcn3_02_soc.clock_limits[0].dcfclk_mhz; - if (!max_dispclk_mhz) - max_dispclk_mhz = dcn3_02_soc.clock_limits[0].dispclk_mhz; - if (!max_dppclk_mhz) - max_dppclk_mhz = dcn3_02_soc.clock_limits[0].dppclk_mhz; - if (!max_phyclk_mhz) - max_phyclk_mhz = dcn3_02_soc.clock_limits[0].phyclk_mhz; - - if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { - /* If max DCFCLK is greater than the max DCFCLK STA target, insert into the DCFCLK STA target array */ - dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz; - num_dcfclk_sta_targets++; - } else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { - /* If max DCFCLK is less than the max DCFCLK STA target, cap values and remove duplicates */ - for (i = 0; i < num_dcfclk_sta_targets; i++) { - if (dcfclk_sta_targets[i] > max_dcfclk_mhz) { - dcfclk_sta_targets[i] = max_dcfclk_mhz; - break; - } - } - /* Update size of array since we "removed" duplicates */ - num_dcfclk_sta_targets = i + 1; - } - - num_uclk_states = bw_params->clk_table.num_entries; - - /* Calculate optimal dcfclk for each uclk */ - for (i = 0; i < num_uclk_states; i++) { - dcn302_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16, - &optimal_dcfclk_for_uclk[i], NULL); - if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) { - optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz; - } - } - - /* Calculate optimal uclk for each dcfclk sta target */ - for (i = 0; i < num_dcfclk_sta_targets; i++) { - for (j = 0; j < num_uclk_states; j++) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { - optimal_uclk_for_dcfclk_sta_targets[i] = - bw_params->clk_table.entries[j].memclk_mhz * 16; - break; - } - } - } - - i = 0; - j = 0; - /* create the final dcfclk and uclk table */ - while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { - if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { - dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; - dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; - } else { - if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { - dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; - dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; - } else { - j = num_uclk_states; - } - } - } - - while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) { - dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; - dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; - } - - while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES && - optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { - dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; - dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; - } - - dcn3_02_soc.num_states = num_states; - for (i = 0; i < dcn3_02_soc.num_states; i++) { - dcn3_02_soc.clock_limits[i].state = i; - dcn3_02_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i]; - dcn3_02_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i]; - dcn3_02_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i]; - - /* Fill all states with max values of all other clocks */ - dcn3_02_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz; - dcn3_02_soc.clock_limits[i].dppclk_mhz = max_dppclk_mhz; - dcn3_02_soc.clock_limits[i].phyclk_mhz = max_phyclk_mhz; - /* Populate from bw_params for DTBCLK, SOCCLK */ - if (!bw_params->clk_table.entries[i].dtbclk_mhz && i > 0) - dcn3_02_soc.clock_limits[i].dtbclk_mhz = dcn3_02_soc.clock_limits[i-1].dtbclk_mhz; - else - dcn3_02_soc.clock_limits[i].dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; - if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0) - dcn3_02_soc.clock_limits[i].socclk_mhz = dcn3_02_soc.clock_limits[i-1].socclk_mhz; - else - dcn3_02_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz; - /* These clocks cannot come from bw_params, always fill from dcn3_02_soc[1] */ - /* FCLK, PHYCLK_D18, DSCCLK */ - dcn3_02_soc.clock_limits[i].phyclk_d18_mhz = dcn3_02_soc.clock_limits[0].phyclk_d18_mhz; - dcn3_02_soc.clock_limits[i].dscclk_mhz = dcn3_02_soc.clock_limits[0].dscclk_mhz; - } - /* re-init DML with updated bb */ - dml_init_instance(&dc->dml, &dcn3_02_soc, &dcn3_02_ip, DML_PROJECT_DCN30); - if (dc->current_state) - dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_02_soc, &dcn3_02_ip, DML_PROJECT_DCN30); - } + DC_FP_START(); + dcn302_fpu_update_bw_bounding_box(dc, bw_params); + DC_FP_END(); } static struct resource_funcs dcn302_res_pool_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.h b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.h index 42d2c73e30bc..9f24e73b92b3 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.h @@ -28,6 +28,9 @@ #include "core_types.h" +extern struct _vcs_dpi_ip_params_st dcn3_02_ip; +extern struct _vcs_dpi_soc_bounding_box_st dcn3_02_soc; + struct resource_pool *dcn302_create_resource_pool(const struct dc_init_data *init_data, struct dc *dc); void dcn302_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c index 2de687f64cf6..36649716e991 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c @@ -48,8 +48,8 @@ #include "sienna_cichlid_ip_offset.h" #include "dcn/dcn_3_0_3_offset.h" #include "dcn/dcn_3_0_3_sh_mask.h" -#include "dcn/dpcs_3_0_3_offset.h" -#include "dcn/dpcs_3_0_3_sh_mask.h" +#include "dpcs/dpcs_3_0_3_offset.h" +#include "dpcs/dpcs_3_0_3_sh_mask.h" #include "nbio/nbio_2_3_offset.h" #define DC_LOGGER_INIT(logger) diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c index ea4f8e06b07c..287a1066b547 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c @@ -121,7 +121,8 @@ static void dccg31_enable_dpstreamclk(struct dccg *dccg, int otg_inst) return; } if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpstream) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + DPSTREAMCLK_GATE_DISABLE, 1, DPSTREAMCLK_ROOT_GATE_DISABLE, 1); } @@ -130,8 +131,9 @@ static void dccg31_disable_dpstreamclk(struct dccg *dccg, int otg_inst) struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpstream) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, - DPSTREAMCLK_ROOT_GATE_DISABLE, 0); + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + DPSTREAMCLK_ROOT_GATE_DISABLE, 0, + DPSTREAMCLK_GATE_DISABLE, 0); switch (otg_inst) { case 0: @@ -180,7 +182,8 @@ void dccg31_enable_symclk32_se( switch (hpo_se_inst) { case 0: if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_SE0_GATE_DISABLE, 1, SYMCLK32_ROOT_SE0_GATE_DISABLE, 1); REG_UPDATE_2(SYMCLK32_SE_CNTL, SYMCLK32_SE0_SRC_SEL, phyd32clk, @@ -188,7 +191,8 @@ void dccg31_enable_symclk32_se( break; case 1: if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_SE1_GATE_DISABLE, 1, SYMCLK32_ROOT_SE1_GATE_DISABLE, 1); REG_UPDATE_2(SYMCLK32_SE_CNTL, SYMCLK32_SE1_SRC_SEL, phyd32clk, @@ -196,7 +200,8 @@ void dccg31_enable_symclk32_se( break; case 2: if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_SE2_GATE_DISABLE, 1, SYMCLK32_ROOT_SE2_GATE_DISABLE, 1); REG_UPDATE_2(SYMCLK32_SE_CNTL, SYMCLK32_SE2_SRC_SEL, phyd32clk, @@ -204,7 +209,8 @@ void dccg31_enable_symclk32_se( break; case 3: if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_SE3_GATE_DISABLE, 1, SYMCLK32_ROOT_SE3_GATE_DISABLE, 1); REG_UPDATE_2(SYMCLK32_SE_CNTL, SYMCLK32_SE3_SRC_SEL, phyd32clk, @@ -229,7 +235,8 @@ void dccg31_disable_symclk32_se( SYMCLK32_SE0_SRC_SEL, 0, SYMCLK32_SE0_EN, 0); if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_SE0_GATE_DISABLE, 0, SYMCLK32_ROOT_SE0_GATE_DISABLE, 0); break; case 1: @@ -237,7 +244,8 @@ void dccg31_disable_symclk32_se( SYMCLK32_SE1_SRC_SEL, 0, SYMCLK32_SE1_EN, 0); if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_SE1_GATE_DISABLE, 0, SYMCLK32_ROOT_SE1_GATE_DISABLE, 0); break; case 2: @@ -245,7 +253,8 @@ void dccg31_disable_symclk32_se( SYMCLK32_SE2_SRC_SEL, 0, SYMCLK32_SE2_EN, 0); if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_SE2_GATE_DISABLE, 0, SYMCLK32_ROOT_SE2_GATE_DISABLE, 0); break; case 3: @@ -253,7 +262,8 @@ void dccg31_disable_symclk32_se( SYMCLK32_SE3_SRC_SEL, 0, SYMCLK32_SE3_EN, 0); if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_SE3_GATE_DISABLE, 0, SYMCLK32_ROOT_SE3_GATE_DISABLE, 0); break; default: @@ -275,7 +285,8 @@ void dccg31_enable_symclk32_le( switch (hpo_le_inst) { case 0: if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_LE0_GATE_DISABLE, 1, SYMCLK32_ROOT_LE0_GATE_DISABLE, 1); REG_UPDATE_2(SYMCLK32_LE_CNTL, SYMCLK32_LE0_SRC_SEL, phyd32clk, @@ -283,7 +294,8 @@ void dccg31_enable_symclk32_le( break; case 1: if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_LE1_GATE_DISABLE, 1, SYMCLK32_ROOT_LE1_GATE_DISABLE, 1); REG_UPDATE_2(SYMCLK32_LE_CNTL, SYMCLK32_LE1_SRC_SEL, phyd32clk, @@ -308,7 +320,8 @@ void dccg31_disable_symclk32_le( SYMCLK32_LE0_SRC_SEL, 0, SYMCLK32_LE0_EN, 0); if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_LE0_GATE_DISABLE, 0, SYMCLK32_ROOT_LE0_GATE_DISABLE, 0); break; case 1: @@ -316,7 +329,8 @@ void dccg31_disable_symclk32_le( SYMCLK32_LE1_SRC_SEL, 0, SYMCLK32_LE1_EN, 0); if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL3, + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_LE1_GATE_DISABLE, 0, SYMCLK32_ROOT_LE1_GATE_DISABLE, 0); break; default: @@ -406,54 +420,89 @@ void dccg31_set_physymclk( /* Force PHYSYMCLK on and Select phyd32clk as the source of clock which is output to PHY through DCIO */ switch (phy_inst) { case 0: - if (force_enable) + if (force_enable) { REG_UPDATE_2(PHYASYMCLK_CLOCK_CNTL, PHYASYMCLK_FORCE_EN, 1, PHYASYMCLK_FORCE_SRC_SEL, clk_src); - else + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYASYMCLK_GATE_DISABLE, 1); + } else { REG_UPDATE_2(PHYASYMCLK_CLOCK_CNTL, PHYASYMCLK_FORCE_EN, 0, PHYASYMCLK_FORCE_SRC_SEL, 0); + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYASYMCLK_GATE_DISABLE, 0); + } break; case 1: - if (force_enable) + if (force_enable) { REG_UPDATE_2(PHYBSYMCLK_CLOCK_CNTL, PHYBSYMCLK_FORCE_EN, 1, PHYBSYMCLK_FORCE_SRC_SEL, clk_src); - else + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYBSYMCLK_GATE_DISABLE, 1); + } else { REG_UPDATE_2(PHYBSYMCLK_CLOCK_CNTL, PHYBSYMCLK_FORCE_EN, 0, PHYBSYMCLK_FORCE_SRC_SEL, 0); + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYBSYMCLK_GATE_DISABLE, 0); + } break; case 2: - if (force_enable) + if (force_enable) { REG_UPDATE_2(PHYCSYMCLK_CLOCK_CNTL, PHYCSYMCLK_FORCE_EN, 1, PHYCSYMCLK_FORCE_SRC_SEL, clk_src); - else + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYCSYMCLK_GATE_DISABLE, 1); + } else { REG_UPDATE_2(PHYCSYMCLK_CLOCK_CNTL, PHYCSYMCLK_FORCE_EN, 0, PHYCSYMCLK_FORCE_SRC_SEL, 0); + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYCSYMCLK_GATE_DISABLE, 0); + } break; case 3: - if (force_enable) + if (force_enable) { REG_UPDATE_2(PHYDSYMCLK_CLOCK_CNTL, PHYDSYMCLK_FORCE_EN, 1, PHYDSYMCLK_FORCE_SRC_SEL, clk_src); - else + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYDSYMCLK_GATE_DISABLE, 1); + } else { REG_UPDATE_2(PHYDSYMCLK_CLOCK_CNTL, PHYDSYMCLK_FORCE_EN, 0, PHYDSYMCLK_FORCE_SRC_SEL, 0); + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYDSYMCLK_GATE_DISABLE, 0); + } break; case 4: - if (force_enable) + if (force_enable) { REG_UPDATE_2(PHYESYMCLK_CLOCK_CNTL, PHYESYMCLK_FORCE_EN, 1, PHYESYMCLK_FORCE_SRC_SEL, clk_src); - else + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYESYMCLK_GATE_DISABLE, 1); + } else { REG_UPDATE_2(PHYESYMCLK_CLOCK_CNTL, PHYESYMCLK_FORCE_EN, 0, PHYESYMCLK_FORCE_SRC_SEL, 0); + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) + REG_UPDATE(DCCG_GATE_DISABLE_CNTL2, + PHYESYMCLK_GATE_DISABLE, 0); + } break; default: BREAK_TO_DEBUGGER(); @@ -615,6 +664,13 @@ void dccg31_init(struct dccg *dccg) dccg31_disable_dpstreamclk(dccg, 3); } + if (dccg->ctx->dc->debug.root_clock_optimization.bits.physymclk) { + dccg31_set_physymclk(dccg, 0, PHYSYMCLK_FORCE_SRC_SYMCLK, false); + dccg31_set_physymclk(dccg, 1, PHYSYMCLK_FORCE_SRC_SYMCLK, false); + dccg31_set_physymclk(dccg, 2, PHYSYMCLK_FORCE_SRC_SYMCLK, false); + dccg31_set_physymclk(dccg, 3, PHYSYMCLK_FORCE_SRC_SYMCLK, false); + dccg31_set_physymclk(dccg, 4, PHYSYMCLK_FORCE_SRC_SYMCLK, false); + } } static const struct dccg_funcs dccg31_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.h index a013a32bbaf7..269cabbea72a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.h @@ -66,6 +66,7 @@ SR(DSCCLK1_DTO_PARAM),\ SR(DSCCLK2_DTO_PARAM),\ SR(DSCCLK_DTO_CTRL),\ + SR(DCCG_GATE_DISABLE_CNTL2),\ SR(DCCG_GATE_DISABLE_CNTL3),\ SR(HDMISTREAMCLK0_DTO_PARAM) @@ -135,6 +136,11 @@ DCCG_SF(DSCCLK_DTO_CTRL, DSCCLK0_DTO_ENABLE, mask_sh),\ DCCG_SF(DSCCLK_DTO_CTRL, DSCCLK1_DTO_ENABLE, mask_sh),\ DCCG_SF(DSCCLK_DTO_CTRL, DSCCLK2_DTO_ENABLE, mask_sh),\ + DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYASYMCLK_GATE_DISABLE, mask_sh),\ + DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYBSYMCLK_GATE_DISABLE, mask_sh),\ + DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYCSYMCLK_GATE_DISABLE, mask_sh),\ + DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYDSYMCLK_GATE_DISABLE, mask_sh),\ + DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYESYMCLK_GATE_DISABLE, mask_sh),\ DCCG_SF(DCCG_GATE_DISABLE_CNTL3, DPSTREAMCLK_ROOT_GATE_DISABLE, mask_sh),\ DCCG_SF(DCCG_GATE_DISABLE_CNTL3, DPSTREAMCLK_GATE_DISABLE, mask_sh),\ DCCG_SF(DCCG_GATE_DISABLE_CNTL3, SYMCLK32_ROOT_SE0_GATE_DISABLE, mask_sh),\ diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.c index 5065904c7833..23621ff08c90 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.c @@ -710,6 +710,16 @@ static void dcn31_hpo_dp_stream_enc_read_state( } } +static void dcn31_set_hblank_min_symbol_width( + struct hpo_dp_stream_encoder *enc, + uint16_t width) +{ + struct dcn31_hpo_dp_stream_encoder *enc3 = DCN3_1_HPO_DP_STREAM_ENC_FROM_HPO_STREAM_ENC(enc); + + REG_SET(DP_SYM32_ENC_HBLANK_CONTROL, 0, + HBLANK_MINIMUM_SYMBOL_WIDTH, width); +} + static const struct hpo_dp_stream_encoder_funcs dcn30_str_enc_funcs = { .enable_stream = dcn31_hpo_dp_stream_enc_enable_stream, .dp_unblank = dcn31_hpo_dp_stream_enc_dp_unblank, @@ -725,6 +735,7 @@ static const struct hpo_dp_stream_encoder_funcs dcn30_str_enc_funcs = { .dp_audio_enable = dcn31_hpo_dp_stream_enc_audio_enable, .dp_audio_disable = dcn31_hpo_dp_stream_enc_audio_disable, .read_state = dcn31_hpo_dp_stream_enc_read_state, + .set_hblank_min_symbol_width = dcn31_set_hblank_min_symbol_width, }; void dcn31_hpo_dp_stream_encoder_construct( diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.h b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.h index 70b94fc25304..7c77c71591a0 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.h @@ -80,7 +80,8 @@ SRI(DP_SYM32_ENC_SDP_GSP_CONTROL11, DP_SYM32_ENC, id),\ SRI(DP_SYM32_ENC_SDP_METADATA_PACKET_CONTROL, DP_SYM32_ENC, id),\ SRI(DP_SYM32_ENC_SDP_AUDIO_CONTROL0, DP_SYM32_ENC, id),\ - SRI(DP_SYM32_ENC_VID_CRC_CONTROL, DP_SYM32_ENC, id) + SRI(DP_SYM32_ENC_VID_CRC_CONTROL, DP_SYM32_ENC, id), \ + SRI(DP_SYM32_ENC_HBLANK_CONTROL, DP_SYM32_ENC, id) #define DCN3_1_HPO_DP_STREAM_ENC_REGS \ uint32_t DP_STREAM_MAPPER_CONTROL0;\ @@ -116,7 +117,8 @@ uint32_t DP_SYM32_ENC_SDP_GSP_CONTROL11;\ uint32_t DP_SYM32_ENC_SDP_METADATA_PACKET_CONTROL;\ uint32_t DP_SYM32_ENC_SDP_AUDIO_CONTROL0;\ - uint32_t DP_SYM32_ENC_VID_CRC_CONTROL + uint32_t DP_SYM32_ENC_VID_CRC_CONTROL;\ + uint32_t DP_SYM32_ENC_HBLANK_CONTROL #define DCN3_1_HPO_DP_STREAM_ENC_MASK_SH_LIST(mask_sh)\ @@ -202,7 +204,8 @@ type GSP_SOF_REFERENCE;\ type METADATA_PACKET_ENABLE;\ type CRC_ENABLE;\ - type CRC_CONT_MODE_ENABLE + type CRC_CONT_MODE_ENABLE;\ + type HBLANK_MINIMUM_SYMBOL_WIDTH struct dcn31_hpo_dp_stream_encoder_registers { diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c index 4206ce5bf9a9..b2cfd277b913 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hwseq.c @@ -197,8 +197,7 @@ void dcn31_init_hw(struct dc *dc) dmub_enable_outbox_notification(dc); /* we want to turn off all dp displays before doing detection */ - if (dc->config.power_down_display_on_boot) - dc_link_blank_all_dp_displays(dc); + dc_link_blank_all_dp_displays(dc); /* If taking control over from VBIOS, we may want to optimize our first * mode set, so we need to skip powering down pipes until we know which @@ -206,7 +205,7 @@ void dcn31_init_hw(struct dc *dc) * Otherwise, if taking control is not possible, we need to power * everything down. */ - if (dcb->funcs->is_accelerated_mode(dcb) || dc->config.power_down_display_on_boot) { + if (dcb->funcs->is_accelerated_mode(dcb) || !dc->config.seamless_boot_edp_requested) { hws->funcs.init_pipes(dc, dc->current_state); if (dc->res_pool->hubbub->funcs->allow_self_refresh_control) dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub, diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_optc.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_optc.c index e8562fa11366..8afe2130d7c5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_optc.c @@ -161,7 +161,7 @@ static bool optc31_immediate_disable_crtc(struct timing_generator *optc) return true; } -static void optc31_set_drr( +void optc31_set_drr( struct timing_generator *optc, const struct drr_params *params) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_optc.h b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_optc.h index d8ef2f0d0c95..a37b16040c1d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_optc.h @@ -256,4 +256,6 @@ void dcn31_timing_generator_init(struct optc *optc1); +void optc31_set_drr(struct timing_generator *optc, const struct drr_params *params); + #endif /* __DC_OPTC_DCN31_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c index 8d64187478e4..1166748b85a7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c @@ -1033,6 +1033,7 @@ static const struct dc_debug_options debug_defaults_drv = { .optimize_edp_link_rate = true, .enable_sw_cntl_psr = true, .apply_vendor_specific_lttpr_wa = true, + .enable_z9_disable_interface = false }; static const struct dc_debug_options debug_defaults_diags = { @@ -1841,7 +1842,8 @@ static int dcn31_populate_dml_pipes_from_context( if (is_dual_plane(pipe->plane_state->format) && pipe->plane_state->src_rect.width <= 1920 && pipe->plane_state->src_rect.height <= 1080) { dc->config.enable_4to1MPC = true; - } else if (!is_dual_plane(pipe->plane_state->format)) { + } else if (!is_dual_plane(pipe->plane_state->format) && pipe->plane_state->src_rect.width <= 5120) { + /* Limit to 5k max to avoid forced pipe split when there is not enough detile for swath */ context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192; pipes[0].pipe.src.unbounded_req_mode = true; } diff --git a/drivers/gpu/drm/amd/display/dc/dm_helpers.h b/drivers/gpu/drm/amd/display/dc/dm_helpers.h index 7f94e3f70d7f..b71b5fb894e2 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_helpers.h +++ b/drivers/gpu/drm/amd/display/dc/dm_helpers.h @@ -113,7 +113,7 @@ bool dm_helpers_dp_mst_start_top_mgr( const struct dc_link *link, bool boot); -void dm_helpers_dp_mst_stop_top_mgr( +bool dm_helpers_dp_mst_stop_top_mgr( struct dc_context *ctx, struct dc_link *link); /** @@ -170,9 +170,9 @@ bool dm_helpers_dmub_outbox_interrupt_control(struct dc_context *ctx, bool enabl void dm_helpers_smu_timeout(struct dc_context *ctx, unsigned int msg_id, unsigned int param, unsigned int timeout_us); -// 0x1 = Result_OK, 0xFE = Result_UnkmownCmd +// 0x1 = Result_OK, 0xFE = Result_UnkmownCmd, 0x0 = Status_Busy #define IS_SMU_TIMEOUT(result) \ - (!(result == 0x1 || result == 0xFE)) + (result == 0x0) int dm_helper_dmub_aux_transfer_sync( struct dc_context *ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index eee6672bd32d..b16c492593e2 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -71,8 +71,12 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_rq_dlg_calc_30.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_mode_vba_31.o := $(dml_ccflags) $(frame_warn_flag) CFLAGS_$(AMDDALPATH)/dc/dml/dcn31/display_rq_dlg_calc_31.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn301/dcn301_fpu.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn302/dcn302_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dsc/rc_calc_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calcs.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_auto.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_ccflags) -Wno-tautological-compare CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn2x/dcn2x.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_rcflags) @@ -93,10 +97,14 @@ CFLAGS_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_ccflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dml1_display_rq_dlg_calc.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/display_rq_dlg_helpers.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calcs.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_auto.o := $(dml_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_rcflags) -DML = display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o \ +DML = calcs/dce_calcs.o calcs/custom_float.o calcs/bw_fixed.o ifdef CONFIG_DRM_AMD_DC_DCN +DML += display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o DML += dcn20/dcn20_fpu.o DML += display_mode_vba.o dcn20/display_rq_dlg_calc_20.o dcn20/display_mode_vba_20.o DML += dcn20/display_rq_dlg_calc_20v2.o dcn20/display_mode_vba_20v2.o @@ -104,7 +112,9 @@ DML += dcn21/display_rq_dlg_calc_21.o dcn21/display_mode_vba_21.o DML += dcn30/display_mode_vba_30.o dcn30/display_rq_dlg_calc_30.o DML += dcn31/display_mode_vba_31.o dcn31/display_rq_dlg_calc_31.o DML += dcn301/dcn301_fpu.o +DML += dcn302/dcn302_fpu.o DML += dsc/rc_calc_fpu.o +DML += calcs/dcn_calcs.o calcs/dcn_calc_math.o calcs/dcn_calc_auto.o endif AMD_DAL_DML = $(addprefix $(AMDDALPATH)/dc/dml/,$(DML)) diff --git a/drivers/gpu/drm/amd/display/dc/calcs/bw_fixed.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/bw_fixed.c index 6ca288fb5fb9..6ca288fb5fb9 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/bw_fixed.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/bw_fixed.c diff --git a/drivers/gpu/drm/amd/display/dc/calcs/calcs_logger.h b/drivers/gpu/drm/amd/display/dc/dml/calcs/calcs_logger.h index 62435bfc274d..62435bfc274d 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/calcs_logger.h +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/calcs_logger.h diff --git a/drivers/gpu/drm/amd/display/dc/calcs/custom_float.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/custom_float.c index 31d167bc548f..31d167bc548f 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/custom_float.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/custom_float.c diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/dce_calcs.c index e6ef36de0825..0100a6053ab6 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dce_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/dce_calcs.c @@ -3411,35 +3411,33 @@ bool bw_calcs(struct dc_context *ctx, calcs_output->stutter_exit_wm_ns[5].c_mark = bw_fixed_to_int(bw_mul(data-> stutter_exit_watermark[9], bw_int_to_fixed(1000))); - - calcs_output->stutter_entry_wm_ns[0].c_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[4], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[1].c_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[5], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[2].c_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[6], bw_int_to_fixed(1000))); - if (ctx->dc->caps.max_slave_planes) { - calcs_output->stutter_entry_wm_ns[3].c_mark = + calcs_output->stutter_entry_wm_ns[0].c_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[0], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[4].c_mark = + stutter_entry_watermark[4], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[1].c_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[1], bw_int_to_fixed(1000))); - } else { - calcs_output->stutter_entry_wm_ns[3].c_mark = + stutter_entry_watermark[5], bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[2].c_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[7], bw_int_to_fixed(1000))); - calcs_output->stutter_entry_wm_ns[4].c_mark = + stutter_entry_watermark[6], bw_int_to_fixed(1000))); + if (ctx->dc->caps.max_slave_planes) { + calcs_output->stutter_entry_wm_ns[3].c_mark = + bw_fixed_to_int(bw_mul(data->stutter_entry_watermark[0], + bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[4].c_mark = + bw_fixed_to_int(bw_mul(data->stutter_entry_watermark[1], + bw_int_to_fixed(1000))); + } else { + calcs_output->stutter_entry_wm_ns[3].c_mark = + bw_fixed_to_int(bw_mul(data->stutter_entry_watermark[7], + bw_int_to_fixed(1000))); + calcs_output->stutter_entry_wm_ns[4].c_mark = + bw_fixed_to_int(bw_mul(data->stutter_entry_watermark[8], + bw_int_to_fixed(1000))); + } + calcs_output->stutter_entry_wm_ns[5].c_mark = bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[8], bw_int_to_fixed(1000))); - } - calcs_output->stutter_entry_wm_ns[5].c_mark = - bw_fixed_to_int(bw_mul(data-> - stutter_entry_watermark[9], bw_int_to_fixed(1000))); - + stutter_entry_watermark[9], bw_int_to_fixed(1000))); calcs_output->urgent_wm_ns[0].c_mark = bw_fixed_to_int(bw_mul(data-> urgent_watermark[4], bw_int_to_fixed(1000))); diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_auto.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calc_auto.c index 41284e263325..41284e263325 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_auto.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calc_auto.c diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_auto.h b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calc_auto.h index ce35de79a6c7..ce35de79a6c7 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_auto.h +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calc_auto.h diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_math.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calc_math.c index 07d18e78de49..07d18e78de49 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calc_math.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calc_math.c diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c index e447c74be713..e447c74be713 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c new file mode 100644 index 000000000000..e2bcd205aa93 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c @@ -0,0 +1,357 @@ +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "resource.h" +#include "clk_mgr.h" +#include "dcn20/dcn20_resource.h" +#include "dcn302/dcn302_resource.h" + +#include "dml/dcn20/dcn20_fpu.h" +#include "dcn302_fpu.h" + +struct _vcs_dpi_ip_params_st dcn3_02_ip = { + .use_min_dcfclk = 0, + .clamp_min_dcfclk = 0, + .odm_capable = 1, + .gpuvm_enable = 1, + .hostvm_enable = 0, + .gpuvm_max_page_table_levels = 4, + .hostvm_max_page_table_levels = 4, + .hostvm_cached_page_table_levels = 0, + .pte_group_size_bytes = 2048, + .num_dsc = 5, + .rob_buffer_size_kbytes = 184, + .det_buffer_size_kbytes = 184, + .dpte_buffer_size_in_pte_reqs_luma = 64, + .dpte_buffer_size_in_pte_reqs_chroma = 34, + .pde_proc_buffer_size_64k_reqs = 48, + .dpp_output_buffer_pixels = 2560, + .opp_output_buffer_lines = 1, + .pixel_chunk_size_kbytes = 8, + .pte_enable = 1, + .max_page_table_levels = 2, + .pte_chunk_size_kbytes = 2, // ? + .meta_chunk_size_kbytes = 2, + .writeback_chunk_size_kbytes = 8, + .line_buffer_size_bits = 789504, + .is_line_buffer_bpp_fixed = 0, // ? + .line_buffer_fixed_bpp = 0, // ? + .dcc_supported = true, + .writeback_interface_buffer_size_kbytes = 90, + .writeback_line_buffer_buffer_size = 0, + .max_line_buffer_lines = 12, + .writeback_luma_buffer_size_kbytes = 12, // writeback_line_buffer_buffer_size = 656640 + .writeback_chroma_buffer_size_kbytes = 8, + .writeback_chroma_line_buffer_width_pixels = 4, + .writeback_max_hscl_ratio = 1, + .writeback_max_vscl_ratio = 1, + .writeback_min_hscl_ratio = 1, + .writeback_min_vscl_ratio = 1, + .writeback_max_hscl_taps = 1, + .writeback_max_vscl_taps = 1, + .writeback_line_buffer_luma_buffer_size = 0, + .writeback_line_buffer_chroma_buffer_size = 14643, + .cursor_buffer_size = 8, + .cursor_chunk_size = 2, + .max_num_otg = 5, + .max_num_dpp = 5, + .max_num_wb = 1, + .max_dchub_pscl_bw_pix_per_clk = 4, + .max_pscl_lb_bw_pix_per_clk = 2, + .max_lb_vscl_bw_pix_per_clk = 4, + .max_vscl_hscl_bw_pix_per_clk = 4, + .max_hscl_ratio = 6, + .max_vscl_ratio = 6, + .hscl_mults = 4, + .vscl_mults = 4, + .max_hscl_taps = 8, + .max_vscl_taps = 8, + .dispclk_ramp_margin_percent = 1, + .underscan_factor = 1.11, + .min_vblank_lines = 32, + .dppclk_delay_subtotal = 46, + .dynamic_metadata_vm_enabled = true, + .dppclk_delay_scl_lb_only = 16, + .dppclk_delay_scl = 50, + .dppclk_delay_cnvc_formatter = 27, + .dppclk_delay_cnvc_cursor = 6, + .dispclk_delay_subtotal = 119, + .dcfclk_cstate_latency = 5.2, // SRExitTime + .max_inter_dcn_tile_repeaters = 8, + .max_num_hdmi_frl_outputs = 1, + .odm_combine_4to1_supported = true, + + .xfc_supported = false, + .xfc_fill_bw_overhead_percent = 10.0, + .xfc_fill_constant_bytes = 0, + .gfx7_compat_tiling_supported = 0, + .number_of_cursors = 1, +}; + +struct _vcs_dpi_soc_bounding_box_st dcn3_02_soc = { + .clock_limits = { + { + .state = 0, + .dispclk_mhz = 562.0, + .dppclk_mhz = 300.0, + .phyclk_mhz = 300.0, + .phyclk_d18_mhz = 667.0, + .dscclk_mhz = 405.6, + }, + }, + + .min_dcfclk = 500.0, /* TODO: set this to actual min DCFCLK */ + .num_states = 1, + .sr_exit_time_us = 26.5, + .sr_enter_plus_exit_time_us = 31, + .urgent_latency_us = 4.0, + .urgent_latency_pixel_data_only_us = 4.0, + .urgent_latency_pixel_mixed_with_vm_data_us = 4.0, + .urgent_latency_vm_data_only_us = 4.0, + .urgent_out_of_order_return_per_channel_pixel_only_bytes = 4096, + .urgent_out_of_order_return_per_channel_pixel_and_vm_bytes = 4096, + .urgent_out_of_order_return_per_channel_vm_only_bytes = 4096, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_only = 80.0, + .pct_ideal_dram_sdp_bw_after_urgent_pixel_and_vm = 60.0, + .pct_ideal_dram_sdp_bw_after_urgent_vm_only = 40.0, + .max_avg_sdp_bw_use_normal_percent = 60.0, + .max_avg_dram_bw_use_normal_percent = 40.0, + .writeback_latency_us = 12.0, + .max_request_size_bytes = 256, + .fabric_datapath_to_dcn_data_return_bytes = 64, + .dcn_downspread_percent = 0.5, + .downspread_percent = 0.38, + .dram_page_open_time_ns = 50.0, + .dram_rw_turnaround_time_ns = 17.5, + .dram_return_buffer_per_channel_bytes = 8192, + .round_trip_ping_latency_dcfclk_cycles = 156, + .urgent_out_of_order_return_per_channel_bytes = 4096, + .channel_interleave_bytes = 256, + .num_banks = 8, + .gpuvm_min_page_size_bytes = 4096, + .hostvm_min_page_size_bytes = 4096, + .dram_clock_change_latency_us = 404, + .dummy_pstate_latency_us = 5, + .writeback_dram_clock_change_latency_us = 23.0, + .return_bus_width_bytes = 64, + .dispclk_dppclk_vco_speed_mhz = 3650, + .xfc_bus_transport_time_us = 20, // ? + .xfc_xbuf_latency_tolerance_us = 4, // ? + .use_urgent_burst_bw = 1, // ? + .do_urgent_latency_adjustment = true, + .urgent_latency_adjustment_fabric_clock_component_us = 1.0, + .urgent_latency_adjustment_fabric_clock_reference_mhz = 1000, +}; + +static void dcn302_get_optimal_dcfclk_fclk_for_uclk(unsigned int uclk_mts, + unsigned int *optimal_dcfclk, + unsigned int *optimal_fclk) +{ + + double bw_from_dram, bw_from_dram1, bw_from_dram2; + + bw_from_dram1 = uclk_mts * dcn3_02_soc.num_chans * + dcn3_02_soc.dram_channel_width_bytes * + (dcn3_02_soc.max_avg_dram_bw_use_normal_percent / 100); + bw_from_dram2 = uclk_mts * dcn3_02_soc.num_chans * + dcn3_02_soc.dram_channel_width_bytes * + (dcn3_02_soc.max_avg_sdp_bw_use_normal_percent / 100); + + bw_from_dram = (bw_from_dram1 < bw_from_dram2) ? bw_from_dram1 : bw_from_dram2; + + if (optimal_fclk) + *optimal_fclk = bw_from_dram / + (dcn3_02_soc.fabric_datapath_to_dcn_data_return_bytes * + (dcn3_02_soc.max_avg_sdp_bw_use_normal_percent / 100)); + + if (optimal_dcfclk) + *optimal_dcfclk = bw_from_dram / + (dcn3_02_soc.return_bus_width_bytes * + (dcn3_02_soc.max_avg_sdp_bw_use_normal_percent / 100)); +} + +void dcn302_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) +{ + unsigned int i, j; + unsigned int num_states = 0; + + unsigned int dcfclk_mhz[DC__VOLTAGE_STATES] = {0}; + unsigned int dram_speed_mts[DC__VOLTAGE_STATES] = {0}; + unsigned int optimal_uclk_for_dcfclk_sta_targets[DC__VOLTAGE_STATES] = {0}; + unsigned int optimal_dcfclk_for_uclk[DC__VOLTAGE_STATES] = {0}; + + unsigned int dcfclk_sta_targets[DC__VOLTAGE_STATES] = {694, 875, 1000, 1200}; + unsigned int num_dcfclk_sta_targets = 4; + unsigned int num_uclk_states; + + dc_assert_fp_enabled(); + + if (dc->ctx->dc_bios->vram_info.num_chans) + dcn3_02_soc.num_chans = dc->ctx->dc_bios->vram_info.num_chans; + + if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) + dcn3_02_soc.dram_channel_width_bytes = dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; + + dcn3_02_soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; + dc->dml.soc.dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; + + if (bw_params->clk_table.entries[0].memclk_mhz) { + int max_dcfclk_mhz = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0, max_phyclk_mhz = 0; + + for (i = 0; i < MAX_NUM_DPM_LVL; i++) { + if (bw_params->clk_table.entries[i].dcfclk_mhz > max_dcfclk_mhz) + max_dcfclk_mhz = bw_params->clk_table.entries[i].dcfclk_mhz; + if (bw_params->clk_table.entries[i].dispclk_mhz > max_dispclk_mhz) + max_dispclk_mhz = bw_params->clk_table.entries[i].dispclk_mhz; + if (bw_params->clk_table.entries[i].dppclk_mhz > max_dppclk_mhz) + max_dppclk_mhz = bw_params->clk_table.entries[i].dppclk_mhz; + if (bw_params->clk_table.entries[i].phyclk_mhz > max_phyclk_mhz) + max_phyclk_mhz = bw_params->clk_table.entries[i].phyclk_mhz; + } + if (!max_dcfclk_mhz) + max_dcfclk_mhz = dcn3_02_soc.clock_limits[0].dcfclk_mhz; + if (!max_dispclk_mhz) + max_dispclk_mhz = dcn3_02_soc.clock_limits[0].dispclk_mhz; + if (!max_dppclk_mhz) + max_dppclk_mhz = dcn3_02_soc.clock_limits[0].dppclk_mhz; + if (!max_phyclk_mhz) + max_phyclk_mhz = dcn3_02_soc.clock_limits[0].phyclk_mhz; + + if (max_dcfclk_mhz > dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { + /* If max DCFCLK is greater than the max DCFCLK STA target, insert into the DCFCLK STA target array */ + dcfclk_sta_targets[num_dcfclk_sta_targets] = max_dcfclk_mhz; + num_dcfclk_sta_targets++; + } else if (max_dcfclk_mhz < dcfclk_sta_targets[num_dcfclk_sta_targets-1]) { + /* If max DCFCLK is less than the max DCFCLK STA target, cap values and remove duplicates */ + for (i = 0; i < num_dcfclk_sta_targets; i++) { + if (dcfclk_sta_targets[i] > max_dcfclk_mhz) { + dcfclk_sta_targets[i] = max_dcfclk_mhz; + break; + } + } + /* Update size of array since we "removed" duplicates */ + num_dcfclk_sta_targets = i + 1; + } + + num_uclk_states = bw_params->clk_table.num_entries; + + /* Calculate optimal dcfclk for each uclk */ + for (i = 0; i < num_uclk_states; i++) { + dcn302_get_optimal_dcfclk_fclk_for_uclk(bw_params->clk_table.entries[i].memclk_mhz * 16, + &optimal_dcfclk_for_uclk[i], NULL); + if (optimal_dcfclk_for_uclk[i] < bw_params->clk_table.entries[0].dcfclk_mhz) + optimal_dcfclk_for_uclk[i] = bw_params->clk_table.entries[0].dcfclk_mhz; + } + + /* Calculate optimal uclk for each dcfclk sta target */ + for (i = 0; i < num_dcfclk_sta_targets; i++) { + for (j = 0; j < num_uclk_states; j++) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) { + optimal_uclk_for_dcfclk_sta_targets[i] = + bw_params->clk_table.entries[j].memclk_mhz * 16; + break; + } + } + } + + i = 0; + j = 0; + /* create the final dcfclk and uclk table */ + while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) { + if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) { + dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; + dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; + } else { + if (j < num_uclk_states && optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { + dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; + dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; + } else { + j = num_uclk_states; + } + } + } + + while (i < num_dcfclk_sta_targets && num_states < DC__VOLTAGE_STATES) { + dcfclk_mhz[num_states] = dcfclk_sta_targets[i]; + dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++]; + } + + while (j < num_uclk_states && num_states < DC__VOLTAGE_STATES && + optimal_dcfclk_for_uclk[j] <= max_dcfclk_mhz) { + dcfclk_mhz[num_states] = optimal_dcfclk_for_uclk[j]; + dram_speed_mts[num_states++] = bw_params->clk_table.entries[j++].memclk_mhz * 16; + } + + dcn3_02_soc.num_states = num_states; + for (i = 0; i < dcn3_02_soc.num_states; i++) { + dcn3_02_soc.clock_limits[i].state = i; + dcn3_02_soc.clock_limits[i].dcfclk_mhz = dcfclk_mhz[i]; + dcn3_02_soc.clock_limits[i].fabricclk_mhz = dcfclk_mhz[i]; + dcn3_02_soc.clock_limits[i].dram_speed_mts = dram_speed_mts[i]; + + /* Fill all states with max values of all other clocks */ + dcn3_02_soc.clock_limits[i].dispclk_mhz = max_dispclk_mhz; + dcn3_02_soc.clock_limits[i].dppclk_mhz = max_dppclk_mhz; + dcn3_02_soc.clock_limits[i].phyclk_mhz = max_phyclk_mhz; + /* Populate from bw_params for DTBCLK, SOCCLK */ + if (!bw_params->clk_table.entries[i].dtbclk_mhz && i > 0) + dcn3_02_soc.clock_limits[i].dtbclk_mhz = dcn3_02_soc.clock_limits[i-1].dtbclk_mhz; + else + dcn3_02_soc.clock_limits[i].dtbclk_mhz = bw_params->clk_table.entries[i].dtbclk_mhz; + if (!bw_params->clk_table.entries[i].socclk_mhz && i > 0) + dcn3_02_soc.clock_limits[i].socclk_mhz = dcn3_02_soc.clock_limits[i-1].socclk_mhz; + else + dcn3_02_soc.clock_limits[i].socclk_mhz = bw_params->clk_table.entries[i].socclk_mhz; + /* These clocks cannot come from bw_params, always fill from dcn3_02_soc[1] */ + /* FCLK, PHYCLK_D18, DSCCLK */ + dcn3_02_soc.clock_limits[i].phyclk_d18_mhz = dcn3_02_soc.clock_limits[0].phyclk_d18_mhz; + dcn3_02_soc.clock_limits[i].dscclk_mhz = dcn3_02_soc.clock_limits[0].dscclk_mhz; + } + /* re-init DML with updated bb */ + dml_init_instance(&dc->dml, &dcn3_02_soc, &dcn3_02_ip, DML_PROJECT_DCN30); + if (dc->current_state) + dml_init_instance(&dc->current_state->bw_ctx.dml, &dcn3_02_soc, &dcn3_02_ip, DML_PROJECT_DCN30); + } +} + +void dcn302_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info) +{ + + dc_assert_fp_enabled(); + + if (bb_info.dram_clock_change_latency_100ns > 0) + dcn3_02_soc.dram_clock_change_latency_us = + bb_info.dram_clock_change_latency_100ns * 10; + + if (bb_info.dram_sr_enter_exit_latency_100ns > 0) + dcn3_02_soc.sr_enter_plus_exit_time_us = + bb_info.dram_sr_enter_exit_latency_100ns * 10; + + if (bb_info.dram_sr_exit_latency_100ns > 0) + dcn3_02_soc.sr_exit_time_us = + bb_info.dram_sr_exit_latency_100ns * 10; +} + + diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.h new file mode 100644 index 000000000000..548305d96cee --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.h @@ -0,0 +1,32 @@ +/* + * Copyright 2019-2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DCN302_FPU_H__ +#define __DCN302_FPU_H__ + +void dcn302_fpu_init_soc_bounding_box(struct bp_soc_bb_info bb_info); +void dcn302_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); + +#endif /* __DCN302_FPU_H__*/ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c index 6feb23432f8d..e4b9fd31223c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c @@ -64,6 +64,8 @@ typedef struct { double DCFCLKDeepSleep; unsigned int DPPPerPlane; bool ScalerEnabled; + double VRatio; + double VRatioChroma; enum scan_direction_class SourceScan; unsigned int BlockWidth256BytesY; unsigned int BlockHeight256BytesY; @@ -942,6 +944,7 @@ static bool CalculatePrefetchSchedule( double dst_y_prefetch_equ; double Tsw_oto; double prefetch_bw_oto; + double prefetch_bw_pr; double Tvm_oto; double Tr0_oto; double Tvm_oto_lines; @@ -971,6 +974,7 @@ static bool CalculatePrefetchSchedule( double min_Lsw; double Tsw_est1 = 0; double Tsw_est3 = 0; + double max_Tsw = 0; if (GPUVMEnable == true && HostVMEnable == true) { HostVMDynamicLevelsTrips = HostVMMaxNonCachedPageTableLevels; @@ -1111,11 +1115,14 @@ static bool CalculatePrefetchSchedule( bytes_pp = myPipe->BytePerPixelY + myPipe->BytePerPixelC / 4; else bytes_pp = myPipe->BytePerPixelY + myPipe->BytePerPixelC; - + /*rev 99*/ + prefetch_bw_pr = dml_min(1, bytes_pp * myPipe->PixelClock / (double) myPipe->DPPPerPlane); + max_Tsw = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime; prefetch_sw_bytes = PrefetchSourceLinesY * swath_width_luma_ub * myPipe->BytePerPixelY + PrefetchSourceLinesC * swath_width_chroma_ub * myPipe->BytePerPixelC; prefetch_bw_oto = dml_max(bytes_pp * myPipe->PixelClock / myPipe->DPPPerPlane, prefetch_sw_bytes / (dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) * LineTime)); + prefetch_bw_oto = dml_max(prefetch_bw_pr, prefetch_sw_bytes / max_Tsw); - min_Lsw = dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) / max_vratio_pre; + min_Lsw = dml_max(1, dml_max(PrefetchSourceLinesY, PrefetchSourceLinesC) / max_vratio_pre); Lsw_oto = dml_ceil(4 * dml_max(prefetch_sw_bytes / prefetch_bw_oto / LineTime, min_Lsw), 1) / 4; Tsw_oto = Lsw_oto * LineTime; @@ -1389,7 +1396,7 @@ static bool CalculatePrefetchSchedule( dml_print("DML::%s: SwathHeightC = %d\n", __func__, SwathHeightC); dml_print("DML::%s: VInitPreFillC = %f\n", __func__, VInitPreFillC); #endif - if ((SwathHeightC > 4)) { + if ((SwathHeightC > 4) || VInitPreFillC > 3) { if (LinesToRequestPrefetchPixelData > (VInitPreFillC - 3.0) / 2.0) { *VRatioPrefetchC = dml_max( *VRatioPrefetchC, @@ -2663,6 +2670,8 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman myPipe.DCFCLKDeepSleep = v->DCFCLKDeepSleep; myPipe.DPPPerPlane = v->DPPPerPlane[k]; myPipe.ScalerEnabled = v->ScalerEnabled[k]; + myPipe.VRatio = v->VRatio[k]; + myPipe.VRatioChroma = v->VRatioChroma[k]; myPipe.SourceScan = v->SourceScan[k]; myPipe.BlockWidth256BytesY = v->BlockWidth256BytesY[k]; myPipe.BlockHeight256BytesY = v->BlockHeight256BytesY[k]; @@ -3911,6 +3920,9 @@ static noinline void CalculatePrefetchSchedulePerPlane( myPipe.DCFCLKDeepSleep = v->ProjectedDCFCLKDeepSleep[i][j]; myPipe.DPPPerPlane = v->NoOfDPP[i][j][k]; myPipe.ScalerEnabled = v->ScalerEnabled[k]; + myPipe.VRatio = mode_lib->vba.VRatio[k]; + myPipe.VRatioChroma = mode_lib->vba.VRatioChroma[k]; + myPipe.SourceScan = v->SourceScan[k]; myPipe.BlockWidth256BytesY = v->Read256BlockWidthY[k]; myPipe.BlockHeight256BytesY = v->Read256BlockHeightY[k]; @@ -4987,6 +4999,17 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l &v->meta_row_bandwidth[i][j][k], &v->dpte_row_bandwidth[i][j][k]); } + /*DCCMetaBufferSizeSupport(i, j) = True + For k = 0 To NumberOfActivePlanes - 1 + If MetaRowBytes(i, j, k) > 24064 Then + DCCMetaBufferSizeSupport(i, j) = False + End If + Next k*/ + v->DCCMetaBufferSizeSupport[i][j] = true; + for (k = 0; k < v->NumberOfActivePlanes; ++k) { + if (v->MetaRowBytes[i][j][k] > 24064) + v->DCCMetaBufferSizeSupport[i][j] = false; + } v->UrgLatency[i] = CalculateUrgentLatency( v->UrgentLatencyPixelDataOnly, v->UrgentLatencyPixelMixedWithVMData, diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c index 0fad15020c74..c0740dbdcc2e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.c @@ -47,6 +47,7 @@ static void recalculate_params( unsigned int num_pipes); static unsigned int CursorBppEnumToBits(enum cursor_bpp ebpp); +static void cache_debug_params(struct display_mode_lib *mode_lib); unsigned int dml_get_voltage_level( struct display_mode_lib *mode_lib, @@ -73,6 +74,7 @@ unsigned int dml_get_voltage_level( PixelClockAdjustmentForProgressiveToInterlaceUnit(mode_lib); } mode_lib->funcs.validate(mode_lib); + cache_debug_params(mode_lib); return mode_lib->vba.VoltageLevel; } @@ -745,6 +747,28 @@ static void fetch_pipe_params(struct display_mode_lib *mode_lib) mode_lib->vba.HostVMEnable = mode_lib->vba.HostVMEnable && !!ip->hostvm_enable; } +/** + * ******************************************************************************************** + * cache_debug_params: Cache any params that needed to be maintained from the initial validation + * for debug purposes. + * + * The DML getters can modify some of the VBA params that we are interested in (for example when + * calculating with dummy p-state latency), so cache any params here that we want for debugging + * + * @param [in] mode_lib: mode_lib input/output of validate call + * + * @return: void + * + * ******************************************************************************************** + */ +static void cache_debug_params(struct display_mode_lib *mode_lib) +{ + int k = 0; + + for (k = 0; k < mode_lib->vba.NumberOfActivePlanes; k++) + mode_lib->vba.CachedActiveDRAMClockChangeLatencyMargin[k] = mode_lib->vba.ActiveDRAMClockChangeLatencyMargin[k]; +} + // in wm mode we pull the parameters needed from the display_e2e_pipe_params_st structs // rather than working them out as in recalculate_ms static void recalculate_params( diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h index 90e87961fe3e..0603b32971a6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h +++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_vba.h @@ -195,6 +195,7 @@ struct vba_vars_st { unsigned int LBLatencyHidingSourceLinesY; unsigned int LBLatencyHidingSourceLinesC; double ActiveDRAMClockChangeLatencyMargin[DC__NUM_DPP__MAX]; + double CachedActiveDRAMClockChangeLatencyMargin[DC__NUM_DPP__MAX]; // Cache in dml_get_voltage_level for debug purposes only double MinActiveDRAMClockChangeMargin; double InitFillLevel; double FinalFillMargin; @@ -544,6 +545,8 @@ struct vba_vars_st { bool DTBCLKRequiredMoreThanSupported[DC__VOLTAGE_STATES]; double UrgentRoundTripAndOutOfOrderLatencyPerState[DC__VOLTAGE_STATES]; bool ROBSupport[DC__VOLTAGE_STATES][2]; + //based on rev 99: Dim DCCMetaBufferSizeSupport(NumberOfStates, 1) As Boolean + bool DCCMetaBufferSizeSupport[DC__VOLTAGE_STATES][2]; bool PTEBufferSizeNotExceeded[DC__VOLTAGE_STATES][2]; bool TotalVerticalActiveBandwidthSupport[DC__VOLTAGE_STATES][2]; double MaxTotalVerticalActiveAvailableBandwidth[DC__VOLTAGE_STATES][2]; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.c index ec636d06e18c..ef75eb7d5adc 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.c @@ -68,7 +68,7 @@ static void get_qp_set(qp_set qps, enum colour_mode cm, enum bits_per_comp bpc, int sel = table_hash(mode, bpc, max_min); int table_size = 0; int index; - const struct qp_entry *table = 0L; + const struct qp_entry *table = NULL; // alias enum enum { min = DAL_MM_MIN, max = DAL_MM_MAX }; diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c index 9c74564cbd8d..efc2339f1fa0 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c @@ -25,7 +25,7 @@ #include <drm/drm_dsc.h> #include "dc_hw_types.h" #include "dsc.h" -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include "dc.h" #include "rc_calc.h" #include "fixed31_32.h" diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c index 5f6ae3edb755..3b7df1ac26be 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_factory_dcn30.c @@ -42,8 +42,8 @@ #include "nbio/nbio_7_4_offset.h" -#include "dcn/dpcs_3_0_0_offset.h" -#include "dcn/dpcs_3_0_0_sh_mask.h" +#include "dpcs/dpcs_3_0_0_offset.h" +#include "dpcs/dpcs_3_0_0_sh_mask.h" #include "mmhub/mmhub_2_0_0_offset.h" #include "mmhub/mmhub_2_0_0_sh_mask.h" diff --git a/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_translate_dcn30.c b/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_translate_dcn30.c index 0046219a1cc7..6b6b7c7bd12f 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_translate_dcn30.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/dcn30/hw_translate_dcn30.c @@ -40,8 +40,8 @@ #include "nbio/nbio_7_4_offset.h" -#include "dcn/dpcs_3_0_0_offset.h" -#include "dcn/dpcs_3_0_0_sh_mask.h" +#include "dpcs/dpcs_3_0_0_offset.h" +#include "dpcs/dpcs_3_0_0_sh_mask.h" #include "mmhub/mmhub_2_0_0_offset.h" #include "mmhub/mmhub_2_0_0_sh_mask.h" diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index 943240e2809e..951c9b60917d 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -54,6 +54,7 @@ void enable_surface_flip_reporting(struct dc_plane_state *plane_state, #ifdef CONFIG_DRM_AMD_DC_HDCP #include "dm_cp_psp.h" #endif +#include "link_hwss.h" /************ link *****************/ struct link_init_data { @@ -249,12 +250,10 @@ struct resource_pool { /* Number of USB4 DPIA (DisplayPort Input Adapter) link objects created.*/ unsigned int usb4_dpia_count; -#if defined(CONFIG_DRM_AMD_DC_DCN) unsigned int hpo_dp_stream_enc_count; struct hpo_dp_stream_encoder *hpo_dp_stream_enc[MAX_HPO_DP2_ENCODERS]; unsigned int hpo_dp_link_enc_count; struct hpo_dp_link_encoder *hpo_dp_link_enc[MAX_HPO_DP2_LINK_ENCODERS]; -#endif #if defined(CONFIG_DRM_AMD_DC_DCN) struct dc_3dlut *mpc_lut[MAX_PIPES]; struct dc_transfer_func *mpc_shaper[MAX_PIPES]; @@ -307,9 +306,7 @@ struct stream_resource { struct display_stream_compressor *dsc; struct timing_generator *tg; struct stream_encoder *stream_enc; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct hpo_dp_stream_encoder *hpo_dp_stream_enc; -#endif struct audio *audio; struct pixel_clk_params pix_clk_params; @@ -334,18 +331,12 @@ struct plane_resource { struct dcn_fe_bandwidth bw; }; -#if defined(CONFIG_DRM_AMD_DC_DCN) #define LINK_RES_HPO_DP_REC_MAP__MASK 0xFFFF #define LINK_RES_HPO_DP_REC_MAP__SHIFT 0 -#endif /* all mappable hardware resources used to enable a link */ struct link_resource { -#if defined(CONFIG_DRM_AMD_DC_DCN) struct hpo_dp_link_encoder *hpo_dp_link_enc; -#else - void *dummy; -#endif }; union pipe_update_flags { @@ -425,11 +416,9 @@ struct resource_context { uint8_t dp_clock_source_ref_count; bool is_dsc_acquired[MAX_PIPES]; struct link_enc_cfg_context link_enc_cfg_ctx; -#if defined(CONFIG_DRM_AMD_DC_DCN) bool is_hpo_dp_stream_enc_acquired[MAX_HPO_DP2_ENCODERS]; unsigned int hpo_dp_link_enc_to_link_idx[MAX_HPO_DP2_LINK_ENCODERS]; int hpo_dp_link_enc_ref_cnts[MAX_HPO_DP2_LINK_ENCODERS]; -#endif #if defined(CONFIG_DRM_AMD_DC_DCN) bool is_mpc_3dlut_acquired[MAX_PIPES]; #endif diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h index cd52813a8432..477c4d9a972f 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h @@ -54,22 +54,13 @@ enum { PEAK_FACTOR_X1000 = 1006, }; -bool dp_verify_link_cap( - struct dc_link *link, - const struct link_resource *link_res, - struct dc_link_settings *known_limit_link_setting, - int *fail_count); +struct dc_link_settings dp_get_max_link_cap(struct dc_link *link); bool dp_verify_link_cap_with_retries( struct dc_link *link, - const struct link_resource *link_res, struct dc_link_settings *known_limit_link_setting, int attempts); -bool dp_verify_mst_link_cap( - struct dc_link *link, - const struct link_resource *link_res); - bool dp_validate_mode_timing( struct dc_link *link, const struct dc_crtc_timing *timing); @@ -114,6 +105,9 @@ void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode); bool dp_overwrite_extended_receiver_cap(struct dc_link *link); void dpcd_set_source_specific_data(struct dc_link *link); + +void dpcd_update_cable_id(struct dc_link *link); + /* Write DPCD link configuration data. */ enum dc_status dpcd_set_link_settings( struct dc_link *link, @@ -173,7 +167,6 @@ uint8_t dc_dp_initialize_scrambling_data_symbols( enum dc_status dp_set_fec_ready(struct dc_link *link, const struct link_resource *link_res, bool ready); void dp_set_fec_enable(struct dc_link *link, bool enable); -struct link_encoder *dp_get_link_enc(struct dc_link *link); bool dp_set_dsc_enable(struct pipe_ctx *pipe_ctx, bool enable); bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable, bool immediate_update); void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable); @@ -222,8 +215,47 @@ void disable_dp_hpo_output(struct dc_link *link, enum signal_type signal); void setup_dp_hpo_stream(struct pipe_ctx *pipe_ctx, bool enable); bool is_dp_128b_132b_signal(struct pipe_ctx *pipe_ctx); -void reset_dp_hpo_stream_encoders_for_link(struct dc_link *link); bool dp_retrieve_lttpr_cap(struct dc_link *link); void edp_panel_backlight_power_on(struct dc_link *link); +void dp_receiver_power_ctrl(struct dc_link *link, bool on); +void dp_source_sequence_trace(struct dc_link *link, uint8_t dp_test_mode); +void dp_enable_link_phy( + struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings); +void edp_add_delay_for_T9(struct dc_link *link); +bool edp_receiver_ready_T9(struct dc_link *link); +bool edp_receiver_ready_T7(struct dc_link *link); + +void dp_disable_link_phy(struct dc_link *link, const struct link_resource *link_res, + enum signal_type signal); + +void dp_disable_link_phy_mst(struct dc_link *link, const struct link_resource *link_res, + enum signal_type signal); + +bool dp_set_hw_training_pattern( + struct dc_link *link, + const struct link_resource *link_res, + enum dc_dp_training_pattern pattern, + uint32_t offset); + +void dp_set_hw_lane_settings( + struct dc_link *link, + const struct link_resource *link_res, + const struct link_training_settings *link_settings, + uint32_t offset); + +void dp_set_hw_test_pattern( + struct dc_link *link, + const struct link_resource *link_res, + enum dp_test_pattern test_pattern, + uint8_t *custom_pattern, + uint32_t custom_pattern_size); + +void dp_retrain_link_dp_test(struct dc_link *link, + struct dc_link_settings *link_setting, + bool skip_video_pattern); #endif /* __DC_LINK_DP_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h index c940fdfda144..b2fa4de47734 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h @@ -79,7 +79,7 @@ struct dccg_funcs { void (*otg_drop_pixel)(struct dccg *dccg, uint32_t otg_inst); void (*dccg_init)(struct dccg *dccg); -#if defined(CONFIG_DRM_AMD_DC_DCN) + void (*set_dpstreamclk)( struct dccg *dccg, enum hdmistreamclk_source src, @@ -102,7 +102,7 @@ struct dccg_funcs { void (*disable_symclk32_le)( struct dccg *dccg, int hpo_le_inst); -#endif + void (*set_physymclk)( struct dccg *dccg, int phy_inst, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h index 10ecbc667ffa..d89bd55f110f 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h @@ -38,10 +38,8 @@ #define MAX_PIPES 6 #define MAX_DIG_LINK_ENCODERS 7 #define MAX_DWB_PIPES 1 -#if defined(CONFIG_DRM_AMD_DC_DCN) #define MAX_HPO_DP2_ENCODERS 4 #define MAX_HPO_DP2_LINK_ENCODERS 2 -#endif struct gamma_curve { uint32_t offset; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h index 2ce15cd10d80..2013a70603ae 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h @@ -162,7 +162,8 @@ struct link_encoder_funcs { void (*disable_output)(struct link_encoder *link_enc, enum signal_type signal); void (*dp_set_lane_settings)(struct link_encoder *enc, - const struct link_training_settings *link_settings); + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]); void (*dp_set_phy_pattern)(struct link_encoder *enc, const struct encoder_set_dp_phy_pattern_param *para); void (*update_mst_stream_allocation_table)( @@ -220,7 +221,6 @@ enum link_enc_cfg_mode { LINK_ENC_CFG_TRANSIENT /* During commit state - use state to be committed. */ }; -#if defined(CONFIG_DRM_AMD_DC_DCN) enum dp2_link_mode { DP2_LINK_TRAINING_TPS1, DP2_LINK_TRAINING_TPS2, @@ -306,6 +306,5 @@ struct hpo_dp_link_encoder_funcs { const struct dc_link_settings *link_settings, uint8_t ffe_preset); }; -#endif #endif /* LINK_ENCODER_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h index c88e113b94d1..678c2065e5e8 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h @@ -245,7 +245,6 @@ struct stream_encoder_funcs { struct stream_encoder *enc); }; -#if defined(CONFIG_DRM_AMD_DC_DCN) struct hpo_dp_stream_encoder_state { uint32_t stream_enc_enabled; uint32_t vid_stream_enabled; @@ -324,7 +323,10 @@ struct hpo_dp_stream_encoder_funcs { void (*read_state)( struct hpo_dp_stream_encoder *enc, struct hpo_dp_stream_encoder_state *state); + + void (*set_hblank_min_symbol_width)( + struct hpo_dp_stream_encoder *enc, + uint16_t width); }; -#endif #endif /* STREAM_ENCODER_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h index c29320b3855d..59a704781e34 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h @@ -100,9 +100,7 @@ enum crc_selection { enum otg_out_mux_dest { OUT_MUX_DIO = 0, -#if defined(CONFIG_DRM_AMD_DC_DCN) OUT_MUX_HPO_DP = 2, -#endif }; enum h_timing_div_mode { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h index c2008258c50a..280c8764b636 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h @@ -41,9 +41,7 @@ struct dce_hwseq_wa { bool DEGVIDCN10_254; bool DEGVIDCN21; bool disallow_self_refresh_during_multi_plane_transition; -#if defined(CONFIG_DRM_AMD_DC_DCN) bool dp_hpo_and_otg_sequence; -#endif }; struct hwseq_wa_state { diff --git a/drivers/gpu/drm/amd/display/dc/inc/link_dpcd.h b/drivers/gpu/drm/amd/display/dc/inc/link_dpcd.h index 3f12b1600d2a..d561f86d503c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link_dpcd.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link_dpcd.h @@ -27,7 +27,7 @@ #define __LINK_DPCD_H__ #include <inc/core_status.h> #include <dc_link.h> -#include <inc/link_hwss.h> +#include <dc_link_dp.h> enum dc_status core_link_read_dpcd( struct dc_link *link, diff --git a/drivers/gpu/drm/amd/display/dc/inc/link_enc_cfg.h b/drivers/gpu/drm/amd/display/dc/inc/link_enc_cfg.h index a4e43b4826e0..c2f08514a1d9 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link_enc_cfg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link_enc_cfg.h @@ -96,6 +96,9 @@ struct link_encoder *link_enc_cfg_get_link_enc_used_by_stream( struct dc *dc, const struct dc_stream_state *stream); +/* Return DIG link encoder. NULL if unused. */ +struct link_encoder *link_enc_cfg_get_link_enc(const struct dc_link *link); + /* Return true if encoder available to use. */ bool link_enc_cfg_is_link_enc_avail(struct dc *dc, enum engine_id eng_id, struct dc_link *link); diff --git a/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h b/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h index 69d63763a10e..3b3090e3d327 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link_hwss.h @@ -26,50 +26,52 @@ #ifndef __DC_LINK_HWSS_H__ #define __DC_LINK_HWSS_H__ -struct gpio *get_hpd_gpio(struct dc_bios *dcb, - struct graphics_object_id link_id, - struct gpio_service *gpio_service); +/* include basic type headers only */ +#include "dc_dp_types.h" +#include "signal_types.h" +#include "grph_object_id.h" +#include "fixed31_32.h" -void dp_enable_link_phy( - struct dc_link *link, - const struct link_resource *link_res, - enum signal_type signal, - enum clock_source_id clock_source, - const struct dc_link_settings *link_settings); +/* forward declare dc core types */ +struct dc_link; +struct link_resource; +struct pipe_ctx; +struct encoder_set_dp_phy_pattern_param; -void dp_receiver_power_ctrl(struct dc_link *link, bool on); -void dp_source_sequence_trace(struct dc_link *link, uint8_t dp_test_mode); -void edp_add_delay_for_T9(struct dc_link *link); -bool edp_receiver_ready_T9(struct dc_link *link); -bool edp_receiver_ready_T7(struct dc_link *link); +struct link_hwss_ext { + /* function pointers below require check for NULL at all time + * ********************************************************************* + */ + void (*set_hblank_min_symbol_width)(struct pipe_ctx *pipe_ctx, + const struct dc_link_settings *link_settings, + struct fixed31_32 throttled_vcp_size); + void (*set_throttled_vcp_size)(struct pipe_ctx *pipe_ctx, + struct fixed31_32 throttled_vcp_size); + void (*enable_dp_link_output)(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings); + void (*disable_dp_link_output)(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal); + void (*set_dp_link_test_pattern)(struct dc_link *link, + const struct link_resource *link_res, + struct encoder_set_dp_phy_pattern_param *tp_params); + void (*set_dp_lane_settings)(struct dc_link *link, + const struct link_resource *link_res, + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]); +}; -void dp_disable_link_phy(struct dc_link *link, const struct link_resource *link_res, - enum signal_type signal); - -void dp_disable_link_phy_mst(struct dc_link *link, const struct link_resource *link_res, - enum signal_type signal); - -bool dp_set_hw_training_pattern( - struct dc_link *link, - const struct link_resource *link_res, - enum dc_dp_training_pattern pattern, - uint32_t offset); - -void dp_set_hw_lane_settings( - struct dc_link *link, - const struct link_resource *link_res, - const struct link_training_settings *link_settings, - uint32_t offset); - -void dp_set_hw_test_pattern( - struct dc_link *link, - const struct link_resource *link_res, - enum dp_test_pattern test_pattern, - uint8_t *custom_pattern, - uint32_t custom_pattern_size); - -void dp_retrain_link_dp_test(struct dc_link *link, - struct dc_link_settings *link_setting, - bool skip_video_pattern); +struct link_hwss { + struct link_hwss_ext ext; + /* function pointers below MUST be assigned to all types of link_hwss + * ********************************************************************* + */ + void (*setup_stream_encoder)(struct pipe_ctx *pipe_ctx); + void (*reset_stream_encoder)(struct pipe_ctx *pipe_ctx); +}; #endif /* __DC_LINK_HWSS_H__ */ + diff --git a/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h b/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h index 2470405e996b..a402df225a76 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h +++ b/drivers/gpu/drm/amd/display/dc/inc/reg_helper.h @@ -498,6 +498,40 @@ uint32_t generic_indirect_reg_update_ex(const struct dc_context *ctx, uint8_t shift1, uint32_t mask1, uint32_t field_value1, ...); +/* indirect register access + * underlying implementation determines which index/data pair to be used + * in a synchronous way + */ +#define IX_REG_SET_N_SYNC(index, n, initial_val, ...) \ + generic_indirect_reg_update_ex_sync(CTX, \ + IND_REG(index), \ + initial_val, \ + n, __VA_ARGS__) + +#define IX_REG_SET_2_SYNC(index, init_value, f1, v1, f2, v2) \ + IX_REG_SET_N_SYNC(index, 2, init_value, \ + FN(reg, f1), v1,\ + FN(reg, f2), v2) + +#define IX_REG_GET_N_SYNC(index, n, ...) \ + generic_indirect_reg_get_sync(CTX, \ + IND_REG(index), \ + n, __VA_ARGS__) + +#define IX_REG_GET_SYNC(index, field, val) \ + IX_REG_GET_N_SYNC(index, 1, \ + FN(data_reg_name, field), val) + +uint32_t generic_indirect_reg_get_sync(const struct dc_context *ctx, + uint32_t index, int n, + uint8_t shift1, uint32_t mask1, uint32_t *field_value1, + ...); + +uint32_t generic_indirect_reg_update_ex_sync(const struct dc_context *ctx, + uint32_t index, uint32_t reg_val, int n, + uint8_t shift1, uint32_t mask1, uint32_t field_value1, + ...); + /* register offload macros * * instead of MMIO to register directly, in some cases we want diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index dbfe6690ded8..2369f38ed06f 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -54,10 +54,8 @@ struct resource_caps { int num_dsc; unsigned int num_dig_link_enc; // Total number of DIGs (digital encoders) in DIO (Display Input/Output). unsigned int num_usb4_dpia; // Total number of USB4 DPIA (DisplayPort Input Adapters). -#if defined(CONFIG_DRM_AMD_DC_DCN) int num_hpo_dp_stream_encoder; int num_hpo_dp_link_encoder; -#endif int num_mpc_3dlut; }; @@ -77,14 +75,12 @@ struct resource_create_funcs { struct stream_encoder *(*create_stream_encoder)( enum engine_id eng_id, struct dc_context *ctx); -#if defined(CONFIG_DRM_AMD_DC_DCN) struct hpo_dp_stream_encoder *(*create_hpo_dp_stream_encoder)( enum engine_id eng_id, struct dc_context *ctx); struct hpo_dp_link_encoder *(*create_hpo_dp_link_encoder)( uint8_t inst, struct dc_context *ctx); -#endif struct dce_hwseq *(*create_hwseq)( struct dc_context *ctx); @@ -205,12 +201,9 @@ int get_num_mpc_splits(struct pipe_ctx *pipe); int get_num_odm_splits(struct pipe_ctx *pipe); -#if defined(CONFIG_DRM_AMD_DC_DCN) -struct hpo_dp_link_encoder *resource_get_hpo_dp_link_enc_for_det_lt( - const struct resource_context *res_ctx, - const struct resource_pool *pool, - const struct dc_link *link); -#endif +bool get_temp_dp_link_res(struct dc_link *link, + struct link_resource *link_res, + struct dc_link_settings *link_settings); void reset_syncd_pipes_from_disabled_pipes(struct dc *dc, struct dc_state *context); @@ -221,4 +214,7 @@ void check_syncd_pipes_for_disabled_master_pipe(struct dc *dc, uint8_t resource_transmitter_to_phy_idx(const struct dc *dc, enum transmitter transmitter); +const struct link_hwss *get_link_hwss(const struct dc_link *link, + const struct link_resource *link_res); + #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn30/irq_service_dcn30.c b/drivers/gpu/drm/amd/display/dc/irq/dcn30/irq_service_dcn30.c index 914ce2ce1c2f..0b68c08fac3f 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dcn30/irq_service_dcn30.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn30/irq_service_dcn30.c @@ -37,8 +37,8 @@ #include "nbio/nbio_7_4_offset.h" -#include "dcn/dpcs_3_0_0_offset.h" -#include "dcn/dpcs_3_0_0_sh_mask.h" +#include "dpcs/dpcs_3_0_0_offset.h" +#include "dpcs/dpcs_3_0_0_sh_mask.h" #include "mmhub/mmhub_2_0_0_offset.h" #include "mmhub/mmhub_2_0_0_sh_mask.h" diff --git a/drivers/gpu/drm/amd/display/dc/link/Makefile b/drivers/gpu/drm/amd/display/dc/link/Makefile new file mode 100644 index 000000000000..c4a69ba7cb62 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/Makefile @@ -0,0 +1,30 @@ +# +# Copyright 2022 Advanced Micro Devices, Inc. +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the "Software"), +# to deal in the Software without restriction, including without limitation +# the rights to use, copy, modify, merge, publish, distribute, sublicense, +# and/or sell copies of the Software, and to permit persons to whom the +# Software is furnished to do so, subject to the following conditions: +# +# The above copyright notice and this permission notice shall be included in +# all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR +# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +# OTHER DEALINGS IN THE SOFTWARE. +# +# Makefile for the link sub-component of DAL. +# It abstracts the control and status of back end pipe such as DIO, HPO, DPIA, +# PHY, HPD, DDC and etc). + +LINK = link_hwss_dio.o link_hwss_dpia.o link_hwss_hpo_dp.o + +AMD_DAL_LINK = $(addprefix $(AMDDALPATH)/dc/link/,$(LINK)) + +AMD_DISPLAY_FILES += $(AMD_DAL_LINK) diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_dio.c b/drivers/gpu/drm/amd/display/dc/link/link_hwss_dio.c new file mode 100644 index 000000000000..0f845113a6aa --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_dio.c @@ -0,0 +1,137 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "link_hwss_dio.h" +#include "core_types.h" +#include "dc_link_dp.h" +#include "link_enc_cfg.h" + +void set_dio_throttled_vcp_size(struct pipe_ctx *pipe_ctx, + struct fixed31_32 throttled_vcp_size) +{ + struct stream_encoder *stream_encoder = pipe_ctx->stream_res.stream_enc; + + stream_encoder->funcs->set_throttled_vcp_size( + stream_encoder, + throttled_vcp_size); +} + +void setup_dio_stream_encoder(struct pipe_ctx *pipe_ctx) +{ + struct link_encoder *link_enc = link_enc_cfg_get_link_enc(pipe_ctx->stream->link); + + link_enc->funcs->connect_dig_be_to_fe(link_enc, + pipe_ctx->stream_res.stream_enc->id, true); + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + dp_source_sequence_trace(pipe_ctx->stream->link, + DPCD_SOURCE_SEQ_AFTER_CONNECT_DIG_FE_BE); +} + +void reset_dio_stream_encoder(struct pipe_ctx *pipe_ctx) +{ + struct link_encoder *link_enc = link_enc_cfg_get_link_enc(pipe_ctx->stream->link); + + link_enc->funcs->connect_dig_be_to_fe( + link_enc, + pipe_ctx->stream_res.stream_enc->id, + false); + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + dp_source_sequence_trace(pipe_ctx->stream->link, + DPCD_SOURCE_SEQ_AFTER_DISCONNECT_DIG_FE_BE); + +} + +void enable_dio_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings) +{ + struct link_encoder *link_enc = link_enc_cfg_get_link_enc(link); + + if (dc_is_dp_sst_signal(signal)) + link_enc->funcs->enable_dp_output( + link_enc, + link_settings, + clock_source); + else + link_enc->funcs->enable_dp_mst_output( + link_enc, + link_settings, + clock_source); + dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_ENABLE_LINK_PHY); +} + +void disable_dio_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal) +{ + struct link_encoder *link_enc = link_enc_cfg_get_link_enc(link); + + link_enc->funcs->disable_output(link_enc, signal); + dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY); +} + +void set_dio_dp_link_test_pattern(struct dc_link *link, + const struct link_resource *link_res, + struct encoder_set_dp_phy_pattern_param *tp_params) +{ + struct link_encoder *link_enc = link_enc_cfg_get_link_enc(link); + + link_enc->funcs->dp_set_phy_pattern(link_enc, tp_params); + dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_SET_SOURCE_PATTERN); +} + +void set_dio_dp_lane_settings(struct dc_link *link, + const struct link_resource *link_res, + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]) +{ + struct link_encoder *link_enc = link_enc_cfg_get_link_enc(link); + + link_enc->funcs->dp_set_lane_settings(link_enc, link_settings, lane_settings); +} + +static const struct link_hwss dio_link_hwss = { + .setup_stream_encoder = setup_dio_stream_encoder, + .reset_stream_encoder = reset_dio_stream_encoder, + .ext = { + .set_throttled_vcp_size = set_dio_throttled_vcp_size, + .enable_dp_link_output = enable_dio_dp_link_output, + .disable_dp_link_output = disable_dio_dp_link_output, + .set_dp_link_test_pattern = set_dio_dp_link_test_pattern, + .set_dp_lane_settings = set_dio_dp_lane_settings, + }, +}; + +bool can_use_dio_link_hwss(const struct dc_link *link, + const struct link_resource *link_res) +{ + return link->link_enc != NULL; +} + +const struct link_hwss *get_dio_link_hwss(void) +{ + return &dio_link_hwss; +} diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_dio.h b/drivers/gpu/drm/amd/display/dc/link/link_hwss_dio.h new file mode 100644 index 000000000000..680df20b1fa3 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_dio.h @@ -0,0 +1,53 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __LINK_HWSS_DIO_H__ +#define __LINK_HWSS_DIO_H__ + +#include "link_hwss.h" + +const struct link_hwss *get_dio_link_hwss(void); +bool can_use_dio_link_hwss(const struct dc_link *link, + const struct link_resource *link_res); +void set_dio_throttled_vcp_size(struct pipe_ctx *pipe_ctx, + struct fixed31_32 throttled_vcp_size); +void setup_dio_stream_encoder(struct pipe_ctx *pipe_ctx); +void reset_dio_stream_encoder(struct pipe_ctx *pipe_ctx); +void enable_dio_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings); +void disable_dio_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal); +void set_dio_dp_link_test_pattern(struct dc_link *link, + const struct link_resource *link_res, + struct encoder_set_dp_phy_pattern_param *tp_params); +void set_dio_dp_lane_settings(struct dc_link *link, + const struct link_resource *link_res, + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]); + +#endif /* __LINK_HWSS_DIO_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_dpia.c b/drivers/gpu/drm/amd/display/dc/link/link_hwss_dpia.c new file mode 100644 index 000000000000..35b206225201 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_dpia.c @@ -0,0 +1,51 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "link_hwss_dpia.h" +#include "core_types.h" +#include "link_hwss_dio.h" + +static const struct link_hwss dpia_link_hwss = { + .setup_stream_encoder = setup_dio_stream_encoder, + .reset_stream_encoder = reset_dio_stream_encoder, + .ext = { + .set_throttled_vcp_size = set_dio_throttled_vcp_size, + .enable_dp_link_output = enable_dio_dp_link_output, + .disable_dp_link_output = disable_dio_dp_link_output, + .set_dp_link_test_pattern = set_dio_dp_link_test_pattern, + .set_dp_lane_settings = set_dio_dp_lane_settings, + }, +}; + +bool can_use_dpia_link_hwss(const struct dc_link *link, + const struct link_resource *link_res) +{ + return link->is_dig_mapping_flexible && + link->dc->res_pool->funcs->link_encs_assign; +} + +const struct link_hwss *get_dpia_link_hwss(void) +{ + return &dpia_link_hwss; +} diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_dpia.h b/drivers/gpu/drm/amd/display/dc/link/link_hwss_dpia.h new file mode 100644 index 000000000000..ad16ec5d9bb7 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_dpia.h @@ -0,0 +1,34 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __LINK_HWSS_DPIA_H__ +#define __LINK_HWSS_DPIA_H__ + +#include "link_hwss.h" + +const struct link_hwss *get_dpia_link_hwss(void); +bool can_use_dpia_link_hwss(const struct dc_link *link, + const struct link_resource *link_res); + +#endif /* __LINK_HWSS_DPIA_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c new file mode 100644 index 000000000000..74919491675f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.c @@ -0,0 +1,254 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "link_hwss_hpo_dp.h" +#include "dm_helpers.h" +#include "core_types.h" +#include "dccg.h" +#include "dc_link_dp.h" + +static enum phyd32clk_clock_source get_phyd32clk_src(struct dc_link *link) +{ + switch (link->link_enc->transmitter) { + case TRANSMITTER_UNIPHY_A: + return PHYD32CLKA; + case TRANSMITTER_UNIPHY_B: + return PHYD32CLKB; + case TRANSMITTER_UNIPHY_C: + return PHYD32CLKC; + case TRANSMITTER_UNIPHY_D: + return PHYD32CLKD; + case TRANSMITTER_UNIPHY_E: + return PHYD32CLKE; + default: + return PHYD32CLKA; + } +} + +static void set_hpo_dp_throttled_vcp_size(struct pipe_ctx *pipe_ctx, + struct fixed31_32 throttled_vcp_size) +{ + struct hpo_dp_stream_encoder *hpo_dp_stream_encoder = + pipe_ctx->stream_res.hpo_dp_stream_enc; + struct hpo_dp_link_encoder *hpo_dp_link_encoder = + pipe_ctx->link_res.hpo_dp_link_enc; + + hpo_dp_link_encoder->funcs->set_throttled_vcp_size(hpo_dp_link_encoder, + hpo_dp_stream_encoder->inst, + throttled_vcp_size); +} + +static void set_hpo_dp_hblank_min_symbol_width(struct pipe_ctx *pipe_ctx, + const struct dc_link_settings *link_settings, + struct fixed31_32 throttled_vcp_size) +{ + struct hpo_dp_stream_encoder *hpo_dp_stream_encoder = + pipe_ctx->stream_res.hpo_dp_stream_enc; + struct dc_crtc_timing *timing = &pipe_ctx->stream->timing; + struct fixed31_32 h_blank_in_ms, time_slot_in_ms, mtp_cnt_per_h_blank; + uint32_t link_bw_in_kbps = + dc_link_bandwidth_kbps(pipe_ctx->stream->link, link_settings); + uint16_t hblank_min_symbol_width = 0; + + if (link_bw_in_kbps > 0) { + h_blank_in_ms = dc_fixpt_div(dc_fixpt_from_int( + timing->h_total - timing->h_addressable), + dc_fixpt_from_fraction(timing->pix_clk_100hz, 10)); + time_slot_in_ms = dc_fixpt_from_fraction(32 * 4, link_bw_in_kbps); + mtp_cnt_per_h_blank = dc_fixpt_div(h_blank_in_ms, + dc_fixpt_mul_int(time_slot_in_ms, 64)); + hblank_min_symbol_width = dc_fixpt_floor( + dc_fixpt_mul(mtp_cnt_per_h_blank, throttled_vcp_size)); + } + + hpo_dp_stream_encoder->funcs->set_hblank_min_symbol_width(hpo_dp_stream_encoder, + hblank_min_symbol_width); +} + +static int get_odm_segment_count(struct pipe_ctx *pipe_ctx) +{ + struct pipe_ctx *odm_pipe = pipe_ctx->next_odm_pipe; + int count = 1; + + while (odm_pipe != NULL) { + count++; + odm_pipe = odm_pipe->next_odm_pipe; + } + + return count; +} + +static void setup_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) +{ + struct dc *dc = pipe_ctx->stream->ctx->dc; + struct hpo_dp_stream_encoder *stream_enc = pipe_ctx->stream_res.hpo_dp_stream_enc; + struct hpo_dp_link_encoder *link_enc = pipe_ctx->link_res.hpo_dp_link_enc; + struct dccg *dccg = dc->res_pool->dccg; + struct timing_generator *tg = pipe_ctx->stream_res.tg; + int odm_segment_count = get_odm_segment_count(pipe_ctx); + enum phyd32clk_clock_source phyd32clk = get_phyd32clk_src(pipe_ctx->stream->link); + + dccg->funcs->set_dpstreamclk(dccg, DTBCLK0, tg->inst); + dccg->funcs->enable_symclk32_se(dccg, stream_enc->inst, phyd32clk); + dccg->funcs->set_dtbclk_dto(dccg, tg->inst, pipe_ctx->stream->phy_pix_clk, + odm_segment_count, + &pipe_ctx->stream->timing); + stream_enc->funcs->enable_stream(stream_enc); + stream_enc->funcs->map_stream_to_link(stream_enc, stream_enc->inst, link_enc->inst); +} + +static void reset_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) +{ + struct dc *dc = pipe_ctx->stream->ctx->dc; + struct hpo_dp_stream_encoder *stream_enc = pipe_ctx->stream_res.hpo_dp_stream_enc; + struct dccg *dccg = dc->res_pool->dccg; + struct timing_generator *tg = pipe_ctx->stream_res.tg; + + stream_enc->funcs->disable(stream_enc); + dccg->funcs->set_dtbclk_dto(dccg, tg->inst, 0, 0, &pipe_ctx->stream->timing); + dccg->funcs->disable_symclk32_se(dccg, stream_enc->inst); + dccg->funcs->set_dpstreamclk(dccg, REFCLK, tg->inst); +} + +static void enable_hpo_dp_fpga_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings) +{ + const struct dc *dc = link->dc; + enum phyd32clk_clock_source phyd32clk = get_phyd32clk_src(link); + int phyd32clk_freq_khz = link_settings->link_rate == LINK_RATE_UHBR10 ? 312500 : + link_settings->link_rate == LINK_RATE_UHBR13_5 ? 412875 : + link_settings->link_rate == LINK_RATE_UHBR20 ? 625000 : 0; + + dm_set_phyd32clk(dc->ctx, phyd32clk_freq_khz); + dc->res_pool->dccg->funcs->set_physymclk( + dc->res_pool->dccg, + link->link_enc_hw_inst, + PHYSYMCLK_FORCE_SRC_PHYD32CLK, + true); + dc->res_pool->dccg->funcs->enable_symclk32_le( + dc->res_pool->dccg, + link_res->hpo_dp_link_enc->inst, + phyd32clk); + link_res->hpo_dp_link_enc->funcs->link_enable( + link_res->hpo_dp_link_enc, + link_settings->lane_count); + +} + +static void enable_hpo_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings) +{ + if (IS_FPGA_MAXIMUS_DC(link->dc->ctx->dce_environment)) + enable_hpo_dp_fpga_link_output(link, link_res, signal, + clock_source, link_settings); + else + link_res->hpo_dp_link_enc->funcs->enable_link_phy( + link_res->hpo_dp_link_enc, + link_settings, + link->link_enc->transmitter, + link->link_enc->hpd_source); +} + + +static void disable_hpo_dp_fpga_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal) +{ + const struct dc *dc = link->dc; + + link_res->hpo_dp_link_enc->funcs->link_disable(link_res->hpo_dp_link_enc); + dc->res_pool->dccg->funcs->disable_symclk32_le( + dc->res_pool->dccg, + link_res->hpo_dp_link_enc->inst); + dc->res_pool->dccg->funcs->set_physymclk( + dc->res_pool->dccg, + link->link_enc_hw_inst, + PHYSYMCLK_FORCE_SRC_SYMCLK, + false); + dm_set_phyd32clk(dc->ctx, 0); +} + +static void disable_hpo_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal) +{ + if (IS_FPGA_MAXIMUS_DC(link->dc->ctx->dce_environment)) { + disable_hpo_dp_fpga_link_output(link, link_res, signal); + } else { + link_res->hpo_dp_link_enc->funcs->link_disable(link_res->hpo_dp_link_enc); + link_res->hpo_dp_link_enc->funcs->disable_link_phy( + link_res->hpo_dp_link_enc, signal); + } +} + +static void set_hpo_dp_link_test_pattern(struct dc_link *link, + const struct link_resource *link_res, + struct encoder_set_dp_phy_pattern_param *tp_params) +{ + link_res->hpo_dp_link_enc->funcs->set_link_test_pattern( + link_res->hpo_dp_link_enc, tp_params); + dp_source_sequence_trace(link, DPCD_SOURCE_SEQ_AFTER_SET_SOURCE_PATTERN); +} + +static void set_hpo_dp_lane_settings(struct dc_link *link, + const struct link_resource *link_res, + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]) +{ + link_res->hpo_dp_link_enc->funcs->set_ffe( + link_res->hpo_dp_link_enc, + link_settings, + lane_settings[0].FFE_PRESET.raw); +} + +static const struct link_hwss hpo_dp_link_hwss = { + .setup_stream_encoder = setup_hpo_dp_stream_encoder, + .reset_stream_encoder = reset_hpo_dp_stream_encoder, + .ext = { + .set_throttled_vcp_size = set_hpo_dp_throttled_vcp_size, + .set_hblank_min_symbol_width = set_hpo_dp_hblank_min_symbol_width, + .enable_dp_link_output = enable_hpo_dp_link_output, + .disable_dp_link_output = disable_hpo_dp_link_output, + .set_dp_link_test_pattern = set_hpo_dp_link_test_pattern, + .set_dp_lane_settings = set_hpo_dp_lane_settings, + }, +}; + +bool can_use_hpo_dp_link_hwss(const struct dc_link *link, + const struct link_resource *link_res) +{ + return link_res->hpo_dp_link_enc != NULL; +} + +const struct link_hwss *get_hpo_dp_link_hwss(void) +{ + return &hpo_dp_link_hwss; +} + diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.h b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.h new file mode 100644 index 000000000000..57d447ec27b8 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_dp.h @@ -0,0 +1,35 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __LINK_HWSS_HPO_DP_H__ +#define __LINK_HWSS_HPO_DP_H__ + +#include "link_hwss.h" + +bool can_use_hpo_dp_link_hwss(const struct dc_link *link, + const struct link_resource *link_res); +const struct link_hwss *get_hpo_dp_link_hwss(void); + + +#endif /* __LINK_HWSS_HPO_DP_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_frl.c b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_frl.c new file mode 100644 index 000000000000..9df273ca699b --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_frl.c @@ -0,0 +1,43 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "link_hwss_hpo_frl.h" +#include "core_types.h" +#include "virtual/virtual_link_hwss.h" + +static const struct link_hwss hpo_frl_link_hwss = { + .setup_stream_encoder = virtual_setup_stream_encoder, + .reset_stream_encoder = virtual_reset_stream_encoder, +}; + +bool can_use_hpo_frl_link_hwss(const struct dc_link *link, + const struct link_resource *link_res) +{ + return link_res->hpo_frl_link_enc != NULL; +} + +const struct link_hwss *get_hpo_frl_link_hwss(void) +{ + return &hpo_frl_link_hwss; +} diff --git a/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_frl.h b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_frl.h new file mode 100644 index 000000000000..ea8d9760132f --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/link_hwss_hpo_frl.h @@ -0,0 +1,34 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __LINK_HWSS_HPO_FRL_H__ +#define __LINK_HWSS_HPO_FRL_H__ + +#include "link_hwss.h" + +bool can_use_hpo_frl_link_hwss(const struct dc_link *link, + const struct link_resource *link_res); +const struct link_hwss *get_hpo_frl_link_hwss(void); + +#endif /* __LINK_HWSS_HPO_FRL_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/os_types.h b/drivers/gpu/drm/amd/display/dc/os_types.h index 5df1d80c8341..17d05071b809 100644 --- a/drivers/gpu/drm/amd/display/dc/os_types.h +++ b/drivers/gpu/drm/amd/display/dc/os_types.h @@ -36,7 +36,7 @@ #include <asm/byteorder.h> #include <drm/drm_print.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include "cgs_common.h" diff --git a/drivers/gpu/drm/amd/display/dc/virtual/Makefile b/drivers/gpu/drm/amd/display/dc/virtual/Makefile index 07326d244d50..931facd4dab5 100644 --- a/drivers/gpu/drm/amd/display/dc/virtual/Makefile +++ b/drivers/gpu/drm/amd/display/dc/virtual/Makefile @@ -23,7 +23,7 @@ # Makefile for the virtual sub-component of DAL. # It provides the control and status of HW CRTC block. -VIRTUAL = virtual_link_encoder.o virtual_stream_encoder.o +VIRTUAL = virtual_link_encoder.o virtual_stream_encoder.o virtual_link_hwss.o AMD_DAL_VIRTUAL = $(addprefix $(AMDDALPATH)/dc/virtual/,$(VIRTUAL)) diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_encoder.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_encoder.c index 348e9a600a72..df8bc44bc4be 100644 --- a/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_encoder.c @@ -63,7 +63,8 @@ static void virtual_link_encoder_disable_output( static void virtual_link_encoder_dp_set_lane_settings( struct link_encoder *enc, - const struct link_training_settings *link_settings) {} + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]) {} static void virtual_link_encoder_dp_set_phy_pattern( struct link_encoder *enc, diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_hwss.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_hwss.c new file mode 100644 index 000000000000..525eba2a3354 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_hwss.c @@ -0,0 +1,43 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "virtual_link_hwss.h" + +void virtual_setup_stream_encoder(struct pipe_ctx *pipe_ctx) +{ +} + +void virtual_reset_stream_encoder(struct pipe_ctx *pipe_ctx) +{ +} +static const struct link_hwss virtual_link_hwss = { + .setup_stream_encoder = virtual_setup_stream_encoder, + .reset_stream_encoder = virtual_reset_stream_encoder, +}; + +const struct link_hwss *get_virtual_link_hwss(void) +{ + return &virtual_link_hwss; +} diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_hwss.h b/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_hwss.h new file mode 100644 index 000000000000..e6bcb4bb0f3a --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_link_hwss.h @@ -0,0 +1,34 @@ +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __DC_VIRTUAL_LINK_HWSS_H__ +#define __DC_VIRTUAL_LINK_HWSS_H__ + +#include "core_types.h" + +void virtual_setup_stream_encoder(struct pipe_ctx *pipe_ctx); +void virtual_reset_stream_encoder(struct pipe_ctx *pipe_ctx); +const struct link_hwss *get_virtual_link_hwss(void); + +#endif /* __DC_VIRTUAL_LINK_HWSS_H__ */ diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index 873ecd04e01d..fa1177638706 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -46,10 +46,10 @@ /* Firmware versioning. */ #ifdef DMUB_EXPOSE_VERSION -#define DMUB_FW_VERSION_GIT_HASH 0xbaf06b95 +#define DMUB_FW_VERSION_GIT_HASH 0x5189adbf #define DMUB_FW_VERSION_MAJOR 0 #define DMUB_FW_VERSION_MINOR 0 -#define DMUB_FW_VERSION_REVISION 98 +#define DMUB_FW_VERSION_REVISION 103 #define DMUB_FW_VERSION_TEST 0 #define DMUB_FW_VERSION_VBIOS 0 #define DMUB_FW_VERSION_HOTFIX 0 @@ -367,8 +367,9 @@ union dmub_fw_boot_options { /**< 1 if all root clock gating is enabled and low power memory is enabled*/ uint32_t power_optimization: 1; uint32_t diag_env: 1; /* 1 if diagnostic environment */ + uint32_t gpint_scratch8: 1; /* 1 if GPINT is in scratch8*/ - uint32_t reserved : 19; /**< reserved */ + uint32_t reserved : 18; /**< reserved */ } bits; /**< boot bits */ uint32_t all; /**< 32-bit access to bits */ }; @@ -523,7 +524,7 @@ union dmub_inbox0_cmd_lock_hw { uint32_t command_code: 8; /* NOTE: Must be have enough bits to match: enum hw_lock_client */ - uint32_t hw_lock_client: 1; + uint32_t hw_lock_client: 2; /* NOTE: Below fields must match with: struct dmub_hw_lock_inst_flags */ uint32_t otg_inst: 3; @@ -538,7 +539,7 @@ union dmub_inbox0_cmd_lock_hw { uint32_t lock: 1; /**< Lock */ uint32_t should_release: 1; /**< Release */ - uint32_t reserved: 8; /**< Reserved for extending more clients, HW, etc. */ + uint32_t reserved: 7; /**< Reserved for extending more clients, HW, etc. */ } bits; uint32_t all; }; @@ -644,6 +645,7 @@ enum dmub_cmd_type { * Command type used for OUTBOX1 notification enable */ DMUB_CMD__OUTBOX1_ENABLE = 71, + /** * Command type used for all idle optimization commands. */ @@ -656,6 +658,7 @@ enum dmub_cmd_type { * Command type used for all panel control commands. */ DMUB_CMD__PANEL_CNTL = 74, + /** * Command type used for interfacing with DPIA. */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c index fa0569174aec..82c651535628 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c @@ -441,7 +441,7 @@ void dmub_dcn31_get_diagnostic_data(struct dmub_srv *dmub, struct dmub_diagnosti bool dmub_dcn31_should_detect(struct dmub_srv *dmub) { uint32_t fw_boot_status = REG_READ(DMCUB_SCRATCH0); - bool should_detect = fw_boot_status & DMUB_FW_BOOT_STATUS_BIT_DETECTION_REQUIRED; + bool should_detect = (fw_boot_status & DMUB_FW_BOOT_STATUS_BIT_DETECTION_REQUIRED) != 0; return should_detect; } diff --git a/drivers/gpu/drm/amd/display/include/bios_parser_types.h b/drivers/gpu/drm/amd/display/include/bios_parser_types.h index b8ffb216ebc4..cf4027cc3f4c 100644 --- a/drivers/gpu/drm/amd/display/include/bios_parser_types.h +++ b/drivers/gpu/drm/amd/display/include/bios_parser_types.h @@ -152,10 +152,8 @@ struct bp_transmitter_control { enum signal_type signal; enum dc_color_depth color_depth; /* not used for DCE6.0 */ enum hpd_source_id hpd_sel; /* ucHPDSel, used for DCe6.0 */ -#if defined(CONFIG_DRM_AMD_DC_DCN) enum tx_ffe_id txffe_sel; /* used for DCN3 */ enum engine_id hpo_engine_id; /* used for DCN3 */ -#endif struct graphics_object_id connector_obj_id; /* symClock; in 10kHz, pixel clock, in HDMI deep color mode, it should * be pixel clock * deep_color_ratio (in KHz) diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h index e4a2dfacab4c..e672be6327cb 100644 --- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h +++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h @@ -212,6 +212,7 @@ enum { #define ASICREV_IS_GREEN_SARDINE(eChipRev) ((eChipRev >= GREEN_SARDINE_A0) && (eChipRev < 0xFF)) #endif #define DEVICE_ID_NV_13FE 0x13FE // CYAN_SKILLFISH +#define DEVICE_ID_NV_143F 0x143F #define FAMILY_VGH 144 #define DEVICE_ID_VGH_163F 0x163F #define VANGOGH_A0 0x01 diff --git a/drivers/gpu/drm/amd/display/include/ddc_service_types.h b/drivers/gpu/drm/amd/display/include/ddc_service_types.h index a2b80514d83e..fb289a5c873a 100644 --- a/drivers/gpu/drm/amd/display/include/ddc_service_types.h +++ b/drivers/gpu/drm/amd/display/include/ddc_service_types.h @@ -34,6 +34,8 @@ #define DP_BRANCH_DEVICE_ID_90CC24 0x90CC24 #define DP_BRANCH_DEVICE_ID_00E04C 0x00E04C #define DP_BRANCH_DEVICE_ID_006037 0x006037 +#define DP_BRANCH_HW_REV_10 0x10 +#define DP_BRANCH_HW_REV_20 0x20 #define DP_DEVICE_ID_38EC11 0x38EC11 enum ddc_result { @@ -121,4 +123,7 @@ struct av_sync_data { static const uint8_t DP_SINK_DEVICE_STR_ID_1[] = {7, 1, 8, 7, 3, 0}; static const uint8_t DP_SINK_DEVICE_STR_ID_2[] = {7, 1, 8, 7, 5, 0}; +/*MST Dock*/ +static const uint8_t SYNAPTICS_DEVICE_ID[] = "SYNA"; + #endif /* __DAL_DDC_SERVICE_TYPES_H__ */ diff --git a/drivers/gpu/drm/amd/display/include/dpcd_defs.h b/drivers/gpu/drm/amd/display/include/dpcd_defs.h index ffd0df1701e6..ac822181359c 100644 --- a/drivers/gpu/drm/amd/display/include/dpcd_defs.h +++ b/drivers/gpu/drm/amd/display/include/dpcd_defs.h @@ -26,7 +26,7 @@ #ifndef __DAL_DPCD_DEFS_H__ #define __DAL_DPCD_DEFS_H__ -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #ifndef DP_SINK_HW_REVISION_START // can remove this once the define gets into linux drm_dp_helper.h #define DP_SINK_HW_REVISION_START 0x409 #endif @@ -144,14 +144,10 @@ enum dpcd_training_patterns { DPCD_TRAINING_PATTERN_1, DPCD_TRAINING_PATTERN_2, DPCD_TRAINING_PATTERN_3, -#if defined(CONFIG_DRM_AMD_DC_DCN) DPCD_TRAINING_PATTERN_4 = 7, DPCD_128b_132b_TPS1 = 1, DPCD_128b_132b_TPS2 = 2, DPCD_128b_132b_TPS2_CDS = 3, -#else - DPCD_TRAINING_PATTERN_4 = 7 -#endif }; /* This enum is for use with PsrSinkPsrStatus.bits.sinkSelfRefreshStatus diff --git a/drivers/gpu/drm/amd/display/include/grph_object_defs.h b/drivers/gpu/drm/amd/display/include/grph_object_defs.h index 84b299ff500a..05ba188a7c3b 100644 --- a/drivers/gpu/drm/amd/display/include/grph_object_defs.h +++ b/drivers/gpu/drm/amd/display/include/grph_object_defs.h @@ -140,7 +140,6 @@ enum sync_source { SYNC_SOURCE_DUAL_GPU_PIN }; -#if defined(CONFIG_DRM_AMD_DC_DCN) enum tx_ffe_id { TX_FFE0 = 0, TX_FFE1, @@ -150,7 +149,6 @@ enum tx_ffe_id { TX_FFE_PreShoot_Only, TX_FFE_No_FFE, }; -#endif /* connector sizes in millimeters - from BiosParserTypes.hpp */ #define CONNECTOR_SIZE_DVI 40 diff --git a/drivers/gpu/drm/amd/display/include/grph_object_id.h b/drivers/gpu/drm/amd/display/include/grph_object_id.h index 01775417cf4b..fed1edc038d8 100644 --- a/drivers/gpu/drm/amd/display/include/grph_object_id.h +++ b/drivers/gpu/drm/amd/display/include/grph_object_id.h @@ -184,14 +184,12 @@ enum engine_id { ENGINE_ID_DACA, ENGINE_ID_DACB, ENGINE_ID_VCE, /* wireless display pseudo-encoder */ -#if defined(CONFIG_DRM_AMD_DC_DCN) ENGINE_ID_HPO_0, ENGINE_ID_HPO_1, ENGINE_ID_HPO_DP_0, ENGINE_ID_HPO_DP_1, ENGINE_ID_HPO_DP_2, ENGINE_ID_HPO_DP_3, -#endif ENGINE_ID_VIRTUAL, ENGINE_ID_COUNT, diff --git a/drivers/gpu/drm/amd/display/include/link_service_types.h b/drivers/gpu/drm/amd/display/include/link_service_types.h index 424bccd36434..447a56286dd0 100644 --- a/drivers/gpu/drm/amd/display/include/link_service_types.h +++ b/drivers/gpu/drm/amd/display/include/link_service_types.h @@ -74,12 +74,10 @@ enum link_training_result { LINK_TRAINING_LINK_LOSS, /* Abort link training (because sink unplugged) */ LINK_TRAINING_ABORT, -#if defined(CONFIG_DRM_AMD_DC_DCN) DP_128b_132b_LT_FAILED, DP_128b_132b_MAX_LOOP_COUNT_REACHED, DP_128b_132b_CHANNEL_EQ_DONE_TIMEOUT, DP_128b_132b_CDS_DONE_TIMEOUT, -#endif }; enum lttpr_mode { @@ -99,23 +97,19 @@ struct link_training_settings { enum dc_pre_emphasis *pre_emphasis; enum dc_post_cursor2 *post_cursor2; bool should_set_fec_ready; -#if defined(CONFIG_DRM_AMD_DC_DCN) /* TODO - factor lane_settings out because it changes during LT */ union dc_dp_ffe_preset *ffe_preset; -#endif uint16_t cr_pattern_time; uint16_t eq_pattern_time; uint16_t cds_pattern_time; enum dc_dp_training_pattern pattern_for_cr; enum dc_dp_training_pattern pattern_for_eq; -#if defined(CONFIG_DRM_AMD_DC_DCN) enum dc_dp_training_pattern pattern_for_cds; uint32_t eq_wait_time_limit; uint8_t eq_loop_count_limit; uint32_t cds_wait_time_limit; -#endif bool enhanced_framing; enum lttpr_mode lttpr_mode; @@ -161,7 +155,6 @@ enum dp_test_pattern { DP_TEST_PATTERN_CP2520_2, DP_TEST_PATTERN_HBR2_COMPLIANCE_EYE = DP_TEST_PATTERN_CP2520_2, DP_TEST_PATTERN_CP2520_3, -#if defined(CONFIG_DRM_AMD_DC_DCN) DP_TEST_PATTERN_128b_132b_TPS1, DP_TEST_PATTERN_128b_132b_TPS2, DP_TEST_PATTERN_PRBS9, @@ -171,20 +164,15 @@ enum dp_test_pattern { DP_TEST_PATTERN_PRBS31, DP_TEST_PATTERN_264BIT_CUSTOM, DP_TEST_PATTERN_SQUARE_PULSE, -#endif /* Link Training Patterns */ DP_TEST_PATTERN_TRAINING_PATTERN1, DP_TEST_PATTERN_TRAINING_PATTERN2, DP_TEST_PATTERN_TRAINING_PATTERN3, DP_TEST_PATTERN_TRAINING_PATTERN4, -#if defined(CONFIG_DRM_AMD_DC_DCN) DP_TEST_PATTERN_128b_132b_TPS1_TRAINING_MODE, DP_TEST_PATTERN_128b_132b_TPS2_TRAINING_MODE, DP_TEST_PATTERN_PHY_PATTERN_END = DP_TEST_PATTERN_128b_132b_TPS2_TRAINING_MODE, -#else - DP_TEST_PATTERN_PHY_PATTERN_END = DP_TEST_PATTERN_TRAINING_PATTERN4, -#endif /* link test patterns*/ DP_TEST_PATTERN_COLOR_SQUARES, diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h index 399fbca8947b..8502263d2968 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.h @@ -30,7 +30,7 @@ #include "hdcp_log.h" #include <drm/drm_hdcp.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> enum mod_hdcp_trans_input_result { UNKNOWN = 0, diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h index 1ab813b4fd14..1d8b746b02f2 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h @@ -34,7 +34,8 @@ struct dc_info_packet; struct mod_vrr_params; void mod_build_vsc_infopacket(const struct dc_stream_state *stream, - struct dc_info_packet *info_packet); + struct dc_info_packet *info_packet, + enum dc_color_space cs); void mod_build_hf_vsif_infopacket(const struct dc_stream_state *stream, struct dc_info_packet *info_packet); diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c index 57f198de5e2c..b691aa45e84f 100644 --- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c +++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c @@ -130,7 +130,8 @@ enum ColorimetryYCCDP { }; void mod_build_vsc_infopacket(const struct dc_stream_state *stream, - struct dc_info_packet *info_packet) + struct dc_info_packet *info_packet, + enum dc_color_space cs) { unsigned int vsc_packet_revision = vsc_packet_undefined; unsigned int i; @@ -331,13 +332,13 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, /* Set Colorimetry format based on pixel encoding */ switch (stream->timing.pixel_encoding) { case PIXEL_ENCODING_RGB: - if ((stream->output_color_space == COLOR_SPACE_SRGB) || - (stream->output_color_space == COLOR_SPACE_SRGB_LIMITED)) + if ((cs == COLOR_SPACE_SRGB) || + (cs == COLOR_SPACE_SRGB_LIMITED)) colorimetryFormat = ColorimetryRGB_DP_sRGB; - else if (stream->output_color_space == COLOR_SPACE_ADOBERGB) + else if (cs == COLOR_SPACE_ADOBERGB) colorimetryFormat = ColorimetryRGB_DP_AdobeRGB; - else if ((stream->output_color_space == COLOR_SPACE_2020_RGB_FULLRANGE) || - (stream->output_color_space == COLOR_SPACE_2020_RGB_LIMITEDRANGE)) + else if ((cs == COLOR_SPACE_2020_RGB_FULLRANGE) || + (cs == COLOR_SPACE_2020_RGB_LIMITEDRANGE)) colorimetryFormat = ColorimetryRGB_DP_ITU_R_BT2020RGB; break; @@ -347,13 +348,13 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, /* Note: xvYCC probably not supported correctly here on DP since colorspace translation * loses distinction between BT601 vs xvYCC601 in translation */ - if (stream->output_color_space == COLOR_SPACE_YCBCR601) + if (cs == COLOR_SPACE_YCBCR601) colorimetryFormat = ColorimetryYCC_DP_ITU601; - else if (stream->output_color_space == COLOR_SPACE_YCBCR709) + else if (cs == COLOR_SPACE_YCBCR709) colorimetryFormat = ColorimetryYCC_DP_ITU709; - else if (stream->output_color_space == COLOR_SPACE_ADOBERGB) + else if (cs == COLOR_SPACE_ADOBERGB) colorimetryFormat = ColorimetryYCC_DP_AdobeYCC; - else if (stream->output_color_space == COLOR_SPACE_2020_YCBCR) + else if (cs == COLOR_SPACE_2020_YCBCR) colorimetryFormat = ColorimetryYCC_DP_ITU2020YCbCr; break; @@ -391,8 +392,8 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, } /* all YCbCr are always limited range */ - if ((stream->output_color_space == COLOR_SPACE_SRGB_LIMITED) || - (stream->output_color_space == COLOR_SPACE_2020_RGB_LIMITEDRANGE) || + if ((cs == COLOR_SPACE_SRGB_LIMITED) || + (cs == COLOR_SPACE_2020_RGB_LIMITEDRANGE) || (pixelEncoding != 0x0)) { info_packet->sb[17] |= 0x80; /* DB17 bit 7 set to 1 for CEA timing. */ } diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index f57a1478f0fe..fe4e585781bb 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -83,6 +83,7 @@ enum amd_apu_flags { * @AMD_IP_BLOCK_TYPE_VCN: Video Core/Codec Next * @AMD_IP_BLOCK_TYPE_MES: Micro-Engine Scheduler * @AMD_IP_BLOCK_TYPE_JPEG: JPEG Engine +* @AMD_IP_BLOCK_TYPE_NUM: Total number of IP block types */ enum amd_ip_block_type { AMD_IP_BLOCK_TYPE_COMMON, @@ -268,7 +269,6 @@ enum amd_dpm_forced_level; * @set_clockgating_state: enable/disable cg for the IP block * @set_powergating_state: enable/disable pg for the IP block * @get_clockgating_state: get current clockgating status - * @enable_umd_pstate: enable UMD powerstate * * These hooks provide an interface for controlling the operational state * of IP blocks. After acquiring a list of IP blocks for the GPU in use, @@ -299,7 +299,6 @@ struct amd_ip_funcs { int (*set_powergating_state)(void *handle, enum amd_powergating_state state); void (*get_clockgating_state)(void *handle, u32 *flags); - int (*enable_umd_pstate)(void *handle, enum amd_dpm_forced_level *level); }; diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dpcs/dpcs_3_0_0_offset.h index 67faaf68e9d7..0bb47e06eee8 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dpcs/dpcs_3_0_0_offset.h @@ -1,3 +1,10 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright (C) 2020 Advanced Micro Devices, Inc. + * + * Authors: AMD + */ + #ifndef _dpcs_3_0_0_OFFSET_HEADER #define _dpcs_3_0_0_OFFSET_HEADER diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dpcs/dpcs_3_0_0_sh_mask.h index b4ef50a72868..23fa1121a967 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dpcs/dpcs_3_0_0_sh_mask.h @@ -1,3 +1,10 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright (C) 2020 Advanced Micro Devices, Inc. + * + * Authors: AMD + */ + #ifndef _dpcs_3_0_0_SH_MASK_HEADER #define _dpcs_3_0_0_SH_MASK_HEADER diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_3_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dpcs/dpcs_3_0_3_offset.h index 6d6db95a391f..6d6db95a391f 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_3_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dpcs/dpcs_3_0_3_offset.h diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_3_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dpcs/dpcs_3_0_3_sh_mask.h index 55743d06f728..55743d06f728 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dpcs_3_0_3_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dpcs/dpcs_3_0_3_sh_mask.h diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index ac941f62cbed..2f60cf35a444 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -269,18 +269,9 @@ struct kfd2kgd_calls { int (*hqd_sdma_destroy)(struct amdgpu_device *adev, void *mqd, unsigned int timeout); - int (*address_watch_disable)(struct amdgpu_device *adev); - int (*address_watch_execute)(struct amdgpu_device *adev, - unsigned int watch_point_id, - uint32_t cntl_val, - uint32_t addr_hi, - uint32_t addr_lo); int (*wave_control_execute)(struct amdgpu_device *adev, uint32_t gfx_index_val, uint32_t sq_cmd); - uint32_t (*address_watch_get_offset)(struct amdgpu_device *adev, - unsigned int watch_point_id, - unsigned int reg_offset); bool (*get_atc_vmid_pasid_mapping_info)(struct amdgpu_device *adev, uint8_t vmid, uint16_t *p_pasid); diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index 5c0867ebcfce..a4c267f15959 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -25,6 +25,13 @@ #define __KGD_PP_INTERFACE_H__ extern const struct amdgpu_ip_block_version pp_smu_ip_block; +extern const struct amdgpu_ip_block_version smu_v11_0_ip_block; +extern const struct amdgpu_ip_block_version smu_v12_0_ip_block; +extern const struct amdgpu_ip_block_version smu_v13_0_ip_block; + +enum smu_event_type { + SMU_EVENT_RESET_COMPLETE = 0, +}; struct amd_vce_state { /* vce clocks */ @@ -308,12 +315,13 @@ struct amd_pm_funcs { void *rps, bool *equal); /* export for sysfs */ - void (*set_fan_control_mode)(void *handle, u32 mode); - u32 (*get_fan_control_mode)(void *handle); + int (*set_fan_control_mode)(void *handle, u32 mode); + int (*get_fan_control_mode)(void *handle, u32 *fan_mode); int (*set_fan_speed_pwm)(void *handle, u32 speed); int (*get_fan_speed_pwm)(void *handle, u32 *speed); int (*force_clock_level)(void *handle, enum pp_clock_type type, uint32_t mask); int (*print_clock_levels)(void *handle, enum pp_clock_type type, char *buf); + int (*emit_clock_levels)(void *handle, enum pp_clock_type type, char *buf, int *offset); int (*force_performance_level)(void *handle, enum amd_dpm_forced_level level); int (*get_sclk_od)(void *handle); int (*set_sclk_od)(void *handle, uint32_t value); @@ -400,6 +408,7 @@ struct amd_pm_funcs { int (*get_dpm_clock_table)(void *handle, struct dpm_clocks *clock_table); int (*get_smu_prv_buf_details)(void *handle, void **addr, size_t *size); + void (*pm_compute_clocks)(void *handle); }; struct metrics_table_header { diff --git a/drivers/gpu/drm/amd/pm/Makefile b/drivers/gpu/drm/amd/pm/Makefile index 8cf6eff1ea93..51751db436b0 100644 --- a/drivers/gpu/drm/amd/pm/Makefile +++ b/drivers/gpu/drm/amd/pm/Makefile @@ -21,26 +21,29 @@ # subdir-ccflags-y += \ - -I$(FULL_AMD_PATH)/pm/inc/ \ -I$(FULL_AMD_PATH)/include/asic_reg \ -I$(FULL_AMD_PATH)/include \ + -I$(FULL_AMD_PATH)/pm/inc/ \ -I$(FULL_AMD_PATH)/pm/swsmu \ + -I$(FULL_AMD_PATH)/pm/swsmu/inc \ + -I$(FULL_AMD_PATH)/pm/swsmu/inc/pmfw_if \ -I$(FULL_AMD_PATH)/pm/swsmu/smu11 \ -I$(FULL_AMD_PATH)/pm/swsmu/smu12 \ -I$(FULL_AMD_PATH)/pm/swsmu/smu13 \ - -I$(FULL_AMD_PATH)/pm/powerplay \ + -I$(FULL_AMD_PATH)/pm/powerplay/inc \ -I$(FULL_AMD_PATH)/pm/powerplay/smumgr\ - -I$(FULL_AMD_PATH)/pm/powerplay/hwmgr + -I$(FULL_AMD_PATH)/pm/powerplay/hwmgr \ + -I$(FULL_AMD_PATH)/pm/legacy-dpm AMD_PM_PATH = ../pm -PM_LIBS = swsmu powerplay +PM_LIBS = swsmu powerplay legacy-dpm AMD_PM = $(addsuffix /Makefile,$(addprefix $(FULL_AMD_PATH)/pm/,$(PM_LIBS))) include $(AMD_PM) -PM_MGR = amdgpu_dpm.o amdgpu_pm.o +PM_MGR = amdgpu_dpm.o amdgpu_pm.o amdgpu_dpm_internal.o AMD_PM_POWER = $(addprefix $(AMD_PM_PATH)/,$(PM_MGR)) diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c index 08362d506534..1d63f1e8884c 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c @@ -31,896 +31,41 @@ #include "amdgpu_display.h" #include "hwmgr.h" #include <linux/power_supply.h> +#include "amdgpu_smu.h" -#define WIDTH_4K 3840 +#define amdgpu_dpm_enable_bapm(adev, e) \ + ((adev)->powerplay.pp_funcs->enable_bapm((adev)->powerplay.pp_handle, (e))) -void amdgpu_dpm_print_class_info(u32 class, u32 class2) -{ - const char *s; - - switch (class & ATOM_PPLIB_CLASSIFICATION_UI_MASK) { - case ATOM_PPLIB_CLASSIFICATION_UI_NONE: - default: - s = "none"; - break; - case ATOM_PPLIB_CLASSIFICATION_UI_BATTERY: - s = "battery"; - break; - case ATOM_PPLIB_CLASSIFICATION_UI_BALANCED: - s = "balanced"; - break; - case ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE: - s = "performance"; - break; - } - printk("\tui class: %s\n", s); - printk("\tinternal class:"); - if (((class & ~ATOM_PPLIB_CLASSIFICATION_UI_MASK) == 0) && - (class2 == 0)) - pr_cont(" none"); - else { - if (class & ATOM_PPLIB_CLASSIFICATION_BOOT) - pr_cont(" boot"); - if (class & ATOM_PPLIB_CLASSIFICATION_THERMAL) - pr_cont(" thermal"); - if (class & ATOM_PPLIB_CLASSIFICATION_LIMITEDPOWERSOURCE) - pr_cont(" limited_pwr"); - if (class & ATOM_PPLIB_CLASSIFICATION_REST) - pr_cont(" rest"); - if (class & ATOM_PPLIB_CLASSIFICATION_FORCED) - pr_cont(" forced"); - if (class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE) - pr_cont(" 3d_perf"); - if (class & ATOM_PPLIB_CLASSIFICATION_OVERDRIVETEMPLATE) - pr_cont(" ovrdrv"); - if (class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE) - pr_cont(" uvd"); - if (class & ATOM_PPLIB_CLASSIFICATION_3DLOW) - pr_cont(" 3d_low"); - if (class & ATOM_PPLIB_CLASSIFICATION_ACPI) - pr_cont(" acpi"); - if (class & ATOM_PPLIB_CLASSIFICATION_HD2STATE) - pr_cont(" uvd_hd2"); - if (class & ATOM_PPLIB_CLASSIFICATION_HDSTATE) - pr_cont(" uvd_hd"); - if (class & ATOM_PPLIB_CLASSIFICATION_SDSTATE) - pr_cont(" uvd_sd"); - if (class2 & ATOM_PPLIB_CLASSIFICATION2_LIMITEDPOWERSOURCE_2) - pr_cont(" limited_pwr2"); - if (class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) - pr_cont(" ulv"); - if (class2 & ATOM_PPLIB_CLASSIFICATION2_MVC) - pr_cont(" uvd_mvc"); - } - pr_cont("\n"); -} - -void amdgpu_dpm_print_cap_info(u32 caps) -{ - printk("\tcaps:"); - if (caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) - pr_cont(" single_disp"); - if (caps & ATOM_PPLIB_SUPPORTS_VIDEO_PLAYBACK) - pr_cont(" video"); - if (caps & ATOM_PPLIB_DISALLOW_ON_DC) - pr_cont(" no_dc"); - pr_cont("\n"); -} - -void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev, - struct amdgpu_ps *rps) -{ - printk("\tstatus:"); - if (rps == adev->pm.dpm.current_ps) - pr_cont(" c"); - if (rps == adev->pm.dpm.requested_ps) - pr_cont(" r"); - if (rps == adev->pm.dpm.boot_ps) - pr_cont(" b"); - pr_cont("\n"); -} - -void amdgpu_dpm_get_active_displays(struct amdgpu_device *adev) -{ - struct drm_device *ddev = adev_to_drm(adev); - struct drm_crtc *crtc; - struct amdgpu_crtc *amdgpu_crtc; - - adev->pm.dpm.new_active_crtcs = 0; - adev->pm.dpm.new_active_crtc_count = 0; - if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) { - list_for_each_entry(crtc, - &ddev->mode_config.crtc_list, head) { - amdgpu_crtc = to_amdgpu_crtc(crtc); - if (amdgpu_crtc->enabled) { - adev->pm.dpm.new_active_crtcs |= (1 << amdgpu_crtc->crtc_id); - adev->pm.dpm.new_active_crtc_count++; - } - } - } -} - - -u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev) -{ - struct drm_device *dev = adev_to_drm(adev); - struct drm_crtc *crtc; - struct amdgpu_crtc *amdgpu_crtc; - u32 vblank_in_pixels; - u32 vblank_time_us = 0xffffffff; /* if the displays are off, vblank time is max */ - - if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) { - list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { - amdgpu_crtc = to_amdgpu_crtc(crtc); - if (crtc->enabled && amdgpu_crtc->enabled && amdgpu_crtc->hw_mode.clock) { - vblank_in_pixels = - amdgpu_crtc->hw_mode.crtc_htotal * - (amdgpu_crtc->hw_mode.crtc_vblank_end - - amdgpu_crtc->hw_mode.crtc_vdisplay + - (amdgpu_crtc->v_border * 2)); - - vblank_time_us = vblank_in_pixels * 1000 / amdgpu_crtc->hw_mode.clock; - break; - } - } - } - - return vblank_time_us; -} - -u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev) -{ - struct drm_device *dev = adev_to_drm(adev); - struct drm_crtc *crtc; - struct amdgpu_crtc *amdgpu_crtc; - u32 vrefresh = 0; - - if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) { - list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { - amdgpu_crtc = to_amdgpu_crtc(crtc); - if (crtc->enabled && amdgpu_crtc->enabled && amdgpu_crtc->hw_mode.clock) { - vrefresh = drm_mode_vrefresh(&amdgpu_crtc->hw_mode); - break; - } - } - } - - return vrefresh; -} - -bool amdgpu_is_internal_thermal_sensor(enum amdgpu_int_thermal_type sensor) -{ - switch (sensor) { - case THERMAL_TYPE_RV6XX: - case THERMAL_TYPE_RV770: - case THERMAL_TYPE_EVERGREEN: - case THERMAL_TYPE_SUMO: - case THERMAL_TYPE_NI: - case THERMAL_TYPE_SI: - case THERMAL_TYPE_CI: - case THERMAL_TYPE_KV: - return true; - case THERMAL_TYPE_ADT7473_WITH_INTERNAL: - case THERMAL_TYPE_EMC2103_WITH_INTERNAL: - return false; /* need special handling */ - case THERMAL_TYPE_NONE: - case THERMAL_TYPE_EXTERNAL: - case THERMAL_TYPE_EXTERNAL_GPIO: - default: - return false; - } -} - -union power_info { - struct _ATOM_POWERPLAY_INFO info; - struct _ATOM_POWERPLAY_INFO_V2 info_2; - struct _ATOM_POWERPLAY_INFO_V3 info_3; - struct _ATOM_PPLIB_POWERPLAYTABLE pplib; - struct _ATOM_PPLIB_POWERPLAYTABLE2 pplib2; - struct _ATOM_PPLIB_POWERPLAYTABLE3 pplib3; - struct _ATOM_PPLIB_POWERPLAYTABLE4 pplib4; - struct _ATOM_PPLIB_POWERPLAYTABLE5 pplib5; -}; - -union fan_info { - struct _ATOM_PPLIB_FANTABLE fan; - struct _ATOM_PPLIB_FANTABLE2 fan2; - struct _ATOM_PPLIB_FANTABLE3 fan3; -}; - -static int amdgpu_parse_clk_voltage_dep_table(struct amdgpu_clock_voltage_dependency_table *amdgpu_table, - ATOM_PPLIB_Clock_Voltage_Dependency_Table *atom_table) -{ - u32 size = atom_table->ucNumEntries * - sizeof(struct amdgpu_clock_voltage_dependency_entry); - int i; - ATOM_PPLIB_Clock_Voltage_Dependency_Record *entry; - - amdgpu_table->entries = kzalloc(size, GFP_KERNEL); - if (!amdgpu_table->entries) - return -ENOMEM; - - entry = &atom_table->entries[0]; - for (i = 0; i < atom_table->ucNumEntries; i++) { - amdgpu_table->entries[i].clk = le16_to_cpu(entry->usClockLow) | - (entry->ucClockHigh << 16); - amdgpu_table->entries[i].v = le16_to_cpu(entry->usVoltage); - entry = (ATOM_PPLIB_Clock_Voltage_Dependency_Record *) - ((u8 *)entry + sizeof(ATOM_PPLIB_Clock_Voltage_Dependency_Record)); - } - amdgpu_table->count = atom_table->ucNumEntries; - - return 0; -} - -int amdgpu_get_platform_caps(struct amdgpu_device *adev) -{ - struct amdgpu_mode_info *mode_info = &adev->mode_info; - union power_info *power_info; - int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo); - u16 data_offset; - u8 frev, crev; - - if (!amdgpu_atom_parse_data_header(mode_info->atom_context, index, NULL, - &frev, &crev, &data_offset)) - return -EINVAL; - power_info = (union power_info *)(mode_info->atom_context->bios + data_offset); - - adev->pm.dpm.platform_caps = le32_to_cpu(power_info->pplib.ulPlatformCaps); - adev->pm.dpm.backbias_response_time = le16_to_cpu(power_info->pplib.usBackbiasTime); - adev->pm.dpm.voltage_response_time = le16_to_cpu(power_info->pplib.usVoltageTime); - - return 0; -} - -/* sizeof(ATOM_PPLIB_EXTENDEDHEADER) */ -#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V2 12 -#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V3 14 -#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V4 16 -#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V5 18 -#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V6 20 -#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V7 22 -#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V8 24 -#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V9 26 - -int amdgpu_parse_extended_power_table(struct amdgpu_device *adev) -{ - struct amdgpu_mode_info *mode_info = &adev->mode_info; - union power_info *power_info; - union fan_info *fan_info; - ATOM_PPLIB_Clock_Voltage_Dependency_Table *dep_table; - int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo); - u16 data_offset; - u8 frev, crev; - int ret, i; - - if (!amdgpu_atom_parse_data_header(mode_info->atom_context, index, NULL, - &frev, &crev, &data_offset)) - return -EINVAL; - power_info = (union power_info *)(mode_info->atom_context->bios + data_offset); - - /* fan table */ - if (le16_to_cpu(power_info->pplib.usTableSize) >= - sizeof(struct _ATOM_PPLIB_POWERPLAYTABLE3)) { - if (power_info->pplib3.usFanTableOffset) { - fan_info = (union fan_info *)(mode_info->atom_context->bios + data_offset + - le16_to_cpu(power_info->pplib3.usFanTableOffset)); - adev->pm.dpm.fan.t_hyst = fan_info->fan.ucTHyst; - adev->pm.dpm.fan.t_min = le16_to_cpu(fan_info->fan.usTMin); - adev->pm.dpm.fan.t_med = le16_to_cpu(fan_info->fan.usTMed); - adev->pm.dpm.fan.t_high = le16_to_cpu(fan_info->fan.usTHigh); - adev->pm.dpm.fan.pwm_min = le16_to_cpu(fan_info->fan.usPWMMin); - adev->pm.dpm.fan.pwm_med = le16_to_cpu(fan_info->fan.usPWMMed); - adev->pm.dpm.fan.pwm_high = le16_to_cpu(fan_info->fan.usPWMHigh); - if (fan_info->fan.ucFanTableFormat >= 2) - adev->pm.dpm.fan.t_max = le16_to_cpu(fan_info->fan2.usTMax); - else - adev->pm.dpm.fan.t_max = 10900; - adev->pm.dpm.fan.cycle_delay = 100000; - if (fan_info->fan.ucFanTableFormat >= 3) { - adev->pm.dpm.fan.control_mode = fan_info->fan3.ucFanControlMode; - adev->pm.dpm.fan.default_max_fan_pwm = - le16_to_cpu(fan_info->fan3.usFanPWMMax); - adev->pm.dpm.fan.default_fan_output_sensitivity = 4836; - adev->pm.dpm.fan.fan_output_sensitivity = - le16_to_cpu(fan_info->fan3.usFanOutputSensitivity); - } - adev->pm.dpm.fan.ucode_fan_control = true; - } - } - - /* clock dependancy tables, shedding tables */ - if (le16_to_cpu(power_info->pplib.usTableSize) >= - sizeof(struct _ATOM_PPLIB_POWERPLAYTABLE4)) { - if (power_info->pplib4.usVddcDependencyOnSCLKOffset) { - dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(power_info->pplib4.usVddcDependencyOnSCLKOffset)); - ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.vddc_dependency_on_sclk, - dep_table); - if (ret) { - amdgpu_free_extended_power_table(adev); - return ret; - } - } - if (power_info->pplib4.usVddciDependencyOnMCLKOffset) { - dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(power_info->pplib4.usVddciDependencyOnMCLKOffset)); - ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.vddci_dependency_on_mclk, - dep_table); - if (ret) { - amdgpu_free_extended_power_table(adev); - return ret; - } - } - if (power_info->pplib4.usVddcDependencyOnMCLKOffset) { - dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(power_info->pplib4.usVddcDependencyOnMCLKOffset)); - ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.vddc_dependency_on_mclk, - dep_table); - if (ret) { - amdgpu_free_extended_power_table(adev); - return ret; - } - } - if (power_info->pplib4.usMvddDependencyOnMCLKOffset) { - dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(power_info->pplib4.usMvddDependencyOnMCLKOffset)); - ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk, - dep_table); - if (ret) { - amdgpu_free_extended_power_table(adev); - return ret; - } - } - if (power_info->pplib4.usMaxClockVoltageOnDCOffset) { - ATOM_PPLIB_Clock_Voltage_Limit_Table *clk_v = - (ATOM_PPLIB_Clock_Voltage_Limit_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(power_info->pplib4.usMaxClockVoltageOnDCOffset)); - if (clk_v->ucNumEntries) { - adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.sclk = - le16_to_cpu(clk_v->entries[0].usSclkLow) | - (clk_v->entries[0].ucSclkHigh << 16); - adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.mclk = - le16_to_cpu(clk_v->entries[0].usMclkLow) | - (clk_v->entries[0].ucMclkHigh << 16); - adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.vddc = - le16_to_cpu(clk_v->entries[0].usVddc); - adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.vddci = - le16_to_cpu(clk_v->entries[0].usVddci); - } - } - if (power_info->pplib4.usVddcPhaseShedLimitsTableOffset) { - ATOM_PPLIB_PhaseSheddingLimits_Table *psl = - (ATOM_PPLIB_PhaseSheddingLimits_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(power_info->pplib4.usVddcPhaseShedLimitsTableOffset)); - ATOM_PPLIB_PhaseSheddingLimits_Record *entry; - - adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries = - kcalloc(psl->ucNumEntries, - sizeof(struct amdgpu_phase_shedding_limits_entry), - GFP_KERNEL); - if (!adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries) { - amdgpu_free_extended_power_table(adev); - return -ENOMEM; - } - - entry = &psl->entries[0]; - for (i = 0; i < psl->ucNumEntries; i++) { - adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries[i].sclk = - le16_to_cpu(entry->usSclkLow) | (entry->ucSclkHigh << 16); - adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries[i].mclk = - le16_to_cpu(entry->usMclkLow) | (entry->ucMclkHigh << 16); - adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries[i].voltage = - le16_to_cpu(entry->usVoltage); - entry = (ATOM_PPLIB_PhaseSheddingLimits_Record *) - ((u8 *)entry + sizeof(ATOM_PPLIB_PhaseSheddingLimits_Record)); - } - adev->pm.dpm.dyn_state.phase_shedding_limits_table.count = - psl->ucNumEntries; - } - } - - /* cac data */ - if (le16_to_cpu(power_info->pplib.usTableSize) >= - sizeof(struct _ATOM_PPLIB_POWERPLAYTABLE5)) { - adev->pm.dpm.tdp_limit = le32_to_cpu(power_info->pplib5.ulTDPLimit); - adev->pm.dpm.near_tdp_limit = le32_to_cpu(power_info->pplib5.ulNearTDPLimit); - adev->pm.dpm.near_tdp_limit_adjusted = adev->pm.dpm.near_tdp_limit; - adev->pm.dpm.tdp_od_limit = le16_to_cpu(power_info->pplib5.usTDPODLimit); - if (adev->pm.dpm.tdp_od_limit) - adev->pm.dpm.power_control = true; - else - adev->pm.dpm.power_control = false; - adev->pm.dpm.tdp_adjustment = 0; - adev->pm.dpm.sq_ramping_threshold = le32_to_cpu(power_info->pplib5.ulSQRampingThreshold); - adev->pm.dpm.cac_leakage = le32_to_cpu(power_info->pplib5.ulCACLeakage); - adev->pm.dpm.load_line_slope = le16_to_cpu(power_info->pplib5.usLoadLineSlope); - if (power_info->pplib5.usCACLeakageTableOffset) { - ATOM_PPLIB_CAC_Leakage_Table *cac_table = - (ATOM_PPLIB_CAC_Leakage_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(power_info->pplib5.usCACLeakageTableOffset)); - ATOM_PPLIB_CAC_Leakage_Record *entry; - u32 size = cac_table->ucNumEntries * sizeof(struct amdgpu_cac_leakage_table); - adev->pm.dpm.dyn_state.cac_leakage_table.entries = kzalloc(size, GFP_KERNEL); - if (!adev->pm.dpm.dyn_state.cac_leakage_table.entries) { - amdgpu_free_extended_power_table(adev); - return -ENOMEM; - } - entry = &cac_table->entries[0]; - for (i = 0; i < cac_table->ucNumEntries; i++) { - if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_EVV) { - adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc1 = - le16_to_cpu(entry->usVddc1); - adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc2 = - le16_to_cpu(entry->usVddc2); - adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc3 = - le16_to_cpu(entry->usVddc3); - } else { - adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc = - le16_to_cpu(entry->usVddc); - adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].leakage = - le32_to_cpu(entry->ulLeakageValue); - } - entry = (ATOM_PPLIB_CAC_Leakage_Record *) - ((u8 *)entry + sizeof(ATOM_PPLIB_CAC_Leakage_Record)); - } - adev->pm.dpm.dyn_state.cac_leakage_table.count = cac_table->ucNumEntries; - } - } - - /* ext tables */ - if (le16_to_cpu(power_info->pplib.usTableSize) >= - sizeof(struct _ATOM_PPLIB_POWERPLAYTABLE3)) { - ATOM_PPLIB_EXTENDEDHEADER *ext_hdr = (ATOM_PPLIB_EXTENDEDHEADER *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(power_info->pplib3.usExtendendedHeaderOffset)); - if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V2) && - ext_hdr->usVCETableOffset) { - VCEClockInfoArray *array = (VCEClockInfoArray *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usVCETableOffset) + 1); - ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table *limits = - (ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usVCETableOffset) + 1 + - 1 + array->ucNumEntries * sizeof(VCEClockInfo)); - ATOM_PPLIB_VCE_State_Table *states = - (ATOM_PPLIB_VCE_State_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usVCETableOffset) + 1 + - 1 + (array->ucNumEntries * sizeof (VCEClockInfo)) + - 1 + (limits->numEntries * sizeof(ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record))); - ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record *entry; - ATOM_PPLIB_VCE_State_Record *state_entry; - VCEClockInfo *vce_clk; - u32 size = limits->numEntries * - sizeof(struct amdgpu_vce_clock_voltage_dependency_entry); - adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries = - kzalloc(size, GFP_KERNEL); - if (!adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries) { - amdgpu_free_extended_power_table(adev); - return -ENOMEM; - } - adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.count = - limits->numEntries; - entry = &limits->entries[0]; - state_entry = &states->entries[0]; - for (i = 0; i < limits->numEntries; i++) { - vce_clk = (VCEClockInfo *) - ((u8 *)&array->entries[0] + - (entry->ucVCEClockInfoIndex * sizeof(VCEClockInfo))); - adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].evclk = - le16_to_cpu(vce_clk->usEVClkLow) | (vce_clk->ucEVClkHigh << 16); - adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].ecclk = - le16_to_cpu(vce_clk->usECClkLow) | (vce_clk->ucECClkHigh << 16); - adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].v = - le16_to_cpu(entry->usVoltage); - entry = (ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record *) - ((u8 *)entry + sizeof(ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record)); - } - adev->pm.dpm.num_of_vce_states = - states->numEntries > AMD_MAX_VCE_LEVELS ? - AMD_MAX_VCE_LEVELS : states->numEntries; - for (i = 0; i < adev->pm.dpm.num_of_vce_states; i++) { - vce_clk = (VCEClockInfo *) - ((u8 *)&array->entries[0] + - (state_entry->ucVCEClockInfoIndex * sizeof(VCEClockInfo))); - adev->pm.dpm.vce_states[i].evclk = - le16_to_cpu(vce_clk->usEVClkLow) | (vce_clk->ucEVClkHigh << 16); - adev->pm.dpm.vce_states[i].ecclk = - le16_to_cpu(vce_clk->usECClkLow) | (vce_clk->ucECClkHigh << 16); - adev->pm.dpm.vce_states[i].clk_idx = - state_entry->ucClockInfoIndex & 0x3f; - adev->pm.dpm.vce_states[i].pstate = - (state_entry->ucClockInfoIndex & 0xc0) >> 6; - state_entry = (ATOM_PPLIB_VCE_State_Record *) - ((u8 *)state_entry + sizeof(ATOM_PPLIB_VCE_State_Record)); - } - } - if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V3) && - ext_hdr->usUVDTableOffset) { - UVDClockInfoArray *array = (UVDClockInfoArray *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usUVDTableOffset) + 1); - ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table *limits = - (ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usUVDTableOffset) + 1 + - 1 + (array->ucNumEntries * sizeof (UVDClockInfo))); - ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record *entry; - u32 size = limits->numEntries * - sizeof(struct amdgpu_uvd_clock_voltage_dependency_entry); - adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries = - kzalloc(size, GFP_KERNEL); - if (!adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries) { - amdgpu_free_extended_power_table(adev); - return -ENOMEM; - } - adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count = - limits->numEntries; - entry = &limits->entries[0]; - for (i = 0; i < limits->numEntries; i++) { - UVDClockInfo *uvd_clk = (UVDClockInfo *) - ((u8 *)&array->entries[0] + - (entry->ucUVDClockInfoIndex * sizeof(UVDClockInfo))); - adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].vclk = - le16_to_cpu(uvd_clk->usVClkLow) | (uvd_clk->ucVClkHigh << 16); - adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].dclk = - le16_to_cpu(uvd_clk->usDClkLow) | (uvd_clk->ucDClkHigh << 16); - adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].v = - le16_to_cpu(entry->usVoltage); - entry = (ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record *) - ((u8 *)entry + sizeof(ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record)); - } - } - if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V4) && - ext_hdr->usSAMUTableOffset) { - ATOM_PPLIB_SAMClk_Voltage_Limit_Table *limits = - (ATOM_PPLIB_SAMClk_Voltage_Limit_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usSAMUTableOffset) + 1); - ATOM_PPLIB_SAMClk_Voltage_Limit_Record *entry; - u32 size = limits->numEntries * - sizeof(struct amdgpu_clock_voltage_dependency_entry); - adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries = - kzalloc(size, GFP_KERNEL); - if (!adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries) { - amdgpu_free_extended_power_table(adev); - return -ENOMEM; - } - adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.count = - limits->numEntries; - entry = &limits->entries[0]; - for (i = 0; i < limits->numEntries; i++) { - adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[i].clk = - le16_to_cpu(entry->usSAMClockLow) | (entry->ucSAMClockHigh << 16); - adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[i].v = - le16_to_cpu(entry->usVoltage); - entry = (ATOM_PPLIB_SAMClk_Voltage_Limit_Record *) - ((u8 *)entry + sizeof(ATOM_PPLIB_SAMClk_Voltage_Limit_Record)); - } - } - if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V5) && - ext_hdr->usPPMTableOffset) { - ATOM_PPLIB_PPM_Table *ppm = (ATOM_PPLIB_PPM_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usPPMTableOffset)); - adev->pm.dpm.dyn_state.ppm_table = - kzalloc(sizeof(struct amdgpu_ppm_table), GFP_KERNEL); - if (!adev->pm.dpm.dyn_state.ppm_table) { - amdgpu_free_extended_power_table(adev); - return -ENOMEM; - } - adev->pm.dpm.dyn_state.ppm_table->ppm_design = ppm->ucPpmDesign; - adev->pm.dpm.dyn_state.ppm_table->cpu_core_number = - le16_to_cpu(ppm->usCpuCoreNumber); - adev->pm.dpm.dyn_state.ppm_table->platform_tdp = - le32_to_cpu(ppm->ulPlatformTDP); - adev->pm.dpm.dyn_state.ppm_table->small_ac_platform_tdp = - le32_to_cpu(ppm->ulSmallACPlatformTDP); - adev->pm.dpm.dyn_state.ppm_table->platform_tdc = - le32_to_cpu(ppm->ulPlatformTDC); - adev->pm.dpm.dyn_state.ppm_table->small_ac_platform_tdc = - le32_to_cpu(ppm->ulSmallACPlatformTDC); - adev->pm.dpm.dyn_state.ppm_table->apu_tdp = - le32_to_cpu(ppm->ulApuTDP); - adev->pm.dpm.dyn_state.ppm_table->dgpu_tdp = - le32_to_cpu(ppm->ulDGpuTDP); - adev->pm.dpm.dyn_state.ppm_table->dgpu_ulv_power = - le32_to_cpu(ppm->ulDGpuUlvPower); - adev->pm.dpm.dyn_state.ppm_table->tj_max = - le32_to_cpu(ppm->ulTjmax); - } - if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V6) && - ext_hdr->usACPTableOffset) { - ATOM_PPLIB_ACPClk_Voltage_Limit_Table *limits = - (ATOM_PPLIB_ACPClk_Voltage_Limit_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usACPTableOffset) + 1); - ATOM_PPLIB_ACPClk_Voltage_Limit_Record *entry; - u32 size = limits->numEntries * - sizeof(struct amdgpu_clock_voltage_dependency_entry); - adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries = - kzalloc(size, GFP_KERNEL); - if (!adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries) { - amdgpu_free_extended_power_table(adev); - return -ENOMEM; - } - adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.count = - limits->numEntries; - entry = &limits->entries[0]; - for (i = 0; i < limits->numEntries; i++) { - adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[i].clk = - le16_to_cpu(entry->usACPClockLow) | (entry->ucACPClockHigh << 16); - adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[i].v = - le16_to_cpu(entry->usVoltage); - entry = (ATOM_PPLIB_ACPClk_Voltage_Limit_Record *) - ((u8 *)entry + sizeof(ATOM_PPLIB_ACPClk_Voltage_Limit_Record)); - } - } - if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V7) && - ext_hdr->usPowerTuneTableOffset) { - u8 rev = *(u8 *)(mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usPowerTuneTableOffset)); - ATOM_PowerTune_Table *pt; - adev->pm.dpm.dyn_state.cac_tdp_table = - kzalloc(sizeof(struct amdgpu_cac_tdp_table), GFP_KERNEL); - if (!adev->pm.dpm.dyn_state.cac_tdp_table) { - amdgpu_free_extended_power_table(adev); - return -ENOMEM; - } - if (rev > 0) { - ATOM_PPLIB_POWERTUNE_Table_V1 *ppt = (ATOM_PPLIB_POWERTUNE_Table_V1 *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usPowerTuneTableOffset)); - adev->pm.dpm.dyn_state.cac_tdp_table->maximum_power_delivery_limit = - ppt->usMaximumPowerDeliveryLimit; - pt = &ppt->power_tune_table; - } else { - ATOM_PPLIB_POWERTUNE_Table *ppt = (ATOM_PPLIB_POWERTUNE_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usPowerTuneTableOffset)); - adev->pm.dpm.dyn_state.cac_tdp_table->maximum_power_delivery_limit = 255; - pt = &ppt->power_tune_table; - } - adev->pm.dpm.dyn_state.cac_tdp_table->tdp = le16_to_cpu(pt->usTDP); - adev->pm.dpm.dyn_state.cac_tdp_table->configurable_tdp = - le16_to_cpu(pt->usConfigurableTDP); - adev->pm.dpm.dyn_state.cac_tdp_table->tdc = le16_to_cpu(pt->usTDC); - adev->pm.dpm.dyn_state.cac_tdp_table->battery_power_limit = - le16_to_cpu(pt->usBatteryPowerLimit); - adev->pm.dpm.dyn_state.cac_tdp_table->small_power_limit = - le16_to_cpu(pt->usSmallPowerLimit); - adev->pm.dpm.dyn_state.cac_tdp_table->low_cac_leakage = - le16_to_cpu(pt->usLowCACLeakage); - adev->pm.dpm.dyn_state.cac_tdp_table->high_cac_leakage = - le16_to_cpu(pt->usHighCACLeakage); - } - if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V8) && - ext_hdr->usSclkVddgfxTableOffset) { - dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) - (mode_info->atom_context->bios + data_offset + - le16_to_cpu(ext_hdr->usSclkVddgfxTableOffset)); - ret = amdgpu_parse_clk_voltage_dep_table( - &adev->pm.dpm.dyn_state.vddgfx_dependency_on_sclk, - dep_table); - if (ret) { - kfree(adev->pm.dpm.dyn_state.vddgfx_dependency_on_sclk.entries); - return ret; - } - } - } - - return 0; -} - -void amdgpu_free_extended_power_table(struct amdgpu_device *adev) -{ - struct amdgpu_dpm_dynamic_state *dyn_state = &adev->pm.dpm.dyn_state; - - kfree(dyn_state->vddc_dependency_on_sclk.entries); - kfree(dyn_state->vddci_dependency_on_mclk.entries); - kfree(dyn_state->vddc_dependency_on_mclk.entries); - kfree(dyn_state->mvdd_dependency_on_mclk.entries); - kfree(dyn_state->cac_leakage_table.entries); - kfree(dyn_state->phase_shedding_limits_table.entries); - kfree(dyn_state->ppm_table); - kfree(dyn_state->cac_tdp_table); - kfree(dyn_state->vce_clock_voltage_dependency_table.entries); - kfree(dyn_state->uvd_clock_voltage_dependency_table.entries); - kfree(dyn_state->samu_clock_voltage_dependency_table.entries); - kfree(dyn_state->acp_clock_voltage_dependency_table.entries); - kfree(dyn_state->vddgfx_dependency_on_sclk.entries); -} - -static const char *pp_lib_thermal_controller_names[] = { - "NONE", - "lm63", - "adm1032", - "adm1030", - "max6649", - "lm64", - "f75375", - "RV6xx", - "RV770", - "adt7473", - "NONE", - "External GPIO", - "Evergreen", - "emc2103", - "Sumo", - "Northern Islands", - "Southern Islands", - "lm96163", - "Sea Islands", - "Kaveri/Kabini", -}; - -void amdgpu_add_thermal_controller(struct amdgpu_device *adev) -{ - struct amdgpu_mode_info *mode_info = &adev->mode_info; - ATOM_PPLIB_POWERPLAYTABLE *power_table; - int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo); - ATOM_PPLIB_THERMALCONTROLLER *controller; - struct amdgpu_i2c_bus_rec i2c_bus; - u16 data_offset; - u8 frev, crev; - - if (!amdgpu_atom_parse_data_header(mode_info->atom_context, index, NULL, - &frev, &crev, &data_offset)) - return; - power_table = (ATOM_PPLIB_POWERPLAYTABLE *) - (mode_info->atom_context->bios + data_offset); - controller = &power_table->sThermalController; - - /* add the i2c bus for thermal/fan chip */ - if (controller->ucType > 0) { - if (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) - adev->pm.no_fan = true; - adev->pm.fan_pulses_per_revolution = - controller->ucFanParameters & ATOM_PP_FANPARAMETERS_TACHOMETER_PULSES_PER_REVOLUTION_MASK; - if (adev->pm.fan_pulses_per_revolution) { - adev->pm.fan_min_rpm = controller->ucFanMinRPM; - adev->pm.fan_max_rpm = controller->ucFanMaxRPM; - } - if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV6xx) { - DRM_INFO("Internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_RV6XX; - } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV770) { - DRM_INFO("Internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_RV770; - } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EVERGREEN) { - DRM_INFO("Internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_EVERGREEN; - } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_SUMO) { - DRM_INFO("Internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_SUMO; - } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_NISLANDS) { - DRM_INFO("Internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_NI; - } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_SISLANDS) { - DRM_INFO("Internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_SI; - } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_CISLANDS) { - DRM_INFO("Internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_CI; - } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_KAVERI) { - DRM_INFO("Internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_KV; - } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO) { - DRM_INFO("External GPIO thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_EXTERNAL_GPIO; - } else if (controller->ucType == - ATOM_PP_THERMALCONTROLLER_ADT7473_WITH_INTERNAL) { - DRM_INFO("ADT7473 with internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_ADT7473_WITH_INTERNAL; - } else if (controller->ucType == - ATOM_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL) { - DRM_INFO("EMC2103 with internal thermal controller %s fan control\n", - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_EMC2103_WITH_INTERNAL; - } else if (controller->ucType < ARRAY_SIZE(pp_lib_thermal_controller_names)) { - DRM_INFO("Possible %s thermal controller at 0x%02x %s fan control\n", - pp_lib_thermal_controller_names[controller->ucType], - controller->ucI2cAddress >> 1, - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - adev->pm.int_thermal_type = THERMAL_TYPE_EXTERNAL; - i2c_bus = amdgpu_atombios_lookup_i2c_gpio(adev, controller->ucI2cLine); - adev->pm.i2c_bus = amdgpu_i2c_lookup(adev, &i2c_bus); - if (adev->pm.i2c_bus) { - struct i2c_board_info info = { }; - const char *name = pp_lib_thermal_controller_names[controller->ucType]; - info.addr = controller->ucI2cAddress >> 1; - strlcpy(info.type, name, sizeof(info.type)); - i2c_new_client_device(&adev->pm.i2c_bus->adapter, &info); - } - } else { - DRM_INFO("Unknown thermal controller type %d at 0x%02x %s fan control\n", - controller->ucType, - controller->ucI2cAddress >> 1, - (controller->ucFanParameters & - ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); - } - } -} - -enum amdgpu_pcie_gen amdgpu_get_pcie_gen_support(struct amdgpu_device *adev, - u32 sys_mask, - enum amdgpu_pcie_gen asic_gen, - enum amdgpu_pcie_gen default_gen) +int amdgpu_dpm_get_sclk(struct amdgpu_device *adev, bool low) { - switch (asic_gen) { - case AMDGPU_PCIE_GEN1: - return AMDGPU_PCIE_GEN1; - case AMDGPU_PCIE_GEN2: - return AMDGPU_PCIE_GEN2; - case AMDGPU_PCIE_GEN3: - return AMDGPU_PCIE_GEN3; - default: - if ((sys_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) && - (default_gen == AMDGPU_PCIE_GEN3)) - return AMDGPU_PCIE_GEN3; - else if ((sys_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2) && - (default_gen == AMDGPU_PCIE_GEN2)) - return AMDGPU_PCIE_GEN2; - else - return AMDGPU_PCIE_GEN1; - } - return AMDGPU_PCIE_GEN1; -} + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; -struct amd_vce_state* -amdgpu_get_vce_clock_state(void *handle, u32 idx) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (!pp_funcs->get_sclk) + return 0; - if (idx < adev->pm.dpm.num_of_vce_states) - return &adev->pm.dpm.vce_states[idx]; + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_sclk((adev)->powerplay.pp_handle, + low); + mutex_unlock(&adev->pm.mutex); - return NULL; + return ret; } -int amdgpu_dpm_get_sclk(struct amdgpu_device *adev, bool low) +int amdgpu_dpm_get_mclk(struct amdgpu_device *adev, bool low) { const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; - return pp_funcs->get_sclk((adev)->powerplay.pp_handle, (low)); -} + if (!pp_funcs->get_mclk) + return 0; -int amdgpu_dpm_get_mclk(struct amdgpu_device *adev, bool low) -{ - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_mclk((adev)->powerplay.pp_handle, + low); + mutex_unlock(&adev->pm.mutex); - return pp_funcs->get_mclk((adev)->powerplay.pp_handle, (low)); + return ret; } int amdgpu_dpm_set_powergating_by_smu(struct amdgpu_device *adev, uint32_t block_type, bool gate) @@ -935,52 +80,20 @@ int amdgpu_dpm_set_powergating_by_smu(struct amdgpu_device *adev, uint32_t block return 0; } + mutex_lock(&adev->pm.mutex); + switch (block_type) { case AMD_IP_BLOCK_TYPE_UVD: case AMD_IP_BLOCK_TYPE_VCE: - if (pp_funcs && pp_funcs->set_powergating_by_smu) { - /* - * TODO: need a better lock mechanism - * - * Here adev->pm.mutex lock protection is enforced on - * UVD and VCE cases only. Since for other cases, there - * may be already lock protection in amdgpu_pm.c. - * This is a quick fix for the deadlock issue below. - * NFO: task ocltst:2028 blocked for more than 120 seconds. - * Tainted: G OE 5.0.0-37-generic #40~18.04.1-Ubuntu - * echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. - * cltst D 0 2028 2026 0x00000000 - * all Trace: - * __schedule+0x2c0/0x870 - * schedule+0x2c/0x70 - * schedule_preempt_disabled+0xe/0x10 - * __mutex_lock.isra.9+0x26d/0x4e0 - * __mutex_lock_slowpath+0x13/0x20 - * ? __mutex_lock_slowpath+0x13/0x20 - * mutex_lock+0x2f/0x40 - * amdgpu_dpm_set_powergating_by_smu+0x64/0xe0 [amdgpu] - * gfx_v8_0_enable_gfx_static_mg_power_gating+0x3c/0x70 [amdgpu] - * gfx_v8_0_set_powergating_state+0x66/0x260 [amdgpu] - * amdgpu_device_ip_set_powergating_state+0x62/0xb0 [amdgpu] - * pp_dpm_force_performance_level+0xe7/0x100 [amdgpu] - * amdgpu_set_dpm_forced_performance_level+0x129/0x330 [amdgpu] - */ - mutex_lock(&adev->pm.mutex); - ret = (pp_funcs->set_powergating_by_smu( - (adev)->powerplay.pp_handle, block_type, gate)); - mutex_unlock(&adev->pm.mutex); - } - break; case AMD_IP_BLOCK_TYPE_GFX: case AMD_IP_BLOCK_TYPE_VCN: case AMD_IP_BLOCK_TYPE_SDMA: case AMD_IP_BLOCK_TYPE_JPEG: case AMD_IP_BLOCK_TYPE_GMC: case AMD_IP_BLOCK_TYPE_ACP: - if (pp_funcs && pp_funcs->set_powergating_by_smu) { + if (pp_funcs && pp_funcs->set_powergating_by_smu) ret = (pp_funcs->set_powergating_by_smu( (adev)->powerplay.pp_handle, block_type, gate)); - } break; default: break; @@ -989,6 +102,8 @@ int amdgpu_dpm_set_powergating_by_smu(struct amdgpu_device *adev, uint32_t block if (!ret) atomic_set(&adev->pm.pwr_state[block_type], pwr_state); + mutex_unlock(&adev->pm.mutex); + return ret; } @@ -1001,9 +116,13 @@ int amdgpu_dpm_baco_enter(struct amdgpu_device *adev) if (!pp_funcs || !pp_funcs->set_asic_baco_state) return -ENOENT; + mutex_lock(&adev->pm.mutex); + /* enter BACO state */ ret = pp_funcs->set_asic_baco_state(pp_handle, 1); + mutex_unlock(&adev->pm.mutex); + return ret; } @@ -1016,9 +135,13 @@ int amdgpu_dpm_baco_exit(struct amdgpu_device *adev) if (!pp_funcs || !pp_funcs->set_asic_baco_state) return -ENOENT; + mutex_lock(&adev->pm.mutex); + /* exit BACO state */ ret = pp_funcs->set_asic_baco_state(pp_handle, 0); + mutex_unlock(&adev->pm.mutex); + return ret; } @@ -1029,9 +152,13 @@ int amdgpu_dpm_set_mp1_state(struct amdgpu_device *adev, const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; if (pp_funcs && pp_funcs->set_mp1_state) { + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_mp1_state( adev->powerplay.pp_handle, mp1_state); + + mutex_unlock(&adev->pm.mutex); } return ret; @@ -1042,25 +169,37 @@ bool amdgpu_dpm_is_baco_supported(struct amdgpu_device *adev) const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; void *pp_handle = adev->powerplay.pp_handle; bool baco_cap; + int ret = 0; if (!pp_funcs || !pp_funcs->get_asic_baco_capability) return false; - if (pp_funcs->get_asic_baco_capability(pp_handle, &baco_cap)) - return false; + mutex_lock(&adev->pm.mutex); - return baco_cap; + ret = pp_funcs->get_asic_baco_capability(pp_handle, + &baco_cap); + + mutex_unlock(&adev->pm.mutex); + + return ret ? false : baco_cap; } int amdgpu_dpm_mode2_reset(struct amdgpu_device *adev) { const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; void *pp_handle = adev->powerplay.pp_handle; + int ret = 0; if (!pp_funcs || !pp_funcs->asic_reset_mode_2) return -ENOENT; - return pp_funcs->asic_reset_mode_2(pp_handle); + mutex_lock(&adev->pm.mutex); + + ret = pp_funcs->asic_reset_mode_2(pp_handle); + + mutex_unlock(&adev->pm.mutex); + + return ret; } int amdgpu_dpm_baco_reset(struct amdgpu_device *adev) @@ -1072,37 +211,47 @@ int amdgpu_dpm_baco_reset(struct amdgpu_device *adev) if (!pp_funcs || !pp_funcs->set_asic_baco_state) return -ENOENT; + mutex_lock(&adev->pm.mutex); + /* enter BACO state */ ret = pp_funcs->set_asic_baco_state(pp_handle, 1); if (ret) - return ret; + goto out; /* exit BACO state */ ret = pp_funcs->set_asic_baco_state(pp_handle, 0); - if (ret) - return ret; - return 0; +out: + mutex_unlock(&adev->pm.mutex); + return ret; } bool amdgpu_dpm_is_mode1_reset_supported(struct amdgpu_device *adev) { - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; + bool support_mode1_reset = false; - if (is_support_sw_smu(adev)) - return smu_mode1_reset_is_support(smu); + if (is_support_sw_smu(adev)) { + mutex_lock(&adev->pm.mutex); + support_mode1_reset = smu_mode1_reset_is_support(smu); + mutex_unlock(&adev->pm.mutex); + } - return false; + return support_mode1_reset; } int amdgpu_dpm_mode1_reset(struct amdgpu_device *adev) { - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; + int ret = -EOPNOTSUPP; - if (is_support_sw_smu(adev)) - return smu_mode1_reset(smu); + if (is_support_sw_smu(adev)) { + mutex_lock(&adev->pm.mutex); + ret = smu_mode1_reset(smu); + mutex_unlock(&adev->pm.mutex); + } - return -EOPNOTSUPP; + return ret; } int amdgpu_dpm_switch_power_profile(struct amdgpu_device *adev, @@ -1115,9 +264,12 @@ int amdgpu_dpm_switch_power_profile(struct amdgpu_device *adev, if (amdgpu_sriov_vf(adev)) return 0; - if (pp_funcs && pp_funcs->switch_power_profile) + if (pp_funcs && pp_funcs->switch_power_profile) { + mutex_lock(&adev->pm.mutex); ret = pp_funcs->switch_power_profile( adev->powerplay.pp_handle, type, en); + mutex_unlock(&adev->pm.mutex); + } return ret; } @@ -1128,9 +280,12 @@ int amdgpu_dpm_set_xgmi_pstate(struct amdgpu_device *adev, const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; int ret = 0; - if (pp_funcs && pp_funcs->set_xgmi_pstate) + if (pp_funcs && pp_funcs->set_xgmi_pstate) { + mutex_lock(&adev->pm.mutex); ret = pp_funcs->set_xgmi_pstate(adev->powerplay.pp_handle, pstate); + mutex_unlock(&adev->pm.mutex); + } return ret; } @@ -1142,20 +297,27 @@ int amdgpu_dpm_set_df_cstate(struct amdgpu_device *adev, const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; void *pp_handle = adev->powerplay.pp_handle; - if (pp_funcs && pp_funcs->set_df_cstate) + if (pp_funcs && pp_funcs->set_df_cstate) { + mutex_lock(&adev->pm.mutex); ret = pp_funcs->set_df_cstate(pp_handle, cstate); + mutex_unlock(&adev->pm.mutex); + } return ret; } int amdgpu_dpm_allow_xgmi_power_down(struct amdgpu_device *adev, bool en) { - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; + int ret = 0; - if (is_support_sw_smu(adev)) - return smu_allow_xgmi_power_down(smu, en); + if (is_support_sw_smu(adev)) { + mutex_lock(&adev->pm.mutex); + ret = smu_allow_xgmi_power_down(smu, en); + mutex_unlock(&adev->pm.mutex); + } - return 0; + return ret; } int amdgpu_dpm_enable_mgpu_fan_boost(struct amdgpu_device *adev) @@ -1165,8 +327,11 @@ int amdgpu_dpm_enable_mgpu_fan_boost(struct amdgpu_device *adev) adev->powerplay.pp_funcs; int ret = 0; - if (pp_funcs && pp_funcs->enable_mgpu_fan_boost) + if (pp_funcs && pp_funcs->enable_mgpu_fan_boost) { + mutex_lock(&adev->pm.mutex); ret = pp_funcs->enable_mgpu_fan_boost(pp_handle); + mutex_unlock(&adev->pm.mutex); + } return ret; } @@ -1179,9 +344,12 @@ int amdgpu_dpm_set_clockgating_by_smu(struct amdgpu_device *adev, adev->powerplay.pp_funcs; int ret = 0; - if (pp_funcs && pp_funcs->set_clockgating_by_smu) + if (pp_funcs && pp_funcs->set_clockgating_by_smu) { + mutex_lock(&adev->pm.mutex); ret = pp_funcs->set_clockgating_by_smu(pp_handle, msg_id); + mutex_unlock(&adev->pm.mutex); + } return ret; } @@ -1194,9 +362,12 @@ int amdgpu_dpm_smu_i2c_bus_access(struct amdgpu_device *adev, adev->powerplay.pp_funcs; int ret = -EOPNOTSUPP; - if (pp_funcs && pp_funcs->smu_i2c_bus_access) + if (pp_funcs && pp_funcs->smu_i2c_bus_access) { + mutex_lock(&adev->pm.mutex); ret = pp_funcs->smu_i2c_bus_access(pp_handle, acquire); + mutex_unlock(&adev->pm.mutex); + } return ret; } @@ -1209,13 +380,15 @@ void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev) adev->pm.ac_power = true; else adev->pm.ac_power = false; + if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->enable_bapm) amdgpu_dpm_enable_bapm(adev, adev->pm.ac_power); - mutex_unlock(&adev->pm.mutex); if (is_support_sw_smu(adev)) - smu_set_ac_dc(&adev->smu); + smu_set_ac_dc(adev->powerplay.pp_handle); + + mutex_unlock(&adev->pm.mutex); } } @@ -1223,394 +396,1219 @@ int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors senso void *data, uint32_t *size) { const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - int ret = 0; + int ret = -EINVAL; if (!data || !size) return -EINVAL; - if (pp_funcs && pp_funcs->read_sensor) - ret = pp_funcs->read_sensor((adev)->powerplay.pp_handle, - sensor, data, size); - else - ret = -EINVAL; + if (pp_funcs && pp_funcs->read_sensor) { + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->read_sensor(adev->powerplay.pp_handle, + sensor, + data, + size); + mutex_unlock(&adev->pm.mutex); + } return ret; } -void amdgpu_dpm_thermal_work_handler(struct work_struct *work) +void amdgpu_dpm_compute_clocks(struct amdgpu_device *adev) { - struct amdgpu_device *adev = - container_of(work, struct amdgpu_device, - pm.dpm.thermal.work); - /* switch to the thermal state */ - enum amd_pm_state_type dpm_state = POWER_STATE_TYPE_INTERNAL_THERMAL; - int temp, size = sizeof(temp); + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; if (!adev->pm.dpm_enabled) return; - if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_TEMP, - (void *)&temp, &size)) { - if (temp < adev->pm.dpm.thermal.min_temp) - /* switch back the user state */ - dpm_state = adev->pm.dpm.user_state; - } else { - if (adev->pm.dpm.thermal.high_to_low) - /* switch back the user state */ - dpm_state = adev->pm.dpm.user_state; - } + if (!pp_funcs->pm_compute_clocks) + return; + mutex_lock(&adev->pm.mutex); - if (dpm_state == POWER_STATE_TYPE_INTERNAL_THERMAL) - adev->pm.dpm.thermal_active = true; - else - adev->pm.dpm.thermal_active = false; - adev->pm.dpm.state = dpm_state; + pp_funcs->pm_compute_clocks(adev->powerplay.pp_handle); mutex_unlock(&adev->pm.mutex); +} + +void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable) +{ + int ret = 0; - amdgpu_pm_compute_clocks(adev); + ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_UVD, !enable); + if (ret) + DRM_ERROR("Dpm %s uvd failed, ret = %d. \n", + enable ? "enable" : "disable", ret); } -static struct amdgpu_ps *amdgpu_dpm_pick_power_state(struct amdgpu_device *adev, - enum amd_pm_state_type dpm_state) +void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable) { - int i; - struct amdgpu_ps *ps; - u32 ui_class; - bool single_display = (adev->pm.dpm.new_active_crtc_count < 2) ? - true : false; + int ret = 0; - /* check if the vblank period is too short to adjust the mclk */ - if (single_display && adev->powerplay.pp_funcs->vblank_too_short) { - if (amdgpu_dpm_vblank_too_short(adev)) - single_display = false; - } + ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_VCE, !enable); + if (ret) + DRM_ERROR("Dpm %s vce failed, ret = %d. \n", + enable ? "enable" : "disable", ret); +} - /* certain older asics have a separare 3D performance state, - * so try that first if the user selected performance - */ - if (dpm_state == POWER_STATE_TYPE_PERFORMANCE) - dpm_state = POWER_STATE_TYPE_INTERNAL_3DPERF; - /* balanced states don't exist at the moment */ - if (dpm_state == POWER_STATE_TYPE_BALANCED) - dpm_state = POWER_STATE_TYPE_PERFORMANCE; - -restart_search: - /* Pick the best power state based on current conditions */ - for (i = 0; i < adev->pm.dpm.num_ps; i++) { - ps = &adev->pm.dpm.ps[i]; - ui_class = ps->class & ATOM_PPLIB_CLASSIFICATION_UI_MASK; - switch (dpm_state) { - /* user states */ - case POWER_STATE_TYPE_BATTERY: - if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_BATTERY) { - if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) { - if (single_display) - return ps; - } else - return ps; - } - break; - case POWER_STATE_TYPE_BALANCED: - if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_BALANCED) { - if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) { - if (single_display) - return ps; - } else - return ps; - } - break; - case POWER_STATE_TYPE_PERFORMANCE: - if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE) { - if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) { - if (single_display) - return ps; - } else - return ps; - } - break; - /* internal states */ - case POWER_STATE_TYPE_INTERNAL_UVD: - if (adev->pm.dpm.uvd_ps) - return adev->pm.dpm.uvd_ps; - else - break; - case POWER_STATE_TYPE_INTERNAL_UVD_SD: - if (ps->class & ATOM_PPLIB_CLASSIFICATION_SDSTATE) - return ps; - break; - case POWER_STATE_TYPE_INTERNAL_UVD_HD: - if (ps->class & ATOM_PPLIB_CLASSIFICATION_HDSTATE) - return ps; - break; - case POWER_STATE_TYPE_INTERNAL_UVD_HD2: - if (ps->class & ATOM_PPLIB_CLASSIFICATION_HD2STATE) - return ps; - break; - case POWER_STATE_TYPE_INTERNAL_UVD_MVC: - if (ps->class2 & ATOM_PPLIB_CLASSIFICATION2_MVC) - return ps; - break; - case POWER_STATE_TYPE_INTERNAL_BOOT: - return adev->pm.dpm.boot_ps; - case POWER_STATE_TYPE_INTERNAL_THERMAL: - if (ps->class & ATOM_PPLIB_CLASSIFICATION_THERMAL) - return ps; - break; - case POWER_STATE_TYPE_INTERNAL_ACPI: - if (ps->class & ATOM_PPLIB_CLASSIFICATION_ACPI) - return ps; - break; - case POWER_STATE_TYPE_INTERNAL_ULV: - if (ps->class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) - return ps; - break; - case POWER_STATE_TYPE_INTERNAL_3DPERF: - if (ps->class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE) - return ps; - break; - default: - break; - } - } - /* use a fallback state if we didn't match */ - switch (dpm_state) { - case POWER_STATE_TYPE_INTERNAL_UVD_SD: - dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD; - goto restart_search; - case POWER_STATE_TYPE_INTERNAL_UVD_HD: - case POWER_STATE_TYPE_INTERNAL_UVD_HD2: - case POWER_STATE_TYPE_INTERNAL_UVD_MVC: - if (adev->pm.dpm.uvd_ps) { - return adev->pm.dpm.uvd_ps; - } else { - dpm_state = POWER_STATE_TYPE_PERFORMANCE; - goto restart_search; - } - case POWER_STATE_TYPE_INTERNAL_THERMAL: - dpm_state = POWER_STATE_TYPE_INTERNAL_ACPI; - goto restart_search; - case POWER_STATE_TYPE_INTERNAL_ACPI: - dpm_state = POWER_STATE_TYPE_BATTERY; - goto restart_search; - case POWER_STATE_TYPE_BATTERY: - case POWER_STATE_TYPE_BALANCED: - case POWER_STATE_TYPE_INTERNAL_3DPERF: - dpm_state = POWER_STATE_TYPE_PERFORMANCE; - goto restart_search; - default: - break; - } +void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable) +{ + int ret = 0; - return NULL; + ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_JPEG, !enable); + if (ret) + DRM_ERROR("Dpm %s jpeg failed, ret = %d. \n", + enable ? "enable" : "disable", ret); } -static void amdgpu_dpm_change_power_state_locked(struct amdgpu_device *adev) +int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version) { - struct amdgpu_ps *ps; - enum amd_pm_state_type dpm_state; - int ret; - bool equal = false; + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int r = 0; - /* if dpm init failed */ - if (!adev->pm.dpm_enabled) - return; + if (!pp_funcs || !pp_funcs->load_firmware) + return 0; - if (adev->pm.dpm.user_state != adev->pm.dpm.state) { - /* add other state override checks here */ - if ((!adev->pm.dpm.thermal_active) && - (!adev->pm.dpm.uvd_active)) - adev->pm.dpm.state = adev->pm.dpm.user_state; + mutex_lock(&adev->pm.mutex); + r = pp_funcs->load_firmware(adev->powerplay.pp_handle); + if (r) { + pr_err("smu firmware loading failed\n"); + goto out; } - dpm_state = adev->pm.dpm.state; - ps = amdgpu_dpm_pick_power_state(adev, dpm_state); - if (ps) - adev->pm.dpm.requested_ps = ps; - else - return; + if (smu_version) + *smu_version = adev->pm.fw_version; - if (amdgpu_dpm == 1 && adev->powerplay.pp_funcs->print_power_state) { - printk("switching from power state:\n"); - amdgpu_dpm_print_power_state(adev, adev->pm.dpm.current_ps); - printk("switching to power state:\n"); - amdgpu_dpm_print_power_state(adev, adev->pm.dpm.requested_ps); +out: + mutex_unlock(&adev->pm.mutex); + return r; +} + +int amdgpu_dpm_handle_passthrough_sbr(struct amdgpu_device *adev, bool enable) +{ + int ret = 0; + + if (is_support_sw_smu(adev)) { + mutex_lock(&adev->pm.mutex); + ret = smu_handle_passthrough_sbr(adev->powerplay.pp_handle, + enable); + mutex_unlock(&adev->pm.mutex); } - /* update whether vce is active */ - ps->vce_active = adev->pm.dpm.vce_active; - if (adev->powerplay.pp_funcs->display_configuration_changed) - amdgpu_dpm_display_configuration_changed(adev); + return ret; +} - ret = amdgpu_dpm_pre_set_power_state(adev); - if (ret) - return; +int amdgpu_dpm_send_hbm_bad_pages_num(struct amdgpu_device *adev, uint32_t size) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + int ret = 0; - if (adev->powerplay.pp_funcs->check_state_equal) { - if (0 != amdgpu_dpm_check_state_equal(adev, adev->pm.dpm.current_ps, adev->pm.dpm.requested_ps, &equal)) - equal = false; - } + mutex_lock(&adev->pm.mutex); + ret = smu_send_hbm_bad_pages_num(smu, size); + mutex_unlock(&adev->pm.mutex); - if (equal) - return; + return ret; +} - amdgpu_dpm_set_power_state(adev); - amdgpu_dpm_post_set_power_state(adev); - - adev->pm.dpm.current_active_crtcs = adev->pm.dpm.new_active_crtcs; - adev->pm.dpm.current_active_crtc_count = adev->pm.dpm.new_active_crtc_count; - - if (adev->powerplay.pp_funcs->force_performance_level) { - if (adev->pm.dpm.thermal_active) { - enum amd_dpm_forced_level level = adev->pm.dpm.forced_level; - /* force low perf level for thermal */ - amdgpu_dpm_force_performance_level(adev, AMD_DPM_FORCED_LEVEL_LOW); - /* save the user's level */ - adev->pm.dpm.forced_level = level; - } else { - /* otherwise, user selected level */ - amdgpu_dpm_force_performance_level(adev, adev->pm.dpm.forced_level); - } +int amdgpu_dpm_get_dpm_freq_range(struct amdgpu_device *adev, + enum pp_clock_type type, + uint32_t *min, + uint32_t *max) +{ + int ret = 0; + + if (type != PP_SCLK) + return -EINVAL; + + if (!is_support_sw_smu(adev)) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = smu_get_dpm_freq_range(adev->powerplay.pp_handle, + SMU_SCLK, + min, + max); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_soft_freq_range(struct amdgpu_device *adev, + enum pp_clock_type type, + uint32_t min, + uint32_t max) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + int ret = 0; + + if (type != PP_SCLK) + return -EINVAL; + + if (!is_support_sw_smu(adev)) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = smu_set_soft_freq_range(smu, + SMU_SCLK, + min, + max); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_write_watermarks_table(struct amdgpu_device *adev) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + int ret = 0; + + if (!is_support_sw_smu(adev)) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = smu_write_watermarks_table(smu); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_wait_for_event(struct amdgpu_device *adev, + enum smu_event_type event, + uint64_t event_arg) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + int ret = 0; + + if (!is_support_sw_smu(adev)) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = smu_wait_for_event(smu, event, event_arg); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_status_gfxoff(struct amdgpu_device *adev, uint32_t *value) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + int ret = 0; + + if (!is_support_sw_smu(adev)) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = smu_get_status_gfxoff(smu, value); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +uint64_t amdgpu_dpm_get_thermal_throttling_counter(struct amdgpu_device *adev) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + + if (!is_support_sw_smu(adev)) + return 0; + + return atomic64_read(&smu->throttle_int_counter); +} + +/* amdgpu_dpm_gfx_state_change - Handle gfx power state change set + * @adev: amdgpu_device pointer + * @state: gfx power state(1 -sGpuChangeState_D0Entry and 2 -sGpuChangeState_D3Entry) + * + */ +void amdgpu_dpm_gfx_state_change(struct amdgpu_device *adev, + enum gfx_change_state state) +{ + mutex_lock(&adev->pm.mutex); + if (adev->powerplay.pp_funcs && + adev->powerplay.pp_funcs->gfx_state_change_set) + ((adev)->powerplay.pp_funcs->gfx_state_change_set( + (adev)->powerplay.pp_handle, state)); + mutex_unlock(&adev->pm.mutex); +} + +int amdgpu_dpm_get_ecc_info(struct amdgpu_device *adev, + void *umc_ecc) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + + if (!is_support_sw_smu(adev)) + return -EOPNOTSUPP; + + return smu_get_ecc_info(smu, umc_ecc); +} + +struct amd_vce_state *amdgpu_dpm_get_vce_clock_state(struct amdgpu_device *adev, + uint32_t idx) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + struct amd_vce_state *vstate = NULL; + + if (!pp_funcs->get_vce_clock_state) + return NULL; + + mutex_lock(&adev->pm.mutex); + vstate = pp_funcs->get_vce_clock_state(adev->powerplay.pp_handle, + idx); + mutex_unlock(&adev->pm.mutex); + + return vstate; +} + +void amdgpu_dpm_get_current_power_state(struct amdgpu_device *adev, + enum amd_pm_state_type *state) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + + mutex_lock(&adev->pm.mutex); + + if (!pp_funcs->get_current_power_state) { + *state = adev->pm.dpm.user_state; + goto out; } + + *state = pp_funcs->get_current_power_state(adev->powerplay.pp_handle); + if (*state < POWER_STATE_TYPE_DEFAULT || + *state > POWER_STATE_TYPE_INTERNAL_3DPERF) + *state = adev->pm.dpm.user_state; + +out: + mutex_unlock(&adev->pm.mutex); } -void amdgpu_pm_compute_clocks(struct amdgpu_device *adev) +void amdgpu_dpm_set_power_state(struct amdgpu_device *adev, + enum amd_pm_state_type state) { - int i = 0; + mutex_lock(&adev->pm.mutex); + adev->pm.dpm.user_state = state; + mutex_unlock(&adev->pm.mutex); - if (!adev->pm.dpm_enabled) + if (is_support_sw_smu(adev)) return; - if (adev->mode_info.num_crtc) - amdgpu_display_bandwidth_update(adev); + if (amdgpu_dpm_dispatch_task(adev, + AMD_PP_TASK_ENABLE_USER_STATE, + &state) == -EOPNOTSUPP) + amdgpu_dpm_compute_clocks(adev); +} - for (i = 0; i < AMDGPU_MAX_RINGS; i++) { - struct amdgpu_ring *ring = adev->rings[i]; - if (ring && ring->sched.ready) - amdgpu_fence_wait_empty(ring); - } +enum amd_dpm_forced_level amdgpu_dpm_get_performance_level(struct amdgpu_device *adev) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + enum amd_dpm_forced_level level; - if (adev->powerplay.pp_funcs->dispatch_tasks) { - if (!amdgpu_device_has_dc_support(adev)) { - mutex_lock(&adev->pm.mutex); - amdgpu_dpm_get_active_displays(adev); - adev->pm.pm_display_cfg.num_display = adev->pm.dpm.new_active_crtc_count; - adev->pm.pm_display_cfg.vrefresh = amdgpu_dpm_get_vrefresh(adev); - adev->pm.pm_display_cfg.min_vblank_time = amdgpu_dpm_get_vblank_time(adev); - /* we have issues with mclk switching with - * refresh rates over 120 hz on the non-DC code. - */ - if (adev->pm.pm_display_cfg.vrefresh > 120) - adev->pm.pm_display_cfg.min_vblank_time = 0; - if (adev->powerplay.pp_funcs->display_configuration_change) - adev->powerplay.pp_funcs->display_configuration_change( - adev->powerplay.pp_handle, - &adev->pm.pm_display_cfg); - mutex_unlock(&adev->pm.mutex); + mutex_lock(&adev->pm.mutex); + if (pp_funcs->get_performance_level) + level = pp_funcs->get_performance_level(adev->powerplay.pp_handle); + else + level = adev->pm.dpm.forced_level; + mutex_unlock(&adev->pm.mutex); + + return level; +} + +int amdgpu_dpm_force_performance_level(struct amdgpu_device *adev, + enum amd_dpm_forced_level level) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + enum amd_dpm_forced_level current_level; + uint32_t profile_mode_mask = AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD | + AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK | + AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK | + AMD_DPM_FORCED_LEVEL_PROFILE_PEAK; + + if (!pp_funcs->force_performance_level) + return 0; + + if (adev->pm.dpm.thermal_active) + return -EINVAL; + + current_level = amdgpu_dpm_get_performance_level(adev); + if (current_level == level) + return 0; + + if (adev->asic_type == CHIP_RAVEN) { + if (!(adev->apu_flags & AMD_APU_IS_RAVEN2)) { + if (current_level != AMD_DPM_FORCED_LEVEL_MANUAL && + level == AMD_DPM_FORCED_LEVEL_MANUAL) + amdgpu_gfx_off_ctrl(adev, false); + else if (current_level == AMD_DPM_FORCED_LEVEL_MANUAL && + level != AMD_DPM_FORCED_LEVEL_MANUAL) + amdgpu_gfx_off_ctrl(adev, true); } - amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_DISPLAY_CONFIG_CHANGE, NULL); - } else { - mutex_lock(&adev->pm.mutex); - amdgpu_dpm_get_active_displays(adev); - amdgpu_dpm_change_power_state_locked(adev); + } + + if (!(current_level & profile_mode_mask) && + (level == AMD_DPM_FORCED_LEVEL_PROFILE_EXIT)) + return -EINVAL; + + if (!(current_level & profile_mode_mask) && + (level & profile_mode_mask)) { + /* enter UMD Pstate */ + amdgpu_device_ip_set_powergating_state(adev, + AMD_IP_BLOCK_TYPE_GFX, + AMD_PG_STATE_UNGATE); + amdgpu_device_ip_set_clockgating_state(adev, + AMD_IP_BLOCK_TYPE_GFX, + AMD_CG_STATE_UNGATE); + } else if ((current_level & profile_mode_mask) && + !(level & profile_mode_mask)) { + /* exit UMD Pstate */ + amdgpu_device_ip_set_clockgating_state(adev, + AMD_IP_BLOCK_TYPE_GFX, + AMD_CG_STATE_GATE); + amdgpu_device_ip_set_powergating_state(adev, + AMD_IP_BLOCK_TYPE_GFX, + AMD_PG_STATE_GATE); + } + + mutex_lock(&adev->pm.mutex); + + if (pp_funcs->force_performance_level(adev->powerplay.pp_handle, + level)) { mutex_unlock(&adev->pm.mutex); + return -EINVAL; } + + adev->pm.dpm.forced_level = level; + + mutex_unlock(&adev->pm.mutex); + + return 0; } -void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable) +int amdgpu_dpm_get_pp_num_states(struct amdgpu_device *adev, + struct pp_states_info *states) { + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; int ret = 0; - if (adev->family == AMDGPU_FAMILY_SI) { - mutex_lock(&adev->pm.mutex); - if (enable) { - adev->pm.dpm.uvd_active = true; - adev->pm.dpm.state = POWER_STATE_TYPE_INTERNAL_UVD; - } else { - adev->pm.dpm.uvd_active = false; - } - mutex_unlock(&adev->pm.mutex); + if (!pp_funcs->get_pp_num_states) + return -EOPNOTSUPP; - amdgpu_pm_compute_clocks(adev); - } else { - ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_UVD, !enable); - if (ret) - DRM_ERROR("Dpm %s uvd failed, ret = %d. \n", - enable ? "enable" : "disable", ret); - - /* enable/disable Low Memory PState for UVD (4k videos) */ - if (adev->asic_type == CHIP_STONEY && - adev->uvd.decode_image_width >= WIDTH_4K) { - struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; - - if (hwmgr && hwmgr->hwmgr_func && - hwmgr->hwmgr_func->update_nbdpm_pstate) - hwmgr->hwmgr_func->update_nbdpm_pstate(hwmgr, - !enable, - true); - } + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_pp_num_states(adev->powerplay.pp_handle, + states); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_dispatch_task(struct amdgpu_device *adev, + enum amd_pp_task task_id, + enum amd_pm_state_type *user_state) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->dispatch_tasks) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->dispatch_tasks(adev->powerplay.pp_handle, + task_id, + user_state); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_pp_table(struct amdgpu_device *adev, char **table) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_pp_table) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_pp_table(adev->powerplay.pp_handle, + table); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_fine_grain_clk_vol(struct amdgpu_device *adev, + uint32_t type, + long *input, + uint32_t size) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_fine_grain_clk_vol) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_fine_grain_clk_vol(adev->powerplay.pp_handle, + type, + input, + size); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_odn_edit_dpm_table(struct amdgpu_device *adev, + uint32_t type, + long *input, + uint32_t size) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->odn_edit_dpm_table) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->odn_edit_dpm_table(adev->powerplay.pp_handle, + type, + input, + size); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_print_clock_levels(struct amdgpu_device *adev, + enum pp_clock_type type, + char *buf) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->print_clock_levels) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->print_clock_levels(adev->powerplay.pp_handle, + type, + buf); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_emit_clock_levels(struct amdgpu_device *adev, + enum pp_clock_type type, + char *buf, + int *offset) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->emit_clock_levels) + return -ENOENT; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->emit_clock_levels(adev->powerplay.pp_handle, + type, + buf, + offset); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_ppfeature_status(struct amdgpu_device *adev, + uint64_t ppfeature_masks) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_ppfeature_status) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_ppfeature_status(adev->powerplay.pp_handle, + ppfeature_masks); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_ppfeature_status(struct amdgpu_device *adev, char *buf) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_ppfeature_status) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_ppfeature_status(adev->powerplay.pp_handle, + buf); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_force_clock_level(struct amdgpu_device *adev, + enum pp_clock_type type, + uint32_t mask) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->force_clock_level) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->force_clock_level(adev->powerplay.pp_handle, + type, + mask); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_sclk_od(struct amdgpu_device *adev) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_sclk_od) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_sclk_od(adev->powerplay.pp_handle); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_sclk_od(struct amdgpu_device *adev, uint32_t value) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + + if (is_support_sw_smu(adev)) + return 0; + + mutex_lock(&adev->pm.mutex); + if (pp_funcs->set_sclk_od) + pp_funcs->set_sclk_od(adev->powerplay.pp_handle, value); + mutex_unlock(&adev->pm.mutex); + + if (amdgpu_dpm_dispatch_task(adev, + AMD_PP_TASK_READJUST_POWER_STATE, + NULL) == -EOPNOTSUPP) { + adev->pm.dpm.current_ps = adev->pm.dpm.boot_ps; + amdgpu_dpm_compute_clocks(adev); } + + return 0; } -void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable) +int amdgpu_dpm_get_mclk_od(struct amdgpu_device *adev) { + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; int ret = 0; - if (adev->family == AMDGPU_FAMILY_SI) { - mutex_lock(&adev->pm.mutex); - if (enable) { - adev->pm.dpm.vce_active = true; - /* XXX select vce level based on ring/task */ - adev->pm.dpm.vce_level = AMD_VCE_LEVEL_AC_ALL; - } else { - adev->pm.dpm.vce_active = false; - } - mutex_unlock(&adev->pm.mutex); + if (!pp_funcs->get_mclk_od) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_mclk_od(adev->powerplay.pp_handle); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_mclk_od(struct amdgpu_device *adev, uint32_t value) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + + if (is_support_sw_smu(adev)) + return 0; - amdgpu_pm_compute_clocks(adev); - } else { - ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_VCE, !enable); - if (ret) - DRM_ERROR("Dpm %s vce failed, ret = %d. \n", - enable ? "enable" : "disable", ret); + mutex_lock(&adev->pm.mutex); + if (pp_funcs->set_mclk_od) + pp_funcs->set_mclk_od(adev->powerplay.pp_handle, value); + mutex_unlock(&adev->pm.mutex); + + if (amdgpu_dpm_dispatch_task(adev, + AMD_PP_TASK_READJUST_POWER_STATE, + NULL) == -EOPNOTSUPP) { + adev->pm.dpm.current_ps = adev->pm.dpm.boot_ps; + amdgpu_dpm_compute_clocks(adev); } + + return 0; } -void amdgpu_pm_print_power_states(struct amdgpu_device *adev) +int amdgpu_dpm_get_power_profile_mode(struct amdgpu_device *adev, + char *buf) { - int i; + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; - if (adev->powerplay.pp_funcs->print_power_state == NULL) - return; + if (!pp_funcs->get_power_profile_mode) + return -EOPNOTSUPP; - for (i = 0; i < adev->pm.dpm.num_ps; i++) - amdgpu_dpm_print_power_state(adev, &adev->pm.dpm.ps[i]); + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_power_profile_mode(adev->powerplay.pp_handle, + buf); + mutex_unlock(&adev->pm.mutex); + return ret; } -void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable) +int amdgpu_dpm_set_power_profile_mode(struct amdgpu_device *adev, + long *input, uint32_t size) { + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; int ret = 0; - ret = amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_JPEG, !enable); - if (ret) - DRM_ERROR("Dpm %s jpeg failed, ret = %d. \n", - enable ? "enable" : "disable", ret); + if (!pp_funcs->set_power_profile_mode) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_power_profile_mode(adev->powerplay.pp_handle, + input, + size); + mutex_unlock(&adev->pm.mutex); + + return ret; } -int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version) +int amdgpu_dpm_get_gpu_metrics(struct amdgpu_device *adev, void **table) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_gpu_metrics) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_gpu_metrics(adev->powerplay.pp_handle, + table); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_fan_control_mode(struct amdgpu_device *adev, + uint32_t *fan_mode) { - int r; + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; - if (adev->powerplay.pp_funcs && adev->powerplay.pp_funcs->load_firmware) { - r = adev->powerplay.pp_funcs->load_firmware(adev->powerplay.pp_handle); - if (r) { - pr_err("smu firmware loading failed\n"); - return r; - } + if (!pp_funcs->get_fan_control_mode) + return -EOPNOTSUPP; - if (smu_version) - *smu_version = adev->pm.fw_version; - } + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_fan_control_mode(adev->powerplay.pp_handle, + fan_mode); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_fan_speed_pwm(struct amdgpu_device *adev, + uint32_t speed) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_fan_speed_pwm) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_fan_speed_pwm(adev->powerplay.pp_handle, + speed); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_fan_speed_pwm(struct amdgpu_device *adev, + uint32_t *speed) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_fan_speed_pwm) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_fan_speed_pwm(adev->powerplay.pp_handle, + speed); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_fan_speed_rpm(struct amdgpu_device *adev, + uint32_t *speed) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_fan_speed_rpm) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_fan_speed_rpm(adev->powerplay.pp_handle, + speed); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_fan_speed_rpm(struct amdgpu_device *adev, + uint32_t speed) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_fan_speed_rpm) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_fan_speed_rpm(adev->powerplay.pp_handle, + speed); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_fan_control_mode(struct amdgpu_device *adev, + uint32_t mode) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_fan_control_mode) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_fan_control_mode(adev->powerplay.pp_handle, + mode); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_power_limit(struct amdgpu_device *adev, + uint32_t *limit, + enum pp_power_limit_level pp_limit_level, + enum pp_power_type power_type) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_power_limit) + return -ENODATA; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_power_limit(adev->powerplay.pp_handle, + limit, + pp_limit_level, + power_type); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_power_limit(struct amdgpu_device *adev, + uint32_t limit) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_power_limit) + return -EINVAL; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_power_limit(adev->powerplay.pp_handle, + limit); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_is_cclk_dpm_supported(struct amdgpu_device *adev) +{ + bool cclk_dpm_supported = false; + + if (!is_support_sw_smu(adev)) + return false; + + mutex_lock(&adev->pm.mutex); + cclk_dpm_supported = is_support_cclk_dpm(adev); + mutex_unlock(&adev->pm.mutex); + + return (int)cclk_dpm_supported; +} + +int amdgpu_dpm_debugfs_print_current_performance_level(struct amdgpu_device *adev, + struct seq_file *m) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + + if (!pp_funcs->debugfs_print_current_performance_level) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + pp_funcs->debugfs_print_current_performance_level(adev->powerplay.pp_handle, + m); + mutex_unlock(&adev->pm.mutex); return 0; } + +int amdgpu_dpm_get_smu_prv_buf_details(struct amdgpu_device *adev, + void **addr, + size_t *size) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_smu_prv_buf_details) + return -ENOSYS; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_smu_prv_buf_details(adev->powerplay.pp_handle, + addr, + size); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_is_overdrive_supported(struct amdgpu_device *adev) +{ + struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; + struct smu_context *smu = adev->powerplay.pp_handle; + + if ((is_support_sw_smu(adev) && smu->od_enabled) || + (is_support_sw_smu(adev) && smu->is_apu) || + (!is_support_sw_smu(adev) && hwmgr->od_enabled)) + return true; + + return false; +} + +int amdgpu_dpm_set_pp_table(struct amdgpu_device *adev, + const char *buf, + size_t size) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_pp_table) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_pp_table(adev->powerplay.pp_handle, + buf, + size); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_num_cpu_cores(struct amdgpu_device *adev) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + + if (!is_support_sw_smu(adev)) + return INT_MAX; + + return smu->cpu_core_num; +} + +void amdgpu_dpm_stb_debug_fs_init(struct amdgpu_device *adev) +{ + if (!is_support_sw_smu(adev)) + return; + + amdgpu_smu_stb_debug_fs_init(adev); +} + +int amdgpu_dpm_display_configuration_change(struct amdgpu_device *adev, + const struct amd_pp_display_configuration *input) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->display_configuration_change) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->display_configuration_change(adev->powerplay.pp_handle, + input); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_clock_by_type(struct amdgpu_device *adev, + enum amd_pp_clock_type type, + struct amd_pp_clocks *clocks) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_clock_by_type) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_clock_by_type(adev->powerplay.pp_handle, + type, + clocks); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_display_mode_validation_clks(struct amdgpu_device *adev, + struct amd_pp_simple_clock_info *clocks) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_display_mode_validation_clocks) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_display_mode_validation_clocks(adev->powerplay.pp_handle, + clocks); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_clock_by_type_with_latency(struct amdgpu_device *adev, + enum amd_pp_clock_type type, + struct pp_clock_levels_with_latency *clocks) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_clock_by_type_with_latency) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_clock_by_type_with_latency(adev->powerplay.pp_handle, + type, + clocks); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_clock_by_type_with_voltage(struct amdgpu_device *adev, + enum amd_pp_clock_type type, + struct pp_clock_levels_with_voltage *clocks) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_clock_by_type_with_voltage) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_clock_by_type_with_voltage(adev->powerplay.pp_handle, + type, + clocks); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_watermarks_for_clocks_ranges(struct amdgpu_device *adev, + void *clock_ranges) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_watermarks_for_clocks_ranges) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_watermarks_for_clocks_ranges(adev->powerplay.pp_handle, + clock_ranges); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_display_clock_voltage_request(struct amdgpu_device *adev, + struct pp_display_clock_request *clock) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->display_clock_voltage_request) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->display_clock_voltage_request(adev->powerplay.pp_handle, + clock); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_current_clocks(struct amdgpu_device *adev, + struct amd_pp_clock_info *clocks) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_current_clocks) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_current_clocks(adev->powerplay.pp_handle, + clocks); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +void amdgpu_dpm_notify_smu_enable_pwe(struct amdgpu_device *adev) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + + if (!pp_funcs->notify_smu_enable_pwe) + return; + + mutex_lock(&adev->pm.mutex); + pp_funcs->notify_smu_enable_pwe(adev->powerplay.pp_handle); + mutex_unlock(&adev->pm.mutex); +} + +int amdgpu_dpm_set_active_display_count(struct amdgpu_device *adev, + uint32_t count) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_active_display_count) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_active_display_count(adev->powerplay.pp_handle, + count); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_set_min_deep_sleep_dcefclk(struct amdgpu_device *adev, + uint32_t clock) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->set_min_deep_sleep_dcefclk) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->set_min_deep_sleep_dcefclk(adev->powerplay.pp_handle, + clock); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +void amdgpu_dpm_set_hard_min_dcefclk_by_freq(struct amdgpu_device *adev, + uint32_t clock) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + + if (!pp_funcs->set_hard_min_dcefclk_by_freq) + return; + + mutex_lock(&adev->pm.mutex); + pp_funcs->set_hard_min_dcefclk_by_freq(adev->powerplay.pp_handle, + clock); + mutex_unlock(&adev->pm.mutex); +} + +void amdgpu_dpm_set_hard_min_fclk_by_freq(struct amdgpu_device *adev, + uint32_t clock) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + + if (!pp_funcs->set_hard_min_fclk_by_freq) + return; + + mutex_lock(&adev->pm.mutex); + pp_funcs->set_hard_min_fclk_by_freq(adev->powerplay.pp_handle, + clock); + mutex_unlock(&adev->pm.mutex); +} + +int amdgpu_dpm_display_disable_memory_clock_switch(struct amdgpu_device *adev, + bool disable_memory_clock_switch) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->display_disable_memory_clock_switch) + return 0; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->display_disable_memory_clock_switch(adev->powerplay.pp_handle, + disable_memory_clock_switch); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_max_sustainable_clocks_by_dc(struct amdgpu_device *adev, + struct pp_smu_nv_clock_table *max_clocks) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_max_sustainable_clocks_by_dc) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_max_sustainable_clocks_by_dc(adev->powerplay.pp_handle, + max_clocks); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +enum pp_smu_status amdgpu_dpm_get_uclk_dpm_states(struct amdgpu_device *adev, + unsigned int *clock_values_in_khz, + unsigned int *num_states) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_uclk_dpm_states) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_uclk_dpm_states(adev->powerplay.pp_handle, + clock_values_in_khz, + num_states); + mutex_unlock(&adev->pm.mutex); + + return ret; +} + +int amdgpu_dpm_get_dpm_clock_table(struct amdgpu_device *adev, + struct dpm_clocks *clock_table) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + int ret = 0; + + if (!pp_funcs->get_dpm_clock_table) + return -EOPNOTSUPP; + + mutex_lock(&adev->pm.mutex); + ret = pp_funcs->get_dpm_clock_table(adev->powerplay.pp_handle, + clock_table); + mutex_unlock(&adev->pm.mutex); + + return ret; +} diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c new file mode 100644 index 000000000000..42efe838fa85 --- /dev/null +++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c @@ -0,0 +1,95 @@ +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#include "amdgpu.h" +#include "amdgpu_display.h" +#include "hwmgr.h" +#include "amdgpu_smu.h" +#include "amdgpu_dpm_internal.h" + +void amdgpu_dpm_get_active_displays(struct amdgpu_device *adev) +{ + struct drm_device *ddev = adev_to_drm(adev); + struct drm_crtc *crtc; + struct amdgpu_crtc *amdgpu_crtc; + + adev->pm.dpm.new_active_crtcs = 0; + adev->pm.dpm.new_active_crtc_count = 0; + if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) { + list_for_each_entry(crtc, + &ddev->mode_config.crtc_list, head) { + amdgpu_crtc = to_amdgpu_crtc(crtc); + if (amdgpu_crtc->enabled) { + adev->pm.dpm.new_active_crtcs |= (1 << amdgpu_crtc->crtc_id); + adev->pm.dpm.new_active_crtc_count++; + } + } + } +} + +u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev) +{ + struct drm_device *dev = adev_to_drm(adev); + struct drm_crtc *crtc; + struct amdgpu_crtc *amdgpu_crtc; + u32 vblank_in_pixels; + u32 vblank_time_us = 0xffffffff; /* if the displays are off, vblank time is max */ + + if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) { + list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { + amdgpu_crtc = to_amdgpu_crtc(crtc); + if (crtc->enabled && amdgpu_crtc->enabled && amdgpu_crtc->hw_mode.clock) { + vblank_in_pixels = + amdgpu_crtc->hw_mode.crtc_htotal * + (amdgpu_crtc->hw_mode.crtc_vblank_end - + amdgpu_crtc->hw_mode.crtc_vdisplay + + (amdgpu_crtc->v_border * 2)); + + vblank_time_us = vblank_in_pixels * 1000 / amdgpu_crtc->hw_mode.clock; + break; + } + } + } + + return vblank_time_us; +} + +u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev) +{ + struct drm_device *dev = adev_to_drm(adev); + struct drm_crtc *crtc; + struct amdgpu_crtc *amdgpu_crtc; + u32 vrefresh = 0; + + if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) { + list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { + amdgpu_crtc = to_amdgpu_crtc(crtc); + if (crtc->enabled && amdgpu_crtc->enabled && amdgpu_crtc->hw_mode.clock) { + vrefresh = drm_mode_vrefresh(&amdgpu_crtc->hw_mode); + break; + } + } + } + + return vrefresh; +} diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index 48cc009d9bdf..b0243068212b 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -34,7 +34,6 @@ #include <linux/nospec.h> #include <linux/pm_runtime.h> #include <asm/processor.h> -#include "hwmgr.h" static const struct cg_flag_name clocks[] = { {AMD_CG_SUPPORT_GFX_FGCG, "Graphics Fine Grain Clock Gating"}, @@ -132,7 +131,6 @@ static ssize_t amdgpu_get_power_dpm_state(struct device *dev, { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; enum amd_pm_state_type pm; int ret; @@ -147,11 +145,7 @@ static ssize_t amdgpu_get_power_dpm_state(struct device *dev, return ret; } - if (pp_funcs->get_current_power_state) { - pm = amdgpu_dpm_get_current_power_state(adev); - } else { - pm = adev->pm.dpm.user_state; - } + amdgpu_dpm_get_current_power_state(adev, &pm); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -191,19 +185,8 @@ static ssize_t amdgpu_set_power_dpm_state(struct device *dev, return ret; } - if (is_support_sw_smu(adev)) { - mutex_lock(&adev->pm.mutex); - adev->pm.dpm.user_state = state; - mutex_unlock(&adev->pm.mutex); - } else if (adev->powerplay.pp_funcs->dispatch_tasks) { - amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_ENABLE_USER_STATE, &state); - } else { - mutex_lock(&adev->pm.mutex); - adev->pm.dpm.user_state = state; - mutex_unlock(&adev->pm.mutex); + amdgpu_dpm_set_power_state(adev, state); - amdgpu_pm_compute_clocks(adev); - } pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -290,10 +273,7 @@ static ssize_t amdgpu_get_power_dpm_force_performance_level(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->get_performance_level) - level = amdgpu_dpm_get_performance_level(adev); - else - level = adev->pm.dpm.forced_level; + level = amdgpu_dpm_get_performance_level(adev); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -318,9 +298,7 @@ static ssize_t amdgpu_set_power_dpm_force_performance_level(struct device *dev, { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; enum amd_dpm_forced_level level; - enum amd_dpm_forced_level current_level; int ret = 0; if (amdgpu_in_reset(adev)) @@ -358,57 +336,17 @@ static ssize_t amdgpu_set_power_dpm_force_performance_level(struct device *dev, return ret; } - if (pp_funcs->get_performance_level) - current_level = amdgpu_dpm_get_performance_level(adev); - else - current_level = adev->pm.dpm.forced_level; - - if (current_level == level) { - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return count; - } - - if (adev->asic_type == CHIP_RAVEN) { - if (!(adev->apu_flags & AMD_APU_IS_RAVEN2)) { - if (current_level != AMD_DPM_FORCED_LEVEL_MANUAL && level == AMD_DPM_FORCED_LEVEL_MANUAL) - amdgpu_gfx_off_ctrl(adev, false); - else if (current_level == AMD_DPM_FORCED_LEVEL_MANUAL && level != AMD_DPM_FORCED_LEVEL_MANUAL) - amdgpu_gfx_off_ctrl(adev, true); - } - } - - /* profile_exit setting is valid only when current mode is in profile mode */ - if (!(current_level & (AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD | - AMD_DPM_FORCED_LEVEL_PROFILE_MIN_SCLK | - AMD_DPM_FORCED_LEVEL_PROFILE_MIN_MCLK | - AMD_DPM_FORCED_LEVEL_PROFILE_PEAK)) && - (level == AMD_DPM_FORCED_LEVEL_PROFILE_EXIT)) { - pr_err("Currently not in any profile mode!\n"); + mutex_lock(&adev->pm.stable_pstate_ctx_lock); + if (amdgpu_dpm_force_performance_level(adev, level)) { pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); + mutex_unlock(&adev->pm.stable_pstate_ctx_lock); return -EINVAL; } + /* override whatever a user ctx may have set */ + adev->pm.stable_pstate_ctx = NULL; + mutex_unlock(&adev->pm.stable_pstate_ctx_lock); - if (pp_funcs->force_performance_level) { - mutex_lock(&adev->pm.mutex); - if (adev->pm.dpm.thermal_active) { - mutex_unlock(&adev->pm.mutex); - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return -EINVAL; - } - ret = amdgpu_dpm_force_performance_level(adev, level); - if (ret) { - mutex_unlock(&adev->pm.mutex); - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return -EINVAL; - } else { - adev->pm.dpm.forced_level = level; - } - mutex_unlock(&adev->pm.mutex); - } pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -421,7 +359,6 @@ static ssize_t amdgpu_get_pp_num_states(struct device *dev, { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; struct pp_states_info data; uint32_t i; int buf_len, ret; @@ -437,11 +374,8 @@ static ssize_t amdgpu_get_pp_num_states(struct device *dev, return ret; } - if (pp_funcs->get_pp_num_states) { - amdgpu_dpm_get_pp_num_states(adev, &data); - } else { + if (amdgpu_dpm_get_pp_num_states(adev, &data)) memset(&data, 0, sizeof(data)); - } pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -463,7 +397,6 @@ static ssize_t amdgpu_get_pp_cur_state(struct device *dev, { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; struct pp_states_info data = {0}; enum amd_pm_state_type pm = 0; int i = 0, ret = 0; @@ -479,15 +412,16 @@ static ssize_t amdgpu_get_pp_cur_state(struct device *dev, return ret; } - if (pp_funcs->get_current_power_state - && pp_funcs->get_pp_num_states) { - pm = amdgpu_dpm_get_current_power_state(adev); - amdgpu_dpm_get_pp_num_states(adev, &data); - } + amdgpu_dpm_get_current_power_state(adev, &pm); + + ret = amdgpu_dpm_get_pp_num_states(adev, &data); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); + if (ret) + return ret; + for (i = 0; i < data.nums; i++) { if (pm == data.states[i]) break; @@ -511,7 +445,7 @@ static ssize_t amdgpu_get_pp_force_state(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - if (adev->pp_force_state_enabled) + if (adev->pm.pp_force_state_enabled) return amdgpu_get_pp_cur_state(dev, attr, buf); else return sysfs_emit(buf, "\n"); @@ -525,6 +459,7 @@ static ssize_t amdgpu_set_pp_force_state(struct device *dev, struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); enum amd_pm_state_type state = 0; + struct pp_states_info data; unsigned long idx; int ret; @@ -533,41 +468,49 @@ static ssize_t amdgpu_set_pp_force_state(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - if (strlen(buf) == 1) - adev->pp_force_state_enabled = false; - else if (is_support_sw_smu(adev)) - adev->pp_force_state_enabled = false; - else if (adev->powerplay.pp_funcs->dispatch_tasks && - adev->powerplay.pp_funcs->get_pp_num_states) { - struct pp_states_info data; - - ret = kstrtoul(buf, 0, &idx); - if (ret || idx >= ARRAY_SIZE(data.states)) - return -EINVAL; + adev->pm.pp_force_state_enabled = false; - idx = array_index_nospec(idx, ARRAY_SIZE(data.states)); + if (strlen(buf) == 1) + return count; - amdgpu_dpm_get_pp_num_states(adev, &data); - state = data.states[idx]; + ret = kstrtoul(buf, 0, &idx); + if (ret || idx >= ARRAY_SIZE(data.states)) + return -EINVAL; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + idx = array_index_nospec(idx, ARRAY_SIZE(data.states)); - /* only set user selected power states */ - if (state != POWER_STATE_TYPE_INTERNAL_BOOT && - state != POWER_STATE_TYPE_DEFAULT) { - amdgpu_dpm_dispatch_task(adev, - AMD_PP_TASK_ENABLE_USER_STATE, &state); - adev->pp_force_state_enabled = true; - } - pm_runtime_mark_last_busy(ddev->dev); + ret = pm_runtime_get_sync(ddev->dev); + if (ret < 0) { pm_runtime_put_autosuspend(ddev->dev); + return ret; + } + + ret = amdgpu_dpm_get_pp_num_states(adev, &data); + if (ret) + goto err_out; + + state = data.states[idx]; + + /* only set user selected power states */ + if (state != POWER_STATE_TYPE_INTERNAL_BOOT && + state != POWER_STATE_TYPE_DEFAULT) { + ret = amdgpu_dpm_dispatch_task(adev, + AMD_PP_TASK_ENABLE_USER_STATE, &state); + if (ret) + goto err_out; + + adev->pm.pp_force_state_enabled = true; } + pm_runtime_mark_last_busy(ddev->dev); + pm_runtime_put_autosuspend(ddev->dev); + return count; + +err_out: + pm_runtime_mark_last_busy(ddev->dev); + pm_runtime_put_autosuspend(ddev->dev); + return ret; } /** @@ -601,17 +544,13 @@ static ssize_t amdgpu_get_pp_table(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->get_pp_table) { - size = amdgpu_dpm_get_pp_table(adev, &table); - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - if (size < 0) - return size; - } else { - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return 0; - } + size = amdgpu_dpm_get_pp_table(adev, &table); + + pm_runtime_mark_last_busy(ddev->dev); + pm_runtime_put_autosuspend(ddev->dev); + + if (size <= 0) + return size; if (size >= PAGE_SIZE) size = PAGE_SIZE - 1; @@ -642,15 +581,13 @@ static ssize_t amdgpu_set_pp_table(struct device *dev, } ret = amdgpu_dpm_set_pp_table(adev, buf, count); - if (ret) { - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); + if (ret) + return ret; + return count; } @@ -866,46 +803,32 @@ static ssize_t amdgpu_set_pp_od_clk_voltage(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->set_fine_grain_clk_vol) { - ret = amdgpu_dpm_set_fine_grain_clk_vol(adev, type, - parameter, - parameter_size); - if (ret) { - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return -EINVAL; - } - } + if (amdgpu_dpm_set_fine_grain_clk_vol(adev, + type, + parameter, + parameter_size)) + goto err_out; - if (adev->powerplay.pp_funcs->odn_edit_dpm_table) { - ret = amdgpu_dpm_odn_edit_dpm_table(adev, type, - parameter, parameter_size); - if (ret) { - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return -EINVAL; - } - } + if (amdgpu_dpm_odn_edit_dpm_table(adev, type, + parameter, parameter_size)) + goto err_out; if (type == PP_OD_COMMIT_DPM_TABLE) { - if (adev->powerplay.pp_funcs->dispatch_tasks) { - amdgpu_dpm_dispatch_task(adev, - AMD_PP_TASK_READJUST_POWER_STATE, - NULL); - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return count; - } else { - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return -EINVAL; - } + if (amdgpu_dpm_dispatch_task(adev, + AMD_PP_TASK_READJUST_POWER_STATE, + NULL)) + goto err_out; } pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return count; + +err_out: + pm_runtime_mark_last_busy(ddev->dev); + pm_runtime_put_autosuspend(ddev->dev); + return -EINVAL; } static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev, @@ -914,8 +837,17 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev, { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - ssize_t size; + int size = 0; int ret; + enum pp_clock_type od_clocks[6] = { + OD_SCLK, + OD_MCLK, + OD_VDDC_CURVE, + OD_RANGE, + OD_VDDGFX_OFFSET, + OD_CCLK, + }; + uint clk_index; if (amdgpu_in_reset(adev)) return -EPERM; @@ -928,16 +860,25 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->print_clock_levels) { + for (clk_index = 0 ; clk_index < 6 ; clk_index++) { + ret = amdgpu_dpm_emit_clock_levels(adev, od_clocks[clk_index], buf, &size); + if (ret) + break; + } + if (ret == -ENOENT) { size = amdgpu_dpm_print_clock_levels(adev, OD_SCLK, buf); - size += amdgpu_dpm_print_clock_levels(adev, OD_MCLK, buf+size); - size += amdgpu_dpm_print_clock_levels(adev, OD_VDDC_CURVE, buf+size); - size += amdgpu_dpm_print_clock_levels(adev, OD_VDDGFX_OFFSET, buf+size); - size += amdgpu_dpm_print_clock_levels(adev, OD_RANGE, buf+size); - size += amdgpu_dpm_print_clock_levels(adev, OD_CCLK, buf+size); - } else { - size = sysfs_emit(buf, "\n"); + if (size > 0) { + size += amdgpu_dpm_print_clock_levels(adev, OD_MCLK, buf + size); + size += amdgpu_dpm_print_clock_levels(adev, OD_VDDC_CURVE, buf + size); + size += amdgpu_dpm_print_clock_levels(adev, OD_VDDGFX_OFFSET, buf + size); + size += amdgpu_dpm_print_clock_levels(adev, OD_RANGE, buf + size); + size += amdgpu_dpm_print_clock_levels(adev, OD_CCLK, buf + size); + } } + + if (size == 0) + size = sysfs_emit(buf, "\n"); + pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -985,17 +926,14 @@ static ssize_t amdgpu_set_pp_features(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->set_ppfeature_status) { - ret = amdgpu_dpm_set_ppfeature_status(adev, featuremask); - if (ret) { - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return -EINVAL; - } - } + ret = amdgpu_dpm_set_ppfeature_status(adev, featuremask); + pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); + if (ret) + return -EINVAL; + return count; } @@ -1019,9 +957,8 @@ static ssize_t amdgpu_get_pp_features(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->get_ppfeature_status) - size = amdgpu_dpm_get_ppfeature_status(adev, buf); - else + size = amdgpu_dpm_get_ppfeature_status(adev, buf); + if (size <= 0) size = sysfs_emit(buf, "\n"); pm_runtime_mark_last_busy(ddev->dev); @@ -1066,8 +1003,8 @@ static ssize_t amdgpu_get_pp_dpm_clock(struct device *dev, { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - ssize_t size; - int ret; + int size = 0; + int ret = 0; if (amdgpu_in_reset(adev)) return -EPERM; @@ -1080,9 +1017,11 @@ static ssize_t amdgpu_get_pp_dpm_clock(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->print_clock_levels) + ret = amdgpu_dpm_emit_clock_levels(adev, type, buf, &size); + if (ret == -ENOENT) size = amdgpu_dpm_print_clock_levels(adev, type, buf); - else + + if (size == 0) size = sysfs_emit(buf, "\n"); pm_runtime_mark_last_busy(ddev->dev); @@ -1151,10 +1090,7 @@ static ssize_t amdgpu_set_pp_dpm_clock(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->force_clock_level) - ret = amdgpu_dpm_force_clock_level(adev, type, mask); - else - ret = 0; + ret = amdgpu_dpm_force_clock_level(adev, type, mask); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -1305,10 +1241,7 @@ static ssize_t amdgpu_get_pp_sclk_od(struct device *dev, return ret; } - if (is_support_sw_smu(adev)) - value = 0; - else if (adev->powerplay.pp_funcs->get_sclk_od) - value = amdgpu_dpm_get_sclk_od(adev); + value = amdgpu_dpm_get_sclk_od(adev); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -1342,19 +1275,7 @@ static ssize_t amdgpu_set_pp_sclk_od(struct device *dev, return ret; } - if (is_support_sw_smu(adev)) { - value = 0; - } else { - if (adev->powerplay.pp_funcs->set_sclk_od) - amdgpu_dpm_set_sclk_od(adev, (uint32_t)value); - - if (adev->powerplay.pp_funcs->dispatch_tasks) { - amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_READJUST_POWER_STATE, NULL); - } else { - adev->pm.dpm.current_ps = adev->pm.dpm.boot_ps; - amdgpu_pm_compute_clocks(adev); - } - } + amdgpu_dpm_set_sclk_od(adev, (uint32_t)value); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -1382,10 +1303,7 @@ static ssize_t amdgpu_get_pp_mclk_od(struct device *dev, return ret; } - if (is_support_sw_smu(adev)) - value = 0; - else if (adev->powerplay.pp_funcs->get_mclk_od) - value = amdgpu_dpm_get_mclk_od(adev); + value = amdgpu_dpm_get_mclk_od(adev); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -1419,19 +1337,7 @@ static ssize_t amdgpu_set_pp_mclk_od(struct device *dev, return ret; } - if (is_support_sw_smu(adev)) { - value = 0; - } else { - if (adev->powerplay.pp_funcs->set_mclk_od) - amdgpu_dpm_set_mclk_od(adev, (uint32_t)value); - - if (adev->powerplay.pp_funcs->dispatch_tasks) { - amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_READJUST_POWER_STATE, NULL); - } else { - adev->pm.dpm.current_ps = adev->pm.dpm.boot_ps; - amdgpu_pm_compute_clocks(adev); - } - } + amdgpu_dpm_set_mclk_od(adev, (uint32_t)value); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -1479,9 +1385,8 @@ static ssize_t amdgpu_get_pp_power_profile_mode(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->get_power_profile_mode) - size = amdgpu_dpm_get_power_profile_mode(adev, buf); - else + size = amdgpu_dpm_get_power_profile_mode(adev, buf); + if (size <= 0) size = sysfs_emit(buf, "\n"); pm_runtime_mark_last_busy(ddev->dev); @@ -1545,8 +1450,7 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->set_power_profile_mode) - ret = amdgpu_dpm_set_power_profile_mode(adev, parameter, parameter_size); + ret = amdgpu_dpm_set_power_profile_mode(adev, parameter, parameter_size); pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); @@ -1812,9 +1716,7 @@ static ssize_t amdgpu_get_gpu_metrics(struct device *dev, return ret; } - if (adev->powerplay.pp_funcs->get_gpu_metrics) - size = amdgpu_dpm_get_gpu_metrics(adev, &gpu_metrics); - + size = amdgpu_dpm_get_gpu_metrics(adev, &gpu_metrics); if (size <= 0) goto out; @@ -2027,8 +1929,8 @@ static struct amdgpu_device_attr amdgpu_device_attrs[] = { AMDGPU_DEVICE_ATTR_RW(pp_dpm_fclk, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), AMDGPU_DEVICE_ATTR_RW(pp_dpm_vclk, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), AMDGPU_DEVICE_ATTR_RW(pp_dpm_dclk, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), - AMDGPU_DEVICE_ATTR_RW(pp_dpm_dcefclk, ATTR_FLAG_BASIC), - AMDGPU_DEVICE_ATTR_RW(pp_dpm_pcie, ATTR_FLAG_BASIC), + AMDGPU_DEVICE_ATTR_RW(pp_dpm_dcefclk, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), + AMDGPU_DEVICE_ATTR_RW(pp_dpm_pcie, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), AMDGPU_DEVICE_ATTR_RW(pp_sclk_od, ATTR_FLAG_BASIC), AMDGPU_DEVICE_ATTR_RW(pp_mclk_od, ATTR_FLAG_BASIC), AMDGPU_DEVICE_ATTR_RW(pp_power_profile_mode, ATTR_FLAG_BASIC|ATTR_FLAG_ONEVF), @@ -2053,7 +1955,6 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_ { struct device_attribute *dev_attr = &attr->dev_attr; const char *attr_name = dev_attr->attr.name; - struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; enum amd_asic_type asic_type = adev->asic_type; if (!(attr->flags & mask)) { @@ -2076,9 +1977,7 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_ *states = ATTR_STATE_UNSUPPORTED; } else if (DEVICE_ATTR_IS(pp_od_clk_voltage)) { *states = ATTR_STATE_UNSUPPORTED; - if ((is_support_sw_smu(adev) && adev->smu.od_enabled) || - (is_support_sw_smu(adev) && adev->smu.is_apu) || - (!is_support_sw_smu(adev) && hwmgr->od_enabled)) + if (amdgpu_dpm_is_overdrive_supported(adev)) *states = ATTR_STATE_SUPPORTED; } else if (DEVICE_ATTR_IS(mem_busy_percent)) { if (adev->flags & AMD_IS_APU || asic_type == CHIP_VEGA10) @@ -2106,8 +2005,7 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_ if (!(asic_type == CHIP_VANGOGH || asic_type == CHIP_SIENNA_CICHLID)) *states = ATTR_STATE_UNSUPPORTED; } else if (DEVICE_ATTR_IS(pp_power_profile_mode)) { - if (!adev->powerplay.pp_funcs->get_power_profile_mode || - amdgpu_dpm_get_power_profile_mode(adev, NULL) == -EOPNOTSUPP) + if (amdgpu_dpm_get_power_profile_mode(adev, NULL) == -EOPNOTSUPP) *states = ATTR_STATE_UNSUPPORTED; } @@ -2396,17 +2294,14 @@ static ssize_t amdgpu_hwmon_get_pwm1_enable(struct device *dev, return ret; } - if (!adev->powerplay.pp_funcs->get_fan_control_mode) { - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return -EINVAL; - } - - pwm_mode = amdgpu_dpm_get_fan_control_mode(adev); + ret = amdgpu_dpm_get_fan_control_mode(adev, &pwm_mode); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + if (ret) + return -EINVAL; + return sysfs_emit(buf, "%u\n", pwm_mode); } @@ -2434,17 +2329,14 @@ static ssize_t amdgpu_hwmon_set_pwm1_enable(struct device *dev, return ret; } - if (!adev->powerplay.pp_funcs->set_fan_control_mode) { - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return -EINVAL; - } - - amdgpu_dpm_set_fan_control_mode(adev, value); + ret = amdgpu_dpm_set_fan_control_mode(adev, value); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + if (ret) + return -EINVAL; + return count; } @@ -2476,32 +2368,29 @@ static ssize_t amdgpu_hwmon_set_pwm1(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; + err = kstrtou32(buf, 10, &value); + if (err) + return err; + err = pm_runtime_get_sync(adev_to_drm(adev)->dev); if (err < 0) { pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); return err; } - pwm_mode = amdgpu_dpm_get_fan_control_mode(adev); + err = amdgpu_dpm_get_fan_control_mode(adev, &pwm_mode); + if (err) + goto out; + if (pwm_mode != AMD_FAN_CTRL_MANUAL) { pr_info("manual fan speed control should be enabled first\n"); - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return -EINVAL; - } - - err = kstrtou32(buf, 10, &value); - if (err) { - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return err; + err = -EINVAL; + goto out; } - if (adev->powerplay.pp_funcs->set_fan_speed_pwm) - err = amdgpu_dpm_set_fan_speed_pwm(adev, value); - else - err = -EINVAL; + err = amdgpu_dpm_set_fan_speed_pwm(adev, value); +out: pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); @@ -2530,10 +2419,7 @@ static ssize_t amdgpu_hwmon_get_pwm1(struct device *dev, return err; } - if (adev->powerplay.pp_funcs->get_fan_speed_pwm) - err = amdgpu_dpm_get_fan_speed_pwm(adev, &speed); - else - err = -EINVAL; + err = amdgpu_dpm_get_fan_speed_pwm(adev, &speed); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); @@ -2563,10 +2449,7 @@ static ssize_t amdgpu_hwmon_get_fan1_input(struct device *dev, return err; } - if (adev->powerplay.pp_funcs->get_fan_speed_rpm) - err = amdgpu_dpm_get_fan_speed_rpm(adev, &speed); - else - err = -EINVAL; + err = amdgpu_dpm_get_fan_speed_rpm(adev, &speed); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); @@ -2660,10 +2543,7 @@ static ssize_t amdgpu_hwmon_get_fan1_target(struct device *dev, return err; } - if (adev->powerplay.pp_funcs->get_fan_speed_rpm) - err = amdgpu_dpm_get_fan_speed_rpm(adev, &rpm); - else - err = -EINVAL; + err = amdgpu_dpm_get_fan_speed_rpm(adev, &rpm); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); @@ -2688,32 +2568,28 @@ static ssize_t amdgpu_hwmon_set_fan1_target(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; + err = kstrtou32(buf, 10, &value); + if (err) + return err; + err = pm_runtime_get_sync(adev_to_drm(adev)->dev); if (err < 0) { pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); return err; } - pwm_mode = amdgpu_dpm_get_fan_control_mode(adev); + err = amdgpu_dpm_get_fan_control_mode(adev, &pwm_mode); + if (err) + goto out; if (pwm_mode != AMD_FAN_CTRL_MANUAL) { - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return -ENODATA; - } - - err = kstrtou32(buf, 10, &value); - if (err) { - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return err; + err = -ENODATA; + goto out; } - if (adev->powerplay.pp_funcs->set_fan_speed_rpm) - err = amdgpu_dpm_set_fan_speed_rpm(adev, value); - else - err = -EINVAL; + err = amdgpu_dpm_set_fan_speed_rpm(adev, value); +out: pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); @@ -2742,17 +2618,14 @@ static ssize_t amdgpu_hwmon_get_fan1_enable(struct device *dev, return ret; } - if (!adev->powerplay.pp_funcs->get_fan_control_mode) { - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return -EINVAL; - } - - pwm_mode = amdgpu_dpm_get_fan_control_mode(adev); + ret = amdgpu_dpm_get_fan_control_mode(adev, &pwm_mode); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + if (ret) + return -EINVAL; + return sysfs_emit(buf, "%i\n", pwm_mode == AMD_FAN_CTRL_AUTO ? 0 : 1); } @@ -2788,16 +2661,14 @@ static ssize_t amdgpu_hwmon_set_fan1_enable(struct device *dev, return err; } - if (!adev->powerplay.pp_funcs->set_fan_control_mode) { - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return -EINVAL; - } - amdgpu_dpm_set_fan_control_mode(adev, pwm_mode); + err = amdgpu_dpm_set_fan_control_mode(adev, pwm_mode); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + if (err) + return -EINVAL; + return count; } @@ -2933,7 +2804,6 @@ static ssize_t amdgpu_hwmon_show_power_cap_generic(struct device *dev, enum pp_power_limit_level pp_limit_level) { struct amdgpu_device *adev = dev_get_drvdata(dev); - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; enum pp_power_type power_type = to_sensor_dev_attr(attr)->index; uint32_t limit; ssize_t size; @@ -2944,16 +2814,13 @@ static ssize_t amdgpu_hwmon_show_power_cap_generic(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - if ( !(pp_funcs && pp_funcs->get_power_limit)) - return -ENODATA; - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); if (r < 0) { pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); return r; } - r = pp_funcs->get_power_limit(adev->powerplay.pp_handle, &limit, + r = amdgpu_dpm_get_power_limit(adev, &limit, pp_limit_level, power_type); if (!r) @@ -2996,10 +2863,14 @@ static ssize_t amdgpu_hwmon_show_power_label(struct device *dev, struct device_attribute *attr, char *buf) { - int limit_type = to_sensor_dev_attr(attr)->index; + struct amdgpu_device *adev = dev_get_drvdata(dev); - return sysfs_emit(buf, "%s\n", - limit_type == SMU_FAST_PPT_LIMIT ? "fastPPT" : "slowPPT"); + if (adev->asic_type == CHIP_VANGOGH) + return sysfs_emit(buf, "%s\n", + to_sensor_dev_attr(attr)->index == PP_PWR_TYPE_FAST ? + "fastPPT" : "slowPPT"); + else + return sysfs_emit(buf, "PPT\n"); } static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev, @@ -3008,7 +2879,6 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev, size_t count) { struct amdgpu_device *adev = dev_get_drvdata(dev); - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; int limit_type = to_sensor_dev_attr(attr)->index; int err; u32 value; @@ -3034,10 +2904,7 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev, return err; } - if (pp_funcs && pp_funcs->set_power_limit) - err = pp_funcs->set_power_limit(adev->powerplay.pp_handle, value); - else - err = -EINVAL; + err = amdgpu_dpm_set_power_limit(adev, value); pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); @@ -3315,19 +3182,6 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) return 0; - /* there is no fan under pp one vf mode */ - if (amdgpu_sriov_is_pp_one_vf(adev) && - (attr == &sensor_dev_attr_pwm1.dev_attr.attr || - attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr || - attr == &sensor_dev_attr_pwm1_max.dev_attr.attr || - attr == &sensor_dev_attr_pwm1_min.dev_attr.attr || - attr == &sensor_dev_attr_fan1_input.dev_attr.attr || - attr == &sensor_dev_attr_fan1_min.dev_attr.attr || - attr == &sensor_dev_attr_fan1_max.dev_attr.attr || - attr == &sensor_dev_attr_fan1_target.dev_attr.attr || - attr == &sensor_dev_attr_fan1_enable.dev_attr.attr)) - return 0; - /* Skip fan attributes if fan is not present */ if (adev->pm.no_fan && (attr == &sensor_dev_attr_pwm1.dev_attr.attr || attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr || @@ -3374,20 +3228,18 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, attr == &sensor_dev_attr_fan1_enable.dev_attr.attr)) return 0; - if (!is_support_sw_smu(adev)) { - /* mask fan attributes if we have no bindings for this asic to expose */ - if ((!adev->powerplay.pp_funcs->get_fan_speed_pwm && - attr == &sensor_dev_attr_pwm1.dev_attr.attr) || /* can't query fan */ - (!adev->powerplay.pp_funcs->get_fan_control_mode && - attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't query state */ - effective_mode &= ~S_IRUGO; + /* mask fan attributes if we have no bindings for this asic to expose */ + if (((amdgpu_dpm_get_fan_speed_pwm(adev, NULL) == -EOPNOTSUPP) && + attr == &sensor_dev_attr_pwm1.dev_attr.attr) || /* can't query fan */ + ((amdgpu_dpm_get_fan_control_mode(adev, NULL) == -EOPNOTSUPP) && + attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't query state */ + effective_mode &= ~S_IRUGO; - if ((!adev->powerplay.pp_funcs->set_fan_speed_pwm && - attr == &sensor_dev_attr_pwm1.dev_attr.attr) || /* can't manage fan */ - (!adev->powerplay.pp_funcs->set_fan_control_mode && - attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't manage state */ - effective_mode &= ~S_IWUSR; - } + if (((amdgpu_dpm_set_fan_speed_pwm(adev, U32_MAX) == -EOPNOTSUPP) && + attr == &sensor_dev_attr_pwm1.dev_attr.attr) || /* can't manage fan */ + ((amdgpu_dpm_set_fan_control_mode(adev, U32_MAX) == -EOPNOTSUPP) && + attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr)) /* can't manage state */ + effective_mode &= ~S_IWUSR; if (((adev->family == AMDGPU_FAMILY_SI) || ((adev->flags & AMD_IS_APU) && @@ -3404,22 +3256,20 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, (attr == &sensor_dev_attr_power1_average.dev_attr.attr)) return 0; - if (!is_support_sw_smu(adev)) { - /* hide max/min values if we can't both query and manage the fan */ - if ((!adev->powerplay.pp_funcs->set_fan_speed_pwm && - !adev->powerplay.pp_funcs->get_fan_speed_pwm) && - (!adev->powerplay.pp_funcs->set_fan_speed_rpm && - !adev->powerplay.pp_funcs->get_fan_speed_rpm) && - (attr == &sensor_dev_attr_pwm1_max.dev_attr.attr || - attr == &sensor_dev_attr_pwm1_min.dev_attr.attr)) - return 0; + /* hide max/min values if we can't both query and manage the fan */ + if (((amdgpu_dpm_set_fan_speed_pwm(adev, U32_MAX) == -EOPNOTSUPP) && + (amdgpu_dpm_get_fan_speed_pwm(adev, NULL) == -EOPNOTSUPP) && + (amdgpu_dpm_set_fan_speed_rpm(adev, U32_MAX) == -EOPNOTSUPP) && + (amdgpu_dpm_get_fan_speed_rpm(adev, NULL) == -EOPNOTSUPP)) && + (attr == &sensor_dev_attr_pwm1_max.dev_attr.attr || + attr == &sensor_dev_attr_pwm1_min.dev_attr.attr)) + return 0; - if ((!adev->powerplay.pp_funcs->set_fan_speed_rpm && - !adev->powerplay.pp_funcs->get_fan_speed_rpm) && - (attr == &sensor_dev_attr_fan1_max.dev_attr.attr || - attr == &sensor_dev_attr_fan1_min.dev_attr.attr)) - return 0; - } + if ((amdgpu_dpm_set_fan_speed_rpm(adev, U32_MAX) == -EOPNOTSUPP) && + (amdgpu_dpm_get_fan_speed_rpm(adev, NULL) == -EOPNOTSUPP) && + (attr == &sensor_dev_attr_fan1_max.dev_attr.attr || + attr == &sensor_dev_attr_fan1_min.dev_attr.attr)) + return 0; if ((adev->family == AMDGPU_FAMILY_SI || /* not implemented yet */ adev->family == AMDGPU_FAMILY_KV) && /* not implemented yet */ @@ -3548,14 +3398,15 @@ static void amdgpu_debugfs_prints_cpu_info(struct seq_file *m, uint16_t *p_val; uint32_t size; int i; + uint32_t num_cpu_cores = amdgpu_dpm_get_num_cpu_cores(adev); - if (is_support_cclk_dpm(adev)) { - p_val = kcalloc(adev->smu.cpu_core_num, sizeof(uint16_t), + if (amdgpu_dpm_is_cclk_dpm_supported(adev)) { + p_val = kcalloc(num_cpu_cores, sizeof(uint16_t), GFP_KERNEL); if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_CPU_CLK, (void *)p_val, &size)) { - for (i = 0; i < adev->smu.cpu_core_num; i++) + for (i = 0; i < num_cpu_cores; i++) seq_printf(m, "\t%u MHz (CPU%d)\n", *(p_val + i), i); } @@ -3683,27 +3534,11 @@ static int amdgpu_debugfs_pm_info_show(struct seq_file *m, void *unused) return r; } - if (!adev->pm.dpm_enabled) { - seq_printf(m, "dpm not enabled\n"); - pm_runtime_mark_last_busy(dev->dev); - pm_runtime_put_autosuspend(dev->dev); - return 0; - } - - if (!is_support_sw_smu(adev) && - adev->powerplay.pp_funcs->debugfs_print_current_performance_level) { - mutex_lock(&adev->pm.mutex); - if (adev->powerplay.pp_funcs->debugfs_print_current_performance_level) - adev->powerplay.pp_funcs->debugfs_print_current_performance_level(adev, m); - else - seq_printf(m, "Debugfs support not implemented for this asic\n"); - mutex_unlock(&adev->pm.mutex); - r = 0; - } else { + if (amdgpu_dpm_debugfs_print_current_performance_level(adev, m)) { r = amdgpu_debugfs_pm_info_pp(m, adev); + if (r) + goto out; } - if (r) - goto out; amdgpu_device_ip_get_clockgating_state(adev, &flags); @@ -3729,21 +3564,18 @@ static ssize_t amdgpu_pm_prv_buffer_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { struct amdgpu_device *adev = file_inode(f)->i_private; - const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; - void *pp_handle = adev->powerplay.pp_handle; size_t smu_prv_buf_size; void *smu_prv_buf; + int ret = 0; if (amdgpu_in_reset(adev)) return -EPERM; if (adev->in_suspend && !adev->in_runpm) return -EPERM; - if (pp_funcs && pp_funcs->get_smu_prv_buf_details) - pp_funcs->get_smu_prv_buf_details(pp_handle, &smu_prv_buf, - &smu_prv_buf_size); - else - return -ENOSYS; + ret = amdgpu_dpm_get_smu_prv_buf_details(adev, &smu_prv_buf, &smu_prv_buf_size); + if (ret) + return ret; if (!smu_prv_buf || !smu_prv_buf_size) return -EINVAL; @@ -3767,6 +3599,9 @@ void amdgpu_debugfs_pm_init(struct amdgpu_device *adev) struct drm_minor *minor = adev_to_drm(adev)->primary; struct dentry *root = minor->debugfs_root; + if (!adev->pm.dpm_enabled) + return; + debugfs_create_file("amdgpu_pm_info", 0444, root, adev, &amdgpu_debugfs_pm_info_fops); @@ -3776,6 +3611,6 @@ void amdgpu_debugfs_pm_init(struct amdgpu_device *adev) &amdgpu_debugfs_pm_prv_buffer_fops, adev->pm.smu_prv_buffer_size); - amdgpu_smu_stb_debug_fs_init(adev); + amdgpu_dpm_stb_debug_fs_init(adev); #endif } diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h index c464a045000d..43d6b57173a3 100644 --- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h +++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h @@ -23,6 +23,12 @@ #ifndef __AMDGPU_DPM_H__ #define __AMDGPU_DPM_H__ +/* Argument for PPSMC_MSG_GpuChangeState */ +enum gfx_change_state { + sGpuChangeState_D0Entry = 1, + sGpuChangeState_D3Entry, +}; + enum amdgpu_int_thermal_type { THERMAL_TYPE_NONE, THERMAL_TYPE_EXTERNAL, @@ -39,19 +45,6 @@ enum amdgpu_int_thermal_type { THERMAL_TYPE_KV, }; -enum amdgpu_dpm_auto_throttle_src { - AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL, - AMDGPU_DPM_AUTO_THROTTLE_SRC_EXTERNAL -}; - -enum amdgpu_dpm_event_src { - AMDGPU_DPM_EVENT_SRC_ANALOG = 0, - AMDGPU_DPM_EVENT_SRC_EXTERNAL = 1, - AMDGPU_DPM_EVENT_SRC_DIGITAL = 2, - AMDGPU_DPM_EVENT_SRC_ANALOG_OR_EXTERNAL = 3, - AMDGPU_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL = 4 -}; - struct amdgpu_ps { u32 caps; /* vbios flags */ u32 class; /* vbios flags */ @@ -95,19 +88,6 @@ struct amdgpu_dpm_thermal { struct amdgpu_irq_src irq; }; -enum amdgpu_clk_action -{ - AMDGPU_SCLK_UP = 1, - AMDGPU_SCLK_DOWN -}; - -struct amdgpu_blacklist_clocks -{ - u32 sclk; - u32 mclk; - enum amdgpu_clk_action action; -}; - struct amdgpu_clock_and_voltage_limits { u32 sclk; u32 mclk; @@ -246,128 +226,6 @@ struct amdgpu_dpm_fan { bool ucode_fan_control; }; -enum amdgpu_pcie_gen { - AMDGPU_PCIE_GEN1 = 0, - AMDGPU_PCIE_GEN2 = 1, - AMDGPU_PCIE_GEN3 = 2, - AMDGPU_PCIE_GEN_INVALID = 0xffff -}; - -#define amdgpu_dpm_pre_set_power_state(adev) \ - ((adev)->powerplay.pp_funcs->pre_set_power_state((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_set_power_state(adev) \ - ((adev)->powerplay.pp_funcs->set_power_state((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_post_set_power_state(adev) \ - ((adev)->powerplay.pp_funcs->post_set_power_state((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_display_configuration_changed(adev) \ - ((adev)->powerplay.pp_funcs->display_configuration_changed((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_print_power_state(adev, ps) \ - ((adev)->powerplay.pp_funcs->print_power_state((adev)->powerplay.pp_handle, (ps))) - -#define amdgpu_dpm_vblank_too_short(adev) \ - ((adev)->powerplay.pp_funcs->vblank_too_short((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_enable_bapm(adev, e) \ - ((adev)->powerplay.pp_funcs->enable_bapm((adev)->powerplay.pp_handle, (e))) - -#define amdgpu_dpm_set_fan_control_mode(adev, m) \ - ((adev)->powerplay.pp_funcs->set_fan_control_mode((adev)->powerplay.pp_handle, (m))) - -#define amdgpu_dpm_get_fan_control_mode(adev) \ - ((adev)->powerplay.pp_funcs->get_fan_control_mode((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_set_fan_speed_pwm(adev, s) \ - ((adev)->powerplay.pp_funcs->set_fan_speed_pwm((adev)->powerplay.pp_handle, (s))) - -#define amdgpu_dpm_get_fan_speed_pwm(adev, s) \ - ((adev)->powerplay.pp_funcs->get_fan_speed_pwm((adev)->powerplay.pp_handle, (s))) - -#define amdgpu_dpm_get_fan_speed_rpm(adev, s) \ - ((adev)->powerplay.pp_funcs->get_fan_speed_rpm)((adev)->powerplay.pp_handle, (s)) - -#define amdgpu_dpm_set_fan_speed_rpm(adev, s) \ - ((adev)->powerplay.pp_funcs->set_fan_speed_rpm)((adev)->powerplay.pp_handle, (s)) - -#define amdgpu_dpm_force_performance_level(adev, l) \ - ((adev)->powerplay.pp_funcs->force_performance_level((adev)->powerplay.pp_handle, (l))) - -#define amdgpu_dpm_get_current_power_state(adev) \ - ((adev)->powerplay.pp_funcs->get_current_power_state((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_get_pp_num_states(adev, data) \ - ((adev)->powerplay.pp_funcs->get_pp_num_states((adev)->powerplay.pp_handle, data)) - -#define amdgpu_dpm_get_pp_table(adev, table) \ - ((adev)->powerplay.pp_funcs->get_pp_table((adev)->powerplay.pp_handle, table)) - -#define amdgpu_dpm_set_pp_table(adev, buf, size) \ - ((adev)->powerplay.pp_funcs->set_pp_table((adev)->powerplay.pp_handle, buf, size)) - -#define amdgpu_dpm_print_clock_levels(adev, type, buf) \ - ((adev)->powerplay.pp_funcs->print_clock_levels((adev)->powerplay.pp_handle, type, buf)) - -#define amdgpu_dpm_force_clock_level(adev, type, level) \ - ((adev)->powerplay.pp_funcs->force_clock_level((adev)->powerplay.pp_handle, type, level)) - -#define amdgpu_dpm_get_sclk_od(adev) \ - ((adev)->powerplay.pp_funcs->get_sclk_od((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_set_sclk_od(adev, value) \ - ((adev)->powerplay.pp_funcs->set_sclk_od((adev)->powerplay.pp_handle, value)) - -#define amdgpu_dpm_get_mclk_od(adev) \ - ((adev)->powerplay.pp_funcs->get_mclk_od((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_set_mclk_od(adev, value) \ - ((adev)->powerplay.pp_funcs->set_mclk_od((adev)->powerplay.pp_handle, value)) - -#define amdgpu_dpm_dispatch_task(adev, task_id, user_state) \ - ((adev)->powerplay.pp_funcs->dispatch_tasks)((adev)->powerplay.pp_handle, (task_id), (user_state)) - -#define amdgpu_dpm_check_state_equal(adev, cps, rps, equal) \ - ((adev)->powerplay.pp_funcs->check_state_equal((adev)->powerplay.pp_handle, (cps), (rps), (equal))) - -#define amdgpu_dpm_get_vce_clock_state(adev, i) \ - ((adev)->powerplay.pp_funcs->get_vce_clock_state((adev)->powerplay.pp_handle, (i))) - -#define amdgpu_dpm_get_performance_level(adev) \ - ((adev)->powerplay.pp_funcs->get_performance_level((adev)->powerplay.pp_handle)) - -#define amdgpu_dpm_reset_power_profile_state(adev, request) \ - ((adev)->powerplay.pp_funcs->reset_power_profile_state(\ - (adev)->powerplay.pp_handle, request)) - -#define amdgpu_dpm_get_power_profile_mode(adev, buf) \ - ((adev)->powerplay.pp_funcs->get_power_profile_mode(\ - (adev)->powerplay.pp_handle, buf)) - -#define amdgpu_dpm_set_power_profile_mode(adev, parameter, size) \ - ((adev)->powerplay.pp_funcs->set_power_profile_mode(\ - (adev)->powerplay.pp_handle, parameter, size)) - -#define amdgpu_dpm_set_fine_grain_clk_vol(adev, type, parameter, size) \ - ((adev)->powerplay.pp_funcs->set_fine_grain_clk_vol(\ - (adev)->powerplay.pp_handle, type, parameter, size)) - -#define amdgpu_dpm_odn_edit_dpm_table(adev, type, parameter, size) \ - ((adev)->powerplay.pp_funcs->odn_edit_dpm_table(\ - (adev)->powerplay.pp_handle, type, parameter, size)) - -#define amdgpu_dpm_get_ppfeature_status(adev, buf) \ - ((adev)->powerplay.pp_funcs->get_ppfeature_status(\ - (adev)->powerplay.pp_handle, (buf))) - -#define amdgpu_dpm_set_ppfeature_status(adev, ppfeatures) \ - ((adev)->powerplay.pp_funcs->set_ppfeature_status(\ - (adev)->powerplay.pp_handle, (ppfeatures))) - -#define amdgpu_dpm_get_gpu_metrics(adev, table) \ - ((adev)->powerplay.pp_funcs->get_gpu_metrics((adev)->powerplay.pp_handle, table)) - struct amdgpu_dpm { struct amdgpu_ps *ps; /* number of valid power states */ @@ -426,6 +284,15 @@ enum ip_power_state { /* Used to mask smu debug modes */ #define SMU_DEBUG_HALT_ON_ERROR 0x1 +#define MAX_SMU_I2C_BUSES 2 + +struct amdgpu_smu_i2c_bus { + struct i2c_adapter adapter; + struct amdgpu_device *adev; + int port; + struct mutex mutex; +}; + struct amdgpu_pm { struct mutex mutex; u32 current_sclk; @@ -458,8 +325,9 @@ struct amdgpu_pm { uint32_t pp_feature; /* Used for I2C access to various EEPROMs on relevant ASICs */ - struct i2c_adapter smu_i2c; - struct mutex smu_i2c_mutex; + struct amdgpu_smu_i2c_bus smu_i2c[MAX_SMU_I2C_BUSES]; + struct i2c_adapter *ras_eeprom_i2c_bus; + struct i2c_adapter *fru_eeprom_i2c_bus; struct list_head pm_attr_list; atomic_t pwr_state[AMD_IP_BLOCK_TYPE_NUM]; @@ -468,64 +336,16 @@ struct amdgpu_pm { * 0 = disabled (default), otherwise enable corresponding debug mode */ uint32_t smu_debug_mask; -}; -#define R600_SSTU_DFLT 0 -#define R600_SST_DFLT 0x00C8 + bool pp_force_state_enabled; -/* XXX are these ok? */ -#define R600_TEMP_RANGE_MIN (90 * 1000) -#define R600_TEMP_RANGE_MAX (120 * 1000) - -#define FDO_PWM_MODE_STATIC 1 -#define FDO_PWM_MODE_STATIC_RPM 5 - -enum amdgpu_td { - AMDGPU_TD_AUTO, - AMDGPU_TD_UP, - AMDGPU_TD_DOWN, + struct mutex stable_pstate_ctx_lock; + struct amdgpu_ctx *stable_pstate_ctx; }; -enum amdgpu_display_watermark { - AMDGPU_DISPLAY_WATERMARK_LOW = 0, - AMDGPU_DISPLAY_WATERMARK_HIGH = 1, -}; - -enum amdgpu_display_gap -{ - AMDGPU_PM_DISPLAY_GAP_VBLANK_OR_WM = 0, - AMDGPU_PM_DISPLAY_GAP_VBLANK = 1, - AMDGPU_PM_DISPLAY_GAP_WATERMARK = 2, - AMDGPU_PM_DISPLAY_GAP_IGNORE = 3, -}; - -void amdgpu_dpm_print_class_info(u32 class, u32 class2); -void amdgpu_dpm_print_cap_info(u32 caps); -void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev, - struct amdgpu_ps *rps); -u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev); -u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev); -void amdgpu_dpm_get_active_displays(struct amdgpu_device *adev); int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors sensor, void *data, uint32_t *size); -bool amdgpu_is_internal_thermal_sensor(enum amdgpu_int_thermal_type sensor); - -int amdgpu_get_platform_caps(struct amdgpu_device *adev); - -int amdgpu_parse_extended_power_table(struct amdgpu_device *adev); -void amdgpu_free_extended_power_table(struct amdgpu_device *adev); - -void amdgpu_add_thermal_controller(struct amdgpu_device *adev); - -enum amdgpu_pcie_gen amdgpu_get_pcie_gen_support(struct amdgpu_device *adev, - u32 sys_mask, - enum amdgpu_pcie_gen asic_gen, - enum amdgpu_pcie_gen default_gen); - -struct amd_vce_state* -amdgpu_get_vce_clock_state(void *handle, u32 idx); - int amdgpu_dpm_set_powergating_by_smu(struct amdgpu_device *adev, uint32_t block_type, bool gate); @@ -571,16 +391,139 @@ int amdgpu_dpm_smu_i2c_bus_access(struct amdgpu_device *adev, void amdgpu_pm_acpi_event_handler(struct amdgpu_device *adev); -int amdgpu_dpm_read_sensor(struct amdgpu_device *adev, enum amd_pp_sensors sensor, - void *data, uint32_t *size); - -void amdgpu_dpm_thermal_work_handler(struct work_struct *work); - -void amdgpu_pm_compute_clocks(struct amdgpu_device *adev); +void amdgpu_dpm_compute_clocks(struct amdgpu_device *adev); void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable); void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable); void amdgpu_dpm_enable_jpeg(struct amdgpu_device *adev, bool enable); -void amdgpu_pm_print_power_states(struct amdgpu_device *adev); int amdgpu_pm_load_smu_firmware(struct amdgpu_device *adev, uint32_t *smu_version); - +int amdgpu_dpm_handle_passthrough_sbr(struct amdgpu_device *adev, bool enable); +int amdgpu_dpm_send_hbm_bad_pages_num(struct amdgpu_device *adev, uint32_t size); +int amdgpu_dpm_get_dpm_freq_range(struct amdgpu_device *adev, + enum pp_clock_type type, + uint32_t *min, + uint32_t *max); +int amdgpu_dpm_set_soft_freq_range(struct amdgpu_device *adev, + enum pp_clock_type type, + uint32_t min, + uint32_t max); +int amdgpu_dpm_write_watermarks_table(struct amdgpu_device *adev); +int amdgpu_dpm_wait_for_event(struct amdgpu_device *adev, enum smu_event_type event, + uint64_t event_arg); +int amdgpu_dpm_get_status_gfxoff(struct amdgpu_device *adev, uint32_t *value); +uint64_t amdgpu_dpm_get_thermal_throttling_counter(struct amdgpu_device *adev); +void amdgpu_dpm_gfx_state_change(struct amdgpu_device *adev, + enum gfx_change_state state); +int amdgpu_dpm_get_ecc_info(struct amdgpu_device *adev, + void *umc_ecc); +struct amd_vce_state *amdgpu_dpm_get_vce_clock_state(struct amdgpu_device *adev, + uint32_t idx); +void amdgpu_dpm_get_current_power_state(struct amdgpu_device *adev, enum amd_pm_state_type *state); +void amdgpu_dpm_set_power_state(struct amdgpu_device *adev, + enum amd_pm_state_type state); +enum amd_dpm_forced_level amdgpu_dpm_get_performance_level(struct amdgpu_device *adev); +int amdgpu_dpm_force_performance_level(struct amdgpu_device *adev, + enum amd_dpm_forced_level level); +int amdgpu_dpm_get_pp_num_states(struct amdgpu_device *adev, + struct pp_states_info *states); +int amdgpu_dpm_dispatch_task(struct amdgpu_device *adev, + enum amd_pp_task task_id, + enum amd_pm_state_type *user_state); +int amdgpu_dpm_get_pp_table(struct amdgpu_device *adev, char **table); +int amdgpu_dpm_set_fine_grain_clk_vol(struct amdgpu_device *adev, + uint32_t type, + long *input, + uint32_t size); +int amdgpu_dpm_odn_edit_dpm_table(struct amdgpu_device *adev, + uint32_t type, + long *input, + uint32_t size); +int amdgpu_dpm_print_clock_levels(struct amdgpu_device *adev, + enum pp_clock_type type, + char *buf); +int amdgpu_dpm_emit_clock_levels(struct amdgpu_device *adev, + enum pp_clock_type type, + char *buf, + int *offset); +int amdgpu_dpm_set_ppfeature_status(struct amdgpu_device *adev, + uint64_t ppfeature_masks); +int amdgpu_dpm_get_ppfeature_status(struct amdgpu_device *adev, char *buf); +int amdgpu_dpm_force_clock_level(struct amdgpu_device *adev, + enum pp_clock_type type, + uint32_t mask); +int amdgpu_dpm_get_sclk_od(struct amdgpu_device *adev); +int amdgpu_dpm_set_sclk_od(struct amdgpu_device *adev, uint32_t value); +int amdgpu_dpm_get_mclk_od(struct amdgpu_device *adev); +int amdgpu_dpm_set_mclk_od(struct amdgpu_device *adev, uint32_t value); +int amdgpu_dpm_get_power_profile_mode(struct amdgpu_device *adev, + char *buf); +int amdgpu_dpm_set_power_profile_mode(struct amdgpu_device *adev, + long *input, uint32_t size); +int amdgpu_dpm_get_gpu_metrics(struct amdgpu_device *adev, void **table); +int amdgpu_dpm_get_fan_control_mode(struct amdgpu_device *adev, + uint32_t *fan_mode); +int amdgpu_dpm_set_fan_speed_pwm(struct amdgpu_device *adev, + uint32_t speed); +int amdgpu_dpm_get_fan_speed_pwm(struct amdgpu_device *adev, + uint32_t *speed); +int amdgpu_dpm_get_fan_speed_rpm(struct amdgpu_device *adev, + uint32_t *speed); +int amdgpu_dpm_set_fan_speed_rpm(struct amdgpu_device *adev, + uint32_t speed); +int amdgpu_dpm_set_fan_control_mode(struct amdgpu_device *adev, + uint32_t mode); +int amdgpu_dpm_get_power_limit(struct amdgpu_device *adev, + uint32_t *limit, + enum pp_power_limit_level pp_limit_level, + enum pp_power_type power_type); +int amdgpu_dpm_set_power_limit(struct amdgpu_device *adev, + uint32_t limit); +int amdgpu_dpm_is_cclk_dpm_supported(struct amdgpu_device *adev); +int amdgpu_dpm_debugfs_print_current_performance_level(struct amdgpu_device *adev, + struct seq_file *m); +int amdgpu_dpm_get_smu_prv_buf_details(struct amdgpu_device *adev, + void **addr, + size_t *size); +int amdgpu_dpm_is_overdrive_supported(struct amdgpu_device *adev); +int amdgpu_dpm_set_pp_table(struct amdgpu_device *adev, + const char *buf, + size_t size); +int amdgpu_dpm_get_num_cpu_cores(struct amdgpu_device *adev); +void amdgpu_dpm_stb_debug_fs_init(struct amdgpu_device *adev); +int amdgpu_dpm_display_configuration_change(struct amdgpu_device *adev, + const struct amd_pp_display_configuration *input); +int amdgpu_dpm_get_clock_by_type(struct amdgpu_device *adev, + enum amd_pp_clock_type type, + struct amd_pp_clocks *clocks); +int amdgpu_dpm_get_display_mode_validation_clks(struct amdgpu_device *adev, + struct amd_pp_simple_clock_info *clocks); +int amdgpu_dpm_get_clock_by_type_with_latency(struct amdgpu_device *adev, + enum amd_pp_clock_type type, + struct pp_clock_levels_with_latency *clocks); +int amdgpu_dpm_get_clock_by_type_with_voltage(struct amdgpu_device *adev, + enum amd_pp_clock_type type, + struct pp_clock_levels_with_voltage *clocks); +int amdgpu_dpm_set_watermarks_for_clocks_ranges(struct amdgpu_device *adev, + void *clock_ranges); +int amdgpu_dpm_display_clock_voltage_request(struct amdgpu_device *adev, + struct pp_display_clock_request *clock); +int amdgpu_dpm_get_current_clocks(struct amdgpu_device *adev, + struct amd_pp_clock_info *clocks); +void amdgpu_dpm_notify_smu_enable_pwe(struct amdgpu_device *adev); +int amdgpu_dpm_set_active_display_count(struct amdgpu_device *adev, + uint32_t count); +int amdgpu_dpm_set_min_deep_sleep_dcefclk(struct amdgpu_device *adev, + uint32_t clock); +void amdgpu_dpm_set_hard_min_dcefclk_by_freq(struct amdgpu_device *adev, + uint32_t clock); +void amdgpu_dpm_set_hard_min_fclk_by_freq(struct amdgpu_device *adev, + uint32_t clock); +int amdgpu_dpm_display_disable_memory_clock_switch(struct amdgpu_device *adev, + bool disable_memory_clock_switch); +int amdgpu_dpm_get_max_sustainable_clocks_by_dc(struct amdgpu_device *adev, + struct pp_smu_nv_clock_table *max_clocks); +enum pp_smu_status amdgpu_dpm_get_uclk_dpm_states(struct amdgpu_device *adev, + unsigned int *clock_values_in_khz, + unsigned int *num_states); +int amdgpu_dpm_get_dpm_clock_table(struct amdgpu_device *adev, + struct dpm_clocks *clock_table); #endif diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h new file mode 100644 index 000000000000..5c2a89f0d5d5 --- /dev/null +++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h @@ -0,0 +1,32 @@ +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef __AMDGPU_DPM_INTERNAL_H__ +#define __AMDGPU_DPM_INTERNAL_H__ + +void amdgpu_dpm_get_active_displays(struct amdgpu_device *adev); + +u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev); + +u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev); + +#endif diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/Makefile b/drivers/gpu/drm/amd/pm/legacy-dpm/Makefile new file mode 100644 index 000000000000..baa4265d1daa --- /dev/null +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/Makefile @@ -0,0 +1,32 @@ +# +# Copyright 2021 Advanced Micro Devices, Inc. +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the "Software"), +# to deal in the Software without restriction, including without limitation +# the rights to use, copy, modify, merge, publish, distribute, sublicense, +# and/or sell copies of the Software, and to permit persons to whom the +# Software is furnished to do so, subject to the following conditions: +# +# The above copyright notice and this permission notice shall be included in +# all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR +# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +# OTHER DEALINGS IN THE SOFTWARE. +# + +AMD_LEGACYDPM_PATH = ../pm/legacy-dpm + +LEGACYDPM_MGR-y = legacy_dpm.o + +LEGACYDPM_MGR-$(CONFIG_DRM_AMDGPU_CIK)+= kv_dpm.o kv_smc.o +LEGACYDPM_MGR-$(CONFIG_DRM_AMDGPU_SI)+= si_dpm.o si_smc.o + +AMD_LEGACYDPM_POWER = $(addprefix $(AMD_LEGACYDPM_PATH)/,$(LEGACYDPM_MGR-y)) + +AMD_POWERPLAY_FILES += $(AMD_LEGACYDPM_POWER) diff --git a/drivers/gpu/drm/amd/pm/powerplay/cik_dpm.h b/drivers/gpu/drm/amd/pm/legacy-dpm/cik_dpm.h index 2fcc4b60153c..2fcc4b60153c 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/cik_dpm.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/cik_dpm.h diff --git a/drivers/gpu/drm/amd/pm/powerplay/kv_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c index bcae42cef374..8b23cc9f098a 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/kv_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c @@ -36,6 +36,7 @@ #include "gca/gfx_7_2_d.h" #include "gca/gfx_7_2_sh_mask.h" +#include "legacy_dpm.h" #define KV_MAX_DEEPSLEEP_DIVIDER_ID 5 #define KV_MINIMUM_ENGINE_CLOCK 800 @@ -1256,6 +1257,19 @@ static void kv_dpm_enable_bapm(void *handle, bool enable) } } +static bool kv_is_internal_thermal_sensor(enum amdgpu_int_thermal_type sensor) +{ + switch (sensor) { + case THERMAL_TYPE_KV: + return true; + case THERMAL_TYPE_NONE: + case THERMAL_TYPE_EXTERNAL: + case THERMAL_TYPE_EXTERNAL_GPIO: + default: + return false; + } +} + static int kv_dpm_enable(struct amdgpu_device *adev) { struct kv_power_info *pi = kv_get_pi(adev); @@ -1352,7 +1366,7 @@ static int kv_dpm_enable(struct amdgpu_device *adev) } if (adev->irq.installed && - amdgpu_is_internal_thermal_sensor(adev->pm.int_thermal_type)) { + kv_is_internal_thermal_sensor(adev->pm.int_thermal_type)) { ret = kv_set_thermal_temperature_range(adev, KV_TEMP_RANGE_MIN, KV_TEMP_RANGE_MAX); if (ret) { DRM_ERROR("kv_set_thermal_temperature_range failed\n"); @@ -3016,21 +3030,18 @@ static int kv_dpm_sw_init(void *handle) return 0; INIT_WORK(&adev->pm.dpm.thermal.work, amdgpu_dpm_thermal_work_handler); - mutex_lock(&adev->pm.mutex); ret = kv_dpm_init(adev); if (ret) goto dpm_failed; adev->pm.dpm.current_ps = adev->pm.dpm.requested_ps = adev->pm.dpm.boot_ps; if (amdgpu_dpm == 1) amdgpu_pm_print_power_states(adev); - mutex_unlock(&adev->pm.mutex); DRM_INFO("amdgpu: dpm initialized\n"); return 0; dpm_failed: kv_dpm_fini(adev); - mutex_unlock(&adev->pm.mutex); DRM_ERROR("amdgpu: dpm initialization failed\n"); return ret; } @@ -3041,9 +3052,7 @@ static int kv_dpm_sw_fini(void *handle) flush_work(&adev->pm.dpm.thermal.work); - mutex_lock(&adev->pm.mutex); kv_dpm_fini(adev); - mutex_unlock(&adev->pm.mutex); return 0; } @@ -3056,15 +3065,13 @@ static int kv_dpm_hw_init(void *handle) if (!amdgpu_dpm) return 0; - mutex_lock(&adev->pm.mutex); kv_dpm_setup_asic(adev); ret = kv_dpm_enable(adev); if (ret) adev->pm.dpm_enabled = false; else adev->pm.dpm_enabled = true; - mutex_unlock(&adev->pm.mutex); - amdgpu_pm_compute_clocks(adev); + amdgpu_legacy_dpm_compute_clocks(adev); return ret; } @@ -3072,11 +3079,8 @@ static int kv_dpm_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->pm.dpm_enabled) { - mutex_lock(&adev->pm.mutex); + if (adev->pm.dpm_enabled) kv_dpm_disable(adev); - mutex_unlock(&adev->pm.mutex); - } return 0; } @@ -3086,12 +3090,10 @@ static int kv_dpm_suspend(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; if (adev->pm.dpm_enabled) { - mutex_lock(&adev->pm.mutex); /* disable dpm */ kv_dpm_disable(adev); /* reset the power state */ adev->pm.dpm.current_ps = adev->pm.dpm.requested_ps = adev->pm.dpm.boot_ps; - mutex_unlock(&adev->pm.mutex); } return 0; } @@ -3103,16 +3105,14 @@ static int kv_dpm_resume(void *handle) if (adev->pm.dpm_enabled) { /* asic init will reset to the boot state */ - mutex_lock(&adev->pm.mutex); kv_dpm_setup_asic(adev); ret = kv_dpm_enable(adev); if (ret) adev->pm.dpm_enabled = false; else adev->pm.dpm_enabled = true; - mutex_unlock(&adev->pm.mutex); if (adev->pm.dpm_enabled) - amdgpu_pm_compute_clocks(adev); + amdgpu_legacy_dpm_compute_clocks(adev); } return 0; } @@ -3366,6 +3366,7 @@ static const struct amd_pm_funcs kv_dpm_funcs = { .get_vce_clock_state = amdgpu_get_vce_clock_state, .check_state_equal = kv_check_state_equal, .read_sensor = &kv_dpm_read_sensor, + .pm_compute_clocks = amdgpu_legacy_dpm_compute_clocks, }; static const struct amdgpu_irq_src_funcs kv_dpm_irq_funcs = { diff --git a/drivers/gpu/drm/amd/pm/powerplay/kv_dpm.h b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.h index 6df0ed41317c..6df0ed41317c 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/kv_dpm.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.h diff --git a/drivers/gpu/drm/amd/pm/powerplay/kv_smc.c b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_smc.c index 2d9ab6b8be66..2d9ab6b8be66 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/kv_smc.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_smc.c diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c new file mode 100644 index 000000000000..9613c6181c17 --- /dev/null +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c @@ -0,0 +1,1081 @@ +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ + +#include "amdgpu.h" +#include "amdgpu_i2c.h" +#include "amdgpu_atombios.h" +#include "atom.h" +#include "amd_pcie.h" +#include "legacy_dpm.h" +#include "amdgpu_dpm_internal.h" +#include "amdgpu_display.h" + +#define amdgpu_dpm_pre_set_power_state(adev) \ + ((adev)->powerplay.pp_funcs->pre_set_power_state((adev)->powerplay.pp_handle)) + +#define amdgpu_dpm_post_set_power_state(adev) \ + ((adev)->powerplay.pp_funcs->post_set_power_state((adev)->powerplay.pp_handle)) + +#define amdgpu_dpm_display_configuration_changed(adev) \ + ((adev)->powerplay.pp_funcs->display_configuration_changed((adev)->powerplay.pp_handle)) + +#define amdgpu_dpm_print_power_state(adev, ps) \ + ((adev)->powerplay.pp_funcs->print_power_state((adev)->powerplay.pp_handle, (ps))) + +#define amdgpu_dpm_vblank_too_short(adev) \ + ((adev)->powerplay.pp_funcs->vblank_too_short((adev)->powerplay.pp_handle)) + +#define amdgpu_dpm_check_state_equal(adev, cps, rps, equal) \ + ((adev)->powerplay.pp_funcs->check_state_equal((adev)->powerplay.pp_handle, (cps), (rps), (equal))) + +void amdgpu_dpm_print_class_info(u32 class, u32 class2) +{ + const char *s; + + switch (class & ATOM_PPLIB_CLASSIFICATION_UI_MASK) { + case ATOM_PPLIB_CLASSIFICATION_UI_NONE: + default: + s = "none"; + break; + case ATOM_PPLIB_CLASSIFICATION_UI_BATTERY: + s = "battery"; + break; + case ATOM_PPLIB_CLASSIFICATION_UI_BALANCED: + s = "balanced"; + break; + case ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE: + s = "performance"; + break; + } + printk("\tui class: %s\n", s); + printk("\tinternal class:"); + if (((class & ~ATOM_PPLIB_CLASSIFICATION_UI_MASK) == 0) && + (class2 == 0)) + pr_cont(" none"); + else { + if (class & ATOM_PPLIB_CLASSIFICATION_BOOT) + pr_cont(" boot"); + if (class & ATOM_PPLIB_CLASSIFICATION_THERMAL) + pr_cont(" thermal"); + if (class & ATOM_PPLIB_CLASSIFICATION_LIMITEDPOWERSOURCE) + pr_cont(" limited_pwr"); + if (class & ATOM_PPLIB_CLASSIFICATION_REST) + pr_cont(" rest"); + if (class & ATOM_PPLIB_CLASSIFICATION_FORCED) + pr_cont(" forced"); + if (class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE) + pr_cont(" 3d_perf"); + if (class & ATOM_PPLIB_CLASSIFICATION_OVERDRIVETEMPLATE) + pr_cont(" ovrdrv"); + if (class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE) + pr_cont(" uvd"); + if (class & ATOM_PPLIB_CLASSIFICATION_3DLOW) + pr_cont(" 3d_low"); + if (class & ATOM_PPLIB_CLASSIFICATION_ACPI) + pr_cont(" acpi"); + if (class & ATOM_PPLIB_CLASSIFICATION_HD2STATE) + pr_cont(" uvd_hd2"); + if (class & ATOM_PPLIB_CLASSIFICATION_HDSTATE) + pr_cont(" uvd_hd"); + if (class & ATOM_PPLIB_CLASSIFICATION_SDSTATE) + pr_cont(" uvd_sd"); + if (class2 & ATOM_PPLIB_CLASSIFICATION2_LIMITEDPOWERSOURCE_2) + pr_cont(" limited_pwr2"); + if (class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) + pr_cont(" ulv"); + if (class2 & ATOM_PPLIB_CLASSIFICATION2_MVC) + pr_cont(" uvd_mvc"); + } + pr_cont("\n"); +} + +void amdgpu_dpm_print_cap_info(u32 caps) +{ + printk("\tcaps:"); + if (caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) + pr_cont(" single_disp"); + if (caps & ATOM_PPLIB_SUPPORTS_VIDEO_PLAYBACK) + pr_cont(" video"); + if (caps & ATOM_PPLIB_DISALLOW_ON_DC) + pr_cont(" no_dc"); + pr_cont("\n"); +} + +void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev, + struct amdgpu_ps *rps) +{ + printk("\tstatus:"); + if (rps == adev->pm.dpm.current_ps) + pr_cont(" c"); + if (rps == adev->pm.dpm.requested_ps) + pr_cont(" r"); + if (rps == adev->pm.dpm.boot_ps) + pr_cont(" b"); + pr_cont("\n"); +} + +void amdgpu_pm_print_power_states(struct amdgpu_device *adev) +{ + int i; + + if (adev->powerplay.pp_funcs->print_power_state == NULL) + return; + + for (i = 0; i < adev->pm.dpm.num_ps; i++) + amdgpu_dpm_print_power_state(adev, &adev->pm.dpm.ps[i]); + +} + +union power_info { + struct _ATOM_POWERPLAY_INFO info; + struct _ATOM_POWERPLAY_INFO_V2 info_2; + struct _ATOM_POWERPLAY_INFO_V3 info_3; + struct _ATOM_PPLIB_POWERPLAYTABLE pplib; + struct _ATOM_PPLIB_POWERPLAYTABLE2 pplib2; + struct _ATOM_PPLIB_POWERPLAYTABLE3 pplib3; + struct _ATOM_PPLIB_POWERPLAYTABLE4 pplib4; + struct _ATOM_PPLIB_POWERPLAYTABLE5 pplib5; +}; + +int amdgpu_get_platform_caps(struct amdgpu_device *adev) +{ + struct amdgpu_mode_info *mode_info = &adev->mode_info; + union power_info *power_info; + int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo); + u16 data_offset; + u8 frev, crev; + + if (!amdgpu_atom_parse_data_header(mode_info->atom_context, index, NULL, + &frev, &crev, &data_offset)) + return -EINVAL; + power_info = (union power_info *)(mode_info->atom_context->bios + data_offset); + + adev->pm.dpm.platform_caps = le32_to_cpu(power_info->pplib.ulPlatformCaps); + adev->pm.dpm.backbias_response_time = le16_to_cpu(power_info->pplib.usBackbiasTime); + adev->pm.dpm.voltage_response_time = le16_to_cpu(power_info->pplib.usVoltageTime); + + return 0; +} + +union fan_info { + struct _ATOM_PPLIB_FANTABLE fan; + struct _ATOM_PPLIB_FANTABLE2 fan2; + struct _ATOM_PPLIB_FANTABLE3 fan3; +}; + +static int amdgpu_parse_clk_voltage_dep_table(struct amdgpu_clock_voltage_dependency_table *amdgpu_table, + ATOM_PPLIB_Clock_Voltage_Dependency_Table *atom_table) +{ + u32 size = atom_table->ucNumEntries * + sizeof(struct amdgpu_clock_voltage_dependency_entry); + int i; + ATOM_PPLIB_Clock_Voltage_Dependency_Record *entry; + + amdgpu_table->entries = kzalloc(size, GFP_KERNEL); + if (!amdgpu_table->entries) + return -ENOMEM; + + entry = &atom_table->entries[0]; + for (i = 0; i < atom_table->ucNumEntries; i++) { + amdgpu_table->entries[i].clk = le16_to_cpu(entry->usClockLow) | + (entry->ucClockHigh << 16); + amdgpu_table->entries[i].v = le16_to_cpu(entry->usVoltage); + entry = (ATOM_PPLIB_Clock_Voltage_Dependency_Record *) + ((u8 *)entry + sizeof(ATOM_PPLIB_Clock_Voltage_Dependency_Record)); + } + amdgpu_table->count = atom_table->ucNumEntries; + + return 0; +} + +/* sizeof(ATOM_PPLIB_EXTENDEDHEADER) */ +#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V2 12 +#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V3 14 +#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V4 16 +#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V5 18 +#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V6 20 +#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V7 22 +#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V8 24 +#define SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V9 26 + +int amdgpu_parse_extended_power_table(struct amdgpu_device *adev) +{ + struct amdgpu_mode_info *mode_info = &adev->mode_info; + union power_info *power_info; + union fan_info *fan_info; + ATOM_PPLIB_Clock_Voltage_Dependency_Table *dep_table; + int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo); + u16 data_offset; + u8 frev, crev; + int ret, i; + + if (!amdgpu_atom_parse_data_header(mode_info->atom_context, index, NULL, + &frev, &crev, &data_offset)) + return -EINVAL; + power_info = (union power_info *)(mode_info->atom_context->bios + data_offset); + + /* fan table */ + if (le16_to_cpu(power_info->pplib.usTableSize) >= + sizeof(struct _ATOM_PPLIB_POWERPLAYTABLE3)) { + if (power_info->pplib3.usFanTableOffset) { + fan_info = (union fan_info *)(mode_info->atom_context->bios + data_offset + + le16_to_cpu(power_info->pplib3.usFanTableOffset)); + adev->pm.dpm.fan.t_hyst = fan_info->fan.ucTHyst; + adev->pm.dpm.fan.t_min = le16_to_cpu(fan_info->fan.usTMin); + adev->pm.dpm.fan.t_med = le16_to_cpu(fan_info->fan.usTMed); + adev->pm.dpm.fan.t_high = le16_to_cpu(fan_info->fan.usTHigh); + adev->pm.dpm.fan.pwm_min = le16_to_cpu(fan_info->fan.usPWMMin); + adev->pm.dpm.fan.pwm_med = le16_to_cpu(fan_info->fan.usPWMMed); + adev->pm.dpm.fan.pwm_high = le16_to_cpu(fan_info->fan.usPWMHigh); + if (fan_info->fan.ucFanTableFormat >= 2) + adev->pm.dpm.fan.t_max = le16_to_cpu(fan_info->fan2.usTMax); + else + adev->pm.dpm.fan.t_max = 10900; + adev->pm.dpm.fan.cycle_delay = 100000; + if (fan_info->fan.ucFanTableFormat >= 3) { + adev->pm.dpm.fan.control_mode = fan_info->fan3.ucFanControlMode; + adev->pm.dpm.fan.default_max_fan_pwm = + le16_to_cpu(fan_info->fan3.usFanPWMMax); + adev->pm.dpm.fan.default_fan_output_sensitivity = 4836; + adev->pm.dpm.fan.fan_output_sensitivity = + le16_to_cpu(fan_info->fan3.usFanOutputSensitivity); + } + adev->pm.dpm.fan.ucode_fan_control = true; + } + } + + /* clock dependancy tables, shedding tables */ + if (le16_to_cpu(power_info->pplib.usTableSize) >= + sizeof(struct _ATOM_PPLIB_POWERPLAYTABLE4)) { + if (power_info->pplib4.usVddcDependencyOnSCLKOffset) { + dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(power_info->pplib4.usVddcDependencyOnSCLKOffset)); + ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.vddc_dependency_on_sclk, + dep_table); + if (ret) { + amdgpu_free_extended_power_table(adev); + return ret; + } + } + if (power_info->pplib4.usVddciDependencyOnMCLKOffset) { + dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(power_info->pplib4.usVddciDependencyOnMCLKOffset)); + ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.vddci_dependency_on_mclk, + dep_table); + if (ret) { + amdgpu_free_extended_power_table(adev); + return ret; + } + } + if (power_info->pplib4.usVddcDependencyOnMCLKOffset) { + dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(power_info->pplib4.usVddcDependencyOnMCLKOffset)); + ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.vddc_dependency_on_mclk, + dep_table); + if (ret) { + amdgpu_free_extended_power_table(adev); + return ret; + } + } + if (power_info->pplib4.usMvddDependencyOnMCLKOffset) { + dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(power_info->pplib4.usMvddDependencyOnMCLKOffset)); + ret = amdgpu_parse_clk_voltage_dep_table(&adev->pm.dpm.dyn_state.mvdd_dependency_on_mclk, + dep_table); + if (ret) { + amdgpu_free_extended_power_table(adev); + return ret; + } + } + if (power_info->pplib4.usMaxClockVoltageOnDCOffset) { + ATOM_PPLIB_Clock_Voltage_Limit_Table *clk_v = + (ATOM_PPLIB_Clock_Voltage_Limit_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(power_info->pplib4.usMaxClockVoltageOnDCOffset)); + if (clk_v->ucNumEntries) { + adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.sclk = + le16_to_cpu(clk_v->entries[0].usSclkLow) | + (clk_v->entries[0].ucSclkHigh << 16); + adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.mclk = + le16_to_cpu(clk_v->entries[0].usMclkLow) | + (clk_v->entries[0].ucMclkHigh << 16); + adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.vddc = + le16_to_cpu(clk_v->entries[0].usVddc); + adev->pm.dpm.dyn_state.max_clock_voltage_on_dc.vddci = + le16_to_cpu(clk_v->entries[0].usVddci); + } + } + if (power_info->pplib4.usVddcPhaseShedLimitsTableOffset) { + ATOM_PPLIB_PhaseSheddingLimits_Table *psl = + (ATOM_PPLIB_PhaseSheddingLimits_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(power_info->pplib4.usVddcPhaseShedLimitsTableOffset)); + ATOM_PPLIB_PhaseSheddingLimits_Record *entry; + + adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries = + kcalloc(psl->ucNumEntries, + sizeof(struct amdgpu_phase_shedding_limits_entry), + GFP_KERNEL); + if (!adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries) { + amdgpu_free_extended_power_table(adev); + return -ENOMEM; + } + + entry = &psl->entries[0]; + for (i = 0; i < psl->ucNumEntries; i++) { + adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries[i].sclk = + le16_to_cpu(entry->usSclkLow) | (entry->ucSclkHigh << 16); + adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries[i].mclk = + le16_to_cpu(entry->usMclkLow) | (entry->ucMclkHigh << 16); + adev->pm.dpm.dyn_state.phase_shedding_limits_table.entries[i].voltage = + le16_to_cpu(entry->usVoltage); + entry = (ATOM_PPLIB_PhaseSheddingLimits_Record *) + ((u8 *)entry + sizeof(ATOM_PPLIB_PhaseSheddingLimits_Record)); + } + adev->pm.dpm.dyn_state.phase_shedding_limits_table.count = + psl->ucNumEntries; + } + } + + /* cac data */ + if (le16_to_cpu(power_info->pplib.usTableSize) >= + sizeof(struct _ATOM_PPLIB_POWERPLAYTABLE5)) { + adev->pm.dpm.tdp_limit = le32_to_cpu(power_info->pplib5.ulTDPLimit); + adev->pm.dpm.near_tdp_limit = le32_to_cpu(power_info->pplib5.ulNearTDPLimit); + adev->pm.dpm.near_tdp_limit_adjusted = adev->pm.dpm.near_tdp_limit; + adev->pm.dpm.tdp_od_limit = le16_to_cpu(power_info->pplib5.usTDPODLimit); + if (adev->pm.dpm.tdp_od_limit) + adev->pm.dpm.power_control = true; + else + adev->pm.dpm.power_control = false; + adev->pm.dpm.tdp_adjustment = 0; + adev->pm.dpm.sq_ramping_threshold = le32_to_cpu(power_info->pplib5.ulSQRampingThreshold); + adev->pm.dpm.cac_leakage = le32_to_cpu(power_info->pplib5.ulCACLeakage); + adev->pm.dpm.load_line_slope = le16_to_cpu(power_info->pplib5.usLoadLineSlope); + if (power_info->pplib5.usCACLeakageTableOffset) { + ATOM_PPLIB_CAC_Leakage_Table *cac_table = + (ATOM_PPLIB_CAC_Leakage_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(power_info->pplib5.usCACLeakageTableOffset)); + ATOM_PPLIB_CAC_Leakage_Record *entry; + u32 size = cac_table->ucNumEntries * sizeof(struct amdgpu_cac_leakage_table); + adev->pm.dpm.dyn_state.cac_leakage_table.entries = kzalloc(size, GFP_KERNEL); + if (!adev->pm.dpm.dyn_state.cac_leakage_table.entries) { + amdgpu_free_extended_power_table(adev); + return -ENOMEM; + } + entry = &cac_table->entries[0]; + for (i = 0; i < cac_table->ucNumEntries; i++) { + if (adev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_EVV) { + adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc1 = + le16_to_cpu(entry->usVddc1); + adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc2 = + le16_to_cpu(entry->usVddc2); + adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc3 = + le16_to_cpu(entry->usVddc3); + } else { + adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].vddc = + le16_to_cpu(entry->usVddc); + adev->pm.dpm.dyn_state.cac_leakage_table.entries[i].leakage = + le32_to_cpu(entry->ulLeakageValue); + } + entry = (ATOM_PPLIB_CAC_Leakage_Record *) + ((u8 *)entry + sizeof(ATOM_PPLIB_CAC_Leakage_Record)); + } + adev->pm.dpm.dyn_state.cac_leakage_table.count = cac_table->ucNumEntries; + } + } + + /* ext tables */ + if (le16_to_cpu(power_info->pplib.usTableSize) >= + sizeof(struct _ATOM_PPLIB_POWERPLAYTABLE3)) { + ATOM_PPLIB_EXTENDEDHEADER *ext_hdr = (ATOM_PPLIB_EXTENDEDHEADER *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(power_info->pplib3.usExtendendedHeaderOffset)); + if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V2) && + ext_hdr->usVCETableOffset) { + VCEClockInfoArray *array = (VCEClockInfoArray *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usVCETableOffset) + 1); + ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table *limits = + (ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usVCETableOffset) + 1 + + 1 + array->ucNumEntries * sizeof(VCEClockInfo)); + ATOM_PPLIB_VCE_State_Table *states = + (ATOM_PPLIB_VCE_State_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usVCETableOffset) + 1 + + 1 + (array->ucNumEntries * sizeof (VCEClockInfo)) + + 1 + (limits->numEntries * sizeof(ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record))); + ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record *entry; + ATOM_PPLIB_VCE_State_Record *state_entry; + VCEClockInfo *vce_clk; + u32 size = limits->numEntries * + sizeof(struct amdgpu_vce_clock_voltage_dependency_entry); + adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries = + kzalloc(size, GFP_KERNEL); + if (!adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries) { + amdgpu_free_extended_power_table(adev); + return -ENOMEM; + } + adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.count = + limits->numEntries; + entry = &limits->entries[0]; + state_entry = &states->entries[0]; + for (i = 0; i < limits->numEntries; i++) { + vce_clk = (VCEClockInfo *) + ((u8 *)&array->entries[0] + + (entry->ucVCEClockInfoIndex * sizeof(VCEClockInfo))); + adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].evclk = + le16_to_cpu(vce_clk->usEVClkLow) | (vce_clk->ucEVClkHigh << 16); + adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].ecclk = + le16_to_cpu(vce_clk->usECClkLow) | (vce_clk->ucECClkHigh << 16); + adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table.entries[i].v = + le16_to_cpu(entry->usVoltage); + entry = (ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record *) + ((u8 *)entry + sizeof(ATOM_PPLIB_VCE_Clock_Voltage_Limit_Record)); + } + adev->pm.dpm.num_of_vce_states = + states->numEntries > AMD_MAX_VCE_LEVELS ? + AMD_MAX_VCE_LEVELS : states->numEntries; + for (i = 0; i < adev->pm.dpm.num_of_vce_states; i++) { + vce_clk = (VCEClockInfo *) + ((u8 *)&array->entries[0] + + (state_entry->ucVCEClockInfoIndex * sizeof(VCEClockInfo))); + adev->pm.dpm.vce_states[i].evclk = + le16_to_cpu(vce_clk->usEVClkLow) | (vce_clk->ucEVClkHigh << 16); + adev->pm.dpm.vce_states[i].ecclk = + le16_to_cpu(vce_clk->usECClkLow) | (vce_clk->ucECClkHigh << 16); + adev->pm.dpm.vce_states[i].clk_idx = + state_entry->ucClockInfoIndex & 0x3f; + adev->pm.dpm.vce_states[i].pstate = + (state_entry->ucClockInfoIndex & 0xc0) >> 6; + state_entry = (ATOM_PPLIB_VCE_State_Record *) + ((u8 *)state_entry + sizeof(ATOM_PPLIB_VCE_State_Record)); + } + } + if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V3) && + ext_hdr->usUVDTableOffset) { + UVDClockInfoArray *array = (UVDClockInfoArray *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usUVDTableOffset) + 1); + ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table *limits = + (ATOM_PPLIB_UVD_Clock_Voltage_Limit_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usUVDTableOffset) + 1 + + 1 + (array->ucNumEntries * sizeof (UVDClockInfo))); + ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record *entry; + u32 size = limits->numEntries * + sizeof(struct amdgpu_uvd_clock_voltage_dependency_entry); + adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries = + kzalloc(size, GFP_KERNEL); + if (!adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries) { + amdgpu_free_extended_power_table(adev); + return -ENOMEM; + } + adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.count = + limits->numEntries; + entry = &limits->entries[0]; + for (i = 0; i < limits->numEntries; i++) { + UVDClockInfo *uvd_clk = (UVDClockInfo *) + ((u8 *)&array->entries[0] + + (entry->ucUVDClockInfoIndex * sizeof(UVDClockInfo))); + adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].vclk = + le16_to_cpu(uvd_clk->usVClkLow) | (uvd_clk->ucVClkHigh << 16); + adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].dclk = + le16_to_cpu(uvd_clk->usDClkLow) | (uvd_clk->ucDClkHigh << 16); + adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table.entries[i].v = + le16_to_cpu(entry->usVoltage); + entry = (ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record *) + ((u8 *)entry + sizeof(ATOM_PPLIB_UVD_Clock_Voltage_Limit_Record)); + } + } + if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V4) && + ext_hdr->usSAMUTableOffset) { + ATOM_PPLIB_SAMClk_Voltage_Limit_Table *limits = + (ATOM_PPLIB_SAMClk_Voltage_Limit_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usSAMUTableOffset) + 1); + ATOM_PPLIB_SAMClk_Voltage_Limit_Record *entry; + u32 size = limits->numEntries * + sizeof(struct amdgpu_clock_voltage_dependency_entry); + adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries = + kzalloc(size, GFP_KERNEL); + if (!adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries) { + amdgpu_free_extended_power_table(adev); + return -ENOMEM; + } + adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.count = + limits->numEntries; + entry = &limits->entries[0]; + for (i = 0; i < limits->numEntries; i++) { + adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[i].clk = + le16_to_cpu(entry->usSAMClockLow) | (entry->ucSAMClockHigh << 16); + adev->pm.dpm.dyn_state.samu_clock_voltage_dependency_table.entries[i].v = + le16_to_cpu(entry->usVoltage); + entry = (ATOM_PPLIB_SAMClk_Voltage_Limit_Record *) + ((u8 *)entry + sizeof(ATOM_PPLIB_SAMClk_Voltage_Limit_Record)); + } + } + if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V5) && + ext_hdr->usPPMTableOffset) { + ATOM_PPLIB_PPM_Table *ppm = (ATOM_PPLIB_PPM_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usPPMTableOffset)); + adev->pm.dpm.dyn_state.ppm_table = + kzalloc(sizeof(struct amdgpu_ppm_table), GFP_KERNEL); + if (!adev->pm.dpm.dyn_state.ppm_table) { + amdgpu_free_extended_power_table(adev); + return -ENOMEM; + } + adev->pm.dpm.dyn_state.ppm_table->ppm_design = ppm->ucPpmDesign; + adev->pm.dpm.dyn_state.ppm_table->cpu_core_number = + le16_to_cpu(ppm->usCpuCoreNumber); + adev->pm.dpm.dyn_state.ppm_table->platform_tdp = + le32_to_cpu(ppm->ulPlatformTDP); + adev->pm.dpm.dyn_state.ppm_table->small_ac_platform_tdp = + le32_to_cpu(ppm->ulSmallACPlatformTDP); + adev->pm.dpm.dyn_state.ppm_table->platform_tdc = + le32_to_cpu(ppm->ulPlatformTDC); + adev->pm.dpm.dyn_state.ppm_table->small_ac_platform_tdc = + le32_to_cpu(ppm->ulSmallACPlatformTDC); + adev->pm.dpm.dyn_state.ppm_table->apu_tdp = + le32_to_cpu(ppm->ulApuTDP); + adev->pm.dpm.dyn_state.ppm_table->dgpu_tdp = + le32_to_cpu(ppm->ulDGpuTDP); + adev->pm.dpm.dyn_state.ppm_table->dgpu_ulv_power = + le32_to_cpu(ppm->ulDGpuUlvPower); + adev->pm.dpm.dyn_state.ppm_table->tj_max = + le32_to_cpu(ppm->ulTjmax); + } + if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V6) && + ext_hdr->usACPTableOffset) { + ATOM_PPLIB_ACPClk_Voltage_Limit_Table *limits = + (ATOM_PPLIB_ACPClk_Voltage_Limit_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usACPTableOffset) + 1); + ATOM_PPLIB_ACPClk_Voltage_Limit_Record *entry; + u32 size = limits->numEntries * + sizeof(struct amdgpu_clock_voltage_dependency_entry); + adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries = + kzalloc(size, GFP_KERNEL); + if (!adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries) { + amdgpu_free_extended_power_table(adev); + return -ENOMEM; + } + adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.count = + limits->numEntries; + entry = &limits->entries[0]; + for (i = 0; i < limits->numEntries; i++) { + adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[i].clk = + le16_to_cpu(entry->usACPClockLow) | (entry->ucACPClockHigh << 16); + adev->pm.dpm.dyn_state.acp_clock_voltage_dependency_table.entries[i].v = + le16_to_cpu(entry->usVoltage); + entry = (ATOM_PPLIB_ACPClk_Voltage_Limit_Record *) + ((u8 *)entry + sizeof(ATOM_PPLIB_ACPClk_Voltage_Limit_Record)); + } + } + if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V7) && + ext_hdr->usPowerTuneTableOffset) { + u8 rev = *(u8 *)(mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usPowerTuneTableOffset)); + ATOM_PowerTune_Table *pt; + adev->pm.dpm.dyn_state.cac_tdp_table = + kzalloc(sizeof(struct amdgpu_cac_tdp_table), GFP_KERNEL); + if (!adev->pm.dpm.dyn_state.cac_tdp_table) { + amdgpu_free_extended_power_table(adev); + return -ENOMEM; + } + if (rev > 0) { + ATOM_PPLIB_POWERTUNE_Table_V1 *ppt = (ATOM_PPLIB_POWERTUNE_Table_V1 *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usPowerTuneTableOffset)); + adev->pm.dpm.dyn_state.cac_tdp_table->maximum_power_delivery_limit = + ppt->usMaximumPowerDeliveryLimit; + pt = &ppt->power_tune_table; + } else { + ATOM_PPLIB_POWERTUNE_Table *ppt = (ATOM_PPLIB_POWERTUNE_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usPowerTuneTableOffset)); + adev->pm.dpm.dyn_state.cac_tdp_table->maximum_power_delivery_limit = 255; + pt = &ppt->power_tune_table; + } + adev->pm.dpm.dyn_state.cac_tdp_table->tdp = le16_to_cpu(pt->usTDP); + adev->pm.dpm.dyn_state.cac_tdp_table->configurable_tdp = + le16_to_cpu(pt->usConfigurableTDP); + adev->pm.dpm.dyn_state.cac_tdp_table->tdc = le16_to_cpu(pt->usTDC); + adev->pm.dpm.dyn_state.cac_tdp_table->battery_power_limit = + le16_to_cpu(pt->usBatteryPowerLimit); + adev->pm.dpm.dyn_state.cac_tdp_table->small_power_limit = + le16_to_cpu(pt->usSmallPowerLimit); + adev->pm.dpm.dyn_state.cac_tdp_table->low_cac_leakage = + le16_to_cpu(pt->usLowCACLeakage); + adev->pm.dpm.dyn_state.cac_tdp_table->high_cac_leakage = + le16_to_cpu(pt->usHighCACLeakage); + } + if ((le16_to_cpu(ext_hdr->usSize) >= SIZE_OF_ATOM_PPLIB_EXTENDEDHEADER_V8) && + ext_hdr->usSclkVddgfxTableOffset) { + dep_table = (ATOM_PPLIB_Clock_Voltage_Dependency_Table *) + (mode_info->atom_context->bios + data_offset + + le16_to_cpu(ext_hdr->usSclkVddgfxTableOffset)); + ret = amdgpu_parse_clk_voltage_dep_table( + &adev->pm.dpm.dyn_state.vddgfx_dependency_on_sclk, + dep_table); + if (ret) { + kfree(adev->pm.dpm.dyn_state.vddgfx_dependency_on_sclk.entries); + return ret; + } + } + } + + return 0; +} + +void amdgpu_free_extended_power_table(struct amdgpu_device *adev) +{ + struct amdgpu_dpm_dynamic_state *dyn_state = &adev->pm.dpm.dyn_state; + + kfree(dyn_state->vddc_dependency_on_sclk.entries); + kfree(dyn_state->vddci_dependency_on_mclk.entries); + kfree(dyn_state->vddc_dependency_on_mclk.entries); + kfree(dyn_state->mvdd_dependency_on_mclk.entries); + kfree(dyn_state->cac_leakage_table.entries); + kfree(dyn_state->phase_shedding_limits_table.entries); + kfree(dyn_state->ppm_table); + kfree(dyn_state->cac_tdp_table); + kfree(dyn_state->vce_clock_voltage_dependency_table.entries); + kfree(dyn_state->uvd_clock_voltage_dependency_table.entries); + kfree(dyn_state->samu_clock_voltage_dependency_table.entries); + kfree(dyn_state->acp_clock_voltage_dependency_table.entries); + kfree(dyn_state->vddgfx_dependency_on_sclk.entries); +} + +static const char *pp_lib_thermal_controller_names[] = { + "NONE", + "lm63", + "adm1032", + "adm1030", + "max6649", + "lm64", + "f75375", + "RV6xx", + "RV770", + "adt7473", + "NONE", + "External GPIO", + "Evergreen", + "emc2103", + "Sumo", + "Northern Islands", + "Southern Islands", + "lm96163", + "Sea Islands", + "Kaveri/Kabini", +}; + +void amdgpu_add_thermal_controller(struct amdgpu_device *adev) +{ + struct amdgpu_mode_info *mode_info = &adev->mode_info; + ATOM_PPLIB_POWERPLAYTABLE *power_table; + int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo); + ATOM_PPLIB_THERMALCONTROLLER *controller; + struct amdgpu_i2c_bus_rec i2c_bus; + u16 data_offset; + u8 frev, crev; + + if (!amdgpu_atom_parse_data_header(mode_info->atom_context, index, NULL, + &frev, &crev, &data_offset)) + return; + power_table = (ATOM_PPLIB_POWERPLAYTABLE *) + (mode_info->atom_context->bios + data_offset); + controller = &power_table->sThermalController; + + /* add the i2c bus for thermal/fan chip */ + if (controller->ucType > 0) { + if (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) + adev->pm.no_fan = true; + adev->pm.fan_pulses_per_revolution = + controller->ucFanParameters & ATOM_PP_FANPARAMETERS_TACHOMETER_PULSES_PER_REVOLUTION_MASK; + if (adev->pm.fan_pulses_per_revolution) { + adev->pm.fan_min_rpm = controller->ucFanMinRPM; + adev->pm.fan_max_rpm = controller->ucFanMaxRPM; + } + if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV6xx) { + DRM_INFO("Internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_RV6XX; + } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV770) { + DRM_INFO("Internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_RV770; + } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EVERGREEN) { + DRM_INFO("Internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_EVERGREEN; + } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_SUMO) { + DRM_INFO("Internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_SUMO; + } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_NISLANDS) { + DRM_INFO("Internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_NI; + } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_SISLANDS) { + DRM_INFO("Internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_SI; + } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_CISLANDS) { + DRM_INFO("Internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_CI; + } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_KAVERI) { + DRM_INFO("Internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_KV; + } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO) { + DRM_INFO("External GPIO thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_EXTERNAL_GPIO; + } else if (controller->ucType == + ATOM_PP_THERMALCONTROLLER_ADT7473_WITH_INTERNAL) { + DRM_INFO("ADT7473 with internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_ADT7473_WITH_INTERNAL; + } else if (controller->ucType == + ATOM_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL) { + DRM_INFO("EMC2103 with internal thermal controller %s fan control\n", + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_EMC2103_WITH_INTERNAL; + } else if (controller->ucType < ARRAY_SIZE(pp_lib_thermal_controller_names)) { + DRM_INFO("Possible %s thermal controller at 0x%02x %s fan control\n", + pp_lib_thermal_controller_names[controller->ucType], + controller->ucI2cAddress >> 1, + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + adev->pm.int_thermal_type = THERMAL_TYPE_EXTERNAL; + i2c_bus = amdgpu_atombios_lookup_i2c_gpio(adev, controller->ucI2cLine); + adev->pm.i2c_bus = amdgpu_i2c_lookup(adev, &i2c_bus); + if (adev->pm.i2c_bus) { + struct i2c_board_info info = { }; + const char *name = pp_lib_thermal_controller_names[controller->ucType]; + info.addr = controller->ucI2cAddress >> 1; + strlcpy(info.type, name, sizeof(info.type)); + i2c_new_client_device(&adev->pm.i2c_bus->adapter, &info); + } + } else { + DRM_INFO("Unknown thermal controller type %d at 0x%02x %s fan control\n", + controller->ucType, + controller->ucI2cAddress >> 1, + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); + } + } +} + +struct amd_vce_state* amdgpu_get_vce_clock_state(void *handle, u32 idx) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + if (idx < adev->pm.dpm.num_of_vce_states) + return &adev->pm.dpm.vce_states[idx]; + + return NULL; +} + +static struct amdgpu_ps *amdgpu_dpm_pick_power_state(struct amdgpu_device *adev, + enum amd_pm_state_type dpm_state) +{ + int i; + struct amdgpu_ps *ps; + u32 ui_class; + bool single_display = (adev->pm.dpm.new_active_crtc_count < 2) ? + true : false; + + /* check if the vblank period is too short to adjust the mclk */ + if (single_display && adev->powerplay.pp_funcs->vblank_too_short) { + if (amdgpu_dpm_vblank_too_short(adev)) + single_display = false; + } + + /* certain older asics have a separare 3D performance state, + * so try that first if the user selected performance + */ + if (dpm_state == POWER_STATE_TYPE_PERFORMANCE) + dpm_state = POWER_STATE_TYPE_INTERNAL_3DPERF; + /* balanced states don't exist at the moment */ + if (dpm_state == POWER_STATE_TYPE_BALANCED) + dpm_state = POWER_STATE_TYPE_PERFORMANCE; + +restart_search: + /* Pick the best power state based on current conditions */ + for (i = 0; i < adev->pm.dpm.num_ps; i++) { + ps = &adev->pm.dpm.ps[i]; + ui_class = ps->class & ATOM_PPLIB_CLASSIFICATION_UI_MASK; + switch (dpm_state) { + /* user states */ + case POWER_STATE_TYPE_BATTERY: + if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_BATTERY) { + if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) { + if (single_display) + return ps; + } else + return ps; + } + break; + case POWER_STATE_TYPE_BALANCED: + if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_BALANCED) { + if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) { + if (single_display) + return ps; + } else + return ps; + } + break; + case POWER_STATE_TYPE_PERFORMANCE: + if (ui_class == ATOM_PPLIB_CLASSIFICATION_UI_PERFORMANCE) { + if (ps->caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) { + if (single_display) + return ps; + } else + return ps; + } + break; + /* internal states */ + case POWER_STATE_TYPE_INTERNAL_UVD: + if (adev->pm.dpm.uvd_ps) + return adev->pm.dpm.uvd_ps; + else + break; + case POWER_STATE_TYPE_INTERNAL_UVD_SD: + if (ps->class & ATOM_PPLIB_CLASSIFICATION_SDSTATE) + return ps; + break; + case POWER_STATE_TYPE_INTERNAL_UVD_HD: + if (ps->class & ATOM_PPLIB_CLASSIFICATION_HDSTATE) + return ps; + break; + case POWER_STATE_TYPE_INTERNAL_UVD_HD2: + if (ps->class & ATOM_PPLIB_CLASSIFICATION_HD2STATE) + return ps; + break; + case POWER_STATE_TYPE_INTERNAL_UVD_MVC: + if (ps->class2 & ATOM_PPLIB_CLASSIFICATION2_MVC) + return ps; + break; + case POWER_STATE_TYPE_INTERNAL_BOOT: + return adev->pm.dpm.boot_ps; + case POWER_STATE_TYPE_INTERNAL_THERMAL: + if (ps->class & ATOM_PPLIB_CLASSIFICATION_THERMAL) + return ps; + break; + case POWER_STATE_TYPE_INTERNAL_ACPI: + if (ps->class & ATOM_PPLIB_CLASSIFICATION_ACPI) + return ps; + break; + case POWER_STATE_TYPE_INTERNAL_ULV: + if (ps->class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) + return ps; + break; + case POWER_STATE_TYPE_INTERNAL_3DPERF: + if (ps->class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE) + return ps; + break; + default: + break; + } + } + /* use a fallback state if we didn't match */ + switch (dpm_state) { + case POWER_STATE_TYPE_INTERNAL_UVD_SD: + dpm_state = POWER_STATE_TYPE_INTERNAL_UVD_HD; + goto restart_search; + case POWER_STATE_TYPE_INTERNAL_UVD_HD: + case POWER_STATE_TYPE_INTERNAL_UVD_HD2: + case POWER_STATE_TYPE_INTERNAL_UVD_MVC: + if (adev->pm.dpm.uvd_ps) { + return adev->pm.dpm.uvd_ps; + } else { + dpm_state = POWER_STATE_TYPE_PERFORMANCE; + goto restart_search; + } + case POWER_STATE_TYPE_INTERNAL_THERMAL: + dpm_state = POWER_STATE_TYPE_INTERNAL_ACPI; + goto restart_search; + case POWER_STATE_TYPE_INTERNAL_ACPI: + dpm_state = POWER_STATE_TYPE_BATTERY; + goto restart_search; + case POWER_STATE_TYPE_BATTERY: + case POWER_STATE_TYPE_BALANCED: + case POWER_STATE_TYPE_INTERNAL_3DPERF: + dpm_state = POWER_STATE_TYPE_PERFORMANCE; + goto restart_search; + default: + break; + } + + return NULL; +} + +static int amdgpu_dpm_change_power_state_locked(struct amdgpu_device *adev) +{ + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + struct amdgpu_ps *ps; + enum amd_pm_state_type dpm_state; + int ret; + bool equal = false; + + /* if dpm init failed */ + if (!adev->pm.dpm_enabled) + return 0; + + if (adev->pm.dpm.user_state != adev->pm.dpm.state) { + /* add other state override checks here */ + if ((!adev->pm.dpm.thermal_active) && + (!adev->pm.dpm.uvd_active)) + adev->pm.dpm.state = adev->pm.dpm.user_state; + } + dpm_state = adev->pm.dpm.state; + + ps = amdgpu_dpm_pick_power_state(adev, dpm_state); + if (ps) + adev->pm.dpm.requested_ps = ps; + else + return -EINVAL; + + if (amdgpu_dpm == 1 && pp_funcs->print_power_state) { + printk("switching from power state:\n"); + amdgpu_dpm_print_power_state(adev, adev->pm.dpm.current_ps); + printk("switching to power state:\n"); + amdgpu_dpm_print_power_state(adev, adev->pm.dpm.requested_ps); + } + + /* update whether vce is active */ + ps->vce_active = adev->pm.dpm.vce_active; + if (pp_funcs->display_configuration_changed) + amdgpu_dpm_display_configuration_changed(adev); + + ret = amdgpu_dpm_pre_set_power_state(adev); + if (ret) + return ret; + + if (pp_funcs->check_state_equal) { + if (0 != amdgpu_dpm_check_state_equal(adev, adev->pm.dpm.current_ps, adev->pm.dpm.requested_ps, &equal)) + equal = false; + } + + if (equal) + return 0; + + if (pp_funcs->set_power_state) + pp_funcs->set_power_state(adev->powerplay.pp_handle); + + amdgpu_dpm_post_set_power_state(adev); + + adev->pm.dpm.current_active_crtcs = adev->pm.dpm.new_active_crtcs; + adev->pm.dpm.current_active_crtc_count = adev->pm.dpm.new_active_crtc_count; + + if (pp_funcs->force_performance_level) { + if (adev->pm.dpm.thermal_active) { + enum amd_dpm_forced_level level = adev->pm.dpm.forced_level; + /* force low perf level for thermal */ + pp_funcs->force_performance_level(adev, AMD_DPM_FORCED_LEVEL_LOW); + /* save the user's level */ + adev->pm.dpm.forced_level = level; + } else { + /* otherwise, user selected level */ + pp_funcs->force_performance_level(adev, adev->pm.dpm.forced_level); + } + } + + return 0; +} + +void amdgpu_legacy_dpm_compute_clocks(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int i = 0; + + if (adev->mode_info.num_crtc) + amdgpu_display_bandwidth_update(adev); + + for (i = 0; i < AMDGPU_MAX_RINGS; i++) { + struct amdgpu_ring *ring = adev->rings[i]; + if (ring && ring->sched.ready) + amdgpu_fence_wait_empty(ring); + } + + amdgpu_dpm_get_active_displays(adev); + + amdgpu_dpm_change_power_state_locked(adev); +} + +void amdgpu_dpm_thermal_work_handler(struct work_struct *work) +{ + struct amdgpu_device *adev = + container_of(work, struct amdgpu_device, + pm.dpm.thermal.work); + const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs; + /* switch to the thermal state */ + enum amd_pm_state_type dpm_state = POWER_STATE_TYPE_INTERNAL_THERMAL; + int temp, size = sizeof(temp); + + if (!adev->pm.dpm_enabled) + return; + + if (!pp_funcs->read_sensor(adev->powerplay.pp_handle, + AMDGPU_PP_SENSOR_GPU_TEMP, + (void *)&temp, + &size)) { + if (temp < adev->pm.dpm.thermal.min_temp) + /* switch back the user state */ + dpm_state = adev->pm.dpm.user_state; + } else { + if (adev->pm.dpm.thermal.high_to_low) + /* switch back the user state */ + dpm_state = adev->pm.dpm.user_state; + } + + if (dpm_state == POWER_STATE_TYPE_INTERNAL_THERMAL) + adev->pm.dpm.thermal_active = true; + else + adev->pm.dpm.thermal_active = false; + + adev->pm.dpm.state = dpm_state; + + amdgpu_legacy_dpm_compute_clocks(adev->powerplay.pp_handle); +} diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h new file mode 100644 index 000000000000..93bd3973330c --- /dev/null +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h @@ -0,0 +1,38 @@ +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef __LEGACY_DPM_H__ +#define __LEGACY_DPM_H__ + +void amdgpu_dpm_print_class_info(u32 class, u32 class2); +void amdgpu_dpm_print_cap_info(u32 caps); +void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev, + struct amdgpu_ps *rps); +int amdgpu_get_platform_caps(struct amdgpu_device *adev); +int amdgpu_parse_extended_power_table(struct amdgpu_device *adev); +void amdgpu_free_extended_power_table(struct amdgpu_device *adev); +void amdgpu_add_thermal_controller(struct amdgpu_device *adev); +struct amd_vce_state* amdgpu_get_vce_clock_state(void *handle, u32 idx); +void amdgpu_pm_print_power_states(struct amdgpu_device *adev); +void amdgpu_legacy_dpm_compute_clocks(void *handle); +void amdgpu_dpm_thermal_work_handler(struct work_struct *work); +#endif diff --git a/drivers/gpu/drm/amd/pm/powerplay/ppsmc.h b/drivers/gpu/drm/amd/pm/legacy-dpm/ppsmc.h index 8463245f424f..8463245f424f 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/ppsmc.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/powerplay/r600_dpm.h b/drivers/gpu/drm/amd/pm/legacy-dpm/r600_dpm.h index 055321f61ca7..055321f61ca7 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/r600_dpm.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/r600_dpm.h diff --git a/drivers/gpu/drm/amd/pm/powerplay/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c index 81f82aa05ec2..caae54487f9c 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/si_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c @@ -28,6 +28,7 @@ #include "amdgpu_pm.h" #include "amdgpu_dpm.h" #include "amdgpu_atombios.h" +#include "amdgpu_dpm_internal.h" #include "amd_pcie.h" #include "sid.h" #include "r600_dpm.h" @@ -37,6 +38,7 @@ #include <linux/math64.h> #include <linux/seq_file.h> #include <linux/firmware.h> +#include <legacy_dpm.h> #define MC_CG_ARB_FREQ_F0 0x0a #define MC_CG_ARB_FREQ_F1 0x0b @@ -96,6 +98,19 @@ union pplib_clock_info { struct _ATOM_PPLIB_SI_CLOCK_INFO si; }; +enum si_dpm_auto_throttle_src { + SI_DPM_AUTO_THROTTLE_SRC_THERMAL, + SI_DPM_AUTO_THROTTLE_SRC_EXTERNAL +}; + +enum si_dpm_event_src { + SI_DPM_EVENT_SRC_ANALOG = 0, + SI_DPM_EVENT_SRC_EXTERNAL = 1, + SI_DPM_EVENT_SRC_DIGITAL = 2, + SI_DPM_EVENT_SRC_ANALOG_OR_EXTERNAL = 3, + SI_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL = 4 +}; + static const u32 r600_utc[R600_PM_NUMBER_OF_TC] = { R600_UTC_DFLT_00, @@ -3718,25 +3733,25 @@ static void si_set_dpm_event_sources(struct amdgpu_device *adev, u32 sources) { struct rv7xx_power_info *pi = rv770_get_pi(adev); bool want_thermal_protection; - enum amdgpu_dpm_event_src dpm_event_src; + enum si_dpm_event_src dpm_event_src; switch (sources) { case 0: default: want_thermal_protection = false; break; - case (1 << AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL): + case (1 << SI_DPM_AUTO_THROTTLE_SRC_THERMAL): want_thermal_protection = true; - dpm_event_src = AMDGPU_DPM_EVENT_SRC_DIGITAL; + dpm_event_src = SI_DPM_EVENT_SRC_DIGITAL; break; - case (1 << AMDGPU_DPM_AUTO_THROTTLE_SRC_EXTERNAL): + case (1 << SI_DPM_AUTO_THROTTLE_SRC_EXTERNAL): want_thermal_protection = true; - dpm_event_src = AMDGPU_DPM_EVENT_SRC_EXTERNAL; + dpm_event_src = SI_DPM_EVENT_SRC_EXTERNAL; break; - case ((1 << AMDGPU_DPM_AUTO_THROTTLE_SRC_EXTERNAL) | - (1 << AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL)): + case ((1 << SI_DPM_AUTO_THROTTLE_SRC_EXTERNAL) | + (1 << SI_DPM_AUTO_THROTTLE_SRC_THERMAL)): want_thermal_protection = true; - dpm_event_src = AMDGPU_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL; + dpm_event_src = SI_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL; break; } @@ -3750,7 +3765,7 @@ static void si_set_dpm_event_sources(struct amdgpu_device *adev, u32 sources) } static void si_enable_auto_throttle_source(struct amdgpu_device *adev, - enum amdgpu_dpm_auto_throttle_src source, + enum si_dpm_auto_throttle_src source, bool enable) { struct rv7xx_power_info *pi = rv770_get_pi(adev); @@ -3877,6 +3892,40 @@ static int si_set_boot_state(struct amdgpu_device *adev) } #endif +static int si_set_powergating_by_smu(void *handle, + uint32_t block_type, + bool gate) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + switch (block_type) { + case AMD_IP_BLOCK_TYPE_UVD: + if (!gate) { + adev->pm.dpm.uvd_active = true; + adev->pm.dpm.state = POWER_STATE_TYPE_INTERNAL_UVD; + } else { + adev->pm.dpm.uvd_active = false; + } + + amdgpu_legacy_dpm_compute_clocks(handle); + break; + case AMD_IP_BLOCK_TYPE_VCE: + if (!gate) { + adev->pm.dpm.vce_active = true; + /* XXX select vce level based on ring/task */ + adev->pm.dpm.vce_level = AMD_VCE_LEVEL_AC_ALL; + } else { + adev->pm.dpm.vce_active = false; + } + + amdgpu_legacy_dpm_compute_clocks(handle); + break; + default: + break; + } + return 0; +} + static int si_set_sw_state(struct amdgpu_device *adev) { return (amdgpu_si_send_msg_to_smc(adev, PPSMC_MSG_SwitchToSwState) == PPSMC_Result_OK) ? @@ -4927,6 +4976,31 @@ static int si_populate_smc_initial_state(struct amdgpu_device *adev, return 0; } +static enum si_pcie_gen si_gen_pcie_gen_support(struct amdgpu_device *adev, + u32 sys_mask, + enum si_pcie_gen asic_gen, + enum si_pcie_gen default_gen) +{ + switch (asic_gen) { + case SI_PCIE_GEN1: + return SI_PCIE_GEN1; + case SI_PCIE_GEN2: + return SI_PCIE_GEN2; + case SI_PCIE_GEN3: + return SI_PCIE_GEN3; + default: + if ((sys_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3) && + (default_gen == SI_PCIE_GEN3)) + return SI_PCIE_GEN3; + else if ((sys_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2) && + (default_gen == SI_PCIE_GEN2)) + return SI_PCIE_GEN2; + else + return SI_PCIE_GEN1; + } + return SI_PCIE_GEN1; +} + static int si_populate_smc_acpi_state(struct amdgpu_device *adev, SISLANDS_SMC_STATETABLE *table) { @@ -4989,10 +5063,10 @@ static int si_populate_smc_acpi_state(struct amdgpu_device *adev, &table->ACPIState.level.std_vddc); } table->ACPIState.level.gen2PCIE = - (u8)amdgpu_get_pcie_gen_support(adev, - si_pi->sys_pcie_mask, - si_pi->boot_pcie_gen, - AMDGPU_PCIE_GEN1); + (u8)si_gen_pcie_gen_support(adev, + si_pi->sys_pcie_mask, + si_pi->boot_pcie_gen, + SI_PCIE_GEN1); if (si_pi->vddc_phase_shed_control) si_populate_phase_shedding_value(adev, @@ -5430,7 +5504,7 @@ static int si_convert_power_level_to_smc(struct amdgpu_device *adev, bool gmc_pg = false; if (eg_pi->pcie_performance_request && - (si_pi->force_pcie_gen != AMDGPU_PCIE_GEN_INVALID)) + (si_pi->force_pcie_gen != SI_PCIE_GEN_INVALID)) level->gen2PCIE = (u8)si_pi->force_pcie_gen; else level->gen2PCIE = (u8)pl->pcie_gen; @@ -6147,8 +6221,8 @@ static void si_enable_voltage_control(struct amdgpu_device *adev, bool enable) WREG32_P(GENERAL_PWRMGT, 0, ~VOLT_PWRMGT_EN); } -static enum amdgpu_pcie_gen si_get_maximum_link_speed(struct amdgpu_device *adev, - struct amdgpu_ps *amdgpu_state) +static enum si_pcie_gen si_get_maximum_link_speed(struct amdgpu_device *adev, + struct amdgpu_ps *amdgpu_state) { struct si_ps *state = si_get_ps(amdgpu_state); int i; @@ -6177,27 +6251,27 @@ static void si_request_link_speed_change_before_state_change(struct amdgpu_devic struct amdgpu_ps *amdgpu_current_state) { struct si_power_info *si_pi = si_get_pi(adev); - enum amdgpu_pcie_gen target_link_speed = si_get_maximum_link_speed(adev, amdgpu_new_state); - enum amdgpu_pcie_gen current_link_speed; + enum si_pcie_gen target_link_speed = si_get_maximum_link_speed(adev, amdgpu_new_state); + enum si_pcie_gen current_link_speed; - if (si_pi->force_pcie_gen == AMDGPU_PCIE_GEN_INVALID) + if (si_pi->force_pcie_gen == SI_PCIE_GEN_INVALID) current_link_speed = si_get_maximum_link_speed(adev, amdgpu_current_state); else current_link_speed = si_pi->force_pcie_gen; - si_pi->force_pcie_gen = AMDGPU_PCIE_GEN_INVALID; + si_pi->force_pcie_gen = SI_PCIE_GEN_INVALID; si_pi->pspp_notify_required = false; if (target_link_speed > current_link_speed) { switch (target_link_speed) { #if defined(CONFIG_ACPI) - case AMDGPU_PCIE_GEN3: + case SI_PCIE_GEN3: if (amdgpu_acpi_pcie_performance_request(adev, PCIE_PERF_REQ_PECI_GEN3, false) == 0) break; - si_pi->force_pcie_gen = AMDGPU_PCIE_GEN2; - if (current_link_speed == AMDGPU_PCIE_GEN2) + si_pi->force_pcie_gen = SI_PCIE_GEN2; + if (current_link_speed == SI_PCIE_GEN2) break; fallthrough; - case AMDGPU_PCIE_GEN2: + case SI_PCIE_GEN2: if (amdgpu_acpi_pcie_performance_request(adev, PCIE_PERF_REQ_PECI_GEN2, false) == 0) break; fallthrough; @@ -6217,13 +6291,13 @@ static void si_notify_link_speed_change_after_state_change(struct amdgpu_device struct amdgpu_ps *amdgpu_current_state) { struct si_power_info *si_pi = si_get_pi(adev); - enum amdgpu_pcie_gen target_link_speed = si_get_maximum_link_speed(adev, amdgpu_new_state); + enum si_pcie_gen target_link_speed = si_get_maximum_link_speed(adev, amdgpu_new_state); u8 request; if (si_pi->pspp_notify_required) { - if (target_link_speed == AMDGPU_PCIE_GEN3) + if (target_link_speed == SI_PCIE_GEN3) request = PCIE_PERF_REQ_PECI_GEN3; - else if (target_link_speed == AMDGPU_PCIE_GEN2) + else if (target_link_speed == SI_PCIE_GEN2) request = PCIE_PERF_REQ_PECI_GEN2; else request = PCIE_PERF_REQ_PECI_GEN1; @@ -6546,6 +6620,9 @@ static int si_dpm_get_fan_speed_pwm(void *handle, u64 tmp64; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (!speed) + return -EINVAL; + if (adev->pm.no_fan) return -ENOENT; @@ -6596,10 +6673,13 @@ static int si_dpm_set_fan_speed_pwm(void *handle, return 0; } -static void si_dpm_set_fan_control_mode(void *handle, u32 mode) +static int si_dpm_set_fan_control_mode(void *handle, u32 mode) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (mode == U32_MAX) + return -EINVAL; + if (mode) { /* stop auto-manage */ if (adev->pm.dpm.fan.ucode_fan_control) @@ -6612,19 +6692,26 @@ static void si_dpm_set_fan_control_mode(void *handle, u32 mode) else si_fan_ctrl_set_default_mode(adev); } + + return 0; } -static u32 si_dpm_get_fan_control_mode(void *handle) +static int si_dpm_get_fan_control_mode(void *handle, u32 *fan_mode) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct si_power_info *si_pi = si_get_pi(adev); u32 tmp; + if (!fan_mode) + return -EINVAL; + if (si_pi->fan_is_controlled_by_smc) return 0; tmp = RREG32(CG_FDO_CTRL2) & FDO_PWM_MODE_MASK; - return (tmp >> FDO_PWM_MODE_SHIFT); + *fan_mode = (tmp >> FDO_PWM_MODE_SHIFT); + + return 0; } #if 0 @@ -6864,7 +6951,7 @@ static int si_dpm_enable(struct amdgpu_device *adev) si_enable_sclk_control(adev, true); si_start_dpm(adev); - si_enable_auto_throttle_source(adev, AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL, true); + si_enable_auto_throttle_source(adev, SI_DPM_AUTO_THROTTLE_SRC_THERMAL, true); si_thermal_start_thermal_controller(adev); ni_update_current_ps(adev, boot_ps); @@ -6904,7 +6991,7 @@ static void si_dpm_disable(struct amdgpu_device *adev) si_enable_power_containment(adev, boot_ps, false); si_enable_smc_cac(adev, boot_ps, false); si_enable_spread_spectrum(adev, false); - si_enable_auto_throttle_source(adev, AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL, false); + si_enable_auto_throttle_source(adev, SI_DPM_AUTO_THROTTLE_SRC_THERMAL, false); si_stop_dpm(adev); si_reset_to_default(adev); si_dpm_stop_smc(adev); @@ -6946,10 +7033,7 @@ static int si_power_control_set_level(struct amdgpu_device *adev) ret = si_resume_smc(adev); if (ret) return ret; - ret = si_set_sw_state(adev); - if (ret) - return ret; - return 0; + return si_set_sw_state(adev); } static void si_set_vce_clock(struct amdgpu_device *adev, @@ -7148,10 +7232,10 @@ static void si_parse_pplib_clock_info(struct amdgpu_device *adev, pl->vddc = le16_to_cpu(clock_info->si.usVDDC); pl->vddci = le16_to_cpu(clock_info->si.usVDDCI); pl->flags = le32_to_cpu(clock_info->si.ulFlags); - pl->pcie_gen = amdgpu_get_pcie_gen_support(adev, - si_pi->sys_pcie_mask, - si_pi->boot_pcie_gen, - clock_info->si.ucPCIEGen); + pl->pcie_gen = si_gen_pcie_gen_support(adev, + si_pi->sys_pcie_mask, + si_pi->boot_pcie_gen, + clock_info->si.ucPCIEGen); /* patch up vddc if necessary */ ret = si_get_leakage_voltage_from_leakage_index(adev, pl->vddc, @@ -7318,7 +7402,7 @@ static int si_dpm_init(struct amdgpu_device *adev) si_pi->sys_pcie_mask = adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_MASK; - si_pi->force_pcie_gen = AMDGPU_PCIE_GEN_INVALID; + si_pi->force_pcie_gen = SI_PCIE_GEN_INVALID; si_pi->boot_pcie_gen = si_get_current_pcie_speed(adev); si_set_max_cu_value(adev); @@ -7713,21 +7797,18 @@ static int si_dpm_sw_init(void *handle) return ret; INIT_WORK(&adev->pm.dpm.thermal.work, amdgpu_dpm_thermal_work_handler); - mutex_lock(&adev->pm.mutex); ret = si_dpm_init(adev); if (ret) goto dpm_failed; adev->pm.dpm.current_ps = adev->pm.dpm.requested_ps = adev->pm.dpm.boot_ps; if (amdgpu_dpm == 1) amdgpu_pm_print_power_states(adev); - mutex_unlock(&adev->pm.mutex); DRM_INFO("amdgpu: dpm initialized\n"); return 0; dpm_failed: si_dpm_fini(adev); - mutex_unlock(&adev->pm.mutex); DRM_ERROR("amdgpu: dpm initialization failed\n"); return ret; } @@ -7738,9 +7819,7 @@ static int si_dpm_sw_fini(void *handle) flush_work(&adev->pm.dpm.thermal.work); - mutex_lock(&adev->pm.mutex); si_dpm_fini(adev); - mutex_unlock(&adev->pm.mutex); return 0; } @@ -7754,15 +7833,13 @@ static int si_dpm_hw_init(void *handle) if (!amdgpu_dpm) return 0; - mutex_lock(&adev->pm.mutex); si_dpm_setup_asic(adev); ret = si_dpm_enable(adev); if (ret) adev->pm.dpm_enabled = false; else adev->pm.dpm_enabled = true; - mutex_unlock(&adev->pm.mutex); - amdgpu_pm_compute_clocks(adev); + amdgpu_legacy_dpm_compute_clocks(adev); return ret; } @@ -7770,11 +7847,8 @@ static int si_dpm_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if (adev->pm.dpm_enabled) { - mutex_lock(&adev->pm.mutex); + if (adev->pm.dpm_enabled) si_dpm_disable(adev); - mutex_unlock(&adev->pm.mutex); - } return 0; } @@ -7784,12 +7858,10 @@ static int si_dpm_suspend(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; if (adev->pm.dpm_enabled) { - mutex_lock(&adev->pm.mutex); /* disable dpm */ si_dpm_disable(adev); /* reset the power state */ adev->pm.dpm.current_ps = adev->pm.dpm.requested_ps = adev->pm.dpm.boot_ps; - mutex_unlock(&adev->pm.mutex); } return 0; } @@ -7801,16 +7873,14 @@ static int si_dpm_resume(void *handle) if (adev->pm.dpm_enabled) { /* asic init will reset to the boot state */ - mutex_lock(&adev->pm.mutex); si_dpm_setup_asic(adev); ret = si_dpm_enable(adev); if (ret) adev->pm.dpm_enabled = false; else adev->pm.dpm_enabled = true; - mutex_unlock(&adev->pm.mutex); if (adev->pm.dpm_enabled) - amdgpu_pm_compute_clocks(adev); + amdgpu_legacy_dpm_compute_clocks(adev); } return 0; } @@ -8055,6 +8125,7 @@ static const struct amd_pm_funcs si_dpm_funcs = { .print_power_state = &si_dpm_print_power_state, .debugfs_print_current_performance_level = &si_dpm_debugfs_print_current_performance_level, .force_performance_level = &si_dpm_force_performance_level, + .set_powergating_by_smu = &si_set_powergating_by_smu, .vblank_too_short = &si_dpm_vblank_too_short, .set_fan_control_mode = &si_dpm_set_fan_control_mode, .get_fan_control_mode = &si_dpm_get_fan_control_mode, @@ -8063,6 +8134,7 @@ static const struct amd_pm_funcs si_dpm_funcs = { .check_state_equal = &si_check_state_equal, .get_vce_clock_state = amdgpu_get_vce_clock_state, .read_sensor = &si_dpm_read_sensor, + .pm_compute_clocks = amdgpu_legacy_dpm_compute_clocks, }; static const struct amdgpu_irq_src_funcs si_dpm_irq_funcs = { diff --git a/drivers/gpu/drm/amd/pm/powerplay/si_dpm.h b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.h index bc0be6818e21..11cb7874a6bb 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/si_dpm.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.h @@ -595,13 +595,20 @@ struct rv7xx_power_info { RV770_SMC_STATETABLE smc_statetable; }; +enum si_pcie_gen { + SI_PCIE_GEN1 = 0, + SI_PCIE_GEN2 = 1, + SI_PCIE_GEN3 = 2, + SI_PCIE_GEN_INVALID = 0xffff +}; + struct rv7xx_pl { u32 sclk; u32 mclk; u16 vddc; u16 vddci; /* eg+ only */ u32 flags; - enum amdgpu_pcie_gen pcie_gen; /* si+ only */ + enum si_pcie_gen pcie_gen; /* si+ only */ }; struct rv7xx_ps { @@ -967,9 +974,9 @@ struct si_power_info { struct si_ulv_param ulv; u32 max_cu; /* pcie gen */ - enum amdgpu_pcie_gen force_pcie_gen; - enum amdgpu_pcie_gen boot_pcie_gen; - enum amdgpu_pcie_gen acpi_pcie_gen; + enum si_pcie_gen force_pcie_gen; + enum si_pcie_gen boot_pcie_gen; + enum si_pcie_gen acpi_pcie_gen; u32 sys_pcie_mask; /* flags */ bool enable_dte; diff --git a/drivers/gpu/drm/amd/pm/powerplay/si_smc.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c index 8f994ffa9cd1..8f994ffa9cd1 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/si_smc.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c diff --git a/drivers/gpu/drm/amd/pm/powerplay/sislands_smc.h b/drivers/gpu/drm/amd/pm/legacy-dpm/sislands_smc.h index c7dc117a688c..c7dc117a688c 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/sislands_smc.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/sislands_smc.h diff --git a/drivers/gpu/drm/amd/pm/powerplay/Makefile b/drivers/gpu/drm/amd/pm/powerplay/Makefile index 0fb114adc79f..795a3624cbbf 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/Makefile +++ b/drivers/gpu/drm/amd/pm/powerplay/Makefile @@ -30,10 +30,6 @@ include $(AMD_POWERPLAY) POWER_MGR-y = amd_powerplay.o -POWER_MGR-$(CONFIG_DRM_AMDGPU_CIK)+= kv_dpm.o kv_smc.o - -POWER_MGR-$(CONFIG_DRM_AMDGPU_SI)+= si_dpm.o si_smc.o - AMD_PP_POWER = $(addprefix $(AMD_PP_PATH)/,$(POWER_MGR-y)) AMD_POWERPLAY_FILES += $(AMD_PP_POWER) diff --git a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c index 3ab67b232cd4..a2da46bf3985 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c @@ -31,7 +31,8 @@ #include "power_state.h" #include "amdgpu.h" #include "hwmgr.h" - +#include "amdgpu_dpm_internal.h" +#include "amdgpu_display.h" static const struct amd_pm_funcs pp_dpm_funcs; @@ -49,7 +50,6 @@ static int amd_powerplay_create(struct amdgpu_device *adev) hwmgr->adev = adev; hwmgr->not_vf = !amdgpu_sriov_vf(adev); hwmgr->device = amdgpu_cgs_create_device(adev); - mutex_init(&hwmgr->smu_lock); mutex_init(&hwmgr->msg_lock); hwmgr->chip_family = adev->family; hwmgr->chip_id = adev->asic_type; @@ -177,12 +177,9 @@ static int pp_late_init(void *handle) struct amdgpu_device *adev = handle; struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; - if (hwmgr && hwmgr->pm_en) { - mutex_lock(&hwmgr->smu_lock); + if (hwmgr && hwmgr->pm_en) hwmgr_handle_task(hwmgr, AMD_PP_TASK_COMPLETE_INIT, NULL); - mutex_unlock(&hwmgr->smu_lock); - } if (adev->pm.smu_prv_buffer_size != 0) pp_reserve_vram_for_smu(adev); @@ -322,12 +319,6 @@ static void pp_dpm_en_umd_pstate(struct pp_hwmgr *hwmgr, if (*level & profile_mode_mask) { hwmgr->saved_dpm_level = hwmgr->dpm_level; hwmgr->en_umd_pstate = true; - amdgpu_device_ip_set_powergating_state(hwmgr->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_PG_STATE_UNGATE); - amdgpu_device_ip_set_clockgating_state(hwmgr->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_CG_STATE_UNGATE); } } else { /* exit umd pstate, restore level, enable gfx cg*/ @@ -335,12 +326,6 @@ static void pp_dpm_en_umd_pstate(struct pp_hwmgr *hwmgr, if (*level == AMD_DPM_FORCED_LEVEL_PROFILE_EXIT) *level = hwmgr->saved_dpm_level; hwmgr->en_umd_pstate = false; - amdgpu_device_ip_set_clockgating_state(hwmgr->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_CG_STATE_GATE); - amdgpu_device_ip_set_powergating_state(hwmgr->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_PG_STATE_GATE); } } } @@ -356,11 +341,9 @@ static int pp_dpm_force_performance_level(void *handle, if (level == hwmgr->dpm_level) return 0; - mutex_lock(&hwmgr->smu_lock); pp_dpm_en_umd_pstate(hwmgr, &level); hwmgr->request_dpm_level = level; hwmgr_handle_task(hwmgr, AMD_PP_TASK_READJUST_POWER_STATE, NULL); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -369,21 +352,16 @@ static enum amd_dpm_forced_level pp_dpm_get_performance_level( void *handle) { struct pp_hwmgr *hwmgr = handle; - enum amd_dpm_forced_level level; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - level = hwmgr->dpm_level; - mutex_unlock(&hwmgr->smu_lock); - return level; + return hwmgr->dpm_level; } static uint32_t pp_dpm_get_sclk(void *handle, bool low) { struct pp_hwmgr *hwmgr = handle; - uint32_t clk = 0; if (!hwmgr || !hwmgr->pm_en) return 0; @@ -392,16 +370,12 @@ static uint32_t pp_dpm_get_sclk(void *handle, bool low) pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } - mutex_lock(&hwmgr->smu_lock); - clk = hwmgr->hwmgr_func->get_sclk(hwmgr, low); - mutex_unlock(&hwmgr->smu_lock); - return clk; + return hwmgr->hwmgr_func->get_sclk(hwmgr, low); } static uint32_t pp_dpm_get_mclk(void *handle, bool low) { struct pp_hwmgr *hwmgr = handle; - uint32_t clk = 0; if (!hwmgr || !hwmgr->pm_en) return 0; @@ -410,10 +384,7 @@ static uint32_t pp_dpm_get_mclk(void *handle, bool low) pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } - mutex_lock(&hwmgr->smu_lock); - clk = hwmgr->hwmgr_func->get_mclk(hwmgr, low); - mutex_unlock(&hwmgr->smu_lock); - return clk; + return hwmgr->hwmgr_func->get_mclk(hwmgr, low); } static void pp_dpm_powergate_vce(void *handle, bool gate) @@ -427,9 +398,7 @@ static void pp_dpm_powergate_vce(void *handle, bool gate) pr_info_ratelimited("%s was not implemented.\n", __func__); return; } - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->powergate_vce(hwmgr, gate); - mutex_unlock(&hwmgr->smu_lock); } static void pp_dpm_powergate_uvd(void *handle, bool gate) @@ -443,25 +412,18 @@ static void pp_dpm_powergate_uvd(void *handle, bool gate) pr_info_ratelimited("%s was not implemented.\n", __func__); return; } - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->powergate_uvd(hwmgr, gate); - mutex_unlock(&hwmgr->smu_lock); } static int pp_dpm_dispatch_tasks(void *handle, enum amd_pp_task task_id, enum amd_pm_state_type *user_state) { - int ret = 0; struct pp_hwmgr *hwmgr = handle; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr_handle_task(hwmgr, task_id, user_state); - mutex_unlock(&hwmgr->smu_lock); - - return ret; + return hwmgr_handle_task(hwmgr, task_id, user_state); } static enum amd_pm_state_type pp_dpm_get_current_power_state(void *handle) @@ -473,8 +435,6 @@ static enum amd_pm_state_type pp_dpm_get_current_power_state(void *handle) if (!hwmgr || !hwmgr->pm_en || !hwmgr->current_ps) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - state = hwmgr->current_ps; switch (state->classification.ui_label) { @@ -494,115 +454,107 @@ static enum amd_pm_state_type pp_dpm_get_current_power_state(void *handle) pm_type = POWER_STATE_TYPE_DEFAULT; break; } - mutex_unlock(&hwmgr->smu_lock); return pm_type; } -static void pp_dpm_set_fan_control_mode(void *handle, uint32_t mode) +static int pp_dpm_set_fan_control_mode(void *handle, uint32_t mode) { struct pp_hwmgr *hwmgr = handle; if (!hwmgr || !hwmgr->pm_en) - return; + return -EOPNOTSUPP; + + if (hwmgr->hwmgr_func->set_fan_control_mode == NULL) + return -EOPNOTSUPP; + + if (mode == U32_MAX) + return -EINVAL; - if (hwmgr->hwmgr_func->set_fan_control_mode == NULL) { - pr_info_ratelimited("%s was not implemented.\n", __func__); - return; - } - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->set_fan_control_mode(hwmgr, mode); - mutex_unlock(&hwmgr->smu_lock); + + return 0; } -static uint32_t pp_dpm_get_fan_control_mode(void *handle) +static int pp_dpm_get_fan_control_mode(void *handle, uint32_t *fan_mode) { struct pp_hwmgr *hwmgr = handle; - uint32_t mode = 0; if (!hwmgr || !hwmgr->pm_en) - return 0; + return -EOPNOTSUPP; - if (hwmgr->hwmgr_func->get_fan_control_mode == NULL) { - pr_info_ratelimited("%s was not implemented.\n", __func__); - return 0; - } - mutex_lock(&hwmgr->smu_lock); - mode = hwmgr->hwmgr_func->get_fan_control_mode(hwmgr); - mutex_unlock(&hwmgr->smu_lock); - return mode; + if (hwmgr->hwmgr_func->get_fan_control_mode == NULL) + return -EOPNOTSUPP; + + if (!fan_mode) + return -EINVAL; + + *fan_mode = hwmgr->hwmgr_func->get_fan_control_mode(hwmgr); + return 0; } static int pp_dpm_set_fan_speed_pwm(void *handle, uint32_t speed) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) + return -EOPNOTSUPP; + + if (hwmgr->hwmgr_func->set_fan_speed_pwm == NULL) + return -EOPNOTSUPP; + + if (speed == U32_MAX) return -EINVAL; - if (hwmgr->hwmgr_func->set_fan_speed_pwm == NULL) { - pr_info_ratelimited("%s was not implemented.\n", __func__); - return 0; - } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->set_fan_speed_pwm(hwmgr, speed); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->set_fan_speed_pwm(hwmgr, speed); } static int pp_dpm_get_fan_speed_pwm(void *handle, uint32_t *speed) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) - return -EINVAL; + return -EOPNOTSUPP; - if (hwmgr->hwmgr_func->get_fan_speed_pwm == NULL) { - pr_info_ratelimited("%s was not implemented.\n", __func__); - return 0; - } + if (hwmgr->hwmgr_func->get_fan_speed_pwm == NULL) + return -EOPNOTSUPP; - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->get_fan_speed_pwm(hwmgr, speed); - mutex_unlock(&hwmgr->smu_lock); - return ret; + if (!speed) + return -EINVAL; + + return hwmgr->hwmgr_func->get_fan_speed_pwm(hwmgr, speed); } static int pp_dpm_get_fan_speed_rpm(void *handle, uint32_t *rpm) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) - return -EINVAL; + return -EOPNOTSUPP; if (hwmgr->hwmgr_func->get_fan_speed_rpm == NULL) + return -EOPNOTSUPP; + + if (!rpm) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->get_fan_speed_rpm(hwmgr, rpm); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->get_fan_speed_rpm(hwmgr, rpm); } static int pp_dpm_set_fan_speed_rpm(void *handle, uint32_t rpm) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) + return -EOPNOTSUPP; + + if (hwmgr->hwmgr_func->set_fan_speed_rpm == NULL) + return -EOPNOTSUPP; + + if (rpm == U32_MAX) return -EINVAL; - if (hwmgr->hwmgr_func->set_fan_speed_rpm == NULL) { - pr_info_ratelimited("%s was not implemented.\n", __func__); - return 0; - } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->set_fan_speed_rpm(hwmgr, rpm); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->set_fan_speed_rpm(hwmgr, rpm); } static int pp_dpm_get_pp_num_states(void *handle, @@ -616,8 +568,6 @@ static int pp_dpm_get_pp_num_states(void *handle, if (!hwmgr || !hwmgr->pm_en ||!hwmgr->ps) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - data->nums = hwmgr->num_ps; for (i = 0; i < hwmgr->num_ps; i++) { @@ -640,23 +590,18 @@ static int pp_dpm_get_pp_num_states(void *handle, data->states[i] = POWER_STATE_TYPE_DEFAULT; } } - mutex_unlock(&hwmgr->smu_lock); return 0; } static int pp_dpm_get_pp_table(void *handle, char **table) { struct pp_hwmgr *hwmgr = handle; - int size = 0; if (!hwmgr || !hwmgr->pm_en ||!hwmgr->soft_pp_table) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); *table = (char *)hwmgr->soft_pp_table; - size = hwmgr->soft_pp_table_size; - mutex_unlock(&hwmgr->smu_lock); - return size; + return hwmgr->soft_pp_table_size; } static int amd_powerplay_reset(void *handle) @@ -683,13 +628,12 @@ static int pp_dpm_set_pp_table(void *handle, const char *buf, size_t size) if (!hwmgr || !hwmgr->pm_en) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); if (!hwmgr->hardcode_pp_table) { hwmgr->hardcode_pp_table = kmemdup(hwmgr->soft_pp_table, hwmgr->soft_pp_table_size, GFP_KERNEL); if (!hwmgr->hardcode_pp_table) - goto err; + return ret; } memcpy(hwmgr->hardcode_pp_table, buf, size); @@ -698,17 +642,11 @@ static int pp_dpm_set_pp_table(void *handle, const char *buf, size_t size) ret = amd_powerplay_reset(handle); if (ret) - goto err; + return ret; - if (hwmgr->hwmgr_func->avfs_control) { + if (hwmgr->hwmgr_func->avfs_control) ret = hwmgr->hwmgr_func->avfs_control(hwmgr, false); - if (ret) - goto err; - } - mutex_unlock(&hwmgr->smu_lock); - return 0; -err: - mutex_unlock(&hwmgr->smu_lock); + return ret; } @@ -716,7 +654,6 @@ static int pp_dpm_force_clock_level(void *handle, enum pp_clock_type type, uint32_t mask) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -731,17 +668,13 @@ static int pp_dpm_force_clock_level(void *handle, return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->force_clock_level(hwmgr, type, mask); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->force_clock_level(hwmgr, type, mask); } static int pp_dpm_print_clock_levels(void *handle, enum pp_clock_type type, char *buf) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -750,16 +683,12 @@ static int pp_dpm_print_clock_levels(void *handle, pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->print_clock_levels(hwmgr, type, buf); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->print_clock_levels(hwmgr, type, buf); } static int pp_dpm_get_sclk_od(void *handle) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -768,16 +697,12 @@ static int pp_dpm_get_sclk_od(void *handle) pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->get_sclk_od(hwmgr); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->get_sclk_od(hwmgr); } static int pp_dpm_set_sclk_od(void *handle, uint32_t value) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -787,16 +712,12 @@ static int pp_dpm_set_sclk_od(void *handle, uint32_t value) return 0; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->set_sclk_od(hwmgr, value); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->set_sclk_od(hwmgr, value); } static int pp_dpm_get_mclk_od(void *handle) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -805,16 +726,12 @@ static int pp_dpm_get_mclk_od(void *handle) pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->get_mclk_od(hwmgr); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->get_mclk_od(hwmgr); } static int pp_dpm_set_mclk_od(void *handle, uint32_t value) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -823,17 +740,13 @@ static int pp_dpm_set_mclk_od(void *handle, uint32_t value) pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->set_mclk_od(hwmgr, value); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->set_mclk_od(hwmgr, value); } static int pp_dpm_read_sensor(void *handle, int idx, void *value, int *size) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en || !value) return -EINVAL; @@ -852,10 +765,7 @@ static int pp_dpm_read_sensor(void *handle, int idx, *((uint32_t *)value) = hwmgr->thermal_controller.fanInfo.ulMaxRPM; return 0; default: - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->read_sensor(hwmgr, idx, value, size); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->read_sensor(hwmgr, idx, value, size); } } @@ -875,36 +785,28 @@ pp_dpm_get_vce_clock_state(void *handle, unsigned idx) static int pp_get_power_profile_mode(void *handle, char *buf) { struct pp_hwmgr *hwmgr = handle; - int ret; if (!hwmgr || !hwmgr->pm_en || !hwmgr->hwmgr_func->get_power_profile_mode) return -EOPNOTSUPP; if (!buf) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->get_power_profile_mode(hwmgr, buf); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->get_power_profile_mode(hwmgr, buf); } static int pp_set_power_profile_mode(void *handle, long *input, uint32_t size) { struct pp_hwmgr *hwmgr = handle; - int ret = -EOPNOTSUPP; if (!hwmgr || !hwmgr->pm_en || !hwmgr->hwmgr_func->set_power_profile_mode) - return ret; + return -EOPNOTSUPP; if (hwmgr->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL) { pr_debug("power profile setting is for manual dpm mode only.\n"); return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->set_power_profile_mode(hwmgr, input, size); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return hwmgr->hwmgr_func->set_power_profile_mode(hwmgr, input, size); } static int pp_set_fine_grain_clk_vol(void *handle, uint32_t type, long *input, uint32_t size) @@ -969,8 +871,6 @@ static int pp_dpm_switch_power_profile(void *handle, if (!(type < PP_SMC_POWER_PROFILE_CUSTOM)) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - if (!en) { hwmgr->workload_mask &= ~(1 << hwmgr->workload_prority[type]); index = fls(hwmgr->workload_mask); @@ -985,15 +885,12 @@ static int pp_dpm_switch_power_profile(void *handle, if (type == PP_SMC_POWER_PROFILE_COMPUTE && hwmgr->hwmgr_func->disable_power_features_for_compute_performance) { - if (hwmgr->hwmgr_func->disable_power_features_for_compute_performance(hwmgr, en)) { - mutex_unlock(&hwmgr->smu_lock); + if (hwmgr->hwmgr_func->disable_power_features_for_compute_performance(hwmgr, en)) return -EINVAL; - } } if (hwmgr->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL) hwmgr->hwmgr_func->set_power_profile_mode(hwmgr, &workload, 0); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1023,10 +920,8 @@ static int pp_set_power_limit(void *handle, uint32_t limit) if (limit > max_power_limit) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->set_power_limit(hwmgr, limit); hwmgr->power_limit = limit; - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1043,8 +938,6 @@ static int pp_get_power_limit(void *handle, uint32_t *limit, if (power_type != PP_PWR_TYPE_SUSTAINED) return -EOPNOTSUPP; - mutex_lock(&hwmgr->smu_lock); - switch (pp_limit_level) { case PP_PWR_LIMIT_CURRENT: *limit = hwmgr->power_limit; @@ -1064,8 +957,6 @@ static int pp_get_power_limit(void *handle, uint32_t *limit, break; } - mutex_unlock(&hwmgr->smu_lock); - return ret; } @@ -1077,9 +968,7 @@ static int pp_display_configuration_change(void *handle, if (!hwmgr || !hwmgr->pm_en) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); phm_store_dal_configuration_data(hwmgr, display_config); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1087,15 +976,11 @@ static int pp_get_display_power_level(void *handle, struct amd_pp_simple_clock_info *output) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en ||!output) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - ret = phm_get_dal_power_level(hwmgr, output); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return phm_get_dal_power_level(hwmgr, output); } static int pp_get_current_clocks(void *handle, @@ -1109,8 +994,6 @@ static int pp_get_current_clocks(void *handle, if (!hwmgr || !hwmgr->pm_en) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - phm_get_dal_power_level(hwmgr, &simple_clocks); if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, @@ -1123,7 +1006,6 @@ static int pp_get_current_clocks(void *handle, if (ret) { pr_debug("Error in phm_get_clock_info \n"); - mutex_unlock(&hwmgr->smu_lock); return -EINVAL; } @@ -1146,14 +1028,12 @@ static int pp_get_current_clocks(void *handle, clocks->max_engine_clock_in_sr = hw_clocks.max_eng_clk; clocks->min_engine_clock_in_sr = hw_clocks.min_eng_clk; } - mutex_unlock(&hwmgr->smu_lock); return 0; } static int pp_get_clock_by_type(void *handle, enum amd_pp_clock_type type, struct amd_pp_clocks *clocks) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -1161,10 +1041,7 @@ static int pp_get_clock_by_type(void *handle, enum amd_pp_clock_type type, struc if (clocks == NULL) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - ret = phm_get_clock_by_type(hwmgr, type, clocks); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return phm_get_clock_by_type(hwmgr, type, clocks); } static int pp_get_clock_by_type_with_latency(void *handle, @@ -1172,15 +1049,11 @@ static int pp_get_clock_by_type_with_latency(void *handle, struct pp_clock_levels_with_latency *clocks) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en ||!clocks) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - ret = phm_get_clock_by_type_with_latency(hwmgr, type, clocks); - mutex_unlock(&hwmgr->smu_lock); - return ret; + return phm_get_clock_by_type_with_latency(hwmgr, type, clocks); } static int pp_get_clock_by_type_with_voltage(void *handle, @@ -1188,50 +1061,34 @@ static int pp_get_clock_by_type_with_voltage(void *handle, struct pp_clock_levels_with_voltage *clocks) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en ||!clocks) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - - ret = phm_get_clock_by_type_with_voltage(hwmgr, type, clocks); - - mutex_unlock(&hwmgr->smu_lock); - return ret; + return phm_get_clock_by_type_with_voltage(hwmgr, type, clocks); } static int pp_set_watermarks_for_clocks_ranges(void *handle, void *clock_ranges) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en || !clock_ranges) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - ret = phm_set_watermarks_for_clocks_ranges(hwmgr, - clock_ranges); - mutex_unlock(&hwmgr->smu_lock); - - return ret; + return phm_set_watermarks_for_clocks_ranges(hwmgr, + clock_ranges); } static int pp_display_clock_voltage_request(void *handle, struct pp_display_clock_request *clock) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en ||!clock) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - ret = phm_display_clock_voltage_request(hwmgr, clock); - mutex_unlock(&hwmgr->smu_lock); - - return ret; + return phm_display_clock_voltage_request(hwmgr, clock); } static int pp_get_display_mode_validation_clocks(void *handle, @@ -1245,12 +1102,9 @@ static int pp_get_display_mode_validation_clocks(void *handle, clocks->level = PP_DAL_POWERLEVEL_7; - mutex_lock(&hwmgr->smu_lock); - if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps, PHM_PlatformCaps_DynamicPatchPowerState)) ret = phm_get_max_high_clocks(hwmgr, clocks); - mutex_unlock(&hwmgr->smu_lock); return ret; } @@ -1362,9 +1216,7 @@ static int pp_notify_smu_enable_pwe(void *handle) return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->smus_notify_pwe(hwmgr); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1380,9 +1232,7 @@ static int pp_enable_mgpu_fan_boost(void *handle) hwmgr->hwmgr_func->enable_mgpu_fan_boost == NULL) return 0; - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->enable_mgpu_fan_boost(hwmgr); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1399,9 +1249,7 @@ static int pp_set_min_deep_sleep_dcefclk(void *handle, uint32_t clock) return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->set_min_deep_sleep_dcefclk(hwmgr, clock); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1418,9 +1266,7 @@ static int pp_set_hard_min_dcefclk_by_freq(void *handle, uint32_t clock) return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->set_hard_min_dcefclk_by_freq(hwmgr, clock); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1437,9 +1283,7 @@ static int pp_set_hard_min_fclk_by_freq(void *handle, uint32_t clock) return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->set_hard_min_fclk_by_freq(hwmgr, clock); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1447,16 +1291,11 @@ static int pp_set_hard_min_fclk_by_freq(void *handle, uint32_t clock) static int pp_set_active_display_count(void *handle, uint32_t count) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; - mutex_lock(&hwmgr->smu_lock); - ret = phm_set_active_display_count(hwmgr, count); - mutex_unlock(&hwmgr->smu_lock); - - return ret; + return phm_set_active_display_count(hwmgr, count); } static int pp_get_asic_baco_capability(void *handle, bool *cap) @@ -1471,9 +1310,7 @@ static int pp_get_asic_baco_capability(void *handle, bool *cap) !hwmgr->hwmgr_func->get_asic_baco_capability) return 0; - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->get_asic_baco_capability(hwmgr, cap); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1488,9 +1325,7 @@ static int pp_get_asic_baco_state(void *handle, int *state) if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_asic_baco_state) return 0; - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->get_asic_baco_state(hwmgr, (enum BACO_STATE *)state); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1506,9 +1341,7 @@ static int pp_set_asic_baco_state(void *handle, int state) !hwmgr->hwmgr_func->set_asic_baco_state) return 0; - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->set_asic_baco_state(hwmgr, (enum BACO_STATE)state); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1516,7 +1349,6 @@ static int pp_set_asic_baco_state(void *handle, int state) static int pp_get_ppfeature_status(void *handle, char *buf) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en || !buf) return -EINVAL; @@ -1526,17 +1358,12 @@ static int pp_get_ppfeature_status(void *handle, char *buf) return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->get_ppfeature_status(hwmgr, buf); - mutex_unlock(&hwmgr->smu_lock); - - return ret; + return hwmgr->hwmgr_func->get_ppfeature_status(hwmgr, buf); } static int pp_set_ppfeature_status(void *handle, uint64_t ppfeature_masks) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -1546,17 +1373,12 @@ static int pp_set_ppfeature_status(void *handle, uint64_t ppfeature_masks) return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->set_ppfeature_status(hwmgr, ppfeature_masks); - mutex_unlock(&hwmgr->smu_lock); - - return ret; + return hwmgr->hwmgr_func->set_ppfeature_status(hwmgr, ppfeature_masks); } static int pp_asic_reset_mode_2(void *handle) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -1566,17 +1388,12 @@ static int pp_asic_reset_mode_2(void *handle) return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->asic_reset(hwmgr, SMU_ASIC_RESET_MODE_2); - mutex_unlock(&hwmgr->smu_lock); - - return ret; + return hwmgr->hwmgr_func->asic_reset(hwmgr, SMU_ASIC_RESET_MODE_2); } static int pp_smu_i2c_bus_access(void *handle, bool acquire) { struct pp_hwmgr *hwmgr = handle; - int ret = 0; if (!hwmgr || !hwmgr->pm_en) return -EINVAL; @@ -1586,11 +1403,7 @@ static int pp_smu_i2c_bus_access(void *handle, bool acquire) return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); - ret = hwmgr->hwmgr_func->smu_i2c_bus_access(hwmgr, acquire); - mutex_unlock(&hwmgr->smu_lock); - - return ret; + return hwmgr->hwmgr_func->smu_i2c_bus_access(hwmgr, acquire); } static int pp_set_df_cstate(void *handle, enum pp_df_cstate state) @@ -1603,9 +1416,7 @@ static int pp_set_df_cstate(void *handle, enum pp_df_cstate state) if (!hwmgr->pm_en || !hwmgr->hwmgr_func->set_df_cstate) return 0; - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->set_df_cstate(hwmgr, state); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1620,9 +1431,7 @@ static int pp_set_xgmi_pstate(void *handle, uint32_t pstate) if (!hwmgr->pm_en || !hwmgr->hwmgr_func->set_xgmi_pstate) return 0; - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->set_xgmi_pstate(hwmgr, pstate); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1630,7 +1439,6 @@ static int pp_set_xgmi_pstate(void *handle, uint32_t pstate) static ssize_t pp_get_gpu_metrics(void *handle, void **table) { struct pp_hwmgr *hwmgr = handle; - ssize_t size; if (!hwmgr) return -EINVAL; @@ -1638,11 +1446,7 @@ static ssize_t pp_get_gpu_metrics(void *handle, void **table) if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_gpu_metrics) return -EOPNOTSUPP; - mutex_lock(&hwmgr->smu_lock); - size = hwmgr->hwmgr_func->get_gpu_metrics(hwmgr, table); - mutex_unlock(&hwmgr->smu_lock); - - return size; + return hwmgr->hwmgr_func->get_gpu_metrics(hwmgr, table); } static int pp_gfx_state_change_set(void *handle, uint32_t state) @@ -1657,9 +1461,7 @@ static int pp_gfx_state_change_set(void *handle, uint32_t state) return -EINVAL; } - mutex_lock(&hwmgr->smu_lock); hwmgr->hwmgr_func->gfx_state_change(hwmgr, state); - mutex_unlock(&hwmgr->smu_lock); return 0; } @@ -1673,16 +1475,49 @@ static int pp_get_prv_buffer_details(void *handle, void **addr, size_t *size) *addr = NULL; *size = 0; - mutex_lock(&hwmgr->smu_lock); if (adev->pm.smu_prv_buffer) { amdgpu_bo_kmap(adev->pm.smu_prv_buffer, addr); *size = adev->pm.smu_prv_buffer_size; } - mutex_unlock(&hwmgr->smu_lock); return 0; } +static void pp_pm_compute_clocks(void *handle) +{ + struct pp_hwmgr *hwmgr = handle; + struct amdgpu_device *adev = hwmgr->adev; + int i = 0; + + if (adev->mode_info.num_crtc) + amdgpu_display_bandwidth_update(adev); + + for (i = 0; i < AMDGPU_MAX_RINGS; i++) { + struct amdgpu_ring *ring = adev->rings[i]; + if (ring && ring->sched.ready) + amdgpu_fence_wait_empty(ring); + } + + if (!amdgpu_device_has_dc_support(adev)) { + amdgpu_dpm_get_active_displays(adev); + adev->pm.pm_display_cfg.num_display = adev->pm.dpm.new_active_crtc_count; + adev->pm.pm_display_cfg.vrefresh = amdgpu_dpm_get_vrefresh(adev); + adev->pm.pm_display_cfg.min_vblank_time = amdgpu_dpm_get_vblank_time(adev); + /* we have issues with mclk switching with + * refresh rates over 120 hz on the non-DC code. + */ + if (adev->pm.pm_display_cfg.vrefresh > 120) + adev->pm.pm_display_cfg.min_vblank_time = 0; + + pp_display_configuration_change(handle, + &adev->pm.pm_display_cfg); + } + + pp_dpm_dispatch_tasks(handle, + AMD_PP_TASK_DISPLAY_CONFIG_CHANGE, + NULL); +} + static const struct amd_pm_funcs pp_dpm_funcs = { .load_firmware = pp_dpm_load_fw, .wait_for_fw_loading_complete = pp_dpm_fw_loading_complete, @@ -1747,4 +1582,5 @@ static const struct amd_pm_funcs pp_dpm_funcs = { .get_gpu_metrics = pp_get_gpu_metrics, .gfx_state_change_set = pp_gfx_state_change_set, .get_smu_prv_buf_details = pp_get_prv_buffer_details, + .pm_compute_clocks = pp_pm_compute_clocks, }; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c index cd99db0dc2be..a1e11037831a 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c @@ -3295,10 +3295,6 @@ static int smu7_apply_state_adjust_rules(struct pp_hwmgr *hwmgr, request_ps->classification.ui_label); data->mclk_ignore_signal = false; - PP_ASSERT_WITH_CODE(smu7_ps->performance_level_count == 2, - "VI should always have 2 performance levels", - ); - max_limits = adev->pm.ac_power ? &(hwmgr->dyn_state.max_clock_voltage_on_ac) : &(hwmgr->dyn_state.max_clock_voltage_on_dc); diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu8_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu8_hwmgr.c index 03bf8f069222..b50fd4a4a3d1 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu8_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu8_hwmgr.c @@ -1950,9 +1950,12 @@ static void smu8_dpm_powergate_acp(struct pp_hwmgr *hwmgr, bool bgate) smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ACPPowerON, NULL); } +#define WIDTH_4K 3840 + static void smu8_dpm_powergate_uvd(struct pp_hwmgr *hwmgr, bool bgate) { struct smu8_hwmgr *data = hwmgr->backend; + struct amdgpu_device *adev = hwmgr->adev; data->uvd_power_gated = bgate; @@ -1976,6 +1979,12 @@ static void smu8_dpm_powergate_uvd(struct pp_hwmgr *hwmgr, bool bgate) smu8_dpm_update_uvd_dpm(hwmgr, false); } + /* enable/disable Low Memory PState for UVD (4k videos) */ + if (adev->asic_type == CHIP_STONEY && + adev->uvd.decode_image_width >= WIDTH_4K) + smu8_nbdpm_pstate_enable_disable(hwmgr, + bgate, + true); } static void smu8_dpm_powergate_vce(struct pp_hwmgr *hwmgr, bool bgate) @@ -2037,7 +2046,6 @@ static const struct pp_hwmgr_func smu8_hwmgr_funcs = { .power_state_set = smu8_set_power_state_tasks, .dynamic_state_management_disable = smu8_disable_dpm_tasks, .notify_cac_buffer_info = smu8_notify_cac_buffer_info, - .update_nbdpm_pstate = smu8_nbdpm_pstate_enable_disable, .get_thermal_temperature_range = smu8_get_thermal_temperature_range, }; diff --git a/drivers/gpu/drm/amd/pm/inc/amd_powerplay.h b/drivers/gpu/drm/amd/pm/powerplay/inc/amd_powerplay.h index fe3665965416..fe3665965416 100644 --- a/drivers/gpu/drm/amd/pm/inc/amd_powerplay.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/amd_powerplay.h diff --git a/drivers/gpu/drm/amd/pm/inc/cz_ppsmc.h b/drivers/gpu/drm/amd/pm/powerplay/inc/cz_ppsmc.h index 9b698780aed8..9b698780aed8 100644 --- a/drivers/gpu/drm/amd/pm/inc/cz_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/cz_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/fiji_ppsmc.h b/drivers/gpu/drm/amd/pm/powerplay/inc/fiji_ppsmc.h index 7ae494569a60..7ae494569a60 100644 --- a/drivers/gpu/drm/amd/pm/inc/fiji_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/fiji_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/hardwaremanager.h b/drivers/gpu/drm/amd/pm/powerplay/inc/hardwaremanager.h index 6e0be6027705..6e0be6027705 100644 --- a/drivers/gpu/drm/amd/pm/inc/hardwaremanager.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/hardwaremanager.h diff --git a/drivers/gpu/drm/amd/pm/inc/hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h index 8ed01071fe5a..4f7f2f455301 100644 --- a/drivers/gpu/drm/amd/pm/inc/hwmgr.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h @@ -331,9 +331,6 @@ struct pp_hwmgr_func { uint32_t mc_addr_low, uint32_t mc_addr_hi, uint32_t size); - int (*update_nbdpm_pstate)(struct pp_hwmgr *hwmgr, - bool enable, - bool lock); int (*get_thermal_temperature_range)(struct pp_hwmgr *hwmgr, struct PP_TemperatureRange *range); int (*get_power_profile_mode)(struct pp_hwmgr *hwmgr, char *buf); @@ -751,7 +748,6 @@ struct pp_hwmgr { bool not_vf; bool pm_en; bool pp_one_vf; - struct mutex smu_lock; struct mutex msg_lock; uint32_t pp_table_version; diff --git a/drivers/gpu/drm/amd/pm/inc/polaris10_pwrvirus.h b/drivers/gpu/drm/amd/pm/powerplay/inc/polaris10_pwrvirus.h index 6a53b7e74ccd..6a53b7e74ccd 100644 --- a/drivers/gpu/drm/amd/pm/inc/polaris10_pwrvirus.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/polaris10_pwrvirus.h diff --git a/drivers/gpu/drm/amd/pm/inc/power_state.h b/drivers/gpu/drm/amd/pm/powerplay/inc/power_state.h index a5f2227a3971..a5f2227a3971 100644 --- a/drivers/gpu/drm/amd/pm/inc/power_state.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/power_state.h diff --git a/drivers/gpu/drm/amd/pm/inc/pp_debug.h b/drivers/gpu/drm/amd/pm/powerplay/inc/pp_debug.h index cea65093b6ad..cea65093b6ad 100644 --- a/drivers/gpu/drm/amd/pm/inc/pp_debug.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/pp_debug.h diff --git a/drivers/gpu/drm/amd/pm/inc/pp_endian.h b/drivers/gpu/drm/amd/pm/powerplay/inc/pp_endian.h index f49d1963fe85..f49d1963fe85 100644 --- a/drivers/gpu/drm/amd/pm/inc/pp_endian.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/pp_endian.h diff --git a/drivers/gpu/drm/amd/pm/inc/pp_thermal.h b/drivers/gpu/drm/amd/pm/powerplay/inc/pp_thermal.h index f7c41185097e..f7c41185097e 100644 --- a/drivers/gpu/drm/amd/pm/inc/pp_thermal.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/pp_thermal.h diff --git a/drivers/gpu/drm/amd/pm/inc/ppinterrupt.h b/drivers/gpu/drm/amd/pm/powerplay/inc/ppinterrupt.h index c067e0925b6b..c067e0925b6b 100644 --- a/drivers/gpu/drm/amd/pm/inc/ppinterrupt.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/ppinterrupt.h diff --git a/drivers/gpu/drm/amd/pm/inc/rv_ppsmc.h b/drivers/gpu/drm/amd/pm/powerplay/inc/rv_ppsmc.h index 171f12b82716..171f12b82716 100644 --- a/drivers/gpu/drm/amd/pm/inc/rv_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/rv_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu10.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu10.h index 9e837a5014c5..9e837a5014c5 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu10.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu10.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu10_driver_if.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu10_driver_if.h index c498158771cc..c498158771cc 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu10_driver_if.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu10_driver_if.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu11_driver_if.h index fdc6b7a57bc9..fdc6b7a57bc9 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu11_driver_if.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu7.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7.h index e14072d45918..e14072d45918 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu7.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu71.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu71.h index 71c9b2d28640..71c9b2d28640 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu71.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu71.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu71_discrete.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu71_discrete.h index c0e3936d5c2e..c0e3936d5c2e 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu71_discrete.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu71_discrete.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu72.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu72.h index 9ad1cefff79f..9ad1cefff79f 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu72.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu72.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu72_discrete.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu72_discrete.h index 2aefbb85f620..2aefbb85f620 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu72_discrete.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu72_discrete.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu73.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu73.h index c6b12a4c00db..c6b12a4c00db 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu73.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu73.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu73_discrete.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu73_discrete.h index 5916be08a7fe..5916be08a7fe 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu73_discrete.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu73_discrete.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu74.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu74.h index fd10a9fa843d..fd10a9fa843d 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu74.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu74.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu74_discrete.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu74_discrete.h index 350889e408d2..350889e408d2 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu74_discrete.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu74_discrete.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu75.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu75.h index 771523001533..771523001533 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu75.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu75.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu75_discrete.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu75_discrete.h index b64e58a22ddf..b64e58a22ddf 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu75_discrete.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu75_discrete.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu7_common.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_common.h index 94bf7b649c20..94bf7b649c20 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu7_common.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_common.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu7_discrete.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_discrete.h index ee876745dd12..ee876745dd12 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu7_discrete.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_discrete.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu7_fusion.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_fusion.h index 78ada9ffd508..78ada9ffd508 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu7_fusion.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_fusion.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu7_ppsmc.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_ppsmc.h index a0a38b8a4b1b..a0a38b8a4b1b 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu7_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu8.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu8.h index d758d07b6a31..d758d07b6a31 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu8.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu8.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu8_fusion.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu8_fusion.h index 0c37c94e9414..0c37c94e9414 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu8_fusion.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu8_fusion.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu9.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu9.h index 70ac4d477be2..70ac4d477be2 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu9.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu9.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu9_driver_if.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu9_driver_if.h index 2818c98ff5ca..2818c98ff5ca 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu9_driver_if.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu9_driver_if.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_ucode_xfer_cz.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu_ucode_xfer_cz.h index eb0f79f9c876..701aae598b58 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_ucode_xfer_cz.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu_ucode_xfer_cz.h @@ -121,7 +121,7 @@ typedef struct SMU_Task SMU_Task; struct TOC { uint8_t JobList[NUM_JOBLIST_ENTRIES]; - SMU_Task tasks[1]; + SMU_Task tasks[]; }; // META DATA COMMAND Definitions diff --git a/drivers/gpu/drm/amd/pm/inc/smu_ucode_xfer_vi.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu_ucode_xfer_vi.h index 880152c0f775..880152c0f775 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_ucode_xfer_vi.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu_ucode_xfer_vi.h diff --git a/drivers/gpu/drm/amd/pm/inc/smumgr.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smumgr.h index 5f46f1a4f38e..5f46f1a4f38e 100644 --- a/drivers/gpu/drm/amd/pm/inc/smumgr.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smumgr.h diff --git a/drivers/gpu/drm/amd/pm/inc/tonga_ppsmc.h b/drivers/gpu/drm/amd/pm/powerplay/inc/tonga_ppsmc.h index 63631296d751..63631296d751 100644 --- a/drivers/gpu/drm/amd/pm/inc/tonga_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/tonga_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/vega10_ppsmc.h b/drivers/gpu/drm/amd/pm/powerplay/inc/vega10_ppsmc.h index 715b5a168831..715b5a168831 100644 --- a/drivers/gpu/drm/amd/pm/inc/vega10_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/vega10_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/vega12/smu9_driver_if.h b/drivers/gpu/drm/amd/pm/powerplay/inc/vega12/smu9_driver_if.h index b6ffd08784e7..b6ffd08784e7 100644 --- a/drivers/gpu/drm/amd/pm/inc/vega12/smu9_driver_if.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/vega12/smu9_driver_if.h diff --git a/drivers/gpu/drm/amd/pm/inc/vega12_ppsmc.h b/drivers/gpu/drm/amd/pm/powerplay/inc/vega12_ppsmc.h index f985c78d746a..f985c78d746a 100644 --- a/drivers/gpu/drm/amd/pm/inc/vega12_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/vega12_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/vega20_ppsmc.h b/drivers/gpu/drm/amd/pm/powerplay/inc/vega20_ppsmc.h index 0c66f0fe1aaf..0c66f0fe1aaf 100644 --- a/drivers/gpu/drm/amd/pm/inc/vega20_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/vega20_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c index 93a1c7248e26..5ca3c422f7d4 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c @@ -208,6 +208,7 @@ static int ci_read_smc_sram_dword(struct pp_hwmgr *hwmgr, uint32_t smc_addr, static int ci_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) { + struct amdgpu_device *adev = hwmgr->adev; int ret; cgs_write_register(hwmgr->device, mmSMC_RESP_0, 0); @@ -218,7 +219,8 @@ static int ci_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) ret = PHM_READ_FIELD(hwmgr->device, SMC_RESP_0, SMC_RESP); if (ret != 1) - pr_info("\n failed to send message %x ret is %d\n", msg, ret); + dev_info(adev->dev, + "failed to send message %x ret is %d\n", msg,ret); return 0; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu10_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu10_smumgr.c index 47b34c6ca924..88a5641465dc 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu10_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu10_smumgr.c @@ -87,7 +87,7 @@ static int smu10_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) smu10_send_msg_to_smc_without_waiting(hwmgr, msg); if (smu10_wait_for_response(hwmgr) == 0) - printk("Failed to send Message %x.\n", msg); + dev_err(adev->dev, "Failed to send Message %x.\n", msg); return 0; } @@ -108,7 +108,7 @@ static int smu10_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr, if (smu10_wait_for_response(hwmgr) == 0) - printk("Failed to send Message %x.\n", msg); + dev_err(adev->dev, "Failed to send Message %x.\n", msg); return 0; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c index aae25243eb10..5a010cd38303 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c @@ -165,6 +165,7 @@ bool smu7_is_smc_ram_running(struct pp_hwmgr *hwmgr) int smu7_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) { + struct amdgpu_device *adev = hwmgr->adev; int ret; PHM_WAIT_FIELD_UNEQUAL(hwmgr, SMC_RESP_0, SMC_RESP, 0); @@ -172,9 +173,10 @@ int smu7_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) ret = PHM_READ_FIELD(hwmgr->device, SMC_RESP_0, SMC_RESP); if (ret == 0xFE) - pr_debug("last message was not supported\n"); + dev_dbg(adev->dev, "last message was not supported\n"); else if (ret != 1) - pr_info("\n last message was failed ret is %d\n", ret); + dev_info(adev->dev, + "\nlast message was failed ret is %d\n", ret); cgs_write_register(hwmgr->device, mmSMC_RESP_0, 0); cgs_write_register(hwmgr->device, mmSMC_MESSAGE_0, msg); @@ -184,9 +186,10 @@ int smu7_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) ret = PHM_READ_FIELD(hwmgr->device, SMC_RESP_0, SMC_RESP); if (ret == 0xFE) - pr_debug("message %x was not supported\n", msg); + dev_dbg(adev->dev, "message %x was not supported\n", msg); else if (ret != 1) - pr_info("\n failed to send message %x ret is %d \n", msg, ret); + dev_dbg(adev->dev, + "failed to send message %x ret is %d \n", msg, ret); return 0; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu9_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu9_smumgr.c index 23e5de3c4ec1..8c9bf4940dc1 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu9_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu9_smumgr.c @@ -126,7 +126,7 @@ int smu9_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) ret = smu9_wait_for_response(hwmgr); if (ret != 1) - pr_err("Failed to send message: 0x%x, ret value: 0x%x\n", msg, ret); + dev_err(adev->dev, "Failed to send message: 0x%x, ret value: 0x%x\n", msg, ret); return 0; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega20_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega20_smumgr.c index 741fbc87467f..a5c95b180672 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega20_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vega20_smumgr.c @@ -115,7 +115,7 @@ static int vega20_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg) ret = vega20_wait_for_response(hwmgr); if (ret != PPSMC_Result_OK) - pr_err("Failed to send message 0x%x, response 0x%x\n", msg, ret); + dev_err(adev->dev, "Failed to send message 0x%x, response 0x%x\n", msg, ret); return (ret == PPSMC_Result_OK) ? 0 : -EIO; } @@ -143,7 +143,7 @@ static int vega20_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr, ret = vega20_wait_for_response(hwmgr); if (ret != PPSMC_Result_OK) - pr_err("Failed to send message 0x%x, response 0x%x\n", msg, ret); + dev_err(adev->dev, "Failed to send message 0x%x, response 0x%x\n", msg, ret); return (ret == PPSMC_Result_OK) ? 0 : -EIO; } @@ -520,7 +520,7 @@ static int vega20_smu_init(struct pp_hwmgr *hwmgr) priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].version = 0x01; priv->smu_tables.entry[TABLE_ACTIVITY_MONITOR_COEFF].size = sizeof(DpmActivityMonitorCoeffInt_t); - ret = smu_v11_0_i2c_control_init(&adev->pm.smu_i2c); + ret = smu_v11_0_i2c_control_init(adev); if (ret) goto err4; @@ -558,7 +558,7 @@ static int vega20_smu_fini(struct pp_hwmgr *hwmgr) (struct vega20_smumgr *)(hwmgr->smu_backend); struct amdgpu_device *adev = hwmgr->adev; - smu_v11_0_i2c_control_fini(&adev->pm.smu_i2c); + smu_v11_0_i2c_control_fini(adev); if (priv) { amdgpu_bo_free_kernel(&priv->smu_tables.entry[TABLE_PPTABLE].handle, diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index d93d28c1af95..e846231412bc 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -55,11 +55,10 @@ static int smu_force_smuclk_levels(struct smu_context *smu, uint32_t mask); static int smu_handle_task(struct smu_context *smu, enum amd_dpm_forced_level level, - enum amd_pp_task task_id, - bool lock_needed); + enum amd_pp_task task_id); static int smu_reset(struct smu_context *smu); static int smu_set_fan_speed_pwm(void *handle, u32 speed); -static int smu_set_fan_control_mode(struct smu_context *smu, int value); +static int smu_set_fan_control_mode(void *handle, u32 value); static int smu_set_power_limit(void *handle, uint32_t limit); static int smu_set_fan_speed_rpm(void *handle, uint32_t speed); static int smu_set_gfx_cgpg(struct smu_context *smu, bool enabled); @@ -68,49 +67,32 @@ static int smu_sys_get_pp_feature_mask(void *handle, char *buf) { struct smu_context *smu = handle; - int size = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - - size = smu_get_pp_feature_mask(smu, buf); - - mutex_unlock(&smu->mutex); - - return size; + return smu_get_pp_feature_mask(smu, buf); } static int smu_sys_set_pp_feature_mask(void *handle, uint64_t new_mask) { struct smu_context *smu = handle; - int ret = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - - ret = smu_set_pp_feature_mask(smu, new_mask); - - mutex_unlock(&smu->mutex); - - return ret; + return smu_set_pp_feature_mask(smu, new_mask); } -int smu_get_status_gfxoff(struct amdgpu_device *adev, uint32_t *value) +int smu_get_status_gfxoff(struct smu_context *smu, uint32_t *value) { - int ret = 0; - struct smu_context *smu = &adev->smu; + if (!smu->ppt_funcs->get_gfx_off_status) + return -EINVAL; - if (is_support_sw_smu(adev) && smu->ppt_funcs->get_gfx_off_status) - *value = smu_get_gfx_off_status(smu); - else - ret = -EINVAL; + *value = smu_get_gfx_off_status(smu); - return ret; + return 0; } int smu_set_soft_freq_range(struct smu_context *smu, @@ -120,16 +102,12 @@ int smu_set_soft_freq_range(struct smu_context *smu, { int ret = 0; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->set_soft_freq_limited_range) ret = smu->ppt_funcs->set_soft_freq_limited_range(smu, clk_type, min, max); - mutex_unlock(&smu->mutex); - return ret; } @@ -138,21 +116,17 @@ int smu_get_dpm_freq_range(struct smu_context *smu, uint32_t *min, uint32_t *max) { - int ret = 0; + int ret = -ENOTSUPP; if (!min && !max) return -EINVAL; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->get_dpm_ultimate_freq) ret = smu->ppt_funcs->get_dpm_ultimate_freq(smu, clk_type, min, max); - mutex_unlock(&smu->mutex); - return ret; } @@ -184,8 +158,8 @@ static u32 smu_get_sclk(void *handle, bool low) return clk_freq * 100; } -static int smu_dpm_set_vcn_enable_locked(struct smu_context *smu, - bool enable) +static int smu_dpm_set_vcn_enable(struct smu_context *smu, + bool enable) { struct smu_power_context *smu_power = &smu->smu_power; struct smu_power_gate *power_gate = &smu_power->power_gate; @@ -204,24 +178,8 @@ static int smu_dpm_set_vcn_enable_locked(struct smu_context *smu, return ret; } -static int smu_dpm_set_vcn_enable(struct smu_context *smu, - bool enable) -{ - struct smu_power_context *smu_power = &smu->smu_power; - struct smu_power_gate *power_gate = &smu_power->power_gate; - int ret = 0; - - mutex_lock(&power_gate->vcn_gate_lock); - - ret = smu_dpm_set_vcn_enable_locked(smu, enable); - - mutex_unlock(&power_gate->vcn_gate_lock); - - return ret; -} - -static int smu_dpm_set_jpeg_enable_locked(struct smu_context *smu, - bool enable) +static int smu_dpm_set_jpeg_enable(struct smu_context *smu, + bool enable) { struct smu_power_context *smu_power = &smu->smu_power; struct smu_power_gate *power_gate = &smu_power->power_gate; @@ -240,22 +198,6 @@ static int smu_dpm_set_jpeg_enable_locked(struct smu_context *smu, return ret; } -static int smu_dpm_set_jpeg_enable(struct smu_context *smu, - bool enable) -{ - struct smu_power_context *smu_power = &smu->smu_power; - struct smu_power_gate *power_gate = &smu_power->power_gate; - int ret = 0; - - mutex_lock(&power_gate->jpeg_gate_lock); - - ret = smu_dpm_set_jpeg_enable_locked(smu, enable); - - mutex_unlock(&power_gate->jpeg_gate_lock); - - return ret; -} - /** * smu_dpm_set_power_gate - power gate/ungate the specific IP block * @@ -410,7 +352,7 @@ static void smu_restore_dpm_user_profile(struct smu_context *smu) if (smu->user_dpm_profile.fan_mode == AMD_FAN_CTRL_MANUAL || smu->user_dpm_profile.fan_mode == AMD_FAN_CTRL_NONE) { ret = smu_set_fan_control_mode(smu, smu->user_dpm_profile.fan_mode); - if (ret) { + if (ret != -EOPNOTSUPP) { smu->user_dpm_profile.fan_speed_pwm = 0; smu->user_dpm_profile.fan_speed_rpm = 0; smu->user_dpm_profile.fan_mode = AMD_FAN_CTRL_AUTO; @@ -419,13 +361,13 @@ static void smu_restore_dpm_user_profile(struct smu_context *smu) if (smu->user_dpm_profile.fan_speed_pwm) { ret = smu_set_fan_speed_pwm(smu, smu->user_dpm_profile.fan_speed_pwm); - if (ret) + if (ret != -EOPNOTSUPP) dev_err(smu->adev->dev, "Failed to set manual fan speed in pwm\n"); } if (smu->user_dpm_profile.fan_speed_rpm) { ret = smu_set_fan_speed_rpm(smu, smu->user_dpm_profile.fan_speed_rpm); - if (ret) + if (ret != -EOPNOTSUPP) dev_err(smu->adev->dev, "Failed to set manual fan speed in rpm\n"); } } @@ -471,10 +413,7 @@ bool is_support_sw_smu(struct amdgpu_device *adev) bool is_support_cclk_dpm(struct amdgpu_device *adev) { - struct smu_context *smu = &adev->smu; - - if (!is_support_sw_smu(adev)) - return false; + struct smu_context *smu = adev->powerplay.pp_handle; if (!smu_feature_is_enabled(smu, SMU_FEATURE_CCLK_DPM_BIT)) return false; @@ -488,7 +427,6 @@ static int smu_sys_get_pp_table(void *handle, { struct smu_context *smu = handle; struct smu_table_context *smu_table = &smu->smu_table; - uint32_t powerplay_table_size; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; @@ -496,18 +434,12 @@ static int smu_sys_get_pp_table(void *handle, if (!smu_table->power_play_table && !smu_table->hardcode_pptable) return -EINVAL; - mutex_lock(&smu->mutex); - if (smu_table->hardcode_pptable) *table = smu_table->hardcode_pptable; else *table = smu_table->power_play_table; - powerplay_table_size = smu_table->power_play_table_size; - - mutex_unlock(&smu->mutex); - - return powerplay_table_size; + return smu_table->power_play_table_size; } static int smu_sys_set_pp_table(void *handle, @@ -527,12 +459,10 @@ static int smu_sys_set_pp_table(void *handle, return -EIO; } - mutex_lock(&smu->mutex); - if (!smu_table->hardcode_pptable) - smu_table->hardcode_pptable = kzalloc(size, GFP_KERNEL); if (!smu_table->hardcode_pptable) { - ret = -ENOMEM; - goto failed; + smu_table->hardcode_pptable = kzalloc(size, GFP_KERNEL); + if (!smu_table->hardcode_pptable) + return -ENOMEM; } memcpy(smu_table->hardcode_pptable, buf, size); @@ -551,8 +481,6 @@ static int smu_sys_set_pp_table(void *handle, smu->uploading_custom_pp_table = false; -failed: - mutex_unlock(&smu->mutex); return ret; } @@ -578,7 +506,7 @@ static int smu_get_driver_allowed_feature_mask(struct smu_context *smu) static int smu_set_funcs(struct amdgpu_device *adev) { - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; if (adev->pm.pp_feature & PP_OVERDRIVE_MASK) smu->od_enabled = true; @@ -630,13 +558,15 @@ static int smu_set_funcs(struct amdgpu_device *adev) static int smu_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct smu_context *smu = &adev->smu; + struct smu_context *smu; + + smu = kzalloc(sizeof(struct smu_context), GFP_KERNEL); + if (!smu) + return -ENOMEM; smu->adev = adev; smu->pm_enabled = !!amdgpu_dpm; smu->is_apu = false; - mutex_init(&smu->mutex); - mutex_init(&smu->smu_baco.mutex); smu->smu_baco.state = SMU_BACO_STATE_EXIT; smu->smu_baco.platform_support = false; smu->user_dpm_profile.fan_mode = -1; @@ -657,32 +587,25 @@ static int smu_set_default_dpm_table(struct smu_context *smu) if (!smu->ppt_funcs->set_default_dpm_table) return 0; - mutex_lock(&power_gate->vcn_gate_lock); - mutex_lock(&power_gate->jpeg_gate_lock); - vcn_gate = atomic_read(&power_gate->vcn_gated); jpeg_gate = atomic_read(&power_gate->jpeg_gated); - ret = smu_dpm_set_vcn_enable_locked(smu, true); + ret = smu_dpm_set_vcn_enable(smu, true); if (ret) - goto err0_out; + return ret; - ret = smu_dpm_set_jpeg_enable_locked(smu, true); + ret = smu_dpm_set_jpeg_enable(smu, true); if (ret) - goto err1_out; + goto err_out; ret = smu->ppt_funcs->set_default_dpm_table(smu); if (ret) dev_err(smu->adev->dev, "Failed to setup default dpm clock tables!\n"); - smu_dpm_set_jpeg_enable_locked(smu, !jpeg_gate); -err1_out: - smu_dpm_set_vcn_enable_locked(smu, !vcn_gate); -err0_out: - mutex_unlock(&power_gate->jpeg_gate_lock); - mutex_unlock(&power_gate->vcn_gate_lock); - + smu_dpm_set_jpeg_enable(smu, !jpeg_gate); +err_out: + smu_dpm_set_vcn_enable(smu, !vcn_gate); return ret; } @@ -690,7 +613,7 @@ err0_out: static int smu_late_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; int ret = 0; smu_set_fine_grain_gfx_freq_parameters(smu); @@ -736,10 +659,9 @@ static int smu_late_init(void *handle) smu_get_fan_parameters(smu); - smu_handle_task(&adev->smu, + smu_handle_task(smu, smu->smu_dpm.dpm_level, - AMD_PP_TASK_COMPLETE_INIT, - false); + AMD_PP_TASK_COMPLETE_INIT); smu_restore_dpm_user_profile(smu); @@ -964,7 +886,7 @@ static int smu_smc_table_sw_init(struct smu_context *smu) if (ret) return ret; - ret = smu_i2c_init(smu, &smu->adev->pm.smu_i2c); + ret = smu_i2c_init(smu); if (ret) return ret; @@ -975,7 +897,7 @@ static int smu_smc_table_sw_fini(struct smu_context *smu) { int ret; - smu_i2c_fini(smu, &smu->adev->pm.smu_i2c); + smu_i2c_fini(smu); smu_free_dummy_read_table(smu); @@ -1015,29 +937,21 @@ static void smu_interrupt_work_fn(struct work_struct *work) struct smu_context *smu = container_of(work, struct smu_context, interrupt_work); - mutex_lock(&smu->mutex); - if (smu->ppt_funcs && smu->ppt_funcs->interrupt_work) smu->ppt_funcs->interrupt_work(smu); - - mutex_unlock(&smu->mutex); } static int smu_sw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; int ret; smu->pool_size = adev->pm.smu_prv_buffer_size; smu->smu_feature.feature_num = SMU_FEATURE_MAX; - mutex_init(&smu->smu_feature.mutex); bitmap_zero(smu->smu_feature.supported, SMU_FEATURE_MAX); - bitmap_zero(smu->smu_feature.enabled, SMU_FEATURE_MAX); bitmap_zero(smu->smu_feature.allowed, SMU_FEATURE_MAX); - mutex_init(&smu->sensor_lock); - mutex_init(&smu->metrics_lock); mutex_init(&smu->message_lock); INIT_WORK(&smu->throttling_logging_work, smu_throttling_logging_work_fn); @@ -1049,8 +963,6 @@ static int smu_sw_init(void *handle) atomic_set(&smu->smu_power.power_gate.vcn_gated, 1); atomic_set(&smu->smu_power.power_gate.jpeg_gated, 1); - mutex_init(&smu->smu_power.power_gate.vcn_gate_lock); - mutex_init(&smu->smu_power.power_gate.jpeg_gate_lock); smu->workload_mask = 1 << smu->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT]; smu->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT] = 0; @@ -1101,7 +1013,7 @@ static int smu_sw_init(void *handle) static int smu_sw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; int ret; ret = smu_smc_table_sw_fini(smu); @@ -1144,8 +1056,10 @@ static int smu_get_thermal_temperature_range(struct smu_context *smu) static int smu_smc_hw_setup(struct smu_context *smu) { + struct smu_feature *feature = &smu->smu_feature; struct amdgpu_device *adev = smu->adev; uint32_t pcie_gen = 0, pcie_width = 0; + uint64_t features_supported; int ret = 0; if (adev->in_suspend && smu_is_dpm_running(smu)) { @@ -1225,6 +1139,15 @@ static int smu_smc_hw_setup(struct smu_context *smu) return ret; } + ret = smu_feature_get_enabled_mask(smu, &features_supported); + if (ret) { + dev_err(adev->dev, "Failed to retrieve supported dpm features!\n"); + return ret; + } + bitmap_copy(feature->supported, + (unsigned long *)&features_supported, + feature->feature_num); + if (!smu_is_dpm_running(smu)) dev_info(adev->dev, "dpm has been disabled\n"); @@ -1336,7 +1259,7 @@ static int smu_hw_init(void *handle) { int ret; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) { smu->pm_enabled = false; @@ -1352,7 +1275,7 @@ static int smu_hw_init(void *handle) if (smu->is_apu) { smu_dpm_set_vcn_enable(smu, true); smu_dpm_set_jpeg_enable(smu, true); - smu_set_gfx_cgpg(&adev->smu, true); + smu_set_gfx_cgpg(smu, true); } if (!smu->pm_enabled) @@ -1437,9 +1360,7 @@ static int smu_disable_dpms(struct smu_context *smu) case IP_VERSION(11, 5, 0): case IP_VERSION(11, 0, 12): case IP_VERSION(11, 0, 13): - return smu_disable_all_features_with_exception(smu, - true, - SMU_FEATURE_COUNT); + return 0; default: break; } @@ -1455,9 +1376,7 @@ static int smu_disable_dpms(struct smu_context *smu) case IP_VERSION(11, 0, 0): case IP_VERSION(11, 0, 5): case IP_VERSION(11, 0, 9): - return smu_disable_all_features_with_exception(smu, - true, - SMU_FEATURE_BACO_BIT); + return 0; default: break; } @@ -1469,7 +1388,6 @@ static int smu_disable_dpms(struct smu_context *smu) */ if (use_baco && smu_feature_is_enabled(smu, SMU_FEATURE_BACO_BIT)) { ret = smu_disable_all_features_with_exception(smu, - false, SMU_FEATURE_BACO_BIT); if (ret) dev_err(adev->dev, "Failed to disable smu features except BACO.\n"); @@ -1512,7 +1430,7 @@ static int smu_smc_hw_cleanup(struct smu_context *smu) static int smu_hw_fini(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev)) return 0; @@ -1531,13 +1449,19 @@ static int smu_hw_fini(void *handle) return smu_smc_hw_cleanup(smu); } +static void smu_late_fini(void *handle) +{ + struct amdgpu_device *adev = handle; + struct smu_context *smu = adev->powerplay.pp_handle; + + kfree(smu); +} + static int smu_reset(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; int ret; - amdgpu_gfx_off_ctrl(smu->adev, false); - ret = smu_hw_fini(adev); if (ret) return ret; @@ -1550,15 +1474,13 @@ static int smu_reset(struct smu_context *smu) if (ret) return ret; - amdgpu_gfx_off_ctrl(smu->adev, true); - return 0; } static int smu_suspend(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; int ret; if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev)) @@ -1575,7 +1497,7 @@ static int smu_suspend(void *handle) smu->watermarks_bitmap &= ~(WATERMARKS_LOADED); - smu_set_gfx_cgpg(&adev->smu, false); + smu_set_gfx_cgpg(smu, false); return 0; } @@ -1584,7 +1506,7 @@ static int smu_resume(void *handle) { int ret; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev)) return 0; @@ -1606,7 +1528,7 @@ static int smu_resume(void *handle) return ret; } - smu_set_gfx_cgpg(&adev->smu, true); + smu_set_gfx_cgpg(smu, true); smu->disable_uclk_switch = 0; @@ -1630,8 +1552,6 @@ static int smu_display_configuration_change(void *handle, if (!display_config) return -EINVAL; - mutex_lock(&smu->mutex); - smu_set_min_dcef_deep_sleep(smu, display_config->min_dcef_deep_sleep_set_clk / 100); @@ -1640,8 +1560,6 @@ static int smu_display_configuration_change(void *handle, num_of_active_display++; } - mutex_unlock(&smu->mutex); - return 0; } @@ -1675,14 +1593,7 @@ static int smu_enable_umd_pstate(void *handle, /* enter umd pstate, save current level, disable gfx cg*/ if (*level & profile_mode_mask) { smu_dpm_ctx->saved_dpm_level = smu_dpm_ctx->dpm_level; - smu_dpm_ctx->enable_umd_pstate = true; smu_gpo_control(smu, false); - amdgpu_device_ip_set_powergating_state(smu->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_PG_STATE_UNGATE); - amdgpu_device_ip_set_clockgating_state(smu->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_CG_STATE_UNGATE); smu_gfx_ulv_control(smu, false); smu_deep_sleep_control(smu, false); amdgpu_asic_update_umd_stable_pstate(smu->adev, true); @@ -1692,16 +1603,9 @@ static int smu_enable_umd_pstate(void *handle, if (!(*level & profile_mode_mask)) { if (*level == AMD_DPM_FORCED_LEVEL_PROFILE_EXIT) *level = smu_dpm_ctx->saved_dpm_level; - smu_dpm_ctx->enable_umd_pstate = false; amdgpu_asic_update_umd_stable_pstate(smu->adev, false); smu_deep_sleep_control(smu, true); smu_gfx_ulv_control(smu, true); - amdgpu_device_ip_set_clockgating_state(smu->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_CG_STATE_GATE); - amdgpu_device_ip_set_powergating_state(smu->adev, - AMD_IP_BLOCK_TYPE_GFX, - AMD_PG_STATE_GATE); smu_gpo_control(smu, true); } } @@ -1778,22 +1682,18 @@ static int smu_adjust_power_state_dynamic(struct smu_context *smu, static int smu_handle_task(struct smu_context *smu, enum amd_dpm_forced_level level, - enum amd_pp_task task_id, - bool lock_needed) + enum amd_pp_task task_id) { int ret = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - if (lock_needed) - mutex_lock(&smu->mutex); - switch (task_id) { case AMD_PP_TASK_DISPLAY_CONFIG_CHANGE: ret = smu_pre_display_config_changed(smu); if (ret) - goto out; + return ret; ret = smu_adjust_power_state_dynamic(smu, level, false); break; case AMD_PP_TASK_COMPLETE_INIT: @@ -1804,10 +1704,6 @@ static int smu_handle_task(struct smu_context *smu, break; } -out: - if (lock_needed) - mutex_unlock(&smu->mutex); - return ret; } @@ -1818,7 +1714,7 @@ static int smu_handle_dpm_task(void *handle, struct smu_context *smu = handle; struct smu_dpm_context *smu_dpm = &smu->smu_dpm; - return smu_handle_task(smu, smu_dpm->dpm_level, task_id, true); + return smu_handle_task(smu, smu_dpm->dpm_level, task_id); } @@ -1837,8 +1733,6 @@ static int smu_switch_power_profile(void *handle, if (!(type < PP_SMC_POWER_PROFILE_CUSTOM)) return -EINVAL; - mutex_lock(&smu->mutex); - if (!en) { smu->workload_mask &= ~(1 << smu->workload_prority[type]); index = fls(smu->workload_mask); @@ -1855,8 +1749,6 @@ static int smu_switch_power_profile(void *handle, smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_PERF_DETERMINISM) smu_bump_power_profile_mode(smu, &workload, 0); - mutex_unlock(&smu->mutex); - return 0; } @@ -1864,7 +1756,6 @@ static enum amd_dpm_forced_level smu_get_performance_level(void *handle) { struct smu_context *smu = handle; struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); - enum amd_dpm_forced_level level; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; @@ -1872,11 +1763,7 @@ static enum amd_dpm_forced_level smu_get_performance_level(void *handle) if (!smu->is_apu && !smu_dpm_ctx->dpm_context) return -EINVAL; - mutex_lock(&(smu->mutex)); - level = smu_dpm_ctx->dpm_level; - mutex_unlock(&(smu->mutex)); - - return level; + return smu_dpm_ctx->dpm_level; } static int smu_force_performance_level(void *handle, @@ -1892,19 +1779,12 @@ static int smu_force_performance_level(void *handle, if (!smu->is_apu && !smu_dpm_ctx->dpm_context) return -EINVAL; - mutex_lock(&smu->mutex); - ret = smu_enable_umd_pstate(smu, &level); - if (ret) { - mutex_unlock(&smu->mutex); + if (ret) return ret; - } ret = smu_handle_task(smu, level, - AMD_PP_TASK_READJUST_POWER_STATE, - false); - - mutex_unlock(&smu->mutex); + AMD_PP_TASK_READJUST_POWER_STATE); /* reset user dpm clock state */ if (!ret && smu_dpm_ctx->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL) { @@ -1918,16 +1798,11 @@ static int smu_force_performance_level(void *handle, static int smu_set_display_count(void *handle, uint32_t count) { struct smu_context *smu = handle; - int ret = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - ret = smu_init_display_count(smu, count); - mutex_unlock(&smu->mutex); - - return ret; + return smu_init_display_count(smu, count); } static int smu_force_smuclk_levels(struct smu_context *smu, @@ -1945,8 +1820,6 @@ static int smu_force_smuclk_levels(struct smu_context *smu, return -EINVAL; } - mutex_lock(&smu->mutex); - if (smu->ppt_funcs && smu->ppt_funcs->force_clk_levels) { ret = smu->ppt_funcs->force_clk_levels(smu, clk_type, mask); if (!ret && !(smu->user_dpm_profile.flags & SMU_DPM_USER_PROFILE_RESTORE)) { @@ -1955,8 +1828,6 @@ static int smu_force_smuclk_levels(struct smu_context *smu, } } - mutex_unlock(&smu->mutex); - return ret; } @@ -2015,14 +1886,10 @@ static int smu_set_mp1_state(void *handle, if (!smu->pm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs && smu->ppt_funcs->set_mp1_state) ret = smu->ppt_funcs->set_mp1_state(smu, mp1_state); - mutex_unlock(&smu->mutex); - return ret; } @@ -2038,14 +1905,10 @@ static int smu_set_df_cstate(void *handle, if (!smu->ppt_funcs || !smu->ppt_funcs->set_df_cstate) return 0; - mutex_lock(&smu->mutex); - ret = smu->ppt_funcs->set_df_cstate(smu, state); if (ret) dev_err(smu->adev->dev, "[SetDfCstate] failed!\n"); - mutex_unlock(&smu->mutex); - return ret; } @@ -2059,38 +1922,25 @@ int smu_allow_xgmi_power_down(struct smu_context *smu, bool en) if (!smu->ppt_funcs || !smu->ppt_funcs->allow_xgmi_power_down) return 0; - mutex_lock(&smu->mutex); - ret = smu->ppt_funcs->allow_xgmi_power_down(smu, en); if (ret) dev_err(smu->adev->dev, "[AllowXgmiPowerDown] failed!\n"); - mutex_unlock(&smu->mutex); - return ret; } int smu_write_watermarks_table(struct smu_context *smu) { - int ret = 0; - if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - - ret = smu_set_watermarks_table(smu, NULL); - - mutex_unlock(&smu->mutex); - - return ret; + return smu_set_watermarks_table(smu, NULL); } static int smu_set_watermarks_for_clock_ranges(void *handle, struct pp_smu_wm_range_sets *clock_ranges) { struct smu_context *smu = handle; - int ret = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; @@ -2098,13 +1948,7 @@ static int smu_set_watermarks_for_clock_ranges(void *handle, if (smu->disable_watermark) return 0; - mutex_lock(&smu->mutex); - - ret = smu_set_watermarks_table(smu, clock_ranges); - - mutex_unlock(&smu->mutex); - - return ret; + return smu_set_watermarks_table(smu, clock_ranges); } int smu_set_ac_dc(struct smu_context *smu) @@ -2118,14 +1962,12 @@ int smu_set_ac_dc(struct smu_context *smu) if (smu->dc_controlled_by_gpio) return 0; - mutex_lock(&smu->mutex); ret = smu_set_power_source(smu, smu->adev->pm.ac_power ? SMU_POWER_SOURCE_AC : SMU_POWER_SOURCE_DC); if (ret) dev_err(smu->adev->dev, "Failed to switch to %s mode!\n", smu->adev->pm.ac_power ? "AC" : "DC"); - mutex_unlock(&smu->mutex); return ret; } @@ -2138,6 +1980,7 @@ const struct amd_ip_funcs smu_ip_funcs = { .sw_fini = smu_sw_fini, .hw_init = smu_hw_init, .hw_fini = smu_hw_fini, + .late_fini = smu_late_fini, .suspend = smu_suspend, .resume = smu_resume, .is_idle = NULL, @@ -2146,7 +1989,6 @@ const struct amd_ip_funcs smu_ip_funcs = { .soft_reset = NULL, .set_clockgating_state = smu_set_clockgating_state, .set_powergating_state = smu_set_powergating_state, - .enable_umd_pstate = smu_enable_umd_pstate, }; const struct amdgpu_ip_block_version smu_v11_0_ip_block = @@ -2212,13 +2054,9 @@ static int smu_set_gfx_cgpg(struct smu_context *smu, bool enabled) { int ret = 0; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->set_gfx_cgpg) ret = smu->ppt_funcs->set_gfx_cgpg(smu, enabled); - mutex_unlock(&smu->mutex); - return ret; } @@ -2230,21 +2068,21 @@ static int smu_set_fan_speed_rpm(void *handle, uint32_t speed) if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); + if (!smu->ppt_funcs->set_fan_speed_rpm) + return -EOPNOTSUPP; - if (smu->ppt_funcs->set_fan_speed_rpm) { - ret = smu->ppt_funcs->set_fan_speed_rpm(smu, speed); - if (!ret && !(smu->user_dpm_profile.flags & SMU_DPM_USER_PROFILE_RESTORE)) { - smu->user_dpm_profile.flags |= SMU_CUSTOM_FAN_SPEED_RPM; - smu->user_dpm_profile.fan_speed_rpm = speed; + if (speed == U32_MAX) + return -EINVAL; - /* Override custom PWM setting as they cannot co-exist */ - smu->user_dpm_profile.flags &= ~SMU_CUSTOM_FAN_SPEED_PWM; - smu->user_dpm_profile.fan_speed_pwm = 0; - } - } + ret = smu->ppt_funcs->set_fan_speed_rpm(smu, speed); + if (!ret && !(smu->user_dpm_profile.flags & SMU_DPM_USER_PROFILE_RESTORE)) { + smu->user_dpm_profile.flags |= SMU_CUSTOM_FAN_SPEED_RPM; + smu->user_dpm_profile.fan_speed_rpm = speed; - mutex_unlock(&smu->mutex); + /* Override custom PWM setting as they cannot co-exist */ + smu->user_dpm_profile.flags &= ~SMU_CUSTOM_FAN_SPEED_PWM; + smu->user_dpm_profile.fan_speed_pwm = 0; + } return ret; } @@ -2301,8 +2139,6 @@ int smu_get_power_limit(void *handle, break; } - mutex_lock(&smu->mutex); - if (limit_type != SMU_DEFAULT_PPT_LIMIT) { if (smu->ppt_funcs->get_ppt_limit) ret = smu->ppt_funcs->get_ppt_limit(smu, limit, limit_type, limit_level); @@ -2336,8 +2172,6 @@ int smu_get_power_limit(void *handle, } } - mutex_unlock(&smu->mutex); - return ret; } @@ -2350,21 +2184,16 @@ static int smu_set_power_limit(void *handle, uint32_t limit) if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - limit &= (1<<24)-1; if (limit_type != SMU_DEFAULT_PPT_LIMIT) - if (smu->ppt_funcs->set_power_limit) { - ret = smu->ppt_funcs->set_power_limit(smu, limit_type, limit); - goto out; - } + if (smu->ppt_funcs->set_power_limit) + return smu->ppt_funcs->set_power_limit(smu, limit_type, limit); if (limit > smu->max_power_limit) { dev_err(smu->adev->dev, "New power limit (%d) is over the max allowed %d\n", limit, smu->max_power_limit); - ret = -EINVAL; - goto out; + return -EINVAL; } if (!limit) @@ -2376,9 +2205,6 @@ static int smu_set_power_limit(void *handle, uint32_t limit) smu->user_dpm_profile.power_limit = limit; } -out: - mutex_unlock(&smu->mutex); - return ret; } @@ -2389,21 +2215,14 @@ static int smu_print_smuclk_levels(struct smu_context *smu, enum smu_clk_type cl if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->print_clk_levels) ret = smu->ppt_funcs->print_clk_levels(smu, clk_type, buf); - mutex_unlock(&smu->mutex); - return ret; } -static int smu_print_ppclk_levels(void *handle, - enum pp_clock_type type, - char *buf) +static enum smu_clk_type smu_convert_to_smuclk(enum pp_clock_type type) { - struct smu_context *smu = handle; enum smu_clk_type clk_type; switch (type) { @@ -2436,12 +2255,45 @@ static int smu_print_ppclk_levels(void *handle, case OD_CCLK: clk_type = SMU_OD_CCLK; break; default: - return -EINVAL; + clk_type = SMU_CLK_COUNT; break; } + return clk_type; +} + +static int smu_print_ppclk_levels(void *handle, + enum pp_clock_type type, + char *buf) +{ + struct smu_context *smu = handle; + enum smu_clk_type clk_type; + + clk_type = smu_convert_to_smuclk(type); + if (clk_type == SMU_CLK_COUNT) + return -EINVAL; + return smu_print_smuclk_levels(smu, clk_type, buf); } +static int smu_emit_ppclk_levels(void *handle, enum pp_clock_type type, char *buf, int *offset) +{ + struct smu_context *smu = handle; + enum smu_clk_type clk_type; + + clk_type = smu_convert_to_smuclk(type); + if (clk_type == SMU_CLK_COUNT) + return -EINVAL; + + if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) + return -EOPNOTSUPP; + + if (!smu->ppt_funcs->emit_clk_levels) + return -ENOENT; + + return smu->ppt_funcs->emit_clk_levels(smu, clk_type, buf, offset); + +} + static int smu_od_edit_dpm_table(void *handle, enum PP_OD_DPM_TABLE_COMMAND type, long *input, uint32_t size) @@ -2452,14 +2304,10 @@ static int smu_od_edit_dpm_table(void *handle, if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->od_edit_dpm_table) { ret = smu->ppt_funcs->od_edit_dpm_table(smu, type, input, size); } - mutex_unlock(&smu->mutex); - return ret; } @@ -2483,8 +2331,6 @@ static int smu_read_sensor(void *handle, size_val = *size_arg; size = &size_val; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->read_sensor) if (!smu->ppt_funcs->read_sensor(smu, sensor, data, size)) goto unlock; @@ -2499,7 +2345,7 @@ static int smu_read_sensor(void *handle, *size = 4; break; case AMDGPU_PP_SENSOR_ENABLED_SMC_FEATURES_MASK: - ret = smu_feature_get_enabled_mask(smu, (uint32_t *)data, 2); + ret = smu_feature_get_enabled_mask(smu, (uint64_t *)data); *size = 8; break; case AMDGPU_PP_SENSOR_UVD_POWER: @@ -2525,8 +2371,6 @@ static int smu_read_sensor(void *handle, } unlock: - mutex_unlock(&smu->mutex); - // assign uint32_t to int *size_arg = size_val; @@ -2536,7 +2380,6 @@ unlock: static int smu_get_power_profile_mode(void *handle, char *buf) { struct smu_context *smu = handle; - int ret = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled || !smu->ppt_funcs->get_power_profile_mode) @@ -2544,13 +2387,7 @@ static int smu_get_power_profile_mode(void *handle, char *buf) if (!buf) return -EINVAL; - mutex_lock(&smu->mutex); - - ret = smu->ppt_funcs->get_power_profile_mode(smu, buf); - - mutex_unlock(&smu->mutex); - - return ret; + return smu->ppt_funcs->get_power_profile_mode(smu, buf); } static int smu_set_power_profile_mode(void *handle, @@ -2558,76 +2395,66 @@ static int smu_set_power_profile_mode(void *handle, uint32_t param_size) { struct smu_context *smu = handle; - int ret = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled || !smu->ppt_funcs->set_power_profile_mode) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - - smu_bump_power_profile_mode(smu, param, param_size); - - mutex_unlock(&smu->mutex); - - return ret; + return smu_bump_power_profile_mode(smu, param, param_size); } -static u32 smu_get_fan_control_mode(void *handle) +static int smu_get_fan_control_mode(void *handle, u32 *fan_mode) { struct smu_context *smu = handle; - u32 ret = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) - return AMD_FAN_CTRL_NONE; + return -EOPNOTSUPP; - mutex_lock(&smu->mutex); + if (!smu->ppt_funcs->get_fan_control_mode) + return -EOPNOTSUPP; - if (smu->ppt_funcs->get_fan_control_mode) - ret = smu->ppt_funcs->get_fan_control_mode(smu); + if (!fan_mode) + return -EINVAL; - mutex_unlock(&smu->mutex); + *fan_mode = smu->ppt_funcs->get_fan_control_mode(smu); - return ret; + return 0; } -static int smu_set_fan_control_mode(struct smu_context *smu, int value) +static int smu_set_fan_control_mode(void *handle, u32 value) { + struct smu_context *smu = handle; int ret = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) - return -EOPNOTSUPP; + return -EOPNOTSUPP; - mutex_lock(&smu->mutex); + if (!smu->ppt_funcs->set_fan_control_mode) + return -EOPNOTSUPP; - if (smu->ppt_funcs->set_fan_control_mode) { - ret = smu->ppt_funcs->set_fan_control_mode(smu, value); - if (!ret && !(smu->user_dpm_profile.flags & SMU_DPM_USER_PROFILE_RESTORE)) - smu->user_dpm_profile.fan_mode = value; - } + if (value == U32_MAX) + return -EINVAL; - mutex_unlock(&smu->mutex); + ret = smu->ppt_funcs->set_fan_control_mode(smu, value); + if (ret) + goto out; - /* reset user dpm fan speed */ - if (!ret && value != AMD_FAN_CTRL_MANUAL && - !(smu->user_dpm_profile.flags & SMU_DPM_USER_PROFILE_RESTORE)) { - smu->user_dpm_profile.fan_speed_pwm = 0; - smu->user_dpm_profile.fan_speed_rpm = 0; - smu->user_dpm_profile.flags &= ~(SMU_CUSTOM_FAN_SPEED_RPM | SMU_CUSTOM_FAN_SPEED_PWM); + if (!(smu->user_dpm_profile.flags & SMU_DPM_USER_PROFILE_RESTORE)) { + smu->user_dpm_profile.fan_mode = value; + + /* reset user dpm fan speed */ + if (value != AMD_FAN_CTRL_MANUAL) { + smu->user_dpm_profile.fan_speed_pwm = 0; + smu->user_dpm_profile.fan_speed_rpm = 0; + smu->user_dpm_profile.flags &= ~(SMU_CUSTOM_FAN_SPEED_RPM | SMU_CUSTOM_FAN_SPEED_PWM); + } } +out: return ret; } -static void smu_pp_set_fan_control_mode(void *handle, u32 value) -{ - struct smu_context *smu = handle; - - smu_set_fan_control_mode(smu, value); -} - - static int smu_get_fan_speed_pwm(void *handle, u32 *speed) { struct smu_context *smu = handle; @@ -2636,12 +2463,13 @@ static int smu_get_fan_speed_pwm(void *handle, u32 *speed) if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); + if (!smu->ppt_funcs->get_fan_speed_pwm) + return -EOPNOTSUPP; - if (smu->ppt_funcs->get_fan_speed_pwm) - ret = smu->ppt_funcs->get_fan_speed_pwm(smu, speed); + if (!speed) + return -EINVAL; - mutex_unlock(&smu->mutex); + ret = smu->ppt_funcs->get_fan_speed_pwm(smu, speed); return ret; } @@ -2654,21 +2482,21 @@ static int smu_set_fan_speed_pwm(void *handle, u32 speed) if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); + if (!smu->ppt_funcs->set_fan_speed_pwm) + return -EOPNOTSUPP; - if (smu->ppt_funcs->set_fan_speed_pwm) { - ret = smu->ppt_funcs->set_fan_speed_pwm(smu, speed); - if (!ret && !(smu->user_dpm_profile.flags & SMU_DPM_USER_PROFILE_RESTORE)) { - smu->user_dpm_profile.flags |= SMU_CUSTOM_FAN_SPEED_PWM; - smu->user_dpm_profile.fan_speed_pwm = speed; + if (speed == U32_MAX) + return -EINVAL; - /* Override custom RPM setting as they cannot co-exist */ - smu->user_dpm_profile.flags &= ~SMU_CUSTOM_FAN_SPEED_RPM; - smu->user_dpm_profile.fan_speed_rpm = 0; - } - } + ret = smu->ppt_funcs->set_fan_speed_pwm(smu, speed); + if (!ret && !(smu->user_dpm_profile.flags & SMU_DPM_USER_PROFILE_RESTORE)) { + smu->user_dpm_profile.flags |= SMU_CUSTOM_FAN_SPEED_PWM; + smu->user_dpm_profile.fan_speed_pwm = speed; - mutex_unlock(&smu->mutex); + /* Override custom RPM setting as they cannot co-exist */ + smu->user_dpm_profile.flags &= ~SMU_CUSTOM_FAN_SPEED_RPM; + smu->user_dpm_profile.fan_speed_rpm = 0; + } return ret; } @@ -2681,12 +2509,13 @@ static int smu_get_fan_speed_rpm(void *handle, uint32_t *speed) if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); + if (!smu->ppt_funcs->get_fan_speed_rpm) + return -EOPNOTSUPP; - if (smu->ppt_funcs->get_fan_speed_rpm) - ret = smu->ppt_funcs->get_fan_speed_rpm(smu, speed); + if (!speed) + return -EINVAL; - mutex_unlock(&smu->mutex); + ret = smu->ppt_funcs->get_fan_speed_rpm(smu, speed); return ret; } @@ -2694,18 +2523,11 @@ static int smu_get_fan_speed_rpm(void *handle, uint32_t *speed) static int smu_set_deep_sleep_dcefclk(void *handle, uint32_t clk) { struct smu_context *smu = handle; - int ret = 0; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - - ret = smu_set_min_dcef_deep_sleep(smu, clk); - - mutex_unlock(&smu->mutex); - - return ret; + return smu_set_min_dcef_deep_sleep(smu, clk); } static int smu_get_clock_by_type_with_latency(void *handle, @@ -2719,8 +2541,6 @@ static int smu_get_clock_by_type_with_latency(void *handle, if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->get_clock_by_type_with_latency) { switch (type) { case amd_pp_sys_clock: @@ -2737,15 +2557,12 @@ static int smu_get_clock_by_type_with_latency(void *handle, break; default: dev_err(smu->adev->dev, "Invalid clock type!\n"); - mutex_unlock(&smu->mutex); return -EINVAL; } ret = smu->ppt_funcs->get_clock_by_type_with_latency(smu, clk_type, clocks); } - mutex_unlock(&smu->mutex); - return ret; } @@ -2758,13 +2575,9 @@ static int smu_display_clock_voltage_request(void *handle, if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->display_clock_voltage_request) ret = smu->ppt_funcs->display_clock_voltage_request(smu, clock_req); - mutex_unlock(&smu->mutex); - return ret; } @@ -2778,13 +2591,9 @@ static int smu_display_disable_memory_clock_switch(void *handle, if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->display_disable_memory_clock_switch) ret = smu->ppt_funcs->display_disable_memory_clock_switch(smu, disable_memory_clock_switch); - mutex_unlock(&smu->mutex); - return ret; } @@ -2797,13 +2606,9 @@ static int smu_set_xgmi_pstate(void *handle, if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->set_xgmi_pstate) ret = smu->ppt_funcs->set_xgmi_pstate(smu, pstate); - mutex_unlock(&smu->mutex); - if(ret) dev_err(smu->adev->dev, "Failed to set XGMI pstate!\n"); @@ -2813,21 +2618,16 @@ static int smu_set_xgmi_pstate(void *handle, static int smu_get_baco_capability(void *handle, bool *cap) { struct smu_context *smu = handle; - int ret = 0; *cap = false; if (!smu->pm_enabled) return 0; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs && smu->ppt_funcs->baco_is_support) *cap = smu->ppt_funcs->baco_is_support(smu); - mutex_unlock(&smu->mutex); - - return ret; + return 0; } static int smu_baco_set_state(void *handle, int state) @@ -2839,20 +2639,11 @@ static int smu_baco_set_state(void *handle, int state) return -EOPNOTSUPP; if (state == 0) { - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->baco_exit) ret = smu->ppt_funcs->baco_exit(smu); - - mutex_unlock(&smu->mutex); } else if (state == 1) { - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->baco_enter) ret = smu->ppt_funcs->baco_enter(smu); - - mutex_unlock(&smu->mutex); - } else { return -EINVAL; } @@ -2871,13 +2662,9 @@ bool smu_mode1_reset_is_support(struct smu_context *smu) if (!smu->pm_enabled) return false; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs && smu->ppt_funcs->mode1_reset_is_support) ret = smu->ppt_funcs->mode1_reset_is_support(smu); - mutex_unlock(&smu->mutex); - return ret; } @@ -2888,13 +2675,9 @@ bool smu_mode2_reset_is_support(struct smu_context *smu) if (!smu->pm_enabled) return false; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs && smu->ppt_funcs->mode2_reset_is_support) ret = smu->ppt_funcs->mode2_reset_is_support(smu); - mutex_unlock(&smu->mutex); - return ret; } @@ -2905,13 +2688,9 @@ int smu_mode1_reset(struct smu_context *smu) if (!smu->pm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->mode1_reset) ret = smu->ppt_funcs->mode1_reset(smu); - mutex_unlock(&smu->mutex); - return ret; } @@ -2923,13 +2702,9 @@ static int smu_mode2_reset(void *handle) if (!smu->pm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->mode2_reset) ret = smu->ppt_funcs->mode2_reset(smu); - mutex_unlock(&smu->mutex); - if (ret) dev_err(smu->adev->dev, "Mode2 reset failed!\n"); @@ -2945,13 +2720,9 @@ static int smu_get_max_sustainable_clocks_by_dc(void *handle, if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->get_max_sustainable_clocks_by_dc) ret = smu->ppt_funcs->get_max_sustainable_clocks_by_dc(smu, max_clocks); - mutex_unlock(&smu->mutex); - return ret; } @@ -2965,13 +2736,9 @@ static int smu_get_uclk_dpm_states(void *handle, if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->get_uclk_dpm_states) ret = smu->ppt_funcs->get_uclk_dpm_states(smu, clock_values_in_khz, num_states); - mutex_unlock(&smu->mutex); - return ret; } @@ -2983,13 +2750,9 @@ static enum amd_pm_state_type smu_get_current_power_state(void *handle) if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->get_current_power_state) pm_state = smu->ppt_funcs->get_current_power_state(smu); - mutex_unlock(&smu->mutex); - return pm_state; } @@ -3002,20 +2765,15 @@ static int smu_get_dpm_clock_table(void *handle, if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->get_dpm_clock_table) ret = smu->ppt_funcs->get_dpm_clock_table(smu, clock_table); - mutex_unlock(&smu->mutex); - return ret; } static ssize_t smu_sys_get_gpu_metrics(void *handle, void **table) { struct smu_context *smu = handle; - ssize_t size; if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; @@ -3023,13 +2781,7 @@ static ssize_t smu_sys_get_gpu_metrics(void *handle, void **table) if (!smu->ppt_funcs->get_gpu_metrics) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - - size = smu->ppt_funcs->get_gpu_metrics(smu, table); - - mutex_unlock(&smu->mutex); - - return size; + return smu->ppt_funcs->get_gpu_metrics(smu, table); } static int smu_enable_mgpu_fan_boost(void *handle) @@ -3040,13 +2792,9 @@ static int smu_enable_mgpu_fan_boost(void *handle) if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - mutex_lock(&smu->mutex); - if (smu->ppt_funcs->enable_mgpu_fan_boost) ret = smu->ppt_funcs->enable_mgpu_fan_boost(smu); - mutex_unlock(&smu->mutex); - return ret; } @@ -3056,10 +2804,8 @@ static int smu_gfx_state_change_set(void *handle, struct smu_context *smu = handle; int ret = 0; - mutex_lock(&smu->mutex); if (smu->ppt_funcs->gfx_state_change_set) ret = smu->ppt_funcs->gfx_state_change_set(smu, state); - mutex_unlock(&smu->mutex); return ret; } @@ -3068,10 +2814,8 @@ int smu_handle_passthrough_sbr(struct smu_context *smu, bool enable) { int ret = 0; - mutex_lock(&smu->mutex); if (smu->ppt_funcs->smu_handle_passthrough_sbr) ret = smu->ppt_funcs->smu_handle_passthrough_sbr(smu, enable); - mutex_unlock(&smu->mutex); return ret; } @@ -3080,11 +2824,9 @@ int smu_get_ecc_info(struct smu_context *smu, void *umc_ecc) { int ret = -EOPNOTSUPP; - mutex_lock(&smu->mutex); if (smu->ppt_funcs && smu->ppt_funcs->get_ecc_info) ret = smu->ppt_funcs->get_ecc_info(smu, umc_ecc); - mutex_unlock(&smu->mutex); return ret; @@ -3101,24 +2843,23 @@ static int smu_get_prv_buffer_details(void *handle, void **addr, size_t *size) *addr = NULL; *size = 0; - mutex_lock(&smu->mutex); if (memory_pool->bo) { *addr = memory_pool->cpu_addr; *size = memory_pool->size; } - mutex_unlock(&smu->mutex); return 0; } static const struct amd_pm_funcs swsmu_pm_funcs = { /* export for sysfs */ - .set_fan_control_mode = smu_pp_set_fan_control_mode, + .set_fan_control_mode = smu_set_fan_control_mode, .get_fan_control_mode = smu_get_fan_control_mode, .set_fan_speed_pwm = smu_set_fan_speed_pwm, .get_fan_speed_pwm = smu_get_fan_speed_pwm, .force_clock_level = smu_force_ppclk_levels, .print_clock_levels = smu_print_ppclk_levels, + .emit_clock_levels = smu_emit_ppclk_levels, .force_performance_level = smu_force_performance_level, .read_sensor = smu_read_sensor, .get_performance_level = smu_get_performance_level, @@ -3165,17 +2906,13 @@ static const struct amd_pm_funcs swsmu_pm_funcs = { .get_smu_prv_buf_details = smu_get_prv_buffer_details, }; -int smu_wait_for_event(struct amdgpu_device *adev, enum smu_event_type event, +int smu_wait_for_event(struct smu_context *smu, enum smu_event_type event, uint64_t event_arg) { int ret = -EINVAL; - struct smu_context *smu = &adev->smu; - if (smu->ppt_funcs->wait_for_event) { - mutex_lock(&smu->mutex); + if (smu->ppt_funcs->wait_for_event) ret = smu->ppt_funcs->wait_for_event(smu, event, event_arg); - mutex_unlock(&smu->mutex); - } return ret; } @@ -3203,7 +2940,7 @@ int smu_stb_collect_info(struct smu_context *smu, void *buf, uint32_t size) static int smu_stb_debugfs_open(struct inode *inode, struct file *filp) { struct amdgpu_device *adev = filp->f_inode->i_private; - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; unsigned char *buf; int r; @@ -3228,7 +2965,7 @@ static ssize_t smu_stb_debugfs_read(struct file *filp, char __user *buf, size_t loff_t *pos) { struct amdgpu_device *adev = filp->f_inode->i_private; - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; if (!filp->private_data) @@ -3269,7 +3006,7 @@ void amdgpu_smu_stb_debug_fs_init(struct amdgpu_device *adev) { #if defined(CONFIG_DEBUG_FS) - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; if (!smu->stb_context.stb_buf_size) return; @@ -3281,5 +3018,14 @@ void amdgpu_smu_stb_debug_fs_init(struct amdgpu_device *adev) &smu_stb_debugfs_fops, smu->stb_context.stb_buf_size); #endif +} + +int smu_send_hbm_bad_pages_num(struct smu_context *smu, uint32_t size) +{ + int ret = 0; + if (smu->ppt_funcs && smu->ppt_funcs->send_hbm_bad_pages_num) + ret = smu->ppt_funcs->send_hbm_bad_pages_num(smu, size); + + return ret; } diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h index ba7565bc8104..17594ceb507e 100644 --- a/drivers/gpu/drm/amd/pm/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h @@ -241,11 +241,6 @@ struct smu_user_dpm_profile { uint32_t clk_dependency; }; -enum smu_event_type { - - SMU_EVENT_RESET_COMPLETE = 0, -}; - #define SMU_TABLE_INIT(tables, table_id, s, a, d) \ do { \ tables[table_id].size = s; \ @@ -342,6 +337,7 @@ struct smu_table_context struct smu_bios_boot_up_values boot_values; void *driver_pptable; void *ecc_table; + void *driver_smu_config_table; struct smu_table tables[SMU_TABLE_COUNT]; /* * The driver table is just a staging buffer for @@ -368,7 +364,6 @@ struct smu_dpm_context { uint32_t dpm_context_size; void *dpm_context; void *golden_dpm_context; - bool enable_umd_pstate; enum amd_dpm_forced_level dpm_level; enum amd_dpm_forced_level saved_dpm_level; enum amd_dpm_forced_level requested_dpm_level; @@ -382,8 +377,6 @@ struct smu_power_gate { bool vce_gated; atomic_t vcn_gated; atomic_t jpeg_gated; - struct mutex vcn_gate_lock; - struct mutex jpeg_gate_lock; }; struct smu_power_context { @@ -398,8 +391,6 @@ struct smu_feature uint32_t feature_num; DECLARE_BITMAP(supported, SMU_FEATURE_MAX); DECLARE_BITMAP(allowed, SMU_FEATURE_MAX); - DECLARE_BITMAP(enabled, SMU_FEATURE_MAX); - struct mutex mutex; }; struct smu_clocks { @@ -436,7 +427,6 @@ enum smu_baco_state struct smu_baco_context { - struct mutex mutex; uint32_t state; bool platform_support; }; @@ -494,9 +484,6 @@ struct smu_context const struct cmn2asic_mapping *table_map; const struct cmn2asic_mapping *pwr_src_map; const struct cmn2asic_mapping *workload_map; - struct mutex mutex; - struct mutex sensor_lock; - struct mutex metrics_lock; struct mutex message_lock; uint64_t pool_size; @@ -618,10 +605,24 @@ struct pptable_funcs { * to buffer. Star current level. * * Used for sysfs interfaces. + * Return: Number of characters written to the buffer */ int (*print_clk_levels)(struct smu_context *smu, enum smu_clk_type clk_type, char *buf); /** + * @emit_clk_levels: Print DPM clock levels for a clock domain + * to buffer using sysfs_emit_at. Star current level. + * + * Used for sysfs interfaces. + * &buf: sysfs buffer + * &offset: offset within buffer to start printing, which is updated by the + * function. + * + * Return: 0 on Success or Negative to indicate an error occurred. + */ + int (*emit_clk_levels)(struct smu_context *smu, enum smu_clk_type clk_type, char *buf, int *offset); + + /** * @force_clk_levels: Set a range of allowed DPM levels for a clock * domain. * &clk_type: Clock domain. @@ -829,12 +830,12 @@ struct pptable_funcs { * other devices. The i2c's EEPROM also stores bad page tables on boards * with ECC. */ - int (*i2c_init)(struct smu_context *smu, struct i2c_adapter *control); + int (*i2c_init)(struct smu_context *smu); /** * @i2c_fini: Tear down i2c. */ - void (*i2c_fini)(struct smu_context *smu, struct i2c_adapter *control); + void (*i2c_fini)(struct smu_context *smu); /** * @get_unique_id: Get the GPU's unique id. Used for asset tracking. @@ -988,10 +989,9 @@ struct pptable_funcs { /** * @get_enabled_mask: Get a mask of features that are currently enabled * on the SMU. - * &feature_mask: Array representing enabled feature mask. - * &num: Elements in &feature_mask. + * &feature_mask: Enabled feature mask. */ - int (*get_enabled_mask)(struct smu_context *smu, uint32_t *feature_mask, uint32_t num); + int (*get_enabled_mask)(struct smu_context *smu, uint64_t *feature_mask); /** * @feature_is_enabled: Test if a feature is enabled. @@ -1005,7 +1005,6 @@ struct pptable_funcs { * exception to those in &mask. */ int (*disable_all_features_with_exception)(struct smu_context *smu, - bool no_hw_disablement, enum smu_feature_mask mask); /** @@ -1395,10 +1394,6 @@ int smu_mode1_reset(struct smu_context *smu); extern const struct amd_ip_funcs smu_ip_funcs; -extern const struct amdgpu_ip_block_version smu_v11_0_ip_block; -extern const struct amdgpu_ip_block_version smu_v12_0_ip_block; -extern const struct amdgpu_ip_block_version smu_v13_0_ip_block; - bool is_support_sw_smu(struct amdgpu_device *adev); bool is_support_cclk_dpm(struct amdgpu_device *adev); int smu_write_watermarks_table(struct smu_context *smu); @@ -1413,15 +1408,15 @@ int smu_set_ac_dc(struct smu_context *smu); int smu_allow_xgmi_power_down(struct smu_context *smu, bool en); -int smu_get_status_gfxoff(struct amdgpu_device *adev, uint32_t *value); +int smu_get_status_gfxoff(struct smu_context *smu, uint32_t *value); int smu_handle_passthrough_sbr(struct smu_context *smu, bool enable); -int smu_wait_for_event(struct amdgpu_device *adev, enum smu_event_type event, +int smu_wait_for_event(struct smu_context *smu, enum smu_event_type event, uint64_t event_arg); int smu_get_ecc_info(struct smu_context *smu, void *umc_ecc); int smu_stb_collect_info(struct smu_context *smu, void *buff, uint32_t size); void amdgpu_smu_stb_debug_fs_init(struct amdgpu_device *adev); - +int smu_send_hbm_bad_pages_num(struct smu_context *smu, uint32_t size); #endif #endif diff --git a/drivers/gpu/drm/amd/pm/inc/aldebaran_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/aldebaran_ppsmc.h index ab66a4b9e438..ab66a4b9e438 100644 --- a/drivers/gpu/drm/amd/pm/inc/aldebaran_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/aldebaran_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/arcturus_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/arcturus_ppsmc.h index 45f5d29bc705..45f5d29bc705 100644 --- a/drivers/gpu/drm/amd/pm/inc/arcturus_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/arcturus_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_arcturus.h index 43d43d6addc0..43d43d6addc0 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_arcturus.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_cyan_skillfish.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_cyan_skillfish.h index 4884a4e1f261..4884a4e1f261 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_cyan_skillfish.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_cyan_skillfish.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_navi10.h index 04752ade1016..04752ade1016 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_navi10.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_sienna_cichlid.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_sienna_cichlid.h index 63b8701fd466..b253be602cc2 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_sienna_cichlid.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_sienna_cichlid.h @@ -27,7 +27,7 @@ // *** IMPORTANT *** // SMU TEAM: Always increment the interface version if // any structure is changed in this file -#define SMU11_DRIVER_IF_VERSION 0x3B +#define SMU11_DRIVER_IF_VERSION 0x40 #define PPTABLE_Sienna_Cichlid_SMU_VERSION 7 @@ -172,6 +172,7 @@ typedef enum { #define DPM_OVERRIDE_DISABLE_FAST_FCLK_TIMER 0x00001000 #define DPM_OVERRIDE_DISABLE_VCN_PG 0x00002000 #define DPM_OVERRIDE_DISABLE_FMAX_VMAX 0x00004000 +#define DPM_OVERRIDE_ENABLE_eGPU_USB_WA 0x00008000 // VR Mapping Bit Defines #define VR_MAPPING_VR_SELECT_MASK 0x01 @@ -263,7 +264,22 @@ typedef enum { #define LED_DISPLAY_ERROR_BIT 2 //RLC Pace Table total number of levels -#define RLC_PACE_TABLE_NUM_LEVELS 16 +#define RLC_PACE_TABLE_NUM_LEVELS 16 +#define SIENNA_CICHLID_UMC_CHANNEL_NUM 16 + +typedef struct { + uint64_t mca_umc_status; + uint64_t mca_umc_addr; + + uint16_t ce_count_lo_chip; + uint16_t ce_count_hi_chip; + + uint32_t eccPadding; +} EccInfo_t; + +typedef struct { + EccInfo_t EccInfo[SIENNA_CICHLID_UMC_CHANNEL_NUM]; +} EccInfoTable_t; typedef enum { DRAM_BIT_WIDTH_DISABLED = 0, @@ -283,6 +299,7 @@ typedef enum { #define MAX_SW_I2C_COMMANDS 24 + typedef enum { I2C_CONTROLLER_PORT_0 = 0, //CKSVII2C0 I2C_CONTROLLER_PORT_1 = 1, //CKSVII2C1 @@ -1672,7 +1689,8 @@ typedef struct { #define TABLE_OVERDRIVE 8 #define TABLE_I2C_COMMANDS 9 #define TABLE_PACE 10 -#define TABLE_COUNT 11 +#define TABLE_ECCINFO 11 +#define TABLE_COUNT 12 typedef struct { float FlopsPerByteTable[RLC_PACE_TABLE_NUM_LEVELS]; diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_vangogh.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_vangogh.h index 8361ebd8d876..8361ebd8d876 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_vangogh.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu11_driver_if_vangogh.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu12_driver_if.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu12_driver_if.h index e9315eb5b48e..e9315eb5b48e 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu12_driver_if.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu12_driver_if.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h index 0f67c56c2863..0f67c56c2863 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_aldebaran.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_yellow_carp.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h index 25540cb28208..25540cb28208 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_yellow_carp.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_yellow_carp.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v11_0_7_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_0_7_ppsmc.h index d2e10a724560..d2e10a724560 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v11_0_7_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_0_7_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v11_0_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_0_ppsmc.h index 26181b679098..26181b679098 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v11_0_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_0_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v11_5_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_5_pmfw.h index 22edd88b8117..22edd88b8117 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v11_5_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_5_pmfw.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v11_5_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_5_ppsmc.h index fe130a497d6c..fe130a497d6c 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v11_5_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_5_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v11_8_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_8_pmfw.h index bd4fcb6b9610..bd4fcb6b9610 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v11_8_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_8_pmfw.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v11_8_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_8_ppsmc.h index 909a86aa60f3..909a86aa60f3 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v11_8_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v11_8_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v12_0_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v12_0_ppsmc.h index 9ac9f3bd3664..9ac9f3bd3664 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v12_0_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v12_0_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_1_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_1_pmfw.h index c5e26d619bf0..c5e26d619bf0 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_1_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_1_pmfw.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_1_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_1_ppsmc.h index fc9198846e70..fc9198846e70 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_1_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_1_ppsmc.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_11_0_cdr_table.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_11_0_cdr_table.h index beab6d7b28b7..beab6d7b28b7 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_11_0_cdr_table.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_11_0_cdr_table.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h index ff8a0bcbd290..ff8a0bcbd290 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_types.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v11_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h index acb3be292096..acb3be292096 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v11_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v11_0_7_pptable.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0_7_pptable.h index 247c6e9632ba..247c6e9632ba 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v11_0_7_pptable.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0_7_pptable.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v11_0_pptable.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0_pptable.h index 7a63cf8e85ed..7a63cf8e85ed 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v11_0_pptable.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0_pptable.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v12_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v12_0.h index 1ad2dff71090..1ad2dff71090 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v12_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v12_0.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h index 44af23ae059e..44af23ae059e 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v13_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_pptable.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_pptable.h index 1f311396b706..1f311396b706 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_pptable.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_pptable.h diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 505d2fb94fd9..201563072189 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -25,6 +25,7 @@ #include <linux/firmware.h> #include "amdgpu.h" +#include "amdgpu_dpm.h" #include "amdgpu_smu.h" #include "atomfirmware.h" #include "amdgpu_atomfirmware.h" @@ -33,7 +34,6 @@ #include "smu11_driver_if_arcturus.h" #include "soc15_common.h" #include "atom.h" -#include "power_state.h" #include "arcturus_ppt.h" #include "smu_v11_0_pptable.h" #include "arcturus_ppsmc.h" @@ -57,8 +57,6 @@ #undef pr_info #undef pr_debug -#define to_amdgpu_device(x) (container_of(x, struct amdgpu_device, pm.smu_i2c)) - #define ARCTURUS_FEA_MAP(smu_feature, arcturus_feature) \ [smu_feature] = {1, (arcturus_feature)} @@ -603,15 +601,11 @@ static int arcturus_get_smu_metrics_data(struct smu_context *smu, SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -694,8 +688,6 @@ static int arcturus_get_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -1120,7 +1112,6 @@ static int arcturus_read_sensor(struct smu_context *smu, if (!data || !size) return -EINVAL; - mutex_lock(&smu->sensor_lock); switch (sensor) { case AMDGPU_PP_SENSOR_MAX_FAN_RPM: *(uint32_t *)data = pptable->FanMaximumRpm; @@ -1181,7 +1172,6 @@ static int arcturus_read_sensor(struct smu_context *smu, ret = -EOPNOTSUPP; break; } - mutex_unlock(&smu->sensor_lock); return ret; } @@ -2031,15 +2021,12 @@ static void arcturus_dump_pptable(struct smu_context *smu) static bool arcturus_is_dpm_running(struct smu_context *smu) { int ret = 0; - uint32_t feature_mask[2]; uint64_t feature_enabled; - ret = smu_cmn_get_enabled_mask(smu, feature_mask, 2); + ret = smu_cmn_get_enabled_mask(smu, &feature_enabled); if (ret) return false; - feature_enabled = (uint64_t)feature_mask[1] << 32 | feature_mask[0]; - return !!(feature_enabled & SMC_DPM_FEATURE); } @@ -2071,18 +2058,23 @@ static int arcturus_dpm_set_vcn_enable(struct smu_context *smu, bool enable) static int arcturus_i2c_xfer(struct i2c_adapter *i2c_adap, struct i2c_msg *msg, int num_msgs) { - struct amdgpu_device *adev = to_amdgpu_device(i2c_adap); - struct smu_table_context *smu_table = &adev->smu.smu_table; + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(i2c_adap); + struct amdgpu_device *adev = smu_i2c->adev; + struct smu_context *smu = adev->powerplay.pp_handle; + struct smu_table_context *smu_table = &smu->smu_table; struct smu_table *table = &smu_table->driver_table; SwI2cRequest_t *req, *res = (SwI2cRequest_t *)table->cpu_addr; int i, j, r, c; u16 dir; + if (!adev->pm.dpm_enabled) + return -EBUSY; + req = kzalloc(sizeof(*req), GFP_KERNEL); if (!req) return -ENOMEM; - req->I2CcontrollerPort = 0; + req->I2CcontrollerPort = smu_i2c->port; req->I2CSpeed = I2C_SPEED_FAST_400K; req->SlaveAddress = msg[0].addr << 1; /* wants an 8-bit address */ dir = msg[0].flags & I2C_M_RD; @@ -2118,9 +2110,9 @@ static int arcturus_i2c_xfer(struct i2c_adapter *i2c_adap, } } } - mutex_lock(&adev->smu.mutex); - r = smu_cmn_update_table(&adev->smu, SMU_TABLE_I2C_COMMANDS, 0, req, true); - mutex_unlock(&adev->smu.mutex); + mutex_lock(&adev->pm.mutex); + r = smu_cmn_update_table(smu, SMU_TABLE_I2C_COMMANDS, 0, req, true); + mutex_unlock(&adev->pm.mutex); if (r) goto fail; @@ -2161,28 +2153,60 @@ static const struct i2c_adapter_quirks arcturus_i2c_control_quirks = { .max_comb_2nd_msg_len = MAX_SW_I2C_COMMANDS - 2, }; -static int arcturus_i2c_control_init(struct smu_context *smu, struct i2c_adapter *control) +static int arcturus_i2c_control_init(struct smu_context *smu) { - struct amdgpu_device *adev = to_amdgpu_device(control); - int res; + struct amdgpu_device *adev = smu->adev; + int res, i; + + for (i = 0; i < MAX_SMU_I2C_BUSES; i++) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + + smu_i2c->adev = adev; + smu_i2c->port = i; + mutex_init(&smu_i2c->mutex); + control->owner = THIS_MODULE; + control->class = I2C_CLASS_HWMON; + control->dev.parent = &adev->pdev->dev; + control->algo = &arcturus_i2c_algo; + control->quirks = &arcturus_i2c_control_quirks; + snprintf(control->name, sizeof(control->name), "AMDGPU SMU %d", i); + i2c_set_adapdata(control, smu_i2c); + + res = i2c_add_adapter(control); + if (res) { + DRM_ERROR("Failed to register hw i2c, err: %d\n", res); + goto Out_err; + } + } - control->owner = THIS_MODULE; - control->class = I2C_CLASS_HWMON; - control->dev.parent = &adev->pdev->dev; - control->algo = &arcturus_i2c_algo; - control->quirks = &arcturus_i2c_control_quirks; - snprintf(control->name, sizeof(control->name), "AMDGPU SMU"); + adev->pm.ras_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter; + adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[1].adapter; - res = i2c_add_adapter(control); - if (res) - DRM_ERROR("Failed to register hw i2c, err: %d\n", res); + return 0; +Out_err: + for ( ; i >= 0; i--) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + i2c_del_adapter(control); + } return res; } -static void arcturus_i2c_control_fini(struct smu_context *smu, struct i2c_adapter *control) +static void arcturus_i2c_control_fini(struct smu_context *smu) { - i2c_del_adapter(control); + struct amdgpu_device *adev = smu->adev; + int i; + + for (i = 0; i < MAX_SMU_I2C_BUSES; i++) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + + i2c_del_adapter(control); + } + adev->pm.ras_eeprom_i2c_bus = NULL; + adev->pm.fru_eeprom_i2c_bus = NULL; } static void arcturus_get_unique_id(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/cyan_skillfish_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/cyan_skillfish_ppt.c index 2238ee19c222..b3a0f3fb3e65 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/cyan_skillfish_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/cyan_skillfish_ppt.c @@ -125,22 +125,6 @@ static int cyan_skillfish_init_smc_tables(struct smu_context *smu) return smu_v11_0_init_smc_tables(smu); } -static int cyan_skillfish_finit_smc_tables(struct smu_context *smu) -{ - struct smu_table_context *smu_table = &smu->smu_table; - - kfree(smu_table->metrics_table); - smu_table->metrics_table = NULL; - - kfree(smu_table->gpu_metrics_table); - smu_table->gpu_metrics_table = NULL; - smu_table->gpu_metrics_table_size = 0; - - smu_table->metrics_time = 0; - - return 0; -} - static int cyan_skillfish_get_smu_metrics_data(struct smu_context *smu, MetricsMember_t member, @@ -150,13 +134,9 @@ cyan_skillfish_get_smu_metrics_data(struct smu_context *smu, SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, NULL, false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, NULL, false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -200,8 +180,6 @@ cyan_skillfish_get_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -215,8 +193,6 @@ static int cyan_skillfish_read_sensor(struct smu_context *smu, if (!data || !size) return -EINVAL; - mutex_lock(&smu->sensor_lock); - switch (sensor) { case AMDGPU_PP_SENSOR_GFX_SCLK: ret = cyan_skillfish_get_smu_metrics_data(smu, @@ -267,8 +243,6 @@ static int cyan_skillfish_read_sensor(struct smu_context *smu, break; } - mutex_unlock(&smu->sensor_lock); - return ret; } @@ -376,20 +350,16 @@ static bool cyan_skillfish_is_dpm_running(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; int ret = 0; - uint32_t feature_mask[2]; uint64_t feature_enabled; /* we need to re-init after suspend so return false */ if (adev->in_suspend) return false; - ret = smu_cmn_get_enabled_32_bits_mask(smu, feature_mask, 2); + ret = smu_cmn_get_enabled_mask(smu, &feature_enabled); if (ret) return false; - feature_enabled = (uint64_t)feature_mask[0] | - ((uint64_t)feature_mask[1] << 32); - /* * cyan_skillfish specific, query default sclk inseted of hard code. */ @@ -552,6 +522,36 @@ static int cyan_skillfish_od_edit_dpm_table(struct smu_context *smu, return ret; } +static int cyan_skillfish_get_dpm_ultimate_freq(struct smu_context *smu, + enum smu_clk_type clk_type, + uint32_t *min, + uint32_t *max) +{ + int ret = 0; + uint32_t low, high; + + switch (clk_type) { + case SMU_GFXCLK: + case SMU_SCLK: + low = CYAN_SKILLFISH_SCLK_MIN; + high = CYAN_SKILLFISH_SCLK_MAX; + break; + default: + ret = cyan_skillfish_get_current_clk_freq(smu, clk_type, &low); + if (ret) + return ret; + high = low; + break; + } + + if (min) + *min = low; + if (max) + *max = high; + + return 0; +} + static const struct pptable_funcs cyan_skillfish_ppt_funcs = { .check_fw_status = smu_v11_0_check_fw_status, @@ -559,12 +559,14 @@ static const struct pptable_funcs cyan_skillfish_ppt_funcs = { .init_power = smu_v11_0_init_power, .fini_power = smu_v11_0_fini_power, .init_smc_tables = cyan_skillfish_init_smc_tables, - .fini_smc_tables = cyan_skillfish_finit_smc_tables, + .fini_smc_tables = smu_v11_0_fini_smc_tables, .read_sensor = cyan_skillfish_read_sensor, .print_clk_levels = cyan_skillfish_print_clk_levels, + .get_enabled_mask = smu_cmn_get_enabled_mask, .is_dpm_running = cyan_skillfish_is_dpm_running, .get_gpu_metrics = cyan_skillfish_get_gpu_metrics, .od_edit_dpm_table = cyan_skillfish_od_edit_dpm_table, + .get_dpm_ultimate_freq = cyan_skillfish_get_dpm_ultimate_freq, .register_irq_handler = smu_v11_0_register_irq_handler, .notify_memory_pool_location = smu_v11_0_notify_memory_pool_location, .send_smc_msg_with_param = smu_cmn_send_smc_msg_with_param, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 2bb7816b245a..66f9276c4499 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -27,6 +27,7 @@ #include <linux/pci.h> #include <linux/i2c.h> #include "amdgpu.h" +#include "amdgpu_dpm.h" #include "amdgpu_smu.h" #include "atomfirmware.h" #include "amdgpu_atomfirmware.h" @@ -57,8 +58,6 @@ #undef pr_info #undef pr_debug -#define to_amdgpu_device(x) (container_of(x, struct amdgpu_device, pm.smu_i2c)) - #define FEATURE_MASK(feature) (1ULL << feature) #define SMC_DPM_FEATURE ( \ FEATURE_MASK(FEATURE_DPM_PREFETCHER_BIT) | \ @@ -511,6 +510,8 @@ static int navi10_tables_init(struct smu_context *smu) SMU_TABLE_INIT(tables, SMU_TABLE_ACTIVITY_MONITOR_COEFF, sizeof(DpmActivityMonitorCoeffInt_t), PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); + SMU_TABLE_INIT(tables, SMU_TABLE_DRIVER_SMU_CONFIG, sizeof(DriverSmuConfig_t), + PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); smu_table->metrics_table = kzalloc(sizeof(SmuMetrics_NV1X_t), GFP_KERNEL); @@ -527,8 +528,15 @@ static int navi10_tables_init(struct smu_context *smu) if (!smu_table->watermarks_table) goto err2_out; + smu_table->driver_smu_config_table = + kzalloc(tables[SMU_TABLE_DRIVER_SMU_CONFIG].size, GFP_KERNEL); + if (!smu_table->driver_smu_config_table) + goto err3_out; + return 0; +err3_out: + kfree(smu_table->watermarks_table); err2_out: kfree(smu_table->gpu_metrics_table); err1_out: @@ -546,15 +554,11 @@ static int navi10_get_legacy_smu_metrics_data(struct smu_context *smu, (SmuMetrics_legacy_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -624,8 +628,6 @@ static int navi10_get_legacy_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -638,15 +640,11 @@ static int navi10_get_smu_metrics_data(struct smu_context *smu, (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -719,8 +717,6 @@ static int navi10_get_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -733,15 +729,11 @@ static int navi12_get_legacy_smu_metrics_data(struct smu_context *smu, (SmuMetrics_NV12_legacy_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -811,8 +803,6 @@ static int navi12_get_legacy_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -825,15 +815,11 @@ static int navi12_get_smu_metrics_data(struct smu_context *smu, (SmuMetrics_NV12_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -906,8 +892,6 @@ static int navi12_get_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -1261,6 +1245,215 @@ static void navi10_od_setting_get_range(struct smu_11_0_overdrive_table *od_tabl *max = od_table->max[setting]; } +static int navi10_emit_clk_levels(struct smu_context *smu, + enum smu_clk_type clk_type, + char *buf, + int *offset) +{ + uint16_t *curve_settings; + int ret = 0; + uint32_t cur_value = 0, value = 0; + uint32_t freq_values[3] = {0}; + uint32_t i, levels, mark_index = 0, count = 0; + struct smu_table_context *table_context = &smu->smu_table; + uint32_t gen_speed, lane_width; + struct smu_dpm_context *smu_dpm = &smu->smu_dpm; + struct smu_11_0_dpm_context *dpm_context = smu_dpm->dpm_context; + PPTable_t *pptable = (PPTable_t *)table_context->driver_pptable; + OverDriveTable_t *od_table = + (OverDriveTable_t *)table_context->overdrive_table; + struct smu_11_0_overdrive_table *od_settings = smu->od_settings; + uint32_t min_value, max_value; + + switch (clk_type) { + case SMU_GFXCLK: + case SMU_SCLK: + case SMU_SOCCLK: + case SMU_MCLK: + case SMU_UCLK: + case SMU_FCLK: + case SMU_VCLK: + case SMU_DCLK: + case SMU_DCEFCLK: + ret = navi10_get_current_clk_freq_by_table(smu, clk_type, &cur_value); + if (ret) + return ret; + + ret = smu_v11_0_get_dpm_level_count(smu, clk_type, &count); + if (ret) + return ret; + + if (!navi10_is_support_fine_grained_dpm(smu, clk_type)) { + for (i = 0; i < count; i++) { + ret = smu_v11_0_get_dpm_freq_by_index(smu, + clk_type, i, &value); + if (ret) + return ret; + + *offset += sysfs_emit_at(buf, *offset, + "%d: %uMhz %s\n", + i, value, + cur_value == value ? "*" : ""); + } + } else { + ret = smu_v11_0_get_dpm_freq_by_index(smu, + clk_type, 0, &freq_values[0]); + if (ret) + return ret; + ret = smu_v11_0_get_dpm_freq_by_index(smu, + clk_type, + count - 1, + &freq_values[2]); + if (ret) + return ret; + + freq_values[1] = cur_value; + mark_index = cur_value == freq_values[0] ? 0 : + cur_value == freq_values[2] ? 2 : 1; + + levels = 3; + if (mark_index != 1) { + levels = 2; + freq_values[1] = freq_values[2]; + } + + for (i = 0; i < levels; i++) { + *offset += sysfs_emit_at(buf, *offset, + "%d: %uMhz %s\n", + i, freq_values[i], + i == mark_index ? "*" : ""); + } + } + break; + case SMU_PCIE: + gen_speed = smu_v11_0_get_current_pcie_link_speed_level(smu); + lane_width = smu_v11_0_get_current_pcie_link_width_level(smu); + for (i = 0; i < NUM_LINK_LEVELS; i++) { + *offset += sysfs_emit_at(buf, *offset, "%d: %s %s %dMhz %s\n", i, + (dpm_context->dpm_tables.pcie_table.pcie_gen[i] == 0) ? "2.5GT/s," : + (dpm_context->dpm_tables.pcie_table.pcie_gen[i] == 1) ? "5.0GT/s," : + (dpm_context->dpm_tables.pcie_table.pcie_gen[i] == 2) ? "8.0GT/s," : + (dpm_context->dpm_tables.pcie_table.pcie_gen[i] == 3) ? "16.0GT/s," : "", + (dpm_context->dpm_tables.pcie_table.pcie_lane[i] == 1) ? "x1" : + (dpm_context->dpm_tables.pcie_table.pcie_lane[i] == 2) ? "x2" : + (dpm_context->dpm_tables.pcie_table.pcie_lane[i] == 3) ? "x4" : + (dpm_context->dpm_tables.pcie_table.pcie_lane[i] == 4) ? "x8" : + (dpm_context->dpm_tables.pcie_table.pcie_lane[i] == 5) ? "x12" : + (dpm_context->dpm_tables.pcie_table.pcie_lane[i] == 6) ? "x16" : "", + pptable->LclkFreq[i], + (gen_speed == dpm_context->dpm_tables.pcie_table.pcie_gen[i]) && + (lane_width == dpm_context->dpm_tables.pcie_table.pcie_lane[i]) ? + "*" : ""); + } + break; + case SMU_OD_SCLK: + if (!smu->od_enabled || !od_table || !od_settings) + return -EOPNOTSUPP; + if (!navi10_od_feature_is_supported(od_settings, SMU_11_0_ODCAP_GFXCLK_LIMITS)) + break; + *offset += sysfs_emit_at(buf, *offset, "OD_SCLK:\n0: %uMhz\n1: %uMhz\n", + od_table->GfxclkFmin, od_table->GfxclkFmax); + break; + case SMU_OD_MCLK: + if (!smu->od_enabled || !od_table || !od_settings) + return -EOPNOTSUPP; + if (!navi10_od_feature_is_supported(od_settings, SMU_11_0_ODCAP_UCLK_MAX)) + break; + *offset += sysfs_emit_at(buf, *offset, "OD_MCLK:\n1: %uMHz\n", od_table->UclkFmax); + break; + case SMU_OD_VDDC_CURVE: + if (!smu->od_enabled || !od_table || !od_settings) + return -EOPNOTSUPP; + if (!navi10_od_feature_is_supported(od_settings, SMU_11_0_ODCAP_GFXCLK_CURVE)) + break; + *offset += sysfs_emit_at(buf, *offset, "OD_VDDC_CURVE:\n"); + for (i = 0; i < 3; i++) { + switch (i) { + case 0: + curve_settings = &od_table->GfxclkFreq1; + break; + case 1: + curve_settings = &od_table->GfxclkFreq2; + break; + case 2: + curve_settings = &od_table->GfxclkFreq3; + break; + default: + break; + } + *offset += sysfs_emit_at(buf, *offset, "%d: %uMHz %umV\n", + i, curve_settings[0], + curve_settings[1] / NAVI10_VOLTAGE_SCALE); + } + break; + case SMU_OD_RANGE: + if (!smu->od_enabled || !od_table || !od_settings) + return -EOPNOTSUPP; + *offset += sysfs_emit_at(buf, *offset, "%s:\n", "OD_RANGE"); + + if (navi10_od_feature_is_supported(od_settings, SMU_11_0_ODCAP_GFXCLK_LIMITS)) { + navi10_od_setting_get_range(od_settings, SMU_11_0_ODSETTING_GFXCLKFMIN, + &min_value, NULL); + navi10_od_setting_get_range(od_settings, SMU_11_0_ODSETTING_GFXCLKFMAX, + NULL, &max_value); + *offset += sysfs_emit_at(buf, *offset, "SCLK: %7uMhz %10uMhz\n", + min_value, max_value); + } + + if (navi10_od_feature_is_supported(od_settings, SMU_11_0_ODCAP_UCLK_MAX)) { + navi10_od_setting_get_range(od_settings, SMU_11_0_ODSETTING_UCLKFMAX, + &min_value, &max_value); + *offset += sysfs_emit_at(buf, *offset, "MCLK: %7uMhz %10uMhz\n", + min_value, max_value); + } + + if (navi10_od_feature_is_supported(od_settings, SMU_11_0_ODCAP_GFXCLK_CURVE)) { + navi10_od_setting_get_range(od_settings, + SMU_11_0_ODSETTING_VDDGFXCURVEFREQ_P1, + &min_value, &max_value); + *offset += sysfs_emit_at(buf, *offset, + "VDDC_CURVE_SCLK[0]: %7uMhz %10uMhz\n", + min_value, max_value); + navi10_od_setting_get_range(od_settings, + SMU_11_0_ODSETTING_VDDGFXCURVEVOLTAGE_P1, + &min_value, &max_value); + *offset += sysfs_emit_at(buf, *offset, + "VDDC_CURVE_VOLT[0]: %7dmV %11dmV\n", + min_value, max_value); + navi10_od_setting_get_range(od_settings, + SMU_11_0_ODSETTING_VDDGFXCURVEFREQ_P2, + &min_value, &max_value); + *offset += sysfs_emit_at(buf, *offset, + "VDDC_CURVE_SCLK[1]: %7uMhz %10uMhz\n", + min_value, max_value); + navi10_od_setting_get_range(od_settings, + SMU_11_0_ODSETTING_VDDGFXCURVEVOLTAGE_P2, + &min_value, &max_value); + *offset += sysfs_emit_at(buf, *offset, + "VDDC_CURVE_VOLT[1]: %7dmV %11dmV\n", + min_value, max_value); + navi10_od_setting_get_range(od_settings, + SMU_11_0_ODSETTING_VDDGFXCURVEFREQ_P3, + &min_value, &max_value); + *offset += sysfs_emit_at(buf, *offset, + "VDDC_CURVE_SCLK[2]: %7uMhz %10uMhz\n", + min_value, max_value); + navi10_od_setting_get_range(od_settings, + SMU_11_0_ODSETTING_VDDGFXCURVEVOLTAGE_P3, + &min_value, &max_value); + *offset += sysfs_emit_at(buf, *offset, + "VDDC_CURVE_VOLT[2]: %7dmV %11dmV\n", + min_value, max_value); + } + + break; + default: + break; + } + + return 0; +} + static int navi10_print_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, char *buf) { @@ -1649,8 +1842,8 @@ static int navi10_display_config_changed(struct smu_context *smu) int ret = 0; if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && - smu_cmn_feature_is_supported(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) && - smu_cmn_feature_is_supported(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { + smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) && + smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, smu->display_config->num_display, NULL); @@ -1664,15 +1857,12 @@ static int navi10_display_config_changed(struct smu_context *smu) static bool navi10_is_dpm_running(struct smu_context *smu) { int ret = 0; - uint32_t feature_mask[2]; uint64_t feature_enabled; - ret = smu_cmn_get_enabled_mask(smu, feature_mask, 2); + ret = smu_cmn_get_enabled_mask(smu, &feature_enabled); if (ret) return false; - feature_enabled = (uint64_t)feature_mask[1] << 32 | feature_mask[0]; - return !!(feature_enabled & SMC_DPM_FEATURE); } @@ -1888,13 +2078,13 @@ static int navi10_notify_smc_display_config(struct smu_context *smu) min_clocks.dcef_clock_in_sr = smu->display_config->min_dcef_deep_sleep_set_clk; min_clocks.memory_clock = smu->display_config->min_mem_set_clock; - if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_DPM_DCEFCLK_BIT)) { + if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT)) { clock_req.clock_type = amd_pp_dcef_clock; clock_req.clock_freq_in_khz = min_clocks.dcef_clock * 10; ret = smu_v11_0_display_clock_voltage_request(smu, &clock_req); if (!ret) { - if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_DS_DCEFCLK_BIT)) { + if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DS_DCEFCLK_BIT)) { ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetMinDeepSleepDcefclk, min_clocks.dcef_clock_in_sr/100, @@ -1988,7 +2178,6 @@ static int navi10_read_sensor(struct smu_context *smu, if(!data || !size) return -EINVAL; - mutex_lock(&smu->sensor_lock); switch (sensor) { case AMDGPU_PP_SENSOR_MAX_FAN_RPM: *(uint32_t *)data = pptable->FanMaximumRpm; @@ -2048,7 +2237,6 @@ static int navi10_read_sensor(struct smu_context *smu, ret = -EOPNOTSUPP; break; } - mutex_unlock(&smu->sensor_lock); return ret; } @@ -2708,20 +2896,14 @@ static ssize_t navi10_get_legacy_gpu_metrics(struct smu_context *smu, SmuMetrics_legacy_t metrics; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - true); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + true); + if (ret) return ret; - } memcpy(&metrics, smu_table->metrics_table, sizeof(SmuMetrics_legacy_t)); - mutex_unlock(&smu->metrics_lock); - smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 3); gpu_metrics->temperature_edge = metrics.TemperatureEdge; @@ -2778,18 +2960,23 @@ static ssize_t navi10_get_legacy_gpu_metrics(struct smu_context *smu, static int navi10_i2c_xfer(struct i2c_adapter *i2c_adap, struct i2c_msg *msg, int num_msgs) { - struct amdgpu_device *adev = to_amdgpu_device(i2c_adap); - struct smu_table_context *smu_table = &adev->smu.smu_table; + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(i2c_adap); + struct amdgpu_device *adev = smu_i2c->adev; + struct smu_context *smu = adev->powerplay.pp_handle; + struct smu_table_context *smu_table = &smu->smu_table; struct smu_table *table = &smu_table->driver_table; SwI2cRequest_t *req, *res = (SwI2cRequest_t *)table->cpu_addr; int i, j, r, c; u16 dir; + if (!adev->pm.dpm_enabled) + return -EBUSY; + req = kzalloc(sizeof(*req), GFP_KERNEL); if (!req) return -ENOMEM; - req->I2CcontrollerPort = 0; + req->I2CcontrollerPort = smu_i2c->port; req->I2CSpeed = I2C_SPEED_FAST_400K; req->SlaveAddress = msg[0].addr << 1; /* wants an 8-bit address */ dir = msg[0].flags & I2C_M_RD; @@ -2825,9 +3012,9 @@ static int navi10_i2c_xfer(struct i2c_adapter *i2c_adap, } } } - mutex_lock(&adev->smu.mutex); - r = smu_cmn_update_table(&adev->smu, SMU_TABLE_I2C_COMMANDS, 0, req, true); - mutex_unlock(&adev->smu.mutex); + mutex_lock(&adev->pm.mutex); + r = smu_cmn_update_table(smu, SMU_TABLE_I2C_COMMANDS, 0, req, true); + mutex_unlock(&adev->pm.mutex); if (r) goto fail; @@ -2867,28 +3054,60 @@ static const struct i2c_adapter_quirks navi10_i2c_control_quirks = { .max_comb_2nd_msg_len = MAX_SW_I2C_COMMANDS - 2, }; -static int navi10_i2c_control_init(struct smu_context *smu, struct i2c_adapter *control) +static int navi10_i2c_control_init(struct smu_context *smu) { - struct amdgpu_device *adev = to_amdgpu_device(control); - int res; + struct amdgpu_device *adev = smu->adev; + int res, i; + + for (i = 0; i < MAX_SMU_I2C_BUSES; i++) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + + smu_i2c->adev = adev; + smu_i2c->port = i; + mutex_init(&smu_i2c->mutex); + control->owner = THIS_MODULE; + control->class = I2C_CLASS_HWMON; + control->dev.parent = &adev->pdev->dev; + control->algo = &navi10_i2c_algo; + snprintf(control->name, sizeof(control->name), "AMDGPU SMU %d", i); + control->quirks = &navi10_i2c_control_quirks; + i2c_set_adapdata(control, smu_i2c); + + res = i2c_add_adapter(control); + if (res) { + DRM_ERROR("Failed to register hw i2c, err: %d\n", res); + goto Out_err; + } + } - control->owner = THIS_MODULE; - control->class = I2C_CLASS_HWMON; - control->dev.parent = &adev->pdev->dev; - control->algo = &navi10_i2c_algo; - snprintf(control->name, sizeof(control->name), "AMDGPU SMU"); - control->quirks = &navi10_i2c_control_quirks; + adev->pm.ras_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter; + adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[1].adapter; - res = i2c_add_adapter(control); - if (res) - DRM_ERROR("Failed to register hw i2c, err: %d\n", res); + return 0; +Out_err: + for ( ; i >= 0; i--) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + i2c_del_adapter(control); + } return res; } -static void navi10_i2c_control_fini(struct smu_context *smu, struct i2c_adapter *control) +static void navi10_i2c_control_fini(struct smu_context *smu) { - i2c_del_adapter(control); + struct amdgpu_device *adev = smu->adev; + int i; + + for (i = 0; i < MAX_SMU_I2C_BUSES; i++) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + + i2c_del_adapter(control); + } + adev->pm.ras_eeprom_i2c_bus = NULL; + adev->pm.fru_eeprom_i2c_bus = NULL; } static ssize_t navi10_get_gpu_metrics(struct smu_context *smu, @@ -2900,20 +3119,14 @@ static ssize_t navi10_get_gpu_metrics(struct smu_context *smu, SmuMetrics_t metrics; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - true); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + true); + if (ret) return ret; - } memcpy(&metrics, smu_table->metrics_table, sizeof(SmuMetrics_t)); - mutex_unlock(&smu->metrics_lock); - smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 3); gpu_metrics->temperature_edge = metrics.TemperatureEdge; @@ -2978,20 +3191,14 @@ static ssize_t navi12_get_legacy_gpu_metrics(struct smu_context *smu, SmuMetrics_NV12_legacy_t metrics; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - true); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + true); + if (ret) return ret; - } memcpy(&metrics, smu_table->metrics_table, sizeof(SmuMetrics_NV12_legacy_t)); - mutex_unlock(&smu->metrics_lock); - smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 3); gpu_metrics->temperature_edge = metrics.TemperatureEdge; @@ -3059,20 +3266,14 @@ static ssize_t navi12_get_gpu_metrics(struct smu_context *smu, SmuMetrics_NV12_t metrics; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - true); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + true); + if (ret) return ret; - } memcpy(&metrics, smu_table->metrics_table, sizeof(SmuMetrics_NV12_t)); - mutex_unlock(&smu->metrics_lock); - smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 3); gpu_metrics->temperature_edge = metrics.TemperatureEdge; @@ -3237,6 +3438,7 @@ static const struct pptable_funcs navi10_ppt_funcs = { .i2c_init = navi10_i2c_control_init, .i2c_fini = navi10_i2c_control_fini, .print_clk_levels = navi10_print_clk_levels, + .emit_clk_levels = navi10_emit_clk_levels, .force_clk_levels = navi10_force_clk_levels, .populate_umd_state_clk = navi10_populate_umd_state_clk, .get_clock_by_type_with_latency = navi10_get_clock_by_type_with_latency, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index 2320bd750876..358031c8c79e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -27,6 +27,7 @@ #include <linux/pci.h> #include <linux/i2c.h> #include "amdgpu.h" +#include "amdgpu_dpm.h" #include "amdgpu_smu.h" #include "atomfirmware.h" #include "amdgpu_atomfirmware.h" @@ -46,6 +47,7 @@ #include "mp/mp_11_0_sh_mask.h" #include "asic_reg/mp/mp_11_0_sh_mask.h" +#include "amdgpu_ras.h" #include "smu_cmn.h" /* @@ -58,8 +60,6 @@ #undef pr_info #undef pr_debug -#define to_amdgpu_device(x) (container_of(x, struct amdgpu_device, pm.smu_i2c)) - #define FEATURE_MASK(feature) (1ULL << feature) #define SMC_DPM_FEATURE ( \ FEATURE_MASK(FEATURE_DPM_PREFETCHER_BIT) | \ @@ -83,6 +83,12 @@ /* STB FIFO depth is in 64bit units */ #define SIENNA_CICHLID_STB_DEPTH_UNIT_BYTES 8 +/* + * SMU support ECCTABLE since version 58.70.0, + * use this to check whether ECCTABLE feature is supported. + */ +#define SUPPORT_ECCTABLE_SMU_VERSION 0x003a4600 + static int get_table_size(struct smu_context *smu) { if (smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 13)) @@ -225,6 +231,7 @@ static struct cmn2asic_mapping sienna_cichlid_table_map[SMU_TABLE_COUNT] = { TAB_MAP(OVERDRIVE), TAB_MAP(I2C_COMMANDS), TAB_MAP(PACE), + TAB_MAP(ECCINFO), }; static struct cmn2asic_mapping sienna_cichlid_pwr_src_map[SMU_POWER_SOURCE_COUNT] = { @@ -466,6 +473,10 @@ static int sienna_cichlid_tables_init(struct smu_context *smu) SMU_TABLE_INIT(tables, SMU_TABLE_ACTIVITY_MONITOR_COEFF, sizeof(DpmActivityMonitorCoeffIntExternal_t), PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); + SMU_TABLE_INIT(tables, SMU_TABLE_ECCINFO, sizeof(EccInfoTable_t), + PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); + SMU_TABLE_INIT(tables, SMU_TABLE_DRIVER_SMU_CONFIG, sizeof(DriverSmuConfigExternal_t), + PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); smu_table->metrics_table = kzalloc(sizeof(SmuMetricsExternal_t), GFP_KERNEL); if (!smu_table->metrics_table) @@ -481,8 +492,21 @@ static int sienna_cichlid_tables_init(struct smu_context *smu) if (!smu_table->watermarks_table) goto err2_out; + smu_table->ecc_table = kzalloc(tables[SMU_TABLE_ECCINFO].size, GFP_KERNEL); + if (!smu_table->ecc_table) + goto err3_out; + + smu_table->driver_smu_config_table = + kzalloc(tables[SMU_TABLE_DRIVER_SMU_CONFIG].size, GFP_KERNEL); + if (!smu_table->driver_smu_config_table) + goto err4_out; + return 0; +err4_out: + kfree(smu_table->ecc_table); +err3_out: + kfree(smu_table->watermarks_table); err2_out: kfree(smu_table->gpu_metrics_table); err1_out: @@ -525,15 +549,11 @@ static int sienna_cichlid_get_smu_metrics_data(struct smu_context *smu, uint16_t average_gfx_activity; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -633,8 +653,6 @@ static int sienna_cichlid_get_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -1036,10 +1054,6 @@ static int sienna_cichlid_print_clk_levels(struct smu_context *smu, if (ret) goto print_clk_out; - /* no need to disable gfxoff when retrieving the current gfxclk */ - if ((clk_type == SMU_GFXCLK) || (clk_type == SMU_SCLK)) - amdgpu_gfx_off_ctrl(adev, false); - ret = smu_v11_0_get_dpm_level_count(smu, clk_type, &count); if (ret) goto print_clk_out; @@ -1168,25 +1182,18 @@ static int sienna_cichlid_print_clk_levels(struct smu_context *smu, } print_clk_out: - if ((clk_type == SMU_GFXCLK) || (clk_type == SMU_SCLK)) - amdgpu_gfx_off_ctrl(adev, true); - return size; } static int sienna_cichlid_force_clk_levels(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t mask) { - struct amdgpu_device *adev = smu->adev; int ret = 0; uint32_t soft_min_level = 0, soft_max_level = 0, min_freq = 0, max_freq = 0; soft_min_level = mask ? (ffs(mask) - 1) : 0; soft_max_level = mask ? (fls(mask) - 1) : 0; - if ((clk_type == SMU_GFXCLK) || (clk_type == SMU_SCLK)) - amdgpu_gfx_off_ctrl(adev, false); - switch (clk_type) { case SMU_GFXCLK: case SMU_SCLK: @@ -1220,9 +1227,6 @@ static int sienna_cichlid_force_clk_levels(struct smu_context *smu, } forec_level_out: - if ((clk_type == SMU_GFXCLK) || (clk_type == SMU_SCLK)) - amdgpu_gfx_off_ctrl(adev, true); - return 0; } @@ -1303,8 +1307,8 @@ static int sienna_cichlid_display_config_changed(struct smu_context *smu) int ret = 0; if ((smu->watermarks_bitmap & WATERMARKS_EXIST) && - smu_cmn_feature_is_supported(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) && - smu_cmn_feature_is_supported(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { + smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT) && + smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_SOCCLK_BIT)) { #if 0 ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_NumOfDisplays, smu->display_config->num_display, @@ -1320,15 +1324,12 @@ static int sienna_cichlid_display_config_changed(struct smu_context *smu) static bool sienna_cichlid_is_dpm_running(struct smu_context *smu) { int ret = 0; - uint32_t feature_mask[2]; uint64_t feature_enabled; - ret = smu_cmn_get_enabled_mask(smu, feature_mask, 2); + ret = smu_cmn_get_enabled_mask(smu, &feature_enabled); if (ret) return false; - feature_enabled = (uint64_t)feature_mask[1] << 32 | feature_mask[0]; - return !!(feature_enabled & SMC_DPM_FEATURE); } @@ -1543,13 +1544,13 @@ static int sienna_cichlid_notify_smc_display_config(struct smu_context *smu) min_clocks.dcef_clock_in_sr = smu->display_config->min_dcef_deep_sleep_set_clk; min_clocks.memory_clock = smu->display_config->min_mem_set_clock; - if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_DPM_DCEFCLK_BIT)) { + if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCEFCLK_BIT)) { clock_req.clock_type = amd_pp_dcef_clock; clock_req.clock_freq_in_khz = min_clocks.dcef_clock * 10; ret = smu_v11_0_display_clock_voltage_request(smu, &clock_req); if (!ret) { - if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_DS_DCEFCLK_BIT)) { + if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DS_DCEFCLK_BIT)) { ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetMinDeepSleepDcefclk, min_clocks.dcef_clock_in_sr/100, @@ -1641,7 +1642,6 @@ static int sienna_cichlid_read_sensor(struct smu_context *smu, if(!data || !size) return -EINVAL; - mutex_lock(&smu->sensor_lock); switch (sensor) { case AMDGPU_PP_SENSOR_MAX_FAN_RPM: GET_PPTABLE_MEMBER(FanMaximumRpm, &temp); @@ -1702,7 +1702,6 @@ static int sienna_cichlid_read_sensor(struct smu_context *smu, ret = -EOPNOTSUPP; break; } - mutex_unlock(&smu->sensor_lock); return ret; } @@ -1881,16 +1880,7 @@ static int sienna_cichlid_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *min, uint32_t *max) { - struct amdgpu_device *adev = smu->adev; - int ret; - - if (clk_type == SMU_GFXCLK) - amdgpu_gfx_off_ctrl(adev, false); - ret = smu_v11_0_get_dpm_ultimate_freq(smu, clk_type, min, max); - if (clk_type == SMU_GFXCLK) - amdgpu_gfx_off_ctrl(adev, true); - - return ret; + return smu_v11_0_get_dpm_ultimate_freq(smu, clk_type, min, max); } static void sienna_cichlid_dump_od_table(struct smu_context *smu, @@ -3474,18 +3464,23 @@ static void sienna_cichlid_dump_pptable(struct smu_context *smu) static int sienna_cichlid_i2c_xfer(struct i2c_adapter *i2c_adap, struct i2c_msg *msg, int num_msgs) { - struct amdgpu_device *adev = to_amdgpu_device(i2c_adap); - struct smu_table_context *smu_table = &adev->smu.smu_table; + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(i2c_adap); + struct amdgpu_device *adev = smu_i2c->adev; + struct smu_context *smu = adev->powerplay.pp_handle; + struct smu_table_context *smu_table = &smu->smu_table; struct smu_table *table = &smu_table->driver_table; SwI2cRequest_t *req, *res = (SwI2cRequest_t *)table->cpu_addr; int i, j, r, c; u16 dir; + if (!adev->pm.dpm_enabled) + return -EBUSY; + req = kzalloc(sizeof(*req), GFP_KERNEL); if (!req) return -ENOMEM; - req->I2CcontrollerPort = 1; + req->I2CcontrollerPort = smu_i2c->port; req->I2CSpeed = I2C_SPEED_FAST_400K; req->SlaveAddress = msg[0].addr << 1; /* wants an 8-bit address */ dir = msg[0].flags & I2C_M_RD; @@ -3521,9 +3516,9 @@ static int sienna_cichlid_i2c_xfer(struct i2c_adapter *i2c_adap, } } } - mutex_lock(&adev->smu.mutex); - r = smu_cmn_update_table(&adev->smu, SMU_TABLE_I2C_COMMANDS, 0, req, true); - mutex_unlock(&adev->smu.mutex); + mutex_lock(&adev->pm.mutex); + r = smu_cmn_update_table(smu, SMU_TABLE_I2C_COMMANDS, 0, req, true); + mutex_unlock(&adev->pm.mutex); if (r) goto fail; @@ -3563,28 +3558,61 @@ static const struct i2c_adapter_quirks sienna_cichlid_i2c_control_quirks = { .max_comb_2nd_msg_len = MAX_SW_I2C_COMMANDS - 2, }; -static int sienna_cichlid_i2c_control_init(struct smu_context *smu, struct i2c_adapter *control) +static int sienna_cichlid_i2c_control_init(struct smu_context *smu) { - struct amdgpu_device *adev = to_amdgpu_device(control); - int res; - - control->owner = THIS_MODULE; - control->class = I2C_CLASS_HWMON; - control->dev.parent = &adev->pdev->dev; - control->algo = &sienna_cichlid_i2c_algo; - snprintf(control->name, sizeof(control->name), "AMDGPU SMU"); - control->quirks = &sienna_cichlid_i2c_control_quirks; + struct amdgpu_device *adev = smu->adev; + int res, i; + + for (i = 0; i < MAX_SMU_I2C_BUSES; i++) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + + smu_i2c->adev = adev; + smu_i2c->port = i; + mutex_init(&smu_i2c->mutex); + control->owner = THIS_MODULE; + control->class = I2C_CLASS_HWMON; + control->dev.parent = &adev->pdev->dev; + control->algo = &sienna_cichlid_i2c_algo; + snprintf(control->name, sizeof(control->name), "AMDGPU SMU %d", i); + control->quirks = &sienna_cichlid_i2c_control_quirks; + i2c_set_adapdata(control, smu_i2c); + + res = i2c_add_adapter(control); + if (res) { + DRM_ERROR("Failed to register hw i2c, err: %d\n", res); + goto Out_err; + } + } + /* assign the buses used for the FRU EEPROM and RAS EEPROM */ + /* XXX ideally this would be something in a vbios data table */ + adev->pm.ras_eeprom_i2c_bus = &adev->pm.smu_i2c[1].adapter; + adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter; - res = i2c_add_adapter(control); - if (res) - DRM_ERROR("Failed to register hw i2c, err: %d\n", res); + return 0; +Out_err: + for ( ; i >= 0; i--) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + i2c_del_adapter(control); + } return res; } -static void sienna_cichlid_i2c_control_fini(struct smu_context *smu, struct i2c_adapter *control) +static void sienna_cichlid_i2c_control_fini(struct smu_context *smu) { - i2c_del_adapter(control); + struct amdgpu_device *adev = smu->adev; + int i; + + for (i = 0; i < MAX_SMU_I2C_BUSES; i++) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + + i2c_del_adapter(control); + } + adev->pm.ras_eeprom_i2c_bus = NULL; + adev->pm.fru_eeprom_i2c_bus = NULL; } static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu, @@ -3604,14 +3632,11 @@ static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu, uint16_t average_gfx_activity; int ret = 0; - mutex_lock(&smu->metrics_lock); - ret = smu_cmn_get_metrics_table_locked(smu, - &metrics_external, - true); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + &metrics_external, + true); + if (ret) return ret; - } smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 3); @@ -3701,8 +3726,6 @@ static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu, smu_v11_0_get_current_pcie_link_speed(smu); } - mutex_unlock(&smu->metrics_lock); - gpu_metrics->system_clock_counter = ktime_get_boottime_ns(); *table = (void *)gpu_metrics; @@ -3710,6 +3733,60 @@ static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu, return sizeof(struct gpu_metrics_v1_3); } +static int sienna_cichlid_check_ecc_table_support(struct smu_context *smu) +{ + uint32_t if_version = 0xff, smu_version = 0xff; + int ret = 0; + + ret = smu_cmn_get_smc_version(smu, &if_version, &smu_version); + if (ret) + return -EOPNOTSUPP; + + if (smu_version < SUPPORT_ECCTABLE_SMU_VERSION) + ret = -EOPNOTSUPP; + + return ret; +} + +static ssize_t sienna_cichlid_get_ecc_info(struct smu_context *smu, + void *table) +{ + struct smu_table_context *smu_table = &smu->smu_table; + EccInfoTable_t *ecc_table = NULL; + struct ecc_info_per_ch *ecc_info_per_channel = NULL; + int i, ret = 0; + struct umc_ecc_info *eccinfo = (struct umc_ecc_info *)table; + + ret = sienna_cichlid_check_ecc_table_support(smu); + if (ret) + return ret; + + ret = smu_cmn_update_table(smu, + SMU_TABLE_ECCINFO, + 0, + smu_table->ecc_table, + false); + if (ret) { + dev_info(smu->adev->dev, "Failed to export SMU ecc table!\n"); + return ret; + } + + ecc_table = (EccInfoTable_t *)smu_table->ecc_table; + + for (i = 0; i < SIENNA_CICHLID_UMC_CHANNEL_NUM; i++) { + ecc_info_per_channel = &(eccinfo->ecc[i]); + ecc_info_per_channel->ce_count_lo_chip = + ecc_table->EccInfo[i].ce_count_lo_chip; + ecc_info_per_channel->ce_count_hi_chip = + ecc_table->EccInfo[i].ce_count_hi_chip; + ecc_info_per_channel->mca_umc_status = + ecc_table->EccInfo[i].mca_umc_status; + ecc_info_per_channel->mca_umc_addr = + ecc_table->EccInfo[i].mca_umc_addr; + } + + return ret; +} static int sienna_cichlid_enable_mgpu_fan_boost(struct smu_context *smu) { uint16_t *mgpu_fan_boost_limit_rpm; @@ -3735,7 +3812,7 @@ static int sienna_cichlid_gpo_control(struct smu_context *smu, int ret = 0; - if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_DPM_GFX_GPO_BIT)) { + if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_GFX_GPO_BIT)) { ret = smu_cmn_get_smc_version(smu, NULL, &smu_version); if (ret) return ret; @@ -3848,9 +3925,9 @@ static void sienna_cichlid_stb_init(struct smu_context *smu) } -int sienna_cichlid_stb_get_data_direct(struct smu_context *smu, - void *buf, - uint32_t size) +static int sienna_cichlid_stb_get_data_direct(struct smu_context *smu, + void *buf, + uint32_t size) { uint32_t *p = buf; struct amdgpu_device *adev = smu->adev; @@ -3961,6 +4038,7 @@ static const struct pptable_funcs sienna_cichlid_ppt_funcs = { .gpo_control = sienna_cichlid_gpo_control, .set_mp1_state = sienna_cichlid_set_mp1_state, .stb_collect_info = sienna_cichlid_stb_get_data_direct, + .get_ecc_info = sienna_cichlid_get_ecc_info, }; void sienna_cichlid_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c index 4e9e2cf39859..b87f550af26b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c @@ -225,15 +225,15 @@ int smu_v11_0_check_fw_version(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; uint32_t if_version = 0xff, smu_version = 0xff; - uint16_t smu_major; - uint8_t smu_minor, smu_debug; + uint8_t smu_program, smu_major, smu_minor, smu_debug; int ret = 0; ret = smu_cmn_get_smc_version(smu, &if_version, &smu_version); if (ret) return ret; - smu_major = (smu_version >> 16) & 0xffff; + smu_program = (smu_version >> 24) & 0xff; + smu_major = (smu_version >> 16) & 0xff; smu_minor = (smu_version >> 8) & 0xff; smu_debug = (smu_version >> 0) & 0xff; if (smu->is_apu) @@ -287,9 +287,9 @@ int smu_v11_0_check_fw_version(struct smu_context *smu) */ if (if_version != smu->smc_driver_if_version) { dev_info(smu->adev->dev, "smu driver if version = 0x%08x, smu fw if version = 0x%08x, " - "smu fw version = 0x%08x (%d.%d.%d)\n", + "smu fw program = %d, version = 0x%08x (%d.%d.%d)\n", smu->smc_driver_if_version, if_version, - smu_version, smu_major, smu_minor, smu_debug); + smu_program, smu_version, smu_major, smu_minor, smu_debug); dev_warn(smu->adev->dev, "SMU driver if version not matched\n"); } @@ -473,8 +473,12 @@ int smu_v11_0_fini_smc_tables(struct smu_context *smu) kfree(smu_table->hardcode_pptable); smu_table->hardcode_pptable = NULL; + kfree(smu_table->driver_smu_config_table); + kfree(smu_table->ecc_table); kfree(smu_table->metrics_table); kfree(smu_table->watermarks_table); + smu_table->driver_smu_config_table = NULL; + smu_table->ecc_table = NULL; smu_table->metrics_table = NULL; smu_table->watermarks_table = NULL; smu_table->metrics_time = 0; @@ -796,30 +800,8 @@ failed: int smu_v11_0_system_features_control(struct smu_context *smu, bool en) { - struct smu_feature *feature = &smu->smu_feature; - uint32_t feature_mask[2]; - int ret = 0; - - ret = smu_cmn_send_smc_msg(smu, (en ? SMU_MSG_EnableAllSmuFeatures : - SMU_MSG_DisableAllSmuFeatures), NULL); - if (ret) - return ret; - - bitmap_zero(feature->enabled, feature->feature_num); - bitmap_zero(feature->supported, feature->feature_num); - - if (en) { - ret = smu_cmn_get_enabled_mask(smu, feature_mask, 2); - if (ret) - return ret; - - bitmap_copy(feature->enabled, (unsigned long *)&feature_mask, - feature->feature_num); - bitmap_copy(feature->supported, (unsigned long *)&feature_mask, - feature->feature_num); - } - - return ret; + return smu_cmn_send_smc_msg(smu, (en ? SMU_MSG_EnableAllSmuFeatures : + SMU_MSG_DisableAllSmuFeatures), NULL); } int smu_v11_0_notify_display_change(struct smu_context *smu) @@ -1372,7 +1354,7 @@ static int smu_v11_0_set_irq_state(struct amdgpu_device *adev, unsigned tyep, enum amdgpu_interrupt_state state) { - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; uint32_t low, high; uint32_t val = 0; @@ -1441,7 +1423,7 @@ static int smu_v11_0_irq_process(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; uint32_t client_id = entry->client_id; uint32_t src_id = entry->src_id; /* @@ -1615,13 +1597,8 @@ bool smu_v11_0_baco_is_support(struct smu_context *smu) enum smu_baco_state smu_v11_0_baco_get_state(struct smu_context *smu) { struct smu_baco_context *smu_baco = &smu->smu_baco; - enum smu_baco_state baco_state; - - mutex_lock(&smu_baco->mutex); - baco_state = smu_baco->state; - mutex_unlock(&smu_baco->mutex); - return baco_state; + return smu_baco->state; } #define D3HOT_BACO_SEQUENCE 0 @@ -1638,8 +1615,6 @@ int smu_v11_0_baco_set_state(struct smu_context *smu, enum smu_baco_state state) if (smu_v11_0_baco_get_state(smu) == state) return 0; - mutex_lock(&smu_baco->mutex); - if (state == SMU_BACO_STATE_ENTER) { switch (adev->ip_versions[MP1_HWIP][0]) { case IP_VERSION(11, 0, 7): @@ -1680,18 +1655,16 @@ int smu_v11_0_baco_set_state(struct smu_context *smu, enum smu_baco_state state) } else { ret = smu_cmn_send_smc_msg(smu, SMU_MSG_ExitBaco, NULL); if (ret) - goto out; + return ret; /* clear vbios scratch 6 and 7 for coming asic reinit */ WREG32(adev->bios_scratch_reg_offset + 6, 0); WREG32(adev->bios_scratch_reg_offset + 7, 0); } - if (ret) - goto out; - smu_baco->state = state; -out: - mutex_unlock(&smu_baco->mutex); + if (!ret) + smu_baco->state = state; + return ret; } @@ -1798,7 +1771,6 @@ int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, uint32_t min, uint32_t max) { - struct amdgpu_device *adev = smu->adev; int ret = 0, clk_id = 0; uint32_t param; @@ -1811,9 +1783,6 @@ int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, if (clk_id < 0) return clk_id; - if (clk_type == SMU_GFXCLK) - amdgpu_gfx_off_ctrl(adev, false); - if (max > 0) { param = (uint32_t)((clk_id << 16) | (max & 0xffff)); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxByFreq, @@ -1831,9 +1800,6 @@ int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, } out: - if (clk_type == SMU_GFXCLK) - amdgpu_gfx_off_ctrl(adev, true); - return ret; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c index 5cb07ed227fb..96a5b31f708d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c @@ -273,15 +273,11 @@ static int vangogh_get_legacy_smu_metrics_data(struct smu_context *smu, SmuMetrics_legacy_t *metrics = (SmuMetrics_legacy_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -335,8 +331,6 @@ static int vangogh_get_legacy_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -348,15 +342,11 @@ static int vangogh_get_smu_metrics_data(struct smu_context *smu, SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -410,8 +400,6 @@ static int vangogh_get_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -512,21 +500,17 @@ static bool vangogh_is_dpm_running(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; int ret = 0; - uint32_t feature_mask[2]; uint64_t feature_enabled; /* we need to re-init after suspend so return false */ if (adev->in_suspend) return false; - ret = smu_cmn_get_enabled_32_bits_mask(smu, feature_mask, 2); + ret = smu_cmn_get_enabled_mask(smu, &feature_enabled); if (ret) return false; - feature_enabled = (unsigned long)((uint64_t)feature_mask[0] | - ((uint64_t)feature_mask[1] << 32)); - return !!(feature_enabled & SMC_DPM_FEATURE); } @@ -1506,7 +1490,6 @@ static int vangogh_read_sensor(struct smu_context *smu, if (!data || !size) return -EINVAL; - mutex_lock(&smu->sensor_lock); switch (sensor) { case AMDGPU_PP_SENSOR_GPU_LOAD: ret = vangogh_common_get_smu_metrics_data(smu, @@ -1568,7 +1551,6 @@ static int vangogh_read_sensor(struct smu_context *smu, ret = -EOPNOTSUPP; break; } - mutex_unlock(&smu->sensor_lock); return ret; } @@ -1965,30 +1947,13 @@ static int vangogh_get_dpm_clock_table(struct smu_context *smu, struct dpm_clock static int vangogh_system_features_control(struct smu_context *smu, bool en) { struct amdgpu_device *adev = smu->adev; - struct smu_feature *feature = &smu->smu_feature; - uint32_t feature_mask[2]; int ret = 0; if (adev->pm.fw_version >= 0x43f1700 && !en) ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_RlcPowerNotify, RLC_STATUS_OFF, NULL); - bitmap_zero(feature->enabled, feature->feature_num); - bitmap_zero(feature->supported, feature->feature_num); - - if (!en) - return ret; - - ret = smu_cmn_get_enabled_32_bits_mask(smu, feature_mask, 2); - if (ret) - return ret; - - bitmap_copy(feature->enabled, (unsigned long *)&feature_mask, - feature->feature_num); - bitmap_copy(feature->supported, (unsigned long *)&feature_mask, - feature->feature_num); - - return 0; + return ret; } static int vangogh_post_smu_init(struct smu_context *smu) @@ -2003,7 +1968,7 @@ static int vangogh_post_smu_init(struct smu_context *smu) adev->gfx.config.max_sh_per_se * adev->gfx.config.max_shader_engines; /* allow message will be sent after enable message on Vangogh*/ - if (smu_cmn_feature_is_supported(smu, SMU_FEATURE_DPM_GFXCLK_BIT) && + if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_GFXCLK_BIT) && (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG)) { ret = smu_cmn_send_smc_msg(smu, SMU_MSG_EnableGfxOff, NULL); if (ret) { @@ -2196,7 +2161,7 @@ static const struct pptable_funcs vangogh_ppt_funcs = { .dpm_set_jpeg_enable = vangogh_dpm_set_jpeg_enable, .is_dpm_running = vangogh_is_dpm_running, .read_sensor = vangogh_read_sensor, - .get_enabled_mask = smu_cmn_get_enabled_32_bits_mask, + .get_enabled_mask = smu_cmn_get_enabled_mask, .get_pp_feature_mask = smu_cmn_get_pp_feature_mask, .set_watermarks_table = vangogh_set_watermarks_table, .set_driver_table_location = smu_v11_0_set_driver_table_location, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c index 25c4b135f830..e99e7b2bf25b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c @@ -1128,15 +1128,11 @@ static int renoir_get_smu_metrics_data(struct smu_context *smu, SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_AVERAGE_GFXCLK: @@ -1201,8 +1197,6 @@ static int renoir_get_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -1215,7 +1209,6 @@ static int renoir_read_sensor(struct smu_context *smu, if (!data || !size) return -EINVAL; - mutex_lock(&smu->sensor_lock); switch (sensor) { case AMDGPU_PP_SENSOR_GPU_LOAD: ret = renoir_get_smu_metrics_data(smu, @@ -1283,7 +1276,6 @@ static int renoir_read_sensor(struct smu_context *smu, ret = -EOPNOTSUPP; break; } - mutex_unlock(&smu->sensor_lock); return ret; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c index 9c91e79c955f..56a02bc60cee 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c @@ -74,15 +74,15 @@ int smu_v12_0_check_fw_version(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; uint32_t if_version = 0xff, smu_version = 0xff; - uint16_t smu_major; - uint8_t smu_minor, smu_debug; + uint8_t smu_program, smu_major, smu_minor, smu_debug; int ret = 0; ret = smu_cmn_get_smc_version(smu, &if_version, &smu_version); if (ret) return ret; - smu_major = (smu_version >> 16) & 0xffff; + smu_program = (smu_version >> 24) & 0xff; + smu_major = (smu_version >> 16) & 0xff; smu_minor = (smu_version >> 8) & 0xff; smu_debug = (smu_version >> 0) & 0xff; if (smu->is_apu) @@ -98,9 +98,9 @@ int smu_v12_0_check_fw_version(struct smu_context *smu) */ if (if_version != smu->smc_driver_if_version) { dev_info(smu->adev->dev, "smu driver if version = 0x%08x, smu fw if version = 0x%08x, " - "smu fw version = 0x%08x (%d.%d.%d)\n", + "smu fw program = %d, smu fw version = 0x%08x (%d.%d.%d)\n", smu->smc_driver_if_version, if_version, - smu_version, smu_major, smu_minor, smu_debug); + smu_program, smu_version, smu_major, smu_minor, smu_debug); dev_warn(smu->adev->dev, "SMU driver if version not matched\n"); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index 4885c4ae78b7..890acc4e2cb8 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -25,6 +25,7 @@ #include <linux/firmware.h> #include "amdgpu.h" +#include "amdgpu_dpm.h" #include "amdgpu_smu.h" #include "atomfirmware.h" #include "amdgpu_atomfirmware.h" @@ -33,7 +34,6 @@ #include "smu13_driver_if_aldebaran.h" #include "soc15_common.h" #include "atom.h" -#include "power_state.h" #include "aldebaran_ppt.h" #include "smu_v13_0_pptable.h" #include "aldebaran_ppsmc.h" @@ -57,8 +57,6 @@ #undef pr_info #undef pr_debug -#define to_amdgpu_device(x) (container_of(x, struct amdgpu_device, pm.smu_i2c)) - #define ALDEBARAN_FEA_MAP(smu_feature, aldebaran_feature) \ [smu_feature] = {1, (aldebaran_feature)} @@ -572,15 +570,11 @@ static int aldebaran_get_smu_metrics_data(struct smu_context *smu, SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, - NULL, - false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, + NULL, + false); + if (ret) return ret; - } switch (member) { case METRICS_CURR_GFXCLK: @@ -654,8 +648,6 @@ static int aldebaran_get_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -1148,7 +1140,6 @@ static int aldebaran_read_sensor(struct smu_context *smu, if (!data || !size) return -EINVAL; - mutex_lock(&smu->sensor_lock); switch (sensor) { case AMDGPU_PP_SENSOR_MEM_LOAD: case AMDGPU_PP_SENSOR_GPU_LOAD: @@ -1187,7 +1178,6 @@ static int aldebaran_read_sensor(struct smu_context *smu, ret = -EOPNOTSUPP; break; } - mutex_unlock(&smu->sensor_lock); return ret; } @@ -1460,32 +1450,34 @@ static int aldebaran_usr_edit_dpm_table(struct smu_context *smu, enum PP_OD_DPM_ static bool aldebaran_is_dpm_running(struct smu_context *smu) { int ret; - uint32_t feature_mask[2]; - unsigned long feature_enabled; + uint64_t feature_enabled; - ret = smu_cmn_get_enabled_mask(smu, feature_mask, 2); + ret = smu_cmn_get_enabled_mask(smu, &feature_enabled); if (ret) return false; - feature_enabled = (unsigned long)((uint64_t)feature_mask[0] | - ((uint64_t)feature_mask[1] << 32)); return !!(feature_enabled & SMC_DPM_FEATURE); } static int aldebaran_i2c_xfer(struct i2c_adapter *i2c_adap, struct i2c_msg *msg, int num_msgs) { - struct amdgpu_device *adev = to_amdgpu_device(i2c_adap); - struct smu_table_context *smu_table = &adev->smu.smu_table; + struct amdgpu_smu_i2c_bus *smu_i2c = i2c_get_adapdata(i2c_adap); + struct amdgpu_device *adev = smu_i2c->adev; + struct smu_context *smu = adev->powerplay.pp_handle; + struct smu_table_context *smu_table = &smu->smu_table; struct smu_table *table = &smu_table->driver_table; SwI2cRequest_t *req, *res = (SwI2cRequest_t *)table->cpu_addr; int i, j, r, c; u16 dir; + if (!adev->pm.dpm_enabled) + return -EBUSY; + req = kzalloc(sizeof(*req), GFP_KERNEL); if (!req) return -ENOMEM; - req->I2CcontrollerPort = 0; + req->I2CcontrollerPort = smu_i2c->port; req->I2CSpeed = I2C_SPEED_FAST_400K; req->SlaveAddress = msg[0].addr << 1; /* wants an 8-bit address */ dir = msg[0].flags & I2C_M_RD; @@ -1521,9 +1513,9 @@ static int aldebaran_i2c_xfer(struct i2c_adapter *i2c_adap, } } } - mutex_lock(&adev->smu.mutex); - r = smu_cmn_update_table(&adev->smu, SMU_TABLE_I2C_COMMANDS, 0, req, true); - mutex_unlock(&adev->smu.mutex); + mutex_lock(&adev->pm.mutex); + r = smu_cmn_update_table(smu, SMU_TABLE_I2C_COMMANDS, 0, req, true); + mutex_unlock(&adev->pm.mutex); if (r) goto fail; @@ -1563,28 +1555,53 @@ static const struct i2c_adapter_quirks aldebaran_i2c_control_quirks = { .max_comb_2nd_msg_len = MAX_SW_I2C_COMMANDS - 2, }; -static int aldebaran_i2c_control_init(struct smu_context *smu, struct i2c_adapter *control) +static int aldebaran_i2c_control_init(struct smu_context *smu) { - struct amdgpu_device *adev = to_amdgpu_device(control); + struct amdgpu_device *adev = smu->adev; + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[0]; + struct i2c_adapter *control = &smu_i2c->adapter; int res; + smu_i2c->adev = adev; + smu_i2c->port = 0; + mutex_init(&smu_i2c->mutex); control->owner = THIS_MODULE; control->class = I2C_CLASS_SPD; control->dev.parent = &adev->pdev->dev; control->algo = &aldebaran_i2c_algo; - snprintf(control->name, sizeof(control->name), "AMDGPU SMU"); + snprintf(control->name, sizeof(control->name), "AMDGPU SMU 0"); control->quirks = &aldebaran_i2c_control_quirks; + i2c_set_adapdata(control, smu_i2c); res = i2c_add_adapter(control); - if (res) + if (res) { DRM_ERROR("Failed to register hw i2c, err: %d\n", res); + goto Out_err; + } + + adev->pm.ras_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter; + adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter; + + return 0; +Out_err: + i2c_del_adapter(control); return res; } -static void aldebaran_i2c_control_fini(struct smu_context *smu, struct i2c_adapter *control) +static void aldebaran_i2c_control_fini(struct smu_context *smu) { - i2c_del_adapter(control); + struct amdgpu_device *adev = smu->adev; + int i; + + for (i = 0; i < MAX_SMU_I2C_BUSES; i++) { + struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i]; + struct i2c_adapter *control = &smu_i2c->adapter; + + i2c_del_adapter(control); + } + adev->pm.ras_eeprom_i2c_bus = NULL; + adev->pm.fru_eeprom_i2c_bus = NULL; } static void aldebaran_get_unique_id(struct smu_context *smu) @@ -1594,17 +1611,14 @@ static void aldebaran_get_unique_id(struct smu_context *smu) uint32_t upper32 = 0, lower32 = 0; int ret; - mutex_lock(&smu->metrics_lock); - ret = smu_cmn_get_metrics_table_locked(smu, NULL, false); + ret = smu_cmn_get_metrics_table(smu, NULL, false); if (ret) - goto out_unlock; + goto out; upper32 = metrics->PublicSerialNumUpper32; lower32 = metrics->PublicSerialNumLower32; -out_unlock: - mutex_unlock(&smu->metrics_lock); - +out: adev->unique_id = ((uint64_t)upper32 << 32) | lower32; if (adev->serial[0] == '\0') sprintf(adev->serial, "%016llx", adev->unique_id); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index b54790d3483e..f0ab1dc3ca59 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -198,15 +198,15 @@ int smu_v13_0_check_fw_version(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; uint32_t if_version = 0xff, smu_version = 0xff; - uint16_t smu_major; - uint8_t smu_minor, smu_debug; + uint8_t smu_program, smu_major, smu_minor, smu_debug; int ret = 0; ret = smu_cmn_get_smc_version(smu, &if_version, &smu_version); if (ret) return ret; - smu_major = (smu_version >> 16) & 0xffff; + smu_program = (smu_version >> 24) & 0xff; + smu_major = (smu_version >> 16) & 0xff; smu_minor = (smu_version >> 8) & 0xff; smu_debug = (smu_version >> 0) & 0xff; if (smu->is_apu) @@ -229,8 +229,8 @@ int smu_v13_0_check_fw_version(struct smu_context *smu) /* only for dGPU w/ SMU13*/ if (adev->pm.fw) - dev_dbg(adev->dev, "smu fw reported version = 0x%08x (%d.%d.%d)\n", - smu_version, smu_major, smu_minor, smu_debug); + dev_dbg(smu->adev->dev, "smu fw reported program %d, version = 0x%08x (%d.%d.%d)\n", + smu_program, smu_version, smu_major, smu_minor, smu_debug); /* * 1. if_version mismatch is not critical as our fw is designed @@ -242,9 +242,9 @@ int smu_v13_0_check_fw_version(struct smu_context *smu) */ if (if_version != smu->smc_driver_if_version) { dev_info(adev->dev, "smu driver if version = 0x%08x, smu fw if version = 0x%08x, " - "smu fw version = 0x%08x (%d.%d.%d)\n", + "smu fw program = %d, smu fw version = 0x%08x (%d.%d.%d)\n", smu->smc_driver_if_version, if_version, - smu_version, smu_major, smu_minor, smu_debug); + smu_program, smu_version, smu_major, smu_minor, smu_debug); dev_warn(adev->dev, "SMU driver if version not matched\n"); } @@ -722,25 +722,21 @@ int smu_v13_0_set_allowed_mask(struct smu_context *smu) int ret = 0; uint32_t feature_mask[2]; - mutex_lock(&feature->mutex); - if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || feature->feature_num < 64) - goto failed; + if (bitmap_empty(feature->allowed, SMU_FEATURE_MAX) || + feature->feature_num < 64) + return -EINVAL; bitmap_copy((unsigned long *)feature_mask, feature->allowed, 64); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskHigh, feature_mask[1], NULL); if (ret) - goto failed; - - ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetAllowedFeaturesMaskLow, - feature_mask[0], NULL); - if (ret) - goto failed; + return ret; -failed: - mutex_unlock(&feature->mutex); - return ret; + return smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_SetAllowedFeaturesMaskLow, + feature_mask[0], + NULL); } int smu_v13_0_gfx_off_control(struct smu_context *smu, bool enable) @@ -768,30 +764,8 @@ int smu_v13_0_gfx_off_control(struct smu_context *smu, bool enable) int smu_v13_0_system_features_control(struct smu_context *smu, bool en) { - struct smu_feature *feature = &smu->smu_feature; - uint32_t feature_mask[2]; - int ret = 0; - - ret = smu_cmn_send_smc_msg(smu, (en ? SMU_MSG_EnableAllSmuFeatures : - SMU_MSG_DisableAllSmuFeatures), NULL); - if (ret) - return ret; - - bitmap_zero(feature->enabled, feature->feature_num); - bitmap_zero(feature->supported, feature->feature_num); - - if (en) { - ret = smu_cmn_get_enabled_mask(smu, feature_mask, 2); - if (ret) - return ret; - - bitmap_copy(feature->enabled, (unsigned long *)&feature_mask, - feature->feature_num); - bitmap_copy(feature->supported, (unsigned long *)&feature_mask, - feature->feature_num); - } - - return ret; + return smu_cmn_send_smc_msg(smu, (en ? SMU_MSG_EnableAllSmuFeatures : + SMU_MSG_DisableAllSmuFeatures), NULL); } int smu_v13_0_notify_display_change(struct smu_context *smu) @@ -1200,7 +1174,7 @@ static int smu_v13_0_set_irq_state(struct amdgpu_device *adev, unsigned tyep, enum amdgpu_interrupt_state state) { - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; uint32_t low, high; uint32_t val = 0; @@ -1275,7 +1249,7 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - struct smu_context *smu = &adev->smu; + struct smu_context *smu = adev->powerplay.pp_handle; uint32_t client_id = entry->client_id; uint32_t src_id = entry->src_id; /* @@ -1321,11 +1295,11 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev, switch (ctxid) { case 0x3: dev_dbg(adev->dev, "Switched to AC mode!\n"); - smu_v13_0_ack_ac_dc_interrupt(&adev->smu); + smu_v13_0_ack_ac_dc_interrupt(smu); break; case 0x4: dev_dbg(adev->dev, "Switched to DC mode!\n"); - smu_v13_0_ack_ac_dc_interrupt(&adev->smu); + smu_v13_0_ack_ac_dc_interrupt(smu); break; case 0x7: /* @@ -1533,7 +1507,6 @@ int smu_v13_0_set_soft_freq_limited_range(struct smu_context *smu, uint32_t min, uint32_t max) { - struct amdgpu_device *adev = smu->adev; int ret = 0, clk_id = 0; uint32_t param; @@ -1546,9 +1519,6 @@ int smu_v13_0_set_soft_freq_limited_range(struct smu_context *smu, if (clk_id < 0) return clk_id; - if (clk_type == SMU_GFXCLK) - amdgpu_gfx_off_ctrl(adev, false); - if (max > 0) { param = (uint32_t)((clk_id << 16) | (max & 0xffff)); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxByFreq, @@ -1566,9 +1536,6 @@ int smu_v13_0_set_soft_freq_limited_range(struct smu_context *smu, } out: - if (clk_type == SMU_GFXCLK) - amdgpu_gfx_off_ctrl(adev, true); - return ret; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c index 0bc84b709a93..e2d099409123 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c @@ -195,30 +195,13 @@ static int yellow_carp_fini_smc_tables(struct smu_context *smu) static int yellow_carp_system_features_control(struct smu_context *smu, bool en) { - struct smu_feature *feature = &smu->smu_feature; struct amdgpu_device *adev = smu->adev; - uint32_t feature_mask[2]; int ret = 0; if (!en && !adev->in_s0ix) ret = smu_cmn_send_smc_msg(smu, SMU_MSG_PrepareMp1ForUnload, NULL); - bitmap_zero(feature->enabled, feature->feature_num); - bitmap_zero(feature->supported, feature->feature_num); - - if (!en) - return ret; - - ret = smu_cmn_get_enabled_32_bits_mask(smu, feature_mask, 2); - if (ret) - return ret; - - bitmap_copy(feature->enabled, (unsigned long *)&feature_mask, - feature->feature_num); - bitmap_copy(feature->supported, (unsigned long *)&feature_mask, - feature->feature_num); - - return 0; + return ret; } static int yellow_carp_dpm_set_vcn_enable(struct smu_context *smu, bool enable) @@ -255,16 +238,13 @@ static int yellow_carp_dpm_set_jpeg_enable(struct smu_context *smu, bool enable) static bool yellow_carp_is_dpm_running(struct smu_context *smu) { int ret = 0; - uint32_t feature_mask[2]; uint64_t feature_enabled; - ret = smu_cmn_get_enabled_32_bits_mask(smu, feature_mask, 2); + ret = smu_cmn_get_enabled_mask(smu, &feature_enabled); if (ret) return false; - feature_enabled = (uint64_t)feature_mask[1] << 32 | feature_mask[0]; - return !!(feature_enabled & SMC_DPM_FEATURE); } @@ -305,13 +285,9 @@ static int yellow_carp_get_smu_metrics_data(struct smu_context *smu, SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; - mutex_lock(&smu->metrics_lock); - - ret = smu_cmn_get_metrics_table_locked(smu, NULL, false); - if (ret) { - mutex_unlock(&smu->metrics_lock); + ret = smu_cmn_get_metrics_table(smu, NULL, false); + if (ret) return ret; - } switch (member) { case METRICS_AVERAGE_GFXCLK: @@ -382,8 +358,6 @@ static int yellow_carp_get_smu_metrics_data(struct smu_context *smu, break; } - mutex_unlock(&smu->metrics_lock); - return ret; } @@ -396,7 +370,6 @@ static int yellow_carp_read_sensor(struct smu_context *smu, if (!data || !size) return -EINVAL; - mutex_lock(&smu->sensor_lock); switch (sensor) { case AMDGPU_PP_SENSOR_GPU_LOAD: ret = yellow_carp_get_smu_metrics_data(smu, @@ -464,7 +437,6 @@ static int yellow_carp_read_sensor(struct smu_context *smu, ret = -EOPNOTSUPP; break; } - mutex_unlock(&smu->sensor_lock); return ret; } @@ -1177,7 +1149,7 @@ static const struct pptable_funcs yellow_carp_ppt_funcs = { .is_dpm_running = yellow_carp_is_dpm_running, .set_watermarks_table = yellow_carp_set_watermarks_table, .get_gpu_metrics = yellow_carp_get_gpu_metrics, - .get_enabled_mask = smu_cmn_get_enabled_32_bits_mask, + .get_enabled_mask = smu_cmn_get_enabled_mask, .get_pp_feature_mask = smu_cmn_get_pp_feature_mask, .set_driver_table_location = smu_v13_0_set_driver_table_location, .gfx_off_control = smu_v13_0_gfx_off_control, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index ee1a312fd497..ea6c2dab5ecc 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -481,7 +481,6 @@ int smu_cmn_feature_is_supported(struct smu_context *smu, { struct smu_feature *feature = &smu->smu_feature; int feature_id; - int ret = 0; feature_id = smu_cmn_to_asic_specific_index(smu, CMN2ASIC_MAPPING_FEATURE, @@ -491,22 +490,27 @@ int smu_cmn_feature_is_supported(struct smu_context *smu, WARN_ON(feature_id > feature->feature_num); - mutex_lock(&feature->mutex); - ret = test_bit(feature_id, feature->supported); - mutex_unlock(&feature->mutex); - - return ret; + return test_bit(feature_id, feature->supported); } int smu_cmn_feature_is_enabled(struct smu_context *smu, enum smu_feature_mask mask) { - struct smu_feature *feature = &smu->smu_feature; struct amdgpu_device *adev = smu->adev; + uint64_t enabled_features; int feature_id; - int ret = 0; - if (smu->is_apu && adev->family < AMDGPU_FAMILY_VGH) + if (smu_cmn_get_enabled_mask(smu, &enabled_features)) { + dev_err(adev->dev, "Failed to retrieve enabled ppfeatures!\n"); + return 0; + } + + /* + * For Renoir and Cyan Skillfish, they are assumed to have all features + * enabled. Also considering they have no feature_map available, the + * check here can avoid unwanted feature_map check below. + */ + if (enabled_features == ULLONG_MAX) return 1; feature_id = smu_cmn_to_asic_specific_index(smu, @@ -515,13 +519,7 @@ int smu_cmn_feature_is_enabled(struct smu_context *smu, if (feature_id < 0) return 0; - WARN_ON(feature_id > feature->feature_num); - - mutex_lock(&feature->mutex); - ret = test_bit(feature_id, feature->enabled); - mutex_unlock(&feature->mutex); - - return ret; + return test_bit(feature_id, (unsigned long *)&enabled_features); } bool smu_cmn_clk_dpm_is_enabled(struct smu_context *smu, @@ -552,70 +550,61 @@ bool smu_cmn_clk_dpm_is_enabled(struct smu_context *smu, } int smu_cmn_get_enabled_mask(struct smu_context *smu, - uint32_t *feature_mask, - uint32_t num) + uint64_t *feature_mask) { - uint32_t feature_mask_high = 0, feature_mask_low = 0; - struct smu_feature *feature = &smu->smu_feature; - int ret = 0; - - if (!feature_mask || num < 2) - return -EINVAL; - - if (bitmap_empty(feature->enabled, feature->feature_num)) { - ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesHigh, &feature_mask_high); - if (ret) - return ret; - - ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetEnabledSmuFeaturesLow, &feature_mask_low); - if (ret) - return ret; - - feature_mask[0] = feature_mask_low; - feature_mask[1] = feature_mask_high; - } else { - bitmap_copy((unsigned long *)feature_mask, feature->enabled, - feature->feature_num); - } - - return ret; -} - -int smu_cmn_get_enabled_32_bits_mask(struct smu_context *smu, - uint32_t *feature_mask, - uint32_t num) -{ - uint32_t feature_mask_en_low = 0; - uint32_t feature_mask_en_high = 0; - struct smu_feature *feature = &smu->smu_feature; + struct amdgpu_device *adev = smu->adev; + uint32_t *feature_mask_high; + uint32_t *feature_mask_low; int ret = 0; - if (!feature_mask || num < 2) + if (!feature_mask) return -EINVAL; - if (bitmap_empty(feature->enabled, feature->feature_num)) { - ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_GetEnabledSmuFeatures, 0, - &feature_mask_en_low); + feature_mask_low = &((uint32_t *)feature_mask)[0]; + feature_mask_high = &((uint32_t *)feature_mask)[1]; + switch (adev->ip_versions[MP1_HWIP][0]) { + /* For Vangogh and Yellow Carp */ + case IP_VERSION(11, 5, 0): + case IP_VERSION(13, 0, 1): + case IP_VERSION(13, 0, 3): + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_GetEnabledSmuFeatures, + 0, + feature_mask_low); if (ret) return ret; - ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_GetEnabledSmuFeatures, 1, - &feature_mask_en_high); - + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_GetEnabledSmuFeatures, + 1, + feature_mask_high); + break; + /* + * For Cyan Skillfish and Renoir, there is no interface provided by PMFW + * to retrieve the enabled features. So, we assume all features are enabled. + * TODO: add other APU ASICs which suffer from the same issue here + */ + case IP_VERSION(11, 0, 8): + case IP_VERSION(12, 0, 0): + case IP_VERSION(12, 0, 1): + memset(feature_mask, 0xff, sizeof(*feature_mask)); + break; + /* other dGPU ASICs */ + default: + ret = smu_cmn_send_smc_msg(smu, + SMU_MSG_GetEnabledSmuFeaturesHigh, + feature_mask_high); if (ret) return ret; - feature_mask[0] = feature_mask_en_low; - feature_mask[1] = feature_mask_en_high; - - } else { - bitmap_copy((unsigned long *)feature_mask, feature->enabled, - feature->feature_num); + ret = smu_cmn_send_smc_msg(smu, + SMU_MSG_GetEnabledSmuFeaturesLow, + feature_mask_low); + break; } return ret; - } uint64_t smu_cmn_get_indep_throttler_status( @@ -635,7 +624,6 @@ int smu_cmn_feature_update_enable_state(struct smu_context *smu, uint64_t feature_mask, bool enabled) { - struct smu_feature *feature = &smu->smu_feature; int ret = 0; if (enabled) { @@ -649,8 +637,6 @@ int smu_cmn_feature_update_enable_state(struct smu_context *smu, SMU_MSG_EnableSmuFeaturesHigh, upper_32_bits(feature_mask), NULL); - if (ret) - return ret; } else { ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_DisableSmuFeaturesLow, @@ -662,19 +648,8 @@ int smu_cmn_feature_update_enable_state(struct smu_context *smu, SMU_MSG_DisableSmuFeaturesHigh, upper_32_bits(feature_mask), NULL); - if (ret) - return ret; } - mutex_lock(&feature->mutex); - if (enabled) - bitmap_or(feature->enabled, feature->enabled, - (unsigned long *)(&feature_mask), SMU_FEATURE_MAX); - else - bitmap_andnot(feature->enabled, feature->enabled, - (unsigned long *)(&feature_mask), SMU_FEATURE_MAX); - mutex_unlock(&feature->mutex); - return ret; } @@ -682,7 +657,6 @@ int smu_cmn_feature_set_enabled(struct smu_context *smu, enum smu_feature_mask mask, bool enable) { - struct smu_feature *feature = &smu->smu_feature; int feature_id; feature_id = smu_cmn_to_asic_specific_index(smu, @@ -691,8 +665,6 @@ int smu_cmn_feature_set_enabled(struct smu_context *smu, if (feature_id < 0) return -EINVAL; - WARN_ON(feature_id > feature->feature_num); - return smu_cmn_feature_update_enable_state(smu, 1ULL << feature_id, enable); @@ -715,29 +687,21 @@ static const char *smu_get_feature_name(struct smu_context *smu, size_t smu_cmn_get_pp_feature_mask(struct smu_context *smu, char *buf) { - uint32_t feature_mask[2] = { 0 }; + uint64_t feature_mask; int feature_index = 0; uint32_t count = 0; int8_t sort_feature[SMU_FEATURE_COUNT]; size_t size = 0; int ret = 0, i; + int feature_id; - if (!smu->is_apu) { - ret = smu_cmn_get_enabled_mask(smu, - feature_mask, - 2); - if (ret) - return 0; - } else { - ret = smu_cmn_get_enabled_32_bits_mask(smu, - feature_mask, - 2); - if (ret) - return 0; - } + ret = smu_cmn_get_enabled_mask(smu, + &feature_mask); + if (ret) + return 0; size = sysfs_emit_at(buf, size, "features high: 0x%08x low: 0x%08x\n", - feature_mask[1], feature_mask[0]); + upper_32_bits(feature_mask), lower_32_bits(feature_mask)); memset(sort_feature, -1, sizeof(sort_feature)); @@ -758,11 +722,18 @@ size_t smu_cmn_get_pp_feature_mask(struct smu_context *smu, if (sort_feature[i] < 0) continue; + /* convert to asic spcific feature ID */ + feature_id = smu_cmn_to_asic_specific_index(smu, + CMN2ASIC_MAPPING_FEATURE, + sort_feature[i]); + if (feature_id < 0) + continue; + size += sysfs_emit_at(buf, size, "%02d. %-20s (%2d) : %s\n", count++, smu_get_feature_name(smu, sort_feature[i]), i, - !!smu_cmn_feature_is_enabled(smu, sort_feature[i]) ? + !!test_bit(feature_id, (unsigned long *)&feature_mask) ? "enabled" : "disabled"); } @@ -773,22 +744,17 @@ int smu_cmn_set_pp_feature_mask(struct smu_context *smu, uint64_t new_mask) { int ret = 0; - uint32_t feature_mask[2] = { 0 }; + uint64_t feature_mask; uint64_t feature_2_enabled = 0; uint64_t feature_2_disabled = 0; - uint64_t feature_enables = 0; ret = smu_cmn_get_enabled_mask(smu, - feature_mask, - 2); + &feature_mask); if (ret) return ret; - feature_enables = ((uint64_t)feature_mask[1] << 32 | - (uint64_t)feature_mask[0]); - - feature_2_enabled = ~feature_enables & new_mask; - feature_2_disabled = feature_enables & ~new_mask; + feature_2_enabled = ~feature_mask & new_mask; + feature_2_disabled = feature_mask & ~new_mask; if (feature_2_enabled) { ret = smu_cmn_feature_update_enable_state(smu, @@ -814,9 +780,6 @@ int smu_cmn_set_pp_feature_mask(struct smu_context *smu, * @mask * * @smu: smu_context pointer - * @no_hw_disablement: whether real dpm disablement should be performed - * true: update the cache(about dpm enablement state) only - * false: real dpm disablement plus cache update * @mask: the dpm feature which should not be disabled * SMU_FEATURE_COUNT: no exception, all dpm features * to disable @@ -825,10 +788,8 @@ int smu_cmn_set_pp_feature_mask(struct smu_context *smu, * 0 on success or a negative error code on failure. */ int smu_cmn_disable_all_features_with_exception(struct smu_context *smu, - bool no_hw_disablement, enum smu_feature_mask mask) { - struct smu_feature *feature = &smu->smu_feature; uint64_t features_to_disable = U64_MAX; int skipped_feature_id; @@ -842,18 +803,9 @@ int smu_cmn_disable_all_features_with_exception(struct smu_context *smu, features_to_disable &= ~(1ULL << skipped_feature_id); } - if (no_hw_disablement) { - mutex_lock(&feature->mutex); - bitmap_andnot(feature->enabled, feature->enabled, - (unsigned long *)(&features_to_disable), SMU_FEATURE_MAX); - mutex_unlock(&feature->mutex); - - return 0; - } else { - return smu_cmn_feature_update_enable_state(smu, - features_to_disable, - 0); - } + return smu_cmn_feature_update_enable_state(smu, + features_to_disable, + 0); } int smu_cmn_get_smc_version(struct smu_context *smu, @@ -964,9 +916,9 @@ int smu_cmn_write_pptable(struct smu_context *smu) true); } -int smu_cmn_get_metrics_table_locked(struct smu_context *smu, - void *metrics_table, - bool bypass_cache) +int smu_cmn_get_metrics_table(struct smu_context *smu, + void *metrics_table, + bool bypass_cache) { struct smu_table_context *smu_table= &smu->smu_table; uint32_t table_size = @@ -994,21 +946,6 @@ int smu_cmn_get_metrics_table_locked(struct smu_context *smu, return 0; } -int smu_cmn_get_metrics_table(struct smu_context *smu, - void *metrics_table, - bool bypass_cache) -{ - int ret = 0; - - mutex_lock(&smu->metrics_lock); - ret = smu_cmn_get_metrics_table_locked(smu, - metrics_table, - bypass_cache); - mutex_unlock(&smu->metrics_lock); - - return ret; -} - void smu_cmn_init_soft_gpu_metrics(void *table, uint8_t frev, uint8_t crev) { struct metrics_table_header *header = (struct metrics_table_header *)table; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h index beea03810bca..a4c593ed8b03 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h @@ -26,6 +26,10 @@ #include "amdgpu_smu.h" #if defined(SWSMU_CODE_LAYER_L2) || defined(SWSMU_CODE_LAYER_L3) || defined(SWSMU_CODE_LAYER_L4) + +#define FDO_PWM_MODE_STATIC 1 +#define FDO_PWM_MODE_STATIC_RPM 5 + int smu_cmn_send_msg_without_waiting(struct smu_context *smu, uint16_t msg_index, uint32_t param); @@ -54,12 +58,7 @@ bool smu_cmn_clk_dpm_is_enabled(struct smu_context *smu, enum smu_clk_type clk_type); int smu_cmn_get_enabled_mask(struct smu_context *smu, - uint32_t *feature_mask, - uint32_t num); - -int smu_cmn_get_enabled_32_bits_mask(struct smu_context *smu, - uint32_t *feature_mask, - uint32_t num); + uint64_t *feature_mask); uint64_t smu_cmn_get_indep_throttler_status( const unsigned long dep_status, @@ -80,7 +79,6 @@ int smu_cmn_set_pp_feature_mask(struct smu_context *smu, uint64_t new_mask); int smu_cmn_disable_all_features_with_exception(struct smu_context *smu, - bool no_hw_disablement, enum smu_feature_mask mask); int smu_cmn_get_smc_version(struct smu_context *smu, @@ -97,10 +95,6 @@ int smu_cmn_write_watermarks_table(struct smu_context *smu); int smu_cmn_write_pptable(struct smu_context *smu); -int smu_cmn_get_metrics_table_locked(struct smu_context *smu, - void *metrics_table, - bool bypass_cache); - int smu_cmn_get_metrics_table(struct smu_context *smu, void *metrics_table, bool bypass_cache); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h index 59f9cfff3d61..15bcf72b8e56 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h @@ -55,9 +55,9 @@ #define smu_send_smc_msg(smu, msg, read_arg) smu_ppt_funcs(send_smc_msg, 0, smu, msg, read_arg) #define smu_init_display_count(smu, count) smu_ppt_funcs(init_display_count, 0, smu, count) #define smu_feature_set_allowed_mask(smu) smu_ppt_funcs(set_allowed_mask, 0, smu) -#define smu_feature_get_enabled_mask(smu, mask, num) smu_ppt_funcs(get_enabled_mask, 0, smu, mask, num) +#define smu_feature_get_enabled_mask(smu, mask) smu_ppt_funcs(get_enabled_mask, -EOPNOTSUPP, smu, mask) #define smu_feature_is_enabled(smu, mask) smu_ppt_funcs(feature_is_enabled, 0, smu, mask) -#define smu_disable_all_features_with_exception(smu, no_hw_disablement, mask) smu_ppt_funcs(disable_all_features_with_exception, 0, smu, no_hw_disablement, mask) +#define smu_disable_all_features_with_exception(smu, mask) smu_ppt_funcs(disable_all_features_with_exception, 0, smu, mask) #define smu_is_dpm_running(smu) smu_ppt_funcs(is_dpm_running, 0 , smu) #define smu_notify_display_change(smu) smu_ppt_funcs(notify_display_change, 0, smu) #define smu_populate_umd_state_clk(smu) smu_ppt_funcs(populate_umd_state_clk, 0, smu) @@ -78,8 +78,8 @@ #define smu_dump_pptable(smu) smu_ppt_funcs(dump_pptable, 0, smu) #define smu_update_pcie_parameters(smu, pcie_gen_cap, pcie_width_cap) smu_ppt_funcs(update_pcie_parameters, 0, smu, pcie_gen_cap, pcie_width_cap) #define smu_set_power_source(smu, power_src) smu_ppt_funcs(set_power_source, 0, smu, power_src) -#define smu_i2c_init(smu, control) smu_ppt_funcs(i2c_init, 0, smu, control) -#define smu_i2c_fini(smu, control) smu_ppt_funcs(i2c_fini, 0, smu, control) +#define smu_i2c_init(smu) smu_ppt_funcs(i2c_init, 0, smu) +#define smu_i2c_fini(smu) smu_ppt_funcs(i2c_fini, 0, smu) #define smu_get_unique_id(smu) smu_ppt_funcs(get_unique_id, 0, smu) #define smu_log_thermal_throttling(smu) smu_ppt_funcs(log_thermal_throttling_event, 0, smu) #define smu_get_asic_power_limits(smu, current, default, max) smu_ppt_funcs(get_power_limit, 0, smu, current, default, max) diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c b/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c index 8a02ade369db..42510fdea27e 100644 --- a/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c +++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c @@ -1078,11 +1078,11 @@ static void d71_improc_update(struct komeda_component *c, mask |= IPS_CTRL_YUV | IPS_CTRL_CHD422 | IPS_CTRL_CHD420; /* config color format */ - if (st->color_format == DRM_COLOR_FORMAT_YCRCB420) + if (st->color_format == DRM_COLOR_FORMAT_YCBCR420) ctrl |= IPS_CTRL_YUV | IPS_CTRL_CHD422 | IPS_CTRL_CHD420; - else if (st->color_format == DRM_COLOR_FORMAT_YCRCB422) + else if (st->color_format == DRM_COLOR_FORMAT_YCBCR422) ctrl |= IPS_CTRL_YUV | IPS_CTRL_CHD422; - else if (st->color_format == DRM_COLOR_FORMAT_YCRCB444) + else if (st->color_format == DRM_COLOR_FORMAT_YCBCR444) ctrl |= IPS_CTRL_YUV; malidp_write32_mask(reg, BLK_CONTROL, mask, ctrl); @@ -1144,11 +1144,11 @@ static int d71_improc_init(struct d71_dev *d71, improc = to_improc(c); improc->supported_color_depths = BIT(8) | BIT(10); improc->supported_color_formats = DRM_COLOR_FORMAT_RGB444 | - DRM_COLOR_FORMAT_YCRCB444 | - DRM_COLOR_FORMAT_YCRCB422; + DRM_COLOR_FORMAT_YCBCR444 | + DRM_COLOR_FORMAT_YCBCR422; value = malidp_read32(reg, BLK_INFO); if (value & IPS_INFO_CHD420) - improc->supported_color_formats |= DRM_COLOR_FORMAT_YCRCB420; + improc->supported_color_formats |= DRM_COLOR_FORMAT_YCBCR420; improc->supports_csc = true; improc->supports_gamma = true; diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c index e7933930a657..51e51ff299b7 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c @@ -9,6 +9,7 @@ #include <linux/platform_device.h> #include <linux/component.h> #include <linux/pm_runtime.h> +#include <drm/drm_module.h> #include <drm/drm_of.h> #include "komeda_dev.h" #include "komeda_kms.h" @@ -198,7 +199,7 @@ static struct platform_driver komeda_platform_driver = { }, }; -module_platform_driver(komeda_platform_driver); +drm_module_platform_driver(komeda_platform_driver); MODULE_AUTHOR("James.Qian.Wang <james.qian.wang@arm.com>"); MODULE_DESCRIPTION("Komeda KMS driver"); diff --git a/drivers/gpu/drm/arm/hdlcd_drv.c b/drivers/gpu/drm/arm/hdlcd_drv.c index 479c2422a2e0..e89ae0ec60eb 100644 --- a/drivers/gpu/drm/arm/hdlcd_drv.c +++ b/drivers/gpu/drm/arm/hdlcd_drv.c @@ -30,6 +30,7 @@ #include <drm/drm_gem_cma_helper.h> #include <drm/drm_gem_framebuffer_helper.h> #include <drm/drm_modeset_helper.h> +#include <drm/drm_module.h> #include <drm/drm_of.h> #include <drm/drm_probe_helper.h> #include <drm/drm_vblank.h> @@ -434,7 +435,7 @@ static struct platform_driver hdlcd_platform_driver = { }, }; -module_platform_driver(hdlcd_platform_driver); +drm_module_platform_driver(hdlcd_platform_driver); MODULE_AUTHOR("Liviu Dudau"); MODULE_DESCRIPTION("ARM HDLCD DRM driver"); diff --git a/drivers/gpu/drm/arm/malidp_drv.c b/drivers/gpu/drm/arm/malidp_drv.c index 78d15b04b105..d5aef21426cf 100644 --- a/drivers/gpu/drm/arm/malidp_drv.c +++ b/drivers/gpu/drm/arm/malidp_drv.c @@ -25,6 +25,7 @@ #include <drm/drm_gem_cma_helper.h> #include <drm/drm_gem_framebuffer_helper.h> #include <drm/drm_modeset_helper.h> +#include <drm/drm_module.h> #include <drm/drm_of.h> #include <drm/drm_probe_helper.h> #include <drm/drm_vblank.h> @@ -1008,7 +1009,7 @@ static struct platform_driver malidp_platform_driver = { }, }; -module_platform_driver(malidp_platform_driver); +drm_module_platform_driver(malidp_platform_driver); MODULE_AUTHOR("Liviu Dudau <Liviu.Dudau@arm.com>"); MODULE_DESCRIPTION("ARM Mali DP DRM driver"); diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c index 6d8613f6fe1c..7465c4f0156a 100644 --- a/drivers/gpu/drm/ast/ast_drv.c +++ b/drivers/gpu/drm/ast/ast_drv.c @@ -34,6 +34,7 @@ #include <drm/drm_crtc_helper.h> #include <drm/drm_drv.h> #include <drm/drm_gem_vram_helper.h> +#include <drm/drm_module.h> #include <drm/drm_probe_helper.h> #include "ast_drv.h" @@ -230,22 +231,7 @@ static struct pci_driver ast_pci_driver = { .driver.pm = &ast_pm_ops, }; -static int __init ast_init(void) -{ - if (drm_firmware_drivers_only() && ast_modeset == -1) - return -EINVAL; - - if (ast_modeset == 0) - return -EINVAL; - return pci_register_driver(&ast_pci_driver); -} -static void __exit ast_exit(void) -{ - pci_unregister_driver(&ast_pci_driver); -} - -module_init(ast_init); -module_exit(ast_exit); +drm_module_pci_driver_if_modeset(ast_pci_driver, ast_modeset); MODULE_AUTHOR(DRIVER_AUTHOR); MODULE_DESCRIPTION(DRIVER_DESC); diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 00bfa41ff7cb..9c8d56b0a41b 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -107,7 +107,7 @@ struct ast_cursor_plane { struct { struct drm_gem_vram_object *gbo; - struct dma_buf_map map; + struct iosys_map map; u64 off; } hwc[AST_DEFAULT_HWC_NUM]; diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index 79a361867955..22e9e2d3c71a 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -209,6 +209,8 @@ static int ast_detect_chip(struct drm_device *dev, bool *need_post) if (ast->chip == AST2500 && scu_rev == 0x100) /* ast2510 */ ast->support_wide_screen = true; + if (ast->chip == AST2600) /* ast2600 */ + ast->support_wide_screen = true; } break; } diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 956c8982192b..2c7115a4d81f 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -471,7 +471,10 @@ static void ast_set_color_reg(struct ast_private *ast, static void ast_set_crtthd_reg(struct ast_private *ast) { /* Set Threshold */ - if (ast->chip == AST2300 || ast->chip == AST2400 || + if (ast->chip == AST2600) { + ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa7, 0xe0); + ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa6, 0xa0); + } else if (ast->chip == AST2300 || ast->chip == AST2400 || ast->chip == AST2500) { ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa7, 0x78); ast_set_index_reg(ast, AST_IO_CRTC_PORT, 0xa6, 0x60); @@ -801,11 +804,11 @@ ast_cursor_plane_helper_atomic_update(struct drm_plane *plane, struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(new_state); struct drm_framebuffer *fb = new_state->fb; struct ast_private *ast = to_ast_private(plane->dev); - struct dma_buf_map dst_map = + struct iosys_map dst_map = ast_cursor_plane->hwc[ast_cursor_plane->next_hwc_index].map; u64 dst_off = ast_cursor_plane->hwc[ast_cursor_plane->next_hwc_index].off; - struct dma_buf_map src_map = shadow_plane_state->data[0]; + struct iosys_map src_map = shadow_plane_state->data[0]; unsigned int offset_x, offset_y; u16 x, y; u8 x_offset, y_offset; @@ -883,7 +886,7 @@ static void ast_cursor_plane_destroy(struct drm_plane *plane) struct ast_cursor_plane *ast_cursor_plane = to_ast_cursor_plane(plane); size_t i; struct drm_gem_vram_object *gbo; - struct dma_buf_map map; + struct iosys_map map; for (i = 0; i < ARRAY_SIZE(ast_cursor_plane->hwc); ++i) { gbo = ast_cursor_plane->hwc[i].gbo; @@ -910,7 +913,7 @@ static int ast_cursor_plane_init(struct ast_private *ast) struct drm_plane *cursor_plane = &ast_cursor_plane->base; size_t size, i; struct drm_gem_vram_object *gbo; - struct dma_buf_map map; + struct iosys_map map; int ret; s64 off; diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig index 61db5a66b493..fcd93f1aec90 100644 --- a/drivers/gpu/drm/bridge/Kconfig +++ b/drivers/gpu/drm/bridge/Kconfig @@ -30,6 +30,7 @@ config DRM_CDNS_DSI config DRM_CHIPONE_ICN6211 tristate "Chipone ICN6211 MIPI-DSI/RGB Converter bridge" depends on OF + depends on DRM_KMS_HELPER select DRM_MIPI_DSI select DRM_PANEL_BRIDGE help @@ -183,6 +184,7 @@ config DRM_PARADE_PS8640 tristate "Parade PS8640 MIPI DSI to eDP Converter" depends on OF select DRM_DP_AUX_BUS + select DRM_DP_HELPER select DRM_KMS_HELPER select DRM_MIPI_DSI select DRM_PANEL @@ -253,6 +255,7 @@ config DRM_TOSHIBA_TC358764 config DRM_TOSHIBA_TC358767 tristate "Toshiba TC358767 eDP bridge" depends on OF + select DRM_DP_HELPER select DRM_KMS_HELPER select REGMAP_I2C select DRM_PANEL @@ -272,6 +275,7 @@ config DRM_TOSHIBA_TC358768 config DRM_TOSHIBA_TC358775 tristate "Toshiba TC358775 DSI/LVDS bridge" depends on OF + select DRM_DP_HELPER select DRM_KMS_HELPER select REGMAP_I2C select DRM_PANEL @@ -299,6 +303,7 @@ config DRM_TI_SN65DSI83 config DRM_TI_SN65DSI86 tristate "TI SN65DSI86 DSI to eDP bridge" depends on OF + select DRM_DP_HELPER select DRM_KMS_HELPER select REGMAP_I2C select DRM_PANEL diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511.h b/drivers/gpu/drm/bridge/adv7511/adv7511.h index 592ecfcf00ca..6a882891d91c 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511.h +++ b/drivers/gpu/drm/bridge/adv7511/adv7511.h @@ -169,6 +169,7 @@ #define ADV7511_PACKET_ENABLE_SPARE2 BIT(1) #define ADV7511_PACKET_ENABLE_SPARE1 BIT(0) +#define ADV7535_REG_POWER2_HPD_OVERRIDE BIT(6) #define ADV7511_REG_POWER2_HPD_SRC_MASK 0xc0 #define ADV7511_REG_POWER2_HPD_SRC_BOTH 0x00 #define ADV7511_REG_POWER2_HPD_SRC_HPD 0x40 diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c index f8e5da148599..005bf18682ff 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c @@ -223,7 +223,7 @@ static void adv7511_set_config_csc(struct adv7511 *adv7511, config.csc_coefficents = adv7511_csc_ycbcr_to_rgb; if ((connector->display_info.color_formats & - DRM_COLOR_FORMAT_YCRCB422) && + DRM_COLOR_FORMAT_YCBCR422) && config.hdmi_mode) { config.csc_enable = false; config.avi_infoframe.colorspace = @@ -351,11 +351,17 @@ static void __adv7511_power_on(struct adv7511 *adv7511) * from standby or are enabled. When the HPD goes low the adv7511 is * reset and the outputs are disabled which might cause the monitor to * go to standby again. To avoid this we ignore the HPD pin for the - * first few seconds after enabling the output. + * first few seconds after enabling the output. On the other hand + * adv7535 require to enable HPD Override bit for proper HPD. */ - regmap_update_bits(adv7511->regmap, ADV7511_REG_POWER2, - ADV7511_REG_POWER2_HPD_SRC_MASK, - ADV7511_REG_POWER2_HPD_SRC_NONE); + if (adv7511->type == ADV7535) + regmap_update_bits(adv7511->regmap, ADV7511_REG_POWER2, + ADV7535_REG_POWER2_HPD_OVERRIDE, + ADV7535_REG_POWER2_HPD_OVERRIDE); + else + regmap_update_bits(adv7511->regmap, ADV7511_REG_POWER2, + ADV7511_REG_POWER2_HPD_SRC_MASK, + ADV7511_REG_POWER2_HPD_SRC_NONE); } static void adv7511_power_on(struct adv7511 *adv7511) @@ -375,6 +381,10 @@ static void adv7511_power_on(struct adv7511 *adv7511) static void __adv7511_power_off(struct adv7511 *adv7511) { /* TODO: setup additional power down modes */ + if (adv7511->type == ADV7535) + regmap_update_bits(adv7511->regmap, ADV7511_REG_POWER2, + ADV7535_REG_POWER2_HPD_OVERRIDE, 0); + regmap_update_bits(adv7511->regmap, ADV7511_REG_POWER, ADV7511_POWER_POWER_DOWN, ADV7511_POWER_POWER_DOWN); @@ -672,9 +682,14 @@ adv7511_detect(struct adv7511 *adv7511, struct drm_connector *connector) status = connector_status_disconnected; } else { /* Renable HPD sensing */ - regmap_update_bits(adv7511->regmap, ADV7511_REG_POWER2, - ADV7511_REG_POWER2_HPD_SRC_MASK, - ADV7511_REG_POWER2_HPD_SRC_BOTH); + if (adv7511->type == ADV7535) + regmap_update_bits(adv7511->regmap, ADV7511_REG_POWER2, + ADV7535_REG_POWER2_HPD_OVERRIDE, + ADV7535_REG_POWER2_HPD_OVERRIDE); + else + regmap_update_bits(adv7511->regmap, ADV7511_REG_POWER2, + ADV7511_REG_POWER2_HPD_SRC_MASK, + ADV7511_REG_POWER2_HPD_SRC_BOTH); } adv7511->status = status; diff --git a/drivers/gpu/drm/bridge/adv7511/adv7533.c b/drivers/gpu/drm/bridge/adv7511/adv7533.c index eb7579dec40a..ef6270806d1d 100644 --- a/drivers/gpu/drm/bridge/adv7511/adv7533.c +++ b/drivers/gpu/drm/bridge/adv7511/adv7533.c @@ -29,7 +29,7 @@ static void adv7511_dsi_config_timing_gen(struct adv7511 *adv) struct mipi_dsi_device *dsi = adv->dsi; struct drm_display_mode *mode = &adv->curr_mode; unsigned int hsw, hfp, hbp, vsw, vfp, vbp; - u8 clock_div_by_lanes[] = { 6, 4, 3 }; /* 2, 3, 4 lanes */ + static const u8 clock_div_by_lanes[] = { 6, 4, 3 }; /* 2, 3, 4 lanes */ hsw = mode->hsync_end - mode->hsync_start; hfp = mode->hsync_start - mode->hdisplay; diff --git a/drivers/gpu/drm/bridge/analogix/Kconfig b/drivers/gpu/drm/bridge/analogix/Kconfig index 2ef6eb2b786c..319ba0df57be 100644 --- a/drivers/gpu/drm/bridge/analogix/Kconfig +++ b/drivers/gpu/drm/bridge/analogix/Kconfig @@ -3,6 +3,7 @@ config DRM_ANALOGIX_ANX6345 tristate "Analogix ANX6345 bridge" depends on OF select DRM_ANALOGIX_DP + select DRM_DP_HELPER select DRM_KMS_HELPER select REGMAP_I2C help @@ -14,6 +15,7 @@ config DRM_ANALOGIX_ANX6345 config DRM_ANALOGIX_ANX78XX tristate "Analogix ANX78XX bridge" select DRM_ANALOGIX_DP + select DRM_DP_HELPER select DRM_KMS_HELPER select REGMAP_I2C help diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c index e33cd077595a..94e56a2e91f2 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c +++ b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c @@ -22,7 +22,7 @@ #include <drm/drm_bridge.h> #include <drm/drm_crtc.h> #include <drm/drm_crtc_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c b/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c index 5e6a0ed39199..2768b41c48e9 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c +++ b/drivers/gpu/drm/bridge/analogix/analogix-anx78xx.c @@ -21,7 +21,7 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> #include <drm/drm_crtc.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_print.h> #include <drm/drm_probe_helper.h> diff --git a/drivers/gpu/drm/bridge/analogix/analogix-i2c-dptx.c b/drivers/gpu/drm/bridge/analogix/analogix-i2c-dptx.c index fe40bab21530..e8297168bfef 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix-i2c-dptx.c +++ b/drivers/gpu/drm/bridge/analogix/analogix-i2c-dptx.c @@ -8,7 +8,7 @@ #include <linux/regmap.h> #include <drm/drm.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_print.h> #include "analogix-i2c-dptx.h" diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c index b7d2e4449cfa..eb590fb8e8d0 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.c @@ -1537,9 +1537,9 @@ static void analogix_dp_bridge_mode_set(struct drm_bridge *bridge, video->color_depth = COLOR_8; break; } - if (display_info->color_formats & DRM_COLOR_FORMAT_YCRCB444) + if (display_info->color_formats & DRM_COLOR_FORMAT_YCBCR444) video->color_space = COLOR_YCBCR444; - else if (display_info->color_formats & DRM_COLOR_FORMAT_YCRCB422) + else if (display_info->color_formats & DRM_COLOR_FORMAT_YCBCR422) video->color_space = COLOR_YCBCR422; else video->color_space = COLOR_RGB; diff --git a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h index c051502d7fbf..32665203a6ae 100644 --- a/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h +++ b/drivers/gpu/drm/bridge/analogix/analogix_dp_core.h @@ -10,7 +10,7 @@ #define _ANALOGIX_DP_CORE_H #include <drm/drm_crtc.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #define DP_TIMEOUT_LOOP_COUNT 100 #define MAX_CR_LOOP 5 diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.c b/drivers/gpu/drm/bridge/analogix/anx7625.c index 2346dbcc505f..76662fce4ce6 100644 --- a/drivers/gpu/drm/bridge/analogix/anx7625.c +++ b/drivers/gpu/drm/bridge/analogix/anx7625.c @@ -24,8 +24,9 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> #include <drm/drm_crtc_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> +#include <drm/drm_hdcp.h> #include <drm/drm_mipi_dsi.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> @@ -213,6 +214,65 @@ static int wait_aux_op_finish(struct anx7625_data *ctx) return 0; } +static int anx7625_aux_dpcd_read(struct anx7625_data *ctx, + u32 address, u8 len, u8 *buf) +{ + struct device *dev = &ctx->client->dev; + int ret; + u8 addrh, addrm, addrl; + u8 cmd; + + if (len > MAX_DPCD_BUFFER_SIZE) { + dev_err(dev, "exceed aux buffer len.\n"); + return -EINVAL; + } + + addrl = address & 0xFF; + addrm = (address >> 8) & 0xFF; + addrh = (address >> 16) & 0xFF; + + cmd = DPCD_CMD(len, DPCD_READ); + cmd = ((len - 1) << 4) | 0x09; + + /* Set command and length */ + ret = anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + AP_AUX_COMMAND, cmd); + + /* Set aux access address */ + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + AP_AUX_ADDR_7_0, addrl); + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + AP_AUX_ADDR_15_8, addrm); + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + AP_AUX_ADDR_19_16, addrh); + + /* Enable aux access */ + ret |= anx7625_write_or(ctx, ctx->i2c.rx_p0_client, + AP_AUX_CTRL_STATUS, AP_AUX_CTRL_OP_EN); + + if (ret < 0) { + dev_err(dev, "cannot access aux related register.\n"); + return -EIO; + } + + usleep_range(2000, 2100); + + ret = wait_aux_op_finish(ctx); + if (ret) { + dev_err(dev, "aux IO error: wait aux op finish.\n"); + return ret; + } + + ret = anx7625_reg_block_read(ctx, ctx->i2c.rx_p0_client, + AP_AUX_BUFF_START, len, buf); + if (ret < 0) { + dev_err(dev, "read dpcd register failed\n"); + return -EIO; + } + + return 0; +} + static int anx7625_video_mute_control(struct anx7625_data *ctx, u8 status) { @@ -669,6 +729,165 @@ static int anx7625_dpi_config(struct anx7625_data *ctx) return ret; } +static int anx7625_read_flash_status(struct anx7625_data *ctx) +{ + return anx7625_reg_read(ctx, ctx->i2c.rx_p0_client, R_RAM_CTRL); +} + +static int anx7625_hdcp_key_probe(struct anx7625_data *ctx) +{ + int ret, val; + struct device *dev = &ctx->client->dev; + u8 ident[FLASH_BUF_LEN]; + + ret = anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + FLASH_ADDR_HIGH, 0x91); + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + FLASH_ADDR_LOW, 0xA0); + if (ret < 0) { + dev_err(dev, "IO error : set key flash address.\n"); + return ret; + } + + ret = anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + FLASH_LEN_HIGH, (FLASH_BUF_LEN - 1) >> 8); + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + FLASH_LEN_LOW, (FLASH_BUF_LEN - 1) & 0xFF); + if (ret < 0) { + dev_err(dev, "IO error : set key flash len.\n"); + return ret; + } + + ret = anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + R_FLASH_RW_CTRL, FLASH_READ); + ret |= readx_poll_timeout(anx7625_read_flash_status, + ctx, val, + ((val & FLASH_DONE) || (val < 0)), + 2000, + 2000 * 150); + if (ret) { + dev_err(dev, "flash read access fail!\n"); + return -EIO; + } + + ret = anx7625_reg_block_read(ctx, ctx->i2c.rx_p0_client, + FLASH_BUF_BASE_ADDR, + FLASH_BUF_LEN, ident); + if (ret < 0) { + dev_err(dev, "read flash data fail!\n"); + return -EIO; + } + + if (ident[29] == 0xFF && ident[30] == 0xFF && ident[31] == 0xFF) + return -EINVAL; + + return 0; +} + +static int anx7625_hdcp_key_load(struct anx7625_data *ctx) +{ + int ret; + struct device *dev = &ctx->client->dev; + + /* Select HDCP 1.4 KEY */ + ret = anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + R_BOOT_RETRY, 0x12); + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + FLASH_ADDR_HIGH, HDCP14KEY_START_ADDR >> 8); + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + FLASH_ADDR_LOW, HDCP14KEY_START_ADDR & 0xFF); + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + R_RAM_LEN_H, HDCP14KEY_SIZE >> 12); + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + R_RAM_LEN_L, HDCP14KEY_SIZE >> 4); + + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + R_RAM_ADDR_H, 0); + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + R_RAM_ADDR_L, 0); + /* Enable HDCP 1.4 KEY load */ + ret |= anx7625_reg_write(ctx, ctx->i2c.rx_p0_client, + R_RAM_CTRL, DECRYPT_EN | LOAD_START); + dev_dbg(dev, "load HDCP 1.4 key done\n"); + return ret; +} + +static int anx7625_hdcp_disable(struct anx7625_data *ctx) +{ + int ret; + struct device *dev = &ctx->client->dev; + + dev_dbg(dev, "disable HDCP 1.4\n"); + + /* Disable HDCP */ + ret = anx7625_write_and(ctx, ctx->i2c.rx_p1_client, 0xee, 0x9f); + /* Try auth flag */ + ret |= anx7625_write_or(ctx, ctx->i2c.rx_p1_client, 0xec, 0x10); + /* Interrupt for DRM */ + ret |= anx7625_write_or(ctx, ctx->i2c.rx_p1_client, 0xff, 0x01); + if (ret < 0) + dev_err(dev, "fail to disable HDCP\n"); + + return anx7625_write_and(ctx, ctx->i2c.tx_p0_client, + TX_HDCP_CTRL0, ~HARD_AUTH_EN & 0xFF); +} + +static int anx7625_hdcp_enable(struct anx7625_data *ctx) +{ + u8 bcap; + int ret; + struct device *dev = &ctx->client->dev; + + ret = anx7625_hdcp_key_probe(ctx); + if (ret) { + dev_dbg(dev, "no key found, not to do hdcp\n"); + return ret; + } + + /* Read downstream capability */ + anx7625_aux_dpcd_read(ctx, 0x68028, 1, &bcap); + if (!(bcap & 0x01)) { + pr_warn("downstream not support HDCP 1.4, cap(%x).\n", bcap); + return 0; + } + + dev_dbg(dev, "enable HDCP 1.4\n"); + + /* First clear HDCP state */ + ret = anx7625_reg_write(ctx, ctx->i2c.tx_p0_client, + TX_HDCP_CTRL0, + KSVLIST_VLD | BKSV_SRM_PASS | RE_AUTHEN); + usleep_range(1000, 1100); + /* Second clear HDCP state */ + ret |= anx7625_reg_write(ctx, ctx->i2c.tx_p0_client, + TX_HDCP_CTRL0, + KSVLIST_VLD | BKSV_SRM_PASS | RE_AUTHEN); + + /* Set time for waiting KSVR */ + ret |= anx7625_reg_write(ctx, ctx->i2c.tx_p0_client, + SP_TX_WAIT_KSVR_TIME, 0xc8); + /* Set time for waiting R0 */ + ret |= anx7625_reg_write(ctx, ctx->i2c.tx_p0_client, + SP_TX_WAIT_R0_TIME, 0xb0); + ret |= anx7625_hdcp_key_load(ctx); + if (ret) { + pr_warn("prepare HDCP key failed.\n"); + return ret; + } + + ret = anx7625_write_or(ctx, ctx->i2c.rx_p1_client, 0xee, 0x20); + + /* Try auth flag */ + ret |= anx7625_write_or(ctx, ctx->i2c.rx_p1_client, 0xec, 0x10); + /* Interrupt for DRM */ + ret |= anx7625_write_or(ctx, ctx->i2c.rx_p1_client, 0xff, 0x01); + if (ret < 0) + dev_err(dev, "fail to enable HDCP\n"); + + return anx7625_write_or(ctx, ctx->i2c.tx_p0_client, + TX_HDCP_CTRL0, HARD_AUTH_EN); +} + static void anx7625_dp_start(struct anx7625_data *ctx) { int ret; @@ -679,6 +898,9 @@ static void anx7625_dp_start(struct anx7625_data *ctx) return; } + /* Disable HDCP */ + anx7625_write_and(ctx, ctx->i2c.rx_p1_client, 0xee, 0x9f); + if (ctx->pdata.is_dpi) ret = anx7625_dpi_config(ctx); else @@ -686,6 +908,10 @@ static void anx7625_dp_start(struct anx7625_data *ctx) if (ret < 0) DRM_DEV_ERROR(dev, "MIPI phy setup error.\n"); + + ctx->hdcp_cp = DRM_MODE_CONTENT_PROTECTION_UNDESIRED; + + ctx->dp_en = 1; } static void anx7625_dp_stop(struct anx7625_data *ctx) @@ -705,6 +931,10 @@ static void anx7625_dp_stop(struct anx7625_data *ctx) ret |= anx7625_video_mute_control(ctx, 1); if (ret < 0) DRM_DEV_ERROR(dev, "IO error : mute video fail\n"); + + ctx->hdcp_cp = DRM_MODE_CONTENT_PROTECTION_UNDESIRED; + + ctx->dp_en = 0; } static int sp_tx_rst_aux(struct anx7625_data *ctx) @@ -1098,9 +1328,18 @@ static void anx7625_init_gpio(struct anx7625_data *platform) /* Gpio for chip power enable */ platform->pdata.gpio_p_on = devm_gpiod_get_optional(dev, "enable", GPIOD_OUT_LOW); + if (IS_ERR_OR_NULL(platform->pdata.gpio_p_on)) { + DRM_DEV_DEBUG_DRIVER(dev, "no enable gpio found\n"); + platform->pdata.gpio_p_on = NULL; + } + /* Gpio for chip reset */ platform->pdata.gpio_reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW); + if (IS_ERR_OR_NULL(platform->pdata.gpio_reset)) { + DRM_DEV_DEBUG_DRIVER(dev, "no reset gpio found\n"); + platform->pdata.gpio_reset = NULL; + } if (platform->pdata.gpio_p_on && platform->pdata.gpio_reset) { platform->pdata.low_power_mode = 1; @@ -1601,9 +1840,27 @@ static int anx7625_audio_hook_plugged_cb(struct device *dev, void *data, return 0; } +static int anx7625_audio_get_eld(struct device *dev, void *data, + u8 *buf, size_t len) +{ + struct anx7625_data *ctx = dev_get_drvdata(dev); + + if (!ctx->connector) { + dev_err(dev, "connector not initial\n"); + return -EINVAL; + } + + dev_dbg(dev, "audio copy eld\n"); + memcpy(buf, ctx->connector->eld, + min(sizeof(ctx->connector->eld), len)); + + return 0; +} + static const struct hdmi_codec_ops anx7625_codec_ops = { .hw_params = anx7625_audio_hw_params, .audio_shutdown = anx7625_audio_shutdown, + .get_eld = anx7625_audio_get_eld, .get_dai_id = anx7625_hdmi_i2s_get_dai_id, .hook_plugged_cb = anx7625_audio_hook_plugged_cb, }; @@ -1660,7 +1917,7 @@ static int anx7625_attach_dsi(struct anx7625_data *ctx) host = of_find_mipi_dsi_host_by_node(ctx->pdata.mipi_host_node); if (!host) { DRM_DEV_ERROR(dev, "fail to find dsi host.\n"); - return -EINVAL; + return -EPROBE_DEFER; } dsi = devm_mipi_dsi_device_register_full(dev, host, &info); @@ -1688,6 +1945,83 @@ static int anx7625_attach_dsi(struct anx7625_data *ctx) return 0; } +static void hdcp_check_work_func(struct work_struct *work) +{ + u8 status; + struct delayed_work *dwork; + struct anx7625_data *ctx; + struct device *dev; + struct drm_device *drm_dev; + + dwork = to_delayed_work(work); + ctx = container_of(dwork, struct anx7625_data, hdcp_work); + dev = &ctx->client->dev; + + if (!ctx->connector) { + dev_err(dev, "HDCP connector is null!"); + return; + } + + drm_dev = ctx->connector->dev; + drm_modeset_lock(&drm_dev->mode_config.connection_mutex, NULL); + mutex_lock(&ctx->hdcp_wq_lock); + + status = anx7625_reg_read(ctx, ctx->i2c.tx_p0_client, 0); + dev_dbg(dev, "sink HDCP status check: %.02x\n", status); + if (status & BIT(1)) { + ctx->hdcp_cp = DRM_MODE_CONTENT_PROTECTION_ENABLED; + drm_hdcp_update_content_protection(ctx->connector, + ctx->hdcp_cp); + dev_dbg(dev, "update CP to ENABLE\n"); + } + + mutex_unlock(&ctx->hdcp_wq_lock); + drm_modeset_unlock(&drm_dev->mode_config.connection_mutex); +} + +static int anx7625_connector_atomic_check(struct anx7625_data *ctx, + struct drm_connector_state *state) +{ + struct device *dev = &ctx->client->dev; + int cp; + + dev_dbg(dev, "hdcp state check\n"); + cp = state->content_protection; + + if (cp == ctx->hdcp_cp) + return 0; + + if (cp == DRM_MODE_CONTENT_PROTECTION_DESIRED) { + if (ctx->dp_en) { + dev_dbg(dev, "enable HDCP\n"); + anx7625_hdcp_enable(ctx); + + queue_delayed_work(ctx->hdcp_workqueue, + &ctx->hdcp_work, + msecs_to_jiffies(2000)); + } + } + + if (cp == DRM_MODE_CONTENT_PROTECTION_UNDESIRED) { + if (ctx->hdcp_cp != DRM_MODE_CONTENT_PROTECTION_ENABLED) { + dev_err(dev, "current CP is not ENABLED\n"); + return -EINVAL; + } + anx7625_hdcp_disable(ctx); + ctx->hdcp_cp = DRM_MODE_CONTENT_PROTECTION_UNDESIRED; + drm_hdcp_update_content_protection(ctx->connector, + ctx->hdcp_cp); + dev_dbg(dev, "update CP to UNDESIRE\n"); + } + + if (cp == DRM_MODE_CONTENT_PROTECTION_ENABLED) { + dev_err(dev, "Userspace illegal set to PROTECTION ENABLE\n"); + return -EINVAL; + } + + return 0; +} + static int anx7625_bridge_attach(struct drm_bridge *bridge, enum drm_bridge_attach_flags flags) { @@ -1902,25 +2236,57 @@ static bool anx7625_bridge_mode_fixup(struct drm_bridge *bridge, return true; } -static void anx7625_bridge_enable(struct drm_bridge *bridge) +static int anx7625_bridge_atomic_check(struct drm_bridge *bridge, + struct drm_bridge_state *bridge_state, + struct drm_crtc_state *crtc_state, + struct drm_connector_state *conn_state) +{ + struct anx7625_data *ctx = bridge_to_anx7625(bridge); + struct device *dev = &ctx->client->dev; + + dev_dbg(dev, "drm bridge atomic check\n"); + + anx7625_bridge_mode_fixup(bridge, &crtc_state->mode, + &crtc_state->adjusted_mode); + + return anx7625_connector_atomic_check(ctx, conn_state); +} + +static void anx7625_bridge_atomic_enable(struct drm_bridge *bridge, + struct drm_bridge_state *state) { struct anx7625_data *ctx = bridge_to_anx7625(bridge); struct device *dev = &ctx->client->dev; + struct drm_connector *connector; - DRM_DEV_DEBUG_DRIVER(dev, "drm enable\n"); + dev_dbg(dev, "drm atomic enable\n"); + + if (!bridge->encoder) { + dev_err(dev, "Parent encoder object not found"); + return; + } + + connector = drm_atomic_get_new_connector_for_encoder(state->base.state, + bridge->encoder); + if (!connector) + return; + + ctx->connector = connector; pm_runtime_get_sync(dev); anx7625_dp_start(ctx); } -static void anx7625_bridge_disable(struct drm_bridge *bridge) +static void anx7625_bridge_atomic_disable(struct drm_bridge *bridge, + struct drm_bridge_state *old) { struct anx7625_data *ctx = bridge_to_anx7625(bridge); struct device *dev = &ctx->client->dev; - DRM_DEV_DEBUG_DRIVER(dev, "drm disable\n"); + dev_dbg(dev, "drm atomic disable\n"); + ctx->connector = NULL; anx7625_dp_stop(ctx); pm_runtime_put_sync(dev); @@ -1950,11 +2316,14 @@ static struct edid *anx7625_bridge_get_edid(struct drm_bridge *bridge, static const struct drm_bridge_funcs anx7625_bridge_funcs = { .attach = anx7625_bridge_attach, - .disable = anx7625_bridge_disable, .mode_valid = anx7625_bridge_mode_valid, .mode_set = anx7625_bridge_mode_set, - .mode_fixup = anx7625_bridge_mode_fixup, - .enable = anx7625_bridge_enable, + .atomic_check = anx7625_bridge_atomic_check, + .atomic_enable = anx7625_bridge_atomic_enable, + .atomic_disable = anx7625_bridge_atomic_disable, + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, .detect = anx7625_bridge_detect, .get_edid = anx7625_bridge_get_edid, }; @@ -1962,40 +2331,54 @@ static const struct drm_bridge_funcs anx7625_bridge_funcs = { static int anx7625_register_i2c_dummy_clients(struct anx7625_data *ctx, struct i2c_client *client) { + int err = 0; + ctx->i2c.tx_p0_client = i2c_new_dummy_device(client->adapter, TX_P0_ADDR >> 1); - if (!ctx->i2c.tx_p0_client) - return -ENOMEM; + if (IS_ERR(ctx->i2c.tx_p0_client)) + return PTR_ERR(ctx->i2c.tx_p0_client); ctx->i2c.tx_p1_client = i2c_new_dummy_device(client->adapter, TX_P1_ADDR >> 1); - if (!ctx->i2c.tx_p1_client) + if (IS_ERR(ctx->i2c.tx_p1_client)) { + err = PTR_ERR(ctx->i2c.tx_p1_client); goto free_tx_p0; + } ctx->i2c.tx_p2_client = i2c_new_dummy_device(client->adapter, TX_P2_ADDR >> 1); - if (!ctx->i2c.tx_p2_client) + if (IS_ERR(ctx->i2c.tx_p2_client)) { + err = PTR_ERR(ctx->i2c.tx_p2_client); goto free_tx_p1; + } ctx->i2c.rx_p0_client = i2c_new_dummy_device(client->adapter, RX_P0_ADDR >> 1); - if (!ctx->i2c.rx_p0_client) + if (IS_ERR(ctx->i2c.rx_p0_client)) { + err = PTR_ERR(ctx->i2c.rx_p0_client); goto free_tx_p2; + } ctx->i2c.rx_p1_client = i2c_new_dummy_device(client->adapter, RX_P1_ADDR >> 1); - if (!ctx->i2c.rx_p1_client) + if (IS_ERR(ctx->i2c.rx_p1_client)) { + err = PTR_ERR(ctx->i2c.rx_p1_client); goto free_rx_p0; + } ctx->i2c.rx_p2_client = i2c_new_dummy_device(client->adapter, RX_P2_ADDR >> 1); - if (!ctx->i2c.rx_p2_client) + if (IS_ERR(ctx->i2c.rx_p2_client)) { + err = PTR_ERR(ctx->i2c.rx_p2_client); goto free_rx_p1; + } ctx->i2c.tcpc_client = i2c_new_dummy_device(client->adapter, TCPC_INTERFACE_ADDR >> 1); - if (!ctx->i2c.tcpc_client) + if (IS_ERR(ctx->i2c.tcpc_client)) { + err = PTR_ERR(ctx->i2c.tcpc_client); goto free_rx_p2; + } return 0; @@ -2012,7 +2395,7 @@ free_tx_p1: free_tx_p0: i2c_unregister_device(ctx->i2c.tx_p0_client); - return -ENOMEM; + return err; } static void anx7625_unregister_i2c_dummy_clients(struct anx7625_data *ctx) @@ -2134,6 +2517,15 @@ static int anx7625_i2c_probe(struct i2c_client *client, anx7625_init_gpio(platform); mutex_init(&platform->lock); + mutex_init(&platform->hdcp_wq_lock); + + INIT_DELAYED_WORK(&platform->hdcp_work, hdcp_check_work_func); + platform->hdcp_workqueue = create_workqueue("hdcp workqueue"); + if (!platform->hdcp_workqueue) { + dev_err(dev, "fail to create work queue\n"); + ret = -ENOMEM; + goto free_platform; + } platform->pdata.intp_irq = client->irq; if (platform->pdata.intp_irq) { @@ -2143,7 +2535,7 @@ static int anx7625_i2c_probe(struct i2c_client *client, if (!platform->workqueue) { DRM_DEV_ERROR(dev, "fail to create work queue\n"); ret = -ENOMEM; - goto free_platform; + goto free_hdcp_wq; } ret = devm_request_threaded_irq(dev, platform->pdata.intp_irq, @@ -2213,6 +2605,10 @@ free_wq: if (platform->workqueue) destroy_workqueue(platform->workqueue); +free_hdcp_wq: + if (platform->hdcp_workqueue) + destroy_workqueue(platform->hdcp_workqueue); + free_platform: kfree(platform); @@ -2228,6 +2624,12 @@ static int anx7625_i2c_remove(struct i2c_client *client) if (platform->pdata.intp_irq) destroy_workqueue(platform->workqueue); + if (platform->hdcp_workqueue) { + cancel_delayed_work(&platform->hdcp_work); + flush_workqueue(platform->workqueue); + destroy_workqueue(platform->workqueue); + } + if (!platform->pdata.low_power_mode) pm_runtime_put_sync_suspend(&client->dev); diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.h b/drivers/gpu/drm/bridge/analogix/anx7625.h index 3d79b6fb13c8..56165f5b254c 100644 --- a/drivers/gpu/drm/bridge/analogix/anx7625.h +++ b/drivers/gpu/drm/bridge/analogix/anx7625.h @@ -59,10 +59,23 @@ /***************************************************************/ /* Register definition of device address 0x70 */ -#define I2C_ADDR_70_DPTX 0x70 - -#define SP_TX_LINK_BW_SET_REG 0xA0 -#define SP_TX_LANE_COUNT_SET_REG 0xA1 +#define TX_HDCP_CTRL0 0x01 +#define STORE_AN BIT(7) +#define RX_REPEATER BIT(6) +#define RE_AUTHEN BIT(5) +#define SW_AUTH_OK BIT(4) +#define HARD_AUTH_EN BIT(3) +#define ENC_EN BIT(2) +#define BKSV_SRM_PASS BIT(1) +#define KSVLIST_VLD BIT(0) + +#define SP_TX_WAIT_R0_TIME 0x40 +#define SP_TX_WAIT_KSVR_TIME 0x42 +#define SP_TX_SYS_CTRL1_REG 0x80 +#define HDCP2TX_FW_EN BIT(4) + +#define SP_TX_LINK_BW_SET_REG 0xA0 +#define SP_TX_LANE_COUNT_SET_REG 0xA1 #define M_VID_0 0xC0 #define M_VID_1 0xC1 @@ -71,6 +84,12 @@ #define N_VID_1 0xC4 #define N_VID_2 0xC5 +#define KEY_START_ADDR 0x9000 +#define KEY_RESERVED 416 + +#define HDCP14KEY_START_ADDR (KEY_START_ADDR + KEY_RESERVED) +#define HDCP14KEY_SIZE 624 + /***************************************************************/ /* Register definition of device address 0x72 */ #define AUX_RST 0x04 @@ -155,9 +174,43 @@ #define I2C_ADDR_7E_FLASH_CONTROLLER 0x7E +#define R_BOOT_RETRY 0x00 +#define R_RAM_ADDR_H 0x01 +#define R_RAM_ADDR_L 0x02 +#define R_RAM_LEN_H 0x03 +#define R_RAM_LEN_L 0x04 #define FLASH_LOAD_STA 0x05 #define FLASH_LOAD_STA_CHK BIT(7) +#define R_RAM_CTRL 0x05 +/* bit positions */ +#define FLASH_DONE BIT(7) +#define BOOT_LOAD_DONE BIT(6) +#define CRC_OK BIT(5) +#define LOAD_DONE BIT(4) +#define O_RW_DONE BIT(3) +#define FUSE_BUSY BIT(2) +#define DECRYPT_EN BIT(1) +#define LOAD_START BIT(0) + +#define FLASH_ADDR_HIGH 0x0F +#define FLASH_ADDR_LOW 0x10 +#define FLASH_LEN_HIGH 0x31 +#define FLASH_LEN_LOW 0x32 +#define R_FLASH_RW_CTRL 0x33 +/* bit positions */ +#define READ_DELAY_SELECT BIT(7) +#define GENERAL_INSTRUCTION_EN BIT(6) +#define FLASH_ERASE_EN BIT(5) +#define RDID_READ_EN BIT(4) +#define REMS_READ_EN BIT(3) +#define WRITE_STATUS_EN BIT(2) +#define FLASH_READ BIT(1) +#define FLASH_WRITE BIT(0) + +#define FLASH_BUF_BASE_ADDR 0x60 +#define FLASH_BUF_LEN 0x20 + #define XTAL_FRQ_SEL 0x3F /* bit field positions */ #define XTAL_FRQ_SEL_POS 5 @@ -184,10 +237,15 @@ #define AP_AUX_CTRL_ADDRONLY 0x20 #define AP_AUX_BUFF_START 0x15 -#define PIXEL_CLOCK_L 0x25 -#define PIXEL_CLOCK_H 0x26 +#define PIXEL_CLOCK_L 0x25 +#define PIXEL_CLOCK_H 0x26 + +#define AP_AUX_COMMAND 0x27 /* com+len */ +#define LENGTH_SHIFT 4 +#define DPCD_READ 0x09 +#define DPCD_WRITE 0x08 +#define DPCD_CMD(len, cmd) ((((len) - 1) << LENGTH_SHIFT) | (cmd)) -#define AP_AUX_COMMAND 0x27 /* com+len */ /* Bit 0&1: 3D video structure */ /* 0x01: frame packing, 0x02:Line alternative, 0x03:Side-by-side(full) */ #define AP_AV_STATUS 0x28 @@ -392,21 +450,29 @@ struct anx7625_data { struct platform_device *audio_pdev; int hpd_status; int hpd_high_cnt; + int dp_en; + int hdcp_cp; /* Lock for work queue */ struct mutex lock; struct i2c_client *client; struct anx7625_i2c_client i2c; struct i2c_client *last_client; + struct timer_list hdcp_timer; struct s_edid_data slimport_edid_p; struct device *codec_dev; hdmi_codec_plugged_cb plugged_cb; struct work_struct work; struct workqueue_struct *workqueue; + struct delayed_work hdcp_work; + struct workqueue_struct *hdcp_workqueue; + /* Lock for hdcp work queue */ + struct mutex hdcp_wq_lock; char edid_block; struct display_timing dt; u8 display_timing_valid; struct drm_bridge bridge; u8 bridge_attached; + struct drm_connector *connector; struct mipi_dsi_device *dsi; }; diff --git a/drivers/gpu/drm/bridge/cadence/Kconfig b/drivers/gpu/drm/bridge/cadence/Kconfig index ef8c230e0f62..de697bade05e 100644 --- a/drivers/gpu/drm/bridge/cadence/Kconfig +++ b/drivers/gpu/drm/bridge/cadence/Kconfig @@ -1,6 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only config DRM_CDNS_MHDP8546 tristate "Cadence DPI/DP bridge" + select DRM_DP_HELPER select DRM_KMS_HELPER select DRM_PANEL_BRIDGE depends on OF diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c index 5530fbf64f1e..ac18e15aa167 100644 --- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c +++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c @@ -41,7 +41,7 @@ #include <drm/drm_bridge.h> #include <drm/drm_connector.h> #include <drm/drm_crtc_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_hdcp.h> #include <drm/drm_modeset_helper_vtables.h> #include <drm/drm_print.h> @@ -1553,13 +1553,13 @@ static u32 cdns_mhdp_get_bpp(struct cdns_mhdp_display_fmt *fmt) switch (fmt->color_format) { case DRM_COLOR_FORMAT_RGB444: - case DRM_COLOR_FORMAT_YCRCB444: + case DRM_COLOR_FORMAT_YCBCR444: bpp = fmt->bpc * 3; break; - case DRM_COLOR_FORMAT_YCRCB422: + case DRM_COLOR_FORMAT_YCBCR422: bpp = fmt->bpc * 2; break; - case DRM_COLOR_FORMAT_YCRCB420: + case DRM_COLOR_FORMAT_YCBCR420: bpp = fmt->bpc * 3 / 2; break; default: @@ -1767,8 +1767,8 @@ static void cdns_mhdp_configure_video(struct cdns_mhdp_device *mhdp, * If YCBCR supported and stream not SD, use ITU709 * Need to handle ITU version with YCBCR420 when supported */ - if ((pxlfmt == DRM_COLOR_FORMAT_YCRCB444 || - pxlfmt == DRM_COLOR_FORMAT_YCRCB422) && mode->crtc_vdisplay >= 720) + if ((pxlfmt == DRM_COLOR_FORMAT_YCBCR444 || + pxlfmt == DRM_COLOR_FORMAT_YCBCR422) && mode->crtc_vdisplay >= 720) misc0 = DP_YCBCR_COEFFICIENTS_ITU709; bpp = cdns_mhdp_get_bpp(&mhdp->display_fmt); @@ -1778,15 +1778,15 @@ static void cdns_mhdp_configure_video(struct cdns_mhdp_device *mhdp, pxl_repr = CDNS_DP_FRAMER_RGB << CDNS_DP_FRAMER_PXL_FORMAT; misc0 |= DP_COLOR_FORMAT_RGB; break; - case DRM_COLOR_FORMAT_YCRCB444: + case DRM_COLOR_FORMAT_YCBCR444: pxl_repr = CDNS_DP_FRAMER_YCBCR444 << CDNS_DP_FRAMER_PXL_FORMAT; misc0 |= DP_COLOR_FORMAT_YCbCr444 | DP_TEST_DYNAMIC_RANGE_CEA; break; - case DRM_COLOR_FORMAT_YCRCB422: + case DRM_COLOR_FORMAT_YCBCR422: pxl_repr = CDNS_DP_FRAMER_YCBCR422 << CDNS_DP_FRAMER_PXL_FORMAT; misc0 |= DP_COLOR_FORMAT_YCbCr422 | DP_TEST_DYNAMIC_RANGE_CEA; break; - case DRM_COLOR_FORMAT_YCRCB420: + case DRM_COLOR_FORMAT_YCBCR420: pxl_repr = CDNS_DP_FRAMER_YCBCR420 << CDNS_DP_FRAMER_PXL_FORMAT; break; default: @@ -1882,7 +1882,7 @@ static void cdns_mhdp_configure_video(struct cdns_mhdp_device *mhdp, if (mhdp->display_fmt.y_only) misc1 |= CDNS_DP_TEST_COLOR_FORMAT_RAW_Y_ONLY; /* Use VSC SDP for Y420 */ - if (pxlfmt == DRM_COLOR_FORMAT_YCRCB420) + if (pxlfmt == DRM_COLOR_FORMAT_YCBCR420) misc1 = CDNS_DP_TEST_VSC_SDP; cdns_mhdp_reg_write(mhdp, CDNS_DP_MSA_MISC(stream_id), diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.h b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.h index c74439d0b1a7..fc77f987c835 100644 --- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.h +++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.h @@ -17,7 +17,7 @@ #include <drm/drm_bridge.h> #include <drm/drm_connector.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> struct clk; struct device; diff --git a/drivers/gpu/drm/bridge/chipone-icn6211.c b/drivers/gpu/drm/bridge/chipone-icn6211.c index a6151db95586..e8f36dca56b3 100644 --- a/drivers/gpu/drm/bridge/chipone-icn6211.c +++ b/drivers/gpu/drm/bridge/chipone-icn6211.c @@ -4,6 +4,7 @@ * Author: Jagan Teki <jagan@amarulasolutions.com> */ +#include <drm/drm_atomic_helper.h> #include <drm/drm_of.h> #include <drm/drm_print.h> #include <drm/drm_mipi_dsi.h> @@ -30,6 +31,7 @@ struct chipone { struct device *dev; struct drm_bridge bridge; + struct drm_display_mode mode; struct drm_bridge *panel_bridge; struct gpio_desc *enable_gpio; struct regulator *vdd1; @@ -42,11 +44,6 @@ static inline struct chipone *bridge_to_chipone(struct drm_bridge *bridge) return container_of(bridge, struct chipone, bridge); } -static struct drm_display_mode *bridge_to_mode(struct drm_bridge *bridge) -{ - return &bridge->encoder->crtc->state->adjusted_mode; -} - static inline int chipone_dsi_write(struct chipone *icn, const void *seq, size_t len) { @@ -61,10 +58,11 @@ static inline int chipone_dsi_write(struct chipone *icn, const void *seq, chipone_dsi_write(icn, d, ARRAY_SIZE(d)); \ } -static void chipone_enable(struct drm_bridge *bridge) +static void chipone_atomic_enable(struct drm_bridge *bridge, + struct drm_bridge_state *old_bridge_state) { struct chipone *icn = bridge_to_chipone(bridge); - struct drm_display_mode *mode = bridge_to_mode(bridge); + struct drm_display_mode *mode = &icn->mode; ICN6211_DSI(icn, 0x7a, 0xc1); @@ -114,7 +112,8 @@ static void chipone_enable(struct drm_bridge *bridge) usleep_range(10000, 11000); } -static void chipone_pre_enable(struct drm_bridge *bridge) +static void chipone_atomic_pre_enable(struct drm_bridge *bridge, + struct drm_bridge_state *old_bridge_state) { struct chipone *icn = bridge_to_chipone(bridge); int ret; @@ -145,7 +144,8 @@ static void chipone_pre_enable(struct drm_bridge *bridge) usleep_range(10000, 11000); } -static void chipone_post_disable(struct drm_bridge *bridge) +static void chipone_atomic_post_disable(struct drm_bridge *bridge, + struct drm_bridge_state *old_bridge_state) { struct chipone *icn = bridge_to_chipone(bridge); @@ -161,6 +161,15 @@ static void chipone_post_disable(struct drm_bridge *bridge) gpiod_set_value(icn->enable_gpio, 0); } +static void chipone_mode_set(struct drm_bridge *bridge, + const struct drm_display_mode *mode, + const struct drm_display_mode *adjusted_mode) +{ + struct chipone *icn = bridge_to_chipone(bridge); + + drm_mode_copy(&icn->mode, adjusted_mode); +} + static int chipone_attach(struct drm_bridge *bridge, enum drm_bridge_attach_flags flags) { struct chipone *icn = bridge_to_chipone(bridge); @@ -169,10 +178,14 @@ static int chipone_attach(struct drm_bridge *bridge, enum drm_bridge_attach_flag } static const struct drm_bridge_funcs chipone_bridge_funcs = { - .attach = chipone_attach, - .post_disable = chipone_post_disable, - .pre_enable = chipone_pre_enable, - .enable = chipone_enable, + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, + .atomic_pre_enable = chipone_atomic_pre_enable, + .atomic_enable = chipone_atomic_enable, + .atomic_post_disable = chipone_atomic_post_disable, + .mode_set = chipone_mode_set, + .attach = chipone_attach, }; static int chipone_parse_dt(struct chipone *icn) diff --git a/drivers/gpu/drm/bridge/ite-it66121.c b/drivers/gpu/drm/bridge/ite-it66121.c index 06b59b422c69..69288cf894b9 100644 --- a/drivers/gpu/drm/bridge/ite-it66121.c +++ b/drivers/gpu/drm/bridge/ite-it66121.c @@ -936,9 +936,6 @@ static int it66121_probe(struct i2c_client *client, return -EPROBE_DEFER; } - if (!ctx->next_bridge) - return -EPROBE_DEFER; - i2c_set_clientdata(client, ctx); mutex_init(&ctx->lock); diff --git a/drivers/gpu/drm/bridge/lontium-lt9611.c b/drivers/gpu/drm/bridge/lontium-lt9611.c index dafb1b47c15f..feb128a4557d 100644 --- a/drivers/gpu/drm/bridge/lontium-lt9611.c +++ b/drivers/gpu/drm/bridge/lontium-lt9611.c @@ -1090,7 +1090,7 @@ static int lt9611_probe(struct i2c_client *client, if (!lt9611) return -ENOMEM; - lt9611->dev = &client->dev; + lt9611->dev = dev; lt9611->client = client; lt9611->sleep = false; @@ -1100,7 +1100,7 @@ static int lt9611_probe(struct i2c_client *client, return PTR_ERR(lt9611->regmap); } - ret = lt9611_parse_dt(&client->dev, lt9611); + ret = lt9611_parse_dt(dev, lt9611); if (ret) { dev_err(dev, "failed to parse device tree\n"); return ret; diff --git a/drivers/gpu/drm/bridge/lontium-lt9611uxc.c b/drivers/gpu/drm/bridge/lontium-lt9611uxc.c index 33f9716da0ee..3d62e6bf6892 100644 --- a/drivers/gpu/drm/bridge/lontium-lt9611uxc.c +++ b/drivers/gpu/drm/bridge/lontium-lt9611uxc.c @@ -860,7 +860,7 @@ static int lt9611uxc_probe(struct i2c_client *client, if (!lt9611uxc) return -ENOMEM; - lt9611uxc->dev = &client->dev; + lt9611uxc->dev = dev; lt9611uxc->client = client; mutex_init(<9611uxc->ocm_lock); @@ -870,7 +870,7 @@ static int lt9611uxc_probe(struct i2c_client *client, return PTR_ERR(lt9611uxc->regmap); } - ret = lt9611uxc_parse_dt(&client->dev, lt9611uxc); + ret = lt9611uxc_parse_dt(dev, lt9611uxc); if (ret) { dev_err(dev, "failed to parse device tree\n"); return ret; diff --git a/drivers/gpu/drm/bridge/nwl-dsi.c b/drivers/gpu/drm/bridge/nwl-dsi.c index af07eeb47ca0..5abb5ec3de46 100644 --- a/drivers/gpu/drm/bridge/nwl-dsi.c +++ b/drivers/gpu/drm/bridge/nwl-dsi.c @@ -66,7 +66,6 @@ struct nwl_dsi_transfer { struct nwl_dsi { struct drm_bridge bridge; struct mipi_dsi_host dsi_host; - struct drm_bridge *panel_bridge; struct device *dev; struct phy *phy; union phy_configure_opts phy_cfg; @@ -922,13 +921,11 @@ static int nwl_dsi_bridge_attach(struct drm_bridge *bridge, if (IS_ERR(panel_bridge)) return PTR_ERR(panel_bridge); } - dsi->panel_bridge = panel_bridge; - if (!dsi->panel_bridge) + if (!panel_bridge) return -EPROBE_DEFER; - return drm_bridge_attach(bridge->encoder, dsi->panel_bridge, bridge, - flags); + return drm_bridge_attach(bridge->encoder, panel_bridge, bridge, flags); } static void nwl_dsi_bridge_detach(struct drm_bridge *bridge) @@ -1204,6 +1201,7 @@ static int nwl_dsi_probe(struct platform_device *pdev) ret = nwl_dsi_select_input(dsi); if (ret < 0) { + pm_runtime_disable(dev); mipi_dsi_host_unregister(&dsi->dsi_host); return ret; } diff --git a/drivers/gpu/drm/bridge/parade-ps8640.c b/drivers/gpu/drm/bridge/parade-ps8640.c index 818704bf5e86..3f17337ee389 100644 --- a/drivers/gpu/drm/bridge/parade-ps8640.c +++ b/drivers/gpu/drm/bridge/parade-ps8640.c @@ -14,8 +14,8 @@ #include <linux/regulator/consumer.h> #include <drm/drm_bridge.h> -#include <drm/drm_dp_aux_bus.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_aux_bus.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_mipi_dsi.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> @@ -102,6 +102,7 @@ struct ps8640 { struct regulator_bulk_data supplies[2]; struct gpio_desc *gpio_reset; struct gpio_desc *gpio_powerdown; + struct device_link *link; bool pre_enabled; }; @@ -456,14 +457,36 @@ static int ps8640_bridge_attach(struct drm_bridge *bridge, return ret; } + ps_bridge->link = device_link_add(bridge->dev->dev, dev, DL_FLAG_STATELESS); + if (!ps_bridge->link) { + dev_err(dev, "failed to create device link"); + ret = -EINVAL; + goto err_devlink; + } + /* Attach the panel-bridge to the dsi bridge */ - return drm_bridge_attach(bridge->encoder, ps_bridge->panel_bridge, - &ps_bridge->bridge, flags); + ret = drm_bridge_attach(bridge->encoder, ps_bridge->panel_bridge, + &ps_bridge->bridge, flags); + if (ret) + goto err_bridge_attach; + + return 0; + +err_bridge_attach: + device_link_del(ps_bridge->link); +err_devlink: + drm_dp_aux_unregister(&ps_bridge->aux); + + return ret; } static void ps8640_bridge_detach(struct drm_bridge *bridge) { - drm_dp_aux_unregister(&bridge_to_ps8640(bridge)->aux); + struct ps8640 *ps_bridge = bridge_to_ps8640(bridge); + + drm_dp_aux_unregister(&ps_bridge->aux); + if (ps_bridge->link) + device_link_del(ps_bridge->link); } static struct edid *ps8640_bridge_get_edid(struct drm_bridge *bridge, diff --git a/drivers/gpu/drm/bridge/sii902x.c b/drivers/gpu/drm/bridge/sii902x.c index 89558e581530..65549fbfdc87 100644 --- a/drivers/gpu/drm/bridge/sii902x.c +++ b/drivers/gpu/drm/bridge/sii902x.c @@ -166,10 +166,12 @@ struct sii902x { struct i2c_client *i2c; struct regmap *regmap; struct drm_bridge bridge; + struct drm_bridge *next_bridge; struct drm_connector connector; struct gpio_desc *reset_gpio; struct i2c_mux_core *i2cmux; struct regulator_bulk_data supplies[2]; + bool sink_is_hdmi; /* * Mutex protects audio and video functions from interfering * each other, by keeping their i2c command sequences atomic. @@ -245,10 +247,8 @@ static void sii902x_reset(struct sii902x *sii902x) gpiod_set_value(sii902x->reset_gpio, 0); } -static enum drm_connector_status -sii902x_connector_detect(struct drm_connector *connector, bool force) +static enum drm_connector_status sii902x_detect(struct sii902x *sii902x) { - struct sii902x *sii902x = connector_to_sii902x(connector); unsigned int status; mutex_lock(&sii902x->mutex); @@ -261,6 +261,14 @@ sii902x_connector_detect(struct drm_connector *connector, bool force) connector_status_connected : connector_status_disconnected; } +static enum drm_connector_status +sii902x_connector_detect(struct drm_connector *connector, bool force) +{ + struct sii902x *sii902x = connector_to_sii902x(connector); + + return sii902x_detect(sii902x); +} + static const struct drm_connector_funcs sii902x_connector_funcs = { .detect = sii902x_connector_detect, .fill_modes = drm_helper_probe_single_connector_modes, @@ -270,42 +278,40 @@ static const struct drm_connector_funcs sii902x_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int sii902x_get_modes(struct drm_connector *connector) +static struct edid *sii902x_get_edid(struct sii902x *sii902x, + struct drm_connector *connector) { - struct sii902x *sii902x = connector_to_sii902x(connector); - u32 bus_format = MEDIA_BUS_FMT_RGB888_1X24; - u8 output_mode = SII902X_SYS_CTRL_OUTPUT_DVI; struct edid *edid; - int num = 0, ret; mutex_lock(&sii902x->mutex); edid = drm_get_edid(connector, sii902x->i2cmux->adapter[0]); - drm_connector_update_edid_property(connector, edid); if (edid) { if (drm_detect_hdmi_monitor(edid)) - output_mode = SII902X_SYS_CTRL_OUTPUT_HDMI; - - num = drm_add_edid_modes(connector, edid); - kfree(edid); + sii902x->sink_is_hdmi = true; + else + sii902x->sink_is_hdmi = false; } - ret = drm_display_info_set_bus_formats(&connector->display_info, - &bus_format, 1); - if (ret) - goto error_out; + mutex_unlock(&sii902x->mutex); - ret = regmap_update_bits(sii902x->regmap, SII902X_SYS_CTRL_DATA, - SII902X_SYS_CTRL_OUTPUT_MODE, output_mode); - if (ret) - goto error_out; + return edid; +} - ret = num; +static int sii902x_get_modes(struct drm_connector *connector) +{ + struct sii902x *sii902x = connector_to_sii902x(connector); + struct edid *edid; + int num = 0; -error_out: - mutex_unlock(&sii902x->mutex); + edid = sii902x_get_edid(sii902x, connector); + drm_connector_update_edid_property(connector, edid); + if (edid) { + num = drm_add_edid_modes(connector, edid); + kfree(edid); + } - return ret; + return num; } static enum drm_mode_status sii902x_mode_valid(struct drm_connector *connector, @@ -354,12 +360,16 @@ static void sii902x_bridge_mode_set(struct drm_bridge *bridge, const struct drm_display_mode *adj) { struct sii902x *sii902x = bridge_to_sii902x(bridge); + u8 output_mode = SII902X_SYS_CTRL_OUTPUT_DVI; struct regmap *regmap = sii902x->regmap; u8 buf[HDMI_INFOFRAME_SIZE(AVI)]; struct hdmi_avi_infoframe frame; u16 pixel_clock_10kHz = adj->clock / 10; int ret; + if (sii902x->sink_is_hdmi) + output_mode = SII902X_SYS_CTRL_OUTPUT_HDMI; + buf[0] = pixel_clock_10kHz & 0xff; buf[1] = pixel_clock_10kHz >> 8; buf[2] = drm_mode_vrefresh(adj); @@ -375,6 +385,11 @@ static void sii902x_bridge_mode_set(struct drm_bridge *bridge, mutex_lock(&sii902x->mutex); + ret = regmap_update_bits(sii902x->regmap, SII902X_SYS_CTRL_DATA, + SII902X_SYS_CTRL_OUTPUT_MODE, output_mode); + if (ret) + goto out; + ret = regmap_bulk_write(regmap, SII902X_TPI_VIDEO_DATA, buf, 10); if (ret) goto out; @@ -405,13 +420,13 @@ static int sii902x_bridge_attach(struct drm_bridge *bridge, enum drm_bridge_attach_flags flags) { struct sii902x *sii902x = bridge_to_sii902x(bridge); + u32 bus_format = MEDIA_BUS_FMT_RGB888_1X24; struct drm_device *drm = bridge->dev; int ret; - if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) { - DRM_ERROR("Fix bridge driver to make connector optional!"); - return -EINVAL; - } + if (flags & DRM_BRIDGE_ATTACH_NO_CONNECTOR) + return drm_bridge_attach(bridge->encoder, sii902x->next_bridge, + bridge, flags); drm_connector_helper_add(&sii902x->connector, &sii902x_connector_helper_funcs); @@ -433,16 +448,38 @@ static int sii902x_bridge_attach(struct drm_bridge *bridge, else sii902x->connector.polled = DRM_CONNECTOR_POLL_CONNECT; + ret = drm_display_info_set_bus_formats(&sii902x->connector.display_info, + &bus_format, 1); + if (ret) + return ret; + drm_connector_attach_encoder(&sii902x->connector, bridge->encoder); return 0; } +static enum drm_connector_status sii902x_bridge_detect(struct drm_bridge *bridge) +{ + struct sii902x *sii902x = bridge_to_sii902x(bridge); + + return sii902x_detect(sii902x); +} + +static struct edid *sii902x_bridge_get_edid(struct drm_bridge *bridge, + struct drm_connector *connector) +{ + struct sii902x *sii902x = bridge_to_sii902x(bridge); + + return sii902x_get_edid(sii902x, connector); +} + static const struct drm_bridge_funcs sii902x_bridge_funcs = { .attach = sii902x_bridge_attach, .mode_set = sii902x_bridge_mode_set, .disable = sii902x_bridge_disable, .enable = sii902x_bridge_enable, + .detect = sii902x_bridge_detect, + .get_edid = sii902x_bridge_get_edid, }; static int sii902x_mute(struct sii902x *sii902x, bool mute) @@ -829,8 +866,12 @@ static irqreturn_t sii902x_interrupt(int irq, void *data) mutex_unlock(&sii902x->mutex); - if ((status & SII902X_HOTPLUG_EVENT) && sii902x->bridge.dev) + if ((status & SII902X_HOTPLUG_EVENT) && sii902x->bridge.dev) { drm_helper_hpd_irq_event(sii902x->bridge.dev); + drm_bridge_hpd_notify(&sii902x->bridge, (status & SII902X_PLUGGED_STATUS) + ? connector_status_connected + : connector_status_disconnected); + } return IRQ_HANDLED; } @@ -1001,6 +1042,11 @@ static int sii902x_init(struct sii902x *sii902x) sii902x->bridge.funcs = &sii902x_bridge_funcs; sii902x->bridge.of_node = dev->of_node; sii902x->bridge.timings = &default_sii902x_timings; + sii902x->bridge.ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID; + + if (sii902x->i2c->irq > 0) + sii902x->bridge.ops |= DRM_BRIDGE_OP_HPD; + drm_bridge_add(&sii902x->bridge); sii902x_audio_codec_init(sii902x, dev); @@ -1022,6 +1068,7 @@ static int sii902x_probe(struct i2c_client *client, const struct i2c_device_id *id) { struct device *dev = &client->dev; + struct device_node *endpoint; struct sii902x *sii902x; int ret; @@ -1049,6 +1096,28 @@ static int sii902x_probe(struct i2c_client *client, return PTR_ERR(sii902x->reset_gpio); } + endpoint = of_graph_get_endpoint_by_regs(dev->of_node, 1, -1); + if (endpoint) { + struct device_node *remote = of_graph_get_remote_port_parent(endpoint); + + of_node_put(endpoint); + if (!remote) { + dev_err(dev, "Endpoint in port@1 unconnected\n"); + return -ENODEV; + } + + if (!of_device_is_available(remote)) { + dev_err(dev, "port@1 remote device is disabled\n"); + of_node_put(remote); + return -ENODEV; + } + + sii902x->next_bridge = of_drm_find_bridge(remote); + of_node_put(remote); + if (!sii902x->next_bridge) + return -EPROBE_DEFER; + } + mutex_init(&sii902x->mutex); sii902x->supplies[0].supply = "iovcc"; diff --git a/drivers/gpu/drm/bridge/sil-sii8620.c b/drivers/gpu/drm/bridge/sil-sii8620.c index 843265d7f1b1..ec7745c31da0 100644 --- a/drivers/gpu/drm/bridge/sil-sii8620.c +++ b/drivers/gpu/drm/bridge/sil-sii8620.c @@ -2120,7 +2120,7 @@ static void sii8620_init_rcp_input_dev(struct sii8620 *ctx) if (ret) { dev_err(ctx->dev, "Failed to register RC device\n"); ctx->error = ret; - rc_free_device(ctx->rc_dev); + rc_free_device(rc_dev); return; } ctx->rc_dev = rc_dev; diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c index 54d8fdad395f..b0d8110dd412 100644 --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi.c @@ -2540,7 +2540,7 @@ static u32 *dw_hdmi_bridge_atomic_get_output_bus_fmts(struct drm_bridge *bridge, struct drm_display_mode *mode = &crtc_state->mode; u8 max_bpc = conn_state->max_requested_bpc; bool is_hdmi2_sink = info->hdmi.scdc.supported || - (info->color_formats & DRM_COLOR_FORMAT_YCRCB420); + (info->color_formats & DRM_COLOR_FORMAT_YCBCR420); u32 *output_fmts; unsigned int i = 0; @@ -2594,36 +2594,36 @@ static u32 *dw_hdmi_bridge_atomic_get_output_bus_fmts(struct drm_bridge *bridge, */ if (max_bpc >= 16 && info->bpc == 16) { - if (info->color_formats & DRM_COLOR_FORMAT_YCRCB444) + if (info->color_formats & DRM_COLOR_FORMAT_YCBCR444) output_fmts[i++] = MEDIA_BUS_FMT_YUV16_1X48; output_fmts[i++] = MEDIA_BUS_FMT_RGB161616_1X48; } if (max_bpc >= 12 && info->bpc >= 12) { - if (info->color_formats & DRM_COLOR_FORMAT_YCRCB422) + if (info->color_formats & DRM_COLOR_FORMAT_YCBCR422) output_fmts[i++] = MEDIA_BUS_FMT_UYVY12_1X24; - if (info->color_formats & DRM_COLOR_FORMAT_YCRCB444) + if (info->color_formats & DRM_COLOR_FORMAT_YCBCR444) output_fmts[i++] = MEDIA_BUS_FMT_YUV12_1X36; output_fmts[i++] = MEDIA_BUS_FMT_RGB121212_1X36; } if (max_bpc >= 10 && info->bpc >= 10) { - if (info->color_formats & DRM_COLOR_FORMAT_YCRCB422) + if (info->color_formats & DRM_COLOR_FORMAT_YCBCR422) output_fmts[i++] = MEDIA_BUS_FMT_UYVY10_1X20; - if (info->color_formats & DRM_COLOR_FORMAT_YCRCB444) + if (info->color_formats & DRM_COLOR_FORMAT_YCBCR444) output_fmts[i++] = MEDIA_BUS_FMT_YUV10_1X30; output_fmts[i++] = MEDIA_BUS_FMT_RGB101010_1X30; } - if (info->color_formats & DRM_COLOR_FORMAT_YCRCB422) + if (info->color_formats & DRM_COLOR_FORMAT_YCBCR422) output_fmts[i++] = MEDIA_BUS_FMT_UYVY8_1X16; - if (info->color_formats & DRM_COLOR_FORMAT_YCRCB444) + if (info->color_formats & DRM_COLOR_FORMAT_YCBCR444) output_fmts[i++] = MEDIA_BUS_FMT_YUV8_1X24; /* Default 8bit RGB fallback */ diff --git a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c index e44e18a0112a..11d20b8638cd 100644 --- a/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c +++ b/drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c @@ -871,7 +871,8 @@ static void dw_mipi_dsi_clear_err(struct dw_mipi_dsi *dsi) dsi_write(dsi, DSI_INT_MSK1, 0); } -static void dw_mipi_dsi_bridge_post_disable(struct drm_bridge *bridge) +static void dw_mipi_dsi_bridge_post_atomic_disable(struct drm_bridge *bridge, + struct drm_bridge_state *old_bridge_state) { struct dw_mipi_dsi *dsi = bridge_to_dsi(bridge); const struct dw_mipi_dsi_phy_ops *phy_ops = dsi->plat_data->phy_ops; @@ -978,7 +979,8 @@ static void dw_mipi_dsi_bridge_mode_set(struct drm_bridge *bridge, dw_mipi_dsi_mode_set(dsi->slave, adjusted_mode); } -static void dw_mipi_dsi_bridge_enable(struct drm_bridge *bridge) +static void dw_mipi_dsi_bridge_atomic_enable(struct drm_bridge *bridge, + struct drm_bridge_state *old_bridge_state) { struct dw_mipi_dsi *dsi = bridge_to_dsi(bridge); @@ -998,7 +1000,10 @@ dw_mipi_dsi_bridge_mode_valid(struct drm_bridge *bridge, enum drm_mode_status mode_status = MODE_OK; if (pdata->mode_valid) - mode_status = pdata->mode_valid(pdata->priv_data, mode); + mode_status = pdata->mode_valid(pdata->priv_data, mode, + dsi->mode_flags, + dw_mipi_dsi_get_lanes(dsi), + dsi->format); return mode_status; } @@ -1032,11 +1037,14 @@ static int dw_mipi_dsi_bridge_attach(struct drm_bridge *bridge, } static const struct drm_bridge_funcs dw_mipi_dsi_bridge_funcs = { - .mode_set = dw_mipi_dsi_bridge_mode_set, - .enable = dw_mipi_dsi_bridge_enable, - .post_disable = dw_mipi_dsi_bridge_post_disable, - .mode_valid = dw_mipi_dsi_bridge_mode_valid, - .attach = dw_mipi_dsi_bridge_attach, + .atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state, + .atomic_destroy_state = drm_atomic_helper_bridge_destroy_state, + .atomic_reset = drm_atomic_helper_bridge_reset, + .atomic_enable = dw_mipi_dsi_bridge_atomic_enable, + .atomic_post_disable = dw_mipi_dsi_bridge_post_atomic_disable, + .mode_set = dw_mipi_dsi_bridge_mode_set, + .mode_valid = dw_mipi_dsi_bridge_mode_valid, + .attach = dw_mipi_dsi_bridge_attach, }; #ifdef CONFIG_DEBUG_FS @@ -1199,6 +1207,7 @@ __dw_mipi_dsi_probe(struct platform_device *pdev, ret = mipi_dsi_host_register(&dsi->dsi_host); if (ret) { dev_err(dev, "Failed to register MIPI host: %d\n", ret); + pm_runtime_disable(dev); dw_mipi_dsi_debugfs_remove(dsi); return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/bridge/tc358767.c b/drivers/gpu/drm/bridge/tc358767.c index 23a6f90b694b..c23e0abc65e8 100644 --- a/drivers/gpu/drm/bridge/tc358767.c +++ b/drivers/gpu/drm/bridge/tc358767.c @@ -27,7 +27,7 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> diff --git a/drivers/gpu/drm/bridge/tc358775.c b/drivers/gpu/drm/bridge/tc358775.c index 2c76331b251d..695af3badcc7 100644 --- a/drivers/gpu/drm/bridge/tc358775.c +++ b/drivers/gpu/drm/bridge/tc358775.c @@ -22,7 +22,7 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> #include <drm/drm_crtc_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_mipi_dsi.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> @@ -241,7 +241,7 @@ static inline u32 TC358775_LVCFG_PCLKDIV(uint32_t val) } #define TC358775_LVCFG_LVDLINK__MASK 0x00000002 -#define TC358775_LVCFG_LVDLINK__SHIFT 0 +#define TC358775_LVCFG_LVDLINK__SHIFT 1 static inline u32 TC358775_LVCFG_LVDLINK(uint32_t val) { return ((val) << TC358775_LVCFG_LVDLINK__SHIFT) & diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi83.c b/drivers/gpu/drm/bridge/ti-sn65dsi83.c index 945f08de45f1..19daaddd29a4 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi83.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi83.c @@ -33,6 +33,7 @@ #include <linux/of_device.h> #include <linux/of_graph.h> #include <linux/regmap.h> +#include <linux/regulator/consumer.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> @@ -143,6 +144,7 @@ struct sn65dsi83 { struct mipi_dsi_device *dsi; struct drm_bridge *panel_bridge; struct gpio_desc *enable_gpio; + struct regulator *vcc; int dsi_lanes; bool lvds_dual_link; bool lvds_dual_link_even_odd_swap; @@ -337,6 +339,12 @@ static void sn65dsi83_atomic_enable(struct drm_bridge *bridge, u16 val; int ret; + ret = regulator_enable(ctx->vcc); + if (ret) { + dev_err(ctx->dev, "Failed to enable vcc: %d\n", ret); + return; + } + /* Deassert reset */ gpiod_set_value(ctx->enable_gpio, 1); usleep_range(1000, 1100); @@ -486,11 +494,16 @@ static void sn65dsi83_atomic_disable(struct drm_bridge *bridge, struct drm_bridge_state *old_bridge_state) { struct sn65dsi83 *ctx = bridge_to_sn65dsi83(bridge); + int ret; /* Put the chip in reset, pull EN line low, and assure 10ms reset low timing. */ gpiod_set_value(ctx->enable_gpio, 0); usleep_range(10000, 11000); + ret = regulator_disable(ctx->vcc); + if (ret) + dev_err(ctx->dev, "Failed to disable vcc: %d\n", ret); + regcache_mark_dirty(ctx->regmap); } @@ -560,10 +573,14 @@ static int sn65dsi83_parse_dt(struct sn65dsi83 *ctx, enum sn65dsi83_model model) ctx->host_node = of_graph_get_remote_port_parent(endpoint); of_node_put(endpoint); - if (ctx->dsi_lanes < 0 || ctx->dsi_lanes > 4) - return -EINVAL; - if (!ctx->host_node) - return -ENODEV; + if (ctx->dsi_lanes < 0 || ctx->dsi_lanes > 4) { + ret = -EINVAL; + goto err_put_node; + } + if (!ctx->host_node) { + ret = -ENODEV; + goto err_put_node; + } ctx->lvds_dual_link = false; ctx->lvds_dual_link_even_odd_swap = false; @@ -590,16 +607,27 @@ static int sn65dsi83_parse_dt(struct sn65dsi83 *ctx, enum sn65dsi83_model model) ret = drm_of_find_panel_or_bridge(dev->of_node, 2, 0, &panel, &panel_bridge); if (ret < 0) - return ret; + goto err_put_node; if (panel) { panel_bridge = devm_drm_panel_bridge_add(dev, panel); - if (IS_ERR(panel_bridge)) - return PTR_ERR(panel_bridge); + if (IS_ERR(panel_bridge)) { + ret = PTR_ERR(panel_bridge); + goto err_put_node; + } } ctx->panel_bridge = panel_bridge; + ctx->vcc = devm_regulator_get(dev, "vcc"); + if (IS_ERR(ctx->vcc)) + return dev_err_probe(dev, PTR_ERR(ctx->vcc), + "Failed to get supply 'vcc'\n"); + return 0; + +err_put_node: + of_node_put(ctx->host_node); + return ret; } static int sn65dsi83_host_attach(struct sn65dsi83 *ctx) @@ -662,7 +690,8 @@ static int sn65dsi83_probe(struct i2c_client *client, } /* Put the chip in reset, pull EN line low, and assure 10ms reset low timing. */ - ctx->enable_gpio = devm_gpiod_get(ctx->dev, "enable", GPIOD_OUT_LOW); + ctx->enable_gpio = devm_gpiod_get_optional(ctx->dev, "enable", + GPIOD_OUT_LOW); if (IS_ERR(ctx->enable_gpio)) return PTR_ERR(ctx->enable_gpio); @@ -673,8 +702,10 @@ static int sn65dsi83_probe(struct i2c_client *client, return ret; ctx->regmap = devm_regmap_init_i2c(client, &sn65dsi83_regmap_config); - if (IS_ERR(ctx->regmap)) - return PTR_ERR(ctx->regmap); + if (IS_ERR(ctx->regmap)) { + ret = PTR_ERR(ctx->regmap); + goto err_put_node; + } dev_set_drvdata(dev, ctx); i2c_set_clientdata(client, ctx); @@ -691,6 +722,8 @@ static int sn65dsi83_probe(struct i2c_client *client, err_remove_bridge: drm_bridge_remove(&ctx->bridge); +err_put_node: + of_node_put(ctx->host_node); return ret; } diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c index dab8f76618f3..ba136a188be7 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c @@ -26,8 +26,8 @@ #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> -#include <drm/drm_dp_aux_bus.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_aux_bus.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_mipi_dsi.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> diff --git a/drivers/gpu/drm/dp/Makefile b/drivers/gpu/drm/dp/Makefile new file mode 100644 index 000000000000..75faffc706b1 --- /dev/null +++ b/drivers/gpu/drm/dp/Makefile @@ -0,0 +1,9 @@ +# SPDX-License-Identifier: MIT + +obj-$(CONFIG_DRM_DP_AUX_BUS) += drm_dp_aux_bus.o + +drm_dp_helper-y := drm_dp.o drm_dp_dual_mode_helper.o drm_dp_helper_mod.o drm_dp_mst_topology.o +drm_dp_helper-$(CONFIG_DRM_DP_AUX_CHARDEV) += drm_dp_aux_dev.o +drm_dp_helper-$(CONFIG_DRM_DP_CEC) += drm_dp_cec.o + +obj-$(CONFIG_DRM_DP_HELPER) += drm_dp_helper.o diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/dp/drm_dp.c index 23f9073bc473..6d43325acca5 100644 --- a/drivers/gpu/drm/drm_dp_helper.c +++ b/drivers/gpu/drm/dp/drm_dp.c @@ -29,13 +29,13 @@ #include <linux/sched.h> #include <linux/seq_file.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_print.h> #include <drm/drm_vblank.h> -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_panel.h> -#include "drm_crtc_helper_internal.h" +#include "drm_dp_helper_internal.h" struct dp_aux_backlight { struct backlight_device *base; diff --git a/drivers/gpu/drm/drm_dp_aux_bus.c b/drivers/gpu/drm/dp/drm_dp_aux_bus.c index 298ea7a49591..415afce3cf96 100644 --- a/drivers/gpu/drm/drm_dp_aux_bus.c +++ b/drivers/gpu/drm/dp/drm_dp_aux_bus.c @@ -19,8 +19,8 @@ #include <linux/pm_domain.h> #include <linux/pm_runtime.h> -#include <drm/drm_dp_aux_bus.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_aux_bus.h> +#include <drm/dp/drm_dp_helper.h> /** * dp_aux_ep_match() - The match function for the dp_aux_bus. diff --git a/drivers/gpu/drm/drm_dp_aux_dev.c b/drivers/gpu/drm/dp/drm_dp_aux_dev.c index 06b374cae956..53ad4e72790b 100644 --- a/drivers/gpu/drm/drm_dp_aux_dev.c +++ b/drivers/gpu/drm/dp/drm_dp_aux_dev.c @@ -36,11 +36,11 @@ #include <linux/uio.h> #include <drm/drm_crtc.h> -#include <drm/drm_dp_helper.h> -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_print.h> -#include "drm_crtc_helper_internal.h" +#include "drm_dp_helper_internal.h" struct drm_dp_aux_dev { unsigned index; diff --git a/drivers/gpu/drm/drm_dp_cec.c b/drivers/gpu/drm/dp/drm_dp_cec.c index 3ab2609f9ec7..f9e927355879 100644 --- a/drivers/gpu/drm/drm_dp_cec.c +++ b/drivers/gpu/drm/dp/drm_dp_cec.c @@ -13,7 +13,7 @@ #include <drm/drm_connector.h> #include <drm/drm_device.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> /* * Unfortunately it turns out that we have a chicken-and-egg situation diff --git a/drivers/gpu/drm/drm_dp_dual_mode_helper.c b/drivers/gpu/drm/dp/drm_dp_dual_mode_helper.c index 9faf49354cab..2049cb0f7ed0 100644 --- a/drivers/gpu/drm/drm_dp_dual_mode_helper.c +++ b/drivers/gpu/drm/dp/drm_dp_dual_mode_helper.c @@ -28,7 +28,7 @@ #include <linux/string.h> #include <drm/drm_device.h> -#include <drm/drm_dp_dual_mode_helper.h> +#include <drm/dp/drm_dp_dual_mode_helper.h> #include <drm/drm_print.h> /** diff --git a/drivers/gpu/drm/dp/drm_dp_helper_internal.h b/drivers/gpu/drm/dp/drm_dp_helper_internal.h new file mode 100644 index 000000000000..8917fc3af9ec --- /dev/null +++ b/drivers/gpu/drm/dp/drm_dp_helper_internal.h @@ -0,0 +1,33 @@ +/* SPDX-License-Identifier: MIT */ + +#ifndef DRM_DP_HELPER_INTERNAL_H +#define DRM_DP_HELPER_INTERNAL_H + +struct drm_dp_aux; + +#ifdef CONFIG_DRM_DP_AUX_CHARDEV +int drm_dp_aux_dev_init(void); +void drm_dp_aux_dev_exit(void); +int drm_dp_aux_register_devnode(struct drm_dp_aux *aux); +void drm_dp_aux_unregister_devnode(struct drm_dp_aux *aux); +#else +static inline int drm_dp_aux_dev_init(void) +{ + return 0; +} + +static inline void drm_dp_aux_dev_exit(void) +{ +} + +static inline int drm_dp_aux_register_devnode(struct drm_dp_aux *aux) +{ + return 0; +} + +static inline void drm_dp_aux_unregister_devnode(struct drm_dp_aux *aux) +{ +} +#endif + +#endif diff --git a/drivers/gpu/drm/dp/drm_dp_helper_mod.c b/drivers/gpu/drm/dp/drm_dp_helper_mod.c new file mode 100644 index 000000000000..db753de24000 --- /dev/null +++ b/drivers/gpu/drm/dp/drm_dp_helper_mod.c @@ -0,0 +1,22 @@ +// SPDX-License-Identifier: MIT + +#include <linux/module.h> + +#include "drm_dp_helper_internal.h" + +MODULE_DESCRIPTION("DRM DisplayPort helper"); +MODULE_LICENSE("GPL and additional rights"); + +static int __init drm_dp_helper_module_init(void) +{ + return drm_dp_aux_dev_init(); +} + +static void __exit drm_dp_helper_module_exit(void) +{ + /* Call exit functions from specific dp helpers here */ + drm_dp_aux_dev_exit(); +} + +module_init(drm_dp_helper_module_init); +module_exit(drm_dp_helper_module_exit); diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/dp/drm_dp_mst_topology.c index 8b3822142fed..11300b53d24f 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/dp/drm_dp_mst_topology.c @@ -38,14 +38,14 @@ #include <linux/math64.h> #endif +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> -#include <drm/drm_dp_mst_helper.h> #include <drm/drm_drv.h> #include <drm/drm_print.h> #include <drm/drm_probe_helper.h> -#include "drm_crtc_helper_internal.h" +#include "drm_dp_helper_internal.h" #include "drm_dp_mst_topology_internal.h" /** @@ -4196,7 +4196,7 @@ int drm_dp_mst_hpd_irq(struct drm_dp_mst_topology_mgr *mgr, u8 *esi, bool *handl int ret = 0; int sc; *handled = false; - sc = esi[0] & 0x3f; + sc = DP_GET_SINK_COUNT(esi[0]); if (sc != mgr->sink_count) { mgr->sink_count = sc; @@ -4811,7 +4811,7 @@ static void drm_dp_mst_dump_mstb(struct seq_file *m, seq_printf(m, "%smstb - [%p]: num_ports: %d\n", prefix, mstb, mstb->num_ports); list_for_each_entry(port, &mstb->ports, next) { - seq_printf(m, "%sport %d - [%p] (%s - %s): ddps: %d, ldps: %d, sdp: %d/%d, fec: %s, conn: %p\n", + seq_printf(m, "%sport %d - [%p] (%s - %s): ddps: %d, ldps: %d, sdp: %d/%d, fec: %s, conn: %p\n", prefix, port->port_num, port, diff --git a/drivers/gpu/drm/drm_dp_mst_topology_internal.h b/drivers/gpu/drm/dp/drm_dp_mst_topology_internal.h index eeda9a61c657..401953b59d45 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology_internal.h +++ b/drivers/gpu/drm/dp/drm_dp_mst_topology_internal.h @@ -10,7 +10,7 @@ #ifndef _DRM_DP_MST_HELPER_INTERNAL_H_ #define _DRM_DP_MST_HELPER_INTERNAL_H_ -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> void drm_dp_encode_sideband_req(const struct drm_dp_sideband_msg_req_body *req, diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c new file mode 100644 index 000000000000..d60878bc9c20 --- /dev/null +++ b/drivers/gpu/drm/drm_buddy.c @@ -0,0 +1,535 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2021 Intel Corporation + */ + +#include <linux/kmemleak.h> +#include <linux/module.h> +#include <linux/sizes.h> + +#include <drm/drm_buddy.h> + +static struct kmem_cache *slab_blocks; + +static struct drm_buddy_block *drm_block_alloc(struct drm_buddy *mm, + struct drm_buddy_block *parent, + unsigned int order, + u64 offset) +{ + struct drm_buddy_block *block; + + BUG_ON(order > DRM_BUDDY_MAX_ORDER); + + block = kmem_cache_zalloc(slab_blocks, GFP_KERNEL); + if (!block) + return NULL; + + block->header = offset; + block->header |= order; + block->parent = parent; + + BUG_ON(block->header & DRM_BUDDY_HEADER_UNUSED); + return block; +} + +static void drm_block_free(struct drm_buddy *mm, + struct drm_buddy_block *block) +{ + kmem_cache_free(slab_blocks, block); +} + +static void mark_allocated(struct drm_buddy_block *block) +{ + block->header &= ~DRM_BUDDY_HEADER_STATE; + block->header |= DRM_BUDDY_ALLOCATED; + + list_del(&block->link); +} + +static void mark_free(struct drm_buddy *mm, + struct drm_buddy_block *block) +{ + block->header &= ~DRM_BUDDY_HEADER_STATE; + block->header |= DRM_BUDDY_FREE; + + list_add(&block->link, + &mm->free_list[drm_buddy_block_order(block)]); +} + +static void mark_split(struct drm_buddy_block *block) +{ + block->header &= ~DRM_BUDDY_HEADER_STATE; + block->header |= DRM_BUDDY_SPLIT; + + list_del(&block->link); +} + +/** + * drm_buddy_init - init memory manager + * + * @mm: DRM buddy manager to initialize + * @size: size in bytes to manage + * @chunk_size: minimum page size in bytes for our allocations + * + * Initializes the memory manager and its resources. + * + * Returns: + * 0 on success, error code on failure. + */ +int drm_buddy_init(struct drm_buddy *mm, u64 size, u64 chunk_size) +{ + unsigned int i; + u64 offset; + + if (size < chunk_size) + return -EINVAL; + + if (chunk_size < PAGE_SIZE) + return -EINVAL; + + if (!is_power_of_2(chunk_size)) + return -EINVAL; + + size = round_down(size, chunk_size); + + mm->size = size; + mm->avail = size; + mm->chunk_size = chunk_size; + mm->max_order = ilog2(size) - ilog2(chunk_size); + + BUG_ON(mm->max_order > DRM_BUDDY_MAX_ORDER); + + mm->free_list = kmalloc_array(mm->max_order + 1, + sizeof(struct list_head), + GFP_KERNEL); + if (!mm->free_list) + return -ENOMEM; + + for (i = 0; i <= mm->max_order; ++i) + INIT_LIST_HEAD(&mm->free_list[i]); + + mm->n_roots = hweight64(size); + + mm->roots = kmalloc_array(mm->n_roots, + sizeof(struct drm_buddy_block *), + GFP_KERNEL); + if (!mm->roots) + goto out_free_list; + + offset = 0; + i = 0; + + /* + * Split into power-of-two blocks, in case we are given a size that is + * not itself a power-of-two. + */ + do { + struct drm_buddy_block *root; + unsigned int order; + u64 root_size; + + root_size = rounddown_pow_of_two(size); + order = ilog2(root_size) - ilog2(chunk_size); + + root = drm_block_alloc(mm, NULL, order, offset); + if (!root) + goto out_free_roots; + + mark_free(mm, root); + + BUG_ON(i > mm->max_order); + BUG_ON(drm_buddy_block_size(mm, root) < chunk_size); + + mm->roots[i] = root; + + offset += root_size; + size -= root_size; + i++; + } while (size); + + return 0; + +out_free_roots: + while (i--) + drm_block_free(mm, mm->roots[i]); + kfree(mm->roots); +out_free_list: + kfree(mm->free_list); + return -ENOMEM; +} +EXPORT_SYMBOL(drm_buddy_init); + +/** + * drm_buddy_fini - tear down the memory manager + * + * @mm: DRM buddy manager to free + * + * Cleanup memory manager resources and the freelist + */ +void drm_buddy_fini(struct drm_buddy *mm) +{ + int i; + + for (i = 0; i < mm->n_roots; ++i) { + WARN_ON(!drm_buddy_block_is_free(mm->roots[i])); + drm_block_free(mm, mm->roots[i]); + } + + WARN_ON(mm->avail != mm->size); + + kfree(mm->roots); + kfree(mm->free_list); +} +EXPORT_SYMBOL(drm_buddy_fini); + +static int split_block(struct drm_buddy *mm, + struct drm_buddy_block *block) +{ + unsigned int block_order = drm_buddy_block_order(block) - 1; + u64 offset = drm_buddy_block_offset(block); + + BUG_ON(!drm_buddy_block_is_free(block)); + BUG_ON(!drm_buddy_block_order(block)); + + block->left = drm_block_alloc(mm, block, block_order, offset); + if (!block->left) + return -ENOMEM; + + block->right = drm_block_alloc(mm, block, block_order, + offset + (mm->chunk_size << block_order)); + if (!block->right) { + drm_block_free(mm, block->left); + return -ENOMEM; + } + + mark_free(mm, block->left); + mark_free(mm, block->right); + + mark_split(block); + + return 0; +} + +static struct drm_buddy_block * +get_buddy(struct drm_buddy_block *block) +{ + struct drm_buddy_block *parent; + + parent = block->parent; + if (!parent) + return NULL; + + if (parent->left == block) + return parent->right; + + return parent->left; +} + +static void __drm_buddy_free(struct drm_buddy *mm, + struct drm_buddy_block *block) +{ + struct drm_buddy_block *parent; + + while ((parent = block->parent)) { + struct drm_buddy_block *buddy; + + buddy = get_buddy(block); + + if (!drm_buddy_block_is_free(buddy)) + break; + + list_del(&buddy->link); + + drm_block_free(mm, block); + drm_block_free(mm, buddy); + + block = parent; + } + + mark_free(mm, block); +} + +/** + * drm_buddy_free_block - free a block + * + * @mm: DRM buddy manager + * @block: block to be freed + */ +void drm_buddy_free_block(struct drm_buddy *mm, + struct drm_buddy_block *block) +{ + BUG_ON(!drm_buddy_block_is_allocated(block)); + mm->avail += drm_buddy_block_size(mm, block); + __drm_buddy_free(mm, block); +} +EXPORT_SYMBOL(drm_buddy_free_block); + +/** + * drm_buddy_free_list - free blocks + * + * @mm: DRM buddy manager + * @objects: input list head to free blocks + */ +void drm_buddy_free_list(struct drm_buddy *mm, struct list_head *objects) +{ + struct drm_buddy_block *block, *on; + + list_for_each_entry_safe(block, on, objects, link) { + drm_buddy_free_block(mm, block); + cond_resched(); + } + INIT_LIST_HEAD(objects); +} +EXPORT_SYMBOL(drm_buddy_free_list); + +/** + * drm_buddy_alloc_blocks - allocate power-of-two blocks + * + * @mm: DRM buddy manager to allocate from + * @order: size of the allocation + * + * The order value here translates to: + * + * 0 = 2^0 * mm->chunk_size + * 1 = 2^1 * mm->chunk_size + * 2 = 2^2 * mm->chunk_size + * + * Returns: + * allocated ptr to the &drm_buddy_block on success + */ +struct drm_buddy_block * +drm_buddy_alloc_blocks(struct drm_buddy *mm, unsigned int order) +{ + struct drm_buddy_block *block = NULL; + unsigned int i; + int err; + + for (i = order; i <= mm->max_order; ++i) { + block = list_first_entry_or_null(&mm->free_list[i], + struct drm_buddy_block, + link); + if (block) + break; + } + + if (!block) + return ERR_PTR(-ENOSPC); + + BUG_ON(!drm_buddy_block_is_free(block)); + + while (i != order) { + err = split_block(mm, block); + if (unlikely(err)) + goto out_free; + + /* Go low */ + block = block->left; + i--; + } + + mark_allocated(block); + mm->avail -= drm_buddy_block_size(mm, block); + kmemleak_update_trace(block); + return block; + +out_free: + if (i != order) + __drm_buddy_free(mm, block); + return ERR_PTR(err); +} +EXPORT_SYMBOL(drm_buddy_alloc_blocks); + +static inline bool overlaps(u64 s1, u64 e1, u64 s2, u64 e2) +{ + return s1 <= e2 && e1 >= s2; +} + +static inline bool contains(u64 s1, u64 e1, u64 s2, u64 e2) +{ + return s1 <= s2 && e1 >= e2; +} + +/** + * drm_buddy_alloc_range - allocate range + * + * @mm: DRM buddy manager to allocate from + * @blocks: output list head to add allocated blocks + * @start: start of the allowed range for this block + * @size: size of the allocation + * + * Intended for pre-allocating portions of the address space, for example to + * reserve a block for the initial framebuffer or similar, hence the expectation + * here is that drm_buddy_alloc_blocks() is still the main vehicle for + * allocations, so if that's not the case then the drm_mm range allocator is + * probably a much better fit, and so you should probably go use that instead. + * + * Note that it's safe to chain together multiple alloc_ranges + * with the same blocks list + * + * Returns: + * 0 on success, error code on failure. + */ +int drm_buddy_alloc_range(struct drm_buddy *mm, + struct list_head *blocks, + u64 start, u64 size) +{ + struct drm_buddy_block *block; + struct drm_buddy_block *buddy; + LIST_HEAD(allocated); + LIST_HEAD(dfs); + u64 end; + int err; + int i; + + if (size < mm->chunk_size) + return -EINVAL; + + if (!IS_ALIGNED(size | start, mm->chunk_size)) + return -EINVAL; + + if (range_overflows(start, size, mm->size)) + return -EINVAL; + + for (i = 0; i < mm->n_roots; ++i) + list_add_tail(&mm->roots[i]->tmp_link, &dfs); + + end = start + size - 1; + + do { + u64 block_start; + u64 block_end; + + block = list_first_entry_or_null(&dfs, + struct drm_buddy_block, + tmp_link); + if (!block) + break; + + list_del(&block->tmp_link); + + block_start = drm_buddy_block_offset(block); + block_end = block_start + drm_buddy_block_size(mm, block) - 1; + + if (!overlaps(start, end, block_start, block_end)) + continue; + + if (drm_buddy_block_is_allocated(block)) { + err = -ENOSPC; + goto err_free; + } + + if (contains(start, end, block_start, block_end)) { + if (!drm_buddy_block_is_free(block)) { + err = -ENOSPC; + goto err_free; + } + + mark_allocated(block); + mm->avail -= drm_buddy_block_size(mm, block); + list_add_tail(&block->link, &allocated); + continue; + } + + if (!drm_buddy_block_is_split(block)) { + err = split_block(mm, block); + if (unlikely(err)) + goto err_undo; + } + + list_add(&block->right->tmp_link, &dfs); + list_add(&block->left->tmp_link, &dfs); + } while (1); + + list_splice_tail(&allocated, blocks); + return 0; + +err_undo: + /* + * We really don't want to leave around a bunch of split blocks, since + * bigger is better, so make sure we merge everything back before we + * free the allocated blocks. + */ + buddy = get_buddy(block); + if (buddy && + (drm_buddy_block_is_free(block) && + drm_buddy_block_is_free(buddy))) + __drm_buddy_free(mm, block); + +err_free: + drm_buddy_free_list(mm, &allocated); + return err; +} +EXPORT_SYMBOL(drm_buddy_alloc_range); + +/** + * drm_buddy_block_print - print block information + * + * @mm: DRM buddy manager + * @block: DRM buddy block + * @p: DRM printer to use + */ +void drm_buddy_block_print(struct drm_buddy *mm, + struct drm_buddy_block *block, + struct drm_printer *p) +{ + u64 start = drm_buddy_block_offset(block); + u64 size = drm_buddy_block_size(mm, block); + + drm_printf(p, "%#018llx-%#018llx: %llu\n", start, start + size, size); +} +EXPORT_SYMBOL(drm_buddy_block_print); + +/** + * drm_buddy_print - print allocator state + * + * @mm: DRM buddy manager + * @p: DRM printer to use + */ +void drm_buddy_print(struct drm_buddy *mm, struct drm_printer *p) +{ + int order; + + drm_printf(p, "chunk_size: %lluKiB, total: %lluMiB, free: %lluMiB\n", + mm->chunk_size >> 10, mm->size >> 20, mm->avail >> 20); + + for (order = mm->max_order; order >= 0; order--) { + struct drm_buddy_block *block; + u64 count = 0, free; + + list_for_each_entry(block, &mm->free_list[order], link) { + BUG_ON(!drm_buddy_block_is_free(block)); + count++; + } + + drm_printf(p, "order-%d ", order); + + free = count * (mm->chunk_size << order); + if (free < SZ_1M) + drm_printf(p, "free: %lluKiB", free >> 10); + else + drm_printf(p, "free: %lluMiB", free >> 20); + + drm_printf(p, ", pages: %llu\n", count); + } +} +EXPORT_SYMBOL(drm_buddy_print); + +static void drm_buddy_module_exit(void) +{ + kmem_cache_destroy(slab_blocks); +} + +static int __init drm_buddy_module_init(void) +{ + slab_blocks = KMEM_CACHE(drm_buddy_block, 0); + if (!slab_blocks) + return -ENOMEM; + + return 0; +} + +module_init(drm_buddy_module_init); +module_exit(drm_buddy_module_exit); + +MODULE_DESCRIPTION("DRM Buddy Allocator"); +MODULE_LICENSE("Dual MIT/GPL"); diff --git a/drivers/gpu/drm/drm_cache.c b/drivers/gpu/drm/drm_cache.c index f19d9acbe959..4bb093ccf1b8 100644 --- a/drivers/gpu/drm/drm_cache.c +++ b/drivers/gpu/drm/drm_cache.c @@ -28,10 +28,10 @@ * Authors: Thomas Hellström <thomas-at-tungstengraphics-dot-com> */ -#include <linux/dma-buf-map.h> +#include <linux/cc_platform.h> #include <linux/export.h> #include <linux/highmem.h> -#include <linux/cc_platform.h> +#include <linux/iosys-map.h> #include <xen/xen.h> #include <drm/drm_cache.h> @@ -214,14 +214,14 @@ bool drm_need_swiotlb(int dma_bits) } EXPORT_SYMBOL(drm_need_swiotlb); -static void memcpy_fallback(struct dma_buf_map *dst, - const struct dma_buf_map *src, +static void memcpy_fallback(struct iosys_map *dst, + const struct iosys_map *src, unsigned long len) { if (!dst->is_iomem && !src->is_iomem) { memcpy(dst->vaddr, src->vaddr, len); } else if (!src->is_iomem) { - dma_buf_map_memcpy_to(dst, src->vaddr, len); + iosys_map_memcpy_to(dst, src->vaddr, len); } else if (!dst->is_iomem) { memcpy_fromio(dst->vaddr, src->vaddr_iomem, len); } else { @@ -305,8 +305,8 @@ static void __drm_memcpy_from_wc(void *dst, const void *src, unsigned long len) * Tries an arch optimized memcpy for prefetching reading out of a WC region, * and if no such beast is available, falls back to a normal memcpy. */ -void drm_memcpy_from_wc(struct dma_buf_map *dst, - const struct dma_buf_map *src, +void drm_memcpy_from_wc(struct iosys_map *dst, + const struct iosys_map *src, unsigned long len) { if (WARN_ON(in_interrupt())) { @@ -343,8 +343,8 @@ void drm_memcpy_init_early(void) static_branch_enable(&has_movntdqa); } #else -void drm_memcpy_from_wc(struct dma_buf_map *dst, - const struct dma_buf_map *src, +void drm_memcpy_from_wc(struct iosys_map *dst, + const struct iosys_map *src, unsigned long len) { WARN_ON(in_interrupt()); diff --git a/drivers/gpu/drm/drm_client.c b/drivers/gpu/drm/drm_client.c index ce45e380f4a2..af3b7395bf69 100644 --- a/drivers/gpu/drm/drm_client.c +++ b/drivers/gpu/drm/drm_client.c @@ -3,7 +3,7 @@ * Copyright 2018 Noralf Trønnes */ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/list.h> #include <linux/module.h> #include <linux/mutex.h> @@ -309,9 +309,10 @@ err_delete: * 0 on success, or a negative errno code otherwise. */ int -drm_client_buffer_vmap(struct drm_client_buffer *buffer, struct dma_buf_map *map_copy) +drm_client_buffer_vmap(struct drm_client_buffer *buffer, + struct iosys_map *map_copy) { - struct dma_buf_map *map = &buffer->map; + struct iosys_map *map = &buffer->map; int ret; /* @@ -342,7 +343,7 @@ EXPORT_SYMBOL(drm_client_buffer_vmap); */ void drm_client_buffer_vunmap(struct drm_client_buffer *buffer) { - struct dma_buf_map *map = &buffer->map; + struct iosys_map *map = &buffer->map; drm_gem_vunmap(buffer->gem, map); } diff --git a/drivers/gpu/drm/drm_color_mgmt.c b/drivers/gpu/drm/drm_color_mgmt.c index bb14f488c8f6..9079fbe21d2f 100644 --- a/drivers/gpu/drm/drm_color_mgmt.c +++ b/drivers/gpu/drm/drm_color_mgmt.c @@ -82,6 +82,10 @@ * driver boot-up state too. Drivers can access this blob through * &drm_crtc_state.gamma_lut. * + * Note that for mostly historical reasons stemming from Xorg heritage, + * this is also used to store the color map (also sometimes color lut, CLUT + * or color palette) for indexed formats like DRM_FORMAT_C8. + * * “GAMMA_LUT_SIZE”: * Unsigned range property to give the size of the lookup table to be set * on the GAMMA_LUT property (the size depends on the underlying hardware). diff --git a/drivers/gpu/drm/drm_crtc_helper_internal.h b/drivers/gpu/drm/drm_crtc_helper_internal.h index 61e09f8a8d0f..28e04e750130 100644 --- a/drivers/gpu/drm/drm_crtc_helper_internal.h +++ b/drivers/gpu/drm/drm_crtc_helper_internal.h @@ -28,36 +28,9 @@ #include <drm/drm_connector.h> #include <drm/drm_crtc.h> -#include <drm/drm_dp_helper.h> #include <drm/drm_encoder.h> #include <drm/drm_modes.h> -/* drm_dp_aux_dev.c */ -#ifdef CONFIG_DRM_DP_AUX_CHARDEV -int drm_dp_aux_dev_init(void); -void drm_dp_aux_dev_exit(void); -int drm_dp_aux_register_devnode(struct drm_dp_aux *aux); -void drm_dp_aux_unregister_devnode(struct drm_dp_aux *aux); -#else -static inline int drm_dp_aux_dev_init(void) -{ - return 0; -} - -static inline void drm_dp_aux_dev_exit(void) -{ -} - -static inline int drm_dp_aux_register_devnode(struct drm_dp_aux *aux) -{ - return 0; -} - -static inline void drm_dp_aux_unregister_devnode(struct drm_dp_aux *aux) -{ -} -#endif - /* drm_probe_helper.c */ enum drm_mode_status drm_crtc_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode *mode); diff --git a/drivers/gpu/drm/drm_dsc.c b/drivers/gpu/drm/drm_dsc.c index 46a3c1b62463..fdd8d5f42622 100644 --- a/drivers/gpu/drm/drm_dsc.c +++ b/drivers/gpu/drm/drm_dsc.c @@ -12,7 +12,7 @@ #include <linux/errno.h> #include <linux/byteorder/generic.h> #include <drm/drm_print.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_dsc.h> /** diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index 12893e7be89b..a504542238ed 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -93,6 +93,8 @@ static int oui(u8 first, u8 second, u8 third) /* Non desktop display (i.e. HMD) */ #define EDID_QUIRK_NON_DESKTOP (1 << 12) +#define MICROSOFT_IEEE_OUI 0xca125c + struct detailed_mode_closure { struct drm_connector *connector; struct edid *edid; @@ -212,9 +214,7 @@ static const struct edid_quirk { /* Windows Mixed Reality Headsets */ EDID_QUIRK('A', 'C', 'R', 0x7fce, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('H', 'P', 'N', 0x3515, EDID_QUIRK_NON_DESKTOP), EDID_QUIRK('L', 'E', 'N', 0x0408, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('L', 'E', 'N', 0xb800, EDID_QUIRK_NON_DESKTOP), EDID_QUIRK('F', 'U', 'J', 0x1970, EDID_QUIRK_NON_DESKTOP), EDID_QUIRK('D', 'E', 'L', 0x7fce, EDID_QUIRK_NON_DESKTOP), EDID_QUIRK('S', 'E', 'C', 0x144a, EDID_QUIRK_NON_DESKTOP), @@ -3776,7 +3776,7 @@ static int do_y420vdb_modes(struct drm_connector *connector, } if (modes > 0) - info->color_formats |= DRM_COLOR_FORMAT_YCRCB420; + info->color_formats |= DRM_COLOR_FORMAT_YCBCR420; return modes; } @@ -4222,6 +4222,17 @@ static bool cea_db_is_hdmi_forum_vsdb(const u8 *db) return oui(db[3], db[2], db[1]) == HDMI_FORUM_IEEE_OUI; } +static bool cea_db_is_microsoft_vsdb(const u8 *db) +{ + if (cea_db_tag(db) != VENDOR_BLOCK) + return false; + + if (cea_db_payload_len(db) != 21) + return false; + + return oui(db[3], db[2], db[1]) == MICROSOFT_IEEE_OUI; +} + static bool cea_db_is_vcdb(const u8 *db) { if (cea_db_tag(db) != USE_EXTENDED_TAG) @@ -4279,7 +4290,7 @@ static void drm_parse_y420cmdb_bitmap(struct drm_connector *connector, if (map_len == 0) { /* All CEA modes support ycbcr420 sampling also.*/ hdmi->y420_cmdb_map = U64_MAX; - info->color_formats |= DRM_COLOR_FORMAT_YCRCB420; + info->color_formats |= DRM_COLOR_FORMAT_YCBCR420; return; } @@ -4302,7 +4313,7 @@ static void drm_parse_y420cmdb_bitmap(struct drm_connector *connector, map |= (u64)db[2 + count] << (8 * count); if (map) - info->color_formats |= DRM_COLOR_FORMAT_YCRCB420; + info->color_formats |= DRM_COLOR_FORMAT_YCBCR420; hdmi->y420_cmdb_map = map; } @@ -5075,21 +5086,21 @@ static void drm_parse_hdmi_deep_color_info(struct drm_connector *connector, if (hdmi[6] & DRM_EDID_HDMI_DC_30) { dc_bpc = 10; - info->edid_hdmi_dc_modes |= DRM_EDID_HDMI_DC_30; + info->edid_hdmi_rgb444_dc_modes |= DRM_EDID_HDMI_DC_30; DRM_DEBUG("%s: HDMI sink does deep color 30.\n", connector->name); } if (hdmi[6] & DRM_EDID_HDMI_DC_36) { dc_bpc = 12; - info->edid_hdmi_dc_modes |= DRM_EDID_HDMI_DC_36; + info->edid_hdmi_rgb444_dc_modes |= DRM_EDID_HDMI_DC_36; DRM_DEBUG("%s: HDMI sink does deep color 36.\n", connector->name); } if (hdmi[6] & DRM_EDID_HDMI_DC_48) { dc_bpc = 16; - info->edid_hdmi_dc_modes |= DRM_EDID_HDMI_DC_48; + info->edid_hdmi_rgb444_dc_modes |= DRM_EDID_HDMI_DC_48; DRM_DEBUG("%s: HDMI sink does deep color 48.\n", connector->name); } @@ -5104,16 +5115,9 @@ static void drm_parse_hdmi_deep_color_info(struct drm_connector *connector, connector->name, dc_bpc); info->bpc = dc_bpc; - /* - * Deep color support mandates RGB444 support for all video - * modes and forbids YCRCB422 support for all video modes per - * HDMI 1.3 spec. - */ - info->color_formats = DRM_COLOR_FORMAT_RGB444; - /* YCRCB444 is optional according to spec. */ if (hdmi[6] & DRM_EDID_HDMI_DC_Y444) { - info->color_formats |= DRM_COLOR_FORMAT_YCRCB444; + info->edid_hdmi_ycbcr444_dc_modes = info->edid_hdmi_rgb444_dc_modes; DRM_DEBUG("%s: HDMI sink does YCRCB444 in deep color.\n", connector->name); } @@ -5149,6 +5153,25 @@ drm_parse_hdmi_vsdb_video(struct drm_connector *connector, const u8 *db) drm_parse_hdmi_deep_color_info(connector, db); } +/* + * See EDID extension for head-mounted and specialized monitors, specified at: + * https://docs.microsoft.com/en-us/windows-hardware/drivers/display/specialized-monitors-edid-extension + */ +static void drm_parse_microsoft_vsdb(struct drm_connector *connector, + const u8 *db) +{ + struct drm_display_info *info = &connector->display_info; + u8 version = db[4]; + bool desktop_usage = db[5] & BIT(6); + + /* Version 1 and 2 for HMDs, version 3 flags desktop usage explicitly */ + if (version == 1 || version == 2 || (version == 3 && !desktop_usage)) + info->non_desktop = true; + + drm_dbg_kms(connector->dev, "HMD or specialized display VSDB version %u: 0x%02x\n", + version, db[5]); +} + static void drm_parse_cea_ext(struct drm_connector *connector, const struct edid *edid) { @@ -5165,9 +5188,9 @@ static void drm_parse_cea_ext(struct drm_connector *connector, /* The existence of a CEA block should imply RGB support */ info->color_formats = DRM_COLOR_FORMAT_RGB444; if (edid_ext[3] & EDID_CEA_YCRCB444) - info->color_formats |= DRM_COLOR_FORMAT_YCRCB444; + info->color_formats |= DRM_COLOR_FORMAT_YCBCR444; if (edid_ext[3] & EDID_CEA_YCRCB422) - info->color_formats |= DRM_COLOR_FORMAT_YCRCB422; + info->color_formats |= DRM_COLOR_FORMAT_YCBCR422; if (cea_db_offsets(edid_ext, &start, &end)) return; @@ -5179,6 +5202,8 @@ static void drm_parse_cea_ext(struct drm_connector *connector, drm_parse_hdmi_vsdb_video(connector, db); if (cea_db_is_hdmi_forum_vsdb(db)) drm_parse_hdmi_forum_vsdb(connector, db); + if (cea_db_is_microsoft_vsdb(db)) + drm_parse_microsoft_vsdb(connector, db); if (cea_db_is_y420cmdb(db)) drm_parse_y420cmdb_bitmap(connector, db); if (cea_db_is_vcdb(db)) @@ -5333,17 +5358,13 @@ u32 drm_add_display_info(struct drm_connector *connector, const struct edid *edi info->width_mm = edid->width_cm * 10; info->height_mm = edid->height_cm * 10; - info->non_desktop = !!(quirks & EDID_QUIRK_NON_DESKTOP); - drm_get_monitor_range(connector, edid); - DRM_DEBUG_KMS("non_desktop set to %d\n", info->non_desktop); - if (edid->revision < 3) - return quirks; + goto out; if (!(edid->input & DRM_EDID_INPUT_DIGITAL)) - return quirks; + goto out; drm_parse_cea_ext(connector, edid); @@ -5363,7 +5384,7 @@ u32 drm_add_display_info(struct drm_connector *connector, const struct edid *edi /* Only defined for 1.4 with digital displays */ if (edid->revision < 4) - return quirks; + goto out; switch (edid->input & DRM_EDID_DIGITAL_DEPTH_MASK) { case DRM_EDID_DIGITAL_DEPTH_6: @@ -5395,17 +5416,25 @@ u32 drm_add_display_info(struct drm_connector *connector, const struct edid *edi info->color_formats |= DRM_COLOR_FORMAT_RGB444; if (edid->features & DRM_EDID_FEATURE_RGB_YCRCB444) - info->color_formats |= DRM_COLOR_FORMAT_YCRCB444; + info->color_formats |= DRM_COLOR_FORMAT_YCBCR444; if (edid->features & DRM_EDID_FEATURE_RGB_YCRCB422) - info->color_formats |= DRM_COLOR_FORMAT_YCRCB422; + info->color_formats |= DRM_COLOR_FORMAT_YCBCR422; drm_update_mso(connector, edid); +out: + if (quirks & EDID_QUIRK_NON_DESKTOP) { + drm_dbg_kms(connector->dev, "Non-desktop display%s\n", + info->non_desktop ? " (redundant quirk)" : ""); + info->non_desktop = true; + } + return quirks; } static struct drm_display_mode *drm_mode_displayid_detailed(struct drm_device *dev, - struct displayid_detailed_timings_1 *timings) + struct displayid_detailed_timings_1 *timings, + bool type_7) { struct drm_display_mode *mode; unsigned pixel_clock = (timings->pixel_clock[0] | @@ -5426,7 +5455,8 @@ static struct drm_display_mode *drm_mode_displayid_detailed(struct drm_device *d if (!mode) return NULL; - mode->clock = pixel_clock * 10; + /* resolution is kHz for type VII, and 10 kHz for type I */ + mode->clock = type_7 ? pixel_clock : pixel_clock * 10; mode->hdisplay = hactive; mode->hsync_start = mode->hdisplay + hsync; mode->hsync_end = mode->hsync_start + hsync_width; @@ -5457,6 +5487,7 @@ static int add_displayid_detailed_1_modes(struct drm_connector *connector, int num_timings; struct drm_display_mode *newmode; int num_modes = 0; + bool type_7 = block->tag == DATA_BLOCK_2_TYPE_7_DETAILED_TIMING; /* blocks must be multiple of 20 bytes length */ if (block->num_bytes % 20) return 0; @@ -5465,7 +5496,7 @@ static int add_displayid_detailed_1_modes(struct drm_connector *connector, for (i = 0; i < num_timings; i++) { struct displayid_detailed_timings_1 *timings = &det->timings[i]; - newmode = drm_mode_displayid_detailed(connector->dev, timings); + newmode = drm_mode_displayid_detailed(connector->dev, timings, type_7); if (!newmode) continue; @@ -5484,7 +5515,8 @@ static int add_displayid_detailed_modes(struct drm_connector *connector, displayid_iter_edid_begin(edid, &iter); displayid_iter_for_each(block, &iter) { - if (block->tag == DATA_BLOCK_TYPE_1_DETAILED_TIMING) + if (block->tag == DATA_BLOCK_TYPE_1_DETAILED_TIMING || + block->tag == DATA_BLOCK_2_TYPE_7_DETAILED_TIMING) num_modes += add_displayid_detailed_1_modes(connector, block); } displayid_iter_end(&iter); @@ -5652,7 +5684,7 @@ static bool is_hdmi2_sink(const struct drm_connector *connector) return true; return connector->display_info.hdmi.scdc.supported || - connector->display_info.color_formats & DRM_COLOR_FORMAT_YCRCB420; + connector->display_info.color_formats & DRM_COLOR_FORMAT_YCBCR420; } static inline bool is_eotf_supported(u8 output_eotf, u8 sink_eotf) @@ -5891,13 +5923,13 @@ static const u32 hdmi_colorimetry_val[] = { #undef ACE /** - * drm_hdmi_avi_infoframe_colorspace() - fill the HDMI AVI infoframe - * colorspace information + * drm_hdmi_avi_infoframe_colorimetry() - fill the HDMI AVI infoframe + * colorimetry information * @frame: HDMI AVI infoframe * @conn_state: connector state */ void -drm_hdmi_avi_infoframe_colorspace(struct hdmi_avi_infoframe *frame, +drm_hdmi_avi_infoframe_colorimetry(struct hdmi_avi_infoframe *frame, const struct drm_connector_state *conn_state) { u32 colorimetry_val; @@ -5916,7 +5948,7 @@ drm_hdmi_avi_infoframe_colorspace(struct hdmi_avi_infoframe *frame, frame->extended_colorimetry = (colorimetry_val >> 2) & EXTENDED_COLORIMETRY_MASK; } -EXPORT_SYMBOL(drm_hdmi_avi_infoframe_colorspace); +EXPORT_SYMBOL(drm_hdmi_avi_infoframe_colorimetry); /** * drm_hdmi_avi_infoframe_quant_range() - fill the HDMI AVI infoframe diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index ed43b987d306..e9a9d35fbf5e 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -373,7 +373,7 @@ static void drm_fb_helper_resume_worker(struct work_struct *work) static void drm_fb_helper_damage_blit_real(struct drm_fb_helper *fb_helper, struct drm_clip_rect *clip, - struct dma_buf_map *dst) + struct iosys_map *dst) { struct drm_framebuffer *fb = fb_helper->fb; unsigned int cpp = fb->format->cpp[0]; @@ -382,11 +382,11 @@ static void drm_fb_helper_damage_blit_real(struct drm_fb_helper *fb_helper, size_t len = (clip->x2 - clip->x1) * cpp; unsigned int y; - dma_buf_map_incr(dst, offset); /* go to first pixel within clip rect */ + iosys_map_incr(dst, offset); /* go to first pixel within clip rect */ for (y = clip->y1; y < clip->y2; y++) { - dma_buf_map_memcpy_to(dst, src, len); - dma_buf_map_incr(dst, fb->pitches[0]); + iosys_map_memcpy_to(dst, src, len); + iosys_map_incr(dst, fb->pitches[0]); src += fb->pitches[0]; } } @@ -395,7 +395,7 @@ static int drm_fb_helper_damage_blit(struct drm_fb_helper *fb_helper, struct drm_clip_rect *clip) { struct drm_client_buffer *buffer = fb_helper->buffer; - struct dma_buf_map map, dst; + struct iosys_map map, dst; int ret; /* @@ -2322,7 +2322,7 @@ static int drm_fb_helper_generic_probe(struct drm_fb_helper *fb_helper, struct drm_framebuffer *fb; struct fb_info *fbi; u32 format; - struct dma_buf_map map; + struct iosys_map map; int ret; drm_dbg_kms(dev, "surface width(%d), height(%d) and bpp(%d)\n", diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index 4dcdec6487bb..8c7b24f4b0e4 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -36,7 +36,7 @@ #include <linux/pagemap.h> #include <linux/shmem_fs.h> #include <linux/dma-buf.h> -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/mem_encrypt.h> #include <linux/pagevec.h> @@ -1165,7 +1165,7 @@ void drm_gem_unpin(struct drm_gem_object *obj) obj->funcs->unpin(obj); } -int drm_gem_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) +int drm_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map) { int ret; @@ -1175,23 +1175,23 @@ int drm_gem_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) ret = obj->funcs->vmap(obj, map); if (ret) return ret; - else if (dma_buf_map_is_null(map)) + else if (iosys_map_is_null(map)) return -ENOMEM; return 0; } EXPORT_SYMBOL(drm_gem_vmap); -void drm_gem_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map) +void drm_gem_vunmap(struct drm_gem_object *obj, struct iosys_map *map) { - if (dma_buf_map_is_null(map)) + if (iosys_map_is_null(map)) return; if (obj->funcs->vunmap) obj->funcs->vunmap(obj, map); /* Always set the mapping to NULL. Callers may rely on this. */ - dma_buf_map_clear(map); + iosys_map_clear(map); } EXPORT_SYMBOL(drm_gem_vunmap); diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c index dc275c466c9c..f36734c2c9e1 100644 --- a/drivers/gpu/drm/drm_gem_cma_helper.c +++ b/drivers/gpu/drm/drm_gem_cma_helper.c @@ -209,7 +209,7 @@ drm_gem_cma_create_with_handle(struct drm_file *file_priv, void drm_gem_cma_free(struct drm_gem_cma_object *cma_obj) { struct drm_gem_object *gem_obj = &cma_obj->base; - struct dma_buf_map map = DMA_BUF_MAP_INIT_VADDR(cma_obj->vaddr); + struct iosys_map map = IOSYS_MAP_INIT_VADDR(cma_obj->vaddr); if (gem_obj->import_attach) { if (cma_obj->vaddr) @@ -480,9 +480,10 @@ EXPORT_SYMBOL_GPL(drm_gem_cma_prime_import_sg_table); * Returns: * 0 on success, or a negative error code otherwise. */ -int drm_gem_cma_vmap(struct drm_gem_cma_object *cma_obj, struct dma_buf_map *map) +int drm_gem_cma_vmap(struct drm_gem_cma_object *cma_obj, + struct iosys_map *map) { - dma_buf_map_set_vaddr(map, cma_obj->vaddr); + iosys_map_set_vaddr(map, cma_obj->vaddr); return 0; } @@ -558,7 +559,7 @@ drm_gem_cma_prime_import_sg_table_vmap(struct drm_device *dev, { struct drm_gem_cma_object *cma_obj; struct drm_gem_object *obj; - struct dma_buf_map map; + struct iosys_map map; int ret; ret = dma_buf_vmap(attach->dmabuf, &map); diff --git a/drivers/gpu/drm/drm_gem_framebuffer_helper.c b/drivers/gpu/drm/drm_gem_framebuffer_helper.c index 746fd8c73845..f4619803acd0 100644 --- a/drivers/gpu/drm/drm_gem_framebuffer_helper.c +++ b/drivers/gpu/drm/drm_gem_framebuffer_helper.c @@ -321,7 +321,7 @@ EXPORT_SYMBOL_GPL(drm_gem_fb_create_with_dirty); * @data: returns the data address for each BO, can be NULL * * This function maps all buffer objects of the given framebuffer into - * kernel address space and stores them in struct dma_buf_map. If the + * kernel address space and stores them in struct iosys_map. If the * mapping operation fails for one of the BOs, the function unmaps the * already established mappings automatically. * @@ -335,8 +335,8 @@ EXPORT_SYMBOL_GPL(drm_gem_fb_create_with_dirty); * 0 on success, or a negative errno code otherwise. */ int drm_gem_fb_vmap(struct drm_framebuffer *fb, - struct dma_buf_map map[static DRM_FORMAT_MAX_PLANES], - struct dma_buf_map data[DRM_FORMAT_MAX_PLANES]) + struct iosys_map map[static DRM_FORMAT_MAX_PLANES], + struct iosys_map data[DRM_FORMAT_MAX_PLANES]) { struct drm_gem_object *obj; unsigned int i; @@ -345,7 +345,7 @@ int drm_gem_fb_vmap(struct drm_framebuffer *fb, for (i = 0; i < DRM_FORMAT_MAX_PLANES; ++i) { obj = drm_gem_fb_get_obj(fb, i); if (!obj) { - dma_buf_map_clear(&map[i]); + iosys_map_clear(&map[i]); continue; } ret = drm_gem_vmap(obj, &map[i]); @@ -356,9 +356,9 @@ int drm_gem_fb_vmap(struct drm_framebuffer *fb, if (data) { for (i = 0; i < DRM_FORMAT_MAX_PLANES; ++i) { memcpy(&data[i], &map[i], sizeof(data[i])); - if (dma_buf_map_is_null(&data[i])) + if (iosys_map_is_null(&data[i])) continue; - dma_buf_map_incr(&data[i], fb->offsets[i]); + iosys_map_incr(&data[i], fb->offsets[i]); } } @@ -386,7 +386,7 @@ EXPORT_SYMBOL(drm_gem_fb_vmap); * See drm_gem_fb_vmap() for more information. */ void drm_gem_fb_vunmap(struct drm_framebuffer *fb, - struct dma_buf_map map[static DRM_FORMAT_MAX_PLANES]) + struct iosys_map map[static DRM_FORMAT_MAX_PLANES]) { unsigned int i = DRM_FORMAT_MAX_PLANES; struct drm_gem_object *obj; @@ -396,7 +396,7 @@ void drm_gem_fb_vunmap(struct drm_framebuffer *fb, obj = drm_gem_fb_get_obj(fb, i); if (!obj) continue; - if (dma_buf_map_is_null(&map[i])) + if (iosys_map_is_null(&map[i])) continue; drm_gem_vunmap(obj, &map[i]); } diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c index 621924116eb4..3e738aea2664 100644 --- a/drivers/gpu/drm/drm_gem_shmem_helper.c +++ b/drivers/gpu/drm/drm_gem_shmem_helper.c @@ -286,13 +286,14 @@ void drm_gem_shmem_unpin(struct drm_gem_shmem_object *shmem) } EXPORT_SYMBOL(drm_gem_shmem_unpin); -static int drm_gem_shmem_vmap_locked(struct drm_gem_shmem_object *shmem, struct dma_buf_map *map) +static int drm_gem_shmem_vmap_locked(struct drm_gem_shmem_object *shmem, + struct iosys_map *map) { struct drm_gem_object *obj = &shmem->base; int ret = 0; if (shmem->vmap_use_count++ > 0) { - dma_buf_map_set_vaddr(map, shmem->vaddr); + iosys_map_set_vaddr(map, shmem->vaddr); return 0; } @@ -319,7 +320,7 @@ static int drm_gem_shmem_vmap_locked(struct drm_gem_shmem_object *shmem, struct if (!shmem->vaddr) ret = -ENOMEM; else - dma_buf_map_set_vaddr(map, shmem->vaddr); + iosys_map_set_vaddr(map, shmem->vaddr); } if (ret) { @@ -353,7 +354,8 @@ err_zero_use: * Returns: * 0 on success or a negative error code on failure. */ -int drm_gem_shmem_vmap(struct drm_gem_shmem_object *shmem, struct dma_buf_map *map) +int drm_gem_shmem_vmap(struct drm_gem_shmem_object *shmem, + struct iosys_map *map) { int ret; @@ -368,7 +370,7 @@ int drm_gem_shmem_vmap(struct drm_gem_shmem_object *shmem, struct dma_buf_map *m EXPORT_SYMBOL(drm_gem_shmem_vmap); static void drm_gem_shmem_vunmap_locked(struct drm_gem_shmem_object *shmem, - struct dma_buf_map *map) + struct iosys_map *map) { struct drm_gem_object *obj = &shmem->base; @@ -400,7 +402,8 @@ static void drm_gem_shmem_vunmap_locked(struct drm_gem_shmem_object *shmem, * This function hides the differences between dma-buf imported and natively * allocated objects. */ -void drm_gem_shmem_vunmap(struct drm_gem_shmem_object *shmem, struct dma_buf_map *map) +void drm_gem_shmem_vunmap(struct drm_gem_shmem_object *shmem, + struct iosys_map *map) { mutex_lock(&shmem->vmap_lock); drm_gem_shmem_vunmap_locked(shmem, map); diff --git a/drivers/gpu/drm/drm_gem_ttm_helper.c b/drivers/gpu/drm/drm_gem_ttm_helper.c index ecf3d2a54a98..d5962a34c01d 100644 --- a/drivers/gpu/drm/drm_gem_ttm_helper.c +++ b/drivers/gpu/drm/drm_gem_ttm_helper.c @@ -61,7 +61,7 @@ EXPORT_SYMBOL(drm_gem_ttm_print_info); * 0 on success, or a negative errno code otherwise. */ int drm_gem_ttm_vmap(struct drm_gem_object *gem, - struct dma_buf_map *map) + struct iosys_map *map) { struct ttm_buffer_object *bo = drm_gem_ttm_of_gem(gem); @@ -78,7 +78,7 @@ EXPORT_SYMBOL(drm_gem_ttm_vmap); * &drm_gem_object_funcs.vmap callback. */ void drm_gem_ttm_vunmap(struct drm_gem_object *gem, - struct dma_buf_map *map) + struct iosys_map *map) { struct ttm_buffer_object *bo = drm_gem_ttm_of_gem(gem); diff --git a/drivers/gpu/drm/drm_gem_vram_helper.c b/drivers/gpu/drm/drm_gem_vram_helper.c index 3f00192215d1..dc7f938bfff2 100644 --- a/drivers/gpu/drm/drm_gem_vram_helper.c +++ b/drivers/gpu/drm/drm_gem_vram_helper.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-or-later -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/module.h> #include <drm/drm_debugfs.h> @@ -116,7 +116,7 @@ static void drm_gem_vram_cleanup(struct drm_gem_vram_object *gbo) */ WARN_ON(gbo->vmap_use_count); - WARN_ON(dma_buf_map_is_set(&gbo->map)); + WARN_ON(iosys_map_is_set(&gbo->map)); drm_gem_object_release(&gbo->bo.base); } @@ -365,7 +365,7 @@ int drm_gem_vram_unpin(struct drm_gem_vram_object *gbo) EXPORT_SYMBOL(drm_gem_vram_unpin); static int drm_gem_vram_kmap_locked(struct drm_gem_vram_object *gbo, - struct dma_buf_map *map) + struct iosys_map *map) { int ret; @@ -377,7 +377,7 @@ static int drm_gem_vram_kmap_locked(struct drm_gem_vram_object *gbo, * page mapping might still be around. Only vmap if the there's * no mapping present. */ - if (dma_buf_map_is_null(&gbo->map)) { + if (iosys_map_is_null(&gbo->map)) { ret = ttm_bo_vmap(&gbo->bo, &gbo->map); if (ret) return ret; @@ -391,14 +391,14 @@ out: } static void drm_gem_vram_kunmap_locked(struct drm_gem_vram_object *gbo, - struct dma_buf_map *map) + struct iosys_map *map) { struct drm_device *dev = gbo->bo.base.dev; if (drm_WARN_ON_ONCE(dev, !gbo->vmap_use_count)) return; - if (drm_WARN_ON_ONCE(dev, !dma_buf_map_is_equal(&gbo->map, map))) + if (drm_WARN_ON_ONCE(dev, !iosys_map_is_equal(&gbo->map, map))) return; /* BUG: map not mapped from this BO */ if (--gbo->vmap_use_count > 0) @@ -428,7 +428,7 @@ static void drm_gem_vram_kunmap_locked(struct drm_gem_vram_object *gbo, * Returns: * 0 on success, or a negative error code otherwise. */ -int drm_gem_vram_vmap(struct drm_gem_vram_object *gbo, struct dma_buf_map *map) +int drm_gem_vram_vmap(struct drm_gem_vram_object *gbo, struct iosys_map *map) { int ret; @@ -463,7 +463,8 @@ EXPORT_SYMBOL(drm_gem_vram_vmap); * A call to drm_gem_vram_vunmap() unmaps and unpins a GEM VRAM buffer. See * the documentation for drm_gem_vram_vmap() for more information. */ -void drm_gem_vram_vunmap(struct drm_gem_vram_object *gbo, struct dma_buf_map *map) +void drm_gem_vram_vunmap(struct drm_gem_vram_object *gbo, + struct iosys_map *map) { int ret; @@ -567,7 +568,7 @@ static void drm_gem_vram_bo_driver_move_notify(struct drm_gem_vram_object *gbo) return; ttm_bo_vunmap(bo, &gbo->map); - dma_buf_map_clear(&gbo->map); /* explicitly clear mapping for next vmap call */ + iosys_map_clear(&gbo->map); /* explicitly clear mapping for next vmap call */ } static int drm_gem_vram_bo_driver_move(struct drm_gem_vram_object *gbo, @@ -802,7 +803,8 @@ static void drm_gem_vram_object_unpin(struct drm_gem_object *gem) * Returns: * 0 on success, or a negative error code otherwise. */ -static int drm_gem_vram_object_vmap(struct drm_gem_object *gem, struct dma_buf_map *map) +static int drm_gem_vram_object_vmap(struct drm_gem_object *gem, + struct iosys_map *map) { struct drm_gem_vram_object *gbo = drm_gem_vram_of_gem(gem); @@ -815,7 +817,8 @@ static int drm_gem_vram_object_vmap(struct drm_gem_object *gem, struct dma_buf_m * @gem: The GEM object to unmap * @map: Kernel virtual address where the VRAM GEM object was mapped */ -static void drm_gem_vram_object_vunmap(struct drm_gem_object *gem, struct dma_buf_map *map) +static void drm_gem_vram_object_vunmap(struct drm_gem_object *gem, + struct iosys_map *map) { struct drm_gem_vram_object *gbo = drm_gem_vram_of_gem(gem); diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h index 17f3548c8ed2..1fbbc19f1ac0 100644 --- a/drivers/gpu/drm/drm_internal.h +++ b/drivers/gpu/drm/drm_internal.h @@ -33,7 +33,7 @@ struct dentry; struct dma_buf; -struct dma_buf_map; +struct iosys_map; struct drm_connector; struct drm_crtc; struct drm_framebuffer; @@ -174,8 +174,8 @@ void drm_gem_print_info(struct drm_printer *p, unsigned int indent, int drm_gem_pin(struct drm_gem_object *obj); void drm_gem_unpin(struct drm_gem_object *obj); -int drm_gem_vmap(struct drm_gem_object *obj, struct dma_buf_map *map); -void drm_gem_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map); +int drm_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map); +void drm_gem_vunmap(struct drm_gem_object *obj, struct iosys_map *map); int drm_gem_dumb_destroy(struct drm_file *file, struct drm_device *dev, u32 handle); diff --git a/drivers/gpu/drm/drm_kms_helper_common.c b/drivers/gpu/drm/drm_kms_helper_common.c index 47e92400548d..8be20080cd8d 100644 --- a/drivers/gpu/drm/drm_kms_helper_common.c +++ b/drivers/gpu/drm/drm_kms_helper_common.c @@ -61,17 +61,3 @@ MODULE_PARM_DESC(edid_firmware, "DEPRECATED. Use drm.edid_firmware module parameter instead."); #endif - -static int __init drm_kms_helper_init(void) -{ - return drm_dp_aux_dev_init(); -} - -static void __exit drm_kms_helper_exit(void) -{ - /* Call exit functions from specific kms helpers here */ - drm_dp_aux_dev_exit(); -} - -module_init(drm_kms_helper_init); -module_exit(drm_kms_helper_exit); diff --git a/drivers/gpu/drm/drm_mipi_dbi.c b/drivers/gpu/drm/drm_mipi_dbi.c index 0327d595e028..9314f2ead79f 100644 --- a/drivers/gpu/drm/drm_mipi_dbi.c +++ b/drivers/gpu/drm/drm_mipi_dbi.c @@ -201,8 +201,8 @@ int mipi_dbi_buf_copy(void *dst, struct drm_framebuffer *fb, struct drm_rect *clip, bool swap) { struct drm_gem_object *gem = drm_gem_fb_get_obj(fb, 0); - struct dma_buf_map map[DRM_FORMAT_MAX_PLANES]; - struct dma_buf_map data[DRM_FORMAT_MAX_PLANES]; + struct iosys_map map[DRM_FORMAT_MAX_PLANES]; + struct iosys_map data[DRM_FORMAT_MAX_PLANES]; void *src; int ret; @@ -258,8 +258,8 @@ static void mipi_dbi_set_window_address(struct mipi_dbi_dev *dbidev, static void mipi_dbi_fb_dirty(struct drm_framebuffer *fb, struct drm_rect *rect) { - struct dma_buf_map map[DRM_FORMAT_MAX_PLANES]; - struct dma_buf_map data[DRM_FORMAT_MAX_PLANES]; + struct iosys_map map[DRM_FORMAT_MAX_PLANES]; + struct iosys_map data[DRM_FORMAT_MAX_PLANES]; struct mipi_dbi_dev *dbidev = drm_to_mipi_dbi_dev(fb->dev); unsigned int height = rect->y2 - rect->y1; unsigned int width = rect->x2 - rect->x1; diff --git a/drivers/gpu/drm/drm_panel_orientation_quirks.c b/drivers/gpu/drm/drm_panel_orientation_quirks.c index b910978d3e48..4e853acfd1e8 100644 --- a/drivers/gpu/drm/drm_panel_orientation_quirks.c +++ b/drivers/gpu/drm/drm_panel_orientation_quirks.c @@ -180,6 +180,12 @@ static const struct dmi_system_id orientation_data[] = { DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "MicroPC"), }, .driver_data = (void *)&lcd720x1280_rightside_up, + }, { /* GPD Win Max */ + .matches = { + DMI_EXACT_MATCH(DMI_SYS_VENDOR, "GPD"), + DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "G1619-01"), + }, + .driver_data = (void *)&lcd800x1280_rightside_up, }, { /* * GPD Pocket, note that the the DMI data is less generic then * it seems, devices with a board-vendor of "AMI Corporation" diff --git a/drivers/gpu/drm/drm_plane.c b/drivers/gpu/drm/drm_plane.c index 82afb854141b..deeec60a3315 100644 --- a/drivers/gpu/drm/drm_plane.c +++ b/drivers/gpu/drm/drm_plane.c @@ -202,17 +202,13 @@ static int create_in_format_blob(struct drm_device *dev, struct drm_plane *plane memcpy(formats_ptr(blob_data), plane->format_types, formats_size); - /* If we can't determine support, just bail */ - if (!plane->funcs->format_mod_supported) - goto done; - mod = modifiers_ptr(blob_data); for (i = 0; i < plane->modifier_count; i++) { for (j = 0; j < plane->format_count; j++) { - if (plane->funcs->format_mod_supported(plane, + if (!plane->funcs->format_mod_supported || + plane->funcs->format_mod_supported(plane, plane->format_types[j], plane->modifiers[i])) { - mod->formats |= 1ULL << j; } } @@ -223,7 +219,6 @@ static int create_in_format_blob(struct drm_device *dev, struct drm_plane *plane mod++; } -done: drm_object_attach_property(&plane->base, config->modifiers_property, blob->base.id); diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c index c773d3dfb1ab..e3f09f18110c 100644 --- a/drivers/gpu/drm/drm_prime.c +++ b/drivers/gpu/drm/drm_prime.c @@ -674,7 +674,7 @@ EXPORT_SYMBOL(drm_gem_unmap_dma_buf); * * Returns 0 on success or a negative errno code otherwise. */ -int drm_gem_dmabuf_vmap(struct dma_buf *dma_buf, struct dma_buf_map *map) +int drm_gem_dmabuf_vmap(struct dma_buf *dma_buf, struct iosys_map *map) { struct drm_gem_object *obj = dma_buf->priv; @@ -690,7 +690,7 @@ EXPORT_SYMBOL(drm_gem_dmabuf_vmap); * Releases a kernel virtual mapping. This can be used as the * &dma_buf_ops.vunmap callback. Calls into &drm_gem_object_funcs.vunmap for device specific handling. */ -void drm_gem_dmabuf_vunmap(struct dma_buf *dma_buf, struct dma_buf_map *map) +void drm_gem_dmabuf_vunmap(struct dma_buf *dma_buf, struct iosys_map *map) { struct drm_gem_object *obj = dma_buf->priv; diff --git a/drivers/gpu/drm/drm_privacy_screen.c b/drivers/gpu/drm/drm_privacy_screen.c index b688841c18e4..2b082e5e9732 100644 --- a/drivers/gpu/drm/drm_privacy_screen.c +++ b/drivers/gpu/drm/drm_privacy_screen.c @@ -387,7 +387,8 @@ static void drm_privacy_screen_device_release(struct device *dev) * * An ERR_PTR(errno) on failure. */ struct drm_privacy_screen *drm_privacy_screen_register( - struct device *parent, const struct drm_privacy_screen_ops *ops) + struct device *parent, const struct drm_privacy_screen_ops *ops, + void *data) { struct drm_privacy_screen *priv; int ret; @@ -404,6 +405,7 @@ struct drm_privacy_screen *drm_privacy_screen_register( priv->dev.parent = parent; priv->dev.release = drm_privacy_screen_device_release; dev_set_name(&priv->dev, "privacy_screen-%s", dev_name(parent)); + priv->drvdata = data; priv->ops = ops; priv->ops->get_hw_state(priv); @@ -439,6 +441,7 @@ void drm_privacy_screen_unregister(struct drm_privacy_screen *priv) mutex_unlock(&drm_privacy_screen_devs_lock); mutex_lock(&priv->lock); + priv->drvdata = NULL; priv->ops = NULL; mutex_unlock(&priv->lock); diff --git a/drivers/gpu/drm/drm_privacy_screen_x86.c b/drivers/gpu/drm/drm_privacy_screen_x86.c index e7aa74ad0b24..72ed40e4997e 100644 --- a/drivers/gpu/drm/drm_privacy_screen_x86.c +++ b/drivers/gpu/drm/drm_privacy_screen_x86.c @@ -50,6 +50,13 @@ static bool __init detect_thinkpad_privacy_screen(void) } #endif +#if IS_ENABLED(CONFIG_CHROMEOS_PRIVACY_SCREEN) +static bool __init detect_chromeos_privacy_screen(void) +{ + return acpi_dev_present("GOOG0010", NULL, -1); +} +#endif + static const struct arch_init_data arch_init_data[] __initconst = { #if IS_ENABLED(CONFIG_THINKPAD_ACPI) { @@ -61,6 +68,16 @@ static const struct arch_init_data arch_init_data[] __initconst = { .detect = detect_thinkpad_privacy_screen, }, #endif +#if IS_ENABLED(CONFIG_CHROMEOS_PRIVACY_SCREEN) + { + .lookup = { + .dev_id = NULL, + .con_id = NULL, + .provider = "privacy_screen-GOOG0010:00", + }, + .detect = detect_chromeos_privacy_screen, + }, +#endif }; void __init drm_privacy_screen_lookup_init(void) diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h index 049ae87de9be..f32f4771dada 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h @@ -49,7 +49,7 @@ int etnaviv_ioctl_gem_submit(struct drm_device *dev, void *data, int etnaviv_gem_mmap_offset(struct drm_gem_object *obj, u64 *offset); struct sg_table *etnaviv_gem_prime_get_sg_table(struct drm_gem_object *obj); -int etnaviv_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map); +int etnaviv_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map); struct drm_gem_object *etnaviv_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sg); int etnaviv_gem_prime_pin(struct drm_gem_object *obj); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c index 6788ea8490d1..3fa2da149639 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c @@ -25,14 +25,14 @@ struct sg_table *etnaviv_gem_prime_get_sg_table(struct drm_gem_object *obj) return drm_prime_pages_to_sg(obj->dev, etnaviv_obj->pages, npages); } -int etnaviv_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) +int etnaviv_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map) { void *vaddr; vaddr = etnaviv_gem_vmap(obj); if (!vaddr) return -ENOMEM; - dma_buf_map_set_vaddr(map, vaddr); + iosys_map_set_vaddr(map, vaddr); return 0; } @@ -62,7 +62,7 @@ void etnaviv_gem_prime_unpin(struct drm_gem_object *obj) static void etnaviv_gem_prime_release(struct etnaviv_gem_object *etnaviv_obj) { - struct dma_buf_map map = DMA_BUF_MAP_INIT_VADDR(etnaviv_obj->vaddr); + struct iosys_map map = IOSYS_MAP_INIT_VADDR(etnaviv_obj->vaddr); if (etnaviv_obj->vaddr) dma_buf_vunmap(etnaviv_obj->base.import_attach->dmabuf, &map); @@ -77,7 +77,7 @@ static void etnaviv_gem_prime_release(struct etnaviv_gem_object *etnaviv_obj) static void *etnaviv_gem_prime_vmap_impl(struct etnaviv_gem_object *etnaviv_obj) { - struct dma_buf_map map; + struct iosys_map map; int ret; lockdep_assert_held(&etnaviv_obj->lock); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c index a17313282e8b..4eb00a0cb650 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c @@ -189,8 +189,7 @@ static int submit_fence_sync(struct etnaviv_gem_submit *submit) continue; if (bo->flags & ETNA_SUBMIT_BO_WRITE) { - ret = dma_resv_get_fences(robj, NULL, - &bo->nr_shared, + ret = dma_resv_get_fences(robj, true, &bo->nr_shared, &bo->shared); if (ret) return ret; diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig index 6a251e3aa779..f27cfd2a9726 100644 --- a/drivers/gpu/drm/exynos/Kconfig +++ b/drivers/gpu/drm/exynos/Kconfig @@ -66,6 +66,7 @@ config DRM_EXYNOS_DP bool "Exynos specific extensions for Analogix DP driver" depends on DRM_EXYNOS_FIMD || DRM_EXYNOS7_DECON select DRM_ANALOGIX_DP + select DRM_DP_HELPER default DRM_EXYNOS select DRM_PANEL help diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c index 32a36572b894..b7d0a4aead0a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c @@ -258,6 +258,7 @@ struct exynos_dsi { struct list_head bridge_chain; struct drm_bridge *out_bridge; struct device *dev; + struct drm_display_mode mode; void __iomem *reg_base; struct phy *phy; @@ -881,7 +882,7 @@ static int exynos_dsi_init_link(struct exynos_dsi *dsi) static void exynos_dsi_set_display_mode(struct exynos_dsi *dsi) { - struct drm_display_mode *m = &dsi->encoder.crtc->state->adjusted_mode; + struct drm_display_mode *m = &dsi->mode; unsigned int num_bits_resol = dsi->driver_data->num_bits_resol; u32 reg; @@ -1446,6 +1447,15 @@ static void exynos_dsi_disable(struct drm_encoder *encoder) pm_runtime_put_sync(dsi->dev); } +static void exynos_dsi_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct exynos_dsi *dsi = encoder_to_dsi(encoder); + + drm_mode_copy(&dsi->mode, adjusted_mode); +} + static enum drm_connector_status exynos_dsi_detect(struct drm_connector *connector, bool force) { @@ -1513,6 +1523,7 @@ static int exynos_dsi_create_connector(struct drm_encoder *encoder) static const struct drm_encoder_helper_funcs exynos_dsi_encoder_helper_funcs = { .enable = exynos_dsi_enable, .disable = exynos_dsi_disable, + .mode_set = exynos_dsi_mode_set, }; MODULE_DEVICE_TABLE(of, exynos_dsi_of_match); diff --git a/drivers/gpu/drm/gma500/cdv_intel_dp.c b/drivers/gpu/drm/gma500/cdv_intel_dp.c index ba6ad1466374..f562e91337c7 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_dp.c +++ b/drivers/gpu/drm/gma500/cdv_intel_dp.c @@ -31,7 +31,7 @@ #include <drm/drm_crtc.h> #include <drm/drm_crtc_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_simple_kms_helper.h> #include "gma_display.h" @@ -82,7 +82,6 @@ i2c_algo_dp_aux_address(struct i2c_adapter *adapter, u16 address, bool reading) { struct i2c_algo_dp_aux_data *algo_data = adapter->algo_data; int mode = MODE_I2C_START; - int ret; if (reading) mode |= MODE_I2C_READ; @@ -90,8 +89,7 @@ i2c_algo_dp_aux_address(struct i2c_adapter *adapter, u16 address, bool reading) mode |= MODE_I2C_WRITE; algo_data->address = address; algo_data->running = true; - ret = i2c_algo_dp_aux_transaction(adapter, mode, 0, NULL); - return ret; + return i2c_algo_dp_aux_transaction(adapter, mode, 0, NULL); } /* @@ -122,13 +120,11 @@ static int i2c_algo_dp_aux_put_byte(struct i2c_adapter *adapter, u8 byte) { struct i2c_algo_dp_aux_data *algo_data = adapter->algo_data; - int ret; if (!algo_data->running) return -EIO; - ret = i2c_algo_dp_aux_transaction(adapter, MODE_I2C_WRITE, byte, NULL); - return ret; + return i2c_algo_dp_aux_transaction(adapter, MODE_I2C_WRITE, byte, NULL); } /* @@ -139,13 +135,11 @@ static int i2c_algo_dp_aux_get_byte(struct i2c_adapter *adapter, u8 *byte_ret) { struct i2c_algo_dp_aux_data *algo_data = adapter->algo_data; - int ret; if (!algo_data->running) return -EIO; - ret = i2c_algo_dp_aux_transaction(adapter, MODE_I2C_READ, 0, byte_ret); - return ret; + return i2c_algo_dp_aux_transaction(adapter, MODE_I2C_READ, 0, byte_ret); } static int diff --git a/drivers/gpu/drm/gma500/gma_display.c b/drivers/gpu/drm/gma500/gma_display.c index 99da3118131a..60ba7de59139 100644 --- a/drivers/gpu/drm/gma500/gma_display.c +++ b/drivers/gpu/drm/gma500/gma_display.c @@ -335,7 +335,7 @@ int gma_crtc_cursor_set(struct drm_crtc *crtc, struct psb_gem_object *pobj; struct psb_gem_object *cursor_pobj = gma_crtc->cursor_pobj; struct drm_gem_object *obj; - void *tmp_dst, *tmp_src; + void *tmp_dst; int ret = 0, i, cursor_pages; /* If we didn't get a handle then turn the cursor off */ @@ -400,9 +400,7 @@ int gma_crtc_cursor_set(struct drm_crtc *crtc, /* Copy the cursor to cursor mem */ tmp_dst = dev_priv->vram_addr + cursor_pobj->offset; for (i = 0; i < cursor_pages; i++) { - tmp_src = kmap(pobj->pages[i]); - memcpy(tmp_dst, tmp_src, PAGE_SIZE); - kunmap(pobj->pages[i]); + memcpy_from_page(tmp_dst, pobj->pages[i], 0, PAGE_SIZE); tmp_dst += PAGE_SIZE; } diff --git a/drivers/gpu/drm/gma500/intel_bios.c b/drivers/gpu/drm/gma500/intel_bios.c index d5ca5f241974..ea7c16f33a0e 100644 --- a/drivers/gpu/drm/gma500/intel_bios.c +++ b/drivers/gpu/drm/gma500/intel_bios.c @@ -6,7 +6,7 @@ * Eric Anholt <eric@anholt.net> */ #include <drm/drm.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include "intel_bios.h" #include "psb_drv.h" diff --git a/drivers/gpu/drm/gma500/mmu.c b/drivers/gpu/drm/gma500/mmu.c index fe9ace2a7967..a70b01ccdf70 100644 --- a/drivers/gpu/drm/gma500/mmu.c +++ b/drivers/gpu/drm/gma500/mmu.c @@ -184,17 +184,17 @@ struct psb_mmu_pd *psb_mmu_alloc_pd(struct psb_mmu_driver *driver, pd->invalid_pte = 0; } - v = kmap(pd->dummy_pt); + v = kmap_local_page(pd->dummy_pt); for (i = 0; i < (PAGE_SIZE / sizeof(uint32_t)); ++i) v[i] = pd->invalid_pte; - kunmap(pd->dummy_pt); + kunmap_local(v); - v = kmap(pd->p); + v = kmap_local_page(pd->p); for (i = 0; i < (PAGE_SIZE / sizeof(uint32_t)); ++i) v[i] = pd->invalid_pde; - kunmap(pd->p); + kunmap_local(v); clear_page(kmap(pd->dummy_page)); kunmap(pd->dummy_page); diff --git a/drivers/gpu/drm/gud/gud_pipe.c b/drivers/gpu/drm/gud/gud_pipe.c index a150a5a4b5d4..4873f9799f41 100644 --- a/drivers/gpu/drm/gud/gud_pipe.c +++ b/drivers/gpu/drm/gud/gud_pipe.c @@ -152,8 +152,8 @@ static int gud_prep_flush(struct gud_device *gdrm, struct drm_framebuffer *fb, { struct dma_buf_attachment *import_attach = fb->obj[0]->import_attach; u8 compression = gdrm->compression; - struct dma_buf_map map[DRM_FORMAT_MAX_PLANES]; - struct dma_buf_map map_data[DRM_FORMAT_MAX_PLANES]; + struct iosys_map map[DRM_FORMAT_MAX_PLANES]; + struct iosys_map map_data[DRM_FORMAT_MAX_PLANES]; void *vaddr, *buf; size_t pitch, len; int ret = 0; diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c index 610fc8e135f9..fe4269c5aa0a 100644 --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c @@ -20,6 +20,7 @@ #include <drm/drm_gem_framebuffer_helper.h> #include <drm/drm_gem_vram_helper.h> #include <drm/drm_managed.h> +#include <drm/drm_module.h> #include <drm/drm_vblank.h> #include "hibmc_drm_drv.h" @@ -379,7 +380,7 @@ static struct pci_driver hibmc_pci_driver = { .driver.pm = &hibmc_pm_ops, }; -module_pci_driver(hibmc_pci_driver); +drm_module_pci_driver(hibmc_pci_driver); MODULE_DEVICE_TABLE(pci, hibmc_pci_table); MODULE_AUTHOR("RongrongZou <zourongrong@huawei.com>"); diff --git a/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c b/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c index 93f51e70a951..e82b815f83a6 100644 --- a/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c +++ b/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c @@ -19,7 +19,7 @@ #include "hyperv_drm.h" static int hyperv_blit_to_vram_rect(struct drm_framebuffer *fb, - const struct dma_buf_map *map, + const struct iosys_map *map, struct drm_rect *rect) { struct hyperv_drm_device *hv = to_hv(fb->dev); @@ -38,7 +38,8 @@ static int hyperv_blit_to_vram_rect(struct drm_framebuffer *fb, return 0; } -static int hyperv_blit_to_vram_fullscreen(struct drm_framebuffer *fb, const struct dma_buf_map *map) +static int hyperv_blit_to_vram_fullscreen(struct drm_framebuffer *fb, + const struct iosys_map *map) { struct drm_rect fullscreen = { .x1 = 0, diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig index cfd932514da2..42b5400f0fef 100644 --- a/drivers/gpu/drm/i915/Kconfig +++ b/drivers/gpu/drm/i915/Kconfig @@ -9,6 +9,7 @@ config DRM_I915 # the shmem_readpage() which depends upon tmpfs select SHMEM select TMPFS + select DRM_DP_HELPER select DRM_KMS_HELPER select DRM_PANEL select DRM_MIPI_DSI @@ -27,6 +28,7 @@ config DRM_I915 select CEC_CORE if CEC_NOTIFIER select VMAP_PFN select DRM_TTM + select DRM_BUDDY help Choose this option if you have a system that has "Intel Graphics Media Accelerator" or "HD Graphics" integrated graphics, diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 1b62b9f65196..139e0e7eba94 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -32,8 +32,9 @@ subdir-ccflags-y += -I$(srctree)/$(src) # core driver code i915-y += i915_driver.o \ i915_config.o \ - i915_irq.o \ i915_getparam.o \ + i915_ioctl.o \ + i915_irq.o \ i915_mitigations.o \ i915_module.o \ i915_params.o \ @@ -161,7 +162,6 @@ gem-y += \ i915-y += \ $(gem-y) \ i915_active.o \ - i915_buddy.o \ i915_cmd_parser.o \ i915_deps.o \ i915_gem_evict.o \ diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c index f37677df6ebf..f67bbaaad8e0 100644 --- a/drivers/gpu/drm/i915/display/g4x_dp.c +++ b/drivers/gpu/drm/i915/display/g4x_dp.c @@ -18,6 +18,7 @@ #include "intel_fifo_underrun.h" #include "intel_hdmi.h" #include "intel_hotplug.h" +#include "intel_pch_display.h" #include "intel_pps.h" #include "vlv_sideband.h" @@ -333,6 +334,21 @@ static bool intel_dp_get_hw_state(struct intel_encoder *encoder, return ret; } +static void g4x_dp_get_m_n(struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + + if (crtc_state->has_pch_encoder) { + intel_pch_transcoder_get_m1_n1(crtc, &crtc_state->dp_m_n); + intel_pch_transcoder_get_m2_n2(crtc, &crtc_state->dp_m2_n2); + } else { + intel_cpu_transcoder_get_m1_n1(crtc, crtc_state->cpu_transcoder, + &crtc_state->dp_m_n); + intel_cpu_transcoder_get_m2_n2(crtc, crtc_state->cpu_transcoder, + &crtc_state->dp_m2_n2); + } +} + static void intel_dp_get_config(struct intel_encoder *encoder, struct intel_crtc_state *pipe_config) { @@ -384,7 +400,7 @@ static void intel_dp_get_config(struct intel_encoder *encoder, pipe_config->lane_count = ((tmp & DP_PORT_WIDTH_MASK) >> DP_PORT_WIDTH_SHIFT) + 1; - intel_dp_get_m_n(crtc, pipe_config); + g4x_dp_get_m_n(pipe_config); if (port == PORT_A) { if ((intel_de_read(dev_priv, DP_A) & DP_PLL_FREQ_MASK) == DP_PLL_FREQ_162MHZ) diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.h b/drivers/gpu/drm/i915/display/g4x_hdmi.h index 7aca14b602c6..db9a93bc9321 100644 --- a/drivers/gpu/drm/i915/display/g4x_hdmi.h +++ b/drivers/gpu/drm/i915/display/g4x_hdmi.h @@ -8,7 +8,7 @@ #include <linux/types.h> -#include "i915_reg.h" +#include "i915_reg_defs.h" enum port; struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c index 85950ff67609..a87b65cd41fd 100644 --- a/drivers/gpu/drm/i915/display/i9xx_plane.c +++ b/drivers/gpu/drm/i915/display/i9xx_plane.c @@ -125,7 +125,7 @@ static struct intel_fbc *i9xx_plane_fbc(struct drm_i915_private *dev_priv, enum i9xx_plane_id i9xx_plane) { if (i9xx_plane_has_fbc(dev_priv, i9xx_plane)) - return dev_priv->fbc; + return dev_priv->fbc[INTEL_FBC_A]; else return NULL; } @@ -155,51 +155,51 @@ static u32 i9xx_plane_ctl(const struct intel_crtc_state *crtc_state, unsigned int rotation = plane_state->hw.rotation; u32 dspcntr; - dspcntr = DISPLAY_PLANE_ENABLE; + dspcntr = DISP_ENABLE; if (IS_G4X(dev_priv) || IS_IRONLAKE(dev_priv) || IS_SANDYBRIDGE(dev_priv) || IS_IVYBRIDGE(dev_priv)) - dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE; + dspcntr |= DISP_TRICKLE_FEED_DISABLE; switch (fb->format->format) { case DRM_FORMAT_C8: - dspcntr |= DISPPLANE_8BPP; + dspcntr |= DISP_FORMAT_8BPP; break; case DRM_FORMAT_XRGB1555: - dspcntr |= DISPPLANE_BGRX555; + dspcntr |= DISP_FORMAT_BGRX555; break; case DRM_FORMAT_ARGB1555: - dspcntr |= DISPPLANE_BGRA555; + dspcntr |= DISP_FORMAT_BGRA555; break; case DRM_FORMAT_RGB565: - dspcntr |= DISPPLANE_BGRX565; + dspcntr |= DISP_FORMAT_BGRX565; break; case DRM_FORMAT_XRGB8888: - dspcntr |= DISPPLANE_BGRX888; + dspcntr |= DISP_FORMAT_BGRX888; break; case DRM_FORMAT_XBGR8888: - dspcntr |= DISPPLANE_RGBX888; + dspcntr |= DISP_FORMAT_RGBX888; break; case DRM_FORMAT_ARGB8888: - dspcntr |= DISPPLANE_BGRA888; + dspcntr |= DISP_FORMAT_BGRA888; break; case DRM_FORMAT_ABGR8888: - dspcntr |= DISPPLANE_RGBA888; + dspcntr |= DISP_FORMAT_RGBA888; break; case DRM_FORMAT_XRGB2101010: - dspcntr |= DISPPLANE_BGRX101010; + dspcntr |= DISP_FORMAT_BGRX101010; break; case DRM_FORMAT_XBGR2101010: - dspcntr |= DISPPLANE_RGBX101010; + dspcntr |= DISP_FORMAT_RGBX101010; break; case DRM_FORMAT_ARGB2101010: - dspcntr |= DISPPLANE_BGRA101010; + dspcntr |= DISP_FORMAT_BGRA101010; break; case DRM_FORMAT_ABGR2101010: - dspcntr |= DISPPLANE_RGBA101010; + dspcntr |= DISP_FORMAT_RGBA101010; break; case DRM_FORMAT_XBGR16161616F: - dspcntr |= DISPPLANE_RGBX161616; + dspcntr |= DISP_FORMAT_RGBX161616; break; default: MISSING_CASE(fb->format->format); @@ -208,13 +208,13 @@ static u32 i9xx_plane_ctl(const struct intel_crtc_state *crtc_state, if (DISPLAY_VER(dev_priv) >= 4 && fb->modifier == I915_FORMAT_MOD_X_TILED) - dspcntr |= DISPPLANE_TILED; + dspcntr |= DISP_TILED; if (rotation & DRM_MODE_ROTATE_180) - dspcntr |= DISPPLANE_ROTATE_180; + dspcntr |= DISP_ROTATE_180; if (rotation & DRM_MODE_REFLECT_X) - dspcntr |= DISPPLANE_MIRROR; + dspcntr |= DISP_MIRROR; return dspcntr; } @@ -354,13 +354,13 @@ static u32 i9xx_plane_ctl_crtc(const struct intel_crtc_state *crtc_state) u32 dspcntr = 0; if (crtc_state->gamma_enable) - dspcntr |= DISPPLANE_GAMMA_ENABLE; + dspcntr |= DISP_PIPE_GAMMA_ENABLE; if (crtc_state->csc_enable) - dspcntr |= DISPPLANE_PIPE_CSC_ENABLE; + dspcntr |= DISP_PIPE_CSC_ENABLE; if (DISPLAY_VER(dev_priv) < 5) - dspcntr |= DISPPLANE_SEL_PIPE(crtc->pipe); + dspcntr |= DISP_PIPE_SEL(crtc->pipe); return dspcntr; } @@ -437,9 +437,9 @@ static void i9xx_plane_update_noarm(struct intel_plane *plane, * program whatever is there. */ intel_de_write_fw(dev_priv, DSPPOS(i9xx_plane), - (crtc_y << 16) | crtc_x); + DISP_POS_Y(crtc_y) | DISP_POS_X(crtc_x)); intel_de_write_fw(dev_priv, DSPSIZE(i9xx_plane), - ((crtc_h - 1) << 16) | (crtc_w - 1)); + DISP_HEIGHT(crtc_h - 1) | DISP_WIDTH(crtc_w - 1)); } spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); @@ -474,20 +474,20 @@ static void i9xx_plane_update_arm(struct intel_plane *plane, int crtc_h = drm_rect_height(&plane_state->uapi.dst); intel_de_write_fw(dev_priv, PRIMPOS(i9xx_plane), - (crtc_y << 16) | crtc_x); + PRIM_POS_Y(crtc_y) | PRIM_POS_X(crtc_x)); intel_de_write_fw(dev_priv, PRIMSIZE(i9xx_plane), - ((crtc_h - 1) << 16) | (crtc_w - 1)); + PRIM_HEIGHT(crtc_h - 1) | PRIM_WIDTH(crtc_w - 1)); intel_de_write_fw(dev_priv, PRIMCNSTALPHA(i9xx_plane), 0); } if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) { intel_de_write_fw(dev_priv, DSPOFFSET(i9xx_plane), - (y << 16) | x); + DISP_OFFSET_Y(y) | DISP_OFFSET_X(x)); } else if (DISPLAY_VER(dev_priv) >= 4) { intel_de_write_fw(dev_priv, DSPLINOFF(i9xx_plane), linear_offset); intel_de_write_fw(dev_priv, DSPTILEOFF(i9xx_plane), - (y << 16) | x); + DISP_OFFSET_Y(y) | DISP_OFFSET_X(x)); } /* @@ -564,7 +564,7 @@ g4x_primary_async_flip(struct intel_plane *plane, unsigned long irqflags; if (async_flip) - dspcntr |= DISPPLANE_ASYNC_FLIP; + dspcntr |= DISP_ASYNC_FLIP; spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); intel_de_write_fw(dev_priv, DSPCNTR(i9xx_plane), dspcntr); @@ -696,13 +696,12 @@ static bool i9xx_plane_get_hw_state(struct intel_plane *plane, val = intel_de_read(dev_priv, DSPCNTR(i9xx_plane)); - ret = val & DISPLAY_PLANE_ENABLE; + ret = val & DISP_ENABLE; if (DISPLAY_VER(dev_priv) >= 5) *pipe = plane->pipe; else - *pipe = (val & DISPPLANE_SEL_PIPE_MASK) >> - DISPPLANE_SEL_PIPE_SHIFT; + *pipe = REG_FIELD_GET(DISP_PIPE_SEL_MASK, val); intel_display_power_put(dev_priv, power_domain, wakeref); @@ -958,32 +957,32 @@ fail: static int i9xx_format_to_fourcc(int format) { switch (format) { - case DISPPLANE_8BPP: + case DISP_FORMAT_8BPP: return DRM_FORMAT_C8; - case DISPPLANE_BGRA555: + case DISP_FORMAT_BGRA555: return DRM_FORMAT_ARGB1555; - case DISPPLANE_BGRX555: + case DISP_FORMAT_BGRX555: return DRM_FORMAT_XRGB1555; - case DISPPLANE_BGRX565: + case DISP_FORMAT_BGRX565: return DRM_FORMAT_RGB565; default: - case DISPPLANE_BGRX888: + case DISP_FORMAT_BGRX888: return DRM_FORMAT_XRGB8888; - case DISPPLANE_RGBX888: + case DISP_FORMAT_RGBX888: return DRM_FORMAT_XBGR8888; - case DISPPLANE_BGRA888: + case DISP_FORMAT_BGRA888: return DRM_FORMAT_ARGB8888; - case DISPPLANE_RGBA888: + case DISP_FORMAT_RGBA888: return DRM_FORMAT_ABGR8888; - case DISPPLANE_BGRX101010: + case DISP_FORMAT_BGRX101010: return DRM_FORMAT_XRGB2101010; - case DISPPLANE_RGBX101010: + case DISP_FORMAT_RGBX101010: return DRM_FORMAT_XBGR2101010; - case DISPPLANE_BGRA101010: + case DISP_FORMAT_BGRA101010: return DRM_FORMAT_ARGB2101010; - case DISPPLANE_RGBA101010: + case DISP_FORMAT_RGBA101010: return DRM_FORMAT_ABGR2101010; - case DISPPLANE_RGBX161616: + case DISP_FORMAT_RGBX161616: return DRM_FORMAT_XBGR16161616F; } } @@ -1021,26 +1020,26 @@ i9xx_get_initial_plane_config(struct intel_crtc *crtc, val = intel_de_read(dev_priv, DSPCNTR(i9xx_plane)); if (DISPLAY_VER(dev_priv) >= 4) { - if (val & DISPPLANE_TILED) { + if (val & DISP_TILED) { plane_config->tiling = I915_TILING_X; fb->modifier = I915_FORMAT_MOD_X_TILED; } - if (val & DISPPLANE_ROTATE_180) + if (val & DISP_ROTATE_180) plane_config->rotation = DRM_MODE_ROTATE_180; } if (IS_CHERRYVIEW(dev_priv) && pipe == PIPE_B && - val & DISPPLANE_MIRROR) + val & DISP_MIRROR) plane_config->rotation |= DRM_MODE_REFLECT_X; - pixel_format = val & DISPPLANE_PIXFORMAT_MASK; + pixel_format = val & DISP_FORMAT_MASK; fourcc = i9xx_format_to_fourcc(pixel_format); fb->format = drm_format_info(fourcc); if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) { offset = intel_de_read(dev_priv, DSPOFFSET(i9xx_plane)); - base = intel_de_read(dev_priv, DSPSURF(i9xx_plane)) & 0xfffff000; + base = intel_de_read(dev_priv, DSPSURF(i9xx_plane)) & DISP_ADDR_MASK; } else if (DISPLAY_VER(dev_priv) >= 4) { if (plane_config->tiling) offset = intel_de_read(dev_priv, @@ -1048,15 +1047,15 @@ i9xx_get_initial_plane_config(struct intel_crtc *crtc, else offset = intel_de_read(dev_priv, DSPLINOFF(i9xx_plane)); - base = intel_de_read(dev_priv, DSPSURF(i9xx_plane)) & 0xfffff000; + base = intel_de_read(dev_priv, DSPSURF(i9xx_plane)) & DISP_ADDR_MASK; } else { base = intel_de_read(dev_priv, DSPADDR(i9xx_plane)); } plane_config->base = base; val = intel_de_read(dev_priv, PIPESRC(pipe)); - fb->width = ((val >> 16) & 0xfff) + 1; - fb->height = ((val >> 0) & 0xfff) + 1; + fb->width = REG_FIELD_GET(PIPESRC_WIDTH_MASK, val) + 1; + fb->height = REG_FIELD_GET(PIPESRC_HEIGHT_MASK, val) + 1; val = intel_de_read(dev_priv, DSPSTRIDE(i9xx_plane)); fb->pitches[0] = val & 0xffffffc0; diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c index 5781e9fac8b4..2d5bb9195b20 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi.c +++ b/drivers/gpu/drm/i915/display/icl_dsi.c @@ -32,6 +32,7 @@ #include "intel_atomic.h" #include "intel_backlight.h" #include "intel_combo_phy.h" +#include "intel_combo_phy_regs.h" #include "intel_connector.h" #include "intel_crtc.h" #include "intel_ddi.h" @@ -1050,7 +1051,7 @@ static void gen11_dsi_enable_transcoder(struct intel_encoder *encoder) /* wait for transcoder to be enabled */ if (intel_de_wait_for_set(dev_priv, PIPECONF(dsi_trans), - I965_PIPECONF_ACTIVE, 10)) + PIPECONF_STATE_ENABLE, 10)) drm_err(&dev_priv->drm, "DSI transcoder not enabled\n"); } @@ -1232,8 +1233,6 @@ static void gen11_dsi_pre_enable(struct intel_atomic_state *state, intel_dsc_dsi_pps_write(encoder, pipe_config); - intel_dsc_enable(pipe_config); - /* step6c: configure transcoder timings */ gen11_dsi_set_transcoder_timings(encoder, pipe_config); } @@ -1320,7 +1319,7 @@ static void gen11_dsi_disable_transcoder(struct intel_encoder *encoder) /* wait for transcoder to be disabled */ if (intel_de_wait_for_clear(dev_priv, PIPECONF(dsi_trans), - I965_PIPECONF_ACTIVE, 50)) + PIPECONF_STATE_ENABLE, 50)) drm_err(&dev_priv->drm, "DSI trancoder not disabled\n"); } diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c b/drivers/gpu/drm/i915/display/intel_atomic.c index a62550711e98..093904065112 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic.c +++ b/drivers/gpu/drm/i915/display/intel_atomic.c @@ -34,6 +34,8 @@ #include <drm/drm_fourcc.h> #include <drm/drm_plane_helper.h> +#include "i915_drv.h" +#include "i915_reg.h" #include "intel_atomic.h" #include "intel_cdclk.h" #include "intel_display_types.h" diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c index c2c512cd8ec0..bec02333bdeb 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c @@ -109,6 +109,7 @@ intel_plane_duplicate_state(struct drm_plane *plane) intel_state->ggtt_vma = NULL; intel_state->dpt_vma = NULL; intel_state->flags = 0; + intel_state->do_async_flip = false; /* add reference to fb */ if (intel_state->hw.fb) @@ -491,7 +492,7 @@ void intel_plane_update_arm(struct intel_plane *plane, trace_intel_plane_update_arm(&plane->base, crtc); - if (crtc_state->uapi.async_flip && plane->async_flip) + if (plane_state->do_async_flip) plane->async_flip(plane, crtc_state, plane_state, true); else plane->update_arm(plane, crtc_state, plane_state); @@ -601,6 +602,7 @@ int intel_atomic_plane_check_clipping(struct intel_plane_state *plane_state, int min_scale, int max_scale, bool can_position) { + struct drm_i915_private *i915 = to_i915(plane_state->uapi.plane->dev); struct drm_framebuffer *fb = plane_state->hw.fb; struct drm_rect *src = &plane_state->uapi.src; struct drm_rect *dst = &plane_state->uapi.dst; @@ -619,7 +621,7 @@ int intel_atomic_plane_check_clipping(struct intel_plane_state *plane_state, hscale = drm_rect_calc_hscale(src, dst, min_scale, max_scale); vscale = drm_rect_calc_vscale(src, dst, min_scale, max_scale); if (hscale < 0 || vscale < 0) { - DRM_DEBUG_KMS("Invalid scaling of plane\n"); + drm_dbg_kms(&i915->drm, "Invalid scaling of plane\n"); drm_rect_debug_print("src: ", src, true); drm_rect_debug_print("dst: ", dst, false); return -ERANGE; @@ -644,7 +646,7 @@ int intel_atomic_plane_check_clipping(struct intel_plane_state *plane_state, if (!can_position && plane_state->uapi.visible && !drm_rect_equals(dst, &clip)) { - DRM_DEBUG_KMS("Plane must cover entire CRTC\n"); + drm_dbg_kms(&i915->drm, "Plane must cover entire CRTC\n"); drm_rect_debug_print("dst: ", dst, false); drm_rect_debug_print("clip: ", &clip, false); return -EINVAL; diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.h b/drivers/gpu/drm/i915/display/intel_atomic_plane.h index 7907f601598e..ead789709477 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic_plane.h +++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.h @@ -16,6 +16,7 @@ struct intel_crtc; struct intel_crtc_state; struct intel_plane; struct intel_plane_state; +enum plane_id; unsigned int intel_adjusted_rate(const struct drm_rect *src, const struct drm_rect *dst, diff --git a/drivers/gpu/drm/i915/display/intel_backlight.c b/drivers/gpu/drm/i915/display/intel_backlight.c index 9523411cddd8..98f7ea44042f 100644 --- a/drivers/gpu/drm/i915/display/intel_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_backlight.c @@ -13,6 +13,7 @@ #include "intel_dp_aux_backlight.h" #include "intel_dsi_dcs_backlight.h" #include "intel_panel.h" +#include "intel_pci_config.h" /** * scale - scale values from one range to another @@ -433,6 +434,8 @@ static void ext_pwm_disable_backlight(const struct drm_connector_state *old_conn struct intel_connector *connector = to_intel_connector(old_conn_state->connector); struct intel_panel *panel = &connector->panel; + intel_backlight_set_pwm_level(old_conn_state, level); + panel->backlight.pwm_state.enabled = false; pwm_apply_state(panel->backlight.pwm, &panel->backlight.pwm_state); } diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c index 9d989c9f5da4..aec0efd5350e 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.c +++ b/drivers/gpu/drm/i915/display/intel_bios.c @@ -25,13 +25,14 @@ * */ -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include "display/intel_display.h" #include "display/intel_display_types.h" #include "display/intel_gmbus.h" #include "i915_drv.h" +#include "i915_reg.h" #define _INTEL_BIOS_PRIVATE #include "intel_vbt_defs.h" @@ -905,26 +906,6 @@ parse_psr(struct drm_i915_private *i915, const struct bdb_header *bdb) i915->vbt.psr.idle_frames = psr_table->idle_frames < 0 ? 0 : psr_table->idle_frames > 15 ? 15 : psr_table->idle_frames; - switch (psr_table->lines_to_wait) { - case 0: - i915->vbt.psr.lines_to_wait = PSR_0_LINES_TO_WAIT; - break; - case 1: - i915->vbt.psr.lines_to_wait = PSR_1_LINE_TO_WAIT; - break; - case 2: - i915->vbt.psr.lines_to_wait = PSR_4_LINES_TO_WAIT; - break; - case 3: - i915->vbt.psr.lines_to_wait = PSR_8_LINES_TO_WAIT; - break; - default: - drm_dbg_kms(&i915->drm, - "VBT has unknown PSR lines to wait %u\n", - psr_table->lines_to_wait); - break; - } - /* * New psr options 0=500us, 1=100us, 2=2500us, 3=0us * Old decimal value is wake up time in multiples of 100 us. @@ -2073,14 +2054,16 @@ static void parse_ddi_port(struct drm_i915_private *i915, i915->vbt.ports[port] = devdata; } +static bool has_ddi_port_info(struct drm_i915_private *i915) +{ + return DISPLAY_VER(i915) >= 5 || IS_G4X(i915); +} + static void parse_ddi_ports(struct drm_i915_private *i915) { struct intel_bios_encoder_data *devdata; - if (!HAS_DDI(i915) && !IS_CHERRYVIEW(i915)) - return; - - if (i915->vbt.version < 155) + if (!has_ddi_port_info(i915)) return; list_for_each_entry(devdata, &i915->vbt.display_devices, node) @@ -2335,6 +2318,63 @@ bool intel_bios_is_valid_vbt(const void *buf, size_t size) return vbt; } +static struct vbt_header *spi_oprom_get_vbt(struct drm_i915_private *i915) +{ + u32 count, data, found, store = 0; + u32 static_region, oprom_offset; + u32 oprom_size = 0x200000; + u16 vbt_size; + u32 *vbt; + + static_region = intel_uncore_read(&i915->uncore, SPI_STATIC_REGIONS); + static_region &= OPTIONROM_SPI_REGIONID_MASK; + intel_uncore_write(&i915->uncore, PRIMARY_SPI_REGIONID, static_region); + + oprom_offset = intel_uncore_read(&i915->uncore, OROM_OFFSET); + oprom_offset &= OROM_OFFSET_MASK; + + for (count = 0; count < oprom_size; count += 4) { + intel_uncore_write(&i915->uncore, PRIMARY_SPI_ADDRESS, oprom_offset + count); + data = intel_uncore_read(&i915->uncore, PRIMARY_SPI_TRIGGER); + + if (data == *((const u32 *)"$VBT")) { + found = oprom_offset + count; + break; + } + } + + if (count >= oprom_size) + goto err_not_found; + + /* Get VBT size and allocate space for the VBT */ + intel_uncore_write(&i915->uncore, PRIMARY_SPI_ADDRESS, found + + offsetof(struct vbt_header, vbt_size)); + vbt_size = intel_uncore_read(&i915->uncore, PRIMARY_SPI_TRIGGER); + vbt_size &= 0xffff; + + vbt = kzalloc(round_up(vbt_size, 4), GFP_KERNEL); + if (!vbt) + goto err_not_found; + + for (count = 0; count < vbt_size; count += 4) { + intel_uncore_write(&i915->uncore, PRIMARY_SPI_ADDRESS, found + count); + data = intel_uncore_read(&i915->uncore, PRIMARY_SPI_TRIGGER); + *(vbt + store++) = data; + } + + if (!intel_bios_is_valid_vbt(vbt, vbt_size)) + goto err_free_vbt; + + drm_dbg_kms(&i915->drm, "Found valid VBT in SPI flash\n"); + + return (struct vbt_header *)vbt; + +err_free_vbt: + kfree(vbt); +err_not_found: + return NULL; +} + static struct vbt_header *oprom_get_vbt(struct drm_i915_private *i915) { struct pci_dev *pdev = to_pci_dev(i915->drm.dev); @@ -2384,6 +2424,8 @@ static struct vbt_header *oprom_get_vbt(struct drm_i915_private *i915) pci_unmap_rom(pdev, oprom); + drm_dbg_kms(&i915->drm, "Found valid VBT in PCI ROM\n"); + return vbt; err_free_vbt: @@ -2418,17 +2460,23 @@ void intel_bios_init(struct drm_i915_private *i915) init_vbt_defaults(i915); - /* If the OpRegion does not have VBT, look in PCI ROM. */ + /* + * If the OpRegion does not have VBT, look in SPI flash through MMIO or + * PCI mapping + */ + if (!vbt && IS_DGFX(i915)) { + oprom_vbt = spi_oprom_get_vbt(i915); + vbt = oprom_vbt; + } + if (!vbt) { oprom_vbt = oprom_get_vbt(i915); - if (!oprom_vbt) - goto out; - vbt = oprom_vbt; - - drm_dbg_kms(&i915->drm, "Found valid VBT in PCI ROM\n"); } + if (!vbt) + goto out; + bdb = get_bdb_header(vbt); i915->vbt.version = bdb->version; @@ -2596,37 +2644,10 @@ bool intel_bios_is_lvds_present(struct drm_i915_private *i915, u8 *i2c_pin) */ bool intel_bios_is_port_present(struct drm_i915_private *i915, enum port port) { - const struct intel_bios_encoder_data *devdata; - const struct child_device_config *child; - static const struct { - u16 dp, hdmi; - } port_mapping[] = { - [PORT_B] = { DVO_PORT_DPB, DVO_PORT_HDMIB, }, - [PORT_C] = { DVO_PORT_DPC, DVO_PORT_HDMIC, }, - [PORT_D] = { DVO_PORT_DPD, DVO_PORT_HDMID, }, - [PORT_E] = { DVO_PORT_DPE, DVO_PORT_HDMIE, }, - [PORT_F] = { DVO_PORT_DPF, DVO_PORT_HDMIF, }, - }; - - if (HAS_DDI(i915)) - return i915->vbt.ports[port]; - - /* FIXME maybe deal with port A as well? */ - if (drm_WARN_ON(&i915->drm, - port == PORT_A) || port >= ARRAY_SIZE(port_mapping)) - return false; - - list_for_each_entry(devdata, &i915->vbt.display_devices, node) { - child = &devdata->child; - - if ((child->dvo_port == port_mapping[port].dp || - child->dvo_port == port_mapping[port].hdmi) && - (child->device_type & (DEVICE_TYPE_TMDS_DVI_SIGNALING | - DEVICE_TYPE_DISPLAYPORT_OUTPUT))) - return true; - } + if (WARN_ON(!has_ddi_port_info(i915))) + return true; - return false; + return i915->vbt.ports[port]; } /** @@ -2638,40 +2659,18 @@ bool intel_bios_is_port_present(struct drm_i915_private *i915, enum port port) */ bool intel_bios_is_port_edp(struct drm_i915_private *i915, enum port port) { - const struct intel_bios_encoder_data *devdata; - const struct child_device_config *child; - static const short port_mapping[] = { - [PORT_B] = DVO_PORT_DPB, - [PORT_C] = DVO_PORT_DPC, - [PORT_D] = DVO_PORT_DPD, - [PORT_E] = DVO_PORT_DPE, - [PORT_F] = DVO_PORT_DPF, - }; - - if (HAS_DDI(i915)) { - const struct intel_bios_encoder_data *devdata; - - devdata = intel_bios_encoder_data_lookup(i915, port); - - return devdata && intel_bios_encoder_supports_edp(devdata); - } - - list_for_each_entry(devdata, &i915->vbt.display_devices, node) { - child = &devdata->child; - - if (child->dvo_port == port_mapping[port] && - (child->device_type & DEVICE_TYPE_eDP_BITS) == - (DEVICE_TYPE_eDP & DEVICE_TYPE_eDP_BITS)) - return true; - } + const struct intel_bios_encoder_data *devdata = + intel_bios_encoder_data_lookup(i915, port); - return false; + return devdata && intel_bios_encoder_supports_edp(devdata); } -static bool child_dev_is_dp_dual_mode(const struct child_device_config *child) +static bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_encoder_data *devdata) { - if ((child->device_type & DEVICE_TYPE_DP_DUAL_MODE_BITS) != - (DEVICE_TYPE_DP_DUAL_MODE & DEVICE_TYPE_DP_DUAL_MODE_BITS)) + const struct child_device_config *child = &devdata->child; + + if (!intel_bios_encoder_supports_dp(devdata) || + !intel_bios_encoder_supports_hdmi(devdata)) return false; if (dvo_port_type(child->dvo_port) == DVO_PORT_DPA) @@ -2688,40 +2687,10 @@ static bool child_dev_is_dp_dual_mode(const struct child_device_config *child) bool intel_bios_is_port_dp_dual_mode(struct drm_i915_private *i915, enum port port) { - static const struct { - u16 dp, hdmi; - } port_mapping[] = { - /* - * Buggy VBTs may declare DP ports as having - * HDMI type dvo_port :( So let's check both. - */ - [PORT_B] = { DVO_PORT_DPB, DVO_PORT_HDMIB, }, - [PORT_C] = { DVO_PORT_DPC, DVO_PORT_HDMIC, }, - [PORT_D] = { DVO_PORT_DPD, DVO_PORT_HDMID, }, - [PORT_E] = { DVO_PORT_DPE, DVO_PORT_HDMIE, }, - [PORT_F] = { DVO_PORT_DPF, DVO_PORT_HDMIF, }, - }; - const struct intel_bios_encoder_data *devdata; - - if (HAS_DDI(i915)) { - const struct intel_bios_encoder_data *devdata; - - devdata = intel_bios_encoder_data_lookup(i915, port); - - return devdata && child_dev_is_dp_dual_mode(&devdata->child); - } - - if (port == PORT_A || port >= ARRAY_SIZE(port_mapping)) - return false; - - list_for_each_entry(devdata, &i915->vbt.display_devices, node) { - if ((devdata->child.dvo_port == port_mapping[port].dp || - devdata->child.dvo_port == port_mapping[port].hdmi) && - child_dev_is_dp_dual_mode(&devdata->child)) - return true; - } + const struct intel_bios_encoder_data *devdata = + intel_bios_encoder_data_lookup(i915, port); - return false; + return devdata && intel_bios_encoder_supports_dp_dual_mode(devdata); } /** diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c index 2da4aacc956b..5dce3cf0ed12 100644 --- a/drivers/gpu/drm/i915/display/intel_bw.c +++ b/drivers/gpu/drm/i915/display/intel_bw.c @@ -5,6 +5,7 @@ #include <drm/drm_atomic_state_helper.h> +#include "i915_reg.h" #include "intel_atomic.h" #include "intel_bw.h" #include "intel_cdclk.h" @@ -75,10 +76,9 @@ static int icl_pcode_read_qgv_point_info(struct drm_i915_private *dev_priv, u16 dclk; int ret; - ret = sandybridge_pcode_read(dev_priv, - ICL_PCODE_MEM_SUBSYSYSTEM_INFO | - ICL_PCODE_MEM_SS_READ_QGV_POINT_INFO(point), - &val, &val2); + ret = snb_pcode_read(dev_priv, ICL_PCODE_MEM_SUBSYSYSTEM_INFO | + ICL_PCODE_MEM_SS_READ_QGV_POINT_INFO(point), + &val, &val2); if (ret) return ret; @@ -102,10 +102,8 @@ static int adls_pcode_read_psf_gv_point_info(struct drm_i915_private *dev_priv, int ret; int i; - ret = sandybridge_pcode_read(dev_priv, - ICL_PCODE_MEM_SUBSYSYSTEM_INFO | - ADL_PCODE_MEM_SS_READ_PSF_GV_INFO, - &val, NULL); + ret = snb_pcode_read(dev_priv, ICL_PCODE_MEM_SUBSYSYSTEM_INFO | + ADL_PCODE_MEM_SS_READ_PSF_GV_INFO, &val, NULL); if (ret) return ret; diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c index c30cf8d2b835..4b140a014ca8 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.c +++ b/drivers/gpu/drm/i915/display/intel_cdclk.c @@ -31,6 +31,7 @@ #include "intel_crtc.h" #include "intel_de.h" #include "intel_display_types.h" +#include "intel_pci_config.h" #include "intel_pcode.h" #include "intel_psr.h" #include "vlv_sideband.h" @@ -63,6 +64,17 @@ * dividers can be programmed correctly. */ +struct intel_cdclk_funcs { + void (*get_cdclk)(struct drm_i915_private *i915, + struct intel_cdclk_config *cdclk_config); + void (*set_cdclk)(struct drm_i915_private *i915, + const struct intel_cdclk_config *cdclk_config, + enum pipe pipe); + int (*bw_calc_min_cdclk)(struct intel_atomic_state *state); + int (*modeset_calc_cdclk)(struct intel_cdclk_state *state); + u8 (*calc_voltage_level)(int cdclk); +}; + void intel_cdclk_get_cdclk(struct drm_i915_private *dev_priv, struct intel_cdclk_config *cdclk_config) { @@ -793,8 +805,7 @@ static void bdw_set_cdclk(struct drm_i915_private *dev_priv, "trying to change cdclk frequency with cdclk not enabled\n")) return; - ret = sandybridge_pcode_write(dev_priv, - BDW_PCODE_DISPLAY_FREQ_CHANGE_REQ, 0x0); + ret = snb_pcode_write(dev_priv, BDW_PCODE_DISPLAY_FREQ_CHANGE_REQ, 0x0); if (ret) { drm_err(&dev_priv->drm, "failed to inform pcode about cdclk change\n"); @@ -822,8 +833,8 @@ static void bdw_set_cdclk(struct drm_i915_private *dev_priv, LCPLL_CD_SOURCE_FCLK_DONE) == 0, 1)) drm_err(&dev_priv->drm, "Switching back to LCPLL failed\n"); - sandybridge_pcode_write(dev_priv, HSW_PCODE_DE_WRITE_FREQ_REQ, - cdclk_config->voltage_level); + snb_pcode_write(dev_priv, HSW_PCODE_DE_WRITE_FREQ_REQ, + cdclk_config->voltage_level); intel_de_write(dev_priv, CDCLK_FREQ, DIV_ROUND_CLOSEST(cdclk, 1000) - 1); @@ -1126,8 +1137,8 @@ static void skl_set_cdclk(struct drm_i915_private *dev_priv, intel_de_posting_read(dev_priv, CDCLK_CTL); /* inform PCU of the change */ - sandybridge_pcode_write(dev_priv, SKL_PCODE_CDCLK_CONTROL, - cdclk_config->voltage_level); + snb_pcode_write(dev_priv, SKL_PCODE_CDCLK_CONTROL, + cdclk_config->voltage_level); intel_update_cdclk(dev_priv); } @@ -1145,7 +1156,7 @@ static void skl_sanitize_cdclk(struct drm_i915_private *dev_priv) goto sanitize; intel_update_cdclk(dev_priv); - intel_dump_cdclk_config(&dev_priv->cdclk.hw, "Current CDCLK"); + intel_cdclk_dump_config(dev_priv, &dev_priv->cdclk.hw, "Current CDCLK"); /* Is PLL enabled and locked ? */ if (dev_priv->cdclk.hw.vco == 0 || @@ -1614,7 +1625,7 @@ static void adlp_cdclk_pll_crawl(struct drm_i915_private *dev_priv, int vco) /* Timeout 200us */ if (intel_de_wait_for_set(dev_priv, BXT_DE_PLL_ENABLE, BXT_DE_PLL_LOCK | BXT_DE_PLL_FREQ_REQ_ACK, 1)) - DRM_ERROR("timeout waiting for FREQ change request ack\n"); + drm_err(&dev_priv->drm, "timeout waiting for FREQ change request ack\n"); val &= ~BXT_DE_PLL_FREQ_REQ; intel_de_write(dev_priv, BXT_DE_PLL_ENABLE, val); @@ -1705,10 +1716,9 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, * BSpec requires us to wait up to 150usec, but that leads to * timeouts; the 2ms used here is based on experiment. */ - ret = sandybridge_pcode_write_timeout(dev_priv, - HSW_PCODE_DE_WRITE_FREQ_REQ, - 0x80000000, 150, 2); - + ret = snb_pcode_write_timeout(dev_priv, + HSW_PCODE_DE_WRITE_FREQ_REQ, + 0x80000000, 150, 2); if (ret) { drm_err(&dev_priv->drm, "Failed to inform PCU about cdclk change (err %d, freq %d)\n", @@ -1769,8 +1779,8 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, intel_crtc_wait_for_next_vblank(intel_crtc_for_pipe(dev_priv, pipe)); if (DISPLAY_VER(dev_priv) >= 11) { - ret = sandybridge_pcode_write(dev_priv, SKL_PCODE_CDCLK_CONTROL, - cdclk_config->voltage_level); + ret = snb_pcode_write(dev_priv, SKL_PCODE_CDCLK_CONTROL, + cdclk_config->voltage_level); } else { /* * The timeout isn't specified, the 2ms used here is based on @@ -1778,10 +1788,10 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, * FIXME: Waiting for the request completion could be delayed * until the next PCODE request based on BSpec. */ - ret = sandybridge_pcode_write_timeout(dev_priv, - HSW_PCODE_DE_WRITE_FREQ_REQ, - cdclk_config->voltage_level, - 150, 2); + ret = snb_pcode_write_timeout(dev_priv, + HSW_PCODE_DE_WRITE_FREQ_REQ, + cdclk_config->voltage_level, + 150, 2); } if (ret) { @@ -1807,7 +1817,7 @@ static void bxt_sanitize_cdclk(struct drm_i915_private *dev_priv) int cdclk, clock, vco; intel_update_cdclk(dev_priv); - intel_dump_cdclk_config(&dev_priv->cdclk.hw, "Current CDCLK"); + intel_cdclk_dump_config(dev_priv, &dev_priv->cdclk.hw, "Current CDCLK"); if (dev_priv->cdclk.hw.vco == 0 || dev_priv->cdclk.hw.cdclk == dev_priv->cdclk.hw.bypass) @@ -2047,13 +2057,14 @@ static bool intel_cdclk_changed(const struct intel_cdclk_config *a, a->voltage_level != b->voltage_level; } -void intel_dump_cdclk_config(const struct intel_cdclk_config *cdclk_config, +void intel_cdclk_dump_config(struct drm_i915_private *i915, + const struct intel_cdclk_config *cdclk_config, const char *context) { - DRM_DEBUG_DRIVER("%s %d kHz, VCO %d kHz, ref %d kHz, bypass %d kHz, voltage level %d\n", - context, cdclk_config->cdclk, cdclk_config->vco, - cdclk_config->ref, cdclk_config->bypass, - cdclk_config->voltage_level); + drm_dbg_kms(&i915->drm, "%s %d kHz, VCO %d kHz, ref %d kHz, bypass %d kHz, voltage level %d\n", + context, cdclk_config->cdclk, cdclk_config->vco, + cdclk_config->ref, cdclk_config->bypass, + cdclk_config->voltage_level); } /** @@ -2077,7 +2088,7 @@ static void intel_set_cdclk(struct drm_i915_private *dev_priv, if (drm_WARN_ON_ONCE(&dev_priv->drm, !dev_priv->cdclk_funcs->set_cdclk)) return; - intel_dump_cdclk_config(cdclk_config, "Changing CDCLK to"); + intel_cdclk_dump_config(dev_priv, cdclk_config, "Changing CDCLK to"); for_each_intel_encoder_with_psr(&dev_priv->drm, encoder) { struct intel_dp *intel_dp = enc_to_intel_dp(encoder); @@ -2120,8 +2131,8 @@ static void intel_set_cdclk(struct drm_i915_private *dev_priv, if (drm_WARN(&dev_priv->drm, intel_cdclk_changed(&dev_priv->cdclk.hw, cdclk_config), "cdclk state doesn't match!\n")) { - intel_dump_cdclk_config(&dev_priv->cdclk.hw, "[hw state]"); - intel_dump_cdclk_config(cdclk_config, "[sw state]"); + intel_cdclk_dump_config(dev_priv, &dev_priv->cdclk.hw, "[hw state]"); + intel_cdclk_dump_config(dev_priv, cdclk_config, "[sw state]"); } } diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.h b/drivers/gpu/drm/i915/display/intel_cdclk.h index fc638522e445..df66f66fbad0 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.h +++ b/drivers/gpu/drm/i915/display/intel_cdclk.h @@ -8,7 +8,6 @@ #include <linux/types.h> -#include "i915_drv.h" #include "intel_display.h" #include "intel_global_state.h" @@ -16,6 +15,11 @@ struct drm_i915_private; struct intel_atomic_state; struct intel_crtc_state; +struct intel_cdclk_config { + unsigned int cdclk, vco, ref, bypass; + u8 voltage_level; +}; + struct intel_cdclk_state { struct intel_global_state base; @@ -58,7 +62,8 @@ bool intel_cdclk_needs_modeset(const struct intel_cdclk_config *a, const struct intel_cdclk_config *b); void intel_set_cdclk_pre_plane_update(struct intel_atomic_state *state); void intel_set_cdclk_post_plane_update(struct intel_atomic_state *state); -void intel_dump_cdclk_config(const struct intel_cdclk_config *cdclk_config, +void intel_cdclk_dump_config(struct drm_i915_private *i915, + const struct intel_cdclk_config *cdclk_config, const char *context); int intel_modeset_calc_cdclk(struct intel_atomic_state *state); void intel_cdclk_get_cdclk(struct drm_i915_private *dev_priv, diff --git a/drivers/gpu/drm/i915/display/intel_combo_phy.c b/drivers/gpu/drm/i915/display/intel_combo_phy.c index f628e0542933..4dfe77351b8b 100644 --- a/drivers/gpu/drm/i915/display/intel_combo_phy.c +++ b/drivers/gpu/drm/i915/display/intel_combo_phy.c @@ -4,6 +4,7 @@ */ #include "intel_combo_phy.h" +#include "intel_combo_phy_regs.h" #include "intel_de.h" #include "intel_display_types.h" diff --git a/drivers/gpu/drm/i915/display/intel_combo_phy_regs.h b/drivers/gpu/drm/i915/display/intel_combo_phy_regs.h new file mode 100644 index 000000000000..2ed65193ca19 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_combo_phy_regs.h @@ -0,0 +1,162 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __INTEL_COMBO_PHY_REGS__ +#define __INTEL_COMBO_PHY_REGS__ + +#include "i915_reg_defs.h" + +#define _ICL_COMBOPHY_A 0x162000 +#define _ICL_COMBOPHY_B 0x6C000 +#define _EHL_COMBOPHY_C 0x160000 +#define _RKL_COMBOPHY_D 0x161000 +#define _ADL_COMBOPHY_E 0x16B000 + +#define _ICL_COMBOPHY(phy) _PICK(phy, _ICL_COMBOPHY_A, \ + _ICL_COMBOPHY_B, \ + _EHL_COMBOPHY_C, \ + _RKL_COMBOPHY_D, \ + _ADL_COMBOPHY_E) + +/* ICL Port CL_DW registers */ +#define _ICL_PORT_CL_DW(dw, phy) (_ICL_COMBOPHY(phy) + \ + 4 * (dw)) + +#define ICL_PORT_CL_DW5(phy) _MMIO(_ICL_PORT_CL_DW(5, phy)) +#define CL_POWER_DOWN_ENABLE (1 << 4) +#define SUS_CLOCK_CONFIG (3 << 0) + +#define ICL_PORT_CL_DW10(phy) _MMIO(_ICL_PORT_CL_DW(10, phy)) +#define PG_SEQ_DELAY_OVERRIDE_MASK (3 << 25) +#define PG_SEQ_DELAY_OVERRIDE_SHIFT 25 +#define PG_SEQ_DELAY_OVERRIDE_ENABLE (1 << 24) +#define PWR_UP_ALL_LANES (0x0 << 4) +#define PWR_DOWN_LN_3_2_1 (0xe << 4) +#define PWR_DOWN_LN_3_2 (0xc << 4) +#define PWR_DOWN_LN_3 (0x8 << 4) +#define PWR_DOWN_LN_2_1_0 (0x7 << 4) +#define PWR_DOWN_LN_1_0 (0x3 << 4) +#define PWR_DOWN_LN_3_1 (0xa << 4) +#define PWR_DOWN_LN_3_1_0 (0xb << 4) +#define PWR_DOWN_LN_MASK (0xf << 4) +#define PWR_DOWN_LN_SHIFT 4 +#define EDP4K2K_MODE_OVRD_EN (1 << 3) +#define EDP4K2K_MODE_OVRD_OPTIMIZED (1 << 2) + +#define ICL_PORT_CL_DW12(phy) _MMIO(_ICL_PORT_CL_DW(12, phy)) +#define ICL_LANE_ENABLE_AUX (1 << 0) + +/* ICL Port COMP_DW registers */ +#define _ICL_PORT_COMP 0x100 +#define _ICL_PORT_COMP_DW(dw, phy) (_ICL_COMBOPHY(phy) + \ + _ICL_PORT_COMP + 4 * (dw)) + +#define ICL_PORT_COMP_DW0(phy) _MMIO(_ICL_PORT_COMP_DW(0, phy)) +#define COMP_INIT (1 << 31) + +#define ICL_PORT_COMP_DW1(phy) _MMIO(_ICL_PORT_COMP_DW(1, phy)) + +#define ICL_PORT_COMP_DW3(phy) _MMIO(_ICL_PORT_COMP_DW(3, phy)) +#define PROCESS_INFO_DOT_0 (0 << 26) +#define PROCESS_INFO_DOT_1 (1 << 26) +#define PROCESS_INFO_DOT_4 (2 << 26) +#define PROCESS_INFO_MASK (7 << 26) +#define PROCESS_INFO_SHIFT 26 +#define VOLTAGE_INFO_0_85V (0 << 24) +#define VOLTAGE_INFO_0_95V (1 << 24) +#define VOLTAGE_INFO_1_05V (2 << 24) +#define VOLTAGE_INFO_MASK (3 << 24) +#define VOLTAGE_INFO_SHIFT 24 + +#define ICL_PORT_COMP_DW8(phy) _MMIO(_ICL_PORT_COMP_DW(8, phy)) +#define IREFGEN (1 << 24) + +#define ICL_PORT_COMP_DW9(phy) _MMIO(_ICL_PORT_COMP_DW(9, phy)) + +#define ICL_PORT_COMP_DW10(phy) _MMIO(_ICL_PORT_COMP_DW(10, phy)) + +/* ICL Port PCS registers */ +#define _ICL_PORT_PCS_AUX 0x300 +#define _ICL_PORT_PCS_GRP 0x600 +#define _ICL_PORT_PCS_LN(ln) (0x800 + (ln) * 0x100) +#define _ICL_PORT_PCS_DW_AUX(dw, phy) (_ICL_COMBOPHY(phy) + \ + _ICL_PORT_PCS_AUX + 4 * (dw)) +#define _ICL_PORT_PCS_DW_GRP(dw, phy) (_ICL_COMBOPHY(phy) + \ + _ICL_PORT_PCS_GRP + 4 * (dw)) +#define _ICL_PORT_PCS_DW_LN(dw, ln, phy) (_ICL_COMBOPHY(phy) + \ + _ICL_PORT_PCS_LN(ln) + 4 * (dw)) +#define ICL_PORT_PCS_DW1_AUX(phy) _MMIO(_ICL_PORT_PCS_DW_AUX(1, phy)) +#define ICL_PORT_PCS_DW1_GRP(phy) _MMIO(_ICL_PORT_PCS_DW_GRP(1, phy)) +#define ICL_PORT_PCS_DW1_LN(ln, phy) _MMIO(_ICL_PORT_PCS_DW_LN(1, ln, phy)) +#define DCC_MODE_SELECT_MASK (0x3 << 20) +#define DCC_MODE_SELECT_CONTINUOSLY (0x3 << 20) +#define COMMON_KEEPER_EN (1 << 26) +#define LATENCY_OPTIM_MASK (0x3 << 2) +#define LATENCY_OPTIM_VAL(x) ((x) << 2) + +/* ICL Port TX registers */ +#define _ICL_PORT_TX_AUX 0x380 +#define _ICL_PORT_TX_GRP 0x680 +#define _ICL_PORT_TX_LN(ln) (0x880 + (ln) * 0x100) + +#define _ICL_PORT_TX_DW_AUX(dw, phy) (_ICL_COMBOPHY(phy) + \ + _ICL_PORT_TX_AUX + 4 * (dw)) +#define _ICL_PORT_TX_DW_GRP(dw, phy) (_ICL_COMBOPHY(phy) + \ + _ICL_PORT_TX_GRP + 4 * (dw)) +#define _ICL_PORT_TX_DW_LN(dw, ln, phy) (_ICL_COMBOPHY(phy) + \ + _ICL_PORT_TX_LN(ln) + 4 * (dw)) + +#define ICL_PORT_TX_DW2_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(2, phy)) +#define ICL_PORT_TX_DW2_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(2, phy)) +#define ICL_PORT_TX_DW2_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(2, ln, phy)) +#define SWING_SEL_UPPER(x) (((x) >> 3) << 15) +#define SWING_SEL_UPPER_MASK (1 << 15) +#define SWING_SEL_LOWER(x) (((x) & 0x7) << 11) +#define SWING_SEL_LOWER_MASK (0x7 << 11) +#define FRC_LATENCY_OPTIM_MASK (0x7 << 8) +#define FRC_LATENCY_OPTIM_VAL(x) ((x) << 8) +#define RCOMP_SCALAR(x) ((x) << 0) +#define RCOMP_SCALAR_MASK (0xFF << 0) + +#define ICL_PORT_TX_DW4_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(4, phy)) +#define ICL_PORT_TX_DW4_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(4, phy)) +#define ICL_PORT_TX_DW4_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(4, ln, phy)) +#define LOADGEN_SELECT (1 << 31) +#define POST_CURSOR_1(x) ((x) << 12) +#define POST_CURSOR_1_MASK (0x3F << 12) +#define POST_CURSOR_2(x) ((x) << 6) +#define POST_CURSOR_2_MASK (0x3F << 6) +#define CURSOR_COEFF(x) ((x) << 0) +#define CURSOR_COEFF_MASK (0x3F << 0) + +#define ICL_PORT_TX_DW5_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(5, phy)) +#define ICL_PORT_TX_DW5_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(5, phy)) +#define ICL_PORT_TX_DW5_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(5, ln, phy)) +#define TX_TRAINING_EN (1 << 31) +#define TAP2_DISABLE (1 << 30) +#define TAP3_DISABLE (1 << 29) +#define SCALING_MODE_SEL(x) ((x) << 18) +#define SCALING_MODE_SEL_MASK (0x7 << 18) +#define RTERM_SELECT(x) ((x) << 3) +#define RTERM_SELECT_MASK (0x7 << 3) + +#define ICL_PORT_TX_DW7_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(7, phy)) +#define ICL_PORT_TX_DW7_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(7, phy)) +#define ICL_PORT_TX_DW7_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(7, ln, phy)) +#define N_SCALAR(x) ((x) << 24) +#define N_SCALAR_MASK (0x7F << 24) + +#define ICL_PORT_TX_DW8_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(8, phy)) +#define ICL_PORT_TX_DW8_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(8, phy)) +#define ICL_PORT_TX_DW8_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(8, ln, phy)) +#define ICL_PORT_TX_DW8_ODCC_CLK_SEL REG_BIT(31) +#define ICL_PORT_TX_DW8_ODCC_CLK_DIV_SEL_MASK REG_GENMASK(30, 29) +#define ICL_PORT_TX_DW8_ODCC_CLK_DIV_SEL_DIV2 REG_FIELD_PREP(ICL_PORT_TX_DW8_ODCC_CLK_DIV_SEL_MASK, 0x1) + +#define _ICL_DPHY_CHKN_REG 0x194 +#define ICL_DPHY_CHKN(port) _MMIO(_ICL_COMBOPHY(port) + _ICL_DPHY_CHKN_REG) +#define ICL_DPHY_CHKN_AFE_OVER_PPI_STRAP REG_BIT(7) + +#endif /* __INTEL_COMBO_PHY_REGS__ */ diff --git a/drivers/gpu/drm/i915/display/intel_crt.h b/drivers/gpu/drm/i915/display/intel_crt.h index 6c5c44600cbd..c6071efd93ce 100644 --- a/drivers/gpu/drm/i915/display/intel_crt.h +++ b/drivers/gpu/drm/i915/display/intel_crt.h @@ -6,7 +6,7 @@ #ifndef __INTEL_CRT_H__ #define __INTEL_CRT_H__ -#include "i915_reg.h" +#include "i915_reg_defs.h" enum pipe; struct drm_encoder; diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c index 16c3ca66d9f0..08ee3e17ee5c 100644 --- a/drivers/gpu/drm/i915/display/intel_crtc.c +++ b/drivers/gpu/drm/i915/display/intel_crtc.c @@ -12,6 +12,7 @@ #include <drm/drm_plane_helper.h> #include <drm/drm_vblank_work.h> +#include "i915_irq.h" #include "i915_vgpu.h" #include "i9xx_plane.h" #include "icl_dsi.h" diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c index 16d34685d83f..2ade8fdd9bdd 100644 --- a/drivers/gpu/drm/i915/display/intel_cursor.c +++ b/drivers/gpu/drm/i915/display/intel_cursor.c @@ -51,16 +51,16 @@ static u32 intel_cursor_position(const struct intel_plane_state *plane_state) u32 pos = 0; if (x < 0) { - pos |= CURSOR_POS_SIGN << CURSOR_X_SHIFT; + pos |= CURSOR_POS_X_SIGN; x = -x; } - pos |= x << CURSOR_X_SHIFT; + pos |= CURSOR_POS_X(x); if (y < 0) { - pos |= CURSOR_POS_SIGN << CURSOR_Y_SHIFT; + pos |= CURSOR_POS_Y_SIGN; y = -y; } - pos |= y << CURSOR_Y_SHIFT; + pos |= CURSOR_POS_Y(y); return pos; } @@ -180,7 +180,7 @@ static u32 i845_cursor_ctl_crtc(const struct intel_crtc_state *crtc_state) u32 cntl = 0; if (crtc_state->gamma_enable) - cntl |= CURSOR_GAMMA_ENABLE; + cntl |= CURSOR_PIPE_GAMMA_ENABLE; return cntl; } @@ -264,7 +264,7 @@ static void i845_cursor_update_arm(struct intel_plane *plane, cntl = plane_state->ctl | i845_cursor_ctl_crtc(crtc_state); - size = (height << 12) | width; + size = CURSOR_HEIGHT(height) | CURSOR_WIDTH(width); base = intel_cursor_base(plane_state); pos = intel_cursor_position(plane_state); @@ -280,7 +280,7 @@ static void i845_cursor_update_arm(struct intel_plane *plane, plane->cursor.cntl != cntl) { intel_de_write_fw(dev_priv, CURCNTR(PIPE_A), 0); intel_de_write_fw(dev_priv, CURBASE(PIPE_A), base); - intel_de_write_fw(dev_priv, CURSIZE, size); + intel_de_write_fw(dev_priv, CURSIZE(PIPE_A), size); intel_de_write_fw(dev_priv, CURPOS(PIPE_A), pos); intel_de_write_fw(dev_priv, CURCNTR(PIPE_A), cntl); @@ -340,13 +340,13 @@ static u32 i9xx_cursor_ctl_crtc(const struct intel_crtc_state *crtc_state) return cntl; if (crtc_state->gamma_enable) - cntl = MCURSOR_GAMMA_ENABLE; + cntl = MCURSOR_PIPE_GAMMA_ENABLE; if (crtc_state->csc_enable) cntl |= MCURSOR_PIPE_CSC_ENABLE; if (DISPLAY_VER(dev_priv) < 5 && !IS_G4X(dev_priv)) - cntl |= MCURSOR_PIPE_SELECT(crtc->pipe); + cntl |= MCURSOR_PIPE_SEL(crtc->pipe); return cntl; } @@ -502,7 +502,7 @@ static void i9xx_cursor_update_arm(struct intel_plane *plane, i9xx_cursor_ctl_crtc(crtc_state); if (width != height) - fbc_ctl = CUR_FBC_CTL_EN | (height - 1); + fbc_ctl = CUR_FBC_EN | CUR_FBC_HEIGHT(height - 1); base = intel_cursor_base(plane_state); pos = intel_cursor_position(plane_state); @@ -586,13 +586,12 @@ static bool i9xx_cursor_get_hw_state(struct intel_plane *plane, val = intel_de_read(dev_priv, CURCNTR(plane->pipe)); - ret = val & MCURSOR_MODE; + ret = val & MCURSOR_MODE_MASK; if (DISPLAY_VER(dev_priv) >= 5 || IS_G4X(dev_priv)) *pipe = plane->pipe; else - *pipe = (val & MCURSOR_PIPE_SELECT_MASK) >> - MCURSOR_PIPE_SELECT_SHIFT; + *pipe = REG_FIELD_GET(MCURSOR_PIPE_SEL_MASK, val); intel_display_power_put(dev_priv, power_domain, wakeref); diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index cab505277595..354b08d6f81d 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -32,6 +32,7 @@ #include "intel_audio.h" #include "intel_backlight.h" #include "intel_combo_phy.h" +#include "intel_combo_phy_regs.h" #include "intel_connector.h" #include "intel_crtc.h" #include "intel_ddi.h" @@ -56,6 +57,7 @@ #include "intel_snps_phy.h" #include "intel_sprite.h" #include "intel_tc.h" +#include "intel_tc_phy_regs.h" #include "intel_vdsc.h" #include "intel_vrr.h" #include "skl_scaler.h" @@ -2287,116 +2289,6 @@ static void intel_ddi_mso_configure(const struct intel_crtc_state *crtc_state) OVERLAP_PIXELS_MASK, dss1); } -static void dg2_ddi_pre_enable_dp(struct intel_atomic_state *state, - struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state) -{ - struct intel_dp *intel_dp = enc_to_intel_dp(encoder); - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - struct intel_digital_port *dig_port = enc_to_dig_port(encoder); - bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST); - - intel_dp_set_link_params(intel_dp, crtc_state->port_clock, - crtc_state->lane_count); - - /* - * We only configure what the register value will be here. Actual - * enabling happens during link training farther down. - */ - intel_ddi_init_dp_buf_reg(encoder, crtc_state); - - /* - * 1. Enable Power Wells - * - * This was handled at the beginning of intel_atomic_commit_tail(), - * before we called down into this function. - */ - - /* 2. Enable Panel Power if PPS is required */ - intel_pps_on(intel_dp); - - /* - * 3. Enable the port PLL. - */ - intel_ddi_enable_clock(encoder, crtc_state); - - /* 4. Enable IO power */ - if (!intel_tc_port_in_tbt_alt_mode(dig_port)) - dig_port->ddi_io_wakeref = intel_display_power_get(dev_priv, - dig_port->ddi_io_power_domain); - - /* - * 5. The rest of the below are substeps under the bspec's "Enable and - * Train Display Port" step. Note that steps that are specific to - * MST will be handled by intel_mst_pre_enable_dp() before/after it - * calls into this function. Also intel_mst_pre_enable_dp() only calls - * us when active_mst_links==0, so any steps designated for "single - * stream or multi-stream master transcoder" can just be performed - * unconditionally here. - */ - - /* - * 5.a Configure Transcoder Clock Select to direct the Port clock to the - * Transcoder. - */ - intel_ddi_enable_pipe_clock(encoder, crtc_state); - - /* 5.b Configure transcoder for DP 2.0 128b/132b */ - intel_ddi_config_transcoder_dp2(encoder, crtc_state); - - /* - * 5.c Configure TRANS_DDI_FUNC_CTL DDI Select, DDI Mode Select & MST - * Transport Select - */ - intel_ddi_config_transcoder_func(encoder, crtc_state); - - /* - * 5.d Configure & enable DP_TP_CTL with link training pattern 1 - * selected - * - * This will be handled by the intel_dp_start_link_train() farther - * down this function. - */ - - /* 5.e Configure voltage swing and related IO settings */ - encoder->set_signal_levels(encoder, crtc_state); - - if (!is_mst) - intel_dp_set_power(intel_dp, DP_SET_POWER_D0); - - intel_dp_configure_protocol_converter(intel_dp, crtc_state); - intel_dp_sink_set_decompression_state(intel_dp, crtc_state, true); - /* - * DDI FEC: "anticipates enabling FEC encoding sets the FEC_READY bit - * in the FEC_CONFIGURATION register to 1 before initiating link - * training - */ - intel_dp_sink_set_fec_ready(intel_dp, crtc_state); - intel_dp_check_frl_training(intel_dp); - intel_dp_pcon_dsc_configure(intel_dp, crtc_state); - - /* - * 5.h Follow DisplayPort specification training sequence (see notes for - * failure handling) - * 5.i If DisplayPort multi-stream - Set DP_TP_CTL link training to Idle - * Pattern, wait for 5 idle patterns (DP_TP_STATUS Min_Idles_Sent) - * (timeout after 800 us) - */ - intel_dp_start_link_train(intel_dp, crtc_state); - - /* 5.j Set DP_TP_CTL link training to Normal */ - if (!is_trans_port_sync_mode(crtc_state)) - intel_dp_stop_link_train(intel_dp, crtc_state); - - /* 5.k Configure and enable FEC if needed */ - intel_ddi_enable_fec(encoder, crtc_state); - - intel_dsc_dp_pps_write(encoder, crtc_state); - - intel_dsc_enable(crtc_state); -} - static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state, struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, @@ -2470,6 +2362,9 @@ static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state, */ intel_ddi_enable_pipe_clock(encoder, crtc_state); + if (HAS_DP20(dev_priv)) + intel_ddi_config_transcoder_dp2(encoder, crtc_state); + /* * 7.b Configure TRANS_DDI_FUNC_CTL DDI Select, DDI Mode Select & MST * Transport Select @@ -2530,9 +2425,6 @@ static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state, intel_ddi_enable_fec(encoder, crtc_state); intel_dsc_dp_pps_write(encoder, crtc_state); - - if (!crtc_state->bigjoiner) - intel_dsc_enable(crtc_state); } static void hsw_ddi_pre_enable_dp(struct intel_atomic_state *state, @@ -2598,9 +2490,6 @@ static void hsw_ddi_pre_enable_dp(struct intel_atomic_state *state, intel_ddi_enable_pipe_clock(encoder, crtc_state); intel_dsc_dp_pps_write(encoder, crtc_state); - - if (!crtc_state->bigjoiner) - intel_dsc_enable(crtc_state); } static void intel_ddi_pre_enable_dp(struct intel_atomic_state *state, @@ -2610,9 +2499,7 @@ static void intel_ddi_pre_enable_dp(struct intel_atomic_state *state, { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - if (IS_DG2(dev_priv)) - dg2_ddi_pre_enable_dp(state, encoder, crtc_state, conn_state); - else if (DISPLAY_VER(dev_priv) >= 12) + if (DISPLAY_VER(dev_priv) >= 12) tgl_ddi_pre_enable_dp(state, encoder, crtc_state, conn_state); else hsw_ddi_pre_enable_dp(state, encoder, crtc_state, conn_state); @@ -2620,11 +2507,8 @@ static void intel_ddi_pre_enable_dp(struct intel_atomic_state *state, /* MST will call a setting of MSA after an allocating of Virtual Channel * from MST encoder pre_enable callback. */ - if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST)) { + if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST)) intel_ddi_set_dp_msa(crtc_state, conn_state); - - intel_dp_set_m_n(crtc_state, M1_N1); - } } static void intel_ddi_pre_enable_hdmi(struct intel_atomic_state *state, @@ -3471,7 +3355,11 @@ static void intel_ddi_read_func_ctl(struct intel_encoder *encoder, pipe_config->output_types |= BIT(INTEL_OUTPUT_DP); pipe_config->lane_count = ((temp & DDI_PORT_WIDTH_MASK) >> DDI_PORT_WIDTH_SHIFT) + 1; - intel_dp_get_m_n(crtc, pipe_config); + + intel_cpu_transcoder_get_m1_n1(crtc, cpu_transcoder, + &pipe_config->dp_m_n); + intel_cpu_transcoder_get_m2_n2(crtc, cpu_transcoder, + &pipe_config->dp_m2_n2); if (DISPLAY_VER(dev_priv) >= 11) { i915_reg_t dp_tp_ctl = dp_tp_ctl_reg(encoder, pipe_config); @@ -3508,7 +3396,8 @@ static void intel_ddi_read_func_ctl(struct intel_encoder *encoder, pipe_config->mst_master_transcoder = REG_FIELD_GET(TRANS_DDI_MST_TRANSPORT_SELECT_MASK, temp); - intel_dp_get_m_n(crtc, pipe_config); + intel_cpu_transcoder_get_m1_n1(crtc, cpu_transcoder, + &pipe_config->dp_m_n); pipe_config->infoframes.enable |= intel_hdmi_infoframes_enabled(encoder, pipe_config); @@ -3797,8 +3686,8 @@ static bool m_n_equal(const struct intel_link_m_n *m_n_1, const struct intel_link_m_n *m_n_2) { return m_n_1->tu == m_n_2->tu && - m_n_1->gmch_m == m_n_2->gmch_m && - m_n_1->gmch_n == m_n_2->gmch_n && + m_n_1->data_m == m_n_2->data_m && + m_n_1->data_n == m_n_2->data_n && m_n_1->link_m == m_n_2->link_m && m_n_1->link_n == m_n_2->link_n; } diff --git a/drivers/gpu/drm/i915/display/intel_ddi.h b/drivers/gpu/drm/i915/display/intel_ddi.h index c2fea6562917..d39076facdce 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.h +++ b/drivers/gpu/drm/i915/display/intel_ddi.h @@ -6,7 +6,7 @@ #ifndef __INTEL_DDI_H__ #define __INTEL_DDI_H__ -#include "i915_reg.h" +#include "i915_reg_defs.h" struct drm_connector_state; struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c index e2dfb93a82bd..0c32210bf503 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c +++ b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c @@ -985,15 +985,15 @@ static const struct intel_ddi_buf_trans adlp_dkl_phy_trans_dp_hbr2_hbr3 = { }; static const union intel_ddi_buf_trans_entry _dg2_snps_trans[] = { - { .snps = { 26, 0, 0 } }, /* VS 0, pre-emph 0 */ - { .snps = { 33, 0, 6 } }, /* VS 0, pre-emph 1 */ - { .snps = { 38, 0, 12 } }, /* VS 0, pre-emph 2 */ - { .snps = { 43, 0, 19 } }, /* VS 0, pre-emph 3 */ - { .snps = { 39, 0, 0 } }, /* VS 1, pre-emph 0 */ - { .snps = { 44, 0, 8 } }, /* VS 1, pre-emph 1 */ - { .snps = { 47, 0, 15 } }, /* VS 1, pre-emph 2 */ - { .snps = { 52, 0, 0 } }, /* VS 2, pre-emph 0 */ - { .snps = { 51, 0, 10 } }, /* VS 2, pre-emph 1 */ + { .snps = { 25, 0, 0 } }, /* VS 0, pre-emph 0 */ + { .snps = { 32, 0, 6 } }, /* VS 0, pre-emph 1 */ + { .snps = { 35, 0, 10 } }, /* VS 0, pre-emph 2 */ + { .snps = { 43, 0, 17 } }, /* VS 0, pre-emph 3 */ + { .snps = { 35, 0, 0 } }, /* VS 1, pre-emph 0 */ + { .snps = { 45, 0, 8 } }, /* VS 1, pre-emph 1 */ + { .snps = { 48, 0, 14 } }, /* VS 1, pre-emph 2 */ + { .snps = { 47, 0, 0 } }, /* VS 2, pre-emph 0 */ + { .snps = { 55, 0, 7 } }, /* VS 2, pre-emph 1 */ { .snps = { 62, 0, 0 } }, /* VS 3, pre-emph 0 */ }; @@ -1005,21 +1005,21 @@ static const struct intel_ddi_buf_trans dg2_snps_trans = { static const union intel_ddi_buf_trans_entry _dg2_snps_trans_uhbr[] = { { .snps = { 62, 0, 0 } }, /* preset 0 */ - { .snps = { 56, 0, 6 } }, /* preset 1 */ - { .snps = { 51, 0, 11 } }, /* preset 2 */ - { .snps = { 48, 0, 14 } }, /* preset 3 */ - { .snps = { 43, 0, 19 } }, /* preset 4 */ + { .snps = { 55, 0, 7 } }, /* preset 1 */ + { .snps = { 50, 0, 12 } }, /* preset 2 */ + { .snps = { 44, 0, 18 } }, /* preset 3 */ + { .snps = { 35, 0, 21 } }, /* preset 4 */ { .snps = { 59, 3, 0 } }, /* preset 5 */ { .snps = { 53, 3, 6 } }, /* preset 6 */ - { .snps = { 49, 3, 10 } }, /* preset 7 */ - { .snps = { 45, 3, 14 } }, /* preset 8 */ - { .snps = { 42, 3, 17 } }, /* preset 9 */ + { .snps = { 48, 3, 11 } }, /* preset 7 */ + { .snps = { 42, 5, 15 } }, /* preset 8 */ + { .snps = { 37, 5, 20 } }, /* preset 9 */ { .snps = { 56, 6, 0 } }, /* preset 10 */ - { .snps = { 50, 6, 6 } }, /* preset 11 */ - { .snps = { 47, 6, 9 } }, /* preset 12 */ - { .snps = { 42, 6, 14 } }, /* preset 13 */ - { .snps = { 46, 8, 8 } }, /* preset 14 */ - { .snps = { 56, 3, 3 } }, /* preset 15 */ + { .snps = { 48, 7, 7 } }, /* preset 11 */ + { .snps = { 45, 7, 10 } }, /* preset 12 */ + { .snps = { 39, 8, 15 } }, /* preset 13 */ + { .snps = { 48, 14, 0 } }, /* preset 14 */ + { .snps = { 45, 4, 4 } }, /* preset 15 */ }; static const struct intel_ddi_buf_trans dg2_snps_trans_uhbr = { diff --git a/drivers/gpu/drm/i915/display/intel_de.h b/drivers/gpu/drm/i915/display/intel_de.h index 9d8c177aa228..9c104f65e4c8 100644 --- a/drivers/gpu/drm/i915/display/intel_de.h +++ b/drivers/gpu/drm/i915/display/intel_de.h @@ -7,7 +7,6 @@ #define __INTEL_DE_H__ #include "i915_drv.h" -#include "i915_reg.h" #include "i915_trace.h" #include "intel_uncore.h" diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index bb4a85445fc6..7f512f9e9e5c 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -38,7 +38,7 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_atomic_uapi.h> #include <drm/drm_damage_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_fourcc.h> #include <drm/drm_plane_helper.h> @@ -118,9 +118,6 @@ static void intel_set_transcoder_timings(const struct intel_crtc_state *crtc_state); static void intel_set_pipe_src_size(const struct intel_crtc_state *crtc_state); -static void intel_cpu_transcoder_set_m_n(const struct intel_crtc_state *crtc_state, - const struct intel_link_m_n *m_n, - const struct intel_link_m_n *m2_n2); static void i9xx_set_pipeconf(const struct intel_crtc_state *crtc_state); static void ilk_set_pipeconf(const struct intel_crtc_state *crtc_state); static void hsw_set_transconf(const struct intel_crtc_state *crtc_state); @@ -353,16 +350,10 @@ static bool pipe_scanline_is_moving(struct drm_i915_private *dev_priv, { i915_reg_t reg = PIPEDSL(pipe); u32 line1, line2; - u32 line_mask; - if (DISPLAY_VER(dev_priv) == 2) - line_mask = DSL_LINEMASK_GEN2; - else - line_mask = DSL_LINEMASK_GEN3; - - line1 = intel_de_read(dev_priv, reg) & line_mask; + line1 = intel_de_read(dev_priv, reg) & PIPEDSL_LINE_MASK; msleep(5); - line2 = intel_de_read(dev_priv, reg) & line_mask; + line2 = intel_de_read(dev_priv, reg) & PIPEDSL_LINE_MASK; return line1 != line2; } @@ -397,13 +388,11 @@ intel_wait_for_pipe_off(const struct intel_crtc_state *old_crtc_state) if (DISPLAY_VER(dev_priv) >= 4) { enum transcoder cpu_transcoder = old_crtc_state->cpu_transcoder; - i915_reg_t reg = PIPECONF(cpu_transcoder); /* Wait for the Pipe State to go off */ - if (intel_de_wait_for_clear(dev_priv, reg, - I965_PIPECONF_ACTIVE, 100)) - drm_WARN(&dev_priv->drm, 1, - "pipe_off wait timed out\n"); + if (intel_de_wait_for_clear(dev_priv, PIPECONF(cpu_transcoder), + PIPECONF_STATE_ENABLE, 100)) + drm_WARN(&dev_priv->drm, 1, "pipe_off wait timed out\n"); } else { intel_wait_for_pipe_scanline_stopped(crtc); } @@ -1118,8 +1107,8 @@ void hsw_enable_ips(const struct intel_crtc_state *crtc_state) drm_WARN_ON(dev, !(crtc_state->active_planes & ~BIT(PLANE_CURSOR))); if (IS_BROADWELL(dev_priv)) { - drm_WARN_ON(dev, sandybridge_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, - IPS_ENABLE | IPS_PCODE_CONTROL)); + drm_WARN_ON(dev, snb_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, + IPS_ENABLE | IPS_PCODE_CONTROL)); /* Quoting Art Runyan: "its not safe to expect any particular * value in IPS_CTL bit 31 after enabling IPS through the * mailbox." Moreover, the mailbox may return a bogus state, @@ -1149,7 +1138,7 @@ void hsw_disable_ips(const struct intel_crtc_state *crtc_state) if (IS_BROADWELL(dev_priv)) { drm_WARN_ON(dev, - sandybridge_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, 0)); + snb_pcode_write(dev_priv, DISPLAY_IPS_CONTROL, 0)); /* * Wait for PCODE to finish disabling IPS. The BSpec specified * 42ms timeout value leads to occasional timeouts so use 100ms @@ -1369,7 +1358,8 @@ static void intel_crtc_enable_flip_done(struct intel_atomic_state *state, for_each_new_intel_plane_in_state(state, plane, plane_state, i) { if (plane->enable_flip_done && plane->pipe == crtc->pipe && - update_planes & BIT(plane->id)) + update_planes & BIT(plane->id) && + plane_state->do_async_flip) plane->enable_flip_done(plane); } } @@ -1387,7 +1377,8 @@ static void intel_crtc_disable_flip_done(struct intel_atomic_state *state, for_each_new_intel_plane_in_state(state, plane, plane_state, i) { if (plane->disable_flip_done && plane->pipe == crtc->pipe && - update_planes & BIT(plane->id)) + update_planes & BIT(plane->id) && + plane_state->do_async_flip) plane->disable_flip_done(plane); } } @@ -1817,6 +1808,26 @@ static void intel_disable_primary_plane(const struct intel_crtc_state *crtc_stat plane->disable_arm(plane, crtc_state); } +static void ilk_configure_cpu_transcoder(const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; + + if (crtc_state->has_pch_encoder) { + intel_cpu_transcoder_set_m1_n1(crtc, cpu_transcoder, + &crtc_state->fdi_m_n); + } else if (intel_crtc_has_dp_encoder(crtc_state)) { + intel_cpu_transcoder_set_m1_n1(crtc, cpu_transcoder, + &crtc_state->dp_m_n); + intel_cpu_transcoder_set_m2_n2(crtc, cpu_transcoder, + &crtc_state->dp_m2_n2); + } + + intel_set_transcoder_timings(crtc_state); + + ilk_set_pipeconf(crtc_state); +} + static void ilk_crtc_enable(struct intel_atomic_state *state, struct intel_crtc *crtc) { @@ -1841,27 +1852,16 @@ static void ilk_crtc_enable(struct intel_atomic_state *state, intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false); intel_set_pch_fifo_underrun_reporting(dev_priv, pipe, false); - if (intel_crtc_has_dp_encoder(new_crtc_state)) - intel_dp_set_m_n(new_crtc_state, M1_N1); + ilk_configure_cpu_transcoder(new_crtc_state); - intel_set_transcoder_timings(new_crtc_state); intel_set_pipe_src_size(new_crtc_state); - if (new_crtc_state->has_pch_encoder) - intel_cpu_transcoder_set_m_n(new_crtc_state, - &new_crtc_state->fdi_m_n, NULL); - - ilk_set_pipeconf(new_crtc_state); - crtc->active = true; intel_encoders_pre_enable(state, crtc); if (new_crtc_state->has_pch_encoder) { - /* Note: FDI PLL enabling _must_ be done before we enable the - * cpu pipes, hence this is separate from all the other fdi/pch - * enabling. */ - ilk_fdi_pll_enable(new_crtc_state); + ilk_pch_pre_enable(state, crtc); } else { assert_fdi_tx_disabled(dev_priv, pipe); assert_fdi_rx_disabled(dev_priv, pipe); @@ -1974,7 +1974,6 @@ static void hsw_set_frame_start_delay(const struct intel_crtc_state *crtc_state) static void icl_ddi_bigjoiner_pre_enable(struct intel_atomic_state *state, const struct intel_crtc_state *crtc_state) { - struct drm_i915_private *dev_priv = to_i915(state->base.dev); struct intel_crtc_state *master_crtc_state; struct intel_crtc *master_crtc; struct drm_connector_state *conn_state; @@ -2004,12 +2003,33 @@ static void icl_ddi_bigjoiner_pre_enable(struct intel_atomic_state *state, if (crtc_state->bigjoiner_slave) intel_encoders_pre_enable(state, master_crtc); +} - /* need to enable VDSC, which we skipped in pre-enable */ - intel_dsc_enable(crtc_state); +static void hsw_configure_cpu_transcoder(const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; - if (DISPLAY_VER(dev_priv) >= 13) - intel_uncompressed_joiner_enable(crtc_state); + if (crtc_state->has_pch_encoder) { + intel_cpu_transcoder_set_m1_n1(crtc, cpu_transcoder, + &crtc_state->fdi_m_n); + } else if (intel_crtc_has_dp_encoder(crtc_state)) { + intel_cpu_transcoder_set_m1_n1(crtc, cpu_transcoder, + &crtc_state->dp_m_n); + intel_cpu_transcoder_set_m2_n2(crtc, cpu_transcoder, + &crtc_state->dp_m2_n2); + } + + intel_set_transcoder_timings(crtc_state); + + if (cpu_transcoder != TRANSCODER_EDP) + intel_de_write(dev_priv, PIPE_MULT(cpu_transcoder), + crtc_state->pixel_multiplier - 1); + + hsw_set_frame_start_delay(crtc_state); + + hsw_set_transconf(crtc_state); } static void hsw_crtc_enable(struct intel_atomic_state *state, @@ -2036,25 +2056,17 @@ static void hsw_crtc_enable(struct intel_atomic_state *state, icl_ddi_bigjoiner_pre_enable(state, new_crtc_state); } + intel_dsc_enable(new_crtc_state); + + if (DISPLAY_VER(dev_priv) >= 13) + intel_uncompressed_joiner_enable(new_crtc_state); + intel_set_pipe_src_size(new_crtc_state); if (DISPLAY_VER(dev_priv) >= 9 || IS_BROADWELL(dev_priv)) bdw_set_pipemisc(new_crtc_state); - if (!new_crtc_state->bigjoiner_slave && !transcoder_is_dsi(cpu_transcoder)) { - intel_set_transcoder_timings(new_crtc_state); - - if (cpu_transcoder != TRANSCODER_EDP) - intel_de_write(dev_priv, PIPE_MULT(cpu_transcoder), - new_crtc_state->pixel_multiplier - 1); - - if (new_crtc_state->has_pch_encoder) - intel_cpu_transcoder_set_m_n(new_crtc_state, - &new_crtc_state->fdi_m_n, NULL); - - hsw_set_frame_start_delay(new_crtc_state); - - hsw_set_transconf(new_crtc_state); - } + if (!new_crtc_state->bigjoiner_slave && !transcoder_is_dsi(cpu_transcoder)) + hsw_configure_cpu_transcoder(new_crtc_state); crtc->active = true; @@ -2441,6 +2453,23 @@ static void modeset_put_crtc_power_domains(struct intel_crtc *crtc, domains); } +static void i9xx_configure_cpu_transcoder(const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; + + if (intel_crtc_has_dp_encoder(crtc_state)) { + intel_cpu_transcoder_set_m1_n1(crtc, cpu_transcoder, + &crtc_state->dp_m_n); + intel_cpu_transcoder_set_m2_n2(crtc, cpu_transcoder, + &crtc_state->dp_m2_n2); + } + + intel_set_transcoder_timings(crtc_state); + + i9xx_set_pipeconf(crtc_state); +} + static void valleyview_crtc_enable(struct intel_atomic_state *state, struct intel_crtc *crtc) { @@ -2452,10 +2481,8 @@ static void valleyview_crtc_enable(struct intel_atomic_state *state, if (drm_WARN_ON(&dev_priv->drm, crtc->active)) return; - if (intel_crtc_has_dp_encoder(new_crtc_state)) - intel_dp_set_m_n(new_crtc_state, M1_N1); + i9xx_configure_cpu_transcoder(new_crtc_state); - intel_set_transcoder_timings(new_crtc_state); intel_set_pipe_src_size(new_crtc_state); if (IS_CHERRYVIEW(dev_priv) && pipe == PIPE_B) { @@ -2463,8 +2490,6 @@ static void valleyview_crtc_enable(struct intel_atomic_state *state, intel_de_write(dev_priv, CHV_CANVAS(pipe), 0); } - i9xx_set_pipeconf(new_crtc_state); - crtc->active = true; intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true); @@ -2504,14 +2529,10 @@ static void i9xx_crtc_enable(struct intel_atomic_state *state, if (drm_WARN_ON(&dev_priv->drm, crtc->active)) return; - if (intel_crtc_has_dp_encoder(new_crtc_state)) - intel_dp_set_m_n(new_crtc_state, M1_N1); + i9xx_configure_cpu_transcoder(new_crtc_state); - intel_set_transcoder_timings(new_crtc_state); intel_set_pipe_src_size(new_crtc_state); - i9xx_set_pipeconf(new_crtc_state); - crtc->active = true; if (DISPLAY_VER(dev_priv) != 2) @@ -3085,7 +3106,7 @@ intel_link_compute_m_n(u16 bits_per_pixel, int nlanes, m_n->tu = 64; compute_m_n(data_clock, link_clock * nlanes * 8, - &m_n->gmch_m, &m_n->gmch_n, + &m_n->data_m, &m_n->data_n, constant_n); compute_m_n(pixel_clock, link_clock, @@ -3116,99 +3137,66 @@ static void intel_panel_sanitize_ssc(struct drm_i915_private *dev_priv) } } -static void intel_pch_transcoder_set_m_n(const struct intel_crtc_state *crtc_state, - const struct intel_link_m_n *m_n) +void intel_zero_m_n(struct intel_link_m_n *m_n) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - enum pipe pipe = crtc->pipe; + /* corresponds to 0 register value */ + memset(m_n, 0, sizeof(*m_n)); + m_n->tu = 1; +} - intel_de_write(dev_priv, PCH_TRANS_DATA_M1(pipe), - TU_SIZE(m_n->tu) | m_n->gmch_m); - intel_de_write(dev_priv, PCH_TRANS_DATA_N1(pipe), m_n->gmch_n); - intel_de_write(dev_priv, PCH_TRANS_LINK_M1(pipe), m_n->link_m); - intel_de_write(dev_priv, PCH_TRANS_LINK_N1(pipe), m_n->link_n); +void intel_set_m_n(struct drm_i915_private *i915, + const struct intel_link_m_n *m_n, + i915_reg_t data_m_reg, i915_reg_t data_n_reg, + i915_reg_t link_m_reg, i915_reg_t link_n_reg) +{ + intel_de_write(i915, data_m_reg, TU_SIZE(m_n->tu) | m_n->data_m); + intel_de_write(i915, data_n_reg, m_n->data_n); + intel_de_write(i915, link_m_reg, m_n->link_m); + /* + * On BDW+ writing LINK_N arms the double buffered update + * of all the M/N registers, so it must be written last. + */ + intel_de_write(i915, link_n_reg, m_n->link_n); } -static bool transcoder_has_m2_n2(struct drm_i915_private *dev_priv, - enum transcoder transcoder) +bool intel_cpu_transcoder_has_m2_n2(struct drm_i915_private *dev_priv, + enum transcoder transcoder) { if (IS_HASWELL(dev_priv)) return transcoder == TRANSCODER_EDP; - /* - * Strictly speaking some registers are available before - * gen7, but we only support DRRS on gen7+ - */ - return DISPLAY_VER(dev_priv) == 7 || IS_CHERRYVIEW(dev_priv); + return IS_DISPLAY_VER(dev_priv, 5, 7) || IS_CHERRYVIEW(dev_priv); } -static void intel_cpu_transcoder_set_m_n(const struct intel_crtc_state *crtc_state, - const struct intel_link_m_n *m_n, - const struct intel_link_m_n *m2_n2) +void intel_cpu_transcoder_set_m1_n1(struct intel_crtc *crtc, + enum transcoder transcoder, + const struct intel_link_m_n *m_n) { - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); enum pipe pipe = crtc->pipe; - enum transcoder transcoder = crtc_state->cpu_transcoder; - if (DISPLAY_VER(dev_priv) >= 5) { - intel_de_write(dev_priv, PIPE_DATA_M1(transcoder), - TU_SIZE(m_n->tu) | m_n->gmch_m); - intel_de_write(dev_priv, PIPE_DATA_N1(transcoder), - m_n->gmch_n); - intel_de_write(dev_priv, PIPE_LINK_M1(transcoder), - m_n->link_m); - intel_de_write(dev_priv, PIPE_LINK_N1(transcoder), - m_n->link_n); - /* - * M2_N2 registers are set only if DRRS is supported - * (to make sure the registers are not unnecessarily accessed). - */ - if (m2_n2 && crtc_state->has_drrs && - transcoder_has_m2_n2(dev_priv, transcoder)) { - intel_de_write(dev_priv, PIPE_DATA_M2(transcoder), - TU_SIZE(m2_n2->tu) | m2_n2->gmch_m); - intel_de_write(dev_priv, PIPE_DATA_N2(transcoder), - m2_n2->gmch_n); - intel_de_write(dev_priv, PIPE_LINK_M2(transcoder), - m2_n2->link_m); - intel_de_write(dev_priv, PIPE_LINK_N2(transcoder), - m2_n2->link_n); - } - } else { - intel_de_write(dev_priv, PIPE_DATA_M_G4X(pipe), - TU_SIZE(m_n->tu) | m_n->gmch_m); - intel_de_write(dev_priv, PIPE_DATA_N_G4X(pipe), m_n->gmch_n); - intel_de_write(dev_priv, PIPE_LINK_M_G4X(pipe), m_n->link_m); - intel_de_write(dev_priv, PIPE_LINK_N_G4X(pipe), m_n->link_n); - } + if (DISPLAY_VER(dev_priv) >= 5) + intel_set_m_n(dev_priv, m_n, + PIPE_DATA_M1(transcoder), PIPE_DATA_N1(transcoder), + PIPE_LINK_M1(transcoder), PIPE_LINK_N1(transcoder)); + else + intel_set_m_n(dev_priv, m_n, + PIPE_DATA_M_G4X(pipe), PIPE_DATA_N_G4X(pipe), + PIPE_LINK_M_G4X(pipe), PIPE_LINK_N_G4X(pipe)); } -void intel_dp_set_m_n(const struct intel_crtc_state *crtc_state, enum link_m_n_set m_n) +void intel_cpu_transcoder_set_m2_n2(struct intel_crtc *crtc, + enum transcoder transcoder, + const struct intel_link_m_n *m_n) { - const struct intel_link_m_n *dp_m_n, *dp_m2_n2 = NULL; - struct drm_i915_private *i915 = to_i915(crtc_state->uapi.crtc->dev); - - if (m_n == M1_N1) { - dp_m_n = &crtc_state->dp_m_n; - dp_m2_n2 = &crtc_state->dp_m2_n2; - } else if (m_n == M2_N2) { + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - /* - * M2_N2 registers are not supported. Hence m2_n2 divider value - * needs to be programmed into M1_N1. - */ - dp_m_n = &crtc_state->dp_m2_n2; - } else { - drm_err(&i915->drm, "Unsupported divider value\n"); + if (!intel_cpu_transcoder_has_m2_n2(dev_priv, transcoder)) return; - } - if (crtc_state->has_pch_encoder) - intel_pch_transcoder_set_m_n(crtc_state, &crtc_state->dp_m_n); - else - intel_cpu_transcoder_set_m_n(crtc_state, dp_m_n, dp_m2_n2); + intel_set_m_n(dev_priv, m_n, + PIPE_DATA_M2(transcoder), PIPE_DATA_N2(transcoder), + PIPE_LINK_M2(transcoder), PIPE_LINK_N2(transcoder)); } static void intel_set_transcoder_timings(const struct intel_crtc_state *crtc_state) @@ -3279,7 +3267,8 @@ static void intel_set_pipe_src_size(const struct intel_crtc_state *crtc_state) * always be the user's requested size. */ intel_de_write(dev_priv, PIPESRC(pipe), - ((crtc_state->pipe_src_w - 1) << 16) | (crtc_state->pipe_src_h - 1)); + PIPESRC_WIDTH(crtc_state->pipe_src_w - 1) | + PIPESRC_HEIGHT(crtc_state->pipe_src_h - 1)); } static bool intel_pipe_is_interlaced(const struct intel_crtc_state *crtc_state) @@ -3350,8 +3339,8 @@ static void intel_get_pipe_src_size(struct intel_crtc *crtc, u32 tmp; tmp = intel_de_read(dev_priv, PIPESRC(crtc->pipe)); - pipe_config->pipe_src_h = (tmp & 0xffff) + 1; - pipe_config->pipe_src_w = ((tmp >> 16) & 0xffff) + 1; + pipe_config->pipe_src_w = REG_FIELD_GET(PIPESRC_WIDTH_MASK, tmp) + 1; + pipe_config->pipe_src_h = REG_FIELD_GET(PIPESRC_HEIGHT_MASK, tmp) + 1; } static void i9xx_set_pipeconf(const struct intel_crtc_state *crtc_state) @@ -3379,13 +3368,13 @@ static void i9xx_set_pipeconf(const struct intel_crtc_state *crtc_state) switch (crtc_state->pipe_bpp) { case 18: - pipeconf |= PIPECONF_6BPC; + pipeconf |= PIPECONF_BPC_6; break; case 24: - pipeconf |= PIPECONF_8BPC; + pipeconf |= PIPECONF_BPC_8; break; case 30: - pipeconf |= PIPECONF_10BPC; + pipeconf |= PIPECONF_BPC_10; break; default: /* Case prevented by intel_choose_pipe_bpp_dither. */ @@ -3400,7 +3389,7 @@ static void i9xx_set_pipeconf(const struct intel_crtc_state *crtc_state) else pipeconf |= PIPECONF_INTERLACE_W_SYNC_SHIFT; } else { - pipeconf |= PIPECONF_PROGRESSIVE; + pipeconf |= PIPECONF_INTERLACE_PROGRESSIVE; } if ((IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) && @@ -3543,11 +3532,11 @@ static void i9xx_get_pipe_color_config(struct intel_crtc_state *crtc_state) tmp = intel_de_read(dev_priv, DSPCNTR(i9xx_plane)); - if (tmp & DISPPLANE_GAMMA_ENABLE) + if (tmp & DISP_PIPE_GAMMA_ENABLE) crtc_state->gamma_enable = true; if (!HAS_GMCH(dev_priv) && - tmp & DISPPLANE_PIPE_CSC_ENABLE) + tmp & DISP_PIPE_CSC_ENABLE) crtc_state->csc_enable = true; } @@ -3578,16 +3567,17 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc, if (IS_G4X(dev_priv) || IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) { switch (tmp & PIPECONF_BPC_MASK) { - case PIPECONF_6BPC: + case PIPECONF_BPC_6: pipe_config->pipe_bpp = 18; break; - case PIPECONF_8BPC: + case PIPECONF_BPC_8: pipe_config->pipe_bpp = 24; break; - case PIPECONF_10BPC: + case PIPECONF_BPC_10: pipe_config->pipe_bpp = 30; break; default: + MISSING_CASE(tmp); break; } } @@ -3596,8 +3586,7 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc, (tmp & PIPECONF_COLOR_RANGE_SELECT)) pipe_config->limited_color_range = true; - pipe_config->gamma_mode = (tmp & PIPECONF_GAMMA_MODE_MASK_I9XX) >> - PIPECONF_GAMMA_MODE_SHIFT; + pipe_config->gamma_mode = REG_FIELD_GET(PIPECONF_GAMMA_MODE_MASK_I9XX, tmp); if (IS_CHERRYVIEW(dev_priv)) pipe_config->cgm_mode = intel_de_read(dev_priv, @@ -3684,16 +3673,16 @@ static void ilk_set_pipeconf(const struct intel_crtc_state *crtc_state) switch (crtc_state->pipe_bpp) { case 18: - val |= PIPECONF_6BPC; + val |= PIPECONF_BPC_6; break; case 24: - val |= PIPECONF_8BPC; + val |= PIPECONF_BPC_8; break; case 30: - val |= PIPECONF_10BPC; + val |= PIPECONF_BPC_10; break; case 36: - val |= PIPECONF_12BPC; + val |= PIPECONF_BPC_12; break; default: /* Case prevented by intel_choose_pipe_bpp_dither. */ @@ -3701,12 +3690,12 @@ static void ilk_set_pipeconf(const struct intel_crtc_state *crtc_state) } if (crtc_state->dither) - val |= (PIPECONF_DITHER_EN | PIPECONF_DITHER_TYPE_SP); + val |= PIPECONF_DITHER_EN | PIPECONF_DITHER_TYPE_SP; if (crtc_state->hw.adjusted_mode.flags & DRM_MODE_FLAG_INTERLACE) - val |= PIPECONF_INTERLACED_ILK; + val |= PIPECONF_INTERLACE_IF_ID_ILK; else - val |= PIPECONF_PROGRESSIVE; + val |= PIPECONF_INTERLACE_PF_PD_ILK; /* * This would end up with an odd purple hue over @@ -3738,12 +3727,12 @@ static void hsw_set_transconf(const struct intel_crtc_state *crtc_state) u32 val = 0; if (IS_HASWELL(dev_priv) && crtc_state->dither) - val |= (PIPECONF_DITHER_EN | PIPECONF_DITHER_TYPE_SP); + val |= PIPECONF_DITHER_EN | PIPECONF_DITHER_TYPE_SP; if (crtc_state->hw.adjusted_mode.flags & DRM_MODE_FLAG_INTERLACE) - val |= PIPECONF_INTERLACED_ILK; + val |= PIPECONF_INTERLACE_IF_ID_ILK; else - val |= PIPECONF_PROGRESSIVE; + val |= PIPECONF_INTERLACE_PF_PD_ILK; if (IS_HASWELL(dev_priv) && crtc_state->output_format != INTEL_OUTPUT_FORMAT_RGB) @@ -3765,18 +3754,18 @@ static void bdw_set_pipemisc(const struct intel_crtc_state *crtc_state) switch (crtc_state->pipe_bpp) { case 18: - val |= PIPEMISC_6_BPC; + val |= PIPEMISC_BPC_6; break; case 24: - val |= PIPEMISC_8_BPC; + val |= PIPEMISC_BPC_8; break; case 30: - val |= PIPEMISC_10_BPC; + val |= PIPEMISC_BPC_10; break; case 36: /* Port output 12BPC defined for ADLP+ */ if (DISPLAY_VER(dev_priv) > 12) - val |= PIPEMISC_12_BPC_ADLP; + val |= PIPEMISC_BPC_12_ADLP; break; default: MISSING_CASE(crtc_state->pipe_bpp); @@ -3812,7 +3801,7 @@ static void bdw_set_pipemisc(const struct intel_crtc_state *crtc_state) } intel_de_rmw(dev_priv, PIPE_MISC2(crtc->pipe), - PIPE_MISC2_UNDERRUN_BUBBLE_COUNTER_MASK, + PIPE_MISC2_BUBBLE_COUNTER_MASK, scaler_in_use ? PIPE_MISC2_BUBBLE_COUNTER_SCALER_EN : PIPE_MISC2_BUBBLE_COUNTER_SCALER_DIS); } @@ -3828,11 +3817,11 @@ int bdw_get_pipemisc_bpp(struct intel_crtc *crtc) tmp = intel_de_read(dev_priv, PIPEMISC(crtc->pipe)); switch (tmp & PIPEMISC_BPC_MASK) { - case PIPEMISC_6_BPC: + case PIPEMISC_BPC_6: return 18; - case PIPEMISC_8_BPC: + case PIPEMISC_BPC_8: return 24; - case PIPEMISC_10_BPC: + case PIPEMISC_BPC_10: return 30; /* * PORT OUTPUT 12 BPC defined for ADLP+. @@ -3844,7 +3833,7 @@ int bdw_get_pipemisc_bpp(struct intel_crtc *crtc) * on older platforms, need to find a workaround for 12 BPC * MIPI DSI HW readout. */ - case PIPEMISC_12_BPC_ADLP: + case PIPEMISC_BPC_12_ADLP: if (DISPLAY_VER(dev_priv) > 12) return 36; fallthrough; @@ -3865,83 +3854,47 @@ int ilk_get_lanes_required(int target_clock, int link_bw, int bpp) return DIV_ROUND_UP(bps, link_bw * 8); } -static void intel_pch_transcoder_get_m_n(struct intel_crtc *crtc, - struct intel_link_m_n *m_n) +void intel_get_m_n(struct drm_i915_private *i915, + struct intel_link_m_n *m_n, + i915_reg_t data_m_reg, i915_reg_t data_n_reg, + i915_reg_t link_m_reg, i915_reg_t link_n_reg) { - struct drm_device *dev = crtc->base.dev; - struct drm_i915_private *dev_priv = to_i915(dev); - enum pipe pipe = crtc->pipe; - - m_n->link_m = intel_de_read(dev_priv, PCH_TRANS_LINK_M1(pipe)); - m_n->link_n = intel_de_read(dev_priv, PCH_TRANS_LINK_N1(pipe)); - m_n->gmch_m = intel_de_read(dev_priv, PCH_TRANS_DATA_M1(pipe)) - & ~TU_SIZE_MASK; - m_n->gmch_n = intel_de_read(dev_priv, PCH_TRANS_DATA_N1(pipe)); - m_n->tu = ((intel_de_read(dev_priv, PCH_TRANS_DATA_M1(pipe)) - & TU_SIZE_MASK) >> TU_SIZE_SHIFT) + 1; + m_n->link_m = intel_de_read(i915, link_m_reg) & DATA_LINK_M_N_MASK; + m_n->link_n = intel_de_read(i915, link_n_reg) & DATA_LINK_M_N_MASK; + m_n->data_m = intel_de_read(i915, data_m_reg) & DATA_LINK_M_N_MASK; + m_n->data_n = intel_de_read(i915, data_n_reg) & DATA_LINK_M_N_MASK; + m_n->tu = REG_FIELD_GET(TU_SIZE_MASK, intel_de_read(i915, data_m_reg)) + 1; } -static void intel_cpu_transcoder_get_m_n(struct intel_crtc *crtc, - enum transcoder transcoder, - struct intel_link_m_n *m_n, - struct intel_link_m_n *m2_n2) +void intel_cpu_transcoder_get_m1_n1(struct intel_crtc *crtc, + enum transcoder transcoder, + struct intel_link_m_n *m_n) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); enum pipe pipe = crtc->pipe; - if (DISPLAY_VER(dev_priv) >= 5) { - m_n->link_m = intel_de_read(dev_priv, - PIPE_LINK_M1(transcoder)); - m_n->link_n = intel_de_read(dev_priv, - PIPE_LINK_N1(transcoder)); - m_n->gmch_m = intel_de_read(dev_priv, - PIPE_DATA_M1(transcoder)) - & ~TU_SIZE_MASK; - m_n->gmch_n = intel_de_read(dev_priv, - PIPE_DATA_N1(transcoder)); - m_n->tu = ((intel_de_read(dev_priv, PIPE_DATA_M1(transcoder)) - & TU_SIZE_MASK) >> TU_SIZE_SHIFT) + 1; - - if (m2_n2 && transcoder_has_m2_n2(dev_priv, transcoder)) { - m2_n2->link_m = intel_de_read(dev_priv, - PIPE_LINK_M2(transcoder)); - m2_n2->link_n = intel_de_read(dev_priv, - PIPE_LINK_N2(transcoder)); - m2_n2->gmch_m = intel_de_read(dev_priv, - PIPE_DATA_M2(transcoder)) - & ~TU_SIZE_MASK; - m2_n2->gmch_n = intel_de_read(dev_priv, - PIPE_DATA_N2(transcoder)); - m2_n2->tu = ((intel_de_read(dev_priv, PIPE_DATA_M2(transcoder)) - & TU_SIZE_MASK) >> TU_SIZE_SHIFT) + 1; - } - } else { - m_n->link_m = intel_de_read(dev_priv, PIPE_LINK_M_G4X(pipe)); - m_n->link_n = intel_de_read(dev_priv, PIPE_LINK_N_G4X(pipe)); - m_n->gmch_m = intel_de_read(dev_priv, PIPE_DATA_M_G4X(pipe)) - & ~TU_SIZE_MASK; - m_n->gmch_n = intel_de_read(dev_priv, PIPE_DATA_N_G4X(pipe)); - m_n->tu = ((intel_de_read(dev_priv, PIPE_DATA_M_G4X(pipe)) - & TU_SIZE_MASK) >> TU_SIZE_SHIFT) + 1; - } -} - -void intel_dp_get_m_n(struct intel_crtc *crtc, - struct intel_crtc_state *pipe_config) -{ - if (pipe_config->has_pch_encoder) - intel_pch_transcoder_get_m_n(crtc, &pipe_config->dp_m_n); + if (DISPLAY_VER(dev_priv) >= 5) + intel_get_m_n(dev_priv, m_n, + PIPE_DATA_M1(transcoder), PIPE_DATA_N1(transcoder), + PIPE_LINK_M1(transcoder), PIPE_LINK_N1(transcoder)); else - intel_cpu_transcoder_get_m_n(crtc, pipe_config->cpu_transcoder, - &pipe_config->dp_m_n, - &pipe_config->dp_m2_n2); + intel_get_m_n(dev_priv, m_n, + PIPE_DATA_M_G4X(pipe), PIPE_DATA_N_G4X(pipe), + PIPE_LINK_M_G4X(pipe), PIPE_LINK_N_G4X(pipe)); } -void ilk_get_fdi_m_n_config(struct intel_crtc *crtc, - struct intel_crtc_state *pipe_config) +void intel_cpu_transcoder_get_m2_n2(struct intel_crtc *crtc, + enum transcoder transcoder, + struct intel_link_m_n *m_n) { - intel_cpu_transcoder_get_m_n(crtc, pipe_config->cpu_transcoder, - &pipe_config->fdi_m_n, NULL); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + + if (!intel_cpu_transcoder_has_m2_n2(dev_priv, transcoder)) + return; + + intel_get_m_n(dev_priv, m_n, + PIPE_DATA_M2(transcoder), PIPE_DATA_N2(transcoder), + PIPE_LINK_M2(transcoder), PIPE_LINK_N2(transcoder)); } static void ilk_get_pfit_pos_size(struct intel_crtc_state *crtc_state, @@ -4037,16 +3990,16 @@ static bool ilk_get_pipe_config(struct intel_crtc *crtc, goto out; switch (tmp & PIPECONF_BPC_MASK) { - case PIPECONF_6BPC: + case PIPECONF_BPC_6: pipe_config->pipe_bpp = 18; break; - case PIPECONF_8BPC: + case PIPECONF_BPC_8: pipe_config->pipe_bpp = 24; break; - case PIPECONF_10BPC: + case PIPECONF_BPC_10: pipe_config->pipe_bpp = 30; break; - case PIPECONF_12BPC: + case PIPECONF_BPC_12: pipe_config->pipe_bpp = 36; break; default: @@ -4066,8 +4019,7 @@ static bool ilk_get_pipe_config(struct intel_crtc *crtc, break; } - pipe_config->gamma_mode = (tmp & PIPECONF_GAMMA_MODE_MASK_ILK) >> - PIPECONF_GAMMA_MODE_SHIFT; + pipe_config->gamma_mode = REG_FIELD_GET(PIPECONF_GAMMA_MODE_MASK_ILK, tmp); pipe_config->csc_mode = intel_de_read(dev_priv, PIPE_CSC_MODE(crtc->pipe)); @@ -4380,13 +4332,13 @@ static bool hsw_get_pipe_config(struct intel_crtc *crtc, active = true; } + if (!active) + goto out; + intel_dsc_get_config(pipe_config); if (DISPLAY_VER(dev_priv) >= 13 && !pipe_config->dsc.compression_enable) intel_uncompressed_joiner_get_config(pipe_config); - if (!active) - goto out; - if (!transcoder_is_dsi(pipe_config->cpu_transcoder) || DISPLAY_VER(dev_priv) >= 11) intel_get_transcoder_timings(crtc, pipe_config); @@ -4908,6 +4860,28 @@ static bool needs_scaling(const struct intel_plane_state *state) return (src_w != dst_w || src_h != dst_h); } +static bool intel_plane_do_async_flip(struct intel_plane *plane, + const struct intel_crtc_state *old_crtc_state, + const struct intel_crtc_state *new_crtc_state) +{ + struct drm_i915_private *i915 = to_i915(plane->base.dev); + + if (!plane->async_flip) + return false; + + if (!new_crtc_state->uapi.async_flip) + return false; + + /* + * In platforms after DISPLAY13, we might need to override + * first async flip in order to change watermark levels + * as part of optimization. + * So for those, we are checking if this is a first async flip. + * For platforms earlier than DISPLAY13 we always do async flip. + */ + return DISPLAY_VER(i915) < 13 || old_crtc_state->uapi.async_flip; +} + int intel_plane_atomic_calc_changes(const struct intel_crtc_state *old_crtc_state, struct intel_crtc_state *new_crtc_state, const struct intel_plane_state *old_plane_state, @@ -5027,6 +5001,9 @@ int intel_plane_atomic_calc_changes(const struct intel_crtc_state *old_crtc_stat needs_scaling(new_plane_state)))) new_crtc_state->disable_lp_wm = true; + if (intel_plane_do_async_flip(plane, old_crtc_state, new_crtc_state)) + new_plane_state->do_async_flip = true; + return 0; } @@ -5491,9 +5468,9 @@ intel_dump_m_n_config(const struct intel_crtc_state *pipe_config, struct drm_i915_private *i915 = to_i915(pipe_config->uapi.crtc->dev); drm_dbg_kms(&i915->drm, - "%s: lanes: %i; gmch_m: %u, gmch_n: %u, link_m: %u, link_n: %u, tu: %u\n", + "%s: lanes: %i; data_m: %u, data_n: %u, link_m: %u, link_n: %u, tu: %u\n", id, lane_count, - m_n->gmch_m, m_n->gmch_n, + m_n->data_m, m_n->data_n, m_n->link_m, m_n->link_n, m_n->tu); } @@ -5658,11 +5635,11 @@ static void intel_dump_pipe_config(const struct intel_crtc_state *pipe_config, if (intel_crtc_has_dp_encoder(pipe_config)) { intel_dump_m_n_config(pipe_config, "dp m_n", - pipe_config->lane_count, &pipe_config->dp_m_n); - if (pipe_config->has_drrs) - intel_dump_m_n_config(pipe_config, "dp m2_n2", - pipe_config->lane_count, - &pipe_config->dp_m2_n2); + pipe_config->lane_count, + &pipe_config->dp_m_n); + intel_dump_m_n_config(pipe_config, "dp m2_n2", + pipe_config->lane_count, + &pipe_config->dp_m2_n2); } drm_dbg_kms(&dev_priv->drm, @@ -6189,8 +6166,8 @@ intel_compare_link_m_n(const struct intel_link_m_n *m_n, bool exact) { return m_n->tu == m2_n2->tu && - intel_compare_m_n(m_n->gmch_m, m_n->gmch_n, - m2_n2->gmch_m, m2_n2->gmch_n, exact) && + intel_compare_m_n(m_n->data_m, m_n->data_n, + m2_n2->data_m, m2_n2->data_n, exact) && intel_compare_m_n(m_n->link_m, m_n->link_n, m2_n2->link_m, m2_n2->link_n, exact); } @@ -6389,16 +6366,16 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, &pipe_config->name,\ !fastset)) { \ pipe_config_mismatch(fastset, crtc, __stringify(name), \ - "(expected tu %i gmch %i/%i link %i/%i, " \ - "found tu %i, gmch %i/%i link %i/%i)", \ + "(expected tu %i data %i/%i link %i/%i, " \ + "found tu %i, data %i/%i link %i/%i)", \ current_config->name.tu, \ - current_config->name.gmch_m, \ - current_config->name.gmch_n, \ + current_config->name.data_m, \ + current_config->name.data_n, \ current_config->name.link_m, \ current_config->name.link_n, \ pipe_config->name.tu, \ - pipe_config->name.gmch_m, \ - pipe_config->name.gmch_n, \ + pipe_config->name.data_m, \ + pipe_config->name.data_n, \ pipe_config->name.link_m, \ pipe_config->name.link_n); \ ret = false; \ @@ -6416,22 +6393,22 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, !intel_compare_link_m_n(¤t_config->alt_name, \ &pipe_config->name, !fastset)) { \ pipe_config_mismatch(fastset, crtc, __stringify(name), \ - "(expected tu %i gmch %i/%i link %i/%i, " \ - "or tu %i gmch %i/%i link %i/%i, " \ - "found tu %i, gmch %i/%i link %i/%i)", \ + "(expected tu %i data %i/%i link %i/%i, " \ + "or tu %i data %i/%i link %i/%i, " \ + "found tu %i, data %i/%i link %i/%i)", \ current_config->name.tu, \ - current_config->name.gmch_m, \ - current_config->name.gmch_n, \ + current_config->name.data_m, \ + current_config->name.data_n, \ current_config->name.link_m, \ current_config->name.link_n, \ current_config->alt_name.tu, \ - current_config->alt_name.gmch_m, \ - current_config->alt_name.gmch_n, \ + current_config->alt_name.data_m, \ + current_config->alt_name.data_n, \ current_config->alt_name.link_m, \ current_config->alt_name.link_n, \ pipe_config->name.tu, \ - pipe_config->name.gmch_m, \ - pipe_config->name.gmch_n, \ + pipe_config->name.data_m, \ + pipe_config->name.data_n, \ pipe_config->name.link_m, \ pipe_config->name.link_n); \ ret = false; \ @@ -6510,13 +6487,12 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, PIPE_CONF_CHECK_I(lane_count); PIPE_CONF_CHECK_X(lane_lat_optim_mask); - if (DISPLAY_VER(dev_priv) < 8) { - PIPE_CONF_CHECK_M_N(dp_m_n); - - if (current_config->has_drrs) - PIPE_CONF_CHECK_M_N(dp_m2_n2); - } else + if (DISPLAY_VER(dev_priv) >= 9 || IS_BROADWELL(dev_priv)) { PIPE_CONF_CHECK_M_N_ALT(dp_m_n, dp_m2_n2); + } else { + PIPE_CONF_CHECK_M_N(dp_m_n); + PIPE_CONF_CHECK_M_N(dp_m2_n2); + } PIPE_CONF_CHECK_X(output_types); @@ -7600,6 +7576,7 @@ static int intel_atomic_check_bigjoiner(struct intel_atomic_state *state, struct intel_crtc_state *old_crtc_state, struct intel_crtc_state *new_crtc_state) { + struct drm_i915_private *i915 = to_i915(state->base.dev); struct intel_crtc_state *slave_crtc_state, *master_crtc_state; struct intel_crtc *slave_crtc, *master_crtc; @@ -7617,9 +7594,10 @@ static int intel_atomic_check_bigjoiner(struct intel_atomic_state *state, slave_crtc = intel_dsc_get_bigjoiner_secondary(crtc); if (!slave_crtc) { - DRM_DEBUG_KMS("[CRTC:%d:%s] Big joiner configuration requires " - "CRTC + 1 to be used, doesn't exist\n", - crtc->base.base.id, crtc->base.name); + drm_dbg_kms(&i915->drm, + "[CRTC:%d:%s] Big joiner configuration requires " + "CRTC + 1 to be used, doesn't exist\n", + crtc->base.base.id, crtc->base.name); return -EINVAL; } @@ -7633,16 +7611,18 @@ static int intel_atomic_check_bigjoiner(struct intel_atomic_state *state, if (slave_crtc_state->uapi.enable) goto claimed; - DRM_DEBUG_KMS("[CRTC:%d:%s] Used as slave for big joiner\n", - slave_crtc->base.base.id, slave_crtc->base.name); + drm_dbg_kms(&i915->drm, + "[CRTC:%d:%s] Used as slave for big joiner\n", + slave_crtc->base.base.id, slave_crtc->base.name); return copy_bigjoiner_crtc_state(slave_crtc_state, new_crtc_state); claimed: - DRM_DEBUG_KMS("[CRTC:%d:%s] Slave is enabled as normal CRTC, but " - "[CRTC:%d:%s] claiming this CRTC for bigjoiner.\n", - slave_crtc->base.base.id, slave_crtc->base.name, - master_crtc->base.base.id, master_crtc->base.name); + drm_dbg_kms(&i915->drm, + "[CRTC:%d:%s] Slave is enabled as normal CRTC, but " + "[CRTC:%d:%s] claiming this CRTC for bigjoiner.\n", + slave_crtc->base.base.id, slave_crtc->base.name, + master_crtc->base.base.id, master_crtc->base.name); return -EINVAL; } @@ -9478,7 +9458,7 @@ void intel_modeset_init_hw(struct drm_i915_private *i915) cdclk_state = to_intel_cdclk_state(i915->cdclk.obj.state); intel_update_cdclk(i915); - intel_dump_cdclk_config(&i915->cdclk.hw, "Current CDCLK"); + intel_cdclk_dump_config(i915, &i915->cdclk.hw, "Current CDCLK"); cdclk_state->logical = cdclk_state->actual = i915->cdclk.hw; } @@ -9980,8 +9960,7 @@ void i830_enable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe) udelay(150); /* wait for warmup */ } - intel_de_write(dev_priv, PIPECONF(pipe), - PIPECONF_ENABLE | PIPECONF_PROGRESSIVE); + intel_de_write(dev_priv, PIPECONF(pipe), PIPECONF_ENABLE); intel_de_posting_read(dev_priv, PIPECONF(pipe)); intel_wait_for_pipe_scanline_moving(crtc); @@ -9995,18 +9974,15 @@ void i830_disable_pipe(struct drm_i915_private *dev_priv, enum pipe pipe) pipe_name(pipe)); drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, DSPCNTR(PLANE_A)) & - DISPLAY_PLANE_ENABLE); + intel_de_read(dev_priv, DSPCNTR(PLANE_A)) & DISP_ENABLE); drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, DSPCNTR(PLANE_B)) & - DISPLAY_PLANE_ENABLE); + intel_de_read(dev_priv, DSPCNTR(PLANE_B)) & DISP_ENABLE); drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, DSPCNTR(PLANE_C)) & - DISPLAY_PLANE_ENABLE); + intel_de_read(dev_priv, DSPCNTR(PLANE_C)) & DISP_ENABLE); drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, CURCNTR(PIPE_A)) & MCURSOR_MODE); + intel_de_read(dev_priv, CURCNTR(PIPE_A)) & MCURSOR_MODE_MASK); drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, CURCNTR(PIPE_B)) & MCURSOR_MODE); + intel_de_read(dev_priv, CURCNTR(PIPE_B)) & MCURSOR_MODE_MASK); intel_de_write(dev_priv, PIPECONF(pipe), 0); intel_de_posting_read(dev_priv, PIPECONF(pipe)); @@ -10689,6 +10665,8 @@ intel_modeset_setup_hw_state(struct drm_device *dev, } intel_display_power_put(dev_priv, POWER_DOMAIN_INIT, wakeref); + + intel_power_domains_sanitize_state(dev_priv); } void intel_display_resume(struct drm_device *dev) diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h index b61b75248ded..457738aeee3e 100644 --- a/drivers/gpu/drm/i915/display/intel_display.h +++ b/drivers/gpu/drm/i915/display/intel_display.h @@ -27,7 +27,8 @@ #include <drm/drm_util.h> -enum link_m_n_set; +#include "i915_reg_defs.h" + enum drm_scaling_filter; struct dpll; struct drm_connector; @@ -317,8 +318,8 @@ enum aux_ch { /* Used by dp and fdi links */ struct intel_link_m_n { u32 tu; - u32 gmch_m; - u32 gmch_n; + u32 data_m; + u32 data_n; u32 link_m; u32 link_n; }; @@ -605,12 +606,29 @@ bool intel_fuzzy_clock_check(int clock1, int clock2); void intel_display_prepare_reset(struct drm_i915_private *dev_priv); void intel_display_finish_reset(struct drm_i915_private *dev_priv); -void intel_dp_get_m_n(struct intel_crtc *crtc, - struct intel_crtc_state *pipe_config); -void intel_dp_set_m_n(const struct intel_crtc_state *crtc_state, - enum link_m_n_set m_n); -void ilk_get_fdi_m_n_config(struct intel_crtc *crtc, - struct intel_crtc_state *pipe_config); +void intel_zero_m_n(struct intel_link_m_n *m_n); +void intel_set_m_n(struct drm_i915_private *i915, + const struct intel_link_m_n *m_n, + i915_reg_t data_m_reg, i915_reg_t data_n_reg, + i915_reg_t link_m_reg, i915_reg_t link_n_reg); +void intel_get_m_n(struct drm_i915_private *i915, + struct intel_link_m_n *m_n, + i915_reg_t data_m_reg, i915_reg_t data_n_reg, + i915_reg_t link_m_reg, i915_reg_t link_n_reg); +bool intel_cpu_transcoder_has_m2_n2(struct drm_i915_private *dev_priv, + enum transcoder transcoder); +void intel_cpu_transcoder_set_m1_n1(struct intel_crtc *crtc, + enum transcoder cpu_transcoder, + const struct intel_link_m_n *m_n); +void intel_cpu_transcoder_set_m2_n2(struct intel_crtc *crtc, + enum transcoder cpu_transcoder, + const struct intel_link_m_n *m_n); +void intel_cpu_transcoder_get_m1_n1(struct intel_crtc *crtc, + enum transcoder cpu_transcoder, + struct intel_link_m_n *m_n); +void intel_cpu_transcoder_get_m2_n2(struct intel_crtc *crtc, + enum transcoder cpu_transcoder, + struct intel_link_m_n *m_n); void i9xx_crtc_clock_get(struct intel_crtc *crtc, struct intel_crtc_state *pipe_config); int intel_dotclock_calculate(int link_freq, const struct intel_link_m_n *m_n); diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c index 572445299b04..f4de004d470f 100644 --- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c @@ -2402,6 +2402,9 @@ void intel_connector_debugfs_add(struct intel_connector *intel_connector) */ void intel_crtc_debugfs_add(struct drm_crtc *crtc) { - if (crtc->debugfs_entry) - crtc_updates_add(crtc); + if (!crtc->debugfs_entry) + return; + + crtc_updates_add(crtc); + intel_fbc_crtc_debugfs_add(to_intel_crtc(crtc)); } diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c index 05babdcf5f2e..d2102cc17bb4 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.c +++ b/drivers/gpu/drm/i915/display/intel_display_power.c @@ -7,6 +7,7 @@ #include "i915_irq.h" #include "intel_cdclk.h" #include "intel_combo_phy.h" +#include "intel_combo_phy_regs.h" #include "intel_crt.h" #include "intel_de.h" #include "intel_display_power.h" @@ -682,9 +683,8 @@ static void icl_tc_cold_exit(struct drm_i915_private *i915) int ret, tries = 0; while (1) { - ret = sandybridge_pcode_write_timeout(i915, - ICL_PCODE_EXIT_TCCOLD, - 0, 250, 1); + ret = snb_pcode_write_timeout(i915, ICL_PCODE_EXIT_TCCOLD, 0, + 250, 1); if (ret != -EAGAIN || ++tries == 3) break; msleep(1); @@ -4052,8 +4052,7 @@ tgl_tc_cold_request(struct drm_i915_private *i915, bool block) * Spec states that we should timeout the request after 200us * but the function below will timeout after 500us */ - ret = sandybridge_pcode_read(i915, TGL_PCODE_TCCOLD, &low_val, - &high_val); + ret = snb_pcode_read(i915, TGL_PCODE_TCCOLD, &low_val, &high_val); if (ret == 0) { if (block && (low_val & TGL_PCODE_EXIT_TCCOLD_DATA_L_EXIT_FAILED)) @@ -5468,8 +5467,7 @@ static u32 hsw_read_dcomp(struct drm_i915_private *dev_priv) static void hsw_write_dcomp(struct drm_i915_private *dev_priv, u32 val) { if (IS_HASWELL(dev_priv)) { - if (sandybridge_pcode_write(dev_priv, - GEN6_PCODE_WRITE_D_COMP, val)) + if (snb_pcode_write(dev_priv, GEN6_PCODE_WRITE_D_COMP, val)) drm_dbg_kms(&dev_priv->drm, "Failed to write to D_COMP\n"); } else { @@ -5582,7 +5580,7 @@ static void hsw_restore_lcpll(struct drm_i915_private *dev_priv) intel_uncore_forcewake_put(&dev_priv->uncore, FORCEWAKE_ALL); intel_update_cdclk(dev_priv); - intel_dump_cdclk_config(&dev_priv->cdclk.hw, "Current CDCLK"); + intel_cdclk_dump_config(dev_priv, &dev_priv->cdclk.hw, "Current CDCLK"); } /* @@ -6216,6 +6214,37 @@ void intel_power_domains_driver_remove(struct drm_i915_private *i915) } /** + * intel_power_domains_sanitize_state - sanitize power domains state + * @i915: i915 device instance + * + * Sanitize the power domains state during driver loading and system resume. + * The function will disable all display power wells that BIOS has enabled + * without a user for it (any user for a power well has taken a reference + * on it by the time this function is called, after the state of all the + * pipe, encoder, etc. HW resources have been sanitized). + */ +void intel_power_domains_sanitize_state(struct drm_i915_private *i915) +{ + struct i915_power_domains *power_domains = &i915->power_domains; + struct i915_power_well *power_well; + + mutex_lock(&power_domains->lock); + + for_each_power_well_reverse(i915, power_well) { + if (power_well->desc->always_on || power_well->count || + !power_well->desc->ops->is_enabled(i915, power_well)) + continue; + + drm_dbg_kms(&i915->drm, + "BIOS left unused %s power well enabled, disabling it\n", + power_well->desc->name); + intel_power_well_disable(i915, power_well); + } + + mutex_unlock(&power_domains->lock); +} + +/** * intel_power_domains_enable - enable toggling of display power wells * @i915: i915 device instance * diff --git a/drivers/gpu/drm/i915/display/intel_display_power.h b/drivers/gpu/drm/i915/display/intel_display_power.h index 686d18eaa24c..f6d0e6e73c6d 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.h +++ b/drivers/gpu/drm/i915/display/intel_display_power.h @@ -7,7 +7,6 @@ #define __INTEL_DISPLAY_POWER_H__ #include "intel_runtime_pm.h" -#include "i915_reg.h" enum dpio_channel; enum dpio_phy; @@ -219,6 +218,7 @@ void intel_power_domains_disable(struct drm_i915_private *dev_priv); void intel_power_domains_suspend(struct drm_i915_private *dev_priv, enum i915_drm_suspend_mode); void intel_power_domains_resume(struct drm_i915_private *dev_priv); +void intel_power_domains_sanitize_state(struct drm_i915_private *dev_priv); void intel_display_power_suspend_late(struct drm_i915_private *i915); void intel_display_power_resume_early(struct drm_i915_private *i915); diff --git a/drivers/gpu/drm/i915/display/intel_display_trace.h b/drivers/gpu/drm/i915/display/intel_display_trace.h index 4043e1276383..f05f0f9b5103 100644 --- a/drivers/gpu/drm/i915/display/intel_display_trace.h +++ b/drivers/gpu/drm/i915/display/intel_display_trace.h @@ -13,6 +13,7 @@ #include <linux/tracepoint.h> #include "i915_drv.h" +#include "i915_irq.h" #include "intel_crtc.h" #include "intel_display_types.h" diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h index c9c6efadf8b4..60e15226a8cb 100644 --- a/drivers/gpu/drm/i915/display/intel_display_types.h +++ b/drivers/gpu/drm/i915/display/intel_display_types.h @@ -32,10 +32,10 @@ #include <linux/pwm.h> #include <linux/sched/clock.h> +#include <drm/dp/drm_dp_dual_mode_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_atomic.h> #include <drm/drm_crtc.h> -#include <drm/drm_dp_dual_mode_helper.h> -#include <drm/drm_dp_mst_helper.h> #include <drm/drm_dsc.h> #include <drm/drm_encoder.h> #include <drm/drm_fb_helper.h> @@ -634,6 +634,9 @@ struct intel_plane_state { struct intel_fb_view view; + /* Indicates if async flip is required */ + bool do_async_flip; + /* Plane pxp decryption state */ bool decrypt; @@ -1442,25 +1445,6 @@ struct intel_hdmi { }; struct intel_dp_mst_encoder; -/* - * enum link_m_n_set: - * When platform provides two set of M_N registers for dp, we can - * program them and switch between them incase of DRRS. - * But When only one such register is provided, we have to program the - * required divider value on that registers itself based on the DRRS state. - * - * M1_N1 : Program dp_m_n on M1_N1 registers - * dp_m2_n2 on M2_N2 registers (If supported) - * - * M2_N2 : Program dp_m2_n2 on M1_N1 registers - * M2_N2 registers are not supported - */ - -enum link_m_n_set { - /* Sets the m1_n1 and m2_n2 */ - M1_N1 = 0, - M2_N2 -}; struct intel_dp_compliance_data { unsigned long edid; diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c index a69b28d65a9b..7616a3906b9e 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc.c +++ b/drivers/gpu/drm/i915/display/intel_dmc.c @@ -43,9 +43,9 @@ __stringify(major) "_" \ __stringify(minor) ".bin" -#define GEN12_DMC_MAX_FW_SIZE ICL_DMC_MAX_FW_SIZE +#define DISPLAY_VER13_DMC_MAX_FW_SIZE 0x20000 -#define GEN13_DMC_MAX_FW_SIZE 0x20000 +#define DISPLAY_VER12_DMC_MAX_FW_SIZE ICL_DMC_MAX_FW_SIZE #define ADLP_DMC_PATH DMC_PATH(adlp, 2, 14) #define ADLP_DMC_VERSION_REQUIRED DMC_VERSION(2, 14) @@ -684,23 +684,23 @@ void intel_dmc_ucode_init(struct drm_i915_private *dev_priv) if (IS_ALDERLAKE_P(dev_priv)) { dmc->fw_path = ADLP_DMC_PATH; dmc->required_version = ADLP_DMC_VERSION_REQUIRED; - dmc->max_fw_size = GEN13_DMC_MAX_FW_SIZE; + dmc->max_fw_size = DISPLAY_VER13_DMC_MAX_FW_SIZE; } else if (IS_ALDERLAKE_S(dev_priv)) { dmc->fw_path = ADLS_DMC_PATH; dmc->required_version = ADLS_DMC_VERSION_REQUIRED; - dmc->max_fw_size = GEN12_DMC_MAX_FW_SIZE; + dmc->max_fw_size = DISPLAY_VER12_DMC_MAX_FW_SIZE; } else if (IS_DG1(dev_priv)) { dmc->fw_path = DG1_DMC_PATH; dmc->required_version = DG1_DMC_VERSION_REQUIRED; - dmc->max_fw_size = GEN12_DMC_MAX_FW_SIZE; + dmc->max_fw_size = DISPLAY_VER12_DMC_MAX_FW_SIZE; } else if (IS_ROCKETLAKE(dev_priv)) { dmc->fw_path = RKL_DMC_PATH; dmc->required_version = RKL_DMC_VERSION_REQUIRED; - dmc->max_fw_size = GEN12_DMC_MAX_FW_SIZE; + dmc->max_fw_size = DISPLAY_VER12_DMC_MAX_FW_SIZE; } else if (DISPLAY_VER(dev_priv) >= 12) { dmc->fw_path = TGL_DMC_PATH; dmc->required_version = TGL_DMC_VERSION_REQUIRED; - dmc->max_fw_size = GEN12_DMC_MAX_FW_SIZE; + dmc->max_fw_size = DISPLAY_VER12_DMC_MAX_FW_SIZE; } else if (DISPLAY_VER(dev_priv) == 11) { dmc->fw_path = ICL_DMC_PATH; dmc->required_version = ICL_DMC_VERSION_REQUIRED; diff --git a/drivers/gpu/drm/i915/display/intel_dmc.h b/drivers/gpu/drm/i915/display/intel_dmc.h index b20f3441ca60..7c590309a3a9 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc.h +++ b/drivers/gpu/drm/i915/display/intel_dmc.h @@ -6,7 +6,7 @@ #ifndef __INTEL_DMC_H__ #define __INTEL_DMC_H__ -#include "i915_reg.h" +#include "i915_reg_defs.h" #include "intel_wakeref.h" #include <linux/workqueue.h> diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index b5e2508db1cf..146b83916005 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -36,7 +36,7 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_probe_helper.h> @@ -46,6 +46,7 @@ #include "intel_atomic.h" #include "intel_audio.h" #include "intel_backlight.h" +#include "intel_combo_phy_regs.h" #include "intel_connector.h" #include "intel_crtc.h" #include "intel_ddi.h" @@ -72,8 +73,6 @@ #include "intel_vdsc.h" #include "intel_vrr.h" -#define DP_DPRX_ESI_LEN 14 - /* DP DSC throughput values used for slice count calculations KPixels/s */ #define DP_DSC_PEAK_PIXEL_RATE 2720000 #define DP_DSC_MAX_ENC_THROUGHPUT_0 340000 @@ -705,7 +704,7 @@ static u16 intel_dp_dsc_get_output_bpp(struct drm_i915_private *i915, i915->max_cdclk_freq * 48 / intel_dp_mode_to_fec_clock(mode_clock); - DRM_DEBUG_KMS("Max big joiner bpp: %u\n", max_bpp_bigjoiner); + drm_dbg_kms(&i915->drm, "Max big joiner bpp: %u\n", max_bpp_bigjoiner); bits_per_pixel = min(bits_per_pixel, max_bpp_bigjoiner); } @@ -1167,14 +1166,13 @@ static bool intel_dp_hdmi_tmds_clock_valid(struct intel_dp *intel_dp, return true; } -static bool intel_dp_hdmi_deep_color_possible(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state, - int bpc) +static bool intel_dp_hdmi_bpc_possible(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state, + int bpc) { - return intel_hdmi_deep_color_possible(crtc_state, bpc, - intel_dp->has_hdmi_sink, - intel_dp_hdmi_ycbcr420(intel_dp, crtc_state)) && + return intel_hdmi_bpc_possible(crtc_state, bpc, intel_dp->has_hdmi_sink, + intel_dp_hdmi_ycbcr420(intel_dp, crtc_state)) && intel_dp_hdmi_tmds_clock_valid(intel_dp, crtc_state, bpc); } @@ -1192,7 +1190,7 @@ static int intel_dp_max_bpp(struct intel_dp *intel_dp, if (intel_dp->dfp.min_tmds_clock) { for (; bpc >= 10; bpc -= 2) { - if (intel_dp_hdmi_deep_color_possible(intel_dp, crtc_state, bpc)) + if (intel_dp_hdmi_bpc_possible(intel_dp, crtc_state, bpc)) break; } } @@ -1897,7 +1895,7 @@ intel_dp_compute_config(struct intel_encoder *encoder, /* FIXME: abstract this better */ if (pipe_config->splitter.enable) - pipe_config->dp_m_n.gmch_m *= pipe_config->splitter.link_count; + pipe_config->dp_m_n.data_m *= pipe_config->splitter.link_count; if (!HAS_DDI(dev_priv)) g4x_dp_set_clock(encoder, pipe_config); @@ -2813,11 +2811,22 @@ intel_dp_configure_mst(struct intel_dp *intel_dp) } static bool -intel_dp_get_sink_irq_esi(struct intel_dp *intel_dp, u8 *sink_irq_vector) +intel_dp_get_sink_irq_esi(struct intel_dp *intel_dp, u8 *esi) { - return drm_dp_dpcd_read(&intel_dp->aux, DP_SINK_COUNT_ESI, - sink_irq_vector, DP_DPRX_ESI_LEN) == - DP_DPRX_ESI_LEN; + return drm_dp_dpcd_read(&intel_dp->aux, DP_SINK_COUNT_ESI, esi, 4) == 4; +} + +static bool intel_dp_ack_sink_irq_esi(struct intel_dp *intel_dp, u8 esi[4]) +{ + int retry; + + for (retry = 0; retry < 3; retry++) { + if (drm_dp_dpcd_write(&intel_dp->aux, DP_SINK_COUNT_ESI + 1, + &esi[1], 3) == 3) + return true; + } + + return false; } bool @@ -2909,7 +2918,8 @@ out: } static ssize_t -intel_dp_hdr_metadata_infoframe_sdp_pack(const struct hdmi_drm_infoframe *drm_infoframe, +intel_dp_hdr_metadata_infoframe_sdp_pack(struct drm_i915_private *i915, + const struct hdmi_drm_infoframe *drm_infoframe, struct dp_sdp *sdp, size_t size) { @@ -2925,12 +2935,12 @@ intel_dp_hdr_metadata_infoframe_sdp_pack(const struct hdmi_drm_infoframe *drm_in len = hdmi_drm_infoframe_pack_only(drm_infoframe, buf, sizeof(buf)); if (len < 0) { - DRM_DEBUG_KMS("buffer size is smaller than hdr metadata infoframe\n"); + drm_dbg_kms(&i915->drm, "buffer size is smaller than hdr metadata infoframe\n"); return -ENOSPC; } if (len != infoframe_size) { - DRM_DEBUG_KMS("wrong static hdr metadata size\n"); + drm_dbg_kms(&i915->drm, "wrong static hdr metadata size\n"); return -ENOSPC; } @@ -3003,7 +3013,8 @@ static void intel_write_dp_sdp(struct intel_encoder *encoder, sizeof(sdp)); break; case HDMI_PACKET_TYPE_GAMUT_METADATA: - len = intel_dp_hdr_metadata_infoframe_sdp_pack(&crtc_state->infoframes.drm.drm, + len = intel_dp_hdr_metadata_infoframe_sdp_pack(dev_priv, + &crtc_state->infoframes.drm.drm, &sdp, sizeof(sdp)); break; default: @@ -3411,22 +3422,22 @@ static void intel_dp_phy_pattern_update(struct intel_dp *intel_dp, switch (data->phy_pattern) { case DP_PHY_TEST_PATTERN_NONE: - DRM_DEBUG_KMS("Disable Phy Test Pattern\n"); + drm_dbg_kms(&dev_priv->drm, "Disable Phy Test Pattern\n"); intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), 0x0); break; case DP_PHY_TEST_PATTERN_D10_2: - DRM_DEBUG_KMS("Set D10.2 Phy Test Pattern\n"); + drm_dbg_kms(&dev_priv->drm, "Set D10.2 Phy Test Pattern\n"); intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), DDI_DP_COMP_CTL_ENABLE | DDI_DP_COMP_CTL_D10_2); break; case DP_PHY_TEST_PATTERN_ERROR_COUNT: - DRM_DEBUG_KMS("Set Error Count Phy Test Pattern\n"); + drm_dbg_kms(&dev_priv->drm, "Set Error Count Phy Test Pattern\n"); intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), DDI_DP_COMP_CTL_ENABLE | DDI_DP_COMP_CTL_SCRAMBLED_0); break; case DP_PHY_TEST_PATTERN_PRBS7: - DRM_DEBUG_KMS("Set PRBS7 Phy Test Pattern\n"); + drm_dbg_kms(&dev_priv->drm, "Set PRBS7 Phy Test Pattern\n"); intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), DDI_DP_COMP_CTL_ENABLE | DDI_DP_COMP_CTL_PRBS7); break; @@ -3436,7 +3447,8 @@ static void intel_dp_phy_pattern_update(struct intel_dp *intel_dp, * current firmware of DPR-100 could not set it, so hardcoding * now for complaince test. */ - DRM_DEBUG_KMS("Set 80Bit Custom Phy Test Pattern 0x3e0f83e0 0x0f83e0f8 0x0000f83e\n"); + drm_dbg_kms(&dev_priv->drm, + "Set 80Bit Custom Phy Test Pattern 0x3e0f83e0 0x0f83e0f8 0x0000f83e\n"); pattern_val = 0x3e0f83e0; intel_de_write(dev_priv, DDI_DP_COMP_PAT(pipe, 0), pattern_val); pattern_val = 0x0f83e0f8; @@ -3453,7 +3465,7 @@ static void intel_dp_phy_pattern_update(struct intel_dp *intel_dp, * current firmware of DPR-100 could not set it, so hardcoding * now for complaince test. */ - DRM_DEBUG_KMS("Set HBR2 compliance Phy Test Pattern\n"); + drm_dbg_kms(&dev_priv->drm, "Set HBR2 compliance Phy Test Pattern\n"); pattern_val = 0xFB; intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), DDI_DP_COMP_CTL_ENABLE | DDI_DP_COMP_CTL_HBR2 | @@ -3522,13 +3534,14 @@ intel_dp_autotest_phy_ddi_enable(struct intel_dp *intel_dp, static void intel_dp_process_phy_request(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { + struct drm_i915_private *i915 = dp_to_i915(intel_dp); struct drm_dp_phy_test_params *data = &intel_dp->compliance.test_data.phytest; u8 link_status[DP_LINK_STATUS_SIZE]; if (drm_dp_dpcd_read_phy_link_status(&intel_dp->aux, DP_PHY_DPRX, link_status) < 0) { - DRM_DEBUG_KMS("failed to get link status\n"); + drm_dbg_kms(&i915->drm, "failed to get link status\n"); return; } @@ -3553,11 +3566,12 @@ static void intel_dp_process_phy_request(struct intel_dp *intel_dp, static u8 intel_dp_autotest_phy_pattern(struct intel_dp *intel_dp) { + struct drm_i915_private *i915 = dp_to_i915(intel_dp); struct drm_dp_phy_test_params *data = &intel_dp->compliance.test_data.phytest; if (drm_dp_get_phy_test_pattern(&intel_dp->aux, data)) { - DRM_DEBUG_KMS("DP Phy Test pattern AUX read failure\n"); + drm_dbg_kms(&i915->drm, "DP Phy Test pattern AUX read failure\n"); return DP_TEST_NAK; } @@ -3615,14 +3629,43 @@ update_status: } static void -intel_dp_mst_hpd_irq(struct intel_dp *intel_dp, u8 *esi, bool *handled) +intel_dp_mst_hpd_irq(struct intel_dp *intel_dp, u8 *esi, u8 *ack) { - drm_dp_mst_hpd_irq(&intel_dp->mst_mgr, esi, handled); + bool handled = false; - if (esi[1] & DP_CP_IRQ) { - intel_hdcp_handle_cp_irq(intel_dp->attached_connector); - *handled = true; - } + drm_dp_mst_hpd_irq(&intel_dp->mst_mgr, esi, &handled); + if (handled) + ack[1] |= esi[1] & (DP_DOWN_REP_MSG_RDY | DP_UP_REQ_MSG_RDY); + + if (esi[1] & DP_CP_IRQ) { + intel_hdcp_handle_cp_irq(intel_dp->attached_connector); + ack[1] |= DP_CP_IRQ; + } +} + +static bool intel_dp_mst_link_status(struct intel_dp *intel_dp) +{ + struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + u8 link_status[DP_LINK_STATUS_SIZE] = {}; + const size_t esi_link_status_size = DP_LINK_STATUS_SIZE - 2; + + if (drm_dp_dpcd_read(&intel_dp->aux, DP_LANE0_1_STATUS_ESI, link_status, + esi_link_status_size) != esi_link_status_size) { + drm_err(&i915->drm, + "[ENCODER:%d:%s] Failed to read link status\n", + encoder->base.base.id, encoder->base.name); + return false; + } + + if (!drm_dp_channel_eq_ok(link_status, intel_dp->lane_count)) { + drm_dbg_kms(&i915->drm, + "[ENCODER:%d:%s] channel EQ not ok, retraining\n", + encoder->base.base.id, encoder->base.name); + return false; + } + + return true; } /** @@ -3647,20 +3690,8 @@ intel_dp_check_mst_status(struct intel_dp *intel_dp) drm_WARN_ON_ONCE(&i915->drm, intel_dp->active_mst_links < 0); for (;;) { - /* - * The +2 is because DP_DPRX_ESI_LEN is 14, but we then - * pass in "esi+10" to drm_dp_channel_eq_ok(), which - * takes a 6-byte array. So we actually need 16 bytes - * here. - * - * Somebody who knows what the limits actually are - * should check this, but for now this is at least - * harmless and avoids a valid compiler warning about - * using more of the array than we have allocated. - */ - u8 esi[DP_DPRX_ESI_LEN+2] = {}; - bool handled; - int retry; + u8 esi[4] = {}; + u8 ack[4] = {}; if (!intel_dp_get_sink_irq_esi(intel_dp, esi)) { drm_dbg_kms(&i915->drm, @@ -3670,30 +3701,22 @@ intel_dp_check_mst_status(struct intel_dp *intel_dp) break; } - /* check link status - esi[10] = 0x200c */ + drm_dbg_kms(&i915->drm, "DPRX ESI: %4ph\n", esi); + if (intel_dp->active_mst_links > 0 && link_ok && - !drm_dp_channel_eq_ok(&esi[10], intel_dp->lane_count)) { - drm_dbg_kms(&i915->drm, - "channel EQ not ok, retraining\n"); - link_ok = false; + esi[3] & LINK_STATUS_CHANGED) { + if (!intel_dp_mst_link_status(intel_dp)) + link_ok = false; + ack[3] |= LINK_STATUS_CHANGED; } - drm_dbg_kms(&i915->drm, "got esi %3ph\n", esi); + intel_dp_mst_hpd_irq(intel_dp, esi, ack); - intel_dp_mst_hpd_irq(intel_dp, esi, &handled); - - if (!handled) + if (!memchr_inv(ack, 0, sizeof(ack))) break; - for (retry = 0; retry < 3; retry++) { - int wret; - - wret = drm_dp_dpcd_write(&intel_dp->aux, - DP_SINK_COUNT_ESI+1, - &esi[1], 3); - if (wret == 3) - break; - } + if (!intel_dp_ack_sink_irq_esi(intel_dp, ack)) + drm_dbg_kms(&i915->drm, "Failed to ack ESI\n"); } return link_ok; @@ -4974,6 +4997,14 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp, mutex_lock(&dev->mode_config.mutex); edid = drm_get_edid(connector, &intel_dp->aux.ddc); + if (!edid) { + /* Fallback to EDID from ACPI OpRegion, if any */ + edid = intel_opregion_get_edid(intel_connector); + if (edid) + drm_dbg_kms(&dev_priv->drm, + "[CONNECTOR:%d:%s] Using OpRegion EDID\n", + connector->base.id, connector->name); + } if (edid) { if (drm_add_edid_modes(connector, edid)) { drm_connector_update_edid_property(connector, edid); @@ -5048,8 +5079,8 @@ static void intel_dp_modeset_retry_work_fn(struct work_struct *work) intel_connector = container_of(work, typeof(*intel_connector), modeset_retry_work); connector = &intel_connector->base; - DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n", connector->base.id, - connector->name); + drm_dbg_kms(connector->dev, "[CONNECTOR:%d:%s]\n", connector->base.id, + connector->name); /* Grab the locks before changing connector property*/ mutex_lock(&connector->dev->mode_config.mutex); diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h index b64145a3869a..d457e17bdc57 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.h +++ b/drivers/gpu/drm/i915/display/intel_dp.h @@ -8,8 +8,6 @@ #include <linux/types.h> -#include "i915_reg.h" - enum intel_output_format; enum pipe; enum port; diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c index 5fbb767fcd63..2bc119374555 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c @@ -10,7 +10,7 @@ #include "intel_pps.h" #include "intel_tc.h" -u32 intel_dp_pack_aux(const u8 *src, int src_bytes) +static u32 intel_dp_aux_pack(const u8 *src, int src_bytes) { int i; u32 v = 0; @@ -22,7 +22,7 @@ u32 intel_dp_pack_aux(const u8 *src, int src_bytes) return v; } -static void intel_dp_unpack_aux(u32 src, u8 *dst, int dst_bytes) +static void intel_dp_aux_unpack(u32 src, u8 *dst, int dst_bytes) { int i; @@ -267,7 +267,7 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp, for (i = 0; i < send_bytes; i += 4) intel_uncore_write(uncore, ch_data[i >> 2], - intel_dp_pack_aux(send + i, + intel_dp_aux_pack(send + i, send_bytes - i)); /* Send the command and wait for it to complete */ @@ -352,7 +352,7 @@ done: recv_bytes = recv_size; for (i = 0; i < recv_bytes; i += 4) - intel_dp_unpack_aux(intel_uncore_read(uncore, ch_data[i >> 2]), + intel_dp_aux_unpack(intel_uncore_read(uncore, ch_data[i >> 2]), recv + i, recv_bytes - i); ret = recv_bytes; diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.h b/drivers/gpu/drm/i915/display/intel_dp_aux.h index 4afbe76217b9..738577537bc7 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux.h +++ b/drivers/gpu/drm/i915/display/intel_dp_aux.h @@ -6,12 +6,8 @@ #ifndef __INTEL_DP_AUX_H__ #define __INTEL_DP_AUX_H__ -#include <linux/types.h> - struct intel_dp; -u32 intel_dp_pack_aux(const u8 *src, int src_bytes); - void intel_dp_aux_fini(struct intel_dp *intel_dp); void intel_dp_aux_init(struct intel_dp *intel_dp); diff --git a/drivers/gpu/drm/i915/display/intel_dp_hdcp.c b/drivers/gpu/drm/i915/display/intel_dp_hdcp.c index 540a669e01dd..82d024dafe7b 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_dp_hdcp.c @@ -6,8 +6,8 @@ * Sean Paul <seanpaul@chromium.org> */ -#include <drm/drm_dp_helper.h> -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_hdcp.h> #include <drm/drm_print.h> diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.h b/drivers/gpu/drm/i915/display/intel_dp_link_training.h index 6a3a7b37349a..dbfb15705aaa 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_link_training.h +++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.h @@ -6,7 +6,7 @@ #ifndef __INTEL_DP_LINK_TRAINING_H__ #define __INTEL_DP_LINK_TRAINING_H__ -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> struct intel_crtc_state; struct intel_dp; diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index b8bc7d397c81..6b6eab507d30 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -522,8 +522,6 @@ static void intel_mst_pre_enable_dp(struct intel_atomic_state *state, intel_ddi_enable_pipe_clock(encoder, pipe_config); intel_ddi_set_dp_msa(pipe_config, conn_state); - - intel_dp_set_m_n(pipe_config, M1_N1); } static void intel_mst_enable_dp(struct intel_atomic_state *state, diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c index fc8fda77483a..6723c3de5a80 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c @@ -28,6 +28,7 @@ #include "intel_dpll_mgr.h" #include "intel_pch_refclk.h" #include "intel_tc.h" +#include "intel_tc_phy_regs.h" /** * DOC: Display PLLs @@ -49,6 +50,41 @@ * commit phase. */ +/* platform specific hooks for managing DPLLs */ +struct intel_shared_dpll_funcs { + /* + * Hook for enabling the pll, called from intel_enable_shared_dpll() if + * the pll is not already enabled. + */ + void (*enable)(struct drm_i915_private *i915, + struct intel_shared_dpll *pll); + + /* + * Hook for disabling the pll, called from intel_disable_shared_dpll() + * only when it is safe to disable the pll, i.e., there are no more + * tracked users for it. + */ + void (*disable)(struct drm_i915_private *i915, + struct intel_shared_dpll *pll); + + /* + * Hook for reading the values currently programmed to the DPLL + * registers. This is used for initial hw state readout and state + * verification after a mode set. + */ + bool (*get_hw_state)(struct drm_i915_private *i915, + struct intel_shared_dpll *pll, + struct intel_dpll_hw_state *hw_state); + + /* + * Hook for calculating the pll's output frequency based on its passed + * in state. + */ + int (*get_freq)(struct drm_i915_private *i915, + const struct intel_shared_dpll *pll, + const struct intel_dpll_hw_state *pll_state); +}; + struct intel_dpll_mgr { const struct dpll_info *dpll_info; diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h index ef2889753807..91fe181462b2 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h @@ -44,6 +44,7 @@ struct intel_crtc; struct intel_crtc_state; struct intel_encoder; struct intel_shared_dpll; +struct intel_shared_dpll_funcs; /** * enum intel_dpll_id - possible DPLL ids @@ -252,51 +253,6 @@ struct intel_shared_dpll_state { }; /** - * struct intel_shared_dpll_funcs - platform specific hooks for managing DPLLs - */ -struct intel_shared_dpll_funcs { - /** - * @enable: - * - * Hook for enabling the pll, called from intel_enable_shared_dpll() - * if the pll is not already enabled. - */ - void (*enable)(struct drm_i915_private *dev_priv, - struct intel_shared_dpll *pll); - - /** - * @disable: - * - * Hook for disabling the pll, called from intel_disable_shared_dpll() - * only when it is safe to disable the pll, i.e., there are no more - * tracked users for it. - */ - void (*disable)(struct drm_i915_private *dev_priv, - struct intel_shared_dpll *pll); - - /** - * @get_hw_state: - * - * Hook for reading the values currently programmed to the DPLL - * registers. This is used for initial hw state readout and state - * verification after a mode set. - */ - bool (*get_hw_state)(struct drm_i915_private *dev_priv, - struct intel_shared_dpll *pll, - struct intel_dpll_hw_state *hw_state); - - /** - * @get_freq: - * - * Hook for calculating the pll's output frequency based on its - * passed in state. - */ - int (*get_freq)(struct drm_i915_private *i915, - const struct intel_shared_dpll *pll, - const struct intel_dpll_hw_state *pll_state); -}; - -/** * struct dpll_info - display PLL platform specific info */ struct dpll_info { diff --git a/drivers/gpu/drm/i915/display/intel_drrs.c b/drivers/gpu/drm/i915/display/intel_drrs.c index 3ff149df4a77..fa715b8ea310 100644 --- a/drivers/gpu/drm/i915/display/intel_drrs.c +++ b/drivers/gpu/drm/i915/display/intel_drrs.c @@ -47,17 +47,13 @@ * requested by userspace. */ -void -intel_drrs_compute_config(struct intel_dp *intel_dp, - struct intel_crtc_state *pipe_config, - int output_bpp, bool constant_n) +static bool can_enable_drrs(struct intel_connector *connector, + const struct intel_crtc_state *pipe_config) { - struct intel_connector *intel_connector = intel_dp->attached_connector; - struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); - int pixel_clock; + const struct drm_i915_private *i915 = to_i915(connector->base.dev); if (pipe_config->vrr.enable) - return; + return false; /* * DRRS and PSR can't be enable together, so giving preference to PSR @@ -66,15 +62,30 @@ intel_drrs_compute_config(struct intel_dp *intel_dp, * after intel_psr_compute_config(). */ if (pipe_config->has_psr) - return; + return false; + + return connector->panel.downclock_mode && + i915->drrs.type == SEAMLESS_DRRS_SUPPORT; +} - if (!intel_connector->panel.downclock_mode || - dev_priv->drrs.type != SEAMLESS_DRRS_SUPPORT) +void +intel_drrs_compute_config(struct intel_dp *intel_dp, + struct intel_crtc_state *pipe_config, + int output_bpp, bool constant_n) +{ + struct intel_connector *connector = intel_dp->attached_connector; + struct drm_i915_private *i915 = to_i915(connector->base.dev); + int pixel_clock; + + if (!can_enable_drrs(connector, pipe_config)) { + if (intel_cpu_transcoder_has_m2_n2(i915, pipe_config->cpu_transcoder)) + intel_zero_m_n(&pipe_config->dp_m2_n2); return; + } pipe_config->has_drrs = true; - pixel_clock = intel_connector->panel.downclock_mode->clock; + pixel_clock = connector->panel.downclock_mode->clock; if (pipe_config->splitter.enable) pixel_clock /= pipe_config->splitter.link_count; @@ -84,7 +95,42 @@ intel_drrs_compute_config(struct intel_dp *intel_dp, /* FIXME: abstract this better */ if (pipe_config->splitter.enable) - pipe_config->dp_m2_n2.gmch_m *= pipe_config->splitter.link_count; + pipe_config->dp_m2_n2.data_m *= pipe_config->splitter.link_count; +} + +static void +intel_drrs_set_refresh_rate_pipeconf(const struct intel_crtc_state *crtc_state, + enum drrs_refresh_rate_type refresh_type) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; + u32 val, bit; + + if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) + bit = PIPECONF_EDP_RR_MODE_SWITCH_VLV; + else + bit = PIPECONF_EDP_RR_MODE_SWITCH; + + val = intel_de_read(dev_priv, PIPECONF(cpu_transcoder)); + + if (refresh_type == DRRS_LOW_RR) + val |= bit; + else + val &= ~bit; + + intel_de_write(dev_priv, PIPECONF(cpu_transcoder), val); +} + +static void +intel_drrs_set_refresh_rate_m_n(const struct intel_crtc_state *crtc_state, + enum drrs_refresh_rate_type refresh_type) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + + intel_cpu_transcoder_set_m1_n1(crtc, crtc_state->cpu_transcoder, + refresh_type == DRRS_LOW_RR ? + &crtc_state->dp_m2_n2 : &crtc_state->dp_m_n); } static void intel_drrs_set_state(struct drm_i915_private *dev_priv, @@ -120,37 +166,10 @@ static void intel_drrs_set_state(struct drm_i915_private *dev_priv, return; } - if (DISPLAY_VER(dev_priv) >= 8 && !IS_CHERRYVIEW(dev_priv)) { - switch (refresh_type) { - case DRRS_HIGH_RR: - intel_dp_set_m_n(crtc_state, M1_N1); - break; - case DRRS_LOW_RR: - intel_dp_set_m_n(crtc_state, M2_N2); - break; - case DRRS_MAX_RR: - default: - drm_err(&dev_priv->drm, - "Unsupported refreshrate type\n"); - } - } else if (DISPLAY_VER(dev_priv) > 6) { - i915_reg_t reg = PIPECONF(crtc_state->cpu_transcoder); - u32 val; - - val = intel_de_read(dev_priv, reg); - if (refresh_type == DRRS_LOW_RR) { - if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) - val |= PIPECONF_EDP_RR_MODE_SWITCH_VLV; - else - val |= PIPECONF_EDP_RR_MODE_SWITCH; - } else { - if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) - val &= ~PIPECONF_EDP_RR_MODE_SWITCH_VLV; - else - val &= ~PIPECONF_EDP_RR_MODE_SWITCH; - } - intel_de_write(dev_priv, reg, val); - } + if (DISPLAY_VER(dev_priv) >= 8 && !IS_CHERRYVIEW(dev_priv)) + intel_drrs_set_refresh_rate_m_n(crtc_state, refresh_type); + else if (DISPLAY_VER(dev_priv) > 6) + intel_drrs_set_refresh_rate_pipeconf(crtc_state, refresh_type); dev_priv->drrs.refresh_rate_type = refresh_type; diff --git a/drivers/gpu/drm/i915/display/intel_dsb.h b/drivers/gpu/drm/i915/display/intel_dsb.h index 654a11f24b80..6cb9c580cdca 100644 --- a/drivers/gpu/drm/i915/display/intel_dsb.h +++ b/drivers/gpu/drm/i915/display/intel_dsb.h @@ -8,7 +8,7 @@ #include <linux/types.h> -#include "i915_reg.h" +#include "i915_reg_defs.h" struct intel_crtc_state; struct i915_vma; diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c index 0da91849efde..a85574c413e8 100644 --- a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c +++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c @@ -39,6 +39,7 @@ #include <video/mipi_display.h> #include "i915_drv.h" +#include "i915_reg.h" #include "intel_display_types.h" #include "intel_dsi.h" #include "intel_dsi_vbt.h" @@ -426,24 +427,16 @@ static void i2c_acpi_find_adapter(struct intel_dsi *intel_dsi, const u16 slave_addr) { struct drm_device *drm_dev = intel_dsi->base.base.dev; - struct device *dev = drm_dev->dev; - struct acpi_device *acpi_dev; - struct list_head resource_list; - struct i2c_adapter_lookup lookup; - - acpi_dev = ACPI_COMPANION(dev); - if (acpi_dev) { - memset(&lookup, 0, sizeof(lookup)); - lookup.slave_addr = slave_addr; - lookup.intel_dsi = intel_dsi; - lookup.dev_handle = acpi_device_handle(acpi_dev); - - INIT_LIST_HEAD(&resource_list); - acpi_dev_get_resources(acpi_dev, &resource_list, - i2c_adapter_lookup, - &lookup); - acpi_dev_free_resource_list(&resource_list); - } + struct acpi_device *adev = ACPI_COMPANION(drm_dev->dev); + struct i2c_adapter_lookup lookup = { + .slave_addr = slave_addr, + .intel_dsi = intel_dsi, + .dev_handle = acpi_device_handle(adev), + }; + LIST_HEAD(resource_list); + + acpi_dev_get_resources(adev, &resource_list, i2c_adapter_lookup, &lookup); + acpi_dev_free_resource_list(&resource_list); } #else static inline void i2c_acpi_find_adapter(struct intel_dsi *intel_dsi, diff --git a/drivers/gpu/drm/i915/display/intel_dvo_dev.h b/drivers/gpu/drm/i915/display/intel_dvo_dev.h index 94a6ae1e0292..d96c3cc46e50 100644 --- a/drivers/gpu/drm/i915/display/intel_dvo_dev.h +++ b/drivers/gpu/drm/i915/display/intel_dvo_dev.h @@ -27,7 +27,7 @@ #include <drm/drm_crtc.h> -#include "i915_reg.h" +#include "i915_reg_defs.h" struct intel_dvo_device { const char *name; diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index 957feeccff3f..3ec34de35460 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -49,6 +49,14 @@ #include "intel_fbc.h" #include "intel_frontbuffer.h" +#define for_each_fbc_id(__dev_priv, __fbc_id) \ + for ((__fbc_id) = INTEL_FBC_A; (__fbc_id) < I915_MAX_FBCS; (__fbc_id)++) \ + for_each_if(INTEL_INFO(__dev_priv)->display.fbc_mask & BIT(__fbc_id)) + +#define for_each_intel_fbc(__dev_priv, __fbc, __fbc_id) \ + for_each_fbc_id((__dev_priv), (__fbc_id)) \ + for_each_if((__fbc) = (__dev_priv)->fbc[(__fbc_id)]) + struct intel_fbc_funcs { void (*activate)(struct intel_fbc *fbc); void (*deactivate)(struct intel_fbc *fbc); @@ -85,6 +93,8 @@ struct intel_fbc { struct drm_mm_node compressed_fb; struct drm_mm_node compressed_llb; + enum intel_fbc_id id; + u8 limit; bool false_color; @@ -454,10 +464,10 @@ static void ilk_fbc_activate(struct intel_fbc *fbc) struct intel_fbc_state *fbc_state = &fbc->state; struct drm_i915_private *i915 = fbc->i915; - intel_de_write(i915, ILK_DPFC_FENCE_YOFF, + intel_de_write(i915, ILK_DPFC_FENCE_YOFF(fbc->id), fbc_state->fence_y_offset); - intel_de_write(i915, ILK_DPFC_CONTROL, + intel_de_write(i915, ILK_DPFC_CONTROL(fbc->id), DPFC_CTL_EN | g4x_dpfc_ctl(fbc)); } @@ -467,28 +477,28 @@ static void ilk_fbc_deactivate(struct intel_fbc *fbc) u32 dpfc_ctl; /* Disable compression */ - dpfc_ctl = intel_de_read(i915, ILK_DPFC_CONTROL); + dpfc_ctl = intel_de_read(i915, ILK_DPFC_CONTROL(fbc->id)); if (dpfc_ctl & DPFC_CTL_EN) { dpfc_ctl &= ~DPFC_CTL_EN; - intel_de_write(i915, ILK_DPFC_CONTROL, dpfc_ctl); + intel_de_write(i915, ILK_DPFC_CONTROL(fbc->id), dpfc_ctl); } } static bool ilk_fbc_is_active(struct intel_fbc *fbc) { - return intel_de_read(fbc->i915, ILK_DPFC_CONTROL) & DPFC_CTL_EN; + return intel_de_read(fbc->i915, ILK_DPFC_CONTROL(fbc->id)) & DPFC_CTL_EN; } static bool ilk_fbc_is_compressing(struct intel_fbc *fbc) { - return intel_de_read(fbc->i915, ILK_DPFC_STATUS) & DPFC_COMP_SEG_MASK; + return intel_de_read(fbc->i915, ILK_DPFC_STATUS(fbc->id)) & DPFC_COMP_SEG_MASK; } static void ilk_fbc_program_cfb(struct intel_fbc *fbc) { struct drm_i915_private *i915 = fbc->i915; - intel_de_write(i915, ILK_DPFC_CB_BASE, fbc->compressed_fb.start); + intel_de_write(i915, ILK_DPFC_CB_BASE(fbc->id), fbc->compressed_fb.start); } static const struct intel_fbc_funcs ilk_fbc_funcs = { @@ -524,8 +534,8 @@ static void snb_fbc_nuke(struct intel_fbc *fbc) { struct drm_i915_private *i915 = fbc->i915; - intel_de_write(i915, MSG_FBC_REND_STATE, FBC_REND_NUKE); - intel_de_posting_read(i915, MSG_FBC_REND_STATE); + intel_de_write(i915, MSG_FBC_REND_STATE(fbc->id), FBC_REND_NUKE); + intel_de_posting_read(i915, MSG_FBC_REND_STATE(fbc->id)); } static const struct intel_fbc_funcs snb_fbc_funcs = { @@ -547,7 +557,7 @@ static void glk_fbc_program_cfb_stride(struct intel_fbc *fbc) val |= FBC_STRIDE_OVERRIDE | FBC_STRIDE(fbc_state->override_cfb_stride / fbc->limit); - intel_de_write(i915, GLK_FBC_STRIDE, val); + intel_de_write(i915, GLK_FBC_STRIDE(fbc->id), val); } static void skl_fbc_program_cfb_stride(struct intel_fbc *fbc) @@ -598,19 +608,19 @@ static void ivb_fbc_activate(struct intel_fbc *fbc) if (i915->ggtt.num_fences) snb_fbc_program_fence(fbc); - intel_de_write(i915, ILK_DPFC_CONTROL, + intel_de_write(i915, ILK_DPFC_CONTROL(fbc->id), DPFC_CTL_EN | ivb_dpfc_ctl(fbc)); } static bool ivb_fbc_is_compressing(struct intel_fbc *fbc) { - return intel_de_read(fbc->i915, ILK_DPFC_STATUS2) & DPFC_COMP_SEG_MASK_IVB; + return intel_de_read(fbc->i915, ILK_DPFC_STATUS2(fbc->id)) & DPFC_COMP_SEG_MASK_IVB; } static void ivb_fbc_set_false_color(struct intel_fbc *fbc, bool enable) { - intel_de_rmw(fbc->i915, ILK_DPFC_CONTROL, + intel_de_rmw(fbc->i915, ILK_DPFC_CONTROL(fbc->id), DPFC_CTL_FALSE_COLOR, enable ? DPFC_CTL_FALSE_COLOR : 0); } @@ -810,16 +820,16 @@ static void __intel_fbc_cleanup_cfb(struct intel_fbc *fbc) void intel_fbc_cleanup(struct drm_i915_private *i915) { - struct intel_fbc *fbc = i915->fbc; - - if (!fbc) - return; + struct intel_fbc *fbc; + enum intel_fbc_id fbc_id; - mutex_lock(&fbc->lock); - __intel_fbc_cleanup_cfb(fbc); - mutex_unlock(&fbc->lock); + for_each_intel_fbc(i915, fbc, fbc_id) { + mutex_lock(&fbc->lock); + __intel_fbc_cleanup_cfb(fbc); + mutex_unlock(&fbc->lock); - kfree(fbc); + kfree(fbc); + } } static bool stride_is_valid(const struct intel_plane_state *plane_state) @@ -1306,15 +1316,10 @@ static unsigned int intel_fbc_get_frontbuffer_bit(struct intel_fbc *fbc) return fbc->possible_framebuffer_bits; } -void intel_fbc_invalidate(struct drm_i915_private *i915, - unsigned int frontbuffer_bits, - enum fb_op_origin origin) +static void __intel_fbc_invalidate(struct intel_fbc *fbc, + unsigned int frontbuffer_bits, + enum fb_op_origin origin) { - struct intel_fbc *fbc = i915->fbc; - - if (!fbc) - return; - if (origin == ORIGIN_FLIP || origin == ORIGIN_CURSOR_UPDATE) return; @@ -1328,14 +1333,22 @@ void intel_fbc_invalidate(struct drm_i915_private *i915, mutex_unlock(&fbc->lock); } -void intel_fbc_flush(struct drm_i915_private *i915, - unsigned int frontbuffer_bits, enum fb_op_origin origin) +void intel_fbc_invalidate(struct drm_i915_private *i915, + unsigned int frontbuffer_bits, + enum fb_op_origin origin) { - struct intel_fbc *fbc = i915->fbc; + struct intel_fbc *fbc; + enum intel_fbc_id fbc_id; - if (!fbc) - return; + for_each_intel_fbc(i915, fbc, fbc_id) + __intel_fbc_invalidate(fbc, frontbuffer_bits, origin); + +} +static void __intel_fbc_flush(struct intel_fbc *fbc, + unsigned int frontbuffer_bits, + enum fb_op_origin origin) +{ mutex_lock(&fbc->lock); fbc->busy_bits &= ~frontbuffer_bits; @@ -1355,6 +1368,17 @@ out: mutex_unlock(&fbc->lock); } +void intel_fbc_flush(struct drm_i915_private *i915, + unsigned int frontbuffer_bits, + enum fb_op_origin origin) +{ + struct intel_fbc *fbc; + enum intel_fbc_id fbc_id; + + for_each_intel_fbc(i915, fbc, fbc_id) + __intel_fbc_flush(fbc, frontbuffer_bits, origin); +} + int intel_fbc_atomic_check(struct intel_atomic_state *state) { struct intel_plane_state *plane_state; @@ -1484,15 +1508,15 @@ void intel_fbc_update(struct intel_atomic_state *state, */ void intel_fbc_global_disable(struct drm_i915_private *i915) { - struct intel_fbc *fbc = i915->fbc; - - if (!fbc) - return; + struct intel_fbc *fbc; + enum intel_fbc_id fbc_id; - mutex_lock(&fbc->lock); - if (fbc->state.plane) - __intel_fbc_disable(fbc); - mutex_unlock(&fbc->lock); + for_each_intel_fbc(i915, fbc, fbc_id) { + mutex_lock(&fbc->lock); + if (fbc->state.plane) + __intel_fbc_disable(fbc); + mutex_unlock(&fbc->lock); + } } static void intel_fbc_underrun_work_fn(struct work_struct *work) @@ -1517,19 +1541,9 @@ out: mutex_unlock(&fbc->lock); } -/* - * intel_fbc_reset_underrun - reset FBC fifo underrun status. - * @i915: the i915 device - * - * See intel_fbc_handle_fifo_underrun_irq(). For automated testing we - * want to re-enable FBC after an underrun to increase test coverage. - */ -void intel_fbc_reset_underrun(struct drm_i915_private *i915) +static void __intel_fbc_reset_underrun(struct intel_fbc *fbc) { - struct intel_fbc *fbc = i915->fbc; - - if (!fbc) - return; + struct drm_i915_private *i915 = fbc->i915; cancel_work_sync(&fbc->underrun_work); @@ -1545,6 +1559,38 @@ void intel_fbc_reset_underrun(struct drm_i915_private *i915) mutex_unlock(&fbc->lock); } +/* + * intel_fbc_reset_underrun - reset FBC fifo underrun status. + * @i915: the i915 device + * + * See intel_fbc_handle_fifo_underrun_irq(). For automated testing we + * want to re-enable FBC after an underrun to increase test coverage. + */ +void intel_fbc_reset_underrun(struct drm_i915_private *i915) +{ + struct intel_fbc *fbc; + enum intel_fbc_id fbc_id; + + for_each_intel_fbc(i915, fbc, fbc_id) + __intel_fbc_reset_underrun(fbc); +} + +static void __intel_fbc_handle_fifo_underrun_irq(struct intel_fbc *fbc) +{ + /* + * There's no guarantee that underrun_detected won't be set to true + * right after this check and before the work is scheduled, but that's + * not a problem since we'll check it again under the work function + * while FBC is locked. This check here is just to prevent us from + * unnecessarily scheduling the work, and it relies on the fact that we + * never switch underrun_detect back to false after it's true. + */ + if (READ_ONCE(fbc->underrun_detected)) + return; + + schedule_work(&fbc->underrun_work); +} + /** * intel_fbc_handle_fifo_underrun_irq - disable FBC when we get a FIFO underrun * @i915: i915 device @@ -1561,21 +1607,11 @@ void intel_fbc_reset_underrun(struct drm_i915_private *i915) */ void intel_fbc_handle_fifo_underrun_irq(struct drm_i915_private *i915) { - struct intel_fbc *fbc = i915->fbc; - - if (!fbc) - return; - - /* There's no guarantee that underrun_detected won't be set to true - * right after this check and before the work is scheduled, but that's - * not a problem since we'll check it again under the work function - * while FBC is locked. This check here is just to prevent us from - * unnecessarily scheduling the work, and it relies on the fact that we - * never switch underrun_detect back to false after it's true. */ - if (READ_ONCE(fbc->underrun_detected)) - return; + struct intel_fbc *fbc; + enum intel_fbc_id fbc_id; - schedule_work(&fbc->underrun_work); + for_each_intel_fbc(i915, fbc, fbc_id) + __intel_fbc_handle_fifo_underrun_irq(fbc); } /* @@ -1623,7 +1659,8 @@ void intel_fbc_add_plane(struct intel_fbc *fbc, struct intel_plane *plane) fbc->possible_framebuffer_bits |= plane->frontbuffer_bit; } -static struct intel_fbc *intel_fbc_create(struct drm_i915_private *i915) +static struct intel_fbc *intel_fbc_create(struct drm_i915_private *i915, + enum intel_fbc_id fbc_id) { struct intel_fbc *fbc; @@ -1631,6 +1668,7 @@ static struct intel_fbc *intel_fbc_create(struct drm_i915_private *i915) if (!fbc) return NULL; + fbc->id = fbc_id; fbc->i915 = i915; INIT_WORK(&fbc->underrun_work, intel_fbc_underrun_work_fn); mutex_init(&fbc->lock); @@ -1659,32 +1697,35 @@ static struct intel_fbc *intel_fbc_create(struct drm_i915_private *i915) */ void intel_fbc_init(struct drm_i915_private *i915) { - struct intel_fbc *fbc; + enum intel_fbc_id fbc_id; if (!drm_mm_initialized(&i915->mm.stolen)) - mkwrite_device_info(i915)->display.has_fbc = false; + mkwrite_device_info(i915)->display.fbc_mask = 0; if (need_fbc_vtd_wa(i915)) - mkwrite_device_info(i915)->display.has_fbc = false; + mkwrite_device_info(i915)->display.fbc_mask = 0; i915->params.enable_fbc = intel_sanitize_fbc_option(i915); drm_dbg_kms(&i915->drm, "Sanitized enable_fbc value: %d\n", i915->params.enable_fbc); - if (!HAS_FBC(i915)) - return; + for_each_fbc_id(i915, fbc_id) { + struct intel_fbc *fbc; - fbc = intel_fbc_create(i915); - if (!fbc) - return; + fbc = intel_fbc_create(i915, fbc_id); + if (!fbc) + continue; - /* We still don't have any sort of hardware state readout for FBC, so - * deactivate it in case the BIOS activated it to make sure software - * matches the hardware state. */ - if (intel_fbc_hw_is_active(fbc)) - intel_fbc_hw_deactivate(fbc); + /* + * We still don't have any sort of hardware state readout + * for FBC, so deactivate it in case the BIOS activated it + * to make sure software matches the hardware state. + */ + if (intel_fbc_hw_is_active(fbc)) + intel_fbc_hw_deactivate(fbc); - i915->fbc = fbc; + i915->fbc[fbc->id] = fbc; + } } static int intel_fbc_debugfs_status_show(struct seq_file *m, void *unused) @@ -1760,25 +1801,32 @@ DEFINE_SIMPLE_ATTRIBUTE(intel_fbc_debugfs_false_color_fops, intel_fbc_debugfs_false_color_set, "%llu\n"); -static void intel_fbc_debugfs_add(struct intel_fbc *fbc) +static void intel_fbc_debugfs_add(struct intel_fbc *fbc, + struct dentry *parent) { - struct drm_i915_private *i915 = fbc->i915; - struct drm_minor *minor = i915->drm.primary; - - debugfs_create_file("i915_fbc_status", 0444, - minor->debugfs_root, fbc, - &intel_fbc_debugfs_status_fops); + debugfs_create_file("i915_fbc_status", 0444, parent, + fbc, &intel_fbc_debugfs_status_fops); if (fbc->funcs->set_false_color) - debugfs_create_file("i915_fbc_false_color", 0644, - minor->debugfs_root, fbc, - &intel_fbc_debugfs_false_color_fops); + debugfs_create_file("i915_fbc_false_color", 0644, parent, + fbc, &intel_fbc_debugfs_false_color_fops); } +void intel_fbc_crtc_debugfs_add(struct intel_crtc *crtc) +{ + struct intel_plane *plane = to_intel_plane(crtc->base.primary); + + if (plane->fbc) + intel_fbc_debugfs_add(plane->fbc, crtc->base.debugfs_entry); +} + +/* FIXME: remove this once igt is on board with per-crtc stuff */ void intel_fbc_debugfs_register(struct drm_i915_private *i915) { - struct intel_fbc *fbc = i915->fbc; + struct drm_minor *minor = i915->drm.primary; + struct intel_fbc *fbc; + fbc = i915->fbc[INTEL_FBC_A]; if (fbc) - intel_fbc_debugfs_add(fbc); + intel_fbc_debugfs_add(fbc, minor->debugfs_root); } diff --git a/drivers/gpu/drm/i915/display/intel_fbc.h b/drivers/gpu/drm/i915/display/intel_fbc.h index 07ad0411fcc3..8c5a7339a27f 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.h +++ b/drivers/gpu/drm/i915/display/intel_fbc.h @@ -17,6 +17,12 @@ struct intel_fbc; struct intel_plane; struct intel_plane_state; +enum intel_fbc_id { + INTEL_FBC_A, + + I915_MAX_FBCS, +}; + int intel_fbc_atomic_check(struct intel_atomic_state *state); bool intel_fbc_pre_update(struct intel_atomic_state *state, struct intel_crtc *crtc); @@ -36,6 +42,7 @@ void intel_fbc_flush(struct drm_i915_private *dev_priv, void intel_fbc_add_plane(struct intel_fbc *fbc, struct intel_plane *plane); void intel_fbc_handle_fifo_underrun_irq(struct drm_i915_private *i915); void intel_fbc_reset_underrun(struct drm_i915_private *i915); +void intel_fbc_crtc_debugfs_add(struct intel_crtc *crtc); void intel_fbc_debugfs_register(struct drm_i915_private *i915); #endif /* __INTEL_FBC_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c index 3b8b84177085..6ce8c10fe975 100644 --- a/drivers/gpu/drm/i915/display/intel_gmbus.c +++ b/drivers/gpu/drm/i915/display/intel_gmbus.c @@ -931,13 +931,6 @@ struct i2c_adapter *intel_gmbus_get_adapter(struct drm_i915_private *dev_priv, return &dev_priv->gmbus[pin].adapter; } -void intel_gmbus_set_speed(struct i2c_adapter *adapter, int speed) -{ - struct intel_gmbus *bus = to_intel_gmbus(adapter); - - bus->reg0 = (bus->reg0 & ~(0x3 << 8)) | speed; -} - void intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit) { struct intel_gmbus *bus = to_intel_gmbus(adapter); diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.h b/drivers/gpu/drm/i915/display/intel_gmbus.h index b96212b85425..8edc2e99cf53 100644 --- a/drivers/gpu/drm/i915/display/intel_gmbus.h +++ b/drivers/gpu/drm/i915/display/intel_gmbus.h @@ -41,7 +41,6 @@ int intel_gmbus_output_aksv(struct i2c_adapter *adapter); struct i2c_adapter * intel_gmbus_get_adapter(struct drm_i915_private *dev_priv, unsigned int pin); -void intel_gmbus_set_speed(struct i2c_adapter *adapter, int speed); void intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit); bool intel_gmbus_is_forced_bit(struct i2c_adapter *adapter); void intel_gmbus_reset(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index 4509fe7438e8..e1ecf38db0ef 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -297,8 +297,7 @@ static int intel_hdcp_load_keys(struct drm_i915_private *dev_priv) * Mailbox interface. */ if (DISPLAY_VER(dev_priv) == 9 && !IS_BROXTON(dev_priv)) { - ret = sandybridge_pcode_write(dev_priv, - SKL_PCODE_LOAD_HDCP_KEYS, 1); + ret = snb_pcode_write(dev_priv, SKL_PCODE_LOAD_HDCP_KEYS, 1); if (ret) { drm_err(&dev_priv->drm, "Failed to initiate HDCP key load (%d)\n", diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c index 3b5b9e7b05b7..6c72f8587240 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.c +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c @@ -730,7 +730,7 @@ intel_hdmi_compute_avi_infoframe(struct intel_encoder *encoder, else frame->colorspace = HDMI_COLORSPACE_RGB; - drm_hdmi_avi_infoframe_colorspace(frame, conn_state); + drm_hdmi_avi_infoframe_colorimetry(frame, conn_state); /* nonsense combination */ drm_WARN_ON(encoder->base.dev, crtc_state->limited_color_range && @@ -1912,7 +1912,7 @@ static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector, if (ycbcr420_output) return hdmi->y420_dc_modes & DRM_EDID_YCBCR420_DC_36; else - return info->edid_hdmi_dc_modes & DRM_EDID_HDMI_DC_36; + return info->edid_hdmi_rgb444_dc_modes & DRM_EDID_HDMI_DC_36; case 10: if (!has_hdmi_sink) return false; @@ -1920,7 +1920,7 @@ static bool intel_hdmi_sink_bpc_possible(struct drm_connector *connector, if (ycbcr420_output) return hdmi->y420_dc_modes & DRM_EDID_YCBCR420_DC_30; else - return info->edid_hdmi_dc_modes & DRM_EDID_HDMI_DC_30; + return info->edid_hdmi_rgb444_dc_modes & DRM_EDID_HDMI_DC_30; case 8: return true; default: @@ -2002,17 +2002,14 @@ intel_hdmi_mode_valid(struct drm_connector *connector, return intel_mode_valid_max_plane_size(dev_priv, mode, false); } -bool intel_hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state, - int bpc, bool has_hdmi_sink, bool ycbcr420_output) +bool intel_hdmi_bpc_possible(const struct intel_crtc_state *crtc_state, + int bpc, bool has_hdmi_sink, bool ycbcr420_output) { struct drm_atomic_state *state = crtc_state->uapi.state; struct drm_connector_state *connector_state; struct drm_connector *connector; int i; - if (crtc_state->pipe_bpp < bpc * 3) - return false; - for_each_new_connector_in_state(state, connector, connector_state, i) { if (connector_state->crtc != crtc_state->uapi.crtc) continue; @@ -2024,8 +2021,7 @@ bool intel_hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state, return true; } -static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state, - int bpc) +static bool hdmi_bpc_possible(const struct intel_crtc_state *crtc_state, int bpc) { struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); @@ -2039,7 +2035,7 @@ static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state, * HDMI deep color affects the clocks, so it's only possible * when not cloning with other encoder types. */ - if (crtc_state->output_types != BIT(INTEL_OUTPUT_HDMI)) + if (bpc > 8 && crtc_state->output_types != BIT(INTEL_OUTPUT_HDMI)) return false; /* Display Wa_1405510057:icl,ehl */ @@ -2049,35 +2045,50 @@ static bool hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state, adjusted_mode->crtc_hblank_start) % 8 == 2) return false; - return intel_hdmi_deep_color_possible(crtc_state, bpc, - crtc_state->has_hdmi_sink, - intel_hdmi_is_ycbcr420(crtc_state)); + return intel_hdmi_bpc_possible(crtc_state, bpc, crtc_state->has_hdmi_sink, + intel_hdmi_is_ycbcr420(crtc_state)); } static int intel_hdmi_compute_bpc(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, - int clock) + int clock, bool respect_downstream_limits) { struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); bool ycbcr420_output = intel_hdmi_is_ycbcr420(crtc_state); int bpc; - for (bpc = 12; bpc >= 10; bpc -= 2) { - if (hdmi_deep_color_possible(crtc_state, bpc) && - hdmi_port_clock_valid(intel_hdmi, - intel_hdmi_tmds_clock(clock, bpc, ycbcr420_output), - true, crtc_state->has_hdmi_sink) == MODE_OK) + /* + * pipe_bpp could already be below 8bpc due to FDI + * bandwidth constraints. HDMI minimum is 8bpc however. + */ + bpc = max(crtc_state->pipe_bpp / 3, 8); + + /* + * We will never exceed downstream TMDS clock limits while + * attempting deep color. If the user insists on forcing an + * out of spec mode they will have to be satisfied with 8bpc. + */ + if (!respect_downstream_limits) + bpc = 8; + + for (; bpc >= 8; bpc -= 2) { + int tmds_clock = intel_hdmi_tmds_clock(clock, bpc, ycbcr420_output); + + if (hdmi_bpc_possible(crtc_state, bpc) && + hdmi_port_clock_valid(intel_hdmi, tmds_clock, + respect_downstream_limits, + crtc_state->has_hdmi_sink) == MODE_OK) return bpc; } - return 8; + return -EINVAL; } static int intel_hdmi_compute_clock(struct intel_encoder *encoder, - struct intel_crtc_state *crtc_state) + struct intel_crtc_state *crtc_state, + bool respect_downstream_limits) { struct drm_i915_private *i915 = to_i915(encoder->base.dev); - struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode; int bpc, clock = adjusted_mode->crtc_clock; @@ -2085,31 +2096,25 @@ static int intel_hdmi_compute_clock(struct intel_encoder *encoder, if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK) clock *= 2; - bpc = intel_hdmi_compute_bpc(encoder, crtc_state, clock); + bpc = intel_hdmi_compute_bpc(encoder, crtc_state, clock, + respect_downstream_limits); + if (bpc < 0) + return bpc; - crtc_state->port_clock = intel_hdmi_tmds_clock(clock, bpc, - intel_hdmi_is_ycbcr420(crtc_state)); + crtc_state->port_clock = + intel_hdmi_tmds_clock(clock, bpc, intel_hdmi_is_ycbcr420(crtc_state)); /* * pipe_bpp could already be below 8bpc due to * FDI bandwidth constraints. We shouldn't bump it - * back up to 8bpc in that case. + * back up to the HDMI minimum 8bpc in that case. */ - if (crtc_state->pipe_bpp > bpc * 3) - crtc_state->pipe_bpp = bpc * 3; + crtc_state->pipe_bpp = min(crtc_state->pipe_bpp, bpc * 3); drm_dbg_kms(&i915->drm, "picking %d bpc for HDMI output (pipe bpp: %d)\n", bpc, crtc_state->pipe_bpp); - if (hdmi_port_clock_valid(intel_hdmi, crtc_state->port_clock, - false, crtc_state->has_hdmi_sink) != MODE_OK) { - drm_dbg_kms(&i915->drm, - "unsupported HDMI clock (%d kHz), rejecting mode\n", - crtc_state->port_clock); - return -EINVAL; - } - return 0; } @@ -2170,7 +2175,8 @@ intel_hdmi_output_format(struct intel_connector *connector, static int intel_hdmi_compute_output_format(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state) + const struct drm_connector_state *conn_state, + bool respect_downstream_limits) { struct intel_connector *connector = to_intel_connector(conn_state->connector); const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode; @@ -2187,7 +2193,7 @@ static int intel_hdmi_compute_output_format(struct intel_encoder *encoder, crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB; } - ret = intel_hdmi_compute_clock(encoder, crtc_state); + ret = intel_hdmi_compute_clock(encoder, crtc_state, respect_downstream_limits); if (ret) { if (intel_hdmi_is_ycbcr420(crtc_state) || !connector->base.ycbcr_420_allowed || @@ -2195,7 +2201,7 @@ static int intel_hdmi_compute_output_format(struct intel_encoder *encoder, return ret; crtc_state->output_format = intel_hdmi_output_format(connector, true); - ret = intel_hdmi_compute_clock(encoder, crtc_state); + ret = intel_hdmi_compute_clock(encoder, crtc_state, respect_downstream_limits); } return ret; @@ -2231,9 +2237,19 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder, pipe_config->has_audio = intel_hdmi_has_audio(encoder, pipe_config, conn_state); - ret = intel_hdmi_compute_output_format(encoder, pipe_config, conn_state); + /* + * Try to respect downstream TMDS clock limits first, if + * that fails assume the user might know something we don't. + */ + ret = intel_hdmi_compute_output_format(encoder, pipe_config, conn_state, true); if (ret) + ret = intel_hdmi_compute_output_format(encoder, pipe_config, conn_state, false); + if (ret) { + drm_dbg_kms(&dev_priv->drm, + "unsupported HDMI clock (%d kHz), rejecting mode\n", + pipe_config->hw.adjusted_mode.crtc_clock); return ret; + } if (intel_hdmi_is_ycbcr420(pipe_config)) { ret = intel_panel_fitting(pipe_config, conn_state); @@ -2359,6 +2375,14 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector, bool has_edid) "DP dual mode adaptor (%s) detected (max TMDS clock: %d kHz)\n", drm_dp_get_dual_mode_type_name(type), hdmi->dp_dual_mode.max_tmds_clock); + + /* Older VBTs are often buggy and can't be trusted :( Play it safe. */ + if ((DISPLAY_VER(dev_priv) >= 8 || IS_HASWELL(dev_priv)) && + !intel_bios_is_port_dp_dual_mode(dev_priv, port)) { + drm_dbg_kms(&dev_priv->drm, + "Ignoring DP dual mode adaptor max TMDS clock for native HDMI port\n"); + hdmi->dp_dual_mode.max_tmds_clock = 0; + } } static bool diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.h b/drivers/gpu/drm/i915/display/intel_hdmi.h index 2bf440eb400a..b0804b862a89 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.h +++ b/drivers/gpu/drm/i915/display/intel_hdmi.h @@ -9,8 +9,6 @@ #include <linux/hdmi.h> #include <linux/types.h> -#include "i915_reg.h" - struct drm_connector; struct drm_encoder; struct drm_i915_private; @@ -46,8 +44,8 @@ void intel_read_infoframe(struct intel_encoder *encoder, union hdmi_infoframe *frame); bool intel_hdmi_limited_color_range(const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state); -bool intel_hdmi_deep_color_possible(const struct intel_crtc_state *crtc_state, int bpc, - bool has_hdmi_sink, bool ycbcr420_output); +bool intel_hdmi_bpc_possible(const struct intel_crtc_state *crtc_state, + int bpc, bool has_hdmi_sink, bool ycbcr420_output); int intel_hdmi_dsc_get_bpp(int src_fractional_bpp, int slice_width, int num_slices, int output_format, bool hdmi_all_bpp, int hdmi_max_chunk_bytes); diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c index 955f6d07b0e1..912b7003dcfa 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug.c +++ b/drivers/gpu/drm/i915/display/intel_hotplug.c @@ -281,13 +281,13 @@ intel_encoder_hotplug(struct intel_encoder *encoder, ret = true; if (ret) { - DRM_DEBUG_KMS("[CONNECTOR:%d:%s] status updated from %s to %s (epoch counter %llu->%llu)\n", - connector->base.base.id, - connector->base.name, - drm_get_connector_status_name(old_status), - drm_get_connector_status_name(connector->base.status), - old_epoch_counter, - connector->base.epoch_counter); + drm_dbg_kms(dev, "[CONNECTOR:%d:%s] status updated from %s to %s (epoch counter %llu->%llu)\n", + connector->base.base.id, + connector->base.name, + drm_get_connector_status_name(old_status), + drm_get_connector_status_name(connector->base.status), + old_epoch_counter, + connector->base.epoch_counter); return INTEL_HOTPLUG_CHANGED; } return INTEL_HOTPLUG_UNCHANGED; diff --git a/drivers/gpu/drm/i915/display/intel_lspcon.c b/drivers/gpu/drm/i915/display/intel_lspcon.c index 05d2d750fa53..76357c9b76e4 100644 --- a/drivers/gpu/drm/i915/display/intel_lspcon.c +++ b/drivers/gpu/drm/i915/display/intel_lspcon.c @@ -24,7 +24,7 @@ */ #include <drm/drm_atomic_helper.h> -#include <drm/drm_dp_dual_mode_helper.h> +#include <drm/dp/drm_dp_dual_mode_helper.h> #include <drm/drm_edid.h> #include "intel_de.h" @@ -78,11 +78,12 @@ static const char *lspcon_mode_name(enum drm_lspcon_mode mode) static bool lspcon_detect_vendor(struct intel_lspcon *lspcon) { struct intel_dp *dp = lspcon_to_intel_dp(lspcon); + struct drm_i915_private *i915 = dp_to_i915(dp); struct drm_dp_dpcd_ident *ident; u32 vendor_oui; if (drm_dp_read_desc(&dp->aux, &dp->desc, drm_dp_is_branch(dp->dpcd))) { - DRM_ERROR("Can't read description\n"); + drm_err(&i915->drm, "Can't read description\n"); return false; } @@ -93,16 +94,16 @@ static bool lspcon_detect_vendor(struct intel_lspcon *lspcon) switch (vendor_oui) { case LSPCON_VENDOR_MCA_OUI: lspcon->vendor = LSPCON_VENDOR_MCA; - DRM_DEBUG_KMS("Vendor: Mega Chips\n"); + drm_dbg_kms(&i915->drm, "Vendor: Mega Chips\n"); break; case LSPCON_VENDOR_PARADE_OUI: lspcon->vendor = LSPCON_VENDOR_PARADE; - DRM_DEBUG_KMS("Vendor: Parade Tech\n"); + drm_dbg_kms(&i915->drm, "Vendor: Parade Tech\n"); break; default: - DRM_ERROR("Invalid/Unknown vendor OUI\n"); + drm_err(&i915->drm, "Invalid/Unknown vendor OUI\n"); return false; } @@ -119,21 +120,19 @@ static u32 get_hdr_status_reg(struct intel_lspcon *lspcon) void lspcon_detect_hdr_capability(struct intel_lspcon *lspcon) { - struct intel_digital_port *dig_port = - container_of(lspcon, struct intel_digital_port, lspcon); - struct drm_device *dev = dig_port->base.base.dev; - struct intel_dp *dp = lspcon_to_intel_dp(lspcon); + struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); u8 hdr_caps; int ret; - ret = drm_dp_dpcd_read(&dp->aux, get_hdr_status_reg(lspcon), + ret = drm_dp_dpcd_read(&intel_dp->aux, get_hdr_status_reg(lspcon), &hdr_caps, 1); if (ret < 0) { - drm_dbg_kms(dev, "HDR capability detection failed\n"); + drm_dbg_kms(&i915->drm, "HDR capability detection failed\n"); lspcon->hdr_supported = false; } else if (hdr_caps & 0x1) { - drm_dbg_kms(dev, "LSPCON capable of HDR\n"); + drm_dbg_kms(&i915->drm, "LSPCON capable of HDR\n"); lspcon->hdr_supported = true; } } @@ -141,11 +140,12 @@ void lspcon_detect_hdr_capability(struct intel_lspcon *lspcon) static enum drm_lspcon_mode lspcon_get_current_mode(struct intel_lspcon *lspcon) { struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); enum drm_lspcon_mode current_mode; struct i2c_adapter *adapter = &intel_dp->aux.ddc; if (drm_lspcon_get_mode(intel_dp->aux.drm_dev, adapter, ¤t_mode)) { - DRM_DEBUG_KMS("Error reading LSPCON mode\n"); + drm_dbg_kms(&i915->drm, "Error reading LSPCON mode\n"); return DRM_LSPCON_MODE_INVALID; } return current_mode; @@ -154,22 +154,24 @@ static enum drm_lspcon_mode lspcon_get_current_mode(struct intel_lspcon *lspcon) static enum drm_lspcon_mode lspcon_wait_mode(struct intel_lspcon *lspcon, enum drm_lspcon_mode mode) { + struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); enum drm_lspcon_mode current_mode; current_mode = lspcon_get_current_mode(lspcon); if (current_mode == mode) goto out; - DRM_DEBUG_KMS("Waiting for LSPCON mode %s to settle\n", - lspcon_mode_name(mode)); + drm_dbg_kms(&i915->drm, "Waiting for LSPCON mode %s to settle\n", + lspcon_mode_name(mode)); wait_for((current_mode = lspcon_get_current_mode(lspcon)) == mode, 400); if (current_mode != mode) - DRM_ERROR("LSPCON mode hasn't settled\n"); + drm_err(&i915->drm, "LSPCON mode hasn't settled\n"); out: - DRM_DEBUG_KMS("Current LSPCON mode %s\n", - lspcon_mode_name(current_mode)); + drm_dbg_kms(&i915->drm, "Current LSPCON mode %s\n", + lspcon_mode_name(current_mode)); return current_mode; } @@ -178,44 +180,47 @@ static int lspcon_change_mode(struct intel_lspcon *lspcon, enum drm_lspcon_mode mode) { struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); int err; enum drm_lspcon_mode current_mode; struct i2c_adapter *adapter = &intel_dp->aux.ddc; err = drm_lspcon_get_mode(intel_dp->aux.drm_dev, adapter, ¤t_mode); if (err) { - DRM_ERROR("Error reading LSPCON mode\n"); + drm_err(&i915->drm, "Error reading LSPCON mode\n"); return err; } if (current_mode == mode) { - DRM_DEBUG_KMS("Current mode = desired LSPCON mode\n"); + drm_dbg_kms(&i915->drm, "Current mode = desired LSPCON mode\n"); return 0; } err = drm_lspcon_set_mode(intel_dp->aux.drm_dev, adapter, mode); if (err < 0) { - DRM_ERROR("LSPCON mode change failed\n"); + drm_err(&i915->drm, "LSPCON mode change failed\n"); return err; } lspcon->mode = mode; - DRM_DEBUG_KMS("LSPCON mode changed done\n"); + drm_dbg_kms(&i915->drm, "LSPCON mode changed done\n"); return 0; } static bool lspcon_wake_native_aux_ch(struct intel_lspcon *lspcon) { + struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); u8 rev; if (drm_dp_dpcd_readb(&lspcon_to_intel_dp(lspcon)->aux, DP_DPCD_REV, &rev) != 1) { - DRM_DEBUG_KMS("Native AUX CH down\n"); + drm_dbg_kms(&i915->drm, "Native AUX CH down\n"); return false; } - DRM_DEBUG_KMS("Native AUX CH up, DPCD version: %d.%d\n", - rev >> 4, rev & 0xf); + drm_dbg_kms(&i915->drm, "Native AUX CH up, DPCD version: %d.%d\n", + rev >> 4, rev & 0xf); return true; } @@ -225,6 +230,7 @@ static bool lspcon_probe(struct intel_lspcon *lspcon) int retry; enum drm_dp_dual_mode_type adaptor_type; struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); struct i2c_adapter *adapter = &intel_dp->aux.ddc; enum drm_lspcon_mode expected_mode; @@ -242,13 +248,13 @@ static bool lspcon_probe(struct intel_lspcon *lspcon) } if (adaptor_type != DRM_DP_DUAL_MODE_LSPCON) { - DRM_DEBUG_KMS("No LSPCON detected, found %s\n", - drm_dp_get_dual_mode_type_name(adaptor_type)); + drm_dbg_kms(&i915->drm, "No LSPCON detected, found %s\n", + drm_dp_get_dual_mode_type_name(adaptor_type)); return false; } /* Yay ... got a LSPCON device */ - DRM_DEBUG_KMS("LSPCON detected\n"); + drm_dbg_kms(&i915->drm, "LSPCON detected\n"); lspcon->mode = lspcon_wait_mode(lspcon, expected_mode); /* @@ -258,7 +264,7 @@ static bool lspcon_probe(struct intel_lspcon *lspcon) */ if (lspcon->mode != DRM_LSPCON_MODE_PCON) { if (lspcon_change_mode(lspcon, DRM_LSPCON_MODE_PCON) < 0) { - DRM_ERROR("LSPCON mode change to PCON failed\n"); + drm_err(&i915->drm, "LSPCON mode change to PCON failed\n"); return false; } } @@ -268,13 +274,14 @@ static bool lspcon_probe(struct intel_lspcon *lspcon) static void lspcon_resume_in_pcon_wa(struct intel_lspcon *lspcon) { struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); unsigned long start = jiffies; while (1) { if (intel_digital_port_connected(&dig_port->base)) { - DRM_DEBUG_KMS("LSPCON recovering in PCON mode after %u ms\n", - jiffies_to_msecs(jiffies - start)); + drm_dbg_kms(&i915->drm, "LSPCON recovering in PCON mode after %u ms\n", + jiffies_to_msecs(jiffies - start)); return; } @@ -284,7 +291,7 @@ static void lspcon_resume_in_pcon_wa(struct intel_lspcon *lspcon) usleep_range(10000, 15000); } - DRM_DEBUG_KMS("LSPCON DP descriptor mismatch after resume\n"); + drm_dbg_kms(&i915->drm, "LSPCON DP descriptor mismatch after resume\n"); } static bool lspcon_parade_fw_ready(struct drm_dp_aux *aux) @@ -301,7 +308,7 @@ static bool lspcon_parade_fw_ready(struct drm_dp_aux *aux) ret = drm_dp_dpcd_read(aux, LSPCON_PARADE_AVI_IF_CTRL, &avi_if_ctrl, 1); if (ret < 0) { - DRM_ERROR("Failed to read AVI IF control\n"); + drm_err(aux->drm_dev, "Failed to read AVI IF control\n"); return false; } @@ -309,7 +316,7 @@ static bool lspcon_parade_fw_ready(struct drm_dp_aux *aux) return true; } - DRM_ERROR("Parade FW not ready to accept AVI IF\n"); + drm_err(aux->drm_dev, "Parade FW not ready to accept AVI IF\n"); return false; } @@ -324,8 +331,8 @@ static bool _lspcon_parade_write_infoframe_blocks(struct drm_dp_aux *aux, while (block_count < 4) { if (!lspcon_parade_fw_ready(aux)) { - DRM_DEBUG_KMS("LSPCON FW not ready, block %d\n", - block_count); + drm_dbg_kms(aux->drm_dev, "LSPCON FW not ready, block %d\n", + block_count); return false; } @@ -333,8 +340,8 @@ static bool _lspcon_parade_write_infoframe_blocks(struct drm_dp_aux *aux, data = avi_buf + block_count * 8; ret = drm_dp_dpcd_write(aux, reg, data, 8); if (ret < 0) { - DRM_ERROR("Failed to write AVI IF block %d\n", - block_count); + drm_err(aux->drm_dev, "Failed to write AVI IF block %d\n", + block_count); return false; } @@ -348,15 +355,15 @@ static bool _lspcon_parade_write_infoframe_blocks(struct drm_dp_aux *aux, avi_if_ctrl = LSPCON_PARADE_AVI_IF_KICKOFF | block_count; ret = drm_dp_dpcd_write(aux, reg, &avi_if_ctrl, 1); if (ret < 0) { - DRM_ERROR("Failed to update (0x%x), block %d\n", - reg, block_count); + drm_err(aux->drm_dev, "Failed to update (0x%x), block %d\n", + reg, block_count); return false; } block_count++; } - DRM_DEBUG_KMS("Wrote AVI IF blocks successfully\n"); + drm_dbg_kms(aux->drm_dev, "Wrote AVI IF blocks successfully\n"); return true; } @@ -378,14 +385,14 @@ static bool _lspcon_write_avi_infoframe_parade(struct drm_dp_aux *aux, */ if (len > LSPCON_PARADE_AVI_IF_DATA_SIZE - 1) { - DRM_ERROR("Invalid length of infoframes\n"); + drm_err(aux->drm_dev, "Invalid length of infoframes\n"); return false; } memcpy(&avi_if[1], frame, len); if (!_lspcon_parade_write_infoframe_blocks(aux, avi_if)) { - DRM_DEBUG_KMS("Failed to write infoframe blocks\n"); + drm_dbg_kms(aux->drm_dev, "Failed to write infoframe blocks\n"); return false; } @@ -412,7 +419,7 @@ static bool _lspcon_write_avi_infoframe_mca(struct drm_dp_aux *aux, mdelay(50); continue; } else { - DRM_ERROR("DPCD write failed at:0x%x\n", reg); + drm_err(aux->drm_dev, "DPCD write failed at:0x%x\n", reg); return false; } } @@ -423,7 +430,7 @@ static bool _lspcon_write_avi_infoframe_mca(struct drm_dp_aux *aux, reg = LSPCON_MCA_AVI_IF_CTRL; ret = drm_dp_dpcd_read(aux, reg, &val, 1); if (ret < 0) { - DRM_ERROR("DPCD read failed, address 0x%x\n", reg); + drm_err(aux->drm_dev, "DPCD read failed, address 0x%x\n", reg); return false; } @@ -433,19 +440,19 @@ static bool _lspcon_write_avi_infoframe_mca(struct drm_dp_aux *aux, ret = drm_dp_dpcd_write(aux, reg, &val, 1); if (ret < 0) { - DRM_ERROR("DPCD read failed, address 0x%x\n", reg); + drm_err(aux->drm_dev, "DPCD read failed, address 0x%x\n", reg); return false; } val = 0; ret = drm_dp_dpcd_read(aux, reg, &val, 1); if (ret < 0) { - DRM_ERROR("DPCD read failed, address 0x%x\n", reg); + drm_err(aux->drm_dev, "DPCD read failed, address 0x%x\n", reg); return false; } if (val == LSPCON_MCA_AVI_IF_HANDLED) - DRM_DEBUG_KMS("AVI IF handled by FW\n"); + drm_dbg_kms(aux->drm_dev, "AVI IF handled by FW\n"); return true; } @@ -457,6 +464,7 @@ void lspcon_write_infoframe(struct intel_encoder *encoder, { bool ret = true; struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); struct intel_lspcon *lspcon = enc_to_intel_lspcon(encoder); switch (type) { @@ -469,7 +477,7 @@ void lspcon_write_infoframe(struct intel_encoder *encoder, frame, len); break; case HDMI_PACKET_TYPE_GAMUT_METADATA: - drm_dbg_kms(encoder->base.dev, "Update HDR metadata for lspcon\n"); + drm_dbg_kms(&i915->drm, "Update HDR metadata for lspcon\n"); /* It uses the legacy hsw implementation for the same */ hsw_write_infoframe(encoder, crtc_state, type, frame, len); break; @@ -478,7 +486,7 @@ void lspcon_write_infoframe(struct intel_encoder *encoder, } if (!ret) { - DRM_ERROR("Failed to write infoframes\n"); + drm_err(&i915->drm, "Failed to write infoframes\n"); return; } } @@ -504,11 +512,12 @@ void lspcon_set_infoframes(struct intel_encoder *encoder, u8 buf[VIDEO_DIP_DATA_SIZE]; struct intel_digital_port *dig_port = enc_to_dig_port(encoder); struct intel_lspcon *lspcon = &dig_port->lspcon; + struct drm_i915_private *i915 = to_i915(encoder->base.dev); const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode; if (!lspcon->active) { - DRM_ERROR("Writing infoframes while LSPCON disabled ?\n"); + drm_err(&i915->drm, "Writing infoframes while LSPCON disabled ?\n"); return; } @@ -518,7 +527,7 @@ void lspcon_set_infoframes(struct intel_encoder *encoder, conn_state->connector, adjusted_mode); if (ret < 0) { - DRM_ERROR("couldn't fill AVI infoframe\n"); + drm_err(&i915->drm, "couldn't fill AVI infoframe\n"); return; } @@ -537,7 +546,7 @@ void lspcon_set_infoframes(struct intel_encoder *encoder, frame.avi.colorspace = HDMI_COLORSPACE_RGB; /* Set the Colorspace as per the HDMI spec */ - drm_hdmi_avi_infoframe_colorspace(&frame.avi, conn_state); + drm_hdmi_avi_infoframe_colorimetry(&frame.avi, conn_state); /* nonsense combination */ drm_WARN_ON(encoder->base.dev, crtc_state->limited_color_range && @@ -559,7 +568,7 @@ void lspcon_set_infoframes(struct intel_encoder *encoder, ret = hdmi_infoframe_pack(&frame, buf, sizeof(buf)); if (ret < 0) { - DRM_ERROR("Failed to pack AVI IF\n"); + drm_err(&i915->drm, "Failed to pack AVI IF\n"); return; } @@ -575,7 +584,7 @@ static bool _lspcon_read_avi_infoframe_enabled_mca(struct drm_dp_aux *aux) ret = drm_dp_dpcd_read(aux, reg, &val, 1); if (ret < 0) { - DRM_ERROR("DPCD read failed, address 0x%x\n", reg); + drm_err(aux->drm_dev, "DPCD read failed, address 0x%x\n", reg); return false; } @@ -590,7 +599,7 @@ static bool _lspcon_read_avi_infoframe_enabled_parade(struct drm_dp_aux *aux) ret = drm_dp_dpcd_read(aux, reg, &val, 1); if (ret < 0) { - DRM_ERROR("DPCD read failed, address 0x%x\n", reg); + drm_err(aux->drm_dev, "DPCD read failed, address 0x%x\n", reg); return false; } @@ -634,31 +643,32 @@ void lspcon_wait_pcon_mode(struct intel_lspcon *lspcon) bool lspcon_init(struct intel_digital_port *dig_port) { - struct intel_dp *dp = &dig_port->dp; + struct intel_dp *intel_dp = &dig_port->dp; struct intel_lspcon *lspcon = &dig_port->lspcon; - struct drm_connector *connector = &dp->attached_connector->base; + struct drm_i915_private *i915 = dp_to_i915(intel_dp); + struct drm_connector *connector = &intel_dp->attached_connector->base; lspcon->active = false; lspcon->mode = DRM_LSPCON_MODE_INVALID; if (!lspcon_probe(lspcon)) { - DRM_ERROR("Failed to probe lspcon\n"); + drm_err(&i915->drm, "Failed to probe lspcon\n"); return false; } - if (drm_dp_read_dpcd_caps(&dp->aux, dp->dpcd) != 0) { - DRM_ERROR("LSPCON DPCD read failed\n"); + if (drm_dp_read_dpcd_caps(&intel_dp->aux, intel_dp->dpcd) != 0) { + drm_err(&i915->drm, "LSPCON DPCD read failed\n"); return false; } if (!lspcon_detect_vendor(lspcon)) { - DRM_ERROR("LSPCON vendor detection failed\n"); + drm_err(&i915->drm, "LSPCON vendor detection failed\n"); return false; } connector->ycbcr_420_allowed = true; lspcon->active = true; - DRM_DEBUG_KMS("Success: LSPCON init\n"); + drm_dbg_kms(&i915->drm, "Success: LSPCON init\n"); return true; } @@ -674,16 +684,16 @@ void lspcon_resume(struct intel_digital_port *dig_port) { struct intel_lspcon *lspcon = &dig_port->lspcon; struct drm_device *dev = dig_port->base.base.dev; - struct drm_i915_private *dev_priv = to_i915(dev); + struct drm_i915_private *i915 = to_i915(dev); enum drm_lspcon_mode expected_mode; - if (!intel_bios_is_lspcon_present(dev_priv, dig_port->base.port)) + if (!intel_bios_is_lspcon_present(i915, dig_port->base.port)) return; if (!lspcon->active) { if (!lspcon_init(dig_port)) { - DRM_ERROR("LSPCON init failed on port %c\n", - port_name(dig_port->base.port)); + drm_err(&i915->drm, "LSPCON init failed on port %c\n", + port_name(dig_port->base.port)); return; } } @@ -699,7 +709,7 @@ void lspcon_resume(struct intel_digital_port *dig_port) return; if (lspcon_change_mode(lspcon, DRM_LSPCON_MODE_PCON)) - DRM_ERROR("LSPCON resume failed\n"); + drm_err(&i915->drm, "LSPCON resume failed\n"); else - DRM_DEBUG_KMS("LSPCON resume success\n"); + drm_dbg_kms(&i915->drm, "LSPCON resume success\n"); } diff --git a/drivers/gpu/drm/i915/display/intel_lvds.h b/drivers/gpu/drm/i915/display/intel_lvds.h index bc9c8b84ba2f..9d3372dc503f 100644 --- a/drivers/gpu/drm/i915/display/intel_lvds.h +++ b/drivers/gpu/drm/i915/display/intel_lvds.h @@ -8,7 +8,7 @@ #include <linux/types.h> -#include "i915_reg.h" +#include "i915_reg_defs.h" enum pipe; struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/display/intel_opregion.c b/drivers/gpu/drm/i915/display/intel_opregion.c index 4a2662838cd8..ad1afe9df6c3 100644 --- a/drivers/gpu/drm/i915/display/intel_opregion.c +++ b/drivers/gpu/drm/i915/display/intel_opregion.c @@ -35,6 +35,7 @@ #include "intel_backlight.h" #include "intel_display_types.h" #include "intel_opregion.h" +#include "intel_pci_config.h" #define OPREGION_HEADER_OFFSET 0 #define OPREGION_ACPI_OFFSET 0x100 @@ -195,6 +196,8 @@ struct opregion_asle_ext { #define ASLE_IUER_WINDOWS_BTN (1 << 1) #define ASLE_IUER_POWER_BTN (1 << 0) +#define ASLE_PHED_EDID_VALID_MASK 0x3 + /* Software System Control Interrupt (SWSCI) */ #define SWSCI_SCIC_INDICATOR (1 << 0) #define SWSCI_SCIC_MAIN_FUNCTION_SHIFT 1 @@ -923,8 +926,10 @@ int intel_opregion_setup(struct drm_i915_private *dev_priv) opregion->asle->ardy = ASLE_ARDY_NOT_READY; } - if (mboxes & MBOX_ASLE_EXT) + if (mboxes & MBOX_ASLE_EXT) { drm_dbg(&dev_priv->drm, "ASLE extension supported\n"); + opregion->asle_ext = base + OPREGION_ASLE_EXT_OFFSET; + } if (intel_load_vbt_firmware(dev_priv) == 0) goto out; @@ -1051,6 +1056,54 @@ intel_opregion_get_panel_type(struct drm_i915_private *dev_priv) return ret - 1; } +/** + * intel_opregion_get_edid - Fetch EDID from ACPI OpRegion mailbox #5 + * @intel_connector: eDP connector + * + * This reads the ACPI Opregion mailbox #5 to extract the EDID that is passed + * to it. + * + * Returns: + * The EDID in the OpRegion, or NULL if there is none or it's invalid. + * + */ +struct edid *intel_opregion_get_edid(struct intel_connector *intel_connector) +{ + struct drm_connector *connector = &intel_connector->base; + struct drm_i915_private *i915 = to_i915(connector->dev); + struct intel_opregion *opregion = &i915->opregion; + const void *in_edid; + const struct edid *edid; + struct edid *new_edid; + int len; + + if (!opregion->asle_ext) + return NULL; + + in_edid = opregion->asle_ext->bddc; + + /* Validity corresponds to number of 128-byte blocks */ + len = (opregion->asle_ext->phed & ASLE_PHED_EDID_VALID_MASK) * 128; + if (!len || !memchr_inv(in_edid, 0, len)) + return NULL; + + edid = in_edid; + + if (len < EDID_LENGTH * (1 + edid->extensions)) { + drm_dbg_kms(&i915->drm, "Invalid EDID in ACPI OpRegion (Mailbox #5): too short\n"); + return NULL; + } + new_edid = drm_edid_duplicate(edid); + if (!new_edid) + return NULL; + if (!drm_edid_is_valid(new_edid)) { + kfree(new_edid); + drm_dbg_kms(&i915->drm, "Invalid EDID in ACPI OpRegion (Mailbox #5)\n"); + return NULL; + } + return new_edid; +} + void intel_opregion_register(struct drm_i915_private *i915) { struct intel_opregion *opregion = &i915->opregion; @@ -1144,6 +1197,7 @@ void intel_opregion_unregister(struct drm_i915_private *i915) opregion->acpi = NULL; opregion->swsci = NULL; opregion->asle = NULL; + opregion->asle_ext = NULL; opregion->vbt = NULL; opregion->lid_state = NULL; } diff --git a/drivers/gpu/drm/i915/display/intel_opregion.h b/drivers/gpu/drm/i915/display/intel_opregion.h index 4aa68ffbd30e..82cc0ba34af7 100644 --- a/drivers/gpu/drm/i915/display/intel_opregion.h +++ b/drivers/gpu/drm/i915/display/intel_opregion.h @@ -29,12 +29,14 @@ #include <linux/pci.h> struct drm_i915_private; +struct intel_connector; struct intel_encoder; struct opregion_header; struct opregion_acpi; struct opregion_swsci; struct opregion_asle; +struct opregion_asle_ext; struct intel_opregion { struct opregion_header *header; @@ -43,6 +45,7 @@ struct intel_opregion { u32 swsci_gbda_sub_functions; u32 swsci_sbcb_sub_functions; struct opregion_asle *asle; + struct opregion_asle_ext *asle_ext; void *rvda; void *vbt_firmware; const void *vbt; @@ -71,6 +74,7 @@ int intel_opregion_notify_encoder(struct intel_encoder *intel_encoder, int intel_opregion_notify_adapter(struct drm_i915_private *dev_priv, pci_power_t state); int intel_opregion_get_panel_type(struct drm_i915_private *dev_priv); +struct edid *intel_opregion_get_edid(struct intel_connector *connector); #else /* CONFIG_ACPI*/ @@ -117,6 +121,12 @@ static inline int intel_opregion_get_panel_type(struct drm_i915_private *dev) return -ENODEV; } +static inline struct edid * +intel_opregion_get_edid(struct intel_connector *connector) +{ + return NULL; +} + #endif /* CONFIG_ACPI */ #endif diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c index d610e48cab94..5358f03b52db 100644 --- a/drivers/gpu/drm/i915/display/intel_overlay.c +++ b/drivers/gpu/drm/i915/display/intel_overlay.c @@ -38,6 +38,7 @@ #include "intel_display_types.h" #include "intel_frontbuffer.h" #include "intel_overlay.h" +#include "intel_pci_config.h" /* Limits for overlay size. According to intel doc, the real limits are: * Y width: 4095, UV width (planar): 2047, Y height: 2047, diff --git a/drivers/gpu/drm/i915/display/intel_pch_display.c b/drivers/gpu/drm/i915/display/intel_pch_display.c index a55c4bfacd0d..9192769e3337 100644 --- a/drivers/gpu/drm/i915/display/intel_pch_display.c +++ b/drivers/gpu/drm/i915/display/intel_pch_display.c @@ -88,6 +88,50 @@ static void assert_pch_transcoder_disabled(struct drm_i915_private *dev_priv, pipe_name(pipe)); } +static void intel_pch_transcoder_set_m1_n1(struct intel_crtc *crtc, + const struct intel_link_m_n *m_n) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + + intel_set_m_n(dev_priv, m_n, + PCH_TRANS_DATA_M1(pipe), PCH_TRANS_DATA_N1(pipe), + PCH_TRANS_LINK_M1(pipe), PCH_TRANS_LINK_N1(pipe)); +} + +static void intel_pch_transcoder_set_m2_n2(struct intel_crtc *crtc, + const struct intel_link_m_n *m_n) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + + intel_set_m_n(dev_priv, m_n, + PCH_TRANS_DATA_M2(pipe), PCH_TRANS_DATA_N2(pipe), + PCH_TRANS_LINK_M2(pipe), PCH_TRANS_LINK_N2(pipe)); +} + +void intel_pch_transcoder_get_m1_n1(struct intel_crtc *crtc, + struct intel_link_m_n *m_n) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + + intel_get_m_n(dev_priv, m_n, + PCH_TRANS_DATA_M1(pipe), PCH_TRANS_DATA_N1(pipe), + PCH_TRANS_LINK_M1(pipe), PCH_TRANS_LINK_N1(pipe)); +} + +void intel_pch_transcoder_get_m2_n2(struct intel_crtc *crtc, + struct intel_link_m_n *m_n) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + + intel_get_m_n(dev_priv, m_n, + PCH_TRANS_DATA_M2(pipe), PCH_TRANS_DATA_N2(pipe), + PCH_TRANS_LINK_M2(pipe), PCH_TRANS_LINK_N2(pipe)); +} + static void ilk_pch_transcoder_set_timings(const struct intel_crtc_state *crtc_state, enum pipe pch_transcoder) { @@ -157,20 +201,20 @@ static void ilk_enable_pch_transcoder(const struct intel_crtc_state *crtc_state) */ val &= ~PIPECONF_BPC_MASK; if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) - val |= PIPECONF_8BPC; + val |= PIPECONF_BPC_8; else val |= pipeconf_val & PIPECONF_BPC_MASK; } val &= ~TRANS_INTERLACE_MASK; - if ((pipeconf_val & PIPECONF_INTERLACE_MASK) == PIPECONF_INTERLACED_ILK) { + if ((pipeconf_val & PIPECONF_INTERLACE_MASK_ILK) == PIPECONF_INTERLACE_IF_ID_ILK) { if (HAS_PCH_IBX(dev_priv) && intel_crtc_has_type(crtc_state, INTEL_OUTPUT_SDVO)) - val |= TRANS_LEGACY_INTERLACED_ILK; + val |= TRANS_INTERLACE_LEGACY_VSYNC_IBX; else - val |= TRANS_INTERLACED; + val |= TRANS_INTERLACE_INTERLACED; } else { - val |= TRANS_PROGRESSIVE; + val |= TRANS_INTERLACE_PROGRESSIVE; } intel_de_write(dev_priv, reg, val | TRANS_ENABLE); @@ -211,6 +255,20 @@ static void ilk_disable_pch_transcoder(struct intel_crtc *crtc) } } +void ilk_pch_pre_enable(struct intel_atomic_state *state, + struct intel_crtc *crtc) +{ + const struct intel_crtc_state *crtc_state = + intel_atomic_get_new_crtc_state(state, crtc); + + /* + * Note: FDI PLL enabling _must_ be done before we enable the + * cpu pipes, hence this is separate from all the other fdi/pch + * enabling. + */ + ilk_fdi_pll_enable(crtc_state); +} + /* * Enable PCH resources required for PCH ports: * - PCH PLLs @@ -264,6 +322,10 @@ void ilk_pch_enable(struct intel_atomic_state *state, /* set transcoder timing, panel must allow it */ assert_pps_unlocked(dev_priv, pipe); + if (intel_crtc_has_dp_encoder(crtc_state)) { + intel_pch_transcoder_set_m1_n1(crtc, &crtc_state->dp_m_n); + intel_pch_transcoder_set_m2_n2(crtc, &crtc_state->dp_m2_n2); + } ilk_pch_transcoder_set_timings(crtc_state, pipe); intel_fdi_normal_train(crtc); @@ -279,7 +341,8 @@ void ilk_pch_enable(struct intel_atomic_state *state, temp = intel_de_read(dev_priv, reg); temp &= ~(TRANS_DP_PORT_SEL_MASK | - TRANS_DP_SYNC_MASK | + TRANS_DP_VSYNC_ACTIVE_HIGH | + TRANS_DP_HSYNC_ACTIVE_HIGH | TRANS_DP_BPC_MASK); temp |= TRANS_DP_OUTPUT_ENABLE; temp |= bpc << 9; /* same format but at 11:9 */ @@ -371,7 +434,8 @@ void ilk_pch_get_config(struct intel_crtc_state *crtc_state) crtc_state->fdi_lanes = ((FDI_DP_PORT_WIDTH_MASK & tmp) >> FDI_DP_PORT_WIDTH_SHIFT) + 1; - ilk_get_fdi_m_n_config(crtc, crtc_state); + intel_cpu_transcoder_get_m1_n1(crtc, crtc_state->cpu_transcoder, + &crtc_state->fdi_m_n); if (HAS_PCH_IBX(dev_priv)) { /* @@ -422,11 +486,10 @@ static void lpt_enable_pch_transcoder(struct drm_i915_private *dev_priv, val = TRANS_ENABLE; pipeconf_val = intel_de_read(dev_priv, PIPECONF(cpu_transcoder)); - if ((pipeconf_val & PIPECONF_INTERLACE_MASK_HSW) == - PIPECONF_INTERLACED_ILK) - val |= TRANS_INTERLACED; + if ((pipeconf_val & PIPECONF_INTERLACE_MASK_HSW) == PIPECONF_INTERLACE_IF_ID_ILK) + val |= TRANS_INTERLACE_INTERLACED; else - val |= TRANS_PROGRESSIVE; + val |= TRANS_INTERLACE_PROGRESSIVE; intel_de_write(dev_priv, LPT_TRANSCONF, val); if (intel_de_wait_for_set(dev_priv, LPT_TRANSCONF, @@ -495,7 +558,8 @@ void lpt_pch_get_config(struct intel_crtc_state *crtc_state) crtc_state->fdi_lanes = ((FDI_DP_PORT_WIDTH_MASK & tmp) >> FDI_DP_PORT_WIDTH_SHIFT) + 1; - ilk_get_fdi_m_n_config(crtc, crtc_state); + intel_cpu_transcoder_get_m1_n1(crtc, crtc_state->cpu_transcoder, + &crtc_state->fdi_m_n); crtc_state->hw.adjusted_mode.crtc_clock = lpt_get_iclkip(dev_priv); } diff --git a/drivers/gpu/drm/i915/display/intel_pch_display.h b/drivers/gpu/drm/i915/display/intel_pch_display.h index 2c387fe3a467..749473d99320 100644 --- a/drivers/gpu/drm/i915/display/intel_pch_display.h +++ b/drivers/gpu/drm/i915/display/intel_pch_display.h @@ -9,7 +9,10 @@ struct intel_atomic_state; struct intel_crtc; struct intel_crtc_state; +struct intel_link_m_n; +void ilk_pch_pre_enable(struct intel_atomic_state *state, + struct intel_crtc *crtc); void ilk_pch_enable(struct intel_atomic_state *state, struct intel_crtc *crtc); void ilk_pch_disable(struct intel_atomic_state *state, @@ -24,4 +27,9 @@ void lpt_pch_disable(struct intel_atomic_state *state, struct intel_crtc *crtc); void lpt_pch_get_config(struct intel_crtc_state *crtc_state); +void intel_pch_transcoder_get_m1_n1(struct intel_crtc *crtc, + struct intel_link_m_n *m_n); +void intel_pch_transcoder_get_m2_n2(struct intel_crtc *crtc, + struct intel_link_m_n *m_n); + #endif diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c index e9c679bb1b2e..9c986e8932f8 100644 --- a/drivers/gpu/drm/i915/display/intel_pps.c +++ b/drivers/gpu/drm/i915/display/intel_pps.c @@ -1131,16 +1131,20 @@ intel_pps_readout_hw_state(struct intel_dp *intel_dp, struct edp_power_seq *seq) } static void -intel_pps_dump_state(const char *state_name, const struct edp_power_seq *seq) +intel_pps_dump_state(struct intel_dp *intel_dp, const char *state_name, + const struct edp_power_seq *seq) { - DRM_DEBUG_KMS("%s t1_t3 %d t8 %d t9 %d t10 %d t11_t12 %d\n", - state_name, - seq->t1_t3, seq->t8, seq->t9, seq->t10, seq->t11_t12); + struct drm_i915_private *i915 = dp_to_i915(intel_dp); + + drm_dbg_kms(&i915->drm, "%s t1_t3 %d t8 %d t9 %d t10 %d t11_t12 %d\n", + state_name, + seq->t1_t3, seq->t8, seq->t9, seq->t10, seq->t11_t12); } static void intel_pps_verify_state(struct intel_dp *intel_dp) { + struct drm_i915_private *i915 = dp_to_i915(intel_dp); struct edp_power_seq hw; struct edp_power_seq *sw = &intel_dp->pps.pps_delays; @@ -1148,9 +1152,9 @@ intel_pps_verify_state(struct intel_dp *intel_dp) if (hw.t1_t3 != sw->t1_t3 || hw.t8 != sw->t8 || hw.t9 != sw->t9 || hw.t10 != sw->t10 || hw.t11_t12 != sw->t11_t12) { - DRM_ERROR("PPS state mismatch\n"); - intel_pps_dump_state("sw", sw); - intel_pps_dump_state("hw", &hw); + drm_err(&i915->drm, "PPS state mismatch\n"); + intel_pps_dump_state(intel_dp, "sw", sw); + intel_pps_dump_state(intel_dp, "hw", &hw); } } @@ -1168,7 +1172,7 @@ static void pps_init_delays(struct intel_dp *intel_dp) intel_pps_readout_hw_state(intel_dp, &cur); - intel_pps_dump_state("cur", &cur); + intel_pps_dump_state(intel_dp, "cur", &cur); vbt = dev_priv->vbt.edp.pps; /* On Toshiba Satellite P50-C-18C system the VBT T12 delay @@ -1200,7 +1204,7 @@ static void pps_init_delays(struct intel_dp *intel_dp) * too. */ spec.t11_t12 = (510 + 100) * 10; - intel_pps_dump_state("vbt", &vbt); + intel_pps_dump_state(intel_dp, "vbt", &vbt); /* Use the max of the register settings and vbt. If both are * unset, fall back to the spec limits. */ diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.h b/drivers/gpu/drm/i915/display/intel_sdvo.h index 72065e4360d5..2868852c85f2 100644 --- a/drivers/gpu/drm/i915/display/intel_sdvo.h +++ b/drivers/gpu/drm/i915/display/intel_sdvo.h @@ -8,7 +8,7 @@ #include <linux/types.h> -#include "i915_reg.h" +#include "i915_reg_defs.h" struct drm_i915_private; enum pipe; diff --git a/drivers/gpu/drm/i915/display/intel_snps_phy.c b/drivers/gpu/drm/i915/display/intel_snps_phy.c index 09f405e4d363..8573a458811a 100644 --- a/drivers/gpu/drm/i915/display/intel_snps_phy.c +++ b/drivers/gpu/drm/i915/display/intel_snps_phy.c @@ -10,6 +10,7 @@ #include "intel_de.h" #include "intel_display_types.h" #include "intel_snps_phy.h" +#include "intel_snps_phy_regs.h" /** * DOC: Synopsis PHY support @@ -23,18 +24,18 @@ * since it is not handled by the shared DPLL framework as on other platforms. */ -void intel_snps_phy_wait_for_calibration(struct drm_i915_private *dev_priv) +void intel_snps_phy_wait_for_calibration(struct drm_i915_private *i915) { enum phy phy; for_each_phy_masked(phy, ~0) { - if (!intel_phy_is_snps(dev_priv, phy)) + if (!intel_phy_is_snps(i915, phy)) continue; - if (intel_de_wait_for_clear(dev_priv, ICL_PHY_MISC(phy), + if (intel_de_wait_for_clear(i915, ICL_PHY_MISC(phy), DG2_PHY_DP_TX_ACK_MASK, 25)) - DRM_ERROR("SNPS PHY %c failed to calibrate after 25ms.\n", - phy); + drm_err(&i915->drm, "SNPS PHY %c failed to calibrate after 25ms.\n", + phy); } } @@ -775,6 +776,7 @@ intel_mpllb_tables_get(struct intel_crtc_state *crtc_state, int intel_mpllb_calc_state(struct intel_crtc_state *crtc_state, struct intel_encoder *encoder) { + struct drm_i915_private *i915 = to_i915(encoder->base.dev); const struct intel_mpllb_state * const *tables; int i; @@ -786,8 +788,8 @@ int intel_mpllb_calc_state(struct intel_crtc_state *crtc_state, * until we have a proper algorithm under a valid * license. */ - DRM_DEBUG_KMS("Can't support HDMI link rate %d\n", - crtc_state->port_clock); + drm_dbg_kms(&i915->drm, "Can't support HDMI link rate %d\n", + crtc_state->port_clock); return -EINVAL; } } @@ -854,7 +856,7 @@ void intel_mpllb_enable(struct intel_encoder *encoder, * dp_mpllb_state interface signal. */ if (intel_de_wait_for_set(dev_priv, enable_reg, PLL_LOCK, 5)) - DRM_ERROR("Port %c PLL not locked\n", phy_name(phy)); + drm_dbg_kms(&dev_priv->drm, "Port %c PLL not locked\n", phy_name(phy)); /* * 11. If the frequency will result in a change to the voltage @@ -867,8 +869,8 @@ void intel_mpllb_enable(struct intel_encoder *encoder, void intel_mpllb_disable(struct intel_encoder *encoder) { - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - enum phy phy = intel_port_to_phy(dev_priv, encoder->port); + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(i915, encoder->port); i915_reg_t enable_reg = (phy <= PHY_D ? DG2_PLL_ENABLE(phy) : MG_PLL_ENABLE(0)); @@ -881,21 +883,21 @@ void intel_mpllb_disable(struct intel_encoder *encoder) */ /* 2. Software programs DPLL_ENABLE [PLL Enable] to "0" */ - intel_uncore_rmw(&dev_priv->uncore, enable_reg, PLL_ENABLE, 0); + intel_uncore_rmw(&i915->uncore, enable_reg, PLL_ENABLE, 0); /* * 4. Software programs SNPS_PHY_MPLLB_DIV dp_mpllb_force_en to "0". * This will allow the PLL to stop running. */ - intel_uncore_rmw(&dev_priv->uncore, SNPS_PHY_MPLLB_DIV(phy), + intel_uncore_rmw(&i915->uncore, SNPS_PHY_MPLLB_DIV(phy), SNPS_PHY_MPLLB_FORCE_EN, 0); /* * 5. Software polls DPLL_ENABLE [PLL Lock] for PHY acknowledgment * (dp_txX_ack) that the new transmitter setting request is completed. */ - if (intel_de_wait_for_clear(dev_priv, enable_reg, PLL_LOCK, 5)) - DRM_ERROR("Port %c PLL not locked\n", phy_name(phy)); + if (intel_de_wait_for_clear(i915, enable_reg, PLL_LOCK, 5)) + drm_err(&i915->drm, "Port %c PLL not locked\n", phy_name(phy)); /* * 6. If the frequency will result in a change to the voltage diff --git a/drivers/gpu/drm/i915/display/intel_snps_phy_regs.h b/drivers/gpu/drm/i915/display/intel_snps_phy_regs.h new file mode 100644 index 000000000000..0543465aaf14 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_snps_phy_regs.h @@ -0,0 +1,75 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __INTEL_SNPS_PHY_REGS__ +#define __INTEL_SNPS_PHY_REGS__ + +#include "i915_reg_defs.h" + +#define _SNPS_PHY_A_BASE 0x168000 +#define _SNPS_PHY_B_BASE 0x169000 +#define _SNPS_PHY(phy) _PHY(phy, \ + _SNPS_PHY_A_BASE, \ + _SNPS_PHY_B_BASE) +#define _SNPS2(phy, reg) (_SNPS_PHY(phy) - \ + _SNPS_PHY_A_BASE + (reg)) +#define _MMIO_SNPS(phy, reg) _MMIO(_SNPS2(phy, reg)) +#define _MMIO_SNPS_LN(ln, phy, reg) _MMIO(_SNPS2(phy, \ + (reg) + (ln) * 0x10)) + +#define SNPS_PHY_MPLLB_CP(phy) _MMIO_SNPS(phy, 0x168000) +#define SNPS_PHY_MPLLB_CP_INT REG_GENMASK(31, 25) +#define SNPS_PHY_MPLLB_CP_INT_GS REG_GENMASK(23, 17) +#define SNPS_PHY_MPLLB_CP_PROP REG_GENMASK(15, 9) +#define SNPS_PHY_MPLLB_CP_PROP_GS REG_GENMASK(7, 1) + +#define SNPS_PHY_MPLLB_DIV(phy) _MMIO_SNPS(phy, 0x168004) +#define SNPS_PHY_MPLLB_FORCE_EN REG_BIT(31) +#define SNPS_PHY_MPLLB_DIV_CLK_EN REG_BIT(30) +#define SNPS_PHY_MPLLB_DIV5_CLK_EN REG_BIT(29) +#define SNPS_PHY_MPLLB_V2I REG_GENMASK(27, 26) +#define SNPS_PHY_MPLLB_FREQ_VCO REG_GENMASK(25, 24) +#define SNPS_PHY_MPLLB_DIV_MULTIPLIER REG_GENMASK(23, 16) +#define SNPS_PHY_MPLLB_PMIX_EN REG_BIT(10) +#define SNPS_PHY_MPLLB_DP2_MODE REG_BIT(9) +#define SNPS_PHY_MPLLB_WORD_DIV2_EN REG_BIT(8) +#define SNPS_PHY_MPLLB_TX_CLK_DIV REG_GENMASK(7, 5) +#define SNPS_PHY_MPLLB_SHIM_DIV32_CLK_SEL REG_BIT(0) + +#define SNPS_PHY_MPLLB_FRACN1(phy) _MMIO_SNPS(phy, 0x168008) +#define SNPS_PHY_MPLLB_FRACN_EN REG_BIT(31) +#define SNPS_PHY_MPLLB_FRACN_CGG_UPDATE_EN REG_BIT(30) +#define SNPS_PHY_MPLLB_FRACN_DEN REG_GENMASK(15, 0) + +#define SNPS_PHY_MPLLB_FRACN2(phy) _MMIO_SNPS(phy, 0x16800C) +#define SNPS_PHY_MPLLB_FRACN_REM REG_GENMASK(31, 16) +#define SNPS_PHY_MPLLB_FRACN_QUOT REG_GENMASK(15, 0) + +#define SNPS_PHY_MPLLB_SSCEN(phy) _MMIO_SNPS(phy, 0x168014) +#define SNPS_PHY_MPLLB_SSC_EN REG_BIT(31) +#define SNPS_PHY_MPLLB_SSC_UP_SPREAD REG_BIT(30) +#define SNPS_PHY_MPLLB_SSC_PEAK REG_GENMASK(29, 10) + +#define SNPS_PHY_MPLLB_SSCSTEP(phy) _MMIO_SNPS(phy, 0x168018) +#define SNPS_PHY_MPLLB_SSC_STEPSIZE REG_GENMASK(31, 11) + +#define SNPS_PHY_MPLLB_DIV2(phy) _MMIO_SNPS(phy, 0x16801C) +#define SNPS_PHY_MPLLB_HDMI_PIXEL_CLK_DIV REG_GENMASK(19, 18) +#define SNPS_PHY_MPLLB_HDMI_DIV REG_GENMASK(17, 15) +#define SNPS_PHY_MPLLB_REF_CLK_DIV REG_GENMASK(14, 12) +#define SNPS_PHY_MPLLB_MULTIPLIER REG_GENMASK(11, 0) + +#define SNPS_PHY_REF_CONTROL(phy) _MMIO_SNPS(phy, 0x168188) +#define SNPS_PHY_REF_CONTROL_REF_RANGE REG_GENMASK(31, 27) + +#define SNPS_PHY_TX_REQ(phy) _MMIO_SNPS(phy, 0x168200) +#define SNPS_PHY_TX_REQ_LN_DIS_PWR_STATE_PSR REG_GENMASK(31, 30) + +#define SNPS_PHY_TX_EQ(ln, phy) _MMIO_SNPS_LN(ln, phy, 0x168300) +#define SNPS_PHY_TX_EQ_MAIN REG_GENMASK(23, 18) +#define SNPS_PHY_TX_EQ_POST REG_GENMASK(15, 10) +#define SNPS_PHY_TX_EQ_PRE REG_GENMASK(7, 2) + +#endif /* __INTEL_SNPS_PHY_REGS__ */ diff --git a/drivers/gpu/drm/i915/display/intel_sprite.c b/drivers/gpu/drm/i915/display/intel_sprite.c index 2357a1301f48..2d71294aaceb 100644 --- a/drivers/gpu/drm/i915/display/intel_sprite.c +++ b/drivers/gpu/drm/i915/display/intel_sprite.c @@ -53,6 +53,7 @@ int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state) { + struct drm_i915_private *i915 = to_i915(plane_state->uapi.plane->dev); const struct drm_framebuffer *fb = plane_state->hw.fb; struct drm_rect *src = &plane_state->uapi.src; u32 src_x, src_y, src_w, src_h, hsub, vsub; @@ -94,14 +95,14 @@ int intel_plane_check_src_coordinates(struct intel_plane_state *plane_state) hsub = vsub = max(hsub, vsub); if (src_x % hsub || src_w % hsub) { - DRM_DEBUG_KMS("src x/w (%u, %u) must be a multiple of %u (rotated: %s)\n", - src_x, src_w, hsub, yesno(rotated)); + drm_dbg_kms(&i915->drm, "src x/w (%u, %u) must be a multiple of %u (rotated: %s)\n", + src_x, src_w, hsub, yesno(rotated)); return -EINVAL; } if (src_y % vsub || src_h % vsub) { - DRM_DEBUG_KMS("src y/h (%u, %u) must be a multiple of %u (rotated: %s)\n", - src_y, src_h, vsub, yesno(rotated)); + drm_dbg_kms(&i915->drm, "src y/h (%u, %u) must be a multiple of %u (rotated: %s)\n", + src_y, src_h, vsub, yesno(rotated)); return -EINVAL; } @@ -313,7 +314,7 @@ static u32 vlv_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state) u32 sprctl = 0; if (crtc_state->gamma_enable) - sprctl |= SP_GAMMA_ENABLE; + sprctl |= SP_PIPE_GAMMA_ENABLE; return sprctl; } @@ -436,9 +437,9 @@ vlv_sprite_update_noarm(struct intel_plane *plane, intel_de_write_fw(dev_priv, SPSTRIDE(pipe, plane_id), plane_state->view.color_plane[0].mapping_stride); intel_de_write_fw(dev_priv, SPPOS(pipe, plane_id), - (crtc_y << 16) | crtc_x); + SP_POS_Y(crtc_y) | SP_POS_X(crtc_x)); intel_de_write_fw(dev_priv, SPSIZE(pipe, plane_id), - ((crtc_h - 1) << 16) | (crtc_w - 1)); + SP_HEIGHT(crtc_h - 1) | SP_WIDTH(crtc_w - 1)); spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); } @@ -479,7 +480,8 @@ vlv_sprite_update_arm(struct intel_plane *plane, intel_de_write_fw(dev_priv, SPCONSTALPHA(pipe, plane_id), 0); intel_de_write_fw(dev_priv, SPLINOFF(pipe, plane_id), linear_offset); - intel_de_write_fw(dev_priv, SPTILEOFF(pipe, plane_id), (y << 16) | x); + intel_de_write_fw(dev_priv, SPTILEOFF(pipe, plane_id), + SP_OFFSET_Y(y) | SP_OFFSET_X(x)); /* * The control register self-arms if the plane was previously @@ -700,7 +702,7 @@ static u32 ivb_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state) u32 sprctl = 0; if (crtc_state->gamma_enable) - sprctl |= SPRITE_GAMMA_ENABLE; + sprctl |= SPRITE_PIPE_GAMMA_ENABLE; if (crtc_state->csc_enable) sprctl |= SPRITE_PIPE_CSC_ENABLE; @@ -770,7 +772,7 @@ static u32 ivb_sprite_ctl(const struct intel_crtc_state *crtc_state, } if (!ivb_need_sprite_gamma(plane_state)) - sprctl |= SPRITE_INT_GAMMA_DISABLE; + sprctl |= SPRITE_PLANE_GAMMA_DISABLE; if (plane_state->hw.color_encoding == DRM_COLOR_YCBCR_BT709) sprctl |= SPRITE_YUV_TO_RGB_CSC_FORMAT_BT709; @@ -863,14 +865,18 @@ ivb_sprite_update_noarm(struct intel_plane *plane, unsigned long irqflags; if (crtc_w != src_w || crtc_h != src_h) - sprscale = SPRITE_SCALE_ENABLE | ((src_w - 1) << 16) | (src_h - 1); + sprscale = SPRITE_SCALE_ENABLE | + SPRITE_SRC_WIDTH(src_w - 1) | + SPRITE_SRC_HEIGHT(src_h - 1); spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); intel_de_write_fw(dev_priv, SPRSTRIDE(pipe), plane_state->view.color_plane[0].mapping_stride); - intel_de_write_fw(dev_priv, SPRPOS(pipe), (crtc_y << 16) | crtc_x); - intel_de_write_fw(dev_priv, SPRSIZE(pipe), ((crtc_h - 1) << 16) | (crtc_w - 1)); + intel_de_write_fw(dev_priv, SPRPOS(pipe), + SPRITE_POS_Y(crtc_y) | SPRITE_POS_X(crtc_x)); + intel_de_write_fw(dev_priv, SPRSIZE(pipe), + SPRITE_HEIGHT(crtc_h - 1) | SPRITE_WIDTH(crtc_w - 1)); if (IS_IVYBRIDGE(dev_priv)) intel_de_write_fw(dev_priv, SPRSCALE(pipe), sprscale); @@ -907,10 +913,12 @@ ivb_sprite_update_arm(struct intel_plane *plane, /* HSW consolidates SPRTILEOFF and SPRLINOFF into a single SPROFFSET * register */ if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) { - intel_de_write_fw(dev_priv, SPROFFSET(pipe), (y << 16) | x); + intel_de_write_fw(dev_priv, SPROFFSET(pipe), + SPRITE_OFFSET_Y(y) | SPRITE_OFFSET_X(x)); } else { intel_de_write_fw(dev_priv, SPRLINOFF(pipe), linear_offset); - intel_de_write_fw(dev_priv, SPRTILEOFF(pipe), (y << 16) | x); + intel_de_write_fw(dev_priv, SPRTILEOFF(pipe), + SPRITE_OFFSET_Y(y) | SPRITE_OFFSET_X(x)); } /* @@ -1047,7 +1055,7 @@ static u32 g4x_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state) u32 dvscntr = 0; if (crtc_state->gamma_enable) - dvscntr |= DVS_GAMMA_ENABLE; + dvscntr |= DVS_PIPE_GAMMA_ENABLE; if (crtc_state->csc_enable) dvscntr |= DVS_PIPE_CSC_ENABLE; @@ -1199,14 +1207,18 @@ g4x_sprite_update_noarm(struct intel_plane *plane, unsigned long irqflags; if (crtc_w != src_w || crtc_h != src_h) - dvsscale = DVS_SCALE_ENABLE | ((src_w - 1) << 16) | (src_h - 1); + dvsscale = DVS_SCALE_ENABLE | + DVS_SRC_WIDTH(src_w - 1) | + DVS_SRC_HEIGHT(src_h - 1); spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); intel_de_write_fw(dev_priv, DVSSTRIDE(pipe), plane_state->view.color_plane[0].mapping_stride); - intel_de_write_fw(dev_priv, DVSPOS(pipe), (crtc_y << 16) | crtc_x); - intel_de_write_fw(dev_priv, DVSSIZE(pipe), ((crtc_h - 1) << 16) | (crtc_w - 1)); + intel_de_write_fw(dev_priv, DVSPOS(pipe), + DVS_POS_Y(crtc_y) | DVS_POS_X(crtc_x)); + intel_de_write_fw(dev_priv, DVSSIZE(pipe), + DVS_HEIGHT(crtc_h - 1) | DVS_WIDTH(crtc_w - 1)); intel_de_write_fw(dev_priv, DVSSCALE(pipe), dvsscale); spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); @@ -1321,6 +1333,7 @@ static int g4x_sprite_check_scaling(struct intel_crtc_state *crtc_state, struct intel_plane_state *plane_state) { + struct drm_i915_private *i915 = to_i915(plane_state->uapi.plane->dev); const struct drm_framebuffer *fb = plane_state->hw.fb; const struct drm_rect *src = &plane_state->uapi.src; const struct drm_rect *dst = &plane_state->uapi.dst; @@ -1346,7 +1359,7 @@ g4x_sprite_check_scaling(struct intel_crtc_state *crtc_state, if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) { if (src_h & 1) { - DRM_DEBUG_KMS("Source height must be even with interlaced modes\n"); + drm_dbg_kms(&i915->drm, "Source height must be even with interlaced modes\n"); return -EINVAL; } min_height = 6; @@ -1358,20 +1371,20 @@ g4x_sprite_check_scaling(struct intel_crtc_state *crtc_state, if (src_w < min_width || src_h < min_height || src_w > 2048 || src_h > 2048) { - DRM_DEBUG_KMS("Source dimensions (%dx%d) exceed hardware limits (%dx%d - %dx%d)\n", - src_w, src_h, min_width, min_height, 2048, 2048); + drm_dbg_kms(&i915->drm, "Source dimensions (%dx%d) exceed hardware limits (%dx%d - %dx%d)\n", + src_w, src_h, min_width, min_height, 2048, 2048); return -EINVAL; } if (width_bytes > 4096) { - DRM_DEBUG_KMS("Fetch width (%d) exceeds hardware max with scaling (%u)\n", - width_bytes, 4096); + drm_dbg_kms(&i915->drm, "Fetch width (%d) exceeds hardware max with scaling (%u)\n", + width_bytes, 4096); return -EINVAL; } if (stride > 4096) { - DRM_DEBUG_KMS("Stride (%u) exceeds hardware max with scaling (%u)\n", - stride, 4096); + drm_dbg_kms(&i915->drm, "Stride (%u) exceeds hardware max with scaling (%u)\n", + stride, 4096); return -EINVAL; } diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c index dbd7d0d83a14..feead08ddf8f 100644 --- a/drivers/gpu/drm/i915/display/intel_tc.c +++ b/drivers/gpu/drm/i915/display/intel_tc.c @@ -4,10 +4,12 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "intel_display.h" #include "intel_display_types.h" #include "intel_dp_mst.h" #include "intel_tc.h" +#include "intel_tc_phy_regs.h" static const char *tc_port_mode_name(enum tc_port_mode mode) { diff --git a/drivers/gpu/drm/i915/display/intel_tc_phy_regs.h b/drivers/gpu/drm/i915/display/intel_tc_phy_regs.h new file mode 100644 index 000000000000..5a545086f959 --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_tc_phy_regs.h @@ -0,0 +1,280 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __INTEL_TC_PHY_REGS__ +#define __INTEL_TC_PHY_REGS__ + +#include "i915_reg_defs.h" + +#define MG_PHY_PORT_LN(ln, tc_port, ln0p1, ln0p2, ln1p1) \ + _MMIO(_PORT(tc_port, ln0p1, ln0p2) + (ln) * ((ln1p1) - (ln0p1))) + +#define MG_TX_LINK_PARAMS_TX1LN0_PORT1 0x16812C +#define MG_TX_LINK_PARAMS_TX1LN1_PORT1 0x16852C +#define MG_TX_LINK_PARAMS_TX1LN0_PORT2 0x16912C +#define MG_TX_LINK_PARAMS_TX1LN1_PORT2 0x16952C +#define MG_TX1_LINK_PARAMS(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_LINK_PARAMS_TX1LN0_PORT1, \ + MG_TX_LINK_PARAMS_TX1LN0_PORT2, \ + MG_TX_LINK_PARAMS_TX1LN1_PORT1) + +#define MG_TX_LINK_PARAMS_TX2LN0_PORT1 0x1680AC +#define MG_TX_LINK_PARAMS_TX2LN1_PORT1 0x1684AC +#define MG_TX_LINK_PARAMS_TX2LN0_PORT2 0x1690AC +#define MG_TX_LINK_PARAMS_TX2LN1_PORT2 0x1694AC +#define MG_TX2_LINK_PARAMS(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_LINK_PARAMS_TX2LN0_PORT1, \ + MG_TX_LINK_PARAMS_TX2LN0_PORT2, \ + MG_TX_LINK_PARAMS_TX2LN1_PORT1) +#define CRI_USE_FS32 (1 << 5) + +#define MG_TX_PISO_READLOAD_TX1LN0_PORT1 0x16814C +#define MG_TX_PISO_READLOAD_TX1LN1_PORT1 0x16854C +#define MG_TX_PISO_READLOAD_TX1LN0_PORT2 0x16914C +#define MG_TX_PISO_READLOAD_TX1LN1_PORT2 0x16954C +#define MG_TX1_PISO_READLOAD(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_PISO_READLOAD_TX1LN0_PORT1, \ + MG_TX_PISO_READLOAD_TX1LN0_PORT2, \ + MG_TX_PISO_READLOAD_TX1LN1_PORT1) + +#define MG_TX_PISO_READLOAD_TX2LN0_PORT1 0x1680CC +#define MG_TX_PISO_READLOAD_TX2LN1_PORT1 0x1684CC +#define MG_TX_PISO_READLOAD_TX2LN0_PORT2 0x1690CC +#define MG_TX_PISO_READLOAD_TX2LN1_PORT2 0x1694CC +#define MG_TX2_PISO_READLOAD(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_PISO_READLOAD_TX2LN0_PORT1, \ + MG_TX_PISO_READLOAD_TX2LN0_PORT2, \ + MG_TX_PISO_READLOAD_TX2LN1_PORT1) +#define CRI_CALCINIT (1 << 1) + +#define MG_TX_SWINGCTRL_TX1LN0_PORT1 0x168148 +#define MG_TX_SWINGCTRL_TX1LN1_PORT1 0x168548 +#define MG_TX_SWINGCTRL_TX1LN0_PORT2 0x169148 +#define MG_TX_SWINGCTRL_TX1LN1_PORT2 0x169548 +#define MG_TX1_SWINGCTRL(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_SWINGCTRL_TX1LN0_PORT1, \ + MG_TX_SWINGCTRL_TX1LN0_PORT2, \ + MG_TX_SWINGCTRL_TX1LN1_PORT1) + +#define MG_TX_SWINGCTRL_TX2LN0_PORT1 0x1680C8 +#define MG_TX_SWINGCTRL_TX2LN1_PORT1 0x1684C8 +#define MG_TX_SWINGCTRL_TX2LN0_PORT2 0x1690C8 +#define MG_TX_SWINGCTRL_TX2LN1_PORT2 0x1694C8 +#define MG_TX2_SWINGCTRL(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_SWINGCTRL_TX2LN0_PORT1, \ + MG_TX_SWINGCTRL_TX2LN0_PORT2, \ + MG_TX_SWINGCTRL_TX2LN1_PORT1) +#define CRI_TXDEEMPH_OVERRIDE_17_12(x) ((x) << 0) +#define CRI_TXDEEMPH_OVERRIDE_17_12_MASK (0x3F << 0) + +#define MG_TX_DRVCTRL_TX1LN0_TXPORT1 0x168144 +#define MG_TX_DRVCTRL_TX1LN1_TXPORT1 0x168544 +#define MG_TX_DRVCTRL_TX1LN0_TXPORT2 0x169144 +#define MG_TX_DRVCTRL_TX1LN1_TXPORT2 0x169544 +#define MG_TX_DRVCTRL_TX1LN0_TXPORT3 0x16A144 +#define MG_TX_DRVCTRL_TX1LN1_TXPORT3 0x16A544 +#define MG_TX_DRVCTRL_TX1LN0_TXPORT4 0x16B144 +#define MG_TX_DRVCTRL_TX1LN1_TXPORT4 0x16B544 +#define MG_TX1_DRVCTRL(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_DRVCTRL_TX1LN0_TXPORT1, \ + MG_TX_DRVCTRL_TX1LN0_TXPORT2, \ + MG_TX_DRVCTRL_TX1LN1_TXPORT1) + +#define MG_TX_DRVCTRL_TX2LN0_PORT1 0x1680C4 +#define MG_TX_DRVCTRL_TX2LN1_PORT1 0x1684C4 +#define MG_TX_DRVCTRL_TX2LN0_PORT2 0x1690C4 +#define MG_TX_DRVCTRL_TX2LN1_PORT2 0x1694C4 +#define MG_TX2_DRVCTRL(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_DRVCTRL_TX2LN0_PORT1, \ + MG_TX_DRVCTRL_TX2LN0_PORT2, \ + MG_TX_DRVCTRL_TX2LN1_PORT1) +#define CRI_TXDEEMPH_OVERRIDE_11_6(x) ((x) << 24) +#define CRI_TXDEEMPH_OVERRIDE_11_6_MASK (0x3F << 24) +#define CRI_TXDEEMPH_OVERRIDE_EN (1 << 22) +#define CRI_TXDEEMPH_OVERRIDE_5_0(x) ((x) << 16) +#define CRI_TXDEEMPH_OVERRIDE_5_0_MASK (0x3F << 16) +#define CRI_LOADGEN_SEL(x) ((x) << 12) +#define CRI_LOADGEN_SEL_MASK (0x3 << 12) + +#define MG_CLKHUB_LN0_PORT1 0x16839C +#define MG_CLKHUB_LN1_PORT1 0x16879C +#define MG_CLKHUB_LN0_PORT2 0x16939C +#define MG_CLKHUB_LN1_PORT2 0x16979C +#define MG_CLKHUB(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_CLKHUB_LN0_PORT1, \ + MG_CLKHUB_LN0_PORT2, \ + MG_CLKHUB_LN1_PORT1) +#define CFG_LOW_RATE_LKREN_EN (1 << 11) + +#define MG_TX_DCC_TX1LN0_PORT1 0x168110 +#define MG_TX_DCC_TX1LN1_PORT1 0x168510 +#define MG_TX_DCC_TX1LN0_PORT2 0x169110 +#define MG_TX_DCC_TX1LN1_PORT2 0x169510 +#define MG_TX1_DCC(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_DCC_TX1LN0_PORT1, \ + MG_TX_DCC_TX1LN0_PORT2, \ + MG_TX_DCC_TX1LN1_PORT1) +#define MG_TX_DCC_TX2LN0_PORT1 0x168090 +#define MG_TX_DCC_TX2LN1_PORT1 0x168490 +#define MG_TX_DCC_TX2LN0_PORT2 0x169090 +#define MG_TX_DCC_TX2LN1_PORT2 0x169490 +#define MG_TX2_DCC(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_TX_DCC_TX2LN0_PORT1, \ + MG_TX_DCC_TX2LN0_PORT2, \ + MG_TX_DCC_TX2LN1_PORT1) +#define CFG_AMI_CK_DIV_OVERRIDE_VAL(x) ((x) << 25) +#define CFG_AMI_CK_DIV_OVERRIDE_VAL_MASK (0x3 << 25) +#define CFG_AMI_CK_DIV_OVERRIDE_EN (1 << 24) + +#define MG_DP_MODE_LN0_ACU_PORT1 0x1683A0 +#define MG_DP_MODE_LN1_ACU_PORT1 0x1687A0 +#define MG_DP_MODE_LN0_ACU_PORT2 0x1693A0 +#define MG_DP_MODE_LN1_ACU_PORT2 0x1697A0 +#define MG_DP_MODE(ln, tc_port) \ + MG_PHY_PORT_LN(ln, tc_port, MG_DP_MODE_LN0_ACU_PORT1, \ + MG_DP_MODE_LN0_ACU_PORT2, \ + MG_DP_MODE_LN1_ACU_PORT1) +#define MG_DP_MODE_CFG_DP_X2_MODE (1 << 7) +#define MG_DP_MODE_CFG_DP_X1_MODE (1 << 6) + +#define FIA1_BASE 0x163000 +#define FIA2_BASE 0x16E000 +#define FIA3_BASE 0x16F000 +#define _FIA(fia) _PICK((fia), FIA1_BASE, FIA2_BASE, FIA3_BASE) +#define _MMIO_FIA(fia, off) _MMIO(_FIA(fia) + (off)) + +/* ICL PHY DFLEX registers */ +#define PORT_TX_DFLEXDPMLE1(fia) _MMIO_FIA((fia), 0x008C0) +#define DFLEXDPMLE1_DPMLETC_MASK(idx) (0xf << (4 * (idx))) +#define DFLEXDPMLE1_DPMLETC_ML0(idx) (1 << (4 * (idx))) +#define DFLEXDPMLE1_DPMLETC_ML1_0(idx) (3 << (4 * (idx))) +#define DFLEXDPMLE1_DPMLETC_ML3(idx) (8 << (4 * (idx))) +#define DFLEXDPMLE1_DPMLETC_ML3_2(idx) (12 << (4 * (idx))) +#define DFLEXDPMLE1_DPMLETC_ML3_0(idx) (15 << (4 * (idx))) + +#define _MG_REFCLKIN_CTL_PORT1 0x16892C +#define _MG_REFCLKIN_CTL_PORT2 0x16992C +#define MG_REFCLKIN_CTL_OD_2_MUX(x) ((x) << 8) +#define MG_REFCLKIN_CTL_OD_2_MUX_MASK (0x7 << 8) +#define MG_REFCLKIN_CTL(tc_port) _MMIO_PORT((tc_port), \ + _MG_REFCLKIN_CTL_PORT1, \ + _MG_REFCLKIN_CTL_PORT2) + +#define _MG_CLKTOP2_CORECLKCTL1_PORT1 0x1688D8 +#define _MG_CLKTOP2_CORECLKCTL1_PORT2 0x1698D8 +#define MG_CLKTOP2_CORECLKCTL1_B_DIVRATIO(x) ((x) << 16) +#define MG_CLKTOP2_CORECLKCTL1_B_DIVRATIO_MASK (0xff << 16) +#define MG_CLKTOP2_CORECLKCTL1_A_DIVRATIO(x) ((x) << 8) +#define MG_CLKTOP2_CORECLKCTL1_A_DIVRATIO_MASK (0xff << 8) +#define MG_CLKTOP2_CORECLKCTL1(tc_port) _MMIO_PORT((tc_port), \ + _MG_CLKTOP2_CORECLKCTL1_PORT1, \ + _MG_CLKTOP2_CORECLKCTL1_PORT2) + +#define _MG_CLKTOP2_HSCLKCTL_PORT1 0x1688D4 +#define _MG_CLKTOP2_HSCLKCTL_PORT2 0x1698D4 +#define MG_CLKTOP2_HSCLKCTL_CORE_INPUTSEL(x) ((x) << 16) +#define MG_CLKTOP2_HSCLKCTL_CORE_INPUTSEL_MASK (0x1 << 16) +#define MG_CLKTOP2_HSCLKCTL_TLINEDRV_CLKSEL(x) ((x) << 14) +#define MG_CLKTOP2_HSCLKCTL_TLINEDRV_CLKSEL_MASK (0x3 << 14) +#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_MASK (0x3 << 12) +#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_2 (0 << 12) +#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_3 (1 << 12) +#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_5 (2 << 12) +#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_7 (3 << 12) +#define MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO(x) ((x) << 8) +#define MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_SHIFT 8 +#define MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_MASK (0xf << 8) +#define MG_CLKTOP2_HSCLKCTL(tc_port) _MMIO_PORT((tc_port), \ + _MG_CLKTOP2_HSCLKCTL_PORT1, \ + _MG_CLKTOP2_HSCLKCTL_PORT2) + +#define _MG_PLL_DIV0_PORT1 0x168A00 +#define _MG_PLL_DIV0_PORT2 0x169A00 +#define MG_PLL_DIV0_FRACNEN_H (1 << 30) +#define MG_PLL_DIV0_FBDIV_FRAC_MASK (0x3fffff << 8) +#define MG_PLL_DIV0_FBDIV_FRAC_SHIFT 8 +#define MG_PLL_DIV0_FBDIV_FRAC(x) ((x) << 8) +#define MG_PLL_DIV0_FBDIV_INT_MASK (0xff << 0) +#define MG_PLL_DIV0_FBDIV_INT(x) ((x) << 0) +#define MG_PLL_DIV0(tc_port) _MMIO_PORT((tc_port), _MG_PLL_DIV0_PORT1, \ + _MG_PLL_DIV0_PORT2) + +#define _MG_PLL_DIV1_PORT1 0x168A04 +#define _MG_PLL_DIV1_PORT2 0x169A04 +#define MG_PLL_DIV1_IREF_NDIVRATIO(x) ((x) << 16) +#define MG_PLL_DIV1_DITHER_DIV_1 (0 << 12) +#define MG_PLL_DIV1_DITHER_DIV_2 (1 << 12) +#define MG_PLL_DIV1_DITHER_DIV_4 (2 << 12) +#define MG_PLL_DIV1_DITHER_DIV_8 (3 << 12) +#define MG_PLL_DIV1_NDIVRATIO(x) ((x) << 4) +#define MG_PLL_DIV1_FBPREDIV_MASK (0xf << 0) +#define MG_PLL_DIV1_FBPREDIV(x) ((x) << 0) +#define MG_PLL_DIV1(tc_port) _MMIO_PORT((tc_port), _MG_PLL_DIV1_PORT1, \ + _MG_PLL_DIV1_PORT2) + +#define _MG_PLL_LF_PORT1 0x168A08 +#define _MG_PLL_LF_PORT2 0x169A08 +#define MG_PLL_LF_TDCTARGETCNT(x) ((x) << 24) +#define MG_PLL_LF_AFCCNTSEL_256 (0 << 20) +#define MG_PLL_LF_AFCCNTSEL_512 (1 << 20) +#define MG_PLL_LF_GAINCTRL(x) ((x) << 16) +#define MG_PLL_LF_INT_COEFF(x) ((x) << 8) +#define MG_PLL_LF_PROP_COEFF(x) ((x) << 0) +#define MG_PLL_LF(tc_port) _MMIO_PORT((tc_port), _MG_PLL_LF_PORT1, \ + _MG_PLL_LF_PORT2) + +#define _MG_PLL_FRAC_LOCK_PORT1 0x168A0C +#define _MG_PLL_FRAC_LOCK_PORT2 0x169A0C +#define MG_PLL_FRAC_LOCK_TRUELOCK_CRIT_32 (1 << 18) +#define MG_PLL_FRAC_LOCK_EARLYLOCK_CRIT_32 (1 << 16) +#define MG_PLL_FRAC_LOCK_LOCKTHRESH(x) ((x) << 11) +#define MG_PLL_FRAC_LOCK_DCODITHEREN (1 << 10) +#define MG_PLL_FRAC_LOCK_FEEDFWRDCAL_EN (1 << 8) +#define MG_PLL_FRAC_LOCK_FEEDFWRDGAIN(x) ((x) << 0) +#define MG_PLL_FRAC_LOCK(tc_port) _MMIO_PORT((tc_port), \ + _MG_PLL_FRAC_LOCK_PORT1, \ + _MG_PLL_FRAC_LOCK_PORT2) + +#define _MG_PLL_SSC_PORT1 0x168A10 +#define _MG_PLL_SSC_PORT2 0x169A10 +#define MG_PLL_SSC_EN (1 << 28) +#define MG_PLL_SSC_TYPE(x) ((x) << 26) +#define MG_PLL_SSC_STEPLENGTH(x) ((x) << 16) +#define MG_PLL_SSC_STEPNUM(x) ((x) << 10) +#define MG_PLL_SSC_FLLEN (1 << 9) +#define MG_PLL_SSC_STEPSIZE(x) ((x) << 0) +#define MG_PLL_SSC(tc_port) _MMIO_PORT((tc_port), _MG_PLL_SSC_PORT1, \ + _MG_PLL_SSC_PORT2) + +#define _MG_PLL_BIAS_PORT1 0x168A14 +#define _MG_PLL_BIAS_PORT2 0x169A14 +#define MG_PLL_BIAS_BIAS_GB_SEL(x) ((x) << 30) +#define MG_PLL_BIAS_BIAS_GB_SEL_MASK (0x3 << 30) +#define MG_PLL_BIAS_INIT_DCOAMP(x) ((x) << 24) +#define MG_PLL_BIAS_INIT_DCOAMP_MASK (0x3f << 24) +#define MG_PLL_BIAS_BIAS_BONUS(x) ((x) << 16) +#define MG_PLL_BIAS_BIAS_BONUS_MASK (0xff << 16) +#define MG_PLL_BIAS_BIASCAL_EN (1 << 15) +#define MG_PLL_BIAS_CTRIM(x) ((x) << 8) +#define MG_PLL_BIAS_CTRIM_MASK (0x1f << 8) +#define MG_PLL_BIAS_VREF_RDAC(x) ((x) << 5) +#define MG_PLL_BIAS_VREF_RDAC_MASK (0x7 << 5) +#define MG_PLL_BIAS_IREFTRIM(x) ((x) << 0) +#define MG_PLL_BIAS_IREFTRIM_MASK (0x1f << 0) +#define MG_PLL_BIAS(tc_port) _MMIO_PORT((tc_port), _MG_PLL_BIAS_PORT1, \ + _MG_PLL_BIAS_PORT2) + +#define _MG_PLL_TDC_COLDST_BIAS_PORT1 0x168A18 +#define _MG_PLL_TDC_COLDST_BIAS_PORT2 0x169A18 +#define MG_PLL_TDC_COLDST_IREFINT_EN (1 << 27) +#define MG_PLL_TDC_COLDST_REFBIAS_START_PULSE_W(x) ((x) << 17) +#define MG_PLL_TDC_COLDST_COLDSTART (1 << 16) +#define MG_PLL_TDC_TDCOVCCORR_EN (1 << 2) +#define MG_PLL_TDC_TDCSEL(x) ((x) << 0) +#define MG_PLL_TDC_COLDST_BIAS(tc_port) _MMIO_PORT((tc_port), \ + _MG_PLL_TDC_COLDST_BIAS_PORT1, \ + _MG_PLL_TDC_COLDST_BIAS_PORT2) + +#endif /* __INTEL_TC_PHY_REGS__ */ diff --git a/drivers/gpu/drm/i915/display/intel_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_vbt_defs.h index f043d85ba64d..a39d6cfea87a 100644 --- a/drivers/gpu/drm/i915/display/intel_vbt_defs.h +++ b/drivers/gpu/drm/i915/display/intel_vbt_defs.h @@ -226,32 +226,6 @@ struct bdb_general_features { #define DEVICE_TYPE_DIGITAL_OUTPUT (1 << 1) #define DEVICE_TYPE_ANALOG_OUTPUT (1 << 0) -/* - * Bits we care about when checking for DEVICE_TYPE_eDP. Depending on the - * system, the other bits may or may not be set for eDP outputs. - */ -#define DEVICE_TYPE_eDP_BITS \ - (DEVICE_TYPE_INTERNAL_CONNECTOR | \ - DEVICE_TYPE_MIPI_OUTPUT | \ - DEVICE_TYPE_COMPOSITE_OUTPUT | \ - DEVICE_TYPE_DUAL_CHANNEL | \ - DEVICE_TYPE_LVDS_SIGNALING | \ - DEVICE_TYPE_TMDS_DVI_SIGNALING | \ - DEVICE_TYPE_VIDEO_SIGNALING | \ - DEVICE_TYPE_DISPLAYPORT_OUTPUT | \ - DEVICE_TYPE_ANALOG_OUTPUT) - -#define DEVICE_TYPE_DP_DUAL_MODE_BITS \ - (DEVICE_TYPE_INTERNAL_CONNECTOR | \ - DEVICE_TYPE_MIPI_OUTPUT | \ - DEVICE_TYPE_COMPOSITE_OUTPUT | \ - DEVICE_TYPE_LVDS_SIGNALING | \ - DEVICE_TYPE_TMDS_DVI_SIGNALING | \ - DEVICE_TYPE_VIDEO_SIGNALING | \ - DEVICE_TYPE_DISPLAYPORT_OUTPUT | \ - DEVICE_TYPE_DIGITAL_OUTPUT | \ - DEVICE_TYPE_ANALOG_OUTPUT) - #define DEVICE_CFG_NONE 0x00 #define DEVICE_CFG_12BIT_DVOB 0x01 #define DEVICE_CFG_12BIT_DVOC 0x02 diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c index 9b05f93ed8bc..3faea903b9ae 100644 --- a/drivers/gpu/drm/i915/display/intel_vdsc.c +++ b/drivers/gpu/drm/i915/display/intel_vdsc.c @@ -341,19 +341,14 @@ bool intel_dsc_source_support(const struct intel_crtc_state *crtc_state) const struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *i915 = to_i915(crtc->base.dev); enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; - enum pipe pipe = crtc->pipe; if (!INTEL_INFO(i915)->display.has_dsc) return false; - /* On TGL, DSC is supported on all Pipes */ if (DISPLAY_VER(i915) >= 12) return true; - if (DISPLAY_VER(i915) >= 11 && - (pipe != PIPE_A || cpu_transcoder == TRANSCODER_EDP || - cpu_transcoder == TRANSCODER_DSI_0 || - cpu_transcoder == TRANSCODER_DSI_1)) + if (DISPLAY_VER(i915) >= 11 && cpu_transcoder != TRANSCODER_A) return true; return false; diff --git a/drivers/gpu/drm/i915/display/intel_vga.c b/drivers/gpu/drm/i915/display/intel_vga.c index fa779f7ea415..b5d058404c14 100644 --- a/drivers/gpu/drm/i915/display/intel_vga.c +++ b/drivers/gpu/drm/i915/display/intel_vga.c @@ -7,6 +7,7 @@ #include <linux/vgaarb.h> #include <drm/i915_drm.h> +#include <video/vga.h> #include "i915_drv.h" #include "intel_de.h" @@ -34,9 +35,9 @@ void intel_vga_disable(struct drm_i915_private *dev_priv) /* WaEnableVGAAccessThroughIOPort:ctg,elk,ilk,snb,ivb,vlv,hsw */ vga_get_uninterruptible(pdev, VGA_RSRC_LEGACY_IO); - outb(SR01, VGA_SR_INDEX); - sr1 = inb(VGA_SR_DATA); - outb(sr1 | 1 << 5, VGA_SR_DATA); + outb(0x01, VGA_SEQ_I); + sr1 = inb(VGA_SEQ_D); + outb(sr1 | VGA_SR01_SCREEN_OFF, VGA_SEQ_D); vga_put(pdev, VGA_RSRC_LEGACY_IO); udelay(300); @@ -92,7 +93,7 @@ void intel_vga_reset_io_mem(struct drm_i915_private *i915) * and error messages. */ vga_get_uninterruptible(pdev, VGA_RSRC_LEGACY_IO); - outb(inb(VGA_MSR_READ), VGA_MSR_WRITE); + outb(inb(VGA_MIS_R), VGA_MIS_W); vga_put(pdev, VGA_RSRC_LEGACY_IO); } diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c index 93a385396512..1223075595ff 100644 --- a/drivers/gpu/drm/i915/display/skl_universal_plane.c +++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c @@ -961,6 +961,7 @@ static u32 glk_plane_color_ctl(const struct intel_crtc_state *crtc_state, static u32 skl_surf_address(const struct intel_plane_state *plane_state, int color_plane) { + struct drm_i915_private *i915 = to_i915(plane_state->uapi.plane->dev); const struct drm_framebuffer *fb = plane_state->hw.fb; u32 offset = plane_state->view.color_plane[color_plane].offset; @@ -969,11 +970,11 @@ static u32 skl_surf_address(const struct intel_plane_state *plane_state, * The DPT object contains only one vma, so the VMA's offset * within the DPT is always 0. */ - WARN_ON(plane_state->dpt_vma->node.start); - WARN_ON(offset & 0x1fffff); + drm_WARN_ON(&i915->drm, plane_state->dpt_vma->node.start); + drm_WARN_ON(&i915->drm, offset & 0x1fffff); return offset >> 9; } else { - WARN_ON(offset & 0xfff); + drm_WARN_ON(&i915->drm, offset & 0xfff); return offset; } } @@ -992,6 +993,54 @@ static u32 skl_plane_surf(const struct intel_plane_state *plane_state, return plane_surf; } +static u32 skl_plane_aux_dist(const struct intel_plane_state *plane_state, + int color_plane) +{ + struct drm_i915_private *i915 = to_i915(plane_state->uapi.plane->dev); + const struct drm_framebuffer *fb = plane_state->hw.fb; + int aux_plane = skl_main_to_aux_plane(fb, color_plane); + u32 aux_dist; + + if (!aux_plane) + return 0; + + aux_dist = skl_surf_address(plane_state, aux_plane) - + skl_surf_address(plane_state, color_plane); + + if (DISPLAY_VER(i915) < 12) + aux_dist |= PLANE_AUX_STRIDE(skl_plane_stride(plane_state, aux_plane)); + + return aux_dist; +} + +static u32 skl_plane_keyval(const struct intel_plane_state *plane_state) +{ + const struct drm_intel_sprite_colorkey *key = &plane_state->ckey; + + return key->min_value; +} + +static u32 skl_plane_keymax(const struct intel_plane_state *plane_state) +{ + const struct drm_intel_sprite_colorkey *key = &plane_state->ckey; + u8 alpha = plane_state->hw.alpha >> 8; + + return (key->max_value & 0xffffff) | PLANE_KEYMAX_ALPHA(alpha); +} + +static u32 skl_plane_keymsk(const struct intel_plane_state *plane_state) +{ + const struct drm_intel_sprite_colorkey *key = &plane_state->ckey; + u8 alpha = plane_state->hw.alpha >> 8; + u32 keymsk; + + keymsk = key->channel_mask & 0x7ffffff; + if (alpha < 0xff) + keymsk |= PLANE_KEYMSK_ALPHA_ENABLE; + + return keymsk; +} + static void icl_plane_csc_load_black(struct intel_plane *plane) { struct drm_i915_private *i915 = to_i915(plane->base.dev); @@ -1016,15 +1065,24 @@ static void icl_plane_csc_load_black(struct intel_plane *plane) intel_de_write_fw(i915, PLANE_CSC_POSTOFF(pipe, plane_id, 2), 0); } +static int skl_plane_color_plane(const struct intel_plane_state *plane_state) +{ + /* Program the UV plane on planar master */ + if (plane_state->planar_linked_plane && !plane_state->planar_slave) + return 1; + else + return 0; +} + static void -skl_program_plane_noarm(struct intel_plane *plane, - const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state, - int color_plane) +skl_plane_update_noarm(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state) { struct drm_i915_private *dev_priv = to_i915(plane->base.dev); enum plane_id plane_id = plane->id; enum pipe pipe = plane->pipe; + int color_plane = skl_plane_color_plane(plane_state); u32 stride = skl_plane_stride(plane_state, color_plane); const struct drm_framebuffer *fb = plane_state->hw.fb; int crtc_x = plane_state->uapi.dst.x1; @@ -1048,11 +1106,12 @@ skl_program_plane_noarm(struct intel_plane *plane, if (plane_state->force_black) icl_plane_csc_load_black(plane); - intel_de_write_fw(dev_priv, PLANE_STRIDE(pipe, plane_id), stride); + intel_de_write_fw(dev_priv, PLANE_STRIDE(pipe, plane_id), + PLANE_STRIDE_(stride)); intel_de_write_fw(dev_priv, PLANE_POS(pipe, plane_id), - (crtc_y << 16) | crtc_x); + PLANE_POS_Y(crtc_y) | PLANE_POS_X(crtc_x)); intel_de_write_fw(dev_priv, PLANE_SIZE(pipe, plane_id), - ((src_h - 1) << 16) | (src_w - 1)); + PLANE_HEIGHT(src_h - 1) | PLANE_WIDTH(src_w - 1)); if (intel_fb_is_rc_ccs_cc_modifier(fb->modifier)) { intel_de_write_fw(dev_priv, PLANE_CC_VAL(pipe, plane_id, 0), @@ -1076,21 +1135,17 @@ skl_program_plane_noarm(struct intel_plane *plane, } static void -skl_program_plane_arm(struct intel_plane *plane, - const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state, - int color_plane) +skl_plane_update_arm(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state) { struct drm_i915_private *dev_priv = to_i915(plane->base.dev); enum plane_id plane_id = plane->id; enum pipe pipe = plane->pipe; - const struct drm_intel_sprite_colorkey *key = &plane_state->ckey; - const struct drm_framebuffer *fb = plane_state->hw.fb; - int aux_plane = skl_main_to_aux_plane(fb, color_plane); + int color_plane = skl_plane_color_plane(plane_state); u32 x = plane_state->view.color_plane[color_plane].x; u32 y = plane_state->view.color_plane[color_plane].y; - u32 keymsk, keymax, aux_dist = 0, plane_color_ctl = 0; - u8 alpha = plane_state->hw.alpha >> 8; + u32 plane_color_ctl = 0; u32 plane_ctl = plane_state->ctl; unsigned long irqflags; @@ -1100,36 +1155,22 @@ skl_program_plane_arm(struct intel_plane *plane, plane_color_ctl = plane_state->color_ctl | glk_plane_color_ctl_crtc(crtc_state); - keymax = (key->max_value & 0xffffff) | PLANE_KEYMAX_ALPHA(alpha); - - keymsk = key->channel_mask & 0x7ffffff; - if (alpha < 0xff) - keymsk |= PLANE_KEYMSK_ALPHA_ENABLE; - - if (aux_plane) { - aux_dist = skl_surf_address(plane_state, aux_plane) - - skl_surf_address(plane_state, color_plane); - - if (DISPLAY_VER(dev_priv) < 12) - aux_dist |= skl_plane_stride(plane_state, aux_plane); - } - spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); - intel_de_write_fw(dev_priv, PLANE_KEYVAL(pipe, plane_id), - key->min_value); - intel_de_write_fw(dev_priv, PLANE_KEYMSK(pipe, plane_id), keymsk); - intel_de_write_fw(dev_priv, PLANE_KEYMAX(pipe, plane_id), keymax); + intel_de_write_fw(dev_priv, PLANE_KEYVAL(pipe, plane_id), skl_plane_keyval(plane_state)); + intel_de_write_fw(dev_priv, PLANE_KEYMSK(pipe, plane_id), skl_plane_keymsk(plane_state)); + intel_de_write_fw(dev_priv, PLANE_KEYMAX(pipe, plane_id), skl_plane_keymax(plane_state)); intel_de_write_fw(dev_priv, PLANE_OFFSET(pipe, plane_id), - (y << 16) | x); + PLANE_OFFSET_Y(y) | PLANE_OFFSET_X(x)); - intel_de_write_fw(dev_priv, PLANE_AUX_DIST(pipe, plane_id), aux_dist); + intel_de_write_fw(dev_priv, PLANE_AUX_DIST(pipe, plane_id), + skl_plane_aux_dist(plane_state, color_plane)); if (DISPLAY_VER(dev_priv) < 11) intel_de_write_fw(dev_priv, PLANE_AUX_OFFSET(pipe, plane_id), - (plane_state->view.color_plane[1].y << 16) | - plane_state->view.color_plane[1].x); + PLANE_OFFSET_Y(plane_state->view.color_plane[1].y) | + PLANE_OFFSET_X(plane_state->view.color_plane[1].x)); if (DISPLAY_VER(dev_priv) >= 10) intel_de_write_fw(dev_priv, PLANE_COLOR_CTL(pipe, plane_id), plane_color_ctl); @@ -1182,34 +1223,6 @@ skl_plane_async_flip(struct intel_plane *plane, spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); } -static void -skl_plane_update_noarm(struct intel_plane *plane, - const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) -{ - int color_plane = 0; - - if (plane_state->planar_linked_plane && !plane_state->planar_slave) - /* Program the UV plane on planar master */ - color_plane = 1; - - skl_program_plane_noarm(plane, crtc_state, plane_state, color_plane); -} - -static void -skl_plane_update_arm(struct intel_plane *plane, - const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) -{ - int color_plane = 0; - - if (plane_state->planar_linked_plane && !plane_state->planar_slave) - /* Program the UV plane on planar master */ - color_plane = 1; - - skl_program_plane_arm(plane, crtc_state, plane_state, color_plane); -} - static bool intel_format_is_p01x(u32 format) { switch (format) { @@ -1338,6 +1351,7 @@ static int skl_plane_check_dst_coordinates(const struct intel_crtc_state *crtc_s static int skl_plane_check_nv12_rotation(const struct intel_plane_state *plane_state) { + struct drm_i915_private *i915 = to_i915(plane_state->uapi.plane->dev); const struct drm_framebuffer *fb = plane_state->hw.fb; unsigned int rotation = plane_state->hw.rotation; int src_w = drm_rect_width(&plane_state->uapi.src) >> 16; @@ -1347,7 +1361,7 @@ static int skl_plane_check_nv12_rotation(const struct intel_plane_state *plane_s src_w & 3 && (rotation == DRM_MODE_ROTATE_270 || rotation == (DRM_MODE_REFLECT_X | DRM_MODE_ROTATE_90))) { - DRM_DEBUG_KMS("src width must be multiple of 4 for rotated planar YUV\n"); + drm_dbg_kms(&i915->drm, "src width must be multiple of 4 for rotated planar YUV\n"); return -EINVAL; } @@ -1816,20 +1830,27 @@ static int skl_plane_check(struct intel_crtc_state *crtc_state, return 0; } +static enum intel_fbc_id skl_fbc_id_for_pipe(enum pipe pipe) +{ + return pipe - PIPE_A + INTEL_FBC_A; +} + static bool skl_plane_has_fbc(struct drm_i915_private *dev_priv, - enum pipe pipe, enum plane_id plane_id) + enum intel_fbc_id fbc_id, enum plane_id plane_id) { - if (!HAS_FBC(dev_priv)) + if ((INTEL_INFO(dev_priv)->display.fbc_mask & BIT(fbc_id)) == 0) return false; - return pipe == PIPE_A && plane_id == PLANE_PRIMARY; + return plane_id == PLANE_PRIMARY; } static struct intel_fbc *skl_plane_fbc(struct drm_i915_private *dev_priv, enum pipe pipe, enum plane_id plane_id) { - if (skl_plane_has_fbc(dev_priv, pipe, plane_id)) - return dev_priv->fbc; + enum intel_fbc_id fbc_id = skl_fbc_id_for_pipe(pipe); + + if (skl_plane_has_fbc(dev_priv, fbc_id, plane_id)) + return dev_priv->fbc[fbc_id]; else return NULL; } @@ -2282,16 +2303,17 @@ skl_get_initial_plane_config(struct intel_crtc *crtc, val = intel_de_read(dev_priv, PLANE_CTL(pipe, plane_id)); if (DISPLAY_VER(dev_priv) >= 11) - pixel_format = val & ICL_PLANE_CTL_FORMAT_MASK; + pixel_format = val & PLANE_CTL_FORMAT_MASK_ICL; else - pixel_format = val & PLANE_CTL_FORMAT_MASK; + pixel_format = val & PLANE_CTL_FORMAT_MASK_SKL; if (DISPLAY_VER(dev_priv) >= 10) { - alpha = intel_de_read(dev_priv, - PLANE_COLOR_CTL(pipe, plane_id)); - alpha &= PLANE_COLOR_ALPHA_MASK; + u32 color_ctl; + + color_ctl = intel_de_read(dev_priv, PLANE_COLOR_CTL(pipe, plane_id)); + alpha = REG_FIELD_GET(PLANE_COLOR_ALPHA_MASK, color_ctl); } else { - alpha = val & PLANE_CTL_ALPHA_MASK; + alpha = REG_FIELD_GET(PLANE_CTL_ALPHA_MASK, val); } fourcc = skl_format_to_fourcc(pixel_format, @@ -2355,22 +2377,19 @@ skl_get_initial_plane_config(struct intel_crtc *crtc, if (drm_rotation_90_or_270(plane_config->rotation)) goto error; - base = intel_de_read(dev_priv, PLANE_SURF(pipe, plane_id)) & 0xfffff000; + base = intel_de_read(dev_priv, PLANE_SURF(pipe, plane_id)) & PLANE_SURF_ADDR_MASK; plane_config->base = base; offset = intel_de_read(dev_priv, PLANE_OFFSET(pipe, plane_id)); val = intel_de_read(dev_priv, PLANE_SIZE(pipe, plane_id)); - fb->height = ((val >> 16) & 0xffff) + 1; - fb->width = ((val >> 0) & 0xffff) + 1; + fb->height = REG_FIELD_GET(PLANE_HEIGHT_MASK, val) + 1; + fb->width = REG_FIELD_GET(PLANE_WIDTH_MASK, val) + 1; val = intel_de_read(dev_priv, PLANE_STRIDE(pipe, plane_id)); stride_mult = skl_plane_stride_mult(fb, 0, DRM_MODE_ROTATE_0); - if (DISPLAY_VER(dev_priv) >= 13) - fb->pitches[0] = (val & PLANE_STRIDE_MASK_XELPD) * stride_mult; - else - fb->pitches[0] = (val & PLANE_STRIDE_MASK) * stride_mult; + fb->pitches[0] = REG_FIELD_GET(PLANE_STRIDE__MASK, val) * stride_mult; aligned_height = intel_fb_align_height(fb, 0, fb->height); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c index 1b526039a60d..14fdb0796c52 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_dmabuf.c @@ -74,7 +74,8 @@ static void i915_gem_unmap_dma_buf(struct dma_buf_attachment *attachment, kfree(sg); } -static int i915_gem_dmabuf_vmap(struct dma_buf *dma_buf, struct dma_buf_map *map) +static int i915_gem_dmabuf_vmap(struct dma_buf *dma_buf, + struct iosys_map *map) { struct drm_i915_gem_object *obj = dma_buf_to_obj(dma_buf); void *vaddr; @@ -83,12 +84,13 @@ static int i915_gem_dmabuf_vmap(struct dma_buf *dma_buf, struct dma_buf_map *map if (IS_ERR(vaddr)) return PTR_ERR(vaddr); - dma_buf_map_set_vaddr(map, vaddr); + iosys_map_set_vaddr(map, vaddr); return 0; } -static void i915_gem_dmabuf_vunmap(struct dma_buf *dma_buf, struct dma_buf_map *map) +static void i915_gem_dmabuf_vunmap(struct dma_buf *dma_buf, + struct iosys_map *map) { struct drm_i915_gem_object *obj = dma_buf_to_obj(dma_buf); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index 1736efa43339..54cae821513b 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -23,9 +23,11 @@ #include "pxp/intel_pxp.h" +#include "i915_cmd_parser.h" #include "i915_drv.h" #include "i915_gem_clflush.h" #include "i915_gem_context.h" +#include "i915_gem_evict.h" #include "i915_gem_ioctls.h" #include "i915_trace.h" #include "i915_user_extensions.h" diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c index 7df50fd6cc7b..4dfed34191c6 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c @@ -14,6 +14,7 @@ #include "gem/i915_gem_region.h" #include "i915_drv.h" #include "i915_gem_stolen.h" +#include "i915_reg.h" #include "i915_vgpu.h" /* diff --git a/drivers/gpu/drm/i915/gem/i915_gem_tiling.c b/drivers/gpu/drm/i915/gem/i915_gem_tiling.c index ef4d0f7dc118..0e0e4805161a 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_tiling.c @@ -12,6 +12,8 @@ #include "i915_gem_ioctls.h" #include "i915_gem_mman.h" #include "i915_gem_object.h" +#include "i915_gem_tiling.h" +#include "i915_reg.h" /** * DOC: buffer object tiling diff --git a/drivers/gpu/drm/i915/gem/i915_gem_tiling.h b/drivers/gpu/drm/i915/gem/i915_gem_tiling.h new file mode 100644 index 000000000000..9924196a8139 --- /dev/null +++ b/drivers/gpu/drm/i915/gem/i915_gem_tiling.h @@ -0,0 +1,18 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __I915_GEM_TILING_H__ +#define __I915_GEM_TILING_H__ + +#include <linux/types.h> + +struct drm_i915_private; + +u32 i915_gem_fence_size(struct drm_i915_private *i915, u32 size, + unsigned int tiling, unsigned int stride); +u32 i915_gem_fence_alignment(struct drm_i915_private *i915, u32 size, + unsigned int tiling, unsigned int stride); + +#endif /* __I915_GEM_TILING_H__ */ diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c index 3cc01c30dd62..6d1a71d6404c 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c @@ -42,6 +42,7 @@ #include "i915_drv.h" #include "i915_gem_ioctls.h" #include "i915_gem_object.h" +#include "i915_gem_userptr.h" #include "i915_scatterlist.h" #ifdef CONFIG_MMU_NOTIFIER diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.h b/drivers/gpu/drm/i915/gem/i915_gem_userptr.h new file mode 100644 index 000000000000..8dadb2f8436d --- /dev/null +++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2021 Intel Corporation + */ + +#ifndef __I915_GEM_USERPTR_H__ +#define __I915_GEM_USERPTR_H__ + +struct drm_i915_private; + +int i915_gem_init_userptr(struct drm_i915_private *dev_priv); +void i915_gem_cleanup_userptr(struct drm_i915_private *dev_priv); + +#endif /* __I915_GEM_USERPTR_H__ */ diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c index 75947e9dada2..0be86ffb7c19 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c @@ -7,8 +7,9 @@ #include "gt/intel_context.h" #include "gt/intel_engine_user.h" -#include "gt/intel_gt.h" #include "gt/intel_gpu_commands.h" +#include "gt/intel_gt.h" +#include "gt/intel_gt_regs.h" #include "gem/i915_gem_lmem.h" #include "selftests/igt_flush_test.h" diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c index 3f41fe5ec9d4..7d327ffd0464 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c @@ -8,6 +8,7 @@ #include "gem/i915_gem_pm.h" #include "gt/intel_engine_pm.h" +#include "gt/intel_engine_regs.h" #include "gt/intel_gt.h" #include "gt/intel_gt_requests.h" #include "gt/intel_reset.h" @@ -894,7 +895,7 @@ static int rpcs_query_batch(struct drm_i915_gem_object *rpcs, struct i915_vma *v return PTR_ERR(cmd); *cmd++ = MI_STORE_REGISTER_MEM_GEN8; - *cmd++ = i915_mmio_reg_offset(GEN8_R_PWR_CLK_STATE); + *cmd++ = i915_mmio_reg_offset(GEN8_R_PWR_CLK_STATE(RENDER_RING_BASE)); *cmd++ = lower_32_bits(vma->node.start); *cmd++ = upper_32_bits(vma->node.start); *cmd = MI_BATCH_BUFFER_END; diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c index 3cc74b0fed06..b071a58dd6da 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_dmabuf.c @@ -266,7 +266,7 @@ static int igt_dmabuf_import(void *arg) struct drm_i915_gem_object *obj; struct dma_buf *dmabuf; void *obj_map, *dma_map; - struct dma_buf_map map; + struct iosys_map map; u32 pattern[] = { 0, 0xaa, 0xcc, 0x55, 0xff }; int err, i; @@ -349,7 +349,7 @@ static int igt_dmabuf_import_ownership(void *arg) struct drm_i915_private *i915 = arg; struct drm_i915_gem_object *obj; struct dma_buf *dmabuf; - struct dma_buf_map map; + struct iosys_map map; void *ptr; int err; @@ -400,7 +400,7 @@ static int igt_dmabuf_export_vmap(void *arg) struct drm_i915_private *i915 = arg; struct drm_i915_gem_object *obj; struct dma_buf *dmabuf; - struct dma_buf_map map; + struct iosys_map map; void *ptr; int err; diff --git a/drivers/gpu/drm/i915/gem/selftests/mock_dmabuf.c b/drivers/gpu/drm/i915/gem/selftests/mock_dmabuf.c index 2855d11c7a51..b2a5882b8f81 100644 --- a/drivers/gpu/drm/i915/gem/selftests/mock_dmabuf.c +++ b/drivers/gpu/drm/i915/gem/selftests/mock_dmabuf.c @@ -61,7 +61,7 @@ static void mock_dmabuf_release(struct dma_buf *dma_buf) kfree(mock); } -static int mock_dmabuf_vmap(struct dma_buf *dma_buf, struct dma_buf_map *map) +static int mock_dmabuf_vmap(struct dma_buf *dma_buf, struct iosys_map *map) { struct mock_dmabuf *mock = to_mock(dma_buf); void *vaddr; @@ -69,12 +69,12 @@ static int mock_dmabuf_vmap(struct dma_buf *dma_buf, struct dma_buf_map *map) vaddr = vm_map_ram(mock->pages, mock->npages, 0); if (!vaddr) return -ENOMEM; - dma_buf_map_set_vaddr(map, vaddr); + iosys_map_set_vaddr(map, vaddr); return 0; } -static void mock_dmabuf_vunmap(struct dma_buf *dma_buf, struct dma_buf_map *map) +static void mock_dmabuf_vunmap(struct dma_buf *dma_buf, struct iosys_map *map) { struct mock_dmabuf *mock = to_mock(dma_buf); diff --git a/drivers/gpu/drm/i915/gt/gen2_engine_cs.c b/drivers/gpu/drm/i915/gt/gen2_engine_cs.c index 61383830505e..1c82caf525c3 100644 --- a/drivers/gpu/drm/i915/gt/gen2_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/gen2_engine_cs.c @@ -5,7 +5,9 @@ #include "gen2_engine_cs.h" #include "i915_drv.h" +#include "i915_reg.h" #include "intel_engine.h" +#include "intel_engine_regs.h" #include "intel_gpu_commands.h" #include "intel_gt.h" #include "intel_gt_irq.h" diff --git a/drivers/gpu/drm/i915/gt/gen6_engine_cs.c b/drivers/gpu/drm/i915/gt/gen6_engine_cs.c index b388ceeeb1c9..5e65550b4dfb 100644 --- a/drivers/gpu/drm/i915/gt/gen6_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/gen6_engine_cs.c @@ -5,6 +5,7 @@ #include "gen6_engine_cs.h" #include "intel_engine.h" +#include "intel_engine_regs.h" #include "intel_gpu_commands.h" #include "intel_gt.h" #include "intel_gt_irq.h" diff --git a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c index 6e9292918bfc..8471758b3ef4 100644 --- a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c +++ b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c @@ -9,6 +9,8 @@ #include "i915_scatterlist.h" #include "i915_trace.h" #include "i915_vgpu.h" +#include "intel_gt_regs.h" +#include "intel_engine_regs.h" #include "intel_gt.h" /* Write pde (index) from the page directory @pd to the page table @pt */ diff --git a/drivers/gpu/drm/i915/gt/gen7_renderclear.c b/drivers/gpu/drm/i915/gt/gen7_renderclear.c index 21f08e53889c..317efb145787 100644 --- a/drivers/gpu/drm/i915/gt/gen7_renderclear.c +++ b/drivers/gpu/drm/i915/gt/gen7_renderclear.c @@ -6,6 +6,7 @@ #include "gen7_renderclear.h" #include "i915_drv.h" #include "intel_gpu_commands.h" +#include "intel_gt_regs.h" #define GT3_INLINE_DATA_DELAYS 0x1E00 #define batch_advance(Y, CS) GEM_BUG_ON((Y)->end != (CS)) diff --git a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c index e320610dd0b8..1f8cf4f790b2 100644 --- a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c @@ -5,8 +5,9 @@ #include "gen8_engine_cs.h" #include "i915_drv.h" -#include "intel_lrc.h" #include "intel_gpu_commands.h" +#include "intel_gt_regs.h" +#include "intel_lrc.h" #include "intel_ring.h" int gen8_emit_flush_rcs(struct i915_request *rq, u32 mode) diff --git a/drivers/gpu/drm/i915/gt/intel_engine.h b/drivers/gpu/drm/i915/gt/intel_engine.h index 08559ace0ada..0e353d8c2bc8 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine.h +++ b/drivers/gpu/drm/i915/gt/intel_engine.h @@ -11,7 +11,6 @@ #include <linux/seqlock.h> #include "i915_pmu.h" -#include "i915_reg.h" #include "i915_request.h" #include "i915_selftest.h" #include "intel_engine_types.h" diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c index 352254e001b4..84ad09fb9b8b 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c @@ -6,13 +6,15 @@ #include <drm/drm_print.h> #include "gem/i915_gem_context.h" +#include "gt/intel_gt_regs.h" +#include "i915_cmd_parser.h" #include "i915_drv.h" - #include "intel_breadcrumbs.h" #include "intel_context.h" #include "intel_engine.h" #include "intel_engine_pm.h" +#include "intel_engine_regs.h" #include "intel_engine_user.h" #include "intel_execlists_submission.h" #include "intel_gt.h" diff --git a/drivers/gpu/drm/i915/gt/intel_engine_regs.h b/drivers/gpu/drm/i915/gt/intel_engine_regs.h new file mode 100644 index 000000000000..e9fec6214073 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/intel_engine_regs.h @@ -0,0 +1,223 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __INTEL_ENGINE_REGS__ +#define __INTEL_ENGINE_REGS__ + +#include "i915_reg_defs.h" + +#define RING_TAIL(base) _MMIO((base) + 0x30) +#define TAIL_ADDR 0x001FFFF8 +#define RING_HEAD(base) _MMIO((base) + 0x34) +#define HEAD_WRAP_COUNT 0xFFE00000 +#define HEAD_WRAP_ONE 0x00200000 +#define HEAD_ADDR 0x001FFFFC +#define RING_START(base) _MMIO((base) + 0x38) +#define RING_CTL(base) _MMIO((base) + 0x3c) +#define RING_CTL_SIZE(size) ((size) - PAGE_SIZE) /* in bytes -> pages */ +#define RING_NR_PAGES 0x001FF000 +#define RING_REPORT_MASK 0x00000006 +#define RING_REPORT_64K 0x00000002 +#define RING_REPORT_128K 0x00000004 +#define RING_NO_REPORT 0x00000000 +#define RING_VALID_MASK 0x00000001 +#define RING_VALID 0x00000001 +#define RING_INVALID 0x00000000 +#define RING_WAIT_I8XX (1 << 0) /* gen2, PRBx_HEAD */ +#define RING_WAIT (1 << 11) /* gen3+, PRBx_CTL */ +#define RING_WAIT_SEMAPHORE (1 << 10) /* gen6+ */ +#define RING_SYNC_0(base) _MMIO((base) + 0x40) +#define RING_SYNC_1(base) _MMIO((base) + 0x44) +#define RING_SYNC_2(base) _MMIO((base) + 0x48) +#define GEN6_RVSYNC (RING_SYNC_0(RENDER_RING_BASE)) +#define GEN6_RBSYNC (RING_SYNC_1(RENDER_RING_BASE)) +#define GEN6_RVESYNC (RING_SYNC_2(RENDER_RING_BASE)) +#define GEN6_VBSYNC (RING_SYNC_0(GEN6_BSD_RING_BASE)) +#define GEN6_VRSYNC (RING_SYNC_1(GEN6_BSD_RING_BASE)) +#define GEN6_VVESYNC (RING_SYNC_2(GEN6_BSD_RING_BASE)) +#define GEN6_BRSYNC (RING_SYNC_0(BLT_RING_BASE)) +#define GEN6_BVSYNC (RING_SYNC_1(BLT_RING_BASE)) +#define GEN6_BVESYNC (RING_SYNC_2(BLT_RING_BASE)) +#define GEN6_VEBSYNC (RING_SYNC_0(VEBOX_RING_BASE)) +#define GEN6_VERSYNC (RING_SYNC_1(VEBOX_RING_BASE)) +#define GEN6_VEVSYNC (RING_SYNC_2(VEBOX_RING_BASE)) +#define RING_PSMI_CTL(base) _MMIO((base) + 0x50) +#define GEN8_RC_SEMA_IDLE_MSG_DISABLE REG_BIT(12) +#define GEN8_FF_DOP_CLOCK_GATE_DISABLE REG_BIT(10) +#define GEN12_WAIT_FOR_EVENT_POWER_DOWN_DISABLE REG_BIT(7) +#define GEN6_BSD_GO_INDICATOR REG_BIT(4) +#define GEN6_BSD_SLEEP_INDICATOR REG_BIT(3) +#define GEN6_BSD_SLEEP_FLUSH_DISABLE REG_BIT(2) +#define GEN6_PSMI_SLEEP_MSG_DISABLE REG_BIT(0) +#define RING_MAX_IDLE(base) _MMIO((base) + 0x54) +#define PWRCTX_MAXCNT(base) _MMIO((base) + 0x54) +#define IDLE_TIME_MASK 0xFFFFF +#define RING_ACTHD_UDW(base) _MMIO((base) + 0x5c) +#define RING_DMA_FADD_UDW(base) _MMIO((base) + 0x60) /* gen8+ */ +#define RING_IPEIR(base) _MMIO((base) + 0x64) +#define RING_IPEHR(base) _MMIO((base) + 0x68) +#define RING_INSTDONE(base) _MMIO((base) + 0x6c) +#define RING_INSTPS(base) _MMIO((base) + 0x70) +#define RING_DMA_FADD(base) _MMIO((base) + 0x78) +#define RING_ACTHD(base) _MMIO((base) + 0x74) +#define RING_HWS_PGA(base) _MMIO((base) + 0x80) +#define RING_CMD_BUF_CCTL(base) _MMIO((base) + 0x84) +#define IPEIR(base) _MMIO((base) + 0x88) +#define IPEHR(base) _MMIO((base) + 0x8c) +#define RING_ID(base) _MMIO((base) + 0x8c) +#define RING_NOPID(base) _MMIO((base) + 0x94) +#define RING_HWSTAM(base) _MMIO((base) + 0x98) +#define RING_MI_MODE(base) _MMIO((base) + 0x9c) +#define RING_IMR(base) _MMIO((base) + 0xa8) +#define RING_EIR(base) _MMIO((base) + 0xb0) +#define RING_EMR(base) _MMIO((base) + 0xb4) +#define RING_ESR(base) _MMIO((base) + 0xb8) +#define RING_INSTPM(base) _MMIO((base) + 0xc0) +#define RING_CMD_CCTL(base) _MMIO((base) + 0xc4) +#define ACTHD(base) _MMIO((base) + 0xc8) +#define GEN8_R_PWR_CLK_STATE(base) _MMIO((base) + 0xc8) +#define GEN8_RPCS_ENABLE (1 << 31) +#define GEN8_RPCS_S_CNT_ENABLE (1 << 18) +#define GEN8_RPCS_S_CNT_SHIFT 15 +#define GEN8_RPCS_S_CNT_MASK (0x7 << GEN8_RPCS_S_CNT_SHIFT) +#define GEN11_RPCS_S_CNT_SHIFT 12 +#define GEN11_RPCS_S_CNT_MASK (0x3f << GEN11_RPCS_S_CNT_SHIFT) +#define GEN8_RPCS_SS_CNT_ENABLE (1 << 11) +#define GEN8_RPCS_SS_CNT_SHIFT 8 +#define GEN8_RPCS_SS_CNT_MASK (0x7 << GEN8_RPCS_SS_CNT_SHIFT) +#define GEN8_RPCS_EU_MAX_SHIFT 4 +#define GEN8_RPCS_EU_MAX_MASK (0xf << GEN8_RPCS_EU_MAX_SHIFT) +#define GEN8_RPCS_EU_MIN_SHIFT 0 +#define GEN8_RPCS_EU_MIN_MASK (0xf << GEN8_RPCS_EU_MIN_SHIFT) + +#define RING_RESET_CTL(base) _MMIO((base) + 0xd0) +#define RESET_CTL_CAT_ERROR REG_BIT(2) +#define RESET_CTL_READY_TO_RESET REG_BIT(1) +#define RESET_CTL_REQUEST_RESET REG_BIT(0) +#define DMA_FADD_I8XX(base) _MMIO((base) + 0xd0) +#define RING_BBSTATE(base) _MMIO((base) + 0x110) +#define RING_BB_PPGTT (1 << 5) +#define RING_SBBADDR(base) _MMIO((base) + 0x114) /* hsw+ */ +#define RING_SBBSTATE(base) _MMIO((base) + 0x118) /* hsw+ */ +#define RING_SBBADDR_UDW(base) _MMIO((base) + 0x11c) /* gen8+ */ +#define RING_BBADDR(base) _MMIO((base) + 0x140) +#define RING_BBADDR_UDW(base) _MMIO((base) + 0x168) /* gen8+ */ +#define CCID(base) _MMIO((base) + 0x180) +#define CCID_EN BIT(0) +#define CCID_EXTENDED_STATE_RESTORE BIT(2) +#define CCID_EXTENDED_STATE_SAVE BIT(3) +#define RING_BB_PER_CTX_PTR(base) _MMIO((base) + 0x1c0) /* gen8+ */ +#define RING_INDIRECT_CTX(base) _MMIO((base) + 0x1c4) /* gen8+ */ +#define RING_INDIRECT_CTX_OFFSET(base) _MMIO((base) + 0x1c8) /* gen8+ */ +#define ECOSKPD(base) _MMIO((base) + 0x1d0) +#define ECO_CONSTANT_BUFFER_SR_DISABLE REG_BIT(4) +#define ECO_GATING_CX_ONLY REG_BIT(3) +#define GEN6_BLITTER_FBC_NOTIFY REG_BIT(3) +#define ECO_FLIP_DONE REG_BIT(0) +#define GEN6_BLITTER_LOCK_SHIFT 16 + +#define BLIT_CCTL(base) _MMIO((base) + 0x204) +#define BLIT_CCTL_DST_MOCS_MASK REG_GENMASK(14, 8) +#define BLIT_CCTL_SRC_MOCS_MASK REG_GENMASK(6, 0) +#define BLIT_CCTL_MASK (BLIT_CCTL_DST_MOCS_MASK | \ + BLIT_CCTL_SRC_MOCS_MASK) +#define BLIT_CCTL_MOCS(dst, src) \ + (REG_FIELD_PREP(BLIT_CCTL_DST_MOCS_MASK, (dst) << 1) | \ + REG_FIELD_PREP(BLIT_CCTL_SRC_MOCS_MASK, (src) << 1)) + +/* + * CMD_CCTL read/write fields take a MOCS value and _not_ a table index. + * The lsb of each can be considered a separate enabling bit for encryption. + * 6:0 == default MOCS value for reads => 6:1 == table index for reads. + * 13:7 == default MOCS value for writes => 13:8 == table index for writes. + * 15:14 == Reserved => 31:30 are set to 0. + */ +#define CMD_CCTL_WRITE_OVERRIDE_MASK REG_GENMASK(13, 7) +#define CMD_CCTL_READ_OVERRIDE_MASK REG_GENMASK(6, 0) +#define CMD_CCTL_MOCS_MASK (CMD_CCTL_WRITE_OVERRIDE_MASK | \ + CMD_CCTL_READ_OVERRIDE_MASK) +#define CMD_CCTL_MOCS_OVERRIDE(write, read) \ + (REG_FIELD_PREP(CMD_CCTL_WRITE_OVERRIDE_MASK, (write) << 1) | \ + REG_FIELD_PREP(CMD_CCTL_READ_OVERRIDE_MASK, (read) << 1)) + +#define MI_PREDICATE_RESULT_2(base) _MMIO((base) + 0x3bc) +#define LOWER_SLICE_ENABLED (1 << 0) +#define LOWER_SLICE_DISABLED (0 << 0) +#define MI_PREDICATE_SRC0(base) _MMIO((base) + 0x400) +#define MI_PREDICATE_SRC0_UDW(base) _MMIO((base) + 0x400 + 4) +#define MI_PREDICATE_SRC1(base) _MMIO((base) + 0x408) +#define MI_PREDICATE_SRC1_UDW(base) _MMIO((base) + 0x408 + 4) +#define MI_PREDICATE_DATA(base) _MMIO((base) + 0x410) +#define MI_PREDICATE_RESULT(base) _MMIO((base) + 0x418) +#define MI_PREDICATE_RESULT_1(base) _MMIO((base) + 0x41c) + +#define RING_PP_DIR_DCLV(base) _MMIO((base) + 0x220) +#define PP_DIR_DCLV_2G 0xffffffff +#define RING_PP_DIR_BASE(base) _MMIO((base) + 0x228) +#define RING_ELSP(base) _MMIO((base) + 0x230) +#define RING_EXECLIST_STATUS_LO(base) _MMIO((base) + 0x234) +#define RING_EXECLIST_STATUS_HI(base) _MMIO((base) + 0x234 + 4) +#define RING_CONTEXT_CONTROL(base) _MMIO((base) + 0x244) +#define CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT REG_BIT(0) +#define CTX_CTRL_RS_CTX_ENABLE REG_BIT(1) +#define CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT REG_BIT(2) +#define CTX_CTRL_INHIBIT_SYN_CTX_SWITCH REG_BIT(3) +#define GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE REG_BIT(8) +#define RING_SEMA_WAIT_POLL(base) _MMIO((base) + 0x24c) +#define GEN8_RING_PDP_UDW(base, n) _MMIO((base) + 0x270 + (n) * 8 + 4) +#define GEN8_RING_PDP_LDW(base, n) _MMIO((base) + 0x270 + (n) * 8) +#define RING_MODE_GEN7(base) _MMIO((base) + 0x29c) +#define GFX_RUN_LIST_ENABLE (1 << 15) +#define GFX_INTERRUPT_STEERING (1 << 14) +#define GFX_TLB_INVALIDATE_EXPLICIT (1 << 13) +#define GFX_SURFACE_FAULT_ENABLE (1 << 12) +#define GFX_REPLAY_MODE (1 << 11) +#define GFX_PSMI_GRANULARITY (1 << 10) +#define GFX_PPGTT_ENABLE (1 << 9) +#define GEN8_GFX_PPGTT_48B (1 << 7) +#define GFX_FORWARD_VBLANK_MASK (3 << 5) +#define GFX_FORWARD_VBLANK_NEVER (0 << 5) +#define GFX_FORWARD_VBLANK_ALWAYS (1 << 5) +#define GFX_FORWARD_VBLANK_COND (2 << 5) +#define GEN11_GFX_DISABLE_LEGACY_MODE (1 << 3) +#define RING_TIMESTAMP(base) _MMIO((base) + 0x358) +#define RING_TIMESTAMP_UDW(base) _MMIO((base) + 0x358 + 4) +#define RING_CONTEXT_STATUS_PTR(base) _MMIO((base) + 0x3a0) +#define RING_CTX_TIMESTAMP(base) _MMIO((base) + 0x3a8) /* gen8+ */ +#define RING_FORCE_TO_NONPRIV(base, i) _MMIO(((base) + 0x4D0) + (i) * 4) +#define RING_FORCE_TO_NONPRIV_ADDRESS_MASK REG_GENMASK(25, 2) +#define RING_FORCE_TO_NONPRIV_ACCESS_RW (0 << 28) /* CFL+ & Gen11+ */ +#define RING_FORCE_TO_NONPRIV_ACCESS_RD (1 << 28) +#define RING_FORCE_TO_NONPRIV_ACCESS_WR (2 << 28) +#define RING_FORCE_TO_NONPRIV_ACCESS_INVALID (3 << 28) +#define RING_FORCE_TO_NONPRIV_ACCESS_MASK (3 << 28) +#define RING_FORCE_TO_NONPRIV_RANGE_1 (0 << 0) /* CFL+ & Gen11+ */ +#define RING_FORCE_TO_NONPRIV_RANGE_4 (1 << 0) +#define RING_FORCE_TO_NONPRIV_RANGE_16 (2 << 0) +#define RING_FORCE_TO_NONPRIV_RANGE_64 (3 << 0) +#define RING_FORCE_TO_NONPRIV_RANGE_MASK (3 << 0) +#define RING_FORCE_TO_NONPRIV_MASK_VALID \ + (RING_FORCE_TO_NONPRIV_RANGE_MASK | RING_FORCE_TO_NONPRIV_ACCESS_MASK) +#define RING_MAX_NONPRIV_SLOTS 12 + +#define RING_EXECLIST_SQ_CONTENTS(base) _MMIO((base) + 0x510) +#define RING_PP_DIR_BASE_READ(base) _MMIO((base) + 0x518) +#define RING_EXECLIST_CONTROL(base) _MMIO((base) + 0x550) +#define EL_CTRL_LOAD REG_BIT(0) + +/* There are 16 64-bit CS General Purpose Registers per-engine on Gen8+ */ +#define GEN8_RING_CS_GPR(base, n) _MMIO((base) + 0x600 + (n) * 8) +#define GEN8_RING_CS_GPR_UDW(base, n) _MMIO((base) + 0x600 + (n) * 8 + 4) + +#define RING_HWS_PGA_GEN6(base) _MMIO((base) + 0x2080) + +#define VDBOX_CGCTL3F10(base) _MMIO((base) + 0x3f10) +#define IECPUNIT_CLKGATE_DIS REG_BIT(22) + +#define VDBOX_CGCTL3F18(base) _MMIO((base) + 0x3f18) +#define ALNUNIT_CLKGATE_DIS REG_BIT(13) + + +#endif /* __INTEL_ENGINE_REGS__ */ diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c index a69df5e9e77a..49e26d23ce1f 100644 --- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c @@ -116,11 +116,13 @@ #include "intel_context.h" #include "intel_engine_heartbeat.h" #include "intel_engine_pm.h" +#include "intel_engine_regs.h" #include "intel_engine_stats.h" #include "intel_execlists_submission.h" #include "intel_gt.h" #include "intel_gt_irq.h" #include "intel_gt_pm.h" +#include "intel_gt_regs.h" #include "intel_gt_requests.h" #include "intel_lrc.h" #include "intel_lrc_reg.h" diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c index 5263dda7f8d5..d2922f64d1c8 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c @@ -15,6 +15,7 @@ #include "gem/i915_gem_lmem.h" #include "intel_gt.h" +#include "intel_gt_regs.h" #include "i915_drv.h" #include "i915_scatterlist.h" #include "i915_vgpu.h" diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c index f8948de72036..95c9145232f5 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c @@ -4,9 +4,11 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "i915_scatterlist.h" #include "i915_pvinfo.h" #include "i915_vgpu.h" +#include "intel_gt_regs.h" /** * DOC: fence register handling diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c index 35d0fcd3a86c..57763aede976 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.c +++ b/drivers/gpu/drm/i915/gt/intel_gt.c @@ -10,10 +10,12 @@ #include "gem/i915_gem_lmem.h" #include "i915_drv.h" #include "intel_context.h" +#include "intel_engine_regs.h" #include "intel_gt.h" #include "intel_gt_buffer_pool.h" #include "intel_gt_clock_utils.h" #include "intel_gt_pm.h" +#include "intel_gt_regs.h" #include "intel_gt_requests.h" #include "intel_migrate.h" #include "intel_mocs.h" @@ -207,7 +209,7 @@ int intel_gt_init_hw(struct intel_gt *gt) if (IS_HASWELL(i915)) intel_uncore_write(uncore, - MI_PREDICATE_RESULT_2, + HSW_MI_PREDICATE_RESULT_2, IS_HSW_GT3(i915) ? LOWER_SLICE_ENABLED : LOWER_SLICE_DISABLED); diff --git a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c b/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c index 3513d6f90747..0db822c3b7e5 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c @@ -4,8 +4,10 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "intel_gt.h" #include "intel_gt_clock_utils.h" +#include "intel_gt_regs.h" static u32 read_reference_ts_freq(struct intel_uncore *uncore) { diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c index 699a74582d32..983264e10e0a 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c @@ -10,7 +10,7 @@ #include "intel_breadcrumbs.h" #include "intel_gt.h" #include "intel_gt_irq.h" -#include "intel_lrc_reg.h" +#include "intel_gt_regs.h" #include "intel_uncore.h" #include "intel_rps.h" #include "pxp/intel_pxp_irq.h" diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm_debugfs.c b/drivers/gpu/drm/i915/gt/intel_gt_pm_debugfs.c index 404dfa7673c6..4e448c13a64c 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_pm_debugfs.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_pm_debugfs.c @@ -7,11 +7,13 @@ #include <linux/seq_file.h> #include "i915_drv.h" +#include "i915_reg.h" #include "intel_gt.h" #include "intel_gt_clock_utils.h" #include "intel_gt_debugfs.h" #include "intel_gt_pm.h" #include "intel_gt_pm_debugfs.h" +#include "intel_gt_regs.h" #include "intel_llc.h" #include "intel_pcode.h" #include "intel_rc6.h" @@ -134,8 +136,7 @@ static int gen6_drpc(struct seq_file *m) } if (GRAPHICS_VER(i915) <= 7) - sandybridge_pcode_read(i915, GEN6_PCODE_READ_RC6VIDS, - &rc6vids, NULL); + snb_pcode_read(i915, GEN6_PCODE_READ_RC6VIDS, &rc6vids, NULL); seq_printf(m, "RC1e Enabled: %s\n", yesno(rcctl1 & GEN6_RC_CTL_RC1e_ENABLE)); @@ -557,9 +558,8 @@ static int llc_show(struct seq_file *m, void *data) wakeref = intel_runtime_pm_get(gt->uncore->rpm); for (gpu_freq = min_gpu_freq; gpu_freq <= max_gpu_freq; gpu_freq++) { ia_freq = gpu_freq; - sandybridge_pcode_read(i915, - GEN6_PCODE_READ_MIN_FREQ_TABLE, - &ia_freq, NULL); + snb_pcode_read(i915, GEN6_PCODE_READ_MIN_FREQ_TABLE, + &ia_freq, NULL); seq_printf(m, "%d\t\t%d\t\t\t\t%d\n", intel_gpu_freq(rps, (gpu_freq * diff --git a/drivers/gpu/drm/i915/gt/intel_gt_pm_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_pm_irq.c index fe51f894b073..11060f5a4c89 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_pm_irq.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_pm_irq.c @@ -8,6 +8,7 @@ #include "intel_gt.h" #include "intel_gt_irq.h" #include "intel_gt_pm_irq.h" +#include "intel_gt_regs.h" static void write_pm_imr(struct intel_gt *gt) { diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h new file mode 100644 index 000000000000..16d98ebee687 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h @@ -0,0 +1,1554 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __INTEL_GT_REGS__ +#define __INTEL_GT_REGS__ + +#include "i915_reg_defs.h" + +#define ILK_GDSR _MMIO(MCHBAR_MIRROR_BASE + 0x2ca4) +#define ILK_GRDOM_FULL (0 << 1) +#define ILK_GRDOM_RENDER (1 << 1) +#define ILK_GRDOM_MEDIA (3 << 1) +#define ILK_GRDOM_MASK (3 << 1) +#define ILK_GRDOM_RESET_ENABLE (1 << 0) + +#define GEN6_MBCUNIT_SNPCR _MMIO(0x900c) /* for LLC config */ +#define GEN6_MBC_SNPCR_SHIFT 21 +#define GEN6_MBC_SNPCR_MASK (3 << 21) +#define GEN6_MBC_SNPCR_MAX (0 << 21) +#define GEN6_MBC_SNPCR_MED (1 << 21) +#define GEN6_MBC_SNPCR_LOW (2 << 21) +#define GEN6_MBC_SNPCR_MIN (3 << 21) /* only 1/16th of the cache is shared */ + +#define VLV_G3DCTL _MMIO(0x9024) +#define VLV_GSCKGCTL _MMIO(0x9028) + +#define FBC_LLC_READ_CTRL _MMIO(0x9044) +#define FBC_LLC_FULLY_OPEN REG_BIT(30) + +#define GEN6_MBCTL _MMIO(0x0907c) +#define GEN6_MBCTL_ENABLE_BOOT_FETCH (1 << 4) +#define GEN6_MBCTL_CTX_FETCH_NEEDED (1 << 3) +#define GEN6_MBCTL_BME_UPDATE_ENABLE (1 << 2) +#define GEN6_MBCTL_MAE_UPDATE_ENABLE (1 << 1) +#define GEN6_MBCTL_BOOT_FETCH_MECH (1 << 0) + +#define GEN6_GDRST _MMIO(0x941c) +#define GEN6_GRDOM_FULL (1 << 0) +#define GEN6_GRDOM_RENDER (1 << 1) +#define GEN6_GRDOM_MEDIA (1 << 2) +#define GEN6_GRDOM_BLT (1 << 3) +#define GEN6_GRDOM_VECS (1 << 4) +#define GEN9_GRDOM_GUC (1 << 5) +#define GEN8_GRDOM_MEDIA2 (1 << 7) +/* GEN11 changed all bit defs except for FULL & RENDER */ +#define GEN11_GRDOM_FULL GEN6_GRDOM_FULL +#define GEN11_GRDOM_RENDER GEN6_GRDOM_RENDER +#define GEN11_GRDOM_BLT (1 << 2) +#define GEN11_GRDOM_GUC (1 << 3) +#define GEN11_GRDOM_MEDIA (1 << 5) +#define GEN11_GRDOM_MEDIA2 (1 << 6) +#define GEN11_GRDOM_MEDIA3 (1 << 7) +#define GEN11_GRDOM_MEDIA4 (1 << 8) +#define GEN11_GRDOM_MEDIA5 (1 << 9) +#define GEN11_GRDOM_MEDIA6 (1 << 10) +#define GEN11_GRDOM_MEDIA7 (1 << 11) +#define GEN11_GRDOM_MEDIA8 (1 << 12) +#define GEN11_GRDOM_VECS (1 << 13) +#define GEN11_GRDOM_VECS2 (1 << 14) +#define GEN11_GRDOM_VECS3 (1 << 15) +#define GEN11_GRDOM_VECS4 (1 << 16) +#define GEN11_GRDOM_SFC0 (1 << 17) +#define GEN11_GRDOM_SFC1 (1 << 18) +#define GEN11_GRDOM_SFC2 (1 << 19) +#define GEN11_GRDOM_SFC3 (1 << 20) +#define GEN11_VCS_SFC_RESET_BIT(instance) (GEN11_GRDOM_SFC0 << ((instance) >> 1)) +#define GEN11_VECS_SFC_RESET_BIT(instance) (GEN11_GRDOM_SFC0 << (instance)) + +#define GEN11_VCS_SFC_FORCED_LOCK(engine) _MMIO((engine)->mmio_base + 0x88C) +#define GEN11_VCS_SFC_FORCED_LOCK_BIT (1 << 0) +#define GEN11_VCS_SFC_LOCK_STATUS(engine) _MMIO((engine)->mmio_base + 0x890) +#define GEN11_VCS_SFC_USAGE_BIT (1 << 0) +#define GEN11_VCS_SFC_LOCK_ACK_BIT (1 << 1) + +#define GEN11_VECS_SFC_FORCED_LOCK(engine) _MMIO((engine)->mmio_base + 0x201C) +#define GEN11_VECS_SFC_FORCED_LOCK_BIT (1 << 0) +#define GEN11_VECS_SFC_LOCK_ACK(engine) _MMIO((engine)->mmio_base + 0x2018) +#define GEN11_VECS_SFC_LOCK_ACK_BIT (1 << 0) +#define GEN11_VECS_SFC_USAGE(engine) _MMIO((engine)->mmio_base + 0x2014) +#define GEN11_VECS_SFC_USAGE_BIT (1 << 0) + +#define GEN12_HCP_SFC_FORCED_LOCK(engine) _MMIO((engine)->mmio_base + 0x2910) +#define GEN12_HCP_SFC_FORCED_LOCK_BIT REG_BIT(0) +#define GEN12_HCP_SFC_LOCK_STATUS(engine) _MMIO((engine)->mmio_base + 0x2914) +#define GEN12_HCP_SFC_LOCK_ACK_BIT REG_BIT(1) +#define GEN12_HCP_SFC_USAGE_BIT REG_BIT(0) + +#define GEN12_SFC_DONE(n) _MMIO(0x1cc000 + (n) * 0x1000) + +#define WAIT_FOR_RC6_EXIT _MMIO(0x20CC) +/* HSW only */ +#define HSW_SELECTIVE_READ_ADDRESSING_SHIFT 2 +#define HSW_SELECTIVE_READ_ADDRESSING_MASK (0x3 << HSW_SLECTIVE_READ_ADDRESSING_SHIFT) +#define HSW_SELECTIVE_WRITE_ADDRESS_SHIFT 4 +#define HSW_SELECTIVE_WRITE_ADDRESS_MASK (0x7 << HSW_SELECTIVE_WRITE_ADDRESS_SHIFT) +/* HSW+ */ +#define HSW_WAIT_FOR_RC6_EXIT_ENABLE (1 << 0) +#define HSW_RCS_CONTEXT_ENABLE (1 << 7) +#define HSW_RCS_INHIBIT (1 << 8) +/* Gen8 */ +#define GEN8_SELECTIVE_WRITE_ADDRESS_SHIFT 4 +#define GEN8_SELECTIVE_WRITE_ADDRESS_MASK (0x3 << GEN8_SELECTIVE_WRITE_ADDRESS_SHIFT) +#define GEN8_SELECTIVE_WRITE_ADDRESS_SHIFT 4 +#define GEN8_SELECTIVE_WRITE_ADDRESS_MASK (0x3 << GEN8_SELECTIVE_WRITE_ADDRESS_SHIFT) +#define GEN8_SELECTIVE_WRITE_ADDRESSING_ENABLE (1 << 6) +#define GEN8_SELECTIVE_READ_SUBSLICE_SELECT_SHIFT 9 +#define GEN8_SELECTIVE_READ_SUBSLICE_SELECT_MASK (0x3 << GEN8_SELECTIVE_READ_SUBSLICE_SELECT_SHIFT) +#define GEN8_SELECTIVE_READ_SLICE_SELECT_SHIFT 11 +#define GEN8_SELECTIVE_READ_SLICE_SELECT_MASK (0x3 << GEN8_SELECTIVE_READ_SLICE_SELECT_SHIFT) +#define GEN8_SELECTIVE_READ_ADDRESSING_ENABLE (1 << 13) + +#define GAM_ECOCHK _MMIO(0x4090) +#define BDW_DISABLE_HDC_INVALIDATION (1 << 25) +#define ECOCHK_SNB_BIT (1 << 10) +#define ECOCHK_DIS_TLB (1 << 8) +#define HSW_ECOCHK_ARB_PRIO_SOL (1 << 6) +#define ECOCHK_PPGTT_CACHE64B (0x3 << 3) +#define ECOCHK_PPGTT_CACHE4B (0x0 << 3) +#define ECOCHK_PPGTT_GFDT_IVB (0x1 << 4) +#define ECOCHK_PPGTT_LLC_IVB (0x1 << 3) +#define ECOCHK_PPGTT_UC_HSW (0x1 << 3) +#define ECOCHK_PPGTT_WT_HSW (0x2 << 3) +#define ECOCHK_PPGTT_WB_HSW (0x3 << 3) + +#define GEN8_RC6_CTX_INFO _MMIO(0x8504) + +#define GAC_ECO_BITS _MMIO(0x14090) +#define ECOBITS_SNB_BIT (1 << 13) +#define ECOBITS_PPGTT_CACHE64B (3 << 8) +#define ECOBITS_PPGTT_CACHE4B (0 << 8) + +#define GEN12_GAMCNTRL_CTRL _MMIO(0xcf54) +#define INVALIDATION_BROADCAST_MODE_DIS REG_BIT(12) +#define GLOBAL_INVALIDATION_MODE REG_BIT(2) + +#define GEN12_GAMSTLB_CTRL _MMIO(0xcf4c) +#define CONTROL_BLOCK_CLKGATE_DIS REG_BIT(12) +#define EGRESS_BLOCK_CLKGATE_DIS REG_BIT(11) +#define TAG_BLOCK_CLKGATE_DIS REG_BIT(7) + +#define GEN12_MERT_MOD_CTRL _MMIO(0xcf28) +#define RENDER_MOD_CTRL _MMIO(0xcf2c) +#define COMP_MOD_CTRL _MMIO(0xcf30) +#define VDBX_MOD_CTRL _MMIO(0xcf34) +#define VEBX_MOD_CTRL _MMIO(0xcf38) +#define FORCE_MISS_FTLB REG_BIT(3) + +#define GAB_CTL _MMIO(0x24000) +#define GAB_CTL_CONT_AFTER_PAGEFAULT (1 << 8) + +#define HSW_MI_PREDICATE_RESULT_2 _MMIO(0x2214) + +/* + * Registers used only by the command parser + */ +#define BCS_SWCTRL _MMIO(0x22200) +#define BCS_SRC_Y REG_BIT(0) +#define BCS_DST_Y REG_BIT(1) + +#define GPGPU_THREADS_DISPATCHED _MMIO(0x2290) +#define GPGPU_THREADS_DISPATCHED_UDW _MMIO(0x2290 + 4) +#define HS_INVOCATION_COUNT _MMIO(0x2300) +#define HS_INVOCATION_COUNT_UDW _MMIO(0x2300 + 4) +#define DS_INVOCATION_COUNT _MMIO(0x2308) +#define DS_INVOCATION_COUNT_UDW _MMIO(0x2308 + 4) +#define IA_VERTICES_COUNT _MMIO(0x2310) +#define IA_VERTICES_COUNT_UDW _MMIO(0x2310 + 4) +#define IA_PRIMITIVES_COUNT _MMIO(0x2318) +#define IA_PRIMITIVES_COUNT_UDW _MMIO(0x2318 + 4) +#define VS_INVOCATION_COUNT _MMIO(0x2320) +#define VS_INVOCATION_COUNT_UDW _MMIO(0x2320 + 4) +#define GS_INVOCATION_COUNT _MMIO(0x2328) +#define GS_INVOCATION_COUNT_UDW _MMIO(0x2328 + 4) +#define GS_PRIMITIVES_COUNT _MMIO(0x2330) +#define GS_PRIMITIVES_COUNT_UDW _MMIO(0x2330 + 4) +#define CL_INVOCATION_COUNT _MMIO(0x2338) +#define CL_INVOCATION_COUNT_UDW _MMIO(0x2338 + 4) +#define CL_PRIMITIVES_COUNT _MMIO(0x2340) +#define CL_PRIMITIVES_COUNT_UDW _MMIO(0x2340 + 4) +#define PS_INVOCATION_COUNT _MMIO(0x2348) +#define PS_INVOCATION_COUNT_UDW _MMIO(0x2348 + 4) +#define PS_DEPTH_COUNT _MMIO(0x2350) +#define PS_DEPTH_COUNT_UDW _MMIO(0x2350 + 4) + +/* There are the 4 64-bit counter registers, one for each stream output */ +#define GEN7_SO_NUM_PRIMS_WRITTEN(n) _MMIO(0x5200 + (n) * 8) +#define GEN7_SO_NUM_PRIMS_WRITTEN_UDW(n) _MMIO(0x5200 + (n) * 8 + 4) + +#define GEN7_SO_PRIM_STORAGE_NEEDED(n) _MMIO(0x5240 + (n) * 8) +#define GEN7_SO_PRIM_STORAGE_NEEDED_UDW(n) _MMIO(0x5240 + (n) * 8 + 4) + +#define GEN7_3DPRIM_END_OFFSET _MMIO(0x2420) +#define GEN7_3DPRIM_START_VERTEX _MMIO(0x2430) +#define GEN7_3DPRIM_VERTEX_COUNT _MMIO(0x2434) +#define GEN7_3DPRIM_INSTANCE_COUNT _MMIO(0x2438) +#define GEN7_3DPRIM_START_INSTANCE _MMIO(0x243C) +#define GEN7_3DPRIM_BASE_VERTEX _MMIO(0x2440) + +#define GEN7_GPGPU_DISPATCHDIMX _MMIO(0x2500) +#define GEN7_GPGPU_DISPATCHDIMY _MMIO(0x2504) +#define GEN7_GPGPU_DISPATCHDIMZ _MMIO(0x2508) + +#define GEN12_SQCM _MMIO(0x8724) +#define EN_32B_ACCESS REG_BIT(30) + +/* + * Flexible, Aggregate EU Counter Registers. + * Note: these aren't contiguous + */ +#define EU_PERF_CNTL0 _MMIO(0xe458) +#define EU_PERF_CNTL1 _MMIO(0xe558) +#define EU_PERF_CNTL2 _MMIO(0xe658) +#define EU_PERF_CNTL3 _MMIO(0xe758) +#define EU_PERF_CNTL4 _MMIO(0xe45c) +#define EU_PERF_CNTL5 _MMIO(0xe55c) +#define EU_PERF_CNTL6 _MMIO(0xe65c) + +#define RT_CTRL _MMIO(0xe530) +#define DIS_NULL_QUERY REG_BIT(10) + +/* RPM unit config (Gen8+) */ +#define RPM_CONFIG0 _MMIO(0x0D00) +#define GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_SHIFT 3 +#define GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_MASK (1 << GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_SHIFT) +#define GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_19_2_MHZ 0 +#define GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_24_MHZ 1 +#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_SHIFT 3 +#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_MASK (0x7 << GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_SHIFT) +#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_24_MHZ 0 +#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_19_2_MHZ 1 +#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_38_4_MHZ 2 +#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_25_MHZ 3 +#define GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_SHIFT 1 +#define GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_MASK (0x3 << GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_SHIFT) + +#define RPM_CONFIG1 _MMIO(0x0D04) +#define GEN10_GT_NOA_ENABLE (1 << 9) + +/* GPM unit config (Gen9+) */ +#define CTC_MODE _MMIO(0xA26C) +#define CTC_SOURCE_PARAMETER_MASK 1 +#define CTC_SOURCE_CRYSTAL_CLOCK 0 +#define CTC_SOURCE_DIVIDE_LOGIC 1 +#define CTC_SHIFT_PARAMETER_SHIFT 1 +#define CTC_SHIFT_PARAMETER_MASK (0x3 << CTC_SHIFT_PARAMETER_SHIFT) + +/* RCP unit config (Gen8+) */ +#define RCP_CONFIG _MMIO(0x0D08) + +#define MICRO_BP0_0 _MMIO(0x9800) +#define MICRO_BP0_2 _MMIO(0x9804) +#define MICRO_BP0_1 _MMIO(0x9808) + +#define MICRO_BP1_0 _MMIO(0x980C) +#define MICRO_BP1_2 _MMIO(0x9810) +#define MICRO_BP1_1 _MMIO(0x9814) + +#define MICRO_BP2_0 _MMIO(0x9818) +#define MICRO_BP2_2 _MMIO(0x981C) +#define MICRO_BP2_1 _MMIO(0x9820) + +#define MICRO_BP3_0 _MMIO(0x9824) +#define MICRO_BP3_2 _MMIO(0x9828) +#define MICRO_BP3_1 _MMIO(0x982C) + +#define MICRO_BP_TRIGGER _MMIO(0x9830) +#define MICRO_BP3_COUNT_STATUS01 _MMIO(0x9834) +#define MICRO_BP3_COUNT_STATUS23 _MMIO(0x9838) +#define MICRO_BP_FIRED_ARMED _MMIO(0x983C) + +#define GAMTARBMODE _MMIO(0x04a08) +#define ARB_MODE_BWGTLB_DISABLE (1 << 9) +#define ARB_MODE_SWIZZLE_BDW (1 << 1) +#define RENDER_HWS_PGA_GEN7 _MMIO(0x04080) + +#define _RING_FAULT_REG_RCS 0x4094 +#define _RING_FAULT_REG_VCS 0x4194 +#define _RING_FAULT_REG_BCS 0x4294 +#define _RING_FAULT_REG_VECS 0x4394 +#define RING_FAULT_REG(engine) _MMIO(_PICK((engine)->class, \ + _RING_FAULT_REG_RCS, \ + _RING_FAULT_REG_VCS, \ + _RING_FAULT_REG_VECS, \ + _RING_FAULT_REG_BCS)) +#define GEN8_RING_FAULT_REG _MMIO(0x4094) +#define GEN12_RING_FAULT_REG _MMIO(0xcec4) +#define GEN8_RING_FAULT_ENGINE_ID(x) (((x) >> 12) & 0x7) +#define RING_FAULT_GTTSEL_MASK (1 << 11) +#define RING_FAULT_SRCID(x) (((x) >> 3) & 0xff) +#define RING_FAULT_FAULT_TYPE(x) (((x) >> 1) & 0x3) +#define RING_FAULT_VALID (1 << 0) +#define DONE_REG _MMIO(0x40b0) +#define GEN12_GAM_DONE _MMIO(0xcf68) +#define GEN8_PRIVATE_PAT_LO _MMIO(0x40e0) +#define GEN8_PRIVATE_PAT_HI _MMIO(0x40e0 + 4) +#define GEN10_PAT_INDEX(index) _MMIO(0x40e0 + (index) * 4) +#define GEN12_PAT_INDEX(index) _MMIO(0x4800 + (index) * 4) +#define BSD_HWS_PGA_GEN7 _MMIO(0x04180) +#define GEN12_GFX_CCS_AUX_NV _MMIO(0x4208) +#define GEN12_VD0_AUX_NV _MMIO(0x4218) +#define GEN12_VD1_AUX_NV _MMIO(0x4228) +#define GEN12_VD2_AUX_NV _MMIO(0x4298) +#define GEN12_VD3_AUX_NV _MMIO(0x42A8) +#define GEN12_VE0_AUX_NV _MMIO(0x4238) +#define GEN12_VE1_AUX_NV _MMIO(0x42B8) +#define AUX_INV REG_BIT(0) +#define BLT_HWS_PGA_GEN7 _MMIO(0x04280) +#define VEBOX_HWS_PGA_GEN7 _MMIO(0x04380) + +#define MISC_STATUS0 _MMIO(0xA500) +#define MISC_STATUS1 _MMIO(0xA504) + +#define GEN7_TLB_RD_ADDR _MMIO(0x4700) + +#define GEN9_GAMT_ECO_REG_RW_IA _MMIO(0x4ab0) +#define GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS (1 << 18) + +#define GEN8_GAMW_ECO_DEV_RW_IA _MMIO(0x4080) +#define GAMW_ECO_ENABLE_64K_IPS_FIELD 0xF +#define GAMW_ECO_DEV_CTX_RELOAD_DISABLE (1 << 7) + +#define GAMT_CHKN_BIT_REG _MMIO(0x4ab8) +#define GAMT_CHKN_DISABLE_L3_COH_PIPE (1 << 31) +#define GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING (1 << 28) +#define GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT (1 << 24) + +#define GEN8_RTCR _MMIO(0x4260) +#define GEN8_M1TCR _MMIO(0x4264) +#define GEN8_M2TCR _MMIO(0x4268) +#define GEN8_BTCR _MMIO(0x426c) +#define GEN8_VTCR _MMIO(0x4270) + +#define IPEIR_I965 _MMIO(0x2064) +#define IPEHR_I965 _MMIO(0x2068) +#define GEN7_SC_INSTDONE _MMIO(0x7100) +#define GEN12_SC_INSTDONE_EXTRA _MMIO(0x7104) +#define GEN12_SC_INSTDONE_EXTRA2 _MMIO(0x7108) +#define GEN7_SAMPLER_INSTDONE _MMIO(0xe160) +#define GEN7_ROW_INSTDONE _MMIO(0xe164) +#define XEHPG_INSTDONE_GEOM_SVG _MMIO(0x666c) +#define MCFG_MCR_SELECTOR _MMIO(0xfd0) +#define SF_MCR_SELECTOR _MMIO(0xfd8) +#define GEN8_MCR_SELECTOR _MMIO(0xfdc) +#define GEN8_MCR_SLICE(slice) (((slice) & 3) << 26) +#define GEN8_MCR_SLICE_MASK GEN8_MCR_SLICE(3) +#define GEN8_MCR_SUBSLICE(subslice) (((subslice) & 3) << 24) +#define GEN8_MCR_SUBSLICE_MASK GEN8_MCR_SUBSLICE(3) +#define GEN11_MCR_SLICE(slice) (((slice) & 0xf) << 27) +#define GEN11_MCR_SLICE_MASK GEN11_MCR_SLICE(0xf) +#define GEN11_MCR_SUBSLICE(subslice) (((subslice) & 0x7) << 24) +#define GEN11_MCR_SUBSLICE_MASK GEN11_MCR_SUBSLICE(0x7) +/* + * On GEN4, only the render ring INSTDONE exists and has a different + * layout than the GEN7+ version. + * The GEN2 counterpart of this register is GEN2_INSTDONE. + */ +#define INSTPS _MMIO(0x2070) /* 965+ only */ +#define GEN4_INSTDONE1 _MMIO(0x207c) /* 965+ only, aka INSTDONE_2 on SNB */ +#define ACTHD_I965 _MMIO(0x2074) +#define HWS_PGA _MMIO(0x2080) +#define HWS_ADDRESS_MASK 0xfffff000 +#define HWS_START_ADDRESS_SHIFT 4 +#define PWRCTXA _MMIO(0x2088) /* 965GM+ only */ +#define PWRCTX_EN (1 << 0) +#define GEN2_INSTDONE _MMIO(0x2090) +#define NOPID _MMIO(0x2094) +#define HWSTAM _MMIO(0x2098) + +#define VDBOX_CGCTL3F18(base) _MMIO((base) + 0x3f18) +#define ALNUNIT_CLKGATE_DIS REG_BIT(13) + +#define ERROR_GEN6 _MMIO(0x40a0) + +#define GEN8_FAULT_TLB_DATA0 _MMIO(0x4b10) +#define GEN8_FAULT_TLB_DATA1 _MMIO(0x4b14) +#define GEN12_FAULT_TLB_DATA0 _MMIO(0xceb8) +#define GEN12_FAULT_TLB_DATA1 _MMIO(0xcebc) +#define FAULT_VA_HIGH_BITS (0xf << 0) +#define FAULT_GTT_SEL (1 << 4) + +#define GEN12_GFX_TLB_INV_CR _MMIO(0xced8) +#define GEN12_VD_TLB_INV_CR _MMIO(0xcedc) +#define GEN12_VE_TLB_INV_CR _MMIO(0xcee0) +#define GEN12_BLT_TLB_INV_CR _MMIO(0xcee4) + +#define GEN12_AUX_ERR_DBG _MMIO(0x43f4) + +/* GM45+ chicken bits -- debug workaround bits that may be required + * for various sorts of correct behavior. The top 16 bits of each are + * the enables for writing to the corresponding low bit. + */ +#define _3D_CHICKEN _MMIO(0x2084) +#define _3D_CHICKEN_HIZ_PLANE_DISABLE_MSAA_4X_SNB (1 << 10) +#define _3D_CHICKEN2 _MMIO(0x208c) + +#define FF_SLICE_CHICKEN _MMIO(0x2088) +#define FF_SLICE_CHICKEN_CL_PROVOKING_VERTEX_FIX (1 << 1) + +/* Disables pipelining of read flushes past the SF-WIZ interface. + * Required on all Ironlake steppings according to the B-Spec, but the + * particular danger of not doing so is not specified. + */ +# define _3D_CHICKEN2_WM_READ_PIPELINED (1 << 14) +#define _3D_CHICKEN3 _MMIO(0x2090) +#define _3D_CHICKEN_SF_PROVOKING_VERTEX_FIX (1 << 12) +#define _3D_CHICKEN_SF_DISABLE_OBJEND_CULL (1 << 10) +#define _3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE (1 << 5) +#define _3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL (1 << 5) +#define _3D_CHICKEN_SDE_LIMIT_FIFO_POLY_DEPTH(x) ((x) << 1) /* gen8+ */ +#define _3D_CHICKEN3_SF_DISABLE_PIPELINED_ATTR_FETCH (1 << 1) /* gen6 */ + +#define MI_MODE _MMIO(0x209c) +# define VS_TIMER_DISPATCH (1 << 6) +# define MI_FLUSH_ENABLE (1 << 12) +# define TGL_NESTED_BB_EN (1 << 12) +# define ASYNC_FLIP_PERF_DISABLE (1 << 14) +# define MODE_IDLE (1 << 9) +# define STOP_RING (1 << 8) + +#define GEN6_GT_MODE _MMIO(0x20d0) +#define GEN7_GT_MODE _MMIO(0x7008) +#define GEN6_WIZ_HASHING(hi, lo) (((hi) << 9) | ((lo) << 7)) +#define GEN6_WIZ_HASHING_8x8 GEN6_WIZ_HASHING(0, 0) +#define GEN6_WIZ_HASHING_8x4 GEN6_WIZ_HASHING(0, 1) +#define GEN6_WIZ_HASHING_16x4 GEN6_WIZ_HASHING(1, 0) +#define GEN6_WIZ_HASHING_MASK GEN6_WIZ_HASHING(1, 1) +#define GEN6_TD_FOUR_ROW_DISPATCH_DISABLE (1 << 5) +#define GEN9_IZ_HASHING_MASK(slice) (0x3 << ((slice) * 2)) +#define GEN9_IZ_HASHING(slice, val) ((val) << ((slice) * 2)) + +/* chicken reg for WaConextSwitchWithConcurrentTLBInvalidate */ +#define GEN9_CSFE_CHICKEN1_RCS _MMIO(0x20D4) +#define GEN9_PREEMPT_GPGPU_SYNC_SWITCH_DISABLE (1 << 2) +#define GEN11_ENABLE_32_PLANE_MODE (1 << 7) + +#define SCCGCTL94DC _MMIO(0x94dc) +#define CG3DDISURB REG_BIT(14) + +#define MLTICTXCTL _MMIO(0xb170) +#define TDONRENDER REG_BIT(2) + +#define L3SQCREG1_CCS0 _MMIO(0xb200) +#define FLUSHALLNONCOH REG_BIT(5) + +/* WaClearTdlStateAckDirtyBits */ +#define GEN8_STATE_ACK _MMIO(0x20F0) +#define GEN9_STATE_ACK_SLICE1 _MMIO(0x20F8) +#define GEN9_STATE_ACK_SLICE2 _MMIO(0x2100) +#define GEN9_STATE_ACK_TDL0 (1 << 12) +#define GEN9_STATE_ACK_TDL1 (1 << 13) +#define GEN9_STATE_ACK_TDL2 (1 << 14) +#define GEN9_STATE_ACK_TDL3 (1 << 15) +#define GEN9_SUBSLICE_TDL_ACK_BITS \ + (GEN9_STATE_ACK_TDL3 | GEN9_STATE_ACK_TDL2 | \ + GEN9_STATE_ACK_TDL1 | GEN9_STATE_ACK_TDL0) + +#define GFX_MODE _MMIO(0x2520) + +#define CACHE_MODE_0 _MMIO(0x2120) /* 915+ only */ +#define CM0_PIPELINED_RENDER_FLUSH_DISABLE (1 << 8) +#define CM0_IZ_OPT_DISABLE (1 << 6) +#define CM0_ZR_OPT_DISABLE (1 << 5) +#define CM0_STC_EVICT_DISABLE_LRA_SNB (1 << 5) +#define CM0_DEPTH_EVICT_DISABLE (1 << 4) +#define CM0_COLOR_EVICT_DISABLE (1 << 3) +#define CM0_DEPTH_WRITE_DISABLE (1 << 1) +#define CM0_RC_OP_FLUSH_DISABLE (1 << 0) +#define GFX_FLSH_CNTL _MMIO(0x2170) /* 915+ only */ +#define GFX_FLSH_CNTL_GEN6 _MMIO(0x101008) +#define GFX_FLSH_CNTL_EN (1 << 0) + +#define CACHE_MODE_0_GEN7 _MMIO(0x7000) /* IVB+ */ +#define RC_OP_FLUSH_ENABLE (1 << 0) +#define HIZ_RAW_STALL_OPT_DISABLE (1 << 2) +#define CACHE_MODE_1 _MMIO(0x7004) /* IVB+ */ +#define PIXEL_SUBSPAN_COLLECT_OPT_DISABLE (1 << 6) +#define GEN8_4x4_STC_OPTIMIZATION_DISABLE (1 << 6) +#define GEN9_PARTIAL_RESOLVE_IN_VC_DISABLE (1 << 1) + +#define GEN6_RCS_PWR_FSM _MMIO(0x22ac) +#define GEN9_RCS_FE_FSM2 _MMIO(0x22a4) + +#define GEN10_CACHE_MODE_SS _MMIO(0xe420) +#define ENABLE_PREFETCH_INTO_IC REG_BIT(3) +#define FLOAT_BLEND_OPTIMIZATION_ENABLE REG_BIT(4) + +/* Fuse readout registers for GT */ +#define HSW_PAVP_FUSE1 _MMIO(0x911C) +#define XEHP_SFC_ENABLE_MASK REG_GENMASK(27, 24) +#define HSW_F1_EU_DIS_MASK REG_GENMASK(17, 16) +#define HSW_F1_EU_DIS_10EUS 0 +#define HSW_F1_EU_DIS_8EUS 1 +#define HSW_F1_EU_DIS_6EUS 2 + +#define CHV_FUSE_GT _MMIO(VLV_DISPLAY_BASE + 0x2168) +#define CHV_FGT_DISABLE_SS0 (1 << 10) +#define CHV_FGT_DISABLE_SS1 (1 << 11) +#define CHV_FGT_EU_DIS_SS0_R0_SHIFT 16 +#define CHV_FGT_EU_DIS_SS0_R0_MASK (0xf << CHV_FGT_EU_DIS_SS0_R0_SHIFT) +#define CHV_FGT_EU_DIS_SS0_R1_SHIFT 20 +#define CHV_FGT_EU_DIS_SS0_R1_MASK (0xf << CHV_FGT_EU_DIS_SS0_R1_SHIFT) +#define CHV_FGT_EU_DIS_SS1_R0_SHIFT 24 +#define CHV_FGT_EU_DIS_SS1_R0_MASK (0xf << CHV_FGT_EU_DIS_SS1_R0_SHIFT) +#define CHV_FGT_EU_DIS_SS1_R1_SHIFT 28 +#define CHV_FGT_EU_DIS_SS1_R1_MASK (0xf << CHV_FGT_EU_DIS_SS1_R1_SHIFT) + +#define GEN8_FUSE2 _MMIO(0x9120) +#define GEN8_F2_SS_DIS_SHIFT 21 +#define GEN8_F2_SS_DIS_MASK (0x7 << GEN8_F2_SS_DIS_SHIFT) +#define GEN8_F2_S_ENA_SHIFT 25 +#define GEN8_F2_S_ENA_MASK (0x7 << GEN8_F2_S_ENA_SHIFT) + +#define GEN9_F2_SS_DIS_SHIFT 20 +#define GEN9_F2_SS_DIS_MASK (0xf << GEN9_F2_SS_DIS_SHIFT) + +#define GEN10_F2_S_ENA_SHIFT 22 +#define GEN10_F2_S_ENA_MASK (0x3f << GEN10_F2_S_ENA_SHIFT) +#define GEN10_F2_SS_DIS_SHIFT 18 +#define GEN10_F2_SS_DIS_MASK (0xf << GEN10_F2_SS_DIS_SHIFT) + +#define GEN10_MIRROR_FUSE3 _MMIO(0x9118) +#define GEN10_L3BANK_PAIR_COUNT 4 +#define GEN10_L3BANK_MASK 0x0F +/* on Xe_HP the same fuses indicates mslices instead of L3 banks */ +#define GEN12_MAX_MSLICES 4 +#define GEN12_MEML3_EN_MASK 0x0F + +#define GEN8_EU_DISABLE0 _MMIO(0x9134) +#define GEN8_EU_DIS0_S0_MASK 0xffffff +#define GEN8_EU_DIS0_S1_SHIFT 24 +#define GEN8_EU_DIS0_S1_MASK (0xff << GEN8_EU_DIS0_S1_SHIFT) + +#define GEN8_EU_DISABLE1 _MMIO(0x9138) +#define GEN8_EU_DIS1_S1_MASK 0xffff +#define GEN8_EU_DIS1_S2_SHIFT 16 +#define GEN8_EU_DIS1_S2_MASK (0xffff << GEN8_EU_DIS1_S2_SHIFT) + +#define GEN8_EU_DISABLE2 _MMIO(0x913c) +#define GEN8_EU_DIS2_S2_MASK 0xff + +#define GEN9_EU_DISABLE(slice) _MMIO(0x9134 + (slice) * 0x4) + +#define GEN10_EU_DISABLE3 _MMIO(0x9140) +#define GEN10_EU_DIS_SS_MASK 0xff + +#define GEN11_GT_VEBOX_VDBOX_DISABLE _MMIO(0x9140) +#define GEN11_GT_VDBOX_DISABLE_MASK 0xff +#define GEN11_GT_VEBOX_DISABLE_SHIFT 16 +#define GEN11_GT_VEBOX_DISABLE_MASK (0x0f << GEN11_GT_VEBOX_DISABLE_SHIFT) + +#define GEN11_EU_DISABLE _MMIO(0x9134) +#define GEN11_EU_DIS_MASK 0xFF + +#define GEN11_GT_SLICE_ENABLE _MMIO(0x9138) +#define GEN11_GT_S_ENA_MASK 0xFF + +#define GEN11_GT_SUBSLICE_DISABLE _MMIO(0x913C) + +#define GEN12_GT_GEOMETRY_DSS_ENABLE _MMIO(0x913C) +#define GEN12_GT_COMPUTE_DSS_ENABLE _MMIO(0x9144) + +#define XEHP_EU_ENABLE _MMIO(0x9134) +#define XEHP_EU_ENA_MASK 0xFF + +#define CRSTANDVID _MMIO(0x11100) +#define PXVFREQ(fstart) _MMIO(0x11110 + (fstart) * 4) /* P[0-15]VIDFREQ (0x1114c) (Ironlake) */ +#define PXVFREQ_PX_MASK 0x7f000000 +#define PXVFREQ_PX_SHIFT 24 +#define VIDFREQ_BASE _MMIO(0x11110) +#define VIDFREQ1 _MMIO(0x11110) /* VIDFREQ1-4 (0x1111c) (Cantiga) */ +#define VIDFREQ2 _MMIO(0x11114) +#define VIDFREQ3 _MMIO(0x11118) +#define VIDFREQ4 _MMIO(0x1111c) +#define VIDFREQ_P0_MASK 0x1f000000 +#define VIDFREQ_P0_SHIFT 24 +#define VIDFREQ_P0_CSCLK_MASK 0x00f00000 +#define VIDFREQ_P0_CSCLK_SHIFT 20 +#define VIDFREQ_P0_CRCLK_MASK 0x000f0000 +#define VIDFREQ_P0_CRCLK_SHIFT 16 +#define VIDFREQ_P1_MASK 0x00001f00 +#define VIDFREQ_P1_SHIFT 8 +#define VIDFREQ_P1_CSCLK_MASK 0x000000f0 +#define VIDFREQ_P1_CSCLK_SHIFT 4 +#define VIDFREQ_P1_CRCLK_MASK 0x0000000f +#define INTTOEXT_BASE_ILK _MMIO(0x11300) +#define INTTOEXT_BASE _MMIO(0x11120) /* INTTOEXT1-8 (0x1113c) */ +#define INTTOEXT_MAP3_SHIFT 24 +#define INTTOEXT_MAP3_MASK (0x1f << INTTOEXT_MAP3_SHIFT) +#define INTTOEXT_MAP2_SHIFT 16 +#define INTTOEXT_MAP2_MASK (0x1f << INTTOEXT_MAP2_SHIFT) +#define INTTOEXT_MAP1_SHIFT 8 +#define INTTOEXT_MAP1_MASK (0x1f << INTTOEXT_MAP1_SHIFT) +#define INTTOEXT_MAP0_SHIFT 0 +#define INTTOEXT_MAP0_MASK (0x1f << INTTOEXT_MAP0_SHIFT) +#define MEMSWCTL _MMIO(0x11170) /* Ironlake only */ +#define MEMCTL_CMD_MASK 0xe000 +#define MEMCTL_CMD_SHIFT 13 +#define MEMCTL_CMD_RCLK_OFF 0 +#define MEMCTL_CMD_RCLK_ON 1 +#define MEMCTL_CMD_CHFREQ 2 +#define MEMCTL_CMD_CHVID 3 +#define MEMCTL_CMD_VMMOFF 4 +#define MEMCTL_CMD_VMMON 5 +#define MEMCTL_CMD_STS (1 << 12) /* write 1 triggers command, clears + when command complete */ +#define MEMCTL_FREQ_MASK 0x0f00 /* jitter, from 0-15 */ +#define MEMCTL_FREQ_SHIFT 8 +#define MEMCTL_SFCAVM (1 << 7) +#define MEMCTL_TGT_VID_MASK 0x007f +#define MEMIHYST _MMIO(0x1117c) +#define MEMINTREN _MMIO(0x11180) /* 16 bits */ +#define MEMINT_RSEXIT_EN (1 << 8) +#define MEMINT_CX_SUPR_EN (1 << 7) +#define MEMINT_CONT_BUSY_EN (1 << 6) +#define MEMINT_AVG_BUSY_EN (1 << 5) +#define MEMINT_EVAL_CHG_EN (1 << 4) +#define MEMINT_MON_IDLE_EN (1 << 3) +#define MEMINT_UP_EVAL_EN (1 << 2) +#define MEMINT_DOWN_EVAL_EN (1 << 1) +#define MEMINT_SW_CMD_EN (1 << 0) +#define MEMINTRSTR _MMIO(0x11182) /* 16 bits */ +#define MEM_RSEXIT_MASK 0xc000 +#define MEM_RSEXIT_SHIFT 14 +#define MEM_CONT_BUSY_MASK 0x3000 +#define MEM_CONT_BUSY_SHIFT 12 +#define MEM_AVG_BUSY_MASK 0x0c00 +#define MEM_AVG_BUSY_SHIFT 10 +#define MEM_EVAL_CHG_MASK 0x0300 +#define MEM_EVAL_BUSY_SHIFT 8 +#define MEM_MON_IDLE_MASK 0x00c0 +#define MEM_MON_IDLE_SHIFT 6 +#define MEM_UP_EVAL_MASK 0x0030 +#define MEM_UP_EVAL_SHIFT 4 +#define MEM_DOWN_EVAL_MASK 0x000c +#define MEM_DOWN_EVAL_SHIFT 2 +#define MEM_SW_CMD_MASK 0x0003 +#define MEM_INT_STEER_GFX 0 +#define MEM_INT_STEER_CMR 1 +#define MEM_INT_STEER_SMI 2 +#define MEM_INT_STEER_SCI 3 +#define MEMINTRSTS _MMIO(0x11184) +#define MEMINT_RSEXIT (1 << 7) +#define MEMINT_CONT_BUSY (1 << 6) +#define MEMINT_AVG_BUSY (1 << 5) +#define MEMINT_EVAL_CHG (1 << 4) +#define MEMINT_MON_IDLE (1 << 3) +#define MEMINT_UP_EVAL (1 << 2) +#define MEMINT_DOWN_EVAL (1 << 1) +#define MEMINT_SW_CMD (1 << 0) +#define MEMMODECTL _MMIO(0x11190) +#define MEMMODE_BOOST_EN (1 << 31) +#define MEMMODE_BOOST_FREQ_MASK 0x0f000000 /* jitter for boost, 0-15 */ +#define MEMMODE_BOOST_FREQ_SHIFT 24 +#define MEMMODE_IDLE_MODE_MASK 0x00030000 +#define MEMMODE_IDLE_MODE_SHIFT 16 +#define MEMMODE_IDLE_MODE_EVAL 0 +#define MEMMODE_IDLE_MODE_CONT 1 +#define MEMMODE_HWIDLE_EN (1 << 15) +#define MEMMODE_SWMODE_EN (1 << 14) +#define MEMMODE_RCLK_GATE (1 << 13) +#define MEMMODE_HW_UPDATE (1 << 12) +#define MEMMODE_FSTART_MASK 0x00000f00 /* starting jitter, 0-15 */ +#define MEMMODE_FSTART_SHIFT 8 +#define MEMMODE_FMAX_MASK 0x000000f0 /* max jitter, 0-15 */ +#define MEMMODE_FMAX_SHIFT 4 +#define MEMMODE_FMIN_MASK 0x0000000f /* min jitter, 0-15 */ +#define RCBMAXAVG _MMIO(0x1119c) +#define MEMSWCTL2 _MMIO(0x1119e) /* Cantiga only */ +#define SWMEMCMD_RENDER_OFF (0 << 13) +#define SWMEMCMD_RENDER_ON (1 << 13) +#define SWMEMCMD_SWFREQ (2 << 13) +#define SWMEMCMD_TARVID (3 << 13) +#define SWMEMCMD_VRM_OFF (4 << 13) +#define SWMEMCMD_VRM_ON (5 << 13) +#define CMDSTS (1 << 12) +#define SFCAVM (1 << 11) +#define SWFREQ_MASK 0x0380 /* P0-7 */ +#define SWFREQ_SHIFT 7 +#define TARVID_MASK 0x001f +#define MEMSTAT_CTG _MMIO(0x111a0) +#define RCBMINAVG _MMIO(0x111a0) +#define RCUPEI _MMIO(0x111b0) +#define RCDNEI _MMIO(0x111b4) +#define RSTDBYCTL _MMIO(0x111b8) +#define RS1EN (1 << 31) +#define RS2EN (1 << 30) +#define RS3EN (1 << 29) +#define D3RS3EN (1 << 28) /* Display D3 imlies RS3 */ +#define SWPROMORSX (1 << 27) /* RSx promotion timers ignored */ +#define RCWAKERW (1 << 26) /* Resetwarn from PCH causes wakeup */ +#define DPRSLPVREN (1 << 25) /* Fast voltage ramp enable */ +#define GFXTGHYST (1 << 24) /* Hysteresis to allow trunk gating */ +#define RCX_SW_EXIT (1 << 23) /* Leave RSx and prevent re-entry */ +#define RSX_STATUS_MASK (7 << 20) +#define RSX_STATUS_ON (0 << 20) +#define RSX_STATUS_RC1 (1 << 20) +#define RSX_STATUS_RC1E (2 << 20) +#define RSX_STATUS_RS1 (3 << 20) +#define RSX_STATUS_RS2 (4 << 20) /* aka rc6 */ +#define RSX_STATUS_RSVD (5 << 20) /* deep rc6 unsupported on ilk */ +#define RSX_STATUS_RS3 (6 << 20) /* rs3 unsupported on ilk */ +#define RSX_STATUS_RSVD2 (7 << 20) +#define UWRCRSXE (1 << 19) /* wake counter limit prevents rsx */ +#define RSCRP (1 << 18) /* rs requests control on rs1/2 reqs */ +#define JRSC (1 << 17) /* rsx coupled to cpu c-state */ +#define RS2INC0 (1 << 16) /* allow rs2 in cpu c0 */ +#define RS1CONTSAV_MASK (3 << 14) +#define RS1CONTSAV_NO_RS1 (0 << 14) /* rs1 doesn't save/restore context */ +#define RS1CONTSAV_RSVD (1 << 14) +#define RS1CONTSAV_SAVE_RS1 (2 << 14) /* rs1 saves context */ +#define RS1CONTSAV_FULL_RS1 (3 << 14) /* rs1 saves and restores context */ +#define NORMSLEXLAT_MASK (3 << 12) +#define SLOW_RS123 (0 << 12) +#define SLOW_RS23 (1 << 12) +#define SLOW_RS3 (2 << 12) +#define NORMAL_RS123 (3 << 12) +#define RCMODE_TIMEOUT (1 << 11) /* 0 is eval interval method */ +#define IMPROMOEN (1 << 10) /* promo is immediate or delayed until next idle interval (only for timeout method above) */ +#define RCENTSYNC (1 << 9) /* rs coupled to cpu c-state (3/6/7) */ +#define STATELOCK (1 << 7) /* locked to rs_cstate if 0 */ +#define RS_CSTATE_MASK (3 << 4) +#define RS_CSTATE_C367_RS1 (0 << 4) +#define RS_CSTATE_C36_RS1_C7_RS2 (1 << 4) +#define RS_CSTATE_RSVD (2 << 4) +#define RS_CSTATE_C367_RS2 (3 << 4) +#define REDSAVES (1 << 3) /* no context save if was idle during rs0 */ +#define REDRESTORES (1 << 2) /* no restore if was idle during rs0 */ +#define VIDCTL _MMIO(0x111c0) +#define VIDSTS _MMIO(0x111c8) +#define VIDSTART _MMIO(0x111cc) /* 8 bits */ +#define MEMSTAT_ILK _MMIO(0x111f8) +#define MEMSTAT_VID_MASK 0x7f00 +#define MEMSTAT_VID_SHIFT 8 +#define MEMSTAT_PSTATE_MASK 0x00f8 +#define MEMSTAT_PSTATE_SHIFT 3 +#define MEMSTAT_MON_ACTV (1 << 2) +#define MEMSTAT_SRC_CTL_MASK 0x0003 +#define MEMSTAT_SRC_CTL_CORE 0 +#define MEMSTAT_SRC_CTL_TRB 1 +#define MEMSTAT_SRC_CTL_THM 2 +#define MEMSTAT_SRC_CTL_STDBY 3 +#define RCPREVBSYTUPAVG _MMIO(0x113b8) +#define RCPREVBSYTDNAVG _MMIO(0x113bc) +#define PMMISC _MMIO(0x11214) +#define MCPPCE_EN (1 << 0) /* enable PM_MSG from PCH->MPC */ +#define SDEW _MMIO(0x1124c) +#define CSIEW0 _MMIO(0x11250) +#define CSIEW1 _MMIO(0x11254) +#define CSIEW2 _MMIO(0x11258) +#define PEW(i) _MMIO(0x1125c + (i) * 4) /* 5 registers */ +#define DEW(i) _MMIO(0x11270 + (i) * 4) /* 3 registers */ +#define MCHAFE _MMIO(0x112c0) +#define CSIEC _MMIO(0x112e0) +#define DMIEC _MMIO(0x112e4) +#define DDREC _MMIO(0x112e8) +#define PEG0EC _MMIO(0x112ec) +#define PEG1EC _MMIO(0x112f0) +#define GFXEC _MMIO(0x112f4) +#define RPPREVBSYTUPAVG _MMIO(0x113b8) +#define RPPREVBSYTDNAVG _MMIO(0x113bc) +#define ECR _MMIO(0x11600) +#define ECR_GPFE (1 << 31) +#define ECR_IMONE (1 << 30) +#define ECR_CAP_MASK 0x0000001f /* Event range, 0-31 */ +#define OGW0 _MMIO(0x11608) +#define OGW1 _MMIO(0x1160c) +#define EG0 _MMIO(0x11610) +#define EG1 _MMIO(0x11614) +#define EG2 _MMIO(0x11618) +#define EG3 _MMIO(0x1161c) +#define EG4 _MMIO(0x11620) +#define EG5 _MMIO(0x11624) +#define EG6 _MMIO(0x11628) +#define EG7 _MMIO(0x1162c) +#define PXW(i) _MMIO(0x11664 + (i) * 4) /* 4 registers */ +#define PXWL(i) _MMIO(0x11680 + (i) * 8) /* 8 registers */ +#define LCFUSE02 _MMIO(0x116c0) +#define LCFUSE_HIV_MASK 0x000000ff + +#define GEN6_GT_THREAD_STATUS_REG _MMIO(0x13805c) +#define GEN6_GT_THREAD_STATUS_CORE_MASK 0x7 + +/* + * Logical Context regs + */ +/* + * Notes on SNB/IVB/VLV context size: + * - Power context is saved elsewhere (LLC or stolen) + * - Ring/execlist context is saved on SNB, not on IVB + * - Extended context size already includes render context size + * - We always need to follow the extended context size. + * SNB BSpec has comments indicating that we should use the + * render context size instead if execlists are disabled, but + * based on empirical testing that's just nonsense. + * - Pipelined/VF state is saved on SNB/IVB respectively + * - GT1 size just indicates how much of render context + * doesn't need saving on GT1 + */ +#define CXT_SIZE _MMIO(0x21a0) +#define GEN6_CXT_POWER_SIZE(cxt_reg) (((cxt_reg) >> 24) & 0x3f) +#define GEN6_CXT_RING_SIZE(cxt_reg) (((cxt_reg) >> 18) & 0x3f) +#define GEN6_CXT_RENDER_SIZE(cxt_reg) (((cxt_reg) >> 12) & 0x3f) +#define GEN6_CXT_EXTENDED_SIZE(cxt_reg) (((cxt_reg) >> 6) & 0x3f) +#define GEN6_CXT_PIPELINE_SIZE(cxt_reg) (((cxt_reg) >> 0) & 0x3f) +#define GEN6_CXT_TOTAL_SIZE(cxt_reg) (GEN6_CXT_RING_SIZE(cxt_reg) + \ + GEN6_CXT_EXTENDED_SIZE(cxt_reg) + \ + GEN6_CXT_PIPELINE_SIZE(cxt_reg)) +#define GEN7_CXT_SIZE _MMIO(0x21a8) +#define GEN7_CXT_POWER_SIZE(ctx_reg) (((ctx_reg) >> 25) & 0x7f) +#define GEN7_CXT_RING_SIZE(ctx_reg) (((ctx_reg) >> 22) & 0x7) +#define GEN7_CXT_RENDER_SIZE(ctx_reg) (((ctx_reg) >> 16) & 0x3f) +#define GEN7_CXT_EXTENDED_SIZE(ctx_reg) (((ctx_reg) >> 9) & 0x7f) +#define GEN7_CXT_GT1_SIZE(ctx_reg) (((ctx_reg) >> 6) & 0x7) +#define GEN7_CXT_VFSTATE_SIZE(ctx_reg) (((ctx_reg) >> 0) & 0x3f) +#define GEN7_CXT_TOTAL_SIZE(ctx_reg) (GEN7_CXT_EXTENDED_SIZE(ctx_reg) + \ + GEN7_CXT_VFSTATE_SIZE(ctx_reg)) + +enum { + INTEL_ADVANCED_CONTEXT = 0, + INTEL_LEGACY_32B_CONTEXT, + INTEL_ADVANCED_AD_CONTEXT, + INTEL_LEGACY_64B_CONTEXT +}; + +enum { + FAULT_AND_HANG = 0, + FAULT_AND_HALT, /* Debug only */ + FAULT_AND_STREAM, + FAULT_AND_CONTINUE /* Unsupported */ +}; + +#define CTX_GTT_ADDRESS_MASK GENMASK(31, 12) +#define GEN8_CTX_VALID (1 << 0) +#define GEN8_CTX_FORCE_PD_RESTORE (1 << 1) +#define GEN8_CTX_FORCE_RESTORE (1 << 2) +#define GEN8_CTX_L3LLC_COHERENT (1 << 5) +#define GEN8_CTX_PRIVILEGE (1 << 8) +#define GEN8_CTX_ADDRESSING_MODE_SHIFT 3 + +#define GEN8_CTX_ID_SHIFT 32 +#define GEN8_CTX_ID_WIDTH 21 +#define GEN11_SW_CTX_ID_SHIFT 37 +#define GEN11_SW_CTX_ID_WIDTH 11 +#define GEN11_ENGINE_CLASS_SHIFT 61 +#define GEN11_ENGINE_CLASS_WIDTH 3 +#define GEN11_ENGINE_INSTANCE_SHIFT 48 +#define GEN11_ENGINE_INSTANCE_WIDTH 6 + +#define XEHP_SW_CTX_ID_SHIFT 39 +#define XEHP_SW_CTX_ID_WIDTH 16 +#define XEHP_SW_COUNTER_SHIFT 58 +#define XEHP_SW_COUNTER_WIDTH 6 + +#define UNSLCGCTL9440 _MMIO(0x9440) +#define GAMTLBOACS_CLKGATE_DIS REG_BIT(28) +#define GAMTLBVDBOX5_CLKGATE_DIS REG_BIT(27) +#define GAMTLBVDBOX6_CLKGATE_DIS REG_BIT(26) +#define GAMTLBVDBOX3_CLKGATE_DIS REG_BIT(24) +#define GAMTLBVDBOX4_CLKGATE_DIS REG_BIT(23) +#define GAMTLBVDBOX7_CLKGATE_DIS REG_BIT(22) +#define GAMTLBVDBOX2_CLKGATE_DIS REG_BIT(21) +#define GAMTLBVDBOX0_CLKGATE_DIS REG_BIT(17) +#define GAMTLBKCR_CLKGATE_DIS REG_BIT(16) +#define GAMTLBGUC_CLKGATE_DIS REG_BIT(15) +#define GAMTLBBLT_CLKGATE_DIS REG_BIT(14) +#define GAMTLBVDBOX1_CLKGATE_DIS REG_BIT(6) + +#define UNSLCGCTL9444 _MMIO(0x9444) +#define GAMTLBGFXA0_CLKGATE_DIS REG_BIT(30) +#define GAMTLBGFXA1_CLKGATE_DIS REG_BIT(29) +#define GAMTLBCOMPA0_CLKGATE_DIS REG_BIT(28) +#define GAMTLBCOMPA1_CLKGATE_DIS REG_BIT(27) +#define GAMTLBCOMPB0_CLKGATE_DIS REG_BIT(26) +#define GAMTLBCOMPB1_CLKGATE_DIS REG_BIT(25) +#define GAMTLBCOMPC0_CLKGATE_DIS REG_BIT(24) +#define GAMTLBCOMPC1_CLKGATE_DIS REG_BIT(23) +#define GAMTLBCOMPD0_CLKGATE_DIS REG_BIT(22) +#define GAMTLBCOMPD1_CLKGATE_DIS REG_BIT(21) +#define GAMTLBMERT_CLKGATE_DIS REG_BIT(20) +#define GAMTLBVEBOX3_CLKGATE_DIS REG_BIT(19) +#define GAMTLBVEBOX2_CLKGATE_DIS REG_BIT(18) +#define GAMTLBVEBOX1_CLKGATE_DIS REG_BIT(17) +#define GAMTLBVEBOX0_CLKGATE_DIS REG_BIT(16) +#define LTCDD_CLKGATE_DIS REG_BIT(10) + +#define SLICE_UNIT_LEVEL_CLKGATE _MMIO(0x94d4) +#define SARBUNIT_CLKGATE_DIS (1 << 5) +#define RCCUNIT_CLKGATE_DIS (1 << 7) +#define MSCUNIT_CLKGATE_DIS (1 << 10) +#define NODEDSS_CLKGATE_DIS REG_BIT(12) +#define L3_CLKGATE_DIS REG_BIT(16) +#define L3_CR2X_CLKGATE_DIS REG_BIT(17) + +#define SUBSLICE_UNIT_LEVEL_CLKGATE _MMIO(0x9524) +#define DSS_ROUTER_CLKGATE_DIS REG_BIT(28) +#define GWUNIT_CLKGATE_DIS REG_BIT(16) + +#define SUBSLICE_UNIT_LEVEL_CLKGATE2 _MMIO(0x9528) +#define CPSSUNIT_CLKGATE_DIS REG_BIT(9) + +#define SSMCGCTL9530 _MMIO(0x9530) +#define RTFUNIT_CLKGATE_DIS REG_BIT(18) + +#define UNSLICE_UNIT_LEVEL_CLKGATE _MMIO(0x9434) +#define VFUNIT_CLKGATE_DIS REG_BIT(20) +#define TSGUNIT_CLKGATE_DIS REG_BIT(17) /* XEHPSDV */ +#define CG3DDISCFEG_CLKGATE_DIS REG_BIT(17) /* DG2 */ +#define GAMEDIA_CLKGATE_DIS REG_BIT(11) +#define HSUNIT_CLKGATE_DIS REG_BIT(8) +#define VSUNIT_CLKGATE_DIS REG_BIT(3) + +#define UNSLICE_UNIT_LEVEL_CLKGATE2 _MMIO(0x94e4) +#define VSUNIT_CLKGATE_DIS_TGL REG_BIT(19) +#define PSDUNIT_CLKGATE_DIS REG_BIT(5) + +#define INF_UNIT_LEVEL_CLKGATE _MMIO(0x9560) +#define CGPSF_CLKGATE_DIS (1 << 3) + +#define GEN11_GT_INTR_DW0 _MMIO(0x190018) +#define GEN11_CSME (31) +#define GEN11_GUNIT (28) +#define GEN11_GUC (25) +#define GEN11_WDPERF (20) +#define GEN11_KCR (19) +#define GEN11_GTPM (16) +#define GEN11_BCS (15) +#define GEN11_RCS0 (0) + +#define GEN11_GT_INTR_DW1 _MMIO(0x19001c) +#define GEN11_VECS(x) (31 - (x)) +#define GEN11_VCS(x) (x) + +#define GEN11_GT_INTR_DW(x) _MMIO(0x190018 + ((x) * 4)) + +#define GEN11_INTR_IDENTITY_REG0 _MMIO(0x190060) +#define GEN11_INTR_IDENTITY_REG1 _MMIO(0x190064) +#define GEN11_INTR_DATA_VALID (1 << 31) +#define GEN11_INTR_ENGINE_CLASS(x) (((x) & GENMASK(18, 16)) >> 16) +#define GEN11_INTR_ENGINE_INSTANCE(x) (((x) & GENMASK(25, 20)) >> 20) +#define GEN11_INTR_ENGINE_INTR(x) ((x) & 0xffff) +/* irq instances for OTHER_CLASS */ +#define OTHER_GUC_INSTANCE 0 +#define OTHER_GTPM_INSTANCE 1 +#define OTHER_KCR_INSTANCE 4 + +#define GEN11_INTR_IDENTITY_REG(x) _MMIO(0x190060 + ((x) * 4)) + +#define GEN11_IIR_REG0_SELECTOR _MMIO(0x190070) +#define GEN11_IIR_REG1_SELECTOR _MMIO(0x190074) + +#define GEN11_IIR_REG_SELECTOR(x) _MMIO(0x190070 + ((x) * 4)) + +#define GEN11_RENDER_COPY_INTR_ENABLE _MMIO(0x190030) +#define GEN11_VCS_VECS_INTR_ENABLE _MMIO(0x190034) +#define GEN11_GUC_SG_INTR_ENABLE _MMIO(0x190038) +#define GEN11_GPM_WGBOXPERF_INTR_ENABLE _MMIO(0x19003c) +#define GEN11_CRYPTO_RSVD_INTR_ENABLE _MMIO(0x190040) +#define GEN11_GUNIT_CSME_INTR_ENABLE _MMIO(0x190044) + +#define GEN11_RCS0_RSVD_INTR_MASK _MMIO(0x190090) +#define GEN11_BCS_RSVD_INTR_MASK _MMIO(0x1900a0) +#define GEN11_VCS0_VCS1_INTR_MASK _MMIO(0x1900a8) +#define GEN11_VCS2_VCS3_INTR_MASK _MMIO(0x1900ac) +#define GEN12_VCS4_VCS5_INTR_MASK _MMIO(0x1900b0) +#define GEN12_VCS6_VCS7_INTR_MASK _MMIO(0x1900b4) +#define GEN11_VECS0_VECS1_INTR_MASK _MMIO(0x1900d0) +#define GEN12_VECS2_VECS3_INTR_MASK _MMIO(0x1900d4) +#define GEN11_GUC_SG_INTR_MASK _MMIO(0x1900e8) +#define GEN11_GPM_WGBOXPERF_INTR_MASK _MMIO(0x1900ec) +#define GEN11_CRYPTO_RSVD_INTR_MASK _MMIO(0x1900f0) +#define GEN11_GUNIT_CSME_INTR_MASK _MMIO(0x1900f4) + +#define ENGINE1_MASK REG_GENMASK(31, 16) +#define ENGINE0_MASK REG_GENMASK(15, 0) + +#define GEN7_FF_SLICE_CS_CHICKEN1 _MMIO(0x20e0) +#define GEN9_FFSC_PERCTX_PREEMPT_CTRL (1 << 14) + +#define FF_SLICE_CS_CHICKEN2 _MMIO(0x20e4) +#define GEN9_TSG_BARRIER_ACK_DISABLE (1 << 8) +#define GEN9_POOLED_EU_LOAD_BALANCING_FIX_DISABLE (1 << 10) + +#define GEN9_CS_DEBUG_MODE1 _MMIO(0x20ec) +#define FF_DOP_CLOCK_GATE_DISABLE REG_BIT(1) +#define GEN9_CTX_PREEMPT_REG _MMIO(0x2248) +#define GEN12_DISABLE_POSH_BUSY_FF_DOP_CG REG_BIT(11) + +#define GEN12_CS_DEBUG_MODE1_CCCSUNIT_BE_COMMON _MMIO(0x20EC) +#define GEN12_REPLAY_MODE_GRANULARITY REG_BIT(0) + +#define GEN8_CS_CHICKEN1 _MMIO(0x2580) +#define GEN9_PREEMPT_3D_OBJECT_LEVEL (1 << 0) +#define GEN9_PREEMPT_GPGPU_LEVEL(hi, lo) (((hi) << 2) | ((lo) << 1)) +#define GEN9_PREEMPT_GPGPU_MID_THREAD_LEVEL GEN9_PREEMPT_GPGPU_LEVEL(0, 0) +#define GEN9_PREEMPT_GPGPU_THREAD_GROUP_LEVEL GEN9_PREEMPT_GPGPU_LEVEL(0, 1) +#define GEN9_PREEMPT_GPGPU_COMMAND_LEVEL GEN9_PREEMPT_GPGPU_LEVEL(1, 0) +#define GEN9_PREEMPT_GPGPU_LEVEL_MASK GEN9_PREEMPT_GPGPU_LEVEL(1, 1) + +/* GEN7 chicken */ +#define GEN7_COMMON_SLICE_CHICKEN1 _MMIO(0x7010) + #define GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC (1 << 10) + #define GEN9_RHWO_OPTIMIZATION_DISABLE (1 << 14) + +#define COMMON_SLICE_CHICKEN2 _MMIO(0x7014) + #define GEN9_PBE_COMPRESSED_HASH_SELECTION (1 << 13) + #define GEN9_DISABLE_GATHER_AT_SET_SHADER_COMMON_SLICE (1 << 12) + #define GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION (1 << 8) + #define GEN8_CSC2_SBE_VUE_CACHE_CONSERVATIVE (1 << 0) + +#define GEN8_L3CNTLREG _MMIO(0x7034) + #define GEN8_ERRDETBCTRL (1 << 9) + +#define GEN11_COMMON_SLICE_CHICKEN3 _MMIO(0x7304) +#define DG1_FLOAT_POINT_BLEND_OPT_STRICT_MODE_EN REG_BIT(12) +#define XEHP_DUAL_SIMD8_SEQ_MERGE_DISABLE REG_BIT(12) +#define GEN11_BLEND_EMB_FIX_DISABLE_IN_RCC REG_BIT(11) +#define GEN12_DISABLE_CPS_AWARE_COLOR_PIPE REG_BIT(9) + +#define HIZ_CHICKEN _MMIO(0x7018) +# define CHV_HZ_8X8_MODE_IN_1X REG_BIT(15) +# define DG1_HZ_READ_SUPPRESSION_OPTIMIZATION_DISABLE REG_BIT(14) +# define BDW_HIZ_POWER_COMPILER_CLOCK_GATING_DISABLE REG_BIT(3) + +#define GEN9_SLICE_COMMON_ECO_CHICKEN0 _MMIO(0x7308) +#define DISABLE_PIXEL_MASK_CAMMING (1 << 14) + +#define GEN9_SLICE_COMMON_ECO_CHICKEN1 _MMIO(0x731c) +#define GEN11_STATE_CACHE_REDIRECT_TO_CS (1 << 11) + +#define GEN7_SARCHKMD _MMIO(0xB000) +#define GEN7_DISABLE_DEMAND_PREFETCH (1 << 31) +#define GEN7_DISABLE_SAMPLER_PREFETCH (1 << 30) + +#define GEN7_L3SQCREG1 _MMIO(0xB010) +#define VLV_B0_WA_L3SQCREG1_VALUE 0x00D30000 + +#define GEN8_L3SQCREG1 _MMIO(0xB100) +/* + * Note that on CHV the following has an off-by-one error wrt. to BSpec. + * Using the formula in BSpec leads to a hang, while the formula here works + * fine and matches the formulas for all other platforms. A BSpec change + * request has been filed to clarify this. + */ +#define L3_GENERAL_PRIO_CREDITS(x) (((x) >> 1) << 19) +#define L3_HIGH_PRIO_CREDITS(x) (((x) >> 1) << 14) +#define L3_PRIO_CREDITS_MASK ((0x1f << 19) | (0x1f << 14)) + +#define GEN7_L3CNTLREG1 _MMIO(0xB01C) +#define GEN7_WA_FOR_GEN7_L3_CONTROL 0x3C47FF8C +#define GEN7_L3AGDIS (1 << 19) +#define GEN7_L3CNTLREG2 _MMIO(0xB020) +#define GEN7_L3CNTLREG3 _MMIO(0xB024) + +#define GEN7_L3_CHICKEN_MODE_REGISTER _MMIO(0xB030) +#define GEN7_WA_L3_CHICKEN_MODE 0x20000000 +#define GEN10_L3_CHICKEN_MODE_REGISTER _MMIO(0xB114) +#define GEN11_I2M_WRITE_DISABLE (1 << 28) + +#define GEN7_L3SQCREG4 _MMIO(0xb034) +#define L3SQ_URB_READ_CAM_MATCH_DISABLE (1 << 27) + +#define GEN11_SCRATCH2 _MMIO(0xb140) +#define GEN11_COHERENT_PARTIAL_WRITE_MERGE_ENABLE (1 << 19) + +#define GEN8_L3SQCREG4 _MMIO(0xb118) +#define GEN11_LQSC_CLEAN_EVICT_DISABLE (1 << 6) +#define GEN8_LQSC_RO_PERF_DIS (1 << 27) +#define GEN8_LQSC_FLUSH_COHERENT_LINES (1 << 21) +#define GEN8_LQSQ_NONIA_COHERENT_ATOMICS_ENABLE REG_BIT(22) + +#define GEN11_L3SQCREG5 _MMIO(0xb158) +#define L3_PWM_TIMER_INIT_VAL_MASK REG_GENMASK(9, 0) + +#define XEHP_L3SCQREG7 _MMIO(0xb188) +#define BLEND_FILL_CACHING_OPT_DIS REG_BIT(3) + +/* GEN8 chicken */ +#define HDC_CHICKEN0 _MMIO(0x7300) +#define ICL_HDC_MODE _MMIO(0xE5F4) +#define HDC_FORCE_CSR_NON_COHERENT_OVR_DISABLE (1 << 15) +#define HDC_FENCE_DEST_SLM_DISABLE (1 << 14) +#define HDC_DONOT_FETCH_MEM_WHEN_MASKED (1 << 11) +#define HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT (1 << 5) +#define HDC_FORCE_NON_COHERENT (1 << 4) +#define HDC_BARRIER_PERFORMANCE_DISABLE (1 << 10) + +#define GEN12_HDC_CHICKEN0 _MMIO(0xE5F0) +#define LSC_L1_FLUSH_CTL_3D_DATAPORT_FLUSH_EVENTS_MASK REG_GENMASK(13, 11) + +#define SARB_CHICKEN1 _MMIO(0xe90c) +#define COMP_CKN_IN REG_GENMASK(30, 29) + +#define GEN8_HDC_CHICKEN1 _MMIO(0x7304) + +/* GEN9 chicken */ +#define SLICE_ECO_CHICKEN0 _MMIO(0x7308) +#define PIXEL_MASK_CAMMING_DISABLE (1 << 14) + +#define GEN9_WM_CHICKEN3 _MMIO(0x5588) +#define GEN9_FACTOR_IN_CLR_VAL_HIZ (1 << 9) + +/* WaCatErrorRejectionIssue */ +#define GEN7_SQ_CHICKEN_MBCUNIT_CONFIG _MMIO(0x9030) +#define GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB (1 << 11) + +#define HSW_SCRATCH1 _MMIO(0xb038) +#define HSW_SCRATCH1_L3_DATA_ATOMICS_DISABLE (1 << 27) + +#define BDW_SCRATCH1 _MMIO(0xb11c) +#define GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE (1 << 2) + +#define VFLSKPD _MMIO(0x62a8) +#define DIS_OVER_FETCH_CACHE REG_BIT(1) +#define DIS_MULT_MISS_RD_SQUASH REG_BIT(0) + +#define FF_MODE2 _MMIO(0x6604) +#define FF_MODE2_GS_TIMER_MASK REG_GENMASK(31, 24) +#define FF_MODE2_GS_TIMER_224 REG_FIELD_PREP(FF_MODE2_GS_TIMER_MASK, 224) +#define FF_MODE2_TDS_TIMER_MASK REG_GENMASK(23, 16) +#define FF_MODE2_TDS_TIMER_128 REG_FIELD_PREP(FF_MODE2_TDS_TIMER_MASK, 4) + +#define RC6_LOCATION _MMIO(0xD40) +#define RC6_CTX_IN_DRAM (1 << 0) +#define RC6_CTX_BASE _MMIO(0xD48) +#define RC6_CTX_BASE_MASK 0xFFFFFFF0 +#define FORCEWAKE _MMIO(0xA18C) +#define FORCEWAKE_VLV _MMIO(0x1300b0) +#define FORCEWAKE_ACK_VLV _MMIO(0x1300b4) +#define FORCEWAKE_MEDIA_VLV _MMIO(0x1300b8) +#define FORCEWAKE_ACK_MEDIA_VLV _MMIO(0x1300bc) +#define FORCEWAKE_ACK_HSW _MMIO(0x130044) +#define FORCEWAKE_ACK _MMIO(0x130090) +#define VLV_GTLC_WAKE_CTRL _MMIO(0x130090) +#define VLV_GTLC_RENDER_CTX_EXISTS (1 << 25) +#define VLV_GTLC_MEDIA_CTX_EXISTS (1 << 24) +#define VLV_GTLC_ALLOWWAKEREQ (1 << 0) + +#define VLV_GTLC_PW_STATUS _MMIO(0x130094) +#define VLV_GTLC_ALLOWWAKEACK (1 << 0) +#define VLV_GTLC_ALLOWWAKEERR (1 << 1) +#define VLV_GTLC_PW_MEDIA_STATUS_MASK (1 << 5) +#define VLV_GTLC_PW_RENDER_STATUS_MASK (1 << 7) +#define FORCEWAKE_MT _MMIO(0xa188) /* multi-threaded */ +#define FORCEWAKE_MEDIA_GEN9 _MMIO(0xa270) +#define FORCEWAKE_MEDIA_VDBOX_GEN11(n) _MMIO(0xa540 + (n) * 4) +#define FORCEWAKE_MEDIA_VEBOX_GEN11(n) _MMIO(0xa560 + (n) * 4) +#define FORCEWAKE_RENDER_GEN9 _MMIO(0xa278) +#define FORCEWAKE_GT_GEN9 _MMIO(0xa188) +#define FORCEWAKE_ACK_MEDIA_GEN9 _MMIO(0x0D88) +#define FORCEWAKE_ACK_MEDIA_VDBOX_GEN11(n) _MMIO(0x0D50 + (n) * 4) +#define FORCEWAKE_ACK_MEDIA_VEBOX_GEN11(n) _MMIO(0x0D70 + (n) * 4) +#define FORCEWAKE_ACK_RENDER_GEN9 _MMIO(0x0D84) +#define FORCEWAKE_ACK_GT_GEN9 _MMIO(0x130044) +#define FORCEWAKE_KERNEL BIT(0) +#define FORCEWAKE_USER BIT(1) +#define FORCEWAKE_KERNEL_FALLBACK BIT(15) +#define FORCEWAKE_MT_ACK _MMIO(0x130040) +#define ECOBUS _MMIO(0xa180) +#define FORCEWAKE_MT_ENABLE (1 << 5) +#define VLV_SPAREG2H _MMIO(0xA194) +#define GEN9_PWRGT_DOMAIN_STATUS _MMIO(0xA2A0) +#define GEN9_PWRGT_MEDIA_STATUS_MASK (1 << 0) +#define GEN9_PWRGT_RENDER_STATUS_MASK (1 << 1) + +#define GTFIFODBG _MMIO(0x120000) +#define GT_FIFO_SBDEDICATE_FREE_ENTRY_CHV (0x1f << 20) +#define GT_FIFO_FREE_ENTRIES_CHV (0x7f << 13) +#define GT_FIFO_SBDROPERR (1 << 6) +#define GT_FIFO_BLOBDROPERR (1 << 5) +#define GT_FIFO_SB_READ_ABORTERR (1 << 4) +#define GT_FIFO_DROPERR (1 << 3) +#define GT_FIFO_OVFERR (1 << 2) +#define GT_FIFO_IAWRERR (1 << 1) +#define GT_FIFO_IARDERR (1 << 0) + +#define GTFIFOCTL _MMIO(0x120008) +#define GT_FIFO_FREE_ENTRIES_MASK 0x7f +#define GT_FIFO_NUM_RESERVED_ENTRIES 20 +#define GT_FIFO_CTL_BLOCK_ALL_POLICY_STALL (1 << 12) +#define GT_FIFO_CTL_RC6_POLICY_STALL (1 << 11) + +#define HSW_IDICR _MMIO(0x9008) +#define IDIHASHMSK(x) (((x) & 0x3f) << 16) + +#define GEN6_UCGCTL1 _MMIO(0x9400) +# define GEN6_GAMUNIT_CLOCK_GATE_DISABLE (1 << 22) +# define GEN6_EU_TCUNIT_CLOCK_GATE_DISABLE (1 << 16) +# define GEN6_BLBUNIT_CLOCK_GATE_DISABLE (1 << 5) +# define GEN6_CSUNIT_CLOCK_GATE_DISABLE (1 << 7) + +#define GEN6_UCGCTL2 _MMIO(0x9404) +# define GEN6_VFUNIT_CLOCK_GATE_DISABLE (1 << 31) +# define GEN7_VDSUNIT_CLOCK_GATE_DISABLE (1 << 30) +# define GEN7_TDLUNIT_CLOCK_GATE_DISABLE (1 << 22) +# define GEN6_RCZUNIT_CLOCK_GATE_DISABLE (1 << 13) +# define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE (1 << 12) +# define GEN6_RCCUNIT_CLOCK_GATE_DISABLE (1 << 11) + +#define GEN6_UCGCTL3 _MMIO(0x9408) +# define GEN6_OACSUNIT_CLOCK_GATE_DISABLE (1 << 20) + +#define GEN7_UCGCTL4 _MMIO(0x940c) +#define GEN7_L3BANK2X_CLOCK_GATE_DISABLE (1 << 25) +#define GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE (1 << 14) + +#define GEN6_RCGCTL1 _MMIO(0x9410) +#define GEN6_RCGCTL2 _MMIO(0x9414) +#define GEN6_RSTCTL _MMIO(0x9420) + +#define GEN8_UCGCTL6 _MMIO(0x9430) +#define GEN8_GAPSUNIT_CLOCK_GATE_DISABLE (1 << 24) +#define GEN8_SDEUNIT_CLOCK_GATE_DISABLE (1 << 14) +#define GEN8_HDCUNIT_CLOCK_GATE_DISABLE_HDCREQ (1 << 28) + +#define UNSLCGCTL9430 _MMIO(0x9430) +#define MSQDUNIT_CLKGATE_DIS REG_BIT(3) + +#define GEN6_GFXPAUSE _MMIO(0xA000) +#define GEN6_RPNSWREQ _MMIO(0xA008) +#define GEN6_TURBO_DISABLE (1 << 31) +#define GEN6_FREQUENCY(x) ((x) << 25) +#define HSW_FREQUENCY(x) ((x) << 24) +#define GEN9_FREQUENCY(x) ((x) << 23) +#define GEN6_OFFSET(x) ((x) << 19) +#define GEN6_AGGRESSIVE_TURBO (0 << 15) +#define GEN9_SW_REQ_UNSLICE_RATIO_SHIFT 23 +#define GEN9_IGNORE_SLICE_RATIO (0 << 0) + +#define GEN6_RC_VIDEO_FREQ _MMIO(0xA00C) +#define GEN6_RC_CONTROL _MMIO(0xA090) +#define GEN6_RC_CTL_RC6pp_ENABLE (1 << 16) +#define GEN6_RC_CTL_RC6p_ENABLE (1 << 17) +#define GEN6_RC_CTL_RC6_ENABLE (1 << 18) +#define GEN6_RC_CTL_RC1e_ENABLE (1 << 20) +#define GEN6_RC_CTL_RC7_ENABLE (1 << 22) +#define VLV_RC_CTL_CTX_RST_PARALLEL (1 << 24) +#define GEN7_RC_CTL_TO_MODE (1 << 28) +#define GEN6_RC_CTL_EI_MODE(x) ((x) << 27) +#define GEN6_RC_CTL_HW_ENABLE (1 << 31) +#define GEN6_RP_DOWN_TIMEOUT _MMIO(0xA010) +#define GEN6_RP_INTERRUPT_LIMITS _MMIO(0xA014) +#define GEN6_RPSTAT1 _MMIO(0xA01C) +#define GEN6_CAGF_SHIFT 8 +#define HSW_CAGF_SHIFT 7 +#define GEN9_CAGF_SHIFT 23 +#define GEN6_CAGF_MASK (0x7f << GEN6_CAGF_SHIFT) +#define HSW_CAGF_MASK (0x7f << HSW_CAGF_SHIFT) +#define GEN9_CAGF_MASK (0x1ff << GEN9_CAGF_SHIFT) +#define GEN6_RP_CONTROL _MMIO(0xA024) +#define GEN6_RP_MEDIA_TURBO (1 << 11) +#define GEN6_RP_MEDIA_MODE_MASK (3 << 9) +#define GEN6_RP_MEDIA_HW_TURBO_MODE (3 << 9) +#define GEN6_RP_MEDIA_HW_NORMAL_MODE (2 << 9) +#define GEN6_RP_MEDIA_HW_MODE (1 << 9) +#define GEN6_RP_MEDIA_SW_MODE (0 << 9) +#define GEN6_RP_MEDIA_IS_GFX (1 << 8) +#define GEN6_RP_ENABLE (1 << 7) +#define GEN6_RP_UP_IDLE_MIN (0x1 << 3) +#define GEN6_RP_UP_BUSY_AVG (0x2 << 3) +#define GEN6_RP_UP_BUSY_CONT (0x4 << 3) +#define GEN6_RP_DOWN_IDLE_AVG (0x2 << 0) +#define GEN6_RP_DOWN_IDLE_CONT (0x1 << 0) +#define GEN6_RPSWCTL_SHIFT 9 +#define GEN9_RPSWCTL_ENABLE (0x2 << GEN6_RPSWCTL_SHIFT) +#define GEN9_RPSWCTL_DISABLE (0x0 << GEN6_RPSWCTL_SHIFT) +#define GEN6_RP_UP_THRESHOLD _MMIO(0xA02C) +#define GEN6_RP_DOWN_THRESHOLD _MMIO(0xA030) +#define GEN6_RP_CUR_UP_EI _MMIO(0xA050) +#define GEN6_RP_EI_MASK 0xffffff +#define GEN6_CURICONT_MASK GEN6_RP_EI_MASK +#define GEN6_RP_CUR_UP _MMIO(0xA054) +#define GEN6_CURBSYTAVG_MASK GEN6_RP_EI_MASK +#define GEN6_RP_PREV_UP _MMIO(0xA058) +#define GEN6_RP_CUR_DOWN_EI _MMIO(0xA05C) +#define GEN6_CURIAVG_MASK GEN6_RP_EI_MASK +#define GEN6_RP_CUR_DOWN _MMIO(0xA060) +#define GEN6_RP_PREV_DOWN _MMIO(0xA064) +#define GEN6_RP_UP_EI _MMIO(0xA068) +#define GEN6_RP_DOWN_EI _MMIO(0xA06C) +#define GEN6_RP_IDLE_HYSTERSIS _MMIO(0xA070) +#define GEN6_RPDEUHWTC _MMIO(0xA080) +#define GEN6_RPDEUC _MMIO(0xA084) +#define GEN6_RPDEUCSW _MMIO(0xA088) +#define GEN6_RC_STATE _MMIO(0xA094) +#define RC_SW_TARGET_STATE_SHIFT 16 +#define RC_SW_TARGET_STATE_MASK (7 << RC_SW_TARGET_STATE_SHIFT) +#define GEN6_RC1_WAKE_RATE_LIMIT _MMIO(0xA098) +#define GEN6_RC6_WAKE_RATE_LIMIT _MMIO(0xA09C) +#define GEN6_RC6pp_WAKE_RATE_LIMIT _MMIO(0xA0A0) +#define GEN10_MEDIA_WAKE_RATE_LIMIT _MMIO(0xA0A0) +#define GEN6_RC_EVALUATION_INTERVAL _MMIO(0xA0A8) +#define GEN6_RC_IDLE_HYSTERSIS _MMIO(0xA0AC) +#define GEN6_RC_SLEEP _MMIO(0xA0B0) +#define GEN6_RCUBMABDTMR _MMIO(0xA0B0) +#define GEN6_RC1e_THRESHOLD _MMIO(0xA0B4) +#define GEN6_RC6_THRESHOLD _MMIO(0xA0B8) +#define GEN6_RC6p_THRESHOLD _MMIO(0xA0BC) +#define VLV_RCEDATA _MMIO(0xA0BC) +#define GEN6_RC6pp_THRESHOLD _MMIO(0xA0C0) +#define GEN6_PMINTRMSK _MMIO(0xA168) +#define GEN8_PMINTR_DISABLE_REDIRECT_TO_GUC (1 << 31) +#define ARAT_EXPIRED_INTRMSK (1 << 9) +#define GEN8_MISC_CTRL0 _MMIO(0xA180) +#define VLV_PWRDWNUPCTL _MMIO(0xA294) +#define GEN9_MEDIA_PG_IDLE_HYSTERESIS _MMIO(0xA0C4) +#define GEN9_RENDER_PG_IDLE_HYSTERESIS _MMIO(0xA0C8) +#define GEN9_PG_ENABLE _MMIO(0xA210) +#define GEN9_RENDER_PG_ENABLE REG_BIT(0) +#define GEN9_MEDIA_PG_ENABLE REG_BIT(1) +#define GEN11_MEDIA_SAMPLER_PG_ENABLE REG_BIT(2) +#define VDN_HCP_POWERGATE_ENABLE(n) REG_BIT(3 + 2 * (n)) +#define VDN_MFX_POWERGATE_ENABLE(n) REG_BIT(4 + 2 * (n)) +#define GEN8_PUSHBUS_CONTROL _MMIO(0xA248) +#define GEN8_PUSHBUS_ENABLE _MMIO(0xA250) +#define GEN8_PUSHBUS_SHIFT _MMIO(0xA25C) + +#define GEN6_PMISR _MMIO(0x44020) +#define GEN6_PMIMR _MMIO(0x44024) /* rps_lock */ +#define GEN6_PMIIR _MMIO(0x44028) +#define GEN6_PMIER _MMIO(0x4402C) +#define GEN6_PM_MBOX_EVENT (1 << 25) +#define GEN6_PM_THERMAL_EVENT (1 << 24) + +/* + * For Gen11 these are in the upper word of the GPM_WGBOXPERF + * registers. Shifting is handled on accessing the imr and ier. + */ +#define GEN6_PM_RP_DOWN_TIMEOUT (1 << 6) +#define GEN6_PM_RP_UP_THRESHOLD (1 << 5) +#define GEN6_PM_RP_DOWN_THRESHOLD (1 << 4) +#define GEN6_PM_RP_UP_EI_EXPIRED (1 << 2) +#define GEN6_PM_RP_DOWN_EI_EXPIRED (1 << 1) +#define GEN6_PM_RPS_EVENTS (GEN6_PM_RP_UP_EI_EXPIRED | \ + GEN6_PM_RP_UP_THRESHOLD | \ + GEN6_PM_RP_DOWN_EI_EXPIRED | \ + GEN6_PM_RP_DOWN_THRESHOLD | \ + GEN6_PM_RP_DOWN_TIMEOUT) + +#define GEN7_GT_SCRATCH(i) _MMIO(0x4F100 + (i) * 4) +#define GEN7_GT_SCRATCH_REG_NUM 8 + +#define VLV_GTLC_SURVIVABILITY_REG _MMIO(0x130098) +#define VLV_GFX_CLK_STATUS_BIT (1 << 3) +#define VLV_GFX_CLK_FORCE_ON_BIT (1 << 2) + +#define GEN6_GT_GFX_RC6_LOCKED _MMIO(0x138104) +#define VLV_COUNTER_CONTROL _MMIO(0x138104) +#define VLV_COUNT_RANGE_HIGH (1 << 15) +#define VLV_MEDIA_RC0_COUNT_EN (1 << 5) +#define VLV_RENDER_RC0_COUNT_EN (1 << 4) +#define VLV_MEDIA_RC6_COUNT_EN (1 << 1) +#define VLV_RENDER_RC6_COUNT_EN (1 << 0) +#define GEN6_GT_GFX_RC6 _MMIO(0x138108) +#define VLV_GT_RENDER_RC6 _MMIO(0x138108) +#define VLV_GT_MEDIA_RC6 _MMIO(0x13810C) + +#define GEN6_GT_GFX_RC6p _MMIO(0x13810C) +#define GEN6_GT_GFX_RC6pp _MMIO(0x138110) +#define VLV_RENDER_C0_COUNT _MMIO(0x138118) +#define VLV_MEDIA_C0_COUNT _MMIO(0x13811C) + +#define GEN6_GT_CORE_STATUS _MMIO(0x138060) +#define GEN6_CORE_CPD_STATE_MASK (7 << 4) +#define GEN6_RCn_MASK 7 +#define GEN6_RC0 0 +#define GEN6_RC3 2 +#define GEN6_RC6 3 +#define GEN6_RC7 4 + +#define GEN8_GT_SLICE_INFO _MMIO(0x138064) +#define GEN8_LSLICESTAT_MASK 0x7 + +#define CHV_POWER_SS0_SIG1 _MMIO(0xa720) +#define CHV_POWER_SS1_SIG1 _MMIO(0xa728) +#define CHV_SS_PG_ENABLE (1 << 1) +#define CHV_EU08_PG_ENABLE (1 << 9) +#define CHV_EU19_PG_ENABLE (1 << 17) +#define CHV_EU210_PG_ENABLE (1 << 25) + +#define CHV_POWER_SS0_SIG2 _MMIO(0xa724) +#define CHV_POWER_SS1_SIG2 _MMIO(0xa72c) +#define CHV_EU311_PG_ENABLE (1 << 1) + +#define GEN9_SLICE_PGCTL_ACK(slice) _MMIO(0x804c + (slice) * 0x4) +#define GEN10_SLICE_PGCTL_ACK(slice) _MMIO(0x804c + ((slice) / 3) * 0x34 + \ + ((slice) % 3) * 0x4) +#define GEN9_PGCTL_SLICE_ACK (1 << 0) +#define GEN9_PGCTL_SS_ACK(subslice) (1 << (2 + (subslice) * 2)) +#define GEN10_PGCTL_VALID_SS_MASK(slice) ((slice) == 0 ? 0x7F : 0x1F) + +#define GEN9_SS01_EU_PGCTL_ACK(slice) _MMIO(0x805c + (slice) * 0x8) +#define GEN10_SS01_EU_PGCTL_ACK(slice) _MMIO(0x805c + ((slice) / 3) * 0x30 + \ + ((slice) % 3) * 0x8) +#define GEN9_SS23_EU_PGCTL_ACK(slice) _MMIO(0x8060 + (slice) * 0x8) +#define GEN10_SS23_EU_PGCTL_ACK(slice) _MMIO(0x8060 + ((slice) / 3) * 0x30 + \ + ((slice) % 3) * 0x8) +#define GEN9_PGCTL_SSA_EU08_ACK (1 << 0) +#define GEN9_PGCTL_SSA_EU19_ACK (1 << 2) +#define GEN9_PGCTL_SSA_EU210_ACK (1 << 4) +#define GEN9_PGCTL_SSA_EU311_ACK (1 << 6) +#define GEN9_PGCTL_SSB_EU08_ACK (1 << 8) +#define GEN9_PGCTL_SSB_EU19_ACK (1 << 10) +#define GEN9_PGCTL_SSB_EU210_ACK (1 << 12) +#define GEN9_PGCTL_SSB_EU311_ACK (1 << 14) + +#define GEN7_MISCCPCTL _MMIO(0x9424) +#define GEN7_DOP_CLOCK_GATE_ENABLE (1 << 0) +#define GEN8_DOP_CLOCK_GATE_CFCLK_ENABLE (1 << 2) +#define GEN8_DOP_CLOCK_GATE_GUC_ENABLE (1 << 4) +#define GEN8_DOP_CLOCK_GATE_MEDIA_ENABLE (1 << 6) + +#define GEN8_GARBCNTL _MMIO(0xB004) +#define GEN9_GAPS_TSV_CREDIT_DISABLE (1 << 7) +#define GEN11_ARBITRATION_PRIO_ORDER_MASK (0x3f << 22) +#define GEN11_HASH_CTRL_EXCL_MASK (0x7f << 0) +#define GEN11_HASH_CTRL_EXCL_BIT0 (1 << 0) + +#define GEN11_GLBLINVL _MMIO(0xB404) +#define GEN11_BANK_HASH_ADDR_EXCL_MASK (0x7f << 5) +#define GEN11_BANK_HASH_ADDR_EXCL_BIT0 (1 << 5) + +#define GEN10_DFR_RATIO_EN_AND_CHICKEN _MMIO(0x9550) +#define DFR_DISABLE (1 << 9) + +#define GEN11_GACB_PERF_CTRL _MMIO(0x4B80) +#define GEN11_HASH_CTRL_MASK (0x3 << 12 | 0xf << 0) +#define GEN11_HASH_CTRL_BIT0 (1 << 0) +#define GEN11_HASH_CTRL_BIT4 (1 << 12) + +#define GEN11_LSN_UNSLCVC _MMIO(0xB43C) +#define GEN11_LSN_UNSLCVC_GAFS_HALF_CL2_MAXALLOC (1 << 9) +#define GEN11_LSN_UNSLCVC_GAFS_HALF_SF_MAXALLOC (1 << 7) + +#define GEN10_SAMPLER_MODE _MMIO(0xE18C) +#define ENABLE_SMALLPL REG_BIT(15) +#define GEN11_SAMPLER_ENABLE_HEADLESS_MSG REG_BIT(5) + +#define GEN7_L3LOG(slice, i) _MMIO(0xB070 + (slice) * 0x200 + (i) * 4) +#define GEN7_L3LOG_SIZE 0x80 + +#define GEN7_HALF_SLICE_CHICKEN1 _MMIO(0xe100) /* IVB GT1 + VLV */ +#define GEN7_HALF_SLICE_CHICKEN1_GT2 _MMIO(0xf100) +#define GEN7_MAX_PS_THREAD_DEP (8 << 12) +#define GEN7_SINGLE_SUBSCAN_DISPATCH_ENABLE (1 << 10) +#define GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE (1 << 4) +#define GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE (1 << 3) + +#define GEN9_HALF_SLICE_CHICKEN5 _MMIO(0xe188) +#define GEN9_DG_MIRROR_FIX_ENABLE (1 << 5) +#define GEN9_CCS_TLB_PREFETCH_ENABLE (1 << 3) + +#define GEN8_ROW_CHICKEN _MMIO(0xe4f0) +#define FLOW_CONTROL_ENABLE REG_BIT(15) +#define UGM_BACKUP_MODE REG_BIT(13) +#define MDQ_ARBITRATION_MODE REG_BIT(12) +#define PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE REG_BIT(8) +#define STALL_DOP_GATING_DISABLE REG_BIT(5) +#define THROTTLE_12_5 REG_GENMASK(4, 2) +#define DISABLE_EARLY_EOT REG_BIT(1) + +#define GEN7_ROW_CHICKEN2 _MMIO(0xe4f4) +#define GEN12_DISABLE_READ_SUPPRESSION REG_BIT(15) +#define GEN12_DISABLE_EARLY_READ REG_BIT(14) +#define GEN12_ENABLE_LARGE_GRF_MODE REG_BIT(12) +#define GEN12_PUSH_CONST_DEREF_HOLD_DIS REG_BIT(8) + +#define LSC_CHICKEN_BIT_0 _MMIO(0xe7c8) +#define FORCE_1_SUB_MESSAGE_PER_FRAGMENT REG_BIT(15) +#define LSC_CHICKEN_BIT_0_UDW _MMIO(0xe7c8 + 4) +#define DIS_CHAIN_2XSIMD8 REG_BIT(55 - 32) +#define FORCE_SLM_FENCE_SCOPE_TO_TILE REG_BIT(42 - 32) +#define FORCE_UGM_FENCE_SCOPE_TO_TILE REG_BIT(41 - 32) +#define MAXREQS_PER_BANK REG_GENMASK(39 - 32, 37 - 32) +#define DISABLE_128B_EVICTION_COMMAND_UDW REG_BIT(36 - 32) + +#define GEN7_ROW_CHICKEN2_GT2 _MMIO(0xf4f4) +#define DOP_CLOCK_GATING_DISABLE (1 << 0) +#define PUSH_CONSTANT_DEREF_DISABLE (1 << 8) +#define GEN11_TDL_CLOCK_GATING_FIX_DISABLE (1 << 1) + +#define GEN9_ROW_CHICKEN4 _MMIO(0xe48c) +#define GEN12_DISABLE_GRF_CLEAR REG_BIT(13) +#define GEN12_DISABLE_TDL_PUSH REG_BIT(9) +#define GEN11_DIS_PICK_2ND_EU REG_BIT(7) +#define GEN12_DISABLE_HDR_PAST_PAYLOAD_HOLD_FIX REG_BIT(4) + +#define HSW_ROW_CHICKEN3 _MMIO(0xe49c) +#define HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE (1 << 6) + +#define HALF_SLICE_CHICKEN2 _MMIO(0xe180) +#define GEN8_ST_PO_DISABLE (1 << 13) + +#define HALF_SLICE_CHICKEN3 _MMIO(0xe184) +#define HSW_SAMPLE_C_PERFORMANCE (1 << 9) +#define GEN8_CENTROID_PIXEL_OPT_DIS (1 << 8) +#define GEN9_DISABLE_OCL_OOB_SUPPRESS_LOGIC (1 << 5) +#define GEN8_SAMPLER_POWER_BYPASS_DIS (1 << 1) + +#define GEN9_HALF_SLICE_CHICKEN7 _MMIO(0xe194) +#define DG2_DISABLE_ROUND_ENABLE_ALLOW_FOR_SSLA REG_BIT(15) +#define GEN9_SAMPLER_HASH_COMPRESSED_READ_ADDR REG_BIT(8) +#define GEN9_ENABLE_YV12_BUGFIX REG_BIT(4) +#define GEN9_ENABLE_GPGPU_PREEMPTION REG_BIT(2) + +/* MOCS (Memory Object Control State) registers */ +#define GEN9_LNCFCMOCS(i) _MMIO(0xb020 + (i) * 4) /* L3 Cache Control */ +#define GEN9_LNCFCMOCS_REG_COUNT 32 + +#define __GEN9_RCS0_MOCS0 0xc800 +#define GEN9_GFX_MOCS(i) _MMIO(__GEN9_RCS0_MOCS0 + (i) * 4) +#define __GEN9_VCS0_MOCS0 0xc900 +#define GEN9_MFX0_MOCS(i) _MMIO(__GEN9_VCS0_MOCS0 + (i) * 4) +#define __GEN9_VCS1_MOCS0 0xca00 +#define GEN9_MFX1_MOCS(i) _MMIO(__GEN9_VCS1_MOCS0 + (i) * 4) +#define __GEN9_VECS0_MOCS0 0xcb00 +#define GEN9_VEBOX_MOCS(i) _MMIO(__GEN9_VECS0_MOCS0 + (i) * 4) +#define __GEN9_BCS0_MOCS0 0xcc00 +#define GEN9_BLT_MOCS(i) _MMIO(__GEN9_BCS0_MOCS0 + (i) * 4) +#define __GEN11_VCS2_MOCS0 0x10000 +#define GEN11_MFX2_MOCS(i) _MMIO(__GEN11_VCS2_MOCS0 + (i) * 4) + +#define GEN9_SCRATCH_LNCF1 _MMIO(0xb008) +#define GEN9_LNCF_NONIA_COHERENT_ATOMICS_ENABLE REG_BIT(0) + +#define GEN9_SCRATCH1 _MMIO(0xb11c) +#define EVICTION_PERF_FIX_ENABLE REG_BIT(8) + +#define GEN10_SCRATCH_LNCF2 _MMIO(0xb0a0) +#define PMFLUSHDONE_LNICRSDROP (1 << 20) +#define PMFLUSH_GAPL3UNBLOCK (1 << 21) +#define PMFLUSHDONE_LNEBLK (1 << 22) + +#define XEHP_L3NODEARBCFG _MMIO(0xb0b4) +#define XEHP_LNESPARE REG_BIT(19) + +#define GEN12_GLOBAL_MOCS(i) _MMIO(0x4000 + (i) * 4) /* Global MOCS regs */ + +/* gamt regs */ +#define GEN8_L3_LRA_1_GPGPU _MMIO(0x4dd4) +#define GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_BDW 0x67F1427F /* max/min for LRA1/2 */ +#define GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_CHV 0x5FF101FF /* max/min for LRA1/2 */ +#define GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_SKL 0x67F1427F /* " " */ +#define GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_BXT 0x5FF101FF /* " " */ + +#define MMCD_MISC_CTRL _MMIO(0x4ddc) /* skl+ */ +#define MMCD_PCLA (1 << 31) +#define MMCD_HOTSPOT_EN (1 << 27) + +#define SLICE_COMMON_ECO_CHICKEN1 _MMIO(0x731C) +#define MSC_MSAA_REODER_BUF_BYPASS_DISABLE REG_BIT(14) + +#endif /* __INTEL_GT_REGS__ */ diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.c b/drivers/gpu/drm/i915/gt/intel_gtt.c index a94be0306464..b5d3cde08aac 100644 --- a/drivers/gpu/drm/i915/gt/intel_gtt.c +++ b/drivers/gpu/drm/i915/gt/intel_gtt.c @@ -13,6 +13,7 @@ #include "gem/i915_gem_lmem.h" #include "i915_trace.h" #include "intel_gt.h" +#include "intel_gt_regs.h" #include "intel_gtt.h" struct drm_i915_gem_object *alloc_pt_lmem(struct i915_address_space *vm, int sz) diff --git a/drivers/gpu/drm/i915/gt/intel_llc.c b/drivers/gpu/drm/i915/gt/intel_llc.c index 08d7d5ae263a..335c65758d6f 100644 --- a/drivers/gpu/drm/i915/gt/intel_llc.c +++ b/drivers/gpu/drm/i915/gt/intel_llc.c @@ -7,6 +7,7 @@ #include <linux/cpufreq.h> #include "i915_drv.h" +#include "i915_reg.h" #include "intel_gt.h" #include "intel_llc.h" #include "intel_pcode.h" @@ -140,11 +141,10 @@ static void gen6_update_ring_freq(struct intel_llc *llc) unsigned int ia_freq, ring_freq; calc_ia_freq(llc, gpu_freq, &consts, &ia_freq, &ring_freq); - sandybridge_pcode_write(i915, - GEN6_PCODE_WRITE_MIN_FREQ_TABLE, - ia_freq << GEN6_PCODE_FREQ_IA_RATIO_SHIFT | - ring_freq << GEN6_PCODE_FREQ_RING_RATIO_SHIFT | - gpu_freq); + snb_pcode_write(i915, GEN6_PCODE_WRITE_MIN_FREQ_TABLE, + ia_freq << GEN6_PCODE_FREQ_IA_RATIO_SHIFT | + ring_freq << GEN6_PCODE_FREQ_RING_RATIO_SHIFT | + gpu_freq); } } diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index b3489599e4de..e46bf1429f2c 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -9,8 +9,10 @@ #include "i915_drv.h" #include "i915_perf.h" #include "intel_engine.h" +#include "intel_engine_regs.h" #include "intel_gpu_commands.h" #include "intel_gt.h" +#include "intel_gt_regs.h" #include "intel_lrc.h" #include "intel_lrc_reg.h" #include "intel_ring.h" diff --git a/drivers/gpu/drm/i915/gt/intel_lrc_reg.h b/drivers/gpu/drm/i915/gt/intel_lrc_reg.h index f785d0ed238f..304000c7e345 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc_reg.h +++ b/drivers/gpu/drm/i915/gt/intel_lrc_reg.h @@ -53,21 +53,6 @@ #define GEN8_EXECLISTS_STATUS_BUF 0x370 #define GEN11_EXECLISTS_STATUS_BUF2 0x3c0 -/* Execlists regs */ -#define RING_ELSP(base) _MMIO((base) + 0x230) -#define RING_EXECLIST_STATUS_LO(base) _MMIO((base) + 0x234) -#define RING_EXECLIST_STATUS_HI(base) _MMIO((base) + 0x234 + 4) -#define RING_CONTEXT_CONTROL(base) _MMIO((base) + 0x244) -#define CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT REG_BIT(0) -#define CTX_CTRL_RS_CTX_ENABLE REG_BIT(1) -#define CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT REG_BIT(2) -#define CTX_CTRL_INHIBIT_SYN_CTX_SWITCH REG_BIT(3) -#define GEN12_CTX_CTRL_OAR_CONTEXT_ENABLE REG_BIT(8) -#define RING_CONTEXT_STATUS_PTR(base) _MMIO((base) + 0x3a0) -#define RING_EXECLIST_SQ_CONTENTS(base) _MMIO((base) + 0x510) -#define RING_EXECLIST_CONTROL(base) _MMIO((base) + 0x550) -#define EL_CTRL_LOAD REG_BIT(0) - /* * The docs specify that the write pointer wraps around after 5h, "After status * is written out to the last available status QW at offset 5h, this pointer diff --git a/drivers/gpu/drm/i915/gt/intel_mocs.c b/drivers/gpu/drm/i915/gt/intel_mocs.c index 9c253ba593c6..c4c37585ae8c 100644 --- a/drivers/gpu/drm/i915/gt/intel_mocs.c +++ b/drivers/gpu/drm/i915/gt/intel_mocs.c @@ -7,7 +7,7 @@ #include "intel_engine.h" #include "intel_gt.h" -#include "intel_lrc_reg.h" +#include "intel_gt_regs.h" #include "intel_mocs.h" #include "intel_ring.h" diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c index c3155ee58689..6df359c534fe 100644 --- a/drivers/gpu/drm/i915/gt/intel_rc6.c +++ b/drivers/gpu/drm/i915/gt/intel_rc6.c @@ -6,9 +6,12 @@ #include <linux/pm_runtime.h> #include "i915_drv.h" +#include "i915_reg.h" #include "i915_vgpu.h" +#include "intel_engine_regs.h" #include "intel_gt.h" #include "intel_gt_pm.h" +#include "intel_gt_regs.h" #include "intel_pcode.h" #include "intel_rc6.h" @@ -267,8 +270,7 @@ static void gen6_rc6_enable(struct intel_rc6 *rc6) GEN6_RC_CTL_HW_ENABLE; rc6vids = 0; - ret = sandybridge_pcode_read(i915, GEN6_PCODE_READ_RC6VIDS, - &rc6vids, NULL); + ret = snb_pcode_read(i915, GEN6_PCODE_READ_RC6VIDS, &rc6vids, NULL); if (GRAPHICS_VER(i915) == 6 && ret) { drm_dbg(&i915->drm, "Couldn't check for BIOS workaround\n"); } else if (GRAPHICS_VER(i915) == 6 && @@ -278,7 +280,7 @@ static void gen6_rc6_enable(struct intel_rc6 *rc6) GEN6_DECODE_RC6_VID(rc6vids & 0xff), 450); rc6vids &= 0xffff00; rc6vids |= GEN6_ENCODE_RC6_VID(450); - ret = sandybridge_pcode_write(i915, GEN6_PCODE_WRITE_RC6VIDS, rc6vids); + ret = snb_pcode_write(i915, GEN6_PCODE_WRITE_RC6VIDS, rc6vids); if (ret) drm_err(&i915->drm, "Couldn't fix incorrect rc6 voltage\n"); @@ -449,10 +451,10 @@ static bool bxt_check_bios_rc6_setup(struct intel_rc6 *rc6) enable_rc6 = false; } - if (!((intel_uncore_read(uncore, PWRCTX_MAXCNT_RCSUNIT) & IDLE_TIME_MASK) > 1 && - (intel_uncore_read(uncore, PWRCTX_MAXCNT_VCSUNIT0) & IDLE_TIME_MASK) > 1 && - (intel_uncore_read(uncore, PWRCTX_MAXCNT_BCSUNIT) & IDLE_TIME_MASK) > 1 && - (intel_uncore_read(uncore, PWRCTX_MAXCNT_VECSUNIT) & IDLE_TIME_MASK) > 1)) { + if (!((intel_uncore_read(uncore, PWRCTX_MAXCNT(RENDER_RING_BASE)) & IDLE_TIME_MASK) > 1 && + (intel_uncore_read(uncore, PWRCTX_MAXCNT(GEN6_BSD_RING_BASE)) & IDLE_TIME_MASK) > 1 && + (intel_uncore_read(uncore, PWRCTX_MAXCNT(BLT_RING_BASE)) & IDLE_TIME_MASK) > 1 && + (intel_uncore_read(uncore, PWRCTX_MAXCNT(VEBOX_RING_BASE)) & IDLE_TIME_MASK) > 1)) { drm_dbg(&i915->drm, "Engine Idle wait time not set properly.\n"); enable_rc6 = false; diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.h b/drivers/gpu/drm/i915/gt/intel_rc6.h index e119ec4a0bcc..b6fea71afc22 100644 --- a/drivers/gpu/drm/i915/gt/intel_rc6.h +++ b/drivers/gpu/drm/i915/gt/intel_rc6.h @@ -6,7 +6,7 @@ #ifndef INTEL_RC6_H #define INTEL_RC6_H -#include "i915_reg.h" +#include "i915_reg_defs.h" struct intel_engine_cs; struct intel_rc6; diff --git a/drivers/gpu/drm/i915/gt/intel_region_lmem.c b/drivers/gpu/drm/i915/gt/intel_region_lmem.c index fde2dcb59809..cb5a67c98f30 100644 --- a/drivers/gpu/drm/i915/gt/intel_region_lmem.c +++ b/drivers/gpu/drm/i915/gt/intel_region_lmem.c @@ -4,6 +4,7 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "intel_memory_region.h" #include "intel_region_lmem.h" #include "intel_region_ttm.h" diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c index 7be0002d9d70..59beb69ff6f2 100644 --- a/drivers/gpu/drm/i915/gt/intel_reset.c +++ b/drivers/gpu/drm/i915/gt/intel_reset.c @@ -11,14 +11,18 @@ #include "gem/i915_gem_context.h" +#include "gt/intel_gt_regs.h" + #include "i915_drv.h" #include "i915_gpu_error.h" #include "i915_irq.h" #include "intel_breadcrumbs.h" #include "intel_engine_pm.h" +#include "intel_engine_regs.h" #include "intel_gt.h" #include "intel_gt_pm.h" #include "intel_gt_requests.h" +#include "intel_pci_config.h" #include "intel_reset.h" #include "uc/intel_guc.h" diff --git a/drivers/gpu/drm/i915/gt/intel_ring.c b/drivers/gpu/drm/i915/gt/intel_ring.c index 2fdd52b62092..723055340c9b 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring.c +++ b/drivers/gpu/drm/i915/gt/intel_ring.c @@ -9,6 +9,7 @@ #include "i915_drv.h" #include "i915_vma.h" #include "intel_engine.h" +#include "intel_engine_regs.h" #include "intel_gpu_commands.h" #include "intel_ring.h" #include "intel_timeline.h" diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c index 3e6fac0340ef..8362eb09092e 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c @@ -11,8 +11,10 @@ #include "i915_mitigations.h" #include "intel_breadcrumbs.h" #include "intel_context.h" +#include "intel_engine_regs.h" #include "intel_gt.h" #include "intel_gt_irq.h" +#include "intel_gt_regs.h" #include "intel_reset.h" #include "intel_ring.h" #include "shmem_utils.h" @@ -1002,15 +1004,15 @@ static void gen6_bsd_submit_request(struct i915_request *request) /* Disable notification that the ring is IDLE. The GT * will then assume that it is busy and bring it out of rc6. */ - intel_uncore_write_fw(uncore, GEN6_BSD_SLEEP_PSMI_CONTROL, - _MASKED_BIT_ENABLE(GEN6_BSD_SLEEP_MSG_DISABLE)); + intel_uncore_write_fw(uncore, RING_PSMI_CTL(GEN6_BSD_RING_BASE), + _MASKED_BIT_ENABLE(GEN6_PSMI_SLEEP_MSG_DISABLE)); /* Clear the context id. Here be magic! */ intel_uncore_write64_fw(uncore, GEN6_BSD_RNCID, 0x0); /* Wait for the ring not to be idle, i.e. for it to wake up. */ if (__intel_wait_for_register_fw(uncore, - GEN6_BSD_SLEEP_PSMI_CONTROL, + RING_PSMI_CTL(GEN6_BSD_RING_BASE), GEN6_BSD_SLEEP_INDICATOR, 0, 1000, 0, NULL)) @@ -1023,8 +1025,8 @@ static void gen6_bsd_submit_request(struct i915_request *request) /* Let the ring send IDLE messages to the GT again, * and so let it sleep to conserve power when idle. */ - intel_uncore_write_fw(uncore, GEN6_BSD_SLEEP_PSMI_CONTROL, - _MASKED_BIT_DISABLE(GEN6_BSD_SLEEP_MSG_DISABLE)); + intel_uncore_write_fw(uncore, RING_PSMI_CTL(GEN6_BSD_RING_BASE), + _MASKED_BIT_DISABLE(GEN6_PSMI_SLEEP_MSG_DISABLE)); intel_uncore_forcewake_put(uncore, FORCEWAKE_ALL); } diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c index 54e7df788dbf..0cb299cdfc8f 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.c +++ b/drivers/gpu/drm/i915/gt/intel_rps.c @@ -6,11 +6,13 @@ #include <drm/i915_drm.h> #include "i915_drv.h" +#include "i915_irq.h" #include "intel_breadcrumbs.h" #include "intel_gt.h" #include "intel_gt_clock_utils.h" #include "intel_gt_irq.h" #include "intel_gt_pm_irq.h" +#include "intel_gt_regs.h" #include "intel_pcode.h" #include "intel_rps.h" #include "vlv_sideband.h" @@ -1090,9 +1092,8 @@ static void gen6_rps_init(struct intel_rps *rps) IS_GEN9_BC(i915) || GRAPHICS_VER(i915) >= 11) { u32 ddcc_status = 0; - if (sandybridge_pcode_read(i915, - HSW_PCODE_DYNAMIC_DUTY_CYCLE_CONTROL, - &ddcc_status, NULL) == 0) + if (snb_pcode_read(i915, HSW_PCODE_DYNAMIC_DUTY_CYCLE_CONTROL, + &ddcc_status, NULL) == 0) rps->efficient_freq = clamp_t(u8, (ddcc_status >> 8) & 0xff, @@ -1940,8 +1941,7 @@ void intel_rps_init(struct intel_rps *rps) if (GRAPHICS_VER(i915) == 6 || IS_IVYBRIDGE(i915) || IS_HASWELL(i915)) { u32 params = 0; - sandybridge_pcode_read(i915, GEN6_READ_OC_PARAMS, - ¶ms, NULL); + snb_pcode_read(i915, GEN6_READ_OC_PARAMS, ¶ms, NULL); if (params & BIT(31)) { /* OC supported */ drm_dbg(&i915->drm, "Overclocking supported, max: %dMHz, overclock: %dMHz\n", diff --git a/drivers/gpu/drm/i915/gt/intel_sseu.c b/drivers/gpu/drm/i915/gt/intel_sseu.c index bdf09051b8a0..29118c652811 100644 --- a/drivers/gpu/drm/i915/gt/intel_sseu.c +++ b/drivers/gpu/drm/i915/gt/intel_sseu.c @@ -4,7 +4,8 @@ */ #include "i915_drv.h" -#include "intel_lrc_reg.h" +#include "intel_engine_regs.h" +#include "intel_gt_regs.h" #include "intel_sseu.h" void intel_sseu_set_info(struct sseu_dev_info *sseu, u8 max_slices, diff --git a/drivers/gpu/drm/i915/gt/intel_sseu_debugfs.c b/drivers/gpu/drm/i915/gt/intel_sseu_debugfs.c index 8bb3a91dad82..903626f106ea 100644 --- a/drivers/gpu/drm/i915/gt/intel_sseu_debugfs.c +++ b/drivers/gpu/drm/i915/gt/intel_sseu_debugfs.c @@ -6,6 +6,7 @@ #include "i915_drv.h" #include "intel_gt_debugfs.h" +#include "intel_gt_regs.h" #include "intel_sseu_debugfs.h" static void sseu_copy_subslices(const struct sseu_dev_info *sseu, diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index ab3277a3d593..59ad23118875 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -6,8 +6,10 @@ #include "i915_drv.h" #include "intel_context.h" #include "intel_engine_pm.h" +#include "intel_engine_regs.h" #include "intel_gpu_commands.h" #include "intel_gt.h" +#include "intel_gt_regs.h" #include "intel_ring.h" #include "intel_workarounds.h" @@ -2208,7 +2210,7 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) * For DG1 this only applies to A0. */ wa_masked_en(wal, - GEN6_RC_SLEEP_PSMI_CONTROL, + RING_PSMI_CTL(RENDER_RING_BASE), GEN12_WAIT_FOR_EVENT_POWER_DOWN_DISABLE | GEN8_RC_SEMA_IDLE_MSG_DISABLE); } @@ -2423,7 +2425,7 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) if (GRAPHICS_VER(i915) == 7) { /* WaBCSVCSTlbInvalidationMode:ivb,vlv,hsw */ wa_masked_en(wal, - GFX_MODE_GEN7, + RING_MODE_GEN7(RENDER_RING_BASE), GFX_TLB_INVALIDATE_EXPLICIT | GFX_REPLAY_MODE); /* WaDisable_RenderCache_OperationalFlush:ivb,vlv,hsw */ @@ -2536,7 +2538,7 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) * they are already accustomed to from before contexts were * enabled. */ - wa_add(wal, ECOSKPD, + wa_add(wal, ECOSKPD(RENDER_RING_BASE), 0, _MASKED_BIT_ENABLE(ECO_CONSTANT_BUFFER_SR_DISABLE), 0 /* XXX bit doesn't stick on Broadwater */, true); diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds_types.h b/drivers/gpu/drm/i915/gt/intel_workarounds_types.h index 1e873681795d..8a4b6de4e754 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds_types.h +++ b/drivers/gpu/drm/i915/gt/intel_workarounds_types.h @@ -8,7 +8,7 @@ #include <linux/types.h> -#include "i915_reg.h" +#include "i915_reg_defs.h" struct i915_wa { i915_reg_t reg; diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_pm.c b/drivers/gpu/drm/i915/gt/selftest_engine_pm.c index 8af261831470..0dcb3ed44a73 100644 --- a/drivers/gpu/drm/i915/gt/selftest_engine_pm.c +++ b/drivers/gpu/drm/i915/gt/selftest_engine_pm.c @@ -6,6 +6,7 @@ #include <linux/sort.h> #include "i915_selftest.h" +#include "intel_engine_regs.h" #include "intel_gpu_commands.h" #include "intel_gt_clock_utils.h" #include "selftest_engine.h" diff --git a/drivers/gpu/drm/i915/gt/selftest_gt_pm.c b/drivers/gpu/drm/i915/gt/selftest_gt_pm.c index 8bf62a5826cc..be94f863bdef 100644 --- a/drivers/gpu/drm/i915/gt/selftest_gt_pm.c +++ b/drivers/gpu/drm/i915/gt/selftest_gt_pm.c @@ -5,6 +5,7 @@ #include <linux/sort.h> +#include "intel_engine_regs.h" #include "intel_gt_clock_utils.h" #include "selftest_llc.h" diff --git a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c index 15d63435ec4d..4a20ba63446c 100644 --- a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c +++ b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c @@ -7,6 +7,7 @@ #include "gem/i915_gem_context.h" +#include "i915_gem_evict.h" #include "intel_gt.h" #include "intel_engine_heartbeat.h" #include "intel_engine_pm.h" diff --git a/drivers/gpu/drm/i915/gt/selftest_llc.c b/drivers/gpu/drm/i915/gt/selftest_llc.c index 459b775f163a..2cd184ab32b1 100644 --- a/drivers/gpu/drm/i915/gt/selftest_llc.c +++ b/drivers/gpu/drm/i915/gt/selftest_llc.c @@ -31,9 +31,8 @@ static int gen6_verify_ring_freq(struct intel_llc *llc) calc_ia_freq(llc, gpu_freq, &consts, &ia_freq, &ring_freq); val = gpu_freq; - if (sandybridge_pcode_read(i915, - GEN6_PCODE_READ_MIN_FREQ_TABLE, - &val, NULL)) { + if (snb_pcode_read(i915, GEN6_PCODE_READ_MIN_FREQ_TABLE, + &val, NULL)) { pr_err("Failed to read freq table[%d], range [%d, %d]\n", gpu_freq, consts.min_gpu_freq, consts.max_gpu_freq); err = -ENXIO; diff --git a/drivers/gpu/drm/i915/gt/selftest_rps.c b/drivers/gpu/drm/i915/gt/selftest_rps.c index 7ee2513e15f9..e1e5dd5f7638 100644 --- a/drivers/gpu/drm/i915/gt/selftest_rps.c +++ b/drivers/gpu/drm/i915/gt/selftest_rps.c @@ -8,6 +8,7 @@ #include "intel_engine_heartbeat.h" #include "intel_engine_pm.h" +#include "intel_engine_regs.h" #include "intel_gpu_commands.h" #include "intel_gt_clock_utils.h" #include "intel_gt_pm.h" @@ -518,9 +519,8 @@ static void show_pcu_config(struct intel_rps *rps) for (gpu_freq = min_gpu_freq; gpu_freq <= max_gpu_freq; gpu_freq++) { int ia_freq = gpu_freq; - sandybridge_pcode_read(i915, - GEN6_PCODE_READ_MIN_FREQ_TABLE, - &ia_freq, NULL); + snb_pcode_read(i915, GEN6_PCODE_READ_MIN_FREQ_TABLE, + &ia_freq, NULL); pr_info("%5d %5d %5d\n", gpu_freq * 50, diff --git a/drivers/gpu/drm/i915/gt/selftest_timeline.c b/drivers/gpu/drm/i915/gt/selftest_timeline.c index e2eb686a9763..0410c402f2a3 100644 --- a/drivers/gpu/drm/i915/gt/selftest_timeline.c +++ b/drivers/gpu/drm/i915/gt/selftest_timeline.c @@ -8,6 +8,7 @@ #include "intel_context.h" #include "intel_engine_heartbeat.h" #include "intel_engine_pm.h" +#include "intel_engine_regs.h" #include "intel_gpu_commands.h" #include "intel_gt.h" #include "intel_gt_requests.h" diff --git a/drivers/gpu/drm/i915/gt/uc/abi/guc_actions_slpc_abi.h b/drivers/gpu/drm/i915/gt/uc/abi/guc_actions_slpc_abi.h index 7a8d4bfc5f6a..62cb4254a77a 100644 --- a/drivers/gpu/drm/i915/gt/uc/abi/guc_actions_slpc_abi.h +++ b/drivers/gpu/drm/i915/gt/uc/abi/guc_actions_slpc_abi.h @@ -7,7 +7,6 @@ #define _GUC_ACTIONS_SLPC_ABI_H_ #include <linux/types.h> -#include "i915_reg.h" /** * DOC: SLPC SHARED DATA STRUCTURE diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c index 6e228343e8cb..5bab32fef120 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c @@ -7,11 +7,13 @@ #include "gt/intel_gt.h" #include "gt/intel_gt_irq.h" #include "gt/intel_gt_pm_irq.h" +#include "gt/intel_gt_regs.h" #include "intel_guc.h" #include "intel_guc_slpc.h" #include "intel_guc_ads.h" #include "intel_guc_submission.h" #include "i915_drv.h" +#include "i915_irq.h" /** * DOC: GuC diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c index 1a1edae67e4e..162b89198567 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c @@ -5,7 +5,9 @@ #include <linux/bsearch.h> +#include "gt/intel_engine_regs.h" #include "gt/intel_gt.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_lrc.h" #include "gt/shmem_utils.h" #include "intel_guc_ads.h" diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c index 31420ce1ce6b..dcb51b53b495 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c @@ -10,6 +10,7 @@ */ #include "gt/intel_gt.h" +#include "gt/intel_gt_regs.h" #include "intel_guc_fw.h" #include "i915_drv.h" diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c index ac0931f0374b..7b0b43e87244 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c @@ -7,6 +7,7 @@ #include "gt/intel_gt.h" #include "i915_drv.h" +#include "i915_irq.h" #include "i915_memcpy.h" #include "intel_guc_log.h" diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h index b37fc2ffaef2..85846c5570c5 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h @@ -9,7 +9,7 @@ #include <linux/compiler.h> #include <linux/types.h> -#include "i915_reg.h" +#include "i915_reg_defs.h" /* Definitions of GuC H/W registers, bits, etc */ diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c index 13b27b8ff74e..ddbea939b1dc 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c @@ -4,8 +4,10 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "intel_guc_slpc.h" #include "gt/intel_gt.h" +#include "gt/intel_gt_regs.h" static inline struct intel_guc *slpc_to_guc(struct intel_guc_slpc *slpc) { diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index 154ad726e266..04b8321fc758 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -9,13 +9,15 @@ #include "gt/gen8_engine_cs.h" #include "gt/intel_breadcrumbs.h" #include "gt/intel_context.h" -#include "gt/intel_engine_pm.h" #include "gt/intel_engine_heartbeat.h" +#include "gt/intel_engine_pm.h" +#include "gt/intel_engine_regs.h" #include "gt/intel_gpu_commands.h" #include "gt/intel_gt.h" #include "gt/intel_gt_clock_utils.h" #include "gt/intel_gt_irq.h" #include "gt/intel_gt_pm.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_gt_requests.h" #include "gt/intel_lrc.h" #include "gt/intel_lrc_reg.h" diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.h b/drivers/gpu/drm/i915/gt/uc/intel_huc.h index ae8c8a6c8cc8..73ec670800f2 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.h @@ -6,7 +6,7 @@ #ifndef _INTEL_HUC_H_ #define _INTEL_HUC_H_ -#include "i915_reg.h" +#include "i915_reg_defs.h" #include "intel_uc_fw.h" #include "intel_huc_fw.h" diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c index a5af05bde6f2..dd588ecf048a 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c @@ -11,6 +11,7 @@ #include "intel_uc_fw.h" #include "intel_uc_fw_abi.h" #include "i915_drv.h" +#include "i915_reg.h" static inline struct intel_gt * ____uc_fw_to_gt(struct intel_uc_fw *uc_fw, enum intel_uc_fw_type type) diff --git a/drivers/gpu/drm/i915/gvt/aperture_gm.c b/drivers/gpu/drm/i915/gvt/aperture_gm.c index 0d6d59871308..6b3dedd321bb 100644 --- a/drivers/gpu/drm/i915/gvt/aperture_gm.c +++ b/drivers/gpu/drm/i915/gvt/aperture_gm.c @@ -35,6 +35,7 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "gt/intel_ggtt_fencing.h" #include "gvt.h" diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index c4118b808268..2459213b6c87 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -37,7 +37,9 @@ #include <linux/slab.h> #include "i915_drv.h" +#include "gt/intel_engine_regs.h" #include "gt/intel_gpu_commands.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_lrc.h" #include "gt/intel_ring.h" #include "gt/intel_gt_requests.h" @@ -1144,7 +1146,7 @@ struct cmd_interrupt_event { int mi_user_interrupt; }; -static struct cmd_interrupt_event cmd_interrupt_events[] = { +static const struct cmd_interrupt_event cmd_interrupt_events[] = { [RCS0] = { .pipe_control_notify = RCS_PIPE_CONTROL, .mi_flush_dw = INTEL_GVT_EVENT_RESERVED, diff --git a/drivers/gpu/drm/i915/gvt/display.c b/drivers/gpu/drm/i915/gvt/display.c index 034c060f89d4..c7722c818b4d 100644 --- a/drivers/gpu/drm/i915/gvt/display.c +++ b/drivers/gpu/drm/i915/gvt/display.c @@ -33,6 +33,7 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "gvt.h" static int get_edp_pipe(struct intel_vgpu *vgpu) @@ -184,10 +185,10 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu) for_each_pipe(dev_priv, pipe) { vgpu_vreg_t(vgpu, PIPECONF(pipe)) &= - ~(PIPECONF_ENABLE | I965_PIPECONF_ACTIVE); - vgpu_vreg_t(vgpu, DSPCNTR(pipe)) &= ~DISPLAY_PLANE_ENABLE; + ~(PIPECONF_ENABLE | PIPECONF_STATE_ENABLE); + vgpu_vreg_t(vgpu, DSPCNTR(pipe)) &= ~DISP_ENABLE; vgpu_vreg_t(vgpu, SPRCTL(pipe)) &= ~SPRITE_ENABLE; - vgpu_vreg_t(vgpu, CURCNTR(pipe)) &= ~MCURSOR_MODE; + vgpu_vreg_t(vgpu, CURCNTR(pipe)) &= ~MCURSOR_MODE_MASK; vgpu_vreg_t(vgpu, CURCNTR(pipe)) |= MCURSOR_MODE_DISABLE; } @@ -245,7 +246,7 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu) * setup_virtual_dp_monitor. */ vgpu_vreg_t(vgpu, PIPECONF(PIPE_A)) |= PIPECONF_ENABLE; - vgpu_vreg_t(vgpu, PIPECONF(PIPE_A)) |= I965_PIPECONF_ACTIVE; + vgpu_vreg_t(vgpu, PIPECONF(PIPE_A)) |= PIPECONF_STATE_ENABLE; /* * Golden M/N are calculated based on: @@ -253,7 +254,7 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu) * DP link clk 1620 MHz and non-constant_n. * TODO: calculate DP link symbol clk and stream clk m/n. */ - vgpu_vreg_t(vgpu, PIPE_DATA_M1(TRANSCODER_A)) = 63 << TU_SIZE_SHIFT; + vgpu_vreg_t(vgpu, PIPE_DATA_M1(TRANSCODER_A)) = TU_SIZE(64); vgpu_vreg_t(vgpu, PIPE_DATA_M1(TRANSCODER_A)) |= 0x5b425e; vgpu_vreg_t(vgpu, PIPE_DATA_N1(TRANSCODER_A)) = 0x800000; vgpu_vreg_t(vgpu, PIPE_LINK_M1(TRANSCODER_A)) = 0x3cd6e; @@ -387,7 +388,7 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu) * DP link clk 1620 MHz and non-constant_n. * TODO: calculate DP link symbol clk and stream clk m/n. */ - vgpu_vreg_t(vgpu, PIPE_DATA_M1(TRANSCODER_A)) = 63 << TU_SIZE_SHIFT; + vgpu_vreg_t(vgpu, PIPE_DATA_M1(TRANSCODER_A)) = TU_SIZE(64); vgpu_vreg_t(vgpu, PIPE_DATA_M1(TRANSCODER_A)) |= 0x5b425e; vgpu_vreg_t(vgpu, PIPE_DATA_N1(TRANSCODER_A)) = 0x800000; vgpu_vreg_t(vgpu, PIPE_LINK_M1(TRANSCODER_A)) = 0x3cd6e; @@ -496,9 +497,9 @@ static void emulate_monitor_status_change(struct intel_vgpu *vgpu) /* Disable Primary/Sprite/Cursor plane */ for_each_pipe(dev_priv, pipe) { - vgpu_vreg_t(vgpu, DSPCNTR(pipe)) &= ~DISPLAY_PLANE_ENABLE; + vgpu_vreg_t(vgpu, DSPCNTR(pipe)) &= ~DISP_ENABLE; vgpu_vreg_t(vgpu, SPRCTL(pipe)) &= ~SPRITE_ENABLE; - vgpu_vreg_t(vgpu, CURCNTR(pipe)) &= ~MCURSOR_MODE; + vgpu_vreg_t(vgpu, CURCNTR(pipe)) &= ~MCURSOR_MODE_MASK; vgpu_vreg_t(vgpu, CURCNTR(pipe)) |= MCURSOR_MODE_DISABLE; } diff --git a/drivers/gpu/drm/i915/gvt/dmabuf.c b/drivers/gpu/drm/i915/gvt/dmabuf.c index 8e65cd8258b9..d02a48a5335a 100644 --- a/drivers/gpu/drm/i915/gvt/dmabuf.c +++ b/drivers/gpu/drm/i915/gvt/dmabuf.c @@ -32,6 +32,7 @@ #include <linux/vfio.h> #include "i915_drv.h" +#include "i915_reg.h" #include "gvt.h" #define GEN8_DECODE_PTE(pte) (pte & GENMASK_ULL(63, 12)) @@ -148,8 +149,7 @@ static void dmabuf_gem_object_free(struct kref *kref) if (vgpu && vgpu->active && !list_empty(&vgpu->dmabuf_obj_list_head)) { list_for_each(pos, &vgpu->dmabuf_obj_list_head) { - dmabuf_obj = container_of(pos, - struct intel_vgpu_dmabuf_obj, list); + dmabuf_obj = list_entry(pos, struct intel_vgpu_dmabuf_obj, list); if (dmabuf_obj == obj) { list_del(pos); intel_gvt_hypervisor_put_vfio_device(vgpu); @@ -357,10 +357,8 @@ pick_dmabuf_by_info(struct intel_vgpu *vgpu, struct intel_vgpu_dmabuf_obj *ret = NULL; list_for_each(pos, &vgpu->dmabuf_obj_list_head) { - dmabuf_obj = container_of(pos, struct intel_vgpu_dmabuf_obj, - list); - if ((dmabuf_obj == NULL) || - (dmabuf_obj->info == NULL)) + dmabuf_obj = list_entry(pos, struct intel_vgpu_dmabuf_obj, list); + if (!dmabuf_obj->info) continue; fb_info = (struct intel_vgpu_fb_info *)dmabuf_obj->info; @@ -387,11 +385,7 @@ pick_dmabuf_by_num(struct intel_vgpu *vgpu, u32 id) struct intel_vgpu_dmabuf_obj *ret = NULL; list_for_each(pos, &vgpu->dmabuf_obj_list_head) { - dmabuf_obj = container_of(pos, struct intel_vgpu_dmabuf_obj, - list); - if (!dmabuf_obj) - continue; - + dmabuf_obj = list_entry(pos, struct intel_vgpu_dmabuf_obj, list); if (dmabuf_obj->dmabuf_id == id) { ret = dmabuf_obj; break; @@ -600,8 +594,7 @@ void intel_vgpu_dmabuf_cleanup(struct intel_vgpu *vgpu) mutex_lock(&vgpu->dmabuf_lock); list_for_each_safe(pos, n, &vgpu->dmabuf_obj_list_head) { - dmabuf_obj = container_of(pos, struct intel_vgpu_dmabuf_obj, - list); + dmabuf_obj = list_entry(pos, struct intel_vgpu_dmabuf_obj, list); dmabuf_obj->vgpu = NULL; idr_remove(&vgpu->object_idr, dmabuf_obj->dmabuf_id); diff --git a/drivers/gpu/drm/i915/gvt/edid.c b/drivers/gpu/drm/i915/gvt/edid.c index 22247805c345..a30ba2d7b7ba 100644 --- a/drivers/gpu/drm/i915/gvt/edid.c +++ b/drivers/gpu/drm/i915/gvt/edid.c @@ -33,6 +33,7 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "gvt.h" #define GMBUS1_TOTAL_BYTES_SHIFT 16 diff --git a/drivers/gpu/drm/i915/gvt/fb_decoder.c b/drivers/gpu/drm/i915/gvt/fb_decoder.c index 9ec064199364..835c3fde8a20 100644 --- a/drivers/gpu/drm/i915/gvt/fb_decoder.c +++ b/drivers/gpu/drm/i915/gvt/fb_decoder.c @@ -37,15 +37,16 @@ #include "i915_drv.h" #include "gvt.h" #include "i915_pvinfo.h" +#include "i915_reg.h" #define PRIMARY_FORMAT_NUM 16 struct pixel_format { - int drm_format; /* Pixel format in DRM definition */ - int bpp; /* Bits per pixel, 0 indicates invalid */ - char *desc; /* The description */ + int drm_format; /* Pixel format in DRM definition */ + int bpp; /* Bits per pixel, 0 indicates invalid */ + const char *desc; /* The description */ }; -static struct pixel_format bdw_pixel_formats[] = { +static const struct pixel_format bdw_pixel_formats[] = { {DRM_FORMAT_C8, 8, "8-bit Indexed"}, {DRM_FORMAT_RGB565, 16, "16-bit BGRX (5:6:5 MSB-R:G:B)"}, {DRM_FORMAT_XRGB8888, 32, "32-bit BGRX (8:8:8:8 MSB-X:R:G:B)"}, @@ -58,7 +59,7 @@ static struct pixel_format bdw_pixel_formats[] = { {0, 0, NULL}, }; -static struct pixel_format skl_pixel_formats[] = { +static const struct pixel_format skl_pixel_formats[] = { {DRM_FORMAT_YUYV, 16, "16-bit packed YUYV (8:8:8:8 MSB-V:Y2:U:Y1)"}, {DRM_FORMAT_UYVY, 16, "16-bit packed UYVY (8:8:8:8 MSB-Y2:V:Y1:U)"}, {DRM_FORMAT_YVYU, 16, "16-bit packed YVYU (8:8:8:8 MSB-U:Y2:V:Y1)"}, @@ -83,22 +84,22 @@ static int bdw_format_to_drm(int format) int bdw_pixel_formats_index = 6; switch (format) { - case DISPPLANE_8BPP: + case DISP_FORMAT_8BPP: bdw_pixel_formats_index = 0; break; - case DISPPLANE_BGRX565: + case DISP_FORMAT_BGRX565: bdw_pixel_formats_index = 1; break; - case DISPPLANE_BGRX888: + case DISP_FORMAT_BGRX888: bdw_pixel_formats_index = 2; break; - case DISPPLANE_RGBX101010: + case DISP_FORMAT_RGBX101010: bdw_pixel_formats_index = 3; break; - case DISPPLANE_BGRX101010: + case DISP_FORMAT_BGRX101010: bdw_pixel_formats_index = 4; break; - case DISPPLANE_RGBX888: + case DISP_FORMAT_RGBX888: bdw_pixel_formats_index = 5; break; @@ -211,14 +212,14 @@ int intel_vgpu_decode_primary_plane(struct intel_vgpu *vgpu, return -ENODEV; val = vgpu_vreg_t(vgpu, DSPCNTR(pipe)); - plane->enabled = !!(val & DISPLAY_PLANE_ENABLE); + plane->enabled = !!(val & DISP_ENABLE); if (!plane->enabled) return -ENODEV; if (GRAPHICS_VER(dev_priv) >= 9) { plane->tiled = val & PLANE_CTL_TILED_MASK; fmt = skl_format_to_drm( - val & PLANE_CTL_FORMAT_MASK, + val & PLANE_CTL_FORMAT_MASK_SKL, val & PLANE_CTL_ORDER_RGBX, val & PLANE_CTL_ALPHA_MASK, val & PLANE_CTL_YUV422_ORDER_MASK); @@ -231,8 +232,8 @@ int intel_vgpu_decode_primary_plane(struct intel_vgpu *vgpu, plane->bpp = skl_pixel_formats[fmt].bpp; plane->drm_format = skl_pixel_formats[fmt].drm_format; } else { - plane->tiled = val & DISPPLANE_TILED; - fmt = bdw_format_to_drm(val & DISPPLANE_PIXFORMAT_MASK); + plane->tiled = val & DISP_TILED; + fmt = bdw_format_to_drm(val & DISP_FORMAT_MASK); plane->bpp = bdw_pixel_formats[fmt].bpp; plane->drm_format = bdw_pixel_formats[fmt].drm_format; } @@ -278,14 +279,14 @@ int intel_vgpu_decode_primary_plane(struct intel_vgpu *vgpu, #define CURSOR_FORMAT_NUM (1 << 6) struct cursor_mode_format { - int drm_format; /* Pixel format in DRM definition */ - u8 bpp; /* Bits per pixel; 0 indicates invalid */ - u32 width; /* In pixel */ - u32 height; /* In lines */ - char *desc; /* The description */ + int drm_format; /* Pixel format in DRM definition */ + u8 bpp; /* Bits per pixel; 0 indicates invalid */ + u32 width; /* In pixel */ + u32 height; /* In lines */ + const char *desc; /* The description */ }; -static struct cursor_mode_format cursor_pixel_formats[] = { +static const struct cursor_mode_format cursor_pixel_formats[] = { {DRM_FORMAT_ARGB8888, 32, 128, 128, "128x128 32bpp ARGB"}, {DRM_FORMAT_ARGB8888, 32, 256, 256, "256x256 32bpp ARGB"}, {DRM_FORMAT_ARGB8888, 32, 64, 64, "64x64 32bpp ARGB"}, @@ -342,7 +343,7 @@ int intel_vgpu_decode_cursor_plane(struct intel_vgpu *vgpu, return -ENODEV; val = vgpu_vreg_t(vgpu, CURCNTR(pipe)); - mode = val & MCURSOR_MODE; + mode = val & MCURSOR_MODE_MASK; plane->enabled = (mode != MCURSOR_MODE_DISABLE); if (!plane->enabled) return -ENODEV; @@ -391,7 +392,7 @@ int intel_vgpu_decode_cursor_plane(struct intel_vgpu *vgpu, #define SPRITE_FORMAT_NUM (1 << 3) -static struct pixel_format sprite_pixel_formats[SPRITE_FORMAT_NUM] = { +static const struct pixel_format sprite_pixel_formats[SPRITE_FORMAT_NUM] = { [0x0] = {DRM_FORMAT_YUV422, 16, "YUV 16-bit 4:2:2 packed"}, [0x1] = {DRM_FORMAT_XRGB2101010, 32, "RGB 32-bit 2:10:10:10"}, [0x2] = {DRM_FORMAT_XRGB8888, 32, "RGB 32-bit 8:8:8:8"}, @@ -430,7 +431,7 @@ int intel_vgpu_decode_sprite_plane(struct intel_vgpu *vgpu, yuv_order = (val & SPRITE_YUV_ORDER_MASK) >> _SPRITE_YUV_ORDER_SHIFT; - fmt = (val & SPRITE_PIXFORMAT_MASK) >> _SPRITE_FMT_SHIFT; + fmt = (val & SPRITE_FORMAT_MASK) >> _SPRITE_FMT_SHIFT; if (!sprite_pixel_formats[fmt].bpp) { gvt_vgpu_err("Non-supported pixel format (0x%x)\n", fmt); return -EINVAL; diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index af79b39048f7..d4082f4b9be1 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -38,6 +38,8 @@ #include "i915_pvinfo.h" #include "trace.h" +#include "gt/intel_gt_regs.h" + #if defined(VERBOSE_DEBUG) #define gvt_vdbg_mm(fmt, args...) gvt_dbg_mm(fmt, ##args) #else @@ -185,7 +187,7 @@ struct gtt_type_table_entry { .pse_entry_type = pse_type, \ } -static struct gtt_type_table_entry gtt_type_table[] = { +static const struct gtt_type_table_entry gtt_type_table[] = { GTT_TYPE_TABLE_ENTRY(GTT_TYPE_PPGTT_ROOT_L4_ENTRY, GTT_TYPE_PPGTT_ROOT_L4_ENTRY, GTT_TYPE_INVALID, @@ -499,7 +501,7 @@ DEFINE_PPGTT_GMA_TO_INDEX(gen8, l3_pdp, (gma >> 30 & 0x3)); DEFINE_PPGTT_GMA_TO_INDEX(gen8, l4_pdp, (gma >> 30 & 0x1ff)); DEFINE_PPGTT_GMA_TO_INDEX(gen8, pml4, (gma >> 39 & 0x1ff)); -static struct intel_gvt_gtt_pte_ops gen8_gtt_pte_ops = { +static const struct intel_gvt_gtt_pte_ops gen8_gtt_pte_ops = { .get_entry = gtt_get_entry64, .set_entry = gtt_set_entry64, .clear_present = gtt_entry_clear_present, @@ -516,7 +518,7 @@ static struct intel_gvt_gtt_pte_ops gen8_gtt_pte_ops = { .set_pfn = gen8_gtt_set_pfn, }; -static struct intel_gvt_gtt_gma_ops gen8_gtt_gma_ops = { +static const struct intel_gvt_gtt_gma_ops gen8_gtt_gma_ops = { .gma_to_ggtt_pte_index = gma_to_ggtt_pte_index, .gma_to_pte_index = gen8_gma_to_pte_index, .gma_to_pde_index = gen8_gma_to_pde_index, @@ -526,7 +528,7 @@ static struct intel_gvt_gtt_gma_ops gen8_gtt_gma_ops = { }; /* Update entry type per pse and ips bit. */ -static void update_entry_type_for_real(struct intel_gvt_gtt_pte_ops *pte_ops, +static void update_entry_type_for_real(const struct intel_gvt_gtt_pte_ops *pte_ops, struct intel_gvt_gtt_entry *entry, bool ips) { switch (entry->type) { @@ -553,7 +555,7 @@ static void _ppgtt_get_root_entry(struct intel_vgpu_mm *mm, struct intel_gvt_gtt_entry *entry, unsigned long index, bool guest) { - struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; GEM_BUG_ON(mm->type != INTEL_GVT_MM_PPGTT); @@ -580,7 +582,7 @@ static void _ppgtt_set_root_entry(struct intel_vgpu_mm *mm, struct intel_gvt_gtt_entry *entry, unsigned long index, bool guest) { - struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; pte_ops->set_entry(guest ? mm->ppgtt_mm.guest_pdps : mm->ppgtt_mm.shadow_pdps, @@ -596,7 +598,7 @@ static inline void ppgtt_set_shadow_root_entry(struct intel_vgpu_mm *mm, static void ggtt_get_guest_entry(struct intel_vgpu_mm *mm, struct intel_gvt_gtt_entry *entry, unsigned long index) { - struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; GEM_BUG_ON(mm->type != INTEL_GVT_MM_GGTT); @@ -608,7 +610,7 @@ static void ggtt_get_guest_entry(struct intel_vgpu_mm *mm, static void ggtt_set_guest_entry(struct intel_vgpu_mm *mm, struct intel_gvt_gtt_entry *entry, unsigned long index) { - struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; GEM_BUG_ON(mm->type != INTEL_GVT_MM_GGTT); @@ -619,7 +621,7 @@ static void ggtt_set_guest_entry(struct intel_vgpu_mm *mm, static void ggtt_get_host_entry(struct intel_vgpu_mm *mm, struct intel_gvt_gtt_entry *entry, unsigned long index) { - struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; GEM_BUG_ON(mm->type != INTEL_GVT_MM_GGTT); @@ -629,7 +631,7 @@ static void ggtt_get_host_entry(struct intel_vgpu_mm *mm, static void ggtt_set_host_entry(struct intel_vgpu_mm *mm, struct intel_gvt_gtt_entry *entry, unsigned long index) { - struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = mm->vgpu->gvt->gtt.pte_ops; unsigned long offset = index; GEM_BUG_ON(mm->type != INTEL_GVT_MM_GGTT); @@ -655,7 +657,7 @@ static inline int ppgtt_spt_get_entry( bool guest) { struct intel_gvt *gvt = spt->vgpu->gvt; - struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; int ret; e->type = get_entry_type(type); @@ -684,7 +686,7 @@ static inline int ppgtt_spt_set_entry( bool guest) { struct intel_gvt *gvt = spt->vgpu->gvt; - struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; if (WARN(!gtt_type_is_entry(e->type), "invalid entry type\n")) return -EINVAL; @@ -947,7 +949,7 @@ static int ppgtt_invalidate_spt_by_shadow_entry(struct intel_vgpu *vgpu, struct intel_gvt_gtt_entry *e) { struct drm_i915_private *i915 = vgpu->gvt->gt->i915; - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; struct intel_vgpu_ppgtt_spt *s; enum intel_gvt_gtt_type cur_pt_type; @@ -984,7 +986,7 @@ static inline void ppgtt_invalidate_pte(struct intel_vgpu_ppgtt_spt *spt, struct intel_gvt_gtt_entry *entry) { struct intel_vgpu *vgpu = spt->vgpu; - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; unsigned long pfn; int type; @@ -1072,7 +1074,7 @@ static int ppgtt_populate_spt(struct intel_vgpu_ppgtt_spt *spt); static struct intel_vgpu_ppgtt_spt *ppgtt_populate_spt_by_guest_entry( struct intel_vgpu *vgpu, struct intel_gvt_gtt_entry *we) { - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; struct intel_vgpu_ppgtt_spt *spt = NULL; bool ips = false; int ret; @@ -1136,7 +1138,7 @@ err: static inline void ppgtt_generate_shadow_entry(struct intel_gvt_gtt_entry *se, struct intel_vgpu_ppgtt_spt *s, struct intel_gvt_gtt_entry *ge) { - struct intel_gvt_gtt_pte_ops *ops = s->vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = s->vgpu->gvt->gtt.pte_ops; se->type = ge->type; se->val64 = ge->val64; @@ -1159,7 +1161,7 @@ static inline void ppgtt_generate_shadow_entry(struct intel_gvt_gtt_entry *se, static int is_2MB_gtt_possible(struct intel_vgpu *vgpu, struct intel_gvt_gtt_entry *entry) { - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; unsigned long pfn; if (!HAS_PAGE_SIZES(vgpu->gvt->gt->i915, I915_GTT_PAGE_SIZE_2M)) @@ -1176,7 +1178,7 @@ static int split_2MB_gtt_entry(struct intel_vgpu *vgpu, struct intel_vgpu_ppgtt_spt *spt, unsigned long index, struct intel_gvt_gtt_entry *se) { - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; struct intel_vgpu_ppgtt_spt *sub_spt; struct intel_gvt_gtt_entry sub_se; unsigned long start_gfn; @@ -1223,7 +1225,7 @@ static int split_64KB_gtt_entry(struct intel_vgpu *vgpu, struct intel_vgpu_ppgtt_spt *spt, unsigned long index, struct intel_gvt_gtt_entry *se) { - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; struct intel_gvt_gtt_entry entry = *se; unsigned long start_gfn; dma_addr_t dma_addr; @@ -1254,7 +1256,7 @@ static int ppgtt_populate_shadow_entry(struct intel_vgpu *vgpu, struct intel_vgpu_ppgtt_spt *spt, unsigned long index, struct intel_gvt_gtt_entry *ge) { - struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops; struct intel_gvt_gtt_entry se = *ge; unsigned long gfn, page_size = PAGE_SIZE; dma_addr_t dma_addr; @@ -1308,7 +1310,7 @@ static int ppgtt_populate_spt(struct intel_vgpu_ppgtt_spt *spt) { struct intel_vgpu *vgpu = spt->vgpu; struct intel_gvt *gvt = vgpu->gvt; - struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; struct intel_vgpu_ppgtt_spt *s; struct intel_gvt_gtt_entry se, ge; unsigned long gfn, i; @@ -1351,7 +1353,7 @@ static int ppgtt_handle_guest_entry_removal(struct intel_vgpu_ppgtt_spt *spt, struct intel_gvt_gtt_entry *se, unsigned long index) { struct intel_vgpu *vgpu = spt->vgpu; - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; int ret; trace_spt_guest_change(spt->vgpu->id, "remove", spt, @@ -1432,7 +1434,7 @@ static int sync_oos_page(struct intel_vgpu *vgpu, { const struct intel_gvt_device_info *info = &vgpu->gvt->device_info; struct intel_gvt *gvt = vgpu->gvt; - struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; struct intel_vgpu_ppgtt_spt *spt = oos_page->spt; struct intel_gvt_gtt_entry old, new; int index; @@ -1603,7 +1605,7 @@ static int ppgtt_handle_guest_write_page_table( { struct intel_vgpu *vgpu = spt->vgpu; int type = spt->shadow_page.type; - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; struct intel_gvt_gtt_entry old_se; int new_present; int i, ret; @@ -1720,7 +1722,7 @@ static int ppgtt_handle_guest_write_page_table_bytes( u64 pa, void *p_data, int bytes) { struct intel_vgpu *vgpu = spt->vgpu; - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; const struct intel_gvt_device_info *info = &vgpu->gvt->device_info; struct intel_gvt_gtt_entry we, se; unsigned long index; @@ -1785,7 +1787,7 @@ static void invalidate_ppgtt_mm(struct intel_vgpu_mm *mm) struct intel_vgpu *vgpu = mm->vgpu; struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_gtt *gtt = &gvt->gtt; - struct intel_gvt_gtt_pte_ops *ops = gtt->pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = gtt->pte_ops; struct intel_gvt_gtt_entry se; int index; @@ -1815,7 +1817,7 @@ static int shadow_ppgtt_mm(struct intel_vgpu_mm *mm) struct intel_vgpu *vgpu = mm->vgpu; struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_gtt *gtt = &gvt->gtt; - struct intel_gvt_gtt_pte_ops *ops = gtt->pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = gtt->pte_ops; struct intel_vgpu_ppgtt_spt *spt; struct intel_gvt_gtt_entry ge, se; int index, ret; @@ -2067,7 +2069,7 @@ static inline int ppgtt_get_next_level_entry(struct intel_vgpu_mm *mm, struct intel_gvt_gtt_entry *e, unsigned long index, bool guest) { struct intel_vgpu *vgpu = mm->vgpu; - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; struct intel_vgpu_ppgtt_spt *s; s = intel_vgpu_find_spt_by_mfn(vgpu, ops->get_pfn(e)); @@ -2096,8 +2098,8 @@ unsigned long intel_vgpu_gma_to_gpa(struct intel_vgpu_mm *mm, unsigned long gma) { struct intel_vgpu *vgpu = mm->vgpu; struct intel_gvt *gvt = vgpu->gvt; - struct intel_gvt_gtt_pte_ops *pte_ops = gvt->gtt.pte_ops; - struct intel_gvt_gtt_gma_ops *gma_ops = gvt->gtt.gma_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = gvt->gtt.pte_ops; + const struct intel_gvt_gtt_gma_ops *gma_ops = gvt->gtt.gma_ops; unsigned long gpa = INTEL_GVT_INVALID_ADDR; unsigned long gma_index[4]; struct intel_gvt_gtt_entry e; @@ -2221,7 +2223,7 @@ int intel_vgpu_emulate_ggtt_mmio_read(struct intel_vgpu *vgpu, unsigned int off, static void ggtt_invalidate_pte(struct intel_vgpu *vgpu, struct intel_gvt_gtt_entry *entry) { - struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops; unsigned long pfn; pfn = pte_ops->get_pfn(entry); @@ -2236,7 +2238,7 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off, struct intel_gvt *gvt = vgpu->gvt; const struct intel_gvt_device_info *info = &gvt->device_info; struct intel_vgpu_mm *ggtt_mm = vgpu->gtt.ggtt_mm; - struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; unsigned long g_gtt_index = off >> info->gtt_entry_size_shift; unsigned long gma, gfn; struct intel_gvt_gtt_entry e = {.val64 = 0, .type = GTT_TYPE_GGTT_PTE}; @@ -2391,7 +2393,7 @@ static int alloc_scratch_pages(struct intel_vgpu *vgpu, { struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_vgpu_gtt *gtt = &vgpu->gtt; - struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; int page_entry_num = I915_GTT_PAGE_SIZE >> vgpu->gvt->device_info.gtt_entry_size_shift; void *scratch_pt; @@ -2822,7 +2824,7 @@ void intel_vgpu_invalidate_ppgtt(struct intel_vgpu *vgpu) void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu, bool invalidate_old) { struct intel_gvt *gvt = vgpu->gvt; - struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops; struct intel_gvt_gtt_entry entry = {.type = GTT_TYPE_GGTT_PTE}; struct intel_gvt_gtt_entry old_entry; u32 index; diff --git a/drivers/gpu/drm/i915/gvt/gtt.h b/drivers/gpu/drm/i915/gvt/gtt.h index 3bf45672ef98..a3b0f59ec8bd 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.h +++ b/drivers/gpu/drm/i915/gvt/gtt.h @@ -91,8 +91,8 @@ struct intel_gvt_gtt_gma_ops { }; struct intel_gvt_gtt { - struct intel_gvt_gtt_pte_ops *pte_ops; - struct intel_gvt_gtt_gma_ops *gma_ops; + const struct intel_gvt_gtt_pte_ops *pte_ops; + const struct intel_gvt_gtt_gma_ops *gma_ops; int (*mm_alloc_page_table)(struct intel_vgpu_mm *mm); void (*mm_free_page_table)(struct intel_vgpu_mm *mm); struct list_head oos_page_use_list_head; diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h index 0c0615602343..0ebffc327528 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.h +++ b/drivers/gpu/drm/i915/gvt/gvt.h @@ -272,7 +272,7 @@ struct intel_gvt_mmio { /* Value of command write of this reg needs to be patched */ #define F_CMD_WRITE_PATCH (1 << 8) - struct gvt_mmio_block *mmio_block; + const struct gvt_mmio_block *mmio_block; unsigned int num_mmio_block; DECLARE_HASHTABLE(mmio_info_table, INTEL_GVT_MMIO_HASH_BITS); diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index cde0a477fb49..c2ae79092b14 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -37,9 +37,12 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "gvt.h" #include "i915_pvinfo.h" #include "display/intel_display_types.h" +#include "display/intel_fbc.h" +#include "gt/intel_gt_regs.h" /* XXX FIXME i915 has changed PP_XXX definition */ #define PCH_PP_STATUS _MMIO(0xc7200) @@ -701,11 +704,11 @@ static int pipeconf_mmio_write(struct intel_vgpu *vgpu, unsigned int offset, data = vgpu_vreg(vgpu, offset); if (data & PIPECONF_ENABLE) { - vgpu_vreg(vgpu, offset) |= I965_PIPECONF_ACTIVE; + vgpu_vreg(vgpu, offset) |= PIPECONF_STATE_ENABLE; vgpu_update_refresh_rate(vgpu); vgpu_update_vblank_emulation(vgpu, true); } else { - vgpu_vreg(vgpu, offset) &= ~I965_PIPECONF_ACTIVE; + vgpu_vreg(vgpu, offset) &= ~PIPECONF_STATE_ENABLE; vgpu_update_vblank_emulation(vgpu, false); } return 0; @@ -2647,12 +2650,12 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_D(_MMIO(_TRANSA_CHICKEN2), D_ALL); MMIO_D(_MMIO(_TRANSB_CHICKEN2), D_ALL); - MMIO_D(ILK_DPFC_CB_BASE, D_ALL); - MMIO_D(ILK_DPFC_CONTROL, D_ALL); - MMIO_D(ILK_DPFC_RECOMP_CTL, D_ALL); - MMIO_D(ILK_DPFC_STATUS, D_ALL); - MMIO_D(ILK_DPFC_FENCE_YOFF, D_ALL); - MMIO_D(ILK_DPFC_CHICKEN, D_ALL); + MMIO_D(ILK_DPFC_CB_BASE(INTEL_FBC_A), D_ALL); + MMIO_D(ILK_DPFC_CONTROL(INTEL_FBC_A), D_ALL); + MMIO_D(ILK_DPFC_RECOMP_CTL(INTEL_FBC_A), D_ALL); + MMIO_D(ILK_DPFC_STATUS(INTEL_FBC_A), D_ALL); + MMIO_D(ILK_DPFC_FENCE_YOFF(INTEL_FBC_A), D_ALL); + MMIO_D(ILK_DPFC_CHICKEN(INTEL_FBC_A), D_ALL); MMIO_D(ILK_FBC_RT_BASE, D_ALL); MMIO_D(IPS_CTL, D_ALL); @@ -2876,9 +2879,9 @@ static int init_generic_mmio_info(struct intel_gvt *gvt) MMIO_D(_MMIO(0x3c), D_ALL); MMIO_D(_MMIO(0x860), D_ALL); - MMIO_D(ECOSKPD, D_ALL); + MMIO_D(ECOSKPD(RENDER_RING_BASE), D_ALL); MMIO_D(_MMIO(0x121d0), D_ALL); - MMIO_D(GEN6_BLITTER_ECOSKPD, D_ALL); + MMIO_D(ECOSKPD(BLT_RING_BASE), D_ALL); MMIO_D(_MMIO(0x41d0), D_ALL); MMIO_D(GAC_ECO_BITS, D_ALL); MMIO_D(_MMIO(0x6200), D_ALL); @@ -3436,6 +3439,7 @@ static int init_skl_mmio_info(struct intel_gvt *gvt) MMIO_DFH(GAMT_CHKN_BIT_REG, D_KBL | D_CFL, F_CMD_ACCESS, NULL, NULL); MMIO_D(GEN9_CTX_PREEMPT_REG, D_SKL_PLUS & ~D_BXT); + MMIO_DFH(_MMIO(0xe4cc), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL); return 0; } @@ -3627,11 +3631,11 @@ static int init_bxt_mmio_info(struct intel_gvt *gvt) return 0; } -static struct gvt_mmio_block *find_mmio_block(struct intel_gvt *gvt, - unsigned int offset) +static const struct gvt_mmio_block *find_mmio_block(struct intel_gvt *gvt, + unsigned int offset) { unsigned long device = intel_gvt_get_device_type(gvt); - struct gvt_mmio_block *block = gvt->mmio.mmio_block; + const struct gvt_mmio_block *block = gvt->mmio.mmio_block; int num = gvt->mmio.num_mmio_block; int i; @@ -3670,7 +3674,7 @@ void intel_gvt_clean_mmio_info(struct intel_gvt *gvt) * accessible (should have no F_CMD_ACCESS flag). * otherwise, need to update cmd_reg_handler in cmd_parser.c */ -static struct gvt_mmio_block mmio_blocks[] = { +static const struct gvt_mmio_block mmio_blocks[] = { {D_SKL_PLUS, _MMIO(DMC_MMIO_START_RANGE), 0x3000, NULL, NULL}, {D_ALL, _MMIO(MCHBAR_MIRROR_BASE_SNB), 0x40000, NULL, NULL}, {D_ALL, _MMIO(VGT_PVINFO_PAGE), VGT_PVINFO_SIZE, @@ -3753,7 +3757,7 @@ int intel_gvt_for_each_tracked_mmio(struct intel_gvt *gvt, int (*handler)(struct intel_gvt *gvt, u32 offset, void *data), void *data) { - struct gvt_mmio_block *block = gvt->mmio.mmio_block; + const struct gvt_mmio_block *block = gvt->mmio.mmio_block; struct intel_gvt_mmio_info *e; int i, j, ret; @@ -3871,7 +3875,7 @@ int intel_vgpu_mmio_reg_rw(struct intel_vgpu *vgpu, unsigned int offset, struct drm_i915_private *i915 = vgpu->gvt->gt->i915; struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_mmio_info *mmio_info; - struct gvt_mmio_block *mmio_block; + const struct gvt_mmio_block *mmio_block; gvt_mmio_func func; int ret; diff --git a/drivers/gpu/drm/i915/gvt/interrupt.c b/drivers/gpu/drm/i915/gvt/interrupt.c index 614b951d919f..228f623d466d 100644 --- a/drivers/gpu/drm/i915/gvt/interrupt.c +++ b/drivers/gpu/drm/i915/gvt/interrupt.c @@ -30,6 +30,7 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "gvt.h" #include "trace.h" @@ -176,7 +177,7 @@ int intel_vgpu_reg_imr_handler(struct intel_vgpu *vgpu, unsigned int reg, void *p_data, unsigned int bytes) { struct intel_gvt *gvt = vgpu->gvt; - struct intel_gvt_irq_ops *ops = gvt->irq.ops; + const struct intel_gvt_irq_ops *ops = gvt->irq.ops; u32 imr = *(u32 *)p_data; trace_write_ir(vgpu->id, "IMR", reg, imr, vgpu_vreg(vgpu, reg), @@ -206,7 +207,7 @@ int intel_vgpu_reg_master_irq_handler(struct intel_vgpu *vgpu, unsigned int reg, void *p_data, unsigned int bytes) { struct intel_gvt *gvt = vgpu->gvt; - struct intel_gvt_irq_ops *ops = gvt->irq.ops; + const struct intel_gvt_irq_ops *ops = gvt->irq.ops; u32 ier = *(u32 *)p_data; u32 virtual_ier = vgpu_vreg(vgpu, reg); @@ -246,7 +247,7 @@ int intel_vgpu_reg_ier_handler(struct intel_vgpu *vgpu, { struct intel_gvt *gvt = vgpu->gvt; struct drm_i915_private *i915 = gvt->gt->i915; - struct intel_gvt_irq_ops *ops = gvt->irq.ops; + const struct intel_gvt_irq_ops *ops = gvt->irq.ops; struct intel_gvt_irq_info *info; u32 ier = *(u32 *)p_data; @@ -604,7 +605,7 @@ static void gen8_init_irq( SET_BIT_INFO(irq, 25, PCU_PCODE2DRIVER_MAILBOX, INTEL_GVT_IRQ_INFO_PCU); } -static struct intel_gvt_irq_ops gen8_irq_ops = { +static const struct intel_gvt_irq_ops gen8_irq_ops = { .init_irq = gen8_init_irq, .check_pending_irq = gen8_check_pending_irq, }; @@ -626,7 +627,7 @@ void intel_vgpu_trigger_virtual_event(struct intel_vgpu *vgpu, struct intel_gvt *gvt = vgpu->gvt; struct intel_gvt_irq *irq = &gvt->irq; gvt_event_virt_handler_t handler; - struct intel_gvt_irq_ops *ops = gvt->irq.ops; + const struct intel_gvt_irq_ops *ops = gvt->irq.ops; handler = get_event_virt_handler(irq, event); drm_WARN_ON(&i915->drm, !handler); diff --git a/drivers/gpu/drm/i915/gvt/interrupt.h b/drivers/gpu/drm/i915/gvt/interrupt.h index 6c47d3e33161..b62f04ab47cb 100644 --- a/drivers/gpu/drm/i915/gvt/interrupt.h +++ b/drivers/gpu/drm/i915/gvt/interrupt.h @@ -35,7 +35,7 @@ #include <linux/hrtimer.h> #include <linux/kernel.h> -#include "i915_reg.h" +#include "i915_reg_defs.h" enum intel_gvt_event_type { RCS_MI_USER_INTERRUPT = 0, @@ -203,7 +203,7 @@ struct intel_gvt_irq_map { /* structure containing device specific IRQ state */ struct intel_gvt_irq { - struct intel_gvt_irq_ops *ops; + const struct intel_gvt_irq_ops *ops; struct intel_gvt_irq_info *info[INTEL_GVT_IRQ_INFO_MAX]; DECLARE_BITMAP(irq_info_bitmap, INTEL_GVT_IRQ_INFO_MAX); struct intel_gvt_event_info events[INTEL_GVT_EVENT_MAX]; diff --git a/drivers/gpu/drm/i915/gvt/mmio.c b/drivers/gpu/drm/i915/gvt/mmio.c index 24210b1eaec5..5db0ef83d522 100644 --- a/drivers/gpu/drm/i915/gvt/mmio.c +++ b/drivers/gpu/drm/i915/gvt/mmio.c @@ -34,8 +34,11 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "gvt.h" +#include "gt/intel_gt_regs.h" + /** * intel_vgpu_gpa_to_mmio_offset - translate a GPA to MMIO offset * @vgpu: a vGPU diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c index f776c470914d..c85bafe7539e 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.c +++ b/drivers/gpu/drm/i915/gvt/mmio_context.c @@ -35,7 +35,9 @@ #include "i915_drv.h" #include "gt/intel_context.h" +#include "gt/intel_engine_regs.h" #include "gt/intel_gpu_commands.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_ring.h" #include "gvt.h" #include "trace.h" @@ -44,7 +46,7 @@ /* Raw offset is appened to each line for convenience. */ static struct engine_mmio gen8_engine_mmio_list[] __cacheline_aligned = { - {RCS0, GFX_MODE_GEN7, 0xffff, false}, /* 0x229c */ + {RCS0, RING_MODE_GEN7(RENDER_RING_BASE), 0xffff, false}, /* 0x229c */ {RCS0, GEN9_CTX_PREEMPT_REG, 0x0, false}, /* 0x2248 */ {RCS0, HWSTAM, 0x0, false}, /* 0x2098 */ {RCS0, INSTPM, 0xffff, true}, /* 0x20c0 */ @@ -76,7 +78,7 @@ static struct engine_mmio gen8_engine_mmio_list[] __cacheline_aligned = { }; static struct engine_mmio gen9_engine_mmio_list[] __cacheline_aligned = { - {RCS0, GFX_MODE_GEN7, 0xffff, false}, /* 0x229c */ + {RCS0, RING_MODE_GEN7(RENDER_RING_BASE), 0xffff, false}, /* 0x229c */ {RCS0, GEN9_CTX_PREEMPT_REG, 0x0, false}, /* 0x2248 */ {RCS0, HWSTAM, 0x0, false}, /* 0x2098 */ {RCS0, INSTPM, 0xffff, true}, /* 0x20c0 */ diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.h b/drivers/gpu/drm/i915/gvt/mmio_context.h index b6b69777af49..9540813b88e5 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.h +++ b/drivers/gpu/drm/i915/gvt/mmio_context.h @@ -38,9 +38,9 @@ #include <linux/types.h> +#include "gt/intel_engine_regs.h" #include "gt/intel_engine_types.h" #include "gt/intel_lrc_reg.h" -#include "i915_reg.h" struct i915_request; struct intel_context; diff --git a/drivers/gpu/drm/i915/gvt/reg.h b/drivers/gpu/drm/i915/gvt/reg.h index 244cc7320b54..7d666d34f9ff 100644 --- a/drivers/gpu/drm/i915/gvt/reg.h +++ b/drivers/gpu/drm/i915/gvt/reg.h @@ -62,7 +62,6 @@ #define SKL_FLIP_EVENT(pipe, plane) (PRIMARY_A_FLIP_DONE + (plane) * 3 + (pipe)) -#define PLANE_CTL_ASYNC_FLIP (1 << 9) #define REG50080_FLIP_TYPE_MASK 0x3 #define REG50080_FLIP_TYPE_ASYNC 0x1 diff --git a/drivers/gpu/drm/i915/gvt/sched_policy.c b/drivers/gpu/drm/i915/gvt/sched_policy.c index 036b74fe9298..c077fb4674f0 100644 --- a/drivers/gpu/drm/i915/gvt/sched_policy.c +++ b/drivers/gpu/drm/i915/gvt/sched_policy.c @@ -368,7 +368,7 @@ static void tbs_sched_stop_schedule(struct intel_vgpu *vgpu) vgpu_data->active = false; } -static struct intel_gvt_sched_policy_ops tbs_schedule_ops = { +static const struct intel_gvt_sched_policy_ops tbs_schedule_ops = { .init = tbs_sched_init, .clean = tbs_sched_clean, .init_vgpu = tbs_sched_init_vgpu, diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index 42a0c9ae0a73..679476da0640 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -38,11 +38,13 @@ #include "gem/i915_gem_pm.h" #include "gt/intel_context.h" #include "gt/intel_execlists_submission.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_lrc.h" #include "gt/intel_ring.h" #include "i915_drv.h" #include "i915_gem_gtt.h" +#include "i915_perf_oa_regs.h" #include "gvt.h" #define RING_CTX_OFF(x) \ diff --git a/drivers/gpu/drm/i915/gvt/scheduler.h b/drivers/gpu/drm/i915/gvt/scheduler.h index 7c86984a842f..1f391b3da2cc 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.h +++ b/drivers/gpu/drm/i915/gvt/scheduler.h @@ -56,7 +56,7 @@ struct intel_gvt_workload_scheduler { wait_queue_head_t waitq[I915_NUM_ENGINES]; void *sched_data; - struct intel_gvt_sched_policy_ops *sched_ops; + const struct intel_gvt_sched_policy_ops *sched_ops; }; #define INDIRECT_CTX_ADDR_MASK 0xffffffc0 diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c index fa6b92615799..8dddd0a940a1 100644 --- a/drivers/gpu/drm/i915/gvt/vgpu.c +++ b/drivers/gpu/drm/i915/gvt/vgpu.c @@ -77,7 +77,7 @@ void populate_pvinfo_page(struct intel_vgpu *vgpu) #define VGPU_WEIGHT(vgpu_num) \ (VGPU_MAX_WEIGHT / (vgpu_num)) -static struct { +static const struct { unsigned int low_mm; unsigned int high_mm; unsigned int fence; @@ -88,7 +88,7 @@ static struct { */ unsigned int weight; enum intel_vgpu_edid edid; - char *name; + const char *name; } vgpu_types[] = { /* Fixed vGPU type table */ { MB_TO_BYTES(64), MB_TO_BYTES(384), 4, VGPU_WEIGHT(8), GVT_EDID_1024_768, "8" }, diff --git a/drivers/gpu/drm/i915/i915_buddy.c b/drivers/gpu/drm/i915/i915_buddy.c deleted file mode 100644 index 6e2ad68f8f3f..000000000000 --- a/drivers/gpu/drm/i915/i915_buddy.c +++ /dev/null @@ -1,466 +0,0 @@ -// SPDX-License-Identifier: MIT -/* - * Copyright © 2021 Intel Corporation - */ - -#include <linux/kmemleak.h> -#include <linux/sizes.h> - -#include "i915_buddy.h" - -#include "i915_gem.h" -#include "i915_utils.h" - -static struct kmem_cache *slab_blocks; - -static struct i915_buddy_block *i915_block_alloc(struct i915_buddy_mm *mm, - struct i915_buddy_block *parent, - unsigned int order, - u64 offset) -{ - struct i915_buddy_block *block; - - GEM_BUG_ON(order > I915_BUDDY_MAX_ORDER); - - block = kmem_cache_zalloc(slab_blocks, GFP_KERNEL); - if (!block) - return NULL; - - block->header = offset; - block->header |= order; - block->parent = parent; - - GEM_BUG_ON(block->header & I915_BUDDY_HEADER_UNUSED); - return block; -} - -static void i915_block_free(struct i915_buddy_mm *mm, - struct i915_buddy_block *block) -{ - kmem_cache_free(slab_blocks, block); -} - -static void mark_allocated(struct i915_buddy_block *block) -{ - block->header &= ~I915_BUDDY_HEADER_STATE; - block->header |= I915_BUDDY_ALLOCATED; - - list_del(&block->link); -} - -static void mark_free(struct i915_buddy_mm *mm, - struct i915_buddy_block *block) -{ - block->header &= ~I915_BUDDY_HEADER_STATE; - block->header |= I915_BUDDY_FREE; - - list_add(&block->link, - &mm->free_list[i915_buddy_block_order(block)]); -} - -static void mark_split(struct i915_buddy_block *block) -{ - block->header &= ~I915_BUDDY_HEADER_STATE; - block->header |= I915_BUDDY_SPLIT; - - list_del(&block->link); -} - -int i915_buddy_init(struct i915_buddy_mm *mm, u64 size, u64 chunk_size) -{ - unsigned int i; - u64 offset; - - if (size < chunk_size) - return -EINVAL; - - if (chunk_size < PAGE_SIZE) - return -EINVAL; - - if (!is_power_of_2(chunk_size)) - return -EINVAL; - - size = round_down(size, chunk_size); - - mm->size = size; - mm->avail = size; - mm->chunk_size = chunk_size; - mm->max_order = ilog2(size) - ilog2(chunk_size); - - GEM_BUG_ON(mm->max_order > I915_BUDDY_MAX_ORDER); - - mm->free_list = kmalloc_array(mm->max_order + 1, - sizeof(struct list_head), - GFP_KERNEL); - if (!mm->free_list) - return -ENOMEM; - - for (i = 0; i <= mm->max_order; ++i) - INIT_LIST_HEAD(&mm->free_list[i]); - - mm->n_roots = hweight64(size); - - mm->roots = kmalloc_array(mm->n_roots, - sizeof(struct i915_buddy_block *), - GFP_KERNEL); - if (!mm->roots) - goto out_free_list; - - offset = 0; - i = 0; - - /* - * Split into power-of-two blocks, in case we are given a size that is - * not itself a power-of-two. - */ - do { - struct i915_buddy_block *root; - unsigned int order; - u64 root_size; - - root_size = rounddown_pow_of_two(size); - order = ilog2(root_size) - ilog2(chunk_size); - - root = i915_block_alloc(mm, NULL, order, offset); - if (!root) - goto out_free_roots; - - mark_free(mm, root); - - GEM_BUG_ON(i > mm->max_order); - GEM_BUG_ON(i915_buddy_block_size(mm, root) < chunk_size); - - mm->roots[i] = root; - - offset += root_size; - size -= root_size; - i++; - } while (size); - - return 0; - -out_free_roots: - while (i--) - i915_block_free(mm, mm->roots[i]); - kfree(mm->roots); -out_free_list: - kfree(mm->free_list); - return -ENOMEM; -} - -void i915_buddy_fini(struct i915_buddy_mm *mm) -{ - int i; - - for (i = 0; i < mm->n_roots; ++i) { - GEM_WARN_ON(!i915_buddy_block_is_free(mm->roots[i])); - i915_block_free(mm, mm->roots[i]); - } - - GEM_WARN_ON(mm->avail != mm->size); - - kfree(mm->roots); - kfree(mm->free_list); -} - -static int split_block(struct i915_buddy_mm *mm, - struct i915_buddy_block *block) -{ - unsigned int block_order = i915_buddy_block_order(block) - 1; - u64 offset = i915_buddy_block_offset(block); - - GEM_BUG_ON(!i915_buddy_block_is_free(block)); - GEM_BUG_ON(!i915_buddy_block_order(block)); - - block->left = i915_block_alloc(mm, block, block_order, offset); - if (!block->left) - return -ENOMEM; - - block->right = i915_block_alloc(mm, block, block_order, - offset + (mm->chunk_size << block_order)); - if (!block->right) { - i915_block_free(mm, block->left); - return -ENOMEM; - } - - mark_free(mm, block->left); - mark_free(mm, block->right); - - mark_split(block); - - return 0; -} - -static struct i915_buddy_block * -get_buddy(struct i915_buddy_block *block) -{ - struct i915_buddy_block *parent; - - parent = block->parent; - if (!parent) - return NULL; - - if (parent->left == block) - return parent->right; - - return parent->left; -} - -static void __i915_buddy_free(struct i915_buddy_mm *mm, - struct i915_buddy_block *block) -{ - struct i915_buddy_block *parent; - - while ((parent = block->parent)) { - struct i915_buddy_block *buddy; - - buddy = get_buddy(block); - - if (!i915_buddy_block_is_free(buddy)) - break; - - list_del(&buddy->link); - - i915_block_free(mm, block); - i915_block_free(mm, buddy); - - block = parent; - } - - mark_free(mm, block); -} - -void i915_buddy_free(struct i915_buddy_mm *mm, - struct i915_buddy_block *block) -{ - GEM_BUG_ON(!i915_buddy_block_is_allocated(block)); - mm->avail += i915_buddy_block_size(mm, block); - __i915_buddy_free(mm, block); -} - -void i915_buddy_free_list(struct i915_buddy_mm *mm, struct list_head *objects) -{ - struct i915_buddy_block *block, *on; - - list_for_each_entry_safe(block, on, objects, link) { - i915_buddy_free(mm, block); - cond_resched(); - } - INIT_LIST_HEAD(objects); -} - -/* - * Allocate power-of-two block. The order value here translates to: - * - * 0 = 2^0 * mm->chunk_size - * 1 = 2^1 * mm->chunk_size - * 2 = 2^2 * mm->chunk_size - * ... - */ -struct i915_buddy_block * -i915_buddy_alloc(struct i915_buddy_mm *mm, unsigned int order) -{ - struct i915_buddy_block *block = NULL; - unsigned int i; - int err; - - for (i = order; i <= mm->max_order; ++i) { - block = list_first_entry_or_null(&mm->free_list[i], - struct i915_buddy_block, - link); - if (block) - break; - } - - if (!block) - return ERR_PTR(-ENOSPC); - - GEM_BUG_ON(!i915_buddy_block_is_free(block)); - - while (i != order) { - err = split_block(mm, block); - if (unlikely(err)) - goto out_free; - - /* Go low */ - block = block->left; - i--; - } - - mark_allocated(block); - mm->avail -= i915_buddy_block_size(mm, block); - kmemleak_update_trace(block); - return block; - -out_free: - if (i != order) - __i915_buddy_free(mm, block); - return ERR_PTR(err); -} - -static inline bool overlaps(u64 s1, u64 e1, u64 s2, u64 e2) -{ - return s1 <= e2 && e1 >= s2; -} - -static inline bool contains(u64 s1, u64 e1, u64 s2, u64 e2) -{ - return s1 <= s2 && e1 >= e2; -} - -/* - * Allocate range. Note that it's safe to chain together multiple alloc_ranges - * with the same blocks list. - * - * Intended for pre-allocating portions of the address space, for example to - * reserve a block for the initial framebuffer or similar, hence the expectation - * here is that i915_buddy_alloc() is still the main vehicle for - * allocations, so if that's not the case then the drm_mm range allocator is - * probably a much better fit, and so you should probably go use that instead. - */ -int i915_buddy_alloc_range(struct i915_buddy_mm *mm, - struct list_head *blocks, - u64 start, u64 size) -{ - struct i915_buddy_block *block; - struct i915_buddy_block *buddy; - LIST_HEAD(allocated); - LIST_HEAD(dfs); - u64 end; - int err; - int i; - - if (size < mm->chunk_size) - return -EINVAL; - - if (!IS_ALIGNED(size | start, mm->chunk_size)) - return -EINVAL; - - if (range_overflows(start, size, mm->size)) - return -EINVAL; - - for (i = 0; i < mm->n_roots; ++i) - list_add_tail(&mm->roots[i]->tmp_link, &dfs); - - end = start + size - 1; - - do { - u64 block_start; - u64 block_end; - - block = list_first_entry_or_null(&dfs, - struct i915_buddy_block, - tmp_link); - if (!block) - break; - - list_del(&block->tmp_link); - - block_start = i915_buddy_block_offset(block); - block_end = block_start + i915_buddy_block_size(mm, block) - 1; - - if (!overlaps(start, end, block_start, block_end)) - continue; - - if (i915_buddy_block_is_allocated(block)) { - err = -ENOSPC; - goto err_free; - } - - if (contains(start, end, block_start, block_end)) { - if (!i915_buddy_block_is_free(block)) { - err = -ENOSPC; - goto err_free; - } - - mark_allocated(block); - mm->avail -= i915_buddy_block_size(mm, block); - list_add_tail(&block->link, &allocated); - continue; - } - - if (!i915_buddy_block_is_split(block)) { - err = split_block(mm, block); - if (unlikely(err)) - goto err_undo; - } - - list_add(&block->right->tmp_link, &dfs); - list_add(&block->left->tmp_link, &dfs); - } while (1); - - list_splice_tail(&allocated, blocks); - return 0; - -err_undo: - /* - * We really don't want to leave around a bunch of split blocks, since - * bigger is better, so make sure we merge everything back before we - * free the allocated blocks. - */ - buddy = get_buddy(block); - if (buddy && - (i915_buddy_block_is_free(block) && - i915_buddy_block_is_free(buddy))) - __i915_buddy_free(mm, block); - -err_free: - i915_buddy_free_list(mm, &allocated); - return err; -} - -void i915_buddy_block_print(struct i915_buddy_mm *mm, - struct i915_buddy_block *block, - struct drm_printer *p) -{ - u64 start = i915_buddy_block_offset(block); - u64 size = i915_buddy_block_size(mm, block); - - drm_printf(p, "%#018llx-%#018llx: %llu\n", start, start + size, size); -} - -void i915_buddy_print(struct i915_buddy_mm *mm, struct drm_printer *p) -{ - int order; - - drm_printf(p, "chunk_size: %lluKiB, total: %lluMiB, free: %lluMiB\n", - mm->chunk_size >> 10, mm->size >> 20, mm->avail >> 20); - - for (order = mm->max_order; order >= 0; order--) { - struct i915_buddy_block *block; - u64 count = 0, free; - - list_for_each_entry(block, &mm->free_list[order], link) { - GEM_BUG_ON(!i915_buddy_block_is_free(block)); - count++; - } - - drm_printf(p, "order-%d ", order); - - free = count * (mm->chunk_size << order); - if (free < SZ_1M) - drm_printf(p, "free: %lluKiB", free >> 10); - else - drm_printf(p, "free: %lluMiB", free >> 20); - - drm_printf(p, ", pages: %llu\n", count); - } -} - -#if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) -#include "selftests/i915_buddy.c" -#endif - -void i915_buddy_module_exit(void) -{ - kmem_cache_destroy(slab_blocks); -} - -int __init i915_buddy_module_init(void) -{ - slab_blocks = KMEM_CACHE(i915_buddy_block, 0); - if (!slab_blocks) - return -ENOMEM; - - return 0; -} diff --git a/drivers/gpu/drm/i915/i915_buddy.h b/drivers/gpu/drm/i915/i915_buddy.h deleted file mode 100644 index 7077742112ac..000000000000 --- a/drivers/gpu/drm/i915/i915_buddy.h +++ /dev/null @@ -1,143 +0,0 @@ -/* SPDX-License-Identifier: MIT */ -/* - * Copyright © 2021 Intel Corporation - */ - -#ifndef __I915_BUDDY_H__ -#define __I915_BUDDY_H__ - -#include <linux/bitops.h> -#include <linux/list.h> -#include <linux/slab.h> - -#include <drm/drm_print.h> - -struct i915_buddy_block { -#define I915_BUDDY_HEADER_OFFSET GENMASK_ULL(63, 12) -#define I915_BUDDY_HEADER_STATE GENMASK_ULL(11, 10) -#define I915_BUDDY_ALLOCATED (1 << 10) -#define I915_BUDDY_FREE (2 << 10) -#define I915_BUDDY_SPLIT (3 << 10) -/* Free to be used, if needed in the future */ -#define I915_BUDDY_HEADER_UNUSED GENMASK_ULL(9, 6) -#define I915_BUDDY_HEADER_ORDER GENMASK_ULL(5, 0) - u64 header; - - struct i915_buddy_block *left; - struct i915_buddy_block *right; - struct i915_buddy_block *parent; - - void *private; /* owned by creator */ - - /* - * While the block is allocated by the user through i915_buddy_alloc*, - * the user has ownership of the link, for example to maintain within - * a list, if so desired. As soon as the block is freed with - * i915_buddy_free* ownership is given back to the mm. - */ - struct list_head link; - struct list_head tmp_link; -}; - -/* Order-zero must be at least PAGE_SIZE */ -#define I915_BUDDY_MAX_ORDER (63 - PAGE_SHIFT) - -/* - * Binary Buddy System. - * - * Locking should be handled by the user, a simple mutex around - * i915_buddy_alloc* and i915_buddy_free* should suffice. - */ -struct i915_buddy_mm { - /* Maintain a free list for each order. */ - struct list_head *free_list; - - /* - * Maintain explicit binary tree(s) to track the allocation of the - * address space. This gives us a simple way of finding a buddy block - * and performing the potentially recursive merge step when freeing a - * block. Nodes are either allocated or free, in which case they will - * also exist on the respective free list. - */ - struct i915_buddy_block **roots; - - /* - * Anything from here is public, and remains static for the lifetime of - * the mm. Everything above is considered do-not-touch. - */ - unsigned int n_roots; - unsigned int max_order; - - /* Must be at least PAGE_SIZE */ - u64 chunk_size; - u64 size; - u64 avail; -}; - -static inline u64 -i915_buddy_block_offset(struct i915_buddy_block *block) -{ - return block->header & I915_BUDDY_HEADER_OFFSET; -} - -static inline unsigned int -i915_buddy_block_order(struct i915_buddy_block *block) -{ - return block->header & I915_BUDDY_HEADER_ORDER; -} - -static inline unsigned int -i915_buddy_block_state(struct i915_buddy_block *block) -{ - return block->header & I915_BUDDY_HEADER_STATE; -} - -static inline bool -i915_buddy_block_is_allocated(struct i915_buddy_block *block) -{ - return i915_buddy_block_state(block) == I915_BUDDY_ALLOCATED; -} - -static inline bool -i915_buddy_block_is_free(struct i915_buddy_block *block) -{ - return i915_buddy_block_state(block) == I915_BUDDY_FREE; -} - -static inline bool -i915_buddy_block_is_split(struct i915_buddy_block *block) -{ - return i915_buddy_block_state(block) == I915_BUDDY_SPLIT; -} - -static inline u64 -i915_buddy_block_size(struct i915_buddy_mm *mm, - struct i915_buddy_block *block) -{ - return mm->chunk_size << i915_buddy_block_order(block); -} - -int i915_buddy_init(struct i915_buddy_mm *mm, u64 size, u64 chunk_size); - -void i915_buddy_fini(struct i915_buddy_mm *mm); - -struct i915_buddy_block * -i915_buddy_alloc(struct i915_buddy_mm *mm, unsigned int order); - -int i915_buddy_alloc_range(struct i915_buddy_mm *mm, - struct list_head *blocks, - u64 start, u64 size); - -void i915_buddy_free(struct i915_buddy_mm *mm, struct i915_buddy_block *block); - -void i915_buddy_free_list(struct i915_buddy_mm *mm, struct list_head *objects); - -void i915_buddy_print(struct i915_buddy_mm *mm, struct drm_printer *p); -void i915_buddy_block_print(struct i915_buddy_mm *mm, - struct i915_buddy_block *block, - struct drm_printer *p); - -void i915_buddy_module_exit(void); -int i915_buddy_module_init(void); - -#endif diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c index e0403ce9ce69..aea4c30645ff 100644 --- a/drivers/gpu/drm/i915/i915_cmd_parser.c +++ b/drivers/gpu/drm/i915/i915_cmd_parser.c @@ -26,10 +26,14 @@ */ #include "gt/intel_engine.h" +#include "gt/intel_engine_regs.h" #include "gt/intel_gpu_commands.h" +#include "gt/intel_gt_regs.h" +#include "i915_cmd_parser.h" #include "i915_drv.h" #include "i915_memcpy.h" +#include "i915_reg.h" /** * DOC: batch buffer command parser @@ -591,6 +595,10 @@ struct drm_i915_reg_descriptor { { .addr = _reg(idx) }, \ { .addr = _reg ## _UDW(idx) } +#define REG64_BASE_IDX(_reg, base, idx) \ + { .addr = _reg(base, idx) }, \ + { .addr = _reg ## _UDW(base, idx) } + static const struct drm_i915_reg_descriptor gen7_render_regs[] = { REG64(GPGPU_THREADS_DISPATCHED), REG64(HS_INVOCATION_COUNT), @@ -605,8 +613,8 @@ static const struct drm_i915_reg_descriptor gen7_render_regs[] = { REG64(PS_INVOCATION_COUNT), REG64(PS_DEPTH_COUNT), REG64_IDX(RING_TIMESTAMP, RENDER_RING_BASE), - REG64(MI_PREDICATE_SRC0), - REG64(MI_PREDICATE_SRC1), + REG64_IDX(MI_PREDICATE_SRC0, RENDER_RING_BASE), + REG64_IDX(MI_PREDICATE_SRC1, RENDER_RING_BASE), REG32(GEN7_3DPRIM_END_OFFSET), REG32(GEN7_3DPRIM_START_VERTEX), REG32(GEN7_3DPRIM_VERTEX_COUNT), @@ -636,22 +644,22 @@ static const struct drm_i915_reg_descriptor gen7_render_regs[] = { }; static const struct drm_i915_reg_descriptor hsw_render_regs[] = { - REG64_IDX(HSW_CS_GPR, 0), - REG64_IDX(HSW_CS_GPR, 1), - REG64_IDX(HSW_CS_GPR, 2), - REG64_IDX(HSW_CS_GPR, 3), - REG64_IDX(HSW_CS_GPR, 4), - REG64_IDX(HSW_CS_GPR, 5), - REG64_IDX(HSW_CS_GPR, 6), - REG64_IDX(HSW_CS_GPR, 7), - REG64_IDX(HSW_CS_GPR, 8), - REG64_IDX(HSW_CS_GPR, 9), - REG64_IDX(HSW_CS_GPR, 10), - REG64_IDX(HSW_CS_GPR, 11), - REG64_IDX(HSW_CS_GPR, 12), - REG64_IDX(HSW_CS_GPR, 13), - REG64_IDX(HSW_CS_GPR, 14), - REG64_IDX(HSW_CS_GPR, 15), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 0), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 1), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 2), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 3), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 4), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 5), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 6), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 7), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 8), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 9), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 10), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 11), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 12), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 13), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 14), + REG64_BASE_IDX(GEN8_RING_CS_GPR, RENDER_RING_BASE, 15), REG32(HSW_SCRATCH1, .mask = ~HSW_SCRATCH1_L3_DATA_ATOMICS_DISABLE, .value = 0), @@ -674,22 +682,22 @@ static const struct drm_i915_reg_descriptor gen9_blt_regs[] = { REG32(BCS_SWCTRL), REG64_IDX(RING_TIMESTAMP, BLT_RING_BASE), REG32_IDX(RING_CTX_TIMESTAMP, BLT_RING_BASE), - REG64_IDX(BCS_GPR, 0), - REG64_IDX(BCS_GPR, 1), - REG64_IDX(BCS_GPR, 2), - REG64_IDX(BCS_GPR, 3), - REG64_IDX(BCS_GPR, 4), - REG64_IDX(BCS_GPR, 5), - REG64_IDX(BCS_GPR, 6), - REG64_IDX(BCS_GPR, 7), - REG64_IDX(BCS_GPR, 8), - REG64_IDX(BCS_GPR, 9), - REG64_IDX(BCS_GPR, 10), - REG64_IDX(BCS_GPR, 11), - REG64_IDX(BCS_GPR, 12), - REG64_IDX(BCS_GPR, 13), - REG64_IDX(BCS_GPR, 14), - REG64_IDX(BCS_GPR, 15), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 0), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 1), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 2), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 3), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 4), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 5), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 6), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 7), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 8), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 9), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 10), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 11), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 12), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 13), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 14), + REG64_BASE_IDX(GEN8_RING_CS_GPR, BLT_RING_BASE, 15), }; #undef REG64 diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.h b/drivers/gpu/drm/i915/i915_cmd_parser.h new file mode 100644 index 000000000000..ba70ac6c97cd --- /dev/null +++ b/drivers/gpu/drm/i915/i915_cmd_parser.h @@ -0,0 +1,26 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2021 Intel Corporation + */ + +#ifndef __I915_CMD_PARSER_H__ +#define __I915_CMD_PARSER_H__ + +#include <linux/types.h> + +struct drm_i915_private; +struct intel_engine_cs; +struct i915_vma; + +int i915_cmd_parser_get_version(struct drm_i915_private *dev_priv); +int intel_engine_init_cmd_parser(struct intel_engine_cs *engine); +void intel_engine_cleanup_cmd_parser(struct intel_engine_cs *engine); +int intel_engine_cmd_parser(struct intel_engine_cs *engine, + struct i915_vma *batch, + unsigned long batch_offset, + unsigned long batch_length, + struct i915_vma *shadow, + bool trampoline); +#define I915_CMD_PARSER_TRAMPOLINE_SIZE 8 + +#endif /* __I915_CMD_PARSER_H__ */ diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index e0e052cdf8b8..4fe1ce1433fe 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -38,6 +38,7 @@ #include "gt/intel_gt_debugfs.h" #include "gt/intel_gt_pm.h" #include "gt/intel_gt_pm_debugfs.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_gt_requests.h" #include "gt/intel_rc6.h" #include "gt/intel_reset.h" diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c index 95174938b160..4b56d4d83e75 100644 --- a/drivers/gpu/drm/i915/i915_driver.c +++ b/drivers/gpu/drm/i915/i915_driver.c @@ -74,7 +74,9 @@ #include "i915_debugfs.h" #include "i915_driver.h" #include "i915_drv.h" +#include "i915_getparam.h" #include "i915_ioc32.h" +#include "i915_ioctl.h" #include "i915_irq.h" #include "i915_memcpy.h" #include "i915_perf.h" @@ -86,6 +88,7 @@ #include "intel_dram.h" #include "intel_gvt.h" #include "intel_memory_region.h" +#include "intel_pci_config.h" #include "intel_pcode.h" #include "intel_pm.h" #include "intel_region_ttm.h" diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 0c70ab08fc0c..f7bdb973c880 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -61,15 +61,16 @@ #include <drm/ttm/ttm_device.h> #include "i915_params.h" -#include "i915_reg.h" #include "i915_utils.h" #include "display/intel_bios.h" +#include "display/intel_cdclk.h" #include "display/intel_display.h" #include "display/intel_display_power.h" #include "display/intel_dmc.h" #include "display/intel_dpll_mgr.h" #include "display/intel_dsb.h" +#include "display/intel_fbc.h" #include "display/intel_frontbuffer.h" #include "display/intel_global_state.h" #include "display/intel_gmbus.h" @@ -104,7 +105,6 @@ #include "i915_scheduler.h" #include "gt/intel_timeline.h" #include "i915_vma.h" -#include "i915_irq.h" /* General customization: @@ -290,6 +290,7 @@ struct intel_connector; struct intel_encoder; struct intel_atomic_state; struct intel_cdclk_config; +struct intel_cdclk_funcs; struct intel_cdclk_state; struct intel_cdclk_vals; struct intel_initial_plane_config; @@ -338,17 +339,6 @@ struct intel_color_funcs { void (*read_luts)(struct intel_crtc_state *crtc_state); }; -struct intel_cdclk_funcs { - void (*get_cdclk)(struct drm_i915_private *dev_priv, - struct intel_cdclk_config *cdclk_config); - void (*set_cdclk)(struct drm_i915_private *dev_priv, - const struct intel_cdclk_config *cdclk_config, - enum pipe pipe); - int (*bw_calc_min_cdclk)(struct intel_atomic_state *state); - int (*modeset_calc_cdclk)(struct intel_cdclk_state *state); - u8 (*calc_voltage_level)(int cdclk); -}; - struct intel_hotplug_funcs { void (*hpd_irq_setup)(struct drm_i915_private *dev_priv); }; @@ -523,13 +513,6 @@ i915_fence_timeout(const struct drm_i915_private *i915) /* Amount of PSF GV points, BSpec precisely defines this */ #define I915_NUM_PSF_GV_POINTS 3 -enum psr_lines_to_wait { - PSR_0_LINES_TO_WAIT = 0, - PSR_1_LINE_TO_WAIT, - PSR_4_LINES_TO_WAIT, - PSR_8_LINES_TO_WAIT -}; - struct intel_vbt_data { /* bdb version */ u16 version; @@ -569,7 +552,6 @@ struct intel_vbt_data { bool full_link; bool require_aux_wakeup; int idle_frames; - enum psr_lines_to_wait lines_to_wait; int tp1_wakeup_time_us; int tp2_tp3_wakeup_time_us; int psr2_tp2_tp3_wakeup_time_us; @@ -625,11 +607,6 @@ struct i915_virtual_gpu { u32 caps; }; -struct intel_cdclk_config { - unsigned int cdclk, vco, ref, bypass; - u8 voltage_level; -}; - struct i915_selftest_stash { atomic_t counter; struct ida mock_region_instances; @@ -749,7 +726,7 @@ struct drm_i915_private { u32 pipestat_irq_mask[I915_MAX_PIPES]; struct i915_hotplug hotplug; - struct intel_fbc *fbc; + struct intel_fbc *fbc[I915_MAX_FBCS]; struct i915_drrs drrs; struct intel_opregion opregion; struct intel_vbt_data vbt; @@ -1288,6 +1265,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, IS_SUBPLATFORM(dev_priv, INTEL_DG2, INTEL_SUBPLATFORM_G11) #define IS_ADLS_RPLS(dev_priv) \ IS_SUBPLATFORM(dev_priv, INTEL_ALDERLAKE_S, INTEL_SUBPLATFORM_RPL_S) +#define IS_ADLP_N(dev_priv) \ + IS_SUBPLATFORM(dev_priv, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_N) #define IS_HSW_EARLY_SDV(dev_priv) (IS_HASWELL(dev_priv) && \ (INTEL_DEVID(dev_priv) & 0xFF00) == 0x0C00) #define IS_BDW_ULT(dev_priv) \ @@ -1485,8 +1464,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define NEEDS_WaRsDisableCoarsePowerGating(dev_priv) \ (IS_SKL_GT3(dev_priv) || IS_SKL_GT4(dev_priv)) -#define HAS_GMBUS_IRQ(dev_priv) (GRAPHICS_VER(dev_priv) >= 4) -#define HAS_GMBUS_BURST_READ(dev_priv) (GRAPHICS_VER(dev_priv) >= 11 || \ +#define HAS_GMBUS_IRQ(dev_priv) (DISPLAY_VER(dev_priv) >= 4) +#define HAS_GMBUS_BURST_READ(dev_priv) (DISPLAY_VER(dev_priv) >= 11 || \ IS_GEMINILAKE(dev_priv) || \ IS_KABYLAKE(dev_priv)) @@ -1498,9 +1477,9 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define SUPPORTS_TV(dev_priv) (INTEL_INFO(dev_priv)->display.supports_tv) #define I915_HAS_HOTPLUG(dev_priv) (INTEL_INFO(dev_priv)->display.has_hotplug) -#define HAS_FW_BLC(dev_priv) (GRAPHICS_VER(dev_priv) > 2) -#define HAS_FBC(dev_priv) (INTEL_INFO(dev_priv)->display.has_fbc) -#define HAS_CUR_FBC(dev_priv) (!HAS_GMCH(dev_priv) && GRAPHICS_VER(dev_priv) >= 7) +#define HAS_FW_BLC(dev_priv) (DISPLAY_VER(dev_priv) > 2) +#define HAS_FBC(dev_priv) (INTEL_INFO(dev_priv)->display.fbc_mask != 0) +#define HAS_CUR_FBC(dev_priv) (!HAS_GMCH(dev_priv) && DISPLAY_VER(dev_priv) >= 7) #define HAS_IPS(dev_priv) (IS_HSW_ULT(dev_priv) || IS_BROADWELL(dev_priv)) @@ -1513,7 +1492,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_PSR(dev_priv) (INTEL_INFO(dev_priv)->display.has_psr) #define HAS_PSR_HW_TRACKING(dev_priv) \ (INTEL_INFO(dev_priv)->display.has_psr_hw_tracking) -#define HAS_PSR2_SEL_FETCH(dev_priv) (GRAPHICS_VER(dev_priv) >= 12) +#define HAS_PSR2_SEL_FETCH(dev_priv) (DISPLAY_VER(dev_priv) >= 12) #define HAS_TRANSCODER(dev_priv, trans) ((INTEL_INFO(dev_priv)->display.cpu_transcoder_mask & BIT(trans)) != 0) #define HAS_RC6(dev_priv) (INTEL_INFO(dev_priv)->has_rc6) @@ -1524,7 +1503,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_DMC(dev_priv) (INTEL_INFO(dev_priv)->display.has_dmc) -#define HAS_MSO(i915) (GRAPHICS_VER(i915) >= 12) +#define HAS_MSO(i915) (DISPLAY_VER(i915) >= 12) #define HAS_RUNTIME_PM(dev_priv) (INTEL_INFO(dev_priv)->has_runtime_pm) #define HAS_64BIT_RELOC(dev_priv) (INTEL_INFO(dev_priv)->has_64bit_reloc) @@ -1557,7 +1536,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_GMCH(dev_priv) (INTEL_INFO(dev_priv)->display.has_gmch) -#define HAS_LSPCON(dev_priv) (IS_GRAPHICS_VER(dev_priv, 9, 10)) +#define HAS_LSPCON(dev_priv) (IS_DISPLAY_VER(dev_priv, 9, 10)) /* DPF == dynamic parity feature */ #define HAS_L3_DPF(dev_priv) (INTEL_INFO(dev_priv)->has_l3_dpf) @@ -1571,7 +1550,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_DISPLAY(dev_priv) (INTEL_INFO(dev_priv)->display.pipe_mask != 0) -#define HAS_VRR(i915) (GRAPHICS_VER(i915) >= 11) +#define HAS_VRR(i915) (DISPLAY_VER(i915) >= 11) #define HAS_ASYNC_FLIPS(i915) (DISPLAY_VER(i915) >= 5) @@ -1601,7 +1580,7 @@ i915_print_iommu_status(struct drm_i915_private *i915, struct drm_printer *p); static inline bool intel_scanout_needs_vtd_wa(struct drm_i915_private *dev_priv) { - return GRAPHICS_VER(dev_priv) >= 6 && intel_vtd_active(dev_priv); + return DISPLAY_VER(dev_priv) >= 6 && intel_vtd_active(dev_priv); } static inline bool @@ -1616,13 +1595,7 @@ intel_vm_no_concurrent_access_wa(struct drm_i915_private *i915) return IS_CHERRYVIEW(i915) || intel_ggtt_update_needs_vtd_wa(i915); } -/* i915_getparam.c */ -int i915_getparam_ioctl(struct drm_device *dev, void *data, - struct drm_file *file_priv); - /* i915_gem.c */ -int i915_gem_init_userptr(struct drm_i915_private *dev_priv); -void i915_gem_cleanup_userptr(struct drm_i915_private *dev_priv); void i915_gem_init_early(struct drm_i915_private *dev_priv); void i915_gem_cleanup_early(struct drm_i915_private *dev_priv); @@ -1734,17 +1707,6 @@ i915_gem_vm_lookup(struct drm_i915_file_private *file_priv, u32 id) return vm; } -/* i915_gem_evict.c */ -int __must_check i915_gem_evict_something(struct i915_address_space *vm, - u64 min_size, u64 alignment, - unsigned long color, - u64 start, u64 end, - unsigned flags); -int __must_check i915_gem_evict_for_node(struct i915_address_space *vm, - struct drm_mm_node *node, - unsigned int flags); -int i915_gem_evict_vm(struct i915_address_space *vm); - /* i915_gem_internal.c */ struct drm_i915_gem_object * i915_gem_object_create_internal(struct drm_i915_private *dev_priv, @@ -1763,25 +1725,8 @@ static inline bool i915_gem_object_needs_bit17_swizzle(struct drm_i915_gem_objec i915_gem_object_is_tiled(obj); } -u32 i915_gem_fence_size(struct drm_i915_private *dev_priv, u32 size, - unsigned int tiling, unsigned int stride); -u32 i915_gem_fence_alignment(struct drm_i915_private *dev_priv, u32 size, - unsigned int tiling, unsigned int stride); - const char *i915_cache_level_str(struct drm_i915_private *i915, int type); -/* i915_cmd_parser.c */ -int i915_cmd_parser_get_version(struct drm_i915_private *dev_priv); -int intel_engine_init_cmd_parser(struct intel_engine_cs *engine); -void intel_engine_cleanup_cmd_parser(struct intel_engine_cs *engine); -int intel_engine_cmd_parser(struct intel_engine_cs *engine, - struct i915_vma *batch, - unsigned long batch_offset, - unsigned long batch_length, - struct i915_vma *shadow, - bool trampoline); -#define I915_CMD_PARSER_TRAMPOLINE_SIZE 8 - /* intel_device_info.c */ static inline struct intel_device_info * mkwrite_device_info(struct drm_i915_private *dev_priv) @@ -1789,9 +1734,6 @@ mkwrite_device_info(struct drm_i915_private *dev_priv) return (struct intel_device_info *)INTEL_INFO(dev_priv); } -int i915_reg_read_ioctl(struct drm_device *dev, void *data, - struct drm_file *file); - static inline int intel_hws_csb_write_index(struct drm_i915_private *i915) { if (GRAPHICS_VER(i915) >= 11) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 915bf431f320..5ef959a9f594 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -45,6 +45,7 @@ #include "gem/i915_gem_ioctls.h" #include "gem/i915_gem_mman.h" #include "gem/i915_gem_region.h" +#include "gem/i915_gem_userptr.h" #include "gt/intel_engine_user.h" #include "gt/intel_gt.h" #include "gt/intel_gt_pm.h" diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c index 2b73ddb11c66..24eee0c2055f 100644 --- a/drivers/gpu/drm/i915/i915_gem_evict.c +++ b/drivers/gpu/drm/i915/i915_gem_evict.c @@ -31,6 +31,7 @@ #include "gt/intel_gt_requests.h" #include "i915_drv.h" +#include "i915_gem_evict.h" #include "i915_trace.h" I915_SELFTEST_DECLARE(static struct igt_evict_ctl { diff --git a/drivers/gpu/drm/i915/i915_gem_evict.h b/drivers/gpu/drm/i915/i915_gem_evict.h new file mode 100644 index 000000000000..d4478b6ad11b --- /dev/null +++ b/drivers/gpu/drm/i915/i915_gem_evict.h @@ -0,0 +1,24 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2021 Intel Corporation + */ + +#ifndef __I915_GEM_EVICT_H__ +#define __I915_GEM_EVICT_H__ + +#include <linux/types.h> + +struct drm_mm_node; +struct i915_address_space; + +int __must_check i915_gem_evict_something(struct i915_address_space *vm, + u64 min_size, u64 alignment, + unsigned long color, + u64 start, u64 end, + unsigned flags); +int __must_check i915_gem_evict_for_node(struct i915_address_space *vm, + struct drm_mm_node *node, + unsigned int flags); +int i915_gem_evict_vm(struct i915_address_space *vm); + +#endif /* __I915_GEM_EVICT_H__ */ diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index cd5f2348a187..8a7f0d92b56f 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -20,6 +20,7 @@ #include "gt/intel_gt_requests.h" #include "i915_drv.h" +#include "i915_gem_evict.h" #include "i915_scatterlist.h" #include "i915_trace.h" #include "i915_vgpu.h" diff --git a/drivers/gpu/drm/i915/i915_getparam.c b/drivers/gpu/drm/i915/i915_getparam.c index 7f80ad247bc8..dbe49fd87283 100644 --- a/drivers/gpu/drm/i915/i915_getparam.c +++ b/drivers/gpu/drm/i915/i915_getparam.c @@ -5,7 +5,9 @@ #include "gem/i915_gem_mman.h" #include "gt/intel_engine_user.h" +#include "i915_cmd_parser.h" #include "i915_drv.h" +#include "i915_getparam.h" #include "i915_perf.h" int i915_getparam_ioctl(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/i915/i915_getparam.h b/drivers/gpu/drm/i915/i915_getparam.h new file mode 100644 index 000000000000..18e4752e8f70 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_getparam.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2021 Intel Corporation + */ + +#ifndef __I915_GETPARAM_H__ +#define __I915_GETPARAM_H__ + +struct drm_device; +struct drm_file; + +int i915_getparam_ioctl(struct drm_device *dev, void *data, + struct drm_file *file_priv); + +#endif /* __I915_GETPARAM_H__ */ diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 0633888a411e..102c7bb747f8 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -41,8 +41,10 @@ #include "gem/i915_gem_context.h" #include "gem/i915_gem_lmem.h" +#include "gt/intel_engine_regs.h" #include "gt/intel_gt.h" #include "gt/intel_gt_pm.h" +#include "gt/intel_gt_regs.h" #include "i915_drv.h" #include "i915_gpu_error.h" diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c index 55b97c3a3dde..33348960f623 100644 --- a/drivers/gpu/drm/i915/i915_ioc32.c +++ b/drivers/gpu/drm/i915/i915_ioc32.c @@ -31,6 +31,7 @@ #include <drm/drm_ioctl.h> #include "i915_drv.h" +#include "i915_getparam.h" #include "i915_ioc32.h" struct drm_i915_getparam32 { diff --git a/drivers/gpu/drm/i915/i915_ioctl.c b/drivers/gpu/drm/i915/i915_ioctl.c new file mode 100644 index 000000000000..06a10ccea80b --- /dev/null +++ b/drivers/gpu/drm/i915/i915_ioctl.c @@ -0,0 +1,94 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2022 Intel Corporation + */ + +#include "gt/intel_engine_regs.h" + +#include "i915_drv.h" +#include "i915_gem.h" +#include "i915_ioctl.h" +#include "i915_reg.h" +#include "intel_runtime_pm.h" +#include "intel_uncore.h" + +/* + * This file is for small ioctl functions that are out of place everywhere else, + * and not big enough to warrant a file of their own. + * + * This is not the dumping ground for random ioctls. + */ + +struct reg_whitelist { + i915_reg_t offset_ldw; + i915_reg_t offset_udw; + u8 min_graphics_ver; + u8 max_graphics_ver; + u8 size; +}; + +static const struct reg_whitelist reg_read_whitelist[] = { + { + .offset_ldw = RING_TIMESTAMP(RENDER_RING_BASE), + .offset_udw = RING_TIMESTAMP_UDW(RENDER_RING_BASE), + .min_graphics_ver = 4, + .max_graphics_ver = 12, + .size = 8 + } +}; + +int i915_reg_read_ioctl(struct drm_device *dev, + void *data, struct drm_file *unused) +{ + struct drm_i915_private *i915 = to_i915(dev); + struct intel_uncore *uncore = &i915->uncore; + struct drm_i915_reg_read *reg = data; + struct reg_whitelist const *entry; + intel_wakeref_t wakeref; + unsigned int flags; + int remain; + int ret = 0; + + entry = reg_read_whitelist; + remain = ARRAY_SIZE(reg_read_whitelist); + while (remain) { + u32 entry_offset = i915_mmio_reg_offset(entry->offset_ldw); + + GEM_BUG_ON(!is_power_of_2(entry->size)); + GEM_BUG_ON(entry->size > 8); + GEM_BUG_ON(entry_offset & (entry->size - 1)); + + if (IS_GRAPHICS_VER(i915, entry->min_graphics_ver, entry->max_graphics_ver) && + entry_offset == (reg->offset & -entry->size)) + break; + entry++; + remain--; + } + + if (!remain) + return -EINVAL; + + flags = reg->offset & (entry->size - 1); + + with_intel_runtime_pm(&i915->runtime_pm, wakeref) { + if (entry->size == 8 && flags == I915_REG_READ_8B_WA) + reg->val = intel_uncore_read64_2x32(uncore, + entry->offset_ldw, + entry->offset_udw); + else if (entry->size == 8 && flags == 0) + reg->val = intel_uncore_read64(uncore, + entry->offset_ldw); + else if (entry->size == 4 && flags == 0) + reg->val = intel_uncore_read(uncore, entry->offset_ldw); + else if (entry->size == 2 && flags == 0) + reg->val = intel_uncore_read16(uncore, + entry->offset_ldw); + else if (entry->size == 1 && flags == 0) + reg->val = intel_uncore_read8(uncore, + entry->offset_ldw); + else + ret = -EINVAL; + } + + return ret; +} diff --git a/drivers/gpu/drm/i915/i915_ioctl.h b/drivers/gpu/drm/i915/i915_ioctl.h new file mode 100644 index 000000000000..f16ae87b8b8a --- /dev/null +++ b/drivers/gpu/drm/i915/i915_ioctl.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __I915_IOCTL_H__ +#define __I915_IOCTL_H__ + +struct drm_device; +struct drm_file; + +int i915_reg_read_ioctl(struct drm_device *dev, void *data, struct drm_file *file); + +#endif /* __I915_IOCTL_H__ */ diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 21f75b069fa8..c05eb09d8a66 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -46,6 +46,7 @@ #include "gt/intel_gt.h" #include "gt/intel_gt_irq.h" #include "gt/intel_gt_pm_irq.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_rps.h" #include "i915_drv.h" @@ -836,10 +837,7 @@ static int __intel_get_crtc_scanline(struct intel_crtc *crtc) if (mode->flags & DRM_MODE_FLAG_INTERLACE) vtotal /= 2; - if (DISPLAY_VER(dev_priv) == 2) - position = intel_de_read_fw(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN2; - else - position = intel_de_read_fw(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3; + position = intel_de_read_fw(dev_priv, PIPEDSL(pipe)) & PIPEDSL_LINE_MASK; /* * On HSW, the DSL reg (0x70000) appears to return 0 if we @@ -858,7 +856,7 @@ static int __intel_get_crtc_scanline(struct intel_crtc *crtc) for (i = 0; i < 100; i++) { udelay(1); - temp = intel_de_read_fw(dev_priv, PIPEDSL(pipe)) & DSL_LINEMASK_GEN3; + temp = intel_de_read_fw(dev_priv, PIPEDSL(pipe)) & PIPEDSL_LINE_MASK; if (temp != position) { position = temp; break; diff --git a/drivers/gpu/drm/i915/i915_mm.h b/drivers/gpu/drm/i915/i915_mm.h index 3ad22bbe80eb..04c8974d822b 100644 --- a/drivers/gpu/drm/i915/i915_mm.h +++ b/drivers/gpu/drm/i915/i915_mm.h @@ -23,8 +23,7 @@ int remap_io_mapping(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn, unsigned long size, struct io_mapping *iomap) { - pr_err("Architecture has no %s() and shouldn't be calling this function\n", __func__); - WARN_ON_ONCE(1); + WARN_ONCE(1, "Architecture has no drm_cache.c support\n"); return 0; } #endif diff --git a/drivers/gpu/drm/i915/i915_module.c b/drivers/gpu/drm/i915/i915_module.c index f6bcd2f89257..8451822637f0 100644 --- a/drivers/gpu/drm/i915/i915_module.c +++ b/drivers/gpu/drm/i915/i915_module.c @@ -9,7 +9,6 @@ #include "gem/i915_gem_context.h" #include "gem/i915_gem_object.h" #include "i915_active.h" -#include "i915_buddy.h" #include "i915_params.h" #include "i915_pci.h" #include "i915_perf.h" @@ -50,8 +49,6 @@ static const struct { { .init = i915_check_nomodeset }, { .init = i915_active_module_init, .exit = i915_active_module_exit }, - { .init = i915_buddy_module_init, - .exit = i915_buddy_module_exit }, { .init = i915_context_module_init, .exit = i915_context_module_exit }, { .init = i915_gem_context_module_init, diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 261294df535c..189d3bb8955a 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -28,6 +28,7 @@ #include "i915_driver.h" #include "i915_drv.h" #include "i915_pci.h" +#include "i915_reg.h" #define PLATFORM(x) .platform = (x) #define GEN(x) \ @@ -214,13 +215,13 @@ static const struct intel_device_info i845g_info = { static const struct intel_device_info i85x_info = { I830_FEATURES, PLATFORM(INTEL_I85X), - .display.has_fbc = 1, + .display.fbc_mask = BIT(INTEL_FBC_A), }; static const struct intel_device_info i865g_info = { I845_FEATURES, PLATFORM(INTEL_I865G), - .display.has_fbc = 1, + .display.fbc_mask = BIT(INTEL_FBC_A), }; #define GEN3_FEATURES \ @@ -258,7 +259,7 @@ static const struct intel_device_info i915gm_info = { .display.has_overlay = 1, .display.overlay_needs_physical = 1, .display.supports_tv = 1, - .display.has_fbc = 1, + .display.fbc_mask = BIT(INTEL_FBC_A), .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -283,7 +284,7 @@ static const struct intel_device_info i945gm_info = { .display.has_overlay = 1, .display.overlay_needs_physical = 1, .display.supports_tv = 1, - .display.has_fbc = 1, + .display.fbc_mask = BIT(INTEL_FBC_A), .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -342,7 +343,7 @@ static const struct intel_device_info i965gm_info = { GEN4_FEATURES, PLATFORM(INTEL_I965GM), .is_mobile = 1, - .display.has_fbc = 1, + .display.fbc_mask = BIT(INTEL_FBC_A), .display.has_overlay = 1, .display.supports_tv = 1, .hws_needs_physical = 1, @@ -360,7 +361,7 @@ static const struct intel_device_info gm45_info = { GEN4_FEATURES, PLATFORM(INTEL_GM45), .is_mobile = 1, - .display.has_fbc = 1, + .display.fbc_mask = BIT(INTEL_FBC_A), .display.supports_tv = 1, .platform_engine_mask = BIT(RCS0) | BIT(VCS0), .gpu_reset_clobbers_display = false, @@ -393,7 +394,7 @@ static const struct intel_device_info ilk_m_info = { PLATFORM(INTEL_IRONLAKE), .is_mobile = 1, .has_rps = true, - .display.has_fbc = 1, + .display.fbc_mask = BIT(INTEL_FBC_A), }; #define GEN6_FEATURES \ @@ -401,7 +402,7 @@ static const struct intel_device_info ilk_m_info = { .display.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ .display.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \ .display.has_hotplug = 1, \ - .display.has_fbc = 1, \ + .display.fbc_mask = BIT(INTEL_FBC_A), \ .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \ .has_coherent_ggtt = true, \ .has_llc = 1, \ @@ -452,7 +453,7 @@ static const struct intel_device_info snb_m_gt2_info = { .display.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \ .display.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), \ .display.has_hotplug = 1, \ - .display.has_fbc = 1, \ + .display.fbc_mask = BIT(INTEL_FBC_A), \ .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \ .has_coherent_ggtt = true, \ .has_llc = 1, \ @@ -693,7 +694,7 @@ static const struct intel_device_info skl_gt4_info = { .has_64bit_reloc = 1, \ .display.has_ddi = 1, \ .display.has_fpga_dbg = 1, \ - .display.has_fbc = 1, \ + .display.fbc_mask = BIT(INTEL_FBC_A), \ .display.has_hdcp = 1, \ .display.has_psr = 1, \ .display.has_psr_hw_tracking = 1, \ @@ -948,7 +949,7 @@ static const struct intel_device_info adl_s_info = { .display.has_dp_mst = 1, \ .display.has_dsb = 1, \ .display.has_dsc = 1, \ - .display.has_fbc = 1, \ + .display.fbc_mask = BIT(INTEL_FBC_A), \ .display.has_fpga_dbg = 1, \ .display.has_hdcp = 1, \ .display.has_hotplug = 1, \ @@ -1131,6 +1132,7 @@ static const struct pci_device_id pciidlist[] = { INTEL_RKL_IDS(&rkl_info), INTEL_ADLS_IDS(&adl_s_info), INTEL_ADLP_IDS(&adl_p_info), + INTEL_ADLN_IDS(&adl_p_info), INTEL_DG1_IDS(&dg1_info), INTEL_RPLS_IDS(&adl_s_info), {0, 0, 0} diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index e27f3b7cf094..590efdb9de8d 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -197,16 +197,19 @@ #include "gem/i915_gem_context.h" #include "gt/intel_engine_pm.h" +#include "gt/intel_engine_regs.h" #include "gt/intel_engine_user.h" #include "gt/intel_execlists_submission.h" #include "gt/intel_gpu_commands.h" #include "gt/intel_gt.h" #include "gt/intel_gt_clock_utils.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_lrc.h" #include "gt/intel_ring.h" #include "i915_drv.h" #include "i915_perf.h" +#include "i915_perf_oa_regs.h" /* HW requires this to be a power of two, between 128k and 16M, though driver * is currently generally designed assuming the largest 16M size is used such @@ -1682,7 +1685,7 @@ retry: stream, cs, true /* save */, CS_GPR(i), INTEL_GT_SCRATCH_FIELD_PERF_CS_GPR + 8 * i, 2); cs = save_restore_register( - stream, cs, true /* save */, MI_PREDICATE_RESULT_1, + stream, cs, true /* save */, MI_PREDICATE_RESULT_1(RENDER_RING_BASE), INTEL_GT_SCRATCH_FIELD_PERF_PREDICATE_RESULT_1, 1); /* First timestamp snapshot location. */ @@ -1736,7 +1739,7 @@ retry: */ *cs++ = MI_LOAD_REGISTER_REG | (3 - 2); *cs++ = i915_mmio_reg_offset(CS_GPR(JUMP_PREDICATE)); - *cs++ = i915_mmio_reg_offset(MI_PREDICATE_RESULT_1); + *cs++ = i915_mmio_reg_offset(MI_PREDICATE_RESULT_1(RENDER_RING_BASE)); /* Restart from the beginning if we had timestamps roll over. */ *cs++ = (GRAPHICS_VER(i915) < 8 ? @@ -1773,7 +1776,7 @@ retry: */ *cs++ = MI_LOAD_REGISTER_REG | (3 - 2); *cs++ = i915_mmio_reg_offset(CS_GPR(JUMP_PREDICATE)); - *cs++ = i915_mmio_reg_offset(MI_PREDICATE_RESULT_1); + *cs++ = i915_mmio_reg_offset(MI_PREDICATE_RESULT_1(RENDER_RING_BASE)); /* Predicate the jump. */ *cs++ = (GRAPHICS_VER(i915) < 8 ? @@ -1789,7 +1792,7 @@ retry: stream, cs, false /* restore */, CS_GPR(i), INTEL_GT_SCRATCH_FIELD_PERF_CS_GPR + 8 * i, 2); cs = save_restore_register( - stream, cs, false /* restore */, MI_PREDICATE_RESULT_1, + stream, cs, false /* restore */, MI_PREDICATE_RESULT_1(RENDER_RING_BASE), INTEL_GT_SCRATCH_FIELD_PERF_PREDICATE_RESULT_1, 1); /* And return to the ring. */ @@ -2418,7 +2421,7 @@ gen12_configure_all_contexts(struct i915_perf_stream *stream, { struct flex regs[] = { { - GEN8_R_PWR_CLK_STATE, + GEN8_R_PWR_CLK_STATE(RENDER_RING_BASE), CTX_R_PWR_CLK_STATE, }, }; @@ -2438,7 +2441,7 @@ lrc_configure_all_contexts(struct i915_perf_stream *stream, #define ctx_flexeuN(N) (ctx_flexeu0 + 2 * (N) + 1) struct flex regs[] = { { - GEN8_R_PWR_CLK_STATE, + GEN8_R_PWR_CLK_STATE(RENDER_RING_BASE), CTX_R_PWR_CLK_STATE, }, { @@ -3862,80 +3865,116 @@ static bool gen8_is_valid_flex_addr(struct i915_perf *perf, u32 addr) return false; } -#define ADDR_IN_RANGE(addr, start, end) \ - ((addr) >= (start) && \ - (addr) <= (end)) +static bool reg_in_range_table(u32 addr, const struct i915_range *table) +{ + while (table->start || table->end) { + if (addr >= table->start && addr <= table->end) + return true; -#define REG_IN_RANGE(addr, start, end) \ - ((addr) >= i915_mmio_reg_offset(start) && \ - (addr) <= i915_mmio_reg_offset(end)) + table++; + } + + return false; +} #define REG_EQUAL(addr, mmio) \ ((addr) == i915_mmio_reg_offset(mmio)) -static bool gen7_is_valid_b_counter_addr(struct i915_perf *perf, u32 addr) -{ - return REG_IN_RANGE(addr, OASTARTTRIG1, OASTARTTRIG8) || - REG_IN_RANGE(addr, OAREPORTTRIG1, OAREPORTTRIG8) || - REG_IN_RANGE(addr, OACEC0_0, OACEC7_1); -} +static const struct i915_range gen7_oa_b_counters[] = { + { .start = 0x2710, .end = 0x272c }, /* OASTARTTRIG[1-8] */ + { .start = 0x2740, .end = 0x275c }, /* OAREPORTTRIG[1-8] */ + { .start = 0x2770, .end = 0x27ac }, /* OACEC[0-7][0-1] */ + {} +}; + +static const struct i915_range gen12_oa_b_counters[] = { + { .start = 0x2b2c, .end = 0x2b2c }, /* GEN12_OAG_OA_PESS */ + { .start = 0xd900, .end = 0xd91c }, /* GEN12_OAG_OASTARTTRIG[1-8] */ + { .start = 0xd920, .end = 0xd93c }, /* GEN12_OAG_OAREPORTTRIG1[1-8] */ + { .start = 0xd940, .end = 0xd97c }, /* GEN12_OAG_CEC[0-7][0-1] */ + { .start = 0xdc00, .end = 0xdc3c }, /* GEN12_OAG_SCEC[0-7][0-1] */ + { .start = 0xdc40, .end = 0xdc40 }, /* GEN12_OAG_SPCTR_CNF */ + { .start = 0xdc44, .end = 0xdc44 }, /* GEN12_OAA_DBG_REG */ + {} +}; + +static const struct i915_range gen7_oa_mux_regs[] = { + { .start = 0x91b8, .end = 0x91cc }, /* OA_PERFCNT[1-2], OA_PERFMATRIX */ + { .start = 0x9800, .end = 0x9888 }, /* MICRO_BP0_0 - NOA_WRITE */ + { .start = 0xe180, .end = 0xe180 }, /* HALF_SLICE_CHICKEN2 */ + {} +}; -static bool gen7_is_valid_mux_addr(struct i915_perf *perf, u32 addr) +static const struct i915_range hsw_oa_mux_regs[] = { + { .start = 0x09e80, .end = 0x09ea4 }, /* HSW_MBVID2_NOA[0-9] */ + { .start = 0x09ec0, .end = 0x09ec0 }, /* HSW_MBVID2_MISR0 */ + { .start = 0x25100, .end = 0x2ff90 }, + {} +}; + +static const struct i915_range chv_oa_mux_regs[] = { + { .start = 0x182300, .end = 0x1823a4 }, + {} +}; + +static const struct i915_range gen8_oa_mux_regs[] = { + { .start = 0x0d00, .end = 0x0d2c }, /* RPM_CONFIG[0-1], NOA_CONFIG[0-8] */ + { .start = 0x20cc, .end = 0x20cc }, /* WAIT_FOR_RC6_EXIT */ + {} +}; + +static const struct i915_range gen11_oa_mux_regs[] = { + { .start = 0x91c8, .end = 0x91dc }, /* OA_PERFCNT[3-4] */ + {} +}; + +static const struct i915_range gen12_oa_mux_regs[] = { + { .start = 0x0d00, .end = 0x0d04 }, /* RPM_CONFIG[0-1] */ + { .start = 0x0d0c, .end = 0x0d2c }, /* NOA_CONFIG[0-8] */ + { .start = 0x9840, .end = 0x9840 }, /* GDT_CHICKEN_BITS */ + { .start = 0x9884, .end = 0x9888 }, /* NOA_WRITE */ + { .start = 0x20cc, .end = 0x20cc }, /* WAIT_FOR_RC6_EXIT */ + {} +}; + +static bool gen7_is_valid_b_counter_addr(struct i915_perf *perf, u32 addr) { - return REG_EQUAL(addr, HALF_SLICE_CHICKEN2) || - REG_IN_RANGE(addr, MICRO_BP0_0, NOA_WRITE) || - REG_IN_RANGE(addr, OA_PERFCNT1_LO, OA_PERFCNT2_HI) || - REG_IN_RANGE(addr, OA_PERFMATRIX_LO, OA_PERFMATRIX_HI); + return reg_in_range_table(addr, gen7_oa_b_counters); } static bool gen8_is_valid_mux_addr(struct i915_perf *perf, u32 addr) { - return gen7_is_valid_mux_addr(perf, addr) || - REG_EQUAL(addr, WAIT_FOR_RC6_EXIT) || - REG_IN_RANGE(addr, RPM_CONFIG0, NOA_CONFIG(8)); + return reg_in_range_table(addr, gen7_oa_mux_regs) || + reg_in_range_table(addr, gen8_oa_mux_regs); } static bool gen11_is_valid_mux_addr(struct i915_perf *perf, u32 addr) { - return gen8_is_valid_mux_addr(perf, addr) || - REG_EQUAL(addr, GEN10_NOA_WRITE_HIGH) || - REG_IN_RANGE(addr, OA_PERFCNT3_LO, OA_PERFCNT4_HI); + return reg_in_range_table(addr, gen7_oa_mux_regs) || + reg_in_range_table(addr, gen8_oa_mux_regs) || + reg_in_range_table(addr, gen11_oa_mux_regs); } static bool hsw_is_valid_mux_addr(struct i915_perf *perf, u32 addr) { - return gen7_is_valid_mux_addr(perf, addr) || - ADDR_IN_RANGE(addr, 0x25100, 0x2FF90) || - REG_IN_RANGE(addr, HSW_MBVID2_NOA0, HSW_MBVID2_NOA9) || - REG_EQUAL(addr, HSW_MBVID2_MISR0); + return reg_in_range_table(addr, gen7_oa_mux_regs) || + reg_in_range_table(addr, hsw_oa_mux_regs); } static bool chv_is_valid_mux_addr(struct i915_perf *perf, u32 addr) { - return gen7_is_valid_mux_addr(perf, addr) || - ADDR_IN_RANGE(addr, 0x182300, 0x1823A4); + return reg_in_range_table(addr, gen7_oa_mux_regs) || + reg_in_range_table(addr, chv_oa_mux_regs); } static bool gen12_is_valid_b_counter_addr(struct i915_perf *perf, u32 addr) { - return REG_IN_RANGE(addr, GEN12_OAG_OASTARTTRIG1, GEN12_OAG_OASTARTTRIG8) || - REG_IN_RANGE(addr, GEN12_OAG_OAREPORTTRIG1, GEN12_OAG_OAREPORTTRIG8) || - REG_IN_RANGE(addr, GEN12_OAG_CEC0_0, GEN12_OAG_CEC7_1) || - REG_IN_RANGE(addr, GEN12_OAG_SCEC0_0, GEN12_OAG_SCEC7_1) || - REG_EQUAL(addr, GEN12_OAA_DBG_REG) || - REG_EQUAL(addr, GEN12_OAG_OA_PESS) || - REG_EQUAL(addr, GEN12_OAG_SPCTR_CNF); + return reg_in_range_table(addr, gen12_oa_b_counters); } static bool gen12_is_valid_mux_addr(struct i915_perf *perf, u32 addr) { - return REG_EQUAL(addr, NOA_WRITE) || - REG_EQUAL(addr, GEN10_NOA_WRITE_HIGH) || - REG_EQUAL(addr, GDT_CHICKEN_BITS) || - REG_EQUAL(addr, WAIT_FOR_RC6_EXIT) || - REG_EQUAL(addr, RPM_CONFIG0) || - REG_EQUAL(addr, RPM_CONFIG1) || - REG_IN_RANGE(addr, NOA_CONFIG(0), NOA_CONFIG(8)); + return reg_in_range_table(addr, gen12_oa_mux_regs); } static u32 mask_reg_value(u32 reg, u32 val) diff --git a/drivers/gpu/drm/i915/i915_perf_oa_regs.h b/drivers/gpu/drm/i915/i915_perf_oa_regs.h new file mode 100644 index 000000000000..f31c9f13a9fc --- /dev/null +++ b/drivers/gpu/drm/i915/i915_perf_oa_regs.h @@ -0,0 +1,137 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __INTEL_PERF_OA_REGS__ +#define __INTEL_PERF_OA_REGS__ + +#include "i915_reg_defs.h" + +#define GEN7_OACONTROL _MMIO(0x2360) +#define GEN7_OACONTROL_CTX_MASK 0xFFFFF000 +#define GEN7_OACONTROL_TIMER_PERIOD_MASK 0x3F +#define GEN7_OACONTROL_TIMER_PERIOD_SHIFT 6 +#define GEN7_OACONTROL_TIMER_ENABLE (1 << 5) +#define GEN7_OACONTROL_FORMAT_A13 (0 << 2) +#define GEN7_OACONTROL_FORMAT_A29 (1 << 2) +#define GEN7_OACONTROL_FORMAT_A13_B8_C8 (2 << 2) +#define GEN7_OACONTROL_FORMAT_A29_B8_C8 (3 << 2) +#define GEN7_OACONTROL_FORMAT_B4_C8 (4 << 2) +#define GEN7_OACONTROL_FORMAT_A45_B8_C8 (5 << 2) +#define GEN7_OACONTROL_FORMAT_B4_C8_A16 (6 << 2) +#define GEN7_OACONTROL_FORMAT_C4_B8 (7 << 2) +#define GEN7_OACONTROL_FORMAT_SHIFT 2 +#define GEN7_OACONTROL_PER_CTX_ENABLE (1 << 1) +#define GEN7_OACONTROL_ENABLE (1 << 0) + +#define GEN8_OACTXID _MMIO(0x2364) + +#define GEN8_OA_DEBUG _MMIO(0x2B04) +#define GEN9_OA_DEBUG_DISABLE_CLK_RATIO_REPORTS (1 << 5) +#define GEN9_OA_DEBUG_INCLUDE_CLK_RATIO (1 << 6) +#define GEN9_OA_DEBUG_DISABLE_GO_1_0_REPORTS (1 << 2) +#define GEN9_OA_DEBUG_DISABLE_CTX_SWITCH_REPORTS (1 << 1) + +#define GEN8_OACONTROL _MMIO(0x2B00) +#define GEN8_OA_REPORT_FORMAT_A12 (0 << 2) +#define GEN8_OA_REPORT_FORMAT_A12_B8_C8 (2 << 2) +#define GEN8_OA_REPORT_FORMAT_A36_B8_C8 (5 << 2) +#define GEN8_OA_REPORT_FORMAT_C4_B8 (7 << 2) +#define GEN8_OA_REPORT_FORMAT_SHIFT 2 +#define GEN8_OA_SPECIFIC_CONTEXT_ENABLE (1 << 1) +#define GEN8_OA_COUNTER_ENABLE (1 << 0) + +#define GEN8_OACTXCONTROL _MMIO(0x2360) +#define GEN8_OA_TIMER_PERIOD_MASK 0x3F +#define GEN8_OA_TIMER_PERIOD_SHIFT 2 +#define GEN8_OA_TIMER_ENABLE (1 << 1) +#define GEN8_OA_COUNTER_RESUME (1 << 0) + +#define GEN7_OABUFFER _MMIO(0x23B0) /* R/W */ +#define GEN7_OABUFFER_OVERRUN_DISABLE (1 << 3) +#define GEN7_OABUFFER_EDGE_TRIGGER (1 << 2) +#define GEN7_OABUFFER_STOP_RESUME_ENABLE (1 << 1) +#define GEN7_OABUFFER_RESUME (1 << 0) + +#define GEN8_OABUFFER_UDW _MMIO(0x23b4) +#define GEN8_OABUFFER _MMIO(0x2b14) +#define GEN8_OABUFFER_MEM_SELECT_GGTT (1 << 0) /* 0: PPGTT, 1: GGTT */ + +#define GEN7_OASTATUS1 _MMIO(0x2364) +#define GEN7_OASTATUS1_TAIL_MASK 0xffffffc0 +#define GEN7_OASTATUS1_COUNTER_OVERFLOW (1 << 2) +#define GEN7_OASTATUS1_OABUFFER_OVERFLOW (1 << 1) +#define GEN7_OASTATUS1_REPORT_LOST (1 << 0) + +#define GEN7_OASTATUS2 _MMIO(0x2368) +#define GEN7_OASTATUS2_HEAD_MASK 0xffffffc0 +#define GEN7_OASTATUS2_MEM_SELECT_GGTT (1 << 0) /* 0: PPGTT, 1: GGTT */ + +#define GEN8_OASTATUS _MMIO(0x2b08) +#define GEN8_OASTATUS_TAIL_POINTER_WRAP (1 << 17) +#define GEN8_OASTATUS_HEAD_POINTER_WRAP (1 << 16) +#define GEN8_OASTATUS_OVERRUN_STATUS (1 << 3) +#define GEN8_OASTATUS_COUNTER_OVERFLOW (1 << 2) +#define GEN8_OASTATUS_OABUFFER_OVERFLOW (1 << 1) +#define GEN8_OASTATUS_REPORT_LOST (1 << 0) + +#define GEN8_OAHEADPTR _MMIO(0x2B0C) +#define GEN8_OAHEADPTR_MASK 0xffffffc0 +#define GEN8_OATAILPTR _MMIO(0x2B10) +#define GEN8_OATAILPTR_MASK 0xffffffc0 + +#define OABUFFER_SIZE_128K (0 << 3) +#define OABUFFER_SIZE_256K (1 << 3) +#define OABUFFER_SIZE_512K (2 << 3) +#define OABUFFER_SIZE_1M (3 << 3) +#define OABUFFER_SIZE_2M (4 << 3) +#define OABUFFER_SIZE_4M (5 << 3) +#define OABUFFER_SIZE_8M (6 << 3) +#define OABUFFER_SIZE_16M (7 << 3) + +#define GEN12_OA_TLB_INV_CR _MMIO(0xceec) + +/* Gen12 OAR unit */ +#define GEN12_OAR_OACONTROL _MMIO(0x2960) +#define GEN12_OAR_OACONTROL_COUNTER_FORMAT_SHIFT 1 +#define GEN12_OAR_OACONTROL_COUNTER_ENABLE (1 << 0) + +#define GEN12_OACTXCONTROL _MMIO(0x2360) +#define GEN12_OAR_OASTATUS _MMIO(0x2968) + +/* Gen12 OAG unit */ +#define GEN12_OAG_OAHEADPTR _MMIO(0xdb00) +#define GEN12_OAG_OAHEADPTR_MASK 0xffffffc0 +#define GEN12_OAG_OATAILPTR _MMIO(0xdb04) +#define GEN12_OAG_OATAILPTR_MASK 0xffffffc0 + +#define GEN12_OAG_OABUFFER _MMIO(0xdb08) +#define GEN12_OAG_OABUFFER_BUFFER_SIZE_MASK (0x7) +#define GEN12_OAG_OABUFFER_BUFFER_SIZE_SHIFT (3) +#define GEN12_OAG_OABUFFER_MEMORY_SELECT (1 << 0) /* 0: PPGTT, 1: GGTT */ + +#define GEN12_OAG_OAGLBCTXCTRL _MMIO(0x2b28) +#define GEN12_OAG_OAGLBCTXCTRL_TIMER_PERIOD_SHIFT 2 +#define GEN12_OAG_OAGLBCTXCTRL_TIMER_ENABLE (1 << 1) +#define GEN12_OAG_OAGLBCTXCTRL_COUNTER_RESUME (1 << 0) + +#define GEN12_OAG_OACONTROL _MMIO(0xdaf4) +#define GEN12_OAG_OACONTROL_OA_COUNTER_FORMAT_SHIFT 2 +#define GEN12_OAG_OACONTROL_OA_COUNTER_ENABLE (1 << 0) + +#define GEN12_OAG_OA_DEBUG _MMIO(0xdaf8) +#define GEN12_OAG_OA_DEBUG_INCLUDE_CLK_RATIO (1 << 6) +#define GEN12_OAG_OA_DEBUG_DISABLE_CLK_RATIO_REPORTS (1 << 5) +#define GEN12_OAG_OA_DEBUG_DISABLE_GO_1_0_REPORTS (1 << 2) +#define GEN12_OAG_OA_DEBUG_DISABLE_CTX_SWITCH_REPORTS (1 << 1) + +#define GEN12_OAG_OASTATUS _MMIO(0xdafc) +#define GEN12_OAG_OASTATUS_COUNTER_OVERFLOW (1 << 2) +#define GEN12_OAG_OASTATUS_BUFFER_OVERFLOW (1 << 1) +#define GEN12_OAG_OASTATUS_REPORT_LOST (1 << 0) + +#define GDT_CHICKEN_BITS _MMIO(0x9840) +#define GT_NOA_ENABLE 0x00000080 + +#endif /* __INTEL_PERF_OA_REGS__ */ diff --git a/drivers/gpu/drm/i915/i915_perf_types.h b/drivers/gpu/drm/i915/i915_perf_types.h index aa14354a5120..473a3c0544bb 100644 --- a/drivers/gpu/drm/i915/i915_perf_types.h +++ b/drivers/gpu/drm/i915/i915_perf_types.h @@ -18,7 +18,7 @@ #include <uapi/drm/i915_drm.h> #include "gt/intel_sseu.h" -#include "i915_reg.h" +#include "i915_reg_defs.h" #include "intel_wakeref.h" struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c index ea655161793e..cfc21042499d 100644 --- a/drivers/gpu/drm/i915/i915_pmu.c +++ b/drivers/gpu/drm/i915/i915_pmu.c @@ -8,8 +8,10 @@ #include "gt/intel_engine.h" #include "gt/intel_engine_pm.h" +#include "gt/intel_engine_regs.h" #include "gt/intel_engine_user.h" #include "gt/intel_gt_pm.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_rc6.h" #include "gt/intel_rps.h" diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index c2bb33febb68..87c92314ee26 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -25,8 +25,7 @@ #ifndef _I915_REG_H_ #define _I915_REG_H_ -#include <linux/bitfield.h> -#include <linux/bits.h> +#include "i915_reg_defs.h" /** * DOC: The i915 register macro definition style guide @@ -116,92 +115,6 @@ * #define GEN8_BAR _MMIO(0xb888) */ -/** - * REG_BIT() - Prepare a u32 bit value - * @__n: 0-based bit number - * - * Local wrapper for BIT() to force u32, with compile time checks. - * - * @return: Value with bit @__n set. - */ -#define REG_BIT(__n) \ - ((u32)(BIT(__n) + \ - BUILD_BUG_ON_ZERO(__is_constexpr(__n) && \ - ((__n) < 0 || (__n) > 31)))) - -/** - * REG_GENMASK() - Prepare a continuous u32 bitmask - * @__high: 0-based high bit - * @__low: 0-based low bit - * - * Local wrapper for GENMASK() to force u32, with compile time checks. - * - * @return: Continuous bitmask from @__high to @__low, inclusive. - */ -#define REG_GENMASK(__high, __low) \ - ((u32)(GENMASK(__high, __low) + \ - BUILD_BUG_ON_ZERO(__is_constexpr(__high) && \ - __is_constexpr(__low) && \ - ((__low) < 0 || (__high) > 31 || (__low) > (__high))))) - -/* - * Local integer constant expression version of is_power_of_2(). - */ -#define IS_POWER_OF_2(__x) ((__x) && (((__x) & ((__x) - 1)) == 0)) - -/** - * REG_FIELD_PREP() - Prepare a u32 bitfield value - * @__mask: shifted mask defining the field's length and position - * @__val: value to put in the field - * - * Local copy of FIELD_PREP() to generate an integer constant expression, force - * u32 and for consistency with REG_FIELD_GET(), REG_BIT() and REG_GENMASK(). - * - * @return: @__val masked and shifted into the field defined by @__mask. - */ -#define REG_FIELD_PREP(__mask, __val) \ - ((u32)((((typeof(__mask))(__val) << __bf_shf(__mask)) & (__mask)) + \ - BUILD_BUG_ON_ZERO(!__is_constexpr(__mask)) + \ - BUILD_BUG_ON_ZERO((__mask) == 0 || (__mask) > U32_MAX) + \ - BUILD_BUG_ON_ZERO(!IS_POWER_OF_2((__mask) + (1ULL << __bf_shf(__mask)))) + \ - BUILD_BUG_ON_ZERO(__builtin_choose_expr(__is_constexpr(__val), (~((__mask) >> __bf_shf(__mask)) & (__val)), 0)))) - -/** - * REG_FIELD_GET() - Extract a u32 bitfield value - * @__mask: shifted mask defining the field's length and position - * @__val: value to extract the bitfield value from - * - * Local wrapper for FIELD_GET() to force u32 and for consistency with - * REG_FIELD_PREP(), REG_BIT() and REG_GENMASK(). - * - * @return: Masked and shifted value of the field defined by @__mask in @__val. - */ -#define REG_FIELD_GET(__mask, __val) ((u32)FIELD_GET(__mask, __val)) - -typedef struct { - u32 reg; -} i915_reg_t; - -#define _MMIO(r) ((const i915_reg_t){ .reg = (r) }) - -#define INVALID_MMIO_REG _MMIO(0) - -static __always_inline u32 i915_mmio_reg_offset(i915_reg_t reg) -{ - return reg.reg; -} - -static inline bool i915_mmio_reg_equal(i915_reg_t a, i915_reg_t b) -{ - return i915_mmio_reg_offset(a) == i915_mmio_reg_offset(b); -} - -static inline bool i915_mmio_reg_valid(i915_reg_t reg) -{ - return !i915_mmio_reg_equal(reg, INVALID_MMIO_REG); -} - -#define VLV_DISPLAY_BASE 0x180000 #define VLV_MIPI_BASE VLV_DISPLAY_BASE #define BXT_MIPI_BASE 0x60000 @@ -275,247 +188,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define _MASKED_BIT_ENABLE(a) ({ typeof(a) _a = (a); _MASKED_FIELD(_a, _a); }) #define _MASKED_BIT_DISABLE(a) (_MASKED_FIELD((a), 0)) -/* PCI config space */ - -#define MCHBAR_I915 0x44 -#define MCHBAR_I965 0x48 -#define MCHBAR_SIZE (4 * 4096) - -#define DEVEN 0x54 -#define DEVEN_MCHBAR_EN (1 << 28) - -/* BSM in include/drm/i915_drm.h */ - -#define HPLLCC 0xc0 /* 85x only */ -#define GC_CLOCK_CONTROL_MASK (0x7 << 0) -#define GC_CLOCK_133_200 (0 << 0) -#define GC_CLOCK_100_200 (1 << 0) -#define GC_CLOCK_100_133 (2 << 0) -#define GC_CLOCK_133_266 (3 << 0) -#define GC_CLOCK_133_200_2 (4 << 0) -#define GC_CLOCK_133_266_2 (5 << 0) -#define GC_CLOCK_166_266 (6 << 0) -#define GC_CLOCK_166_250 (7 << 0) - -#define I915_GDRST 0xc0 /* PCI config register */ -#define GRDOM_FULL (0 << 2) -#define GRDOM_RENDER (1 << 2) -#define GRDOM_MEDIA (3 << 2) -#define GRDOM_MASK (3 << 2) -#define GRDOM_RESET_STATUS (1 << 1) -#define GRDOM_RESET_ENABLE (1 << 0) - -/* BSpec only has register offset, PCI device and bit found empirically */ -#define I830_CLOCK_GATE 0xc8 /* device 0 */ -#define I830_L2_CACHE_CLOCK_GATE_DISABLE (1 << 2) - -#define GCDGMBUS 0xcc - -#define GCFGC2 0xda -#define GCFGC 0xf0 /* 915+ only */ -#define GC_LOW_FREQUENCY_ENABLE (1 << 7) -#define GC_DISPLAY_CLOCK_190_200_MHZ (0 << 4) -#define GC_DISPLAY_CLOCK_333_320_MHZ (4 << 4) -#define GC_DISPLAY_CLOCK_267_MHZ_PNV (0 << 4) -#define GC_DISPLAY_CLOCK_333_MHZ_PNV (1 << 4) -#define GC_DISPLAY_CLOCK_444_MHZ_PNV (2 << 4) -#define GC_DISPLAY_CLOCK_200_MHZ_PNV (5 << 4) -#define GC_DISPLAY_CLOCK_133_MHZ_PNV (6 << 4) -#define GC_DISPLAY_CLOCK_167_MHZ_PNV (7 << 4) -#define GC_DISPLAY_CLOCK_MASK (7 << 4) -#define GM45_GC_RENDER_CLOCK_MASK (0xf << 0) -#define GM45_GC_RENDER_CLOCK_266_MHZ (8 << 0) -#define GM45_GC_RENDER_CLOCK_320_MHZ (9 << 0) -#define GM45_GC_RENDER_CLOCK_400_MHZ (0xb << 0) -#define GM45_GC_RENDER_CLOCK_533_MHZ (0xc << 0) -#define I965_GC_RENDER_CLOCK_MASK (0xf << 0) -#define I965_GC_RENDER_CLOCK_267_MHZ (2 << 0) -#define I965_GC_RENDER_CLOCK_333_MHZ (3 << 0) -#define I965_GC_RENDER_CLOCK_444_MHZ (4 << 0) -#define I965_GC_RENDER_CLOCK_533_MHZ (5 << 0) -#define I945_GC_RENDER_CLOCK_MASK (7 << 0) -#define I945_GC_RENDER_CLOCK_166_MHZ (0 << 0) -#define I945_GC_RENDER_CLOCK_200_MHZ (1 << 0) -#define I945_GC_RENDER_CLOCK_250_MHZ (3 << 0) -#define I945_GC_RENDER_CLOCK_400_MHZ (5 << 0) -#define I915_GC_RENDER_CLOCK_MASK (7 << 0) -#define I915_GC_RENDER_CLOCK_166_MHZ (0 << 0) -#define I915_GC_RENDER_CLOCK_200_MHZ (1 << 0) -#define I915_GC_RENDER_CLOCK_333_MHZ (4 << 0) - -#define ASLE 0xe4 -#define ASLS 0xfc - -#define SWSCI 0xe8 -#define SWSCI_SCISEL (1 << 15) -#define SWSCI_GSSCIE (1 << 0) - -#define LBPC 0xf4 /* legacy/combination backlight modes, also called LBB */ - - -#define ILK_GDSR _MMIO(MCHBAR_MIRROR_BASE + 0x2ca4) -#define ILK_GRDOM_FULL (0 << 1) -#define ILK_GRDOM_RENDER (1 << 1) -#define ILK_GRDOM_MEDIA (3 << 1) -#define ILK_GRDOM_MASK (3 << 1) -#define ILK_GRDOM_RESET_ENABLE (1 << 0) - -#define GEN6_MBCUNIT_SNPCR _MMIO(0x900c) /* for LLC config */ -#define GEN6_MBC_SNPCR_SHIFT 21 -#define GEN6_MBC_SNPCR_MASK (3 << 21) -#define GEN6_MBC_SNPCR_MAX (0 << 21) -#define GEN6_MBC_SNPCR_MED (1 << 21) -#define GEN6_MBC_SNPCR_LOW (2 << 21) -#define GEN6_MBC_SNPCR_MIN (3 << 21) /* only 1/16th of the cache is shared */ - -#define VLV_G3DCTL _MMIO(0x9024) -#define VLV_GSCKGCTL _MMIO(0x9028) - -#define FBC_LLC_READ_CTRL _MMIO(0x9044) -#define FBC_LLC_FULLY_OPEN REG_BIT(30) - -#define GEN6_MBCTL _MMIO(0x0907c) -#define GEN6_MBCTL_ENABLE_BOOT_FETCH (1 << 4) -#define GEN6_MBCTL_CTX_FETCH_NEEDED (1 << 3) -#define GEN6_MBCTL_BME_UPDATE_ENABLE (1 << 2) -#define GEN6_MBCTL_MAE_UPDATE_ENABLE (1 << 1) -#define GEN6_MBCTL_BOOT_FETCH_MECH (1 << 0) - -#define GEN6_GDRST _MMIO(0x941c) -#define GEN6_GRDOM_FULL (1 << 0) -#define GEN6_GRDOM_RENDER (1 << 1) -#define GEN6_GRDOM_MEDIA (1 << 2) -#define GEN6_GRDOM_BLT (1 << 3) -#define GEN6_GRDOM_VECS (1 << 4) -#define GEN9_GRDOM_GUC (1 << 5) -#define GEN8_GRDOM_MEDIA2 (1 << 7) -/* GEN11 changed all bit defs except for FULL & RENDER */ -#define GEN11_GRDOM_FULL GEN6_GRDOM_FULL -#define GEN11_GRDOM_RENDER GEN6_GRDOM_RENDER -#define GEN11_GRDOM_BLT (1 << 2) -#define GEN11_GRDOM_GUC (1 << 3) -#define GEN11_GRDOM_MEDIA (1 << 5) -#define GEN11_GRDOM_MEDIA2 (1 << 6) -#define GEN11_GRDOM_MEDIA3 (1 << 7) -#define GEN11_GRDOM_MEDIA4 (1 << 8) -#define GEN11_GRDOM_MEDIA5 (1 << 9) -#define GEN11_GRDOM_MEDIA6 (1 << 10) -#define GEN11_GRDOM_MEDIA7 (1 << 11) -#define GEN11_GRDOM_MEDIA8 (1 << 12) -#define GEN11_GRDOM_VECS (1 << 13) -#define GEN11_GRDOM_VECS2 (1 << 14) -#define GEN11_GRDOM_VECS3 (1 << 15) -#define GEN11_GRDOM_VECS4 (1 << 16) -#define GEN11_GRDOM_SFC0 (1 << 17) -#define GEN11_GRDOM_SFC1 (1 << 18) -#define GEN11_GRDOM_SFC2 (1 << 19) -#define GEN11_GRDOM_SFC3 (1 << 20) - -#define GEN11_VCS_SFC_RESET_BIT(instance) (GEN11_GRDOM_SFC0 << ((instance) >> 1)) -#define GEN11_VECS_SFC_RESET_BIT(instance) (GEN11_GRDOM_SFC0 << (instance)) - -#define GEN11_VCS_SFC_FORCED_LOCK(engine) _MMIO((engine)->mmio_base + 0x88C) -#define GEN11_VCS_SFC_FORCED_LOCK_BIT (1 << 0) -#define GEN11_VCS_SFC_LOCK_STATUS(engine) _MMIO((engine)->mmio_base + 0x890) -#define GEN11_VCS_SFC_USAGE_BIT (1 << 0) -#define GEN11_VCS_SFC_LOCK_ACK_BIT (1 << 1) - -#define GEN11_VECS_SFC_FORCED_LOCK(engine) _MMIO((engine)->mmio_base + 0x201C) -#define GEN11_VECS_SFC_FORCED_LOCK_BIT (1 << 0) -#define GEN11_VECS_SFC_LOCK_ACK(engine) _MMIO((engine)->mmio_base + 0x2018) -#define GEN11_VECS_SFC_LOCK_ACK_BIT (1 << 0) -#define GEN11_VECS_SFC_USAGE(engine) _MMIO((engine)->mmio_base + 0x2014) -#define GEN11_VECS_SFC_USAGE_BIT (1 << 0) - -#define GEN12_HCP_SFC_FORCED_LOCK(engine) _MMIO((engine)->mmio_base + 0x2910) -#define GEN12_HCP_SFC_FORCED_LOCK_BIT REG_BIT(0) -#define GEN12_HCP_SFC_LOCK_STATUS(engine) _MMIO((engine)->mmio_base + 0x2914) -#define GEN12_HCP_SFC_LOCK_ACK_BIT REG_BIT(1) -#define GEN12_HCP_SFC_USAGE_BIT REG_BIT(0) - -#define GEN12_SFC_DONE(n) _MMIO(0x1cc000 + (n) * 0x1000) -#define GEN12_SFC_DONE_MAX 4 - -#define RING_PP_DIR_BASE(base) _MMIO((base) + 0x228) -#define RING_PP_DIR_BASE_READ(base) _MMIO((base) + 0x518) -#define RING_PP_DIR_DCLV(base) _MMIO((base) + 0x220) -#define PP_DIR_DCLV_2G 0xffffffff - -#define GEN8_RING_PDP_UDW(base, n) _MMIO((base) + 0x270 + (n) * 8 + 4) -#define GEN8_RING_PDP_LDW(base, n) _MMIO((base) + 0x270 + (n) * 8) - -#define GEN8_R_PWR_CLK_STATE _MMIO(0x20C8) -#define GEN8_RPCS_ENABLE (1 << 31) -#define GEN8_RPCS_S_CNT_ENABLE (1 << 18) -#define GEN8_RPCS_S_CNT_SHIFT 15 -#define GEN8_RPCS_S_CNT_MASK (0x7 << GEN8_RPCS_S_CNT_SHIFT) -#define GEN11_RPCS_S_CNT_SHIFT 12 -#define GEN11_RPCS_S_CNT_MASK (0x3f << GEN11_RPCS_S_CNT_SHIFT) -#define GEN8_RPCS_SS_CNT_ENABLE (1 << 11) -#define GEN8_RPCS_SS_CNT_SHIFT 8 -#define GEN8_RPCS_SS_CNT_MASK (0x7 << GEN8_RPCS_SS_CNT_SHIFT) -#define GEN8_RPCS_EU_MAX_SHIFT 4 -#define GEN8_RPCS_EU_MAX_MASK (0xf << GEN8_RPCS_EU_MAX_SHIFT) -#define GEN8_RPCS_EU_MIN_SHIFT 0 -#define GEN8_RPCS_EU_MIN_MASK (0xf << GEN8_RPCS_EU_MIN_SHIFT) - -#define WAIT_FOR_RC6_EXIT _MMIO(0x20CC) -/* HSW only */ -#define HSW_SELECTIVE_READ_ADDRESSING_SHIFT 2 -#define HSW_SELECTIVE_READ_ADDRESSING_MASK (0x3 << HSW_SLECTIVE_READ_ADDRESSING_SHIFT) -#define HSW_SELECTIVE_WRITE_ADDRESS_SHIFT 4 -#define HSW_SELECTIVE_WRITE_ADDRESS_MASK (0x7 << HSW_SELECTIVE_WRITE_ADDRESS_SHIFT) -/* HSW+ */ -#define HSW_WAIT_FOR_RC6_EXIT_ENABLE (1 << 0) -#define HSW_RCS_CONTEXT_ENABLE (1 << 7) -#define HSW_RCS_INHIBIT (1 << 8) -/* Gen8 */ -#define GEN8_SELECTIVE_WRITE_ADDRESS_SHIFT 4 -#define GEN8_SELECTIVE_WRITE_ADDRESS_MASK (0x3 << GEN8_SELECTIVE_WRITE_ADDRESS_SHIFT) -#define GEN8_SELECTIVE_WRITE_ADDRESS_SHIFT 4 -#define GEN8_SELECTIVE_WRITE_ADDRESS_MASK (0x3 << GEN8_SELECTIVE_WRITE_ADDRESS_SHIFT) -#define GEN8_SELECTIVE_WRITE_ADDRESSING_ENABLE (1 << 6) -#define GEN8_SELECTIVE_READ_SUBSLICE_SELECT_SHIFT 9 -#define GEN8_SELECTIVE_READ_SUBSLICE_SELECT_MASK (0x3 << GEN8_SELECTIVE_READ_SUBSLICE_SELECT_SHIFT) -#define GEN8_SELECTIVE_READ_SLICE_SELECT_SHIFT 11 -#define GEN8_SELECTIVE_READ_SLICE_SELECT_MASK (0x3 << GEN8_SELECTIVE_READ_SLICE_SELECT_SHIFT) -#define GEN8_SELECTIVE_READ_ADDRESSING_ENABLE (1 << 13) - -#define GAM_ECOCHK _MMIO(0x4090) -#define BDW_DISABLE_HDC_INVALIDATION (1 << 25) -#define ECOCHK_SNB_BIT (1 << 10) -#define ECOCHK_DIS_TLB (1 << 8) -#define HSW_ECOCHK_ARB_PRIO_SOL (1 << 6) -#define ECOCHK_PPGTT_CACHE64B (0x3 << 3) -#define ECOCHK_PPGTT_CACHE4B (0x0 << 3) -#define ECOCHK_PPGTT_GFDT_IVB (0x1 << 4) -#define ECOCHK_PPGTT_LLC_IVB (0x1 << 3) -#define ECOCHK_PPGTT_UC_HSW (0x1 << 3) -#define ECOCHK_PPGTT_WT_HSW (0x2 << 3) -#define ECOCHK_PPGTT_WB_HSW (0x3 << 3) - -#define GEN8_RC6_CTX_INFO _MMIO(0x8504) - -#define GAC_ECO_BITS _MMIO(0x14090) -#define ECOBITS_SNB_BIT (1 << 13) -#define ECOBITS_PPGTT_CACHE64B (3 << 8) -#define ECOBITS_PPGTT_CACHE4B (0 << 8) - -#define GEN12_GAMCNTRL_CTRL _MMIO(0xcf54) -#define INVALIDATION_BROADCAST_MODE_DIS REG_BIT(12) -#define GLOBAL_INVALIDATION_MODE REG_BIT(2) - -#define GEN12_GAMSTLB_CTRL _MMIO(0xcf4c) -#define CONTROL_BLOCK_CLKGATE_DIS REG_BIT(12) -#define EGRESS_BLOCK_CLKGATE_DIS REG_BIT(11) -#define TAG_BLOCK_CLKGATE_DIS REG_BIT(7) - -#define GEN12_MERT_MOD_CTRL _MMIO(0xcf28) -#define FORCE_MISS_FTLB REG_BIT(3) - -#define GAB_CTL _MMIO(0x24000) -#define GAB_CTL_CONT_AFTER_PAGEFAULT (1 << 8) - #define GU_CNTL _MMIO(0x101010) #define LMEM_INIT REG_BIT(7) @@ -538,671 +210,7 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define GEN6_STOLEN_RESERVED_ENABLE (1 << 0) #define GEN11_STOLEN_RESERVED_ADDR_MASK (0xFFFFFFFFFFFULL << 20) -/* VGA stuff */ - -#define VGA_ST01_MDA 0x3ba -#define VGA_ST01_CGA 0x3da - #define _VGA_MSR_WRITE _MMIO(0x3c2) -#define VGA_MSR_WRITE 0x3c2 -#define VGA_MSR_READ 0x3cc -#define VGA_MSR_MEM_EN (1 << 1) -#define VGA_MSR_CGA_MODE (1 << 0) - -#define VGA_SR_INDEX 0x3c4 -#define SR01 1 -#define VGA_SR_DATA 0x3c5 - -#define VGA_AR_INDEX 0x3c0 -#define VGA_AR_VID_EN (1 << 5) -#define VGA_AR_DATA_WRITE 0x3c0 -#define VGA_AR_DATA_READ 0x3c1 - -#define VGA_GR_INDEX 0x3ce -#define VGA_GR_DATA 0x3cf -/* GR05 */ -#define VGA_GR_MEM_READ_MODE_SHIFT 3 -#define VGA_GR_MEM_READ_MODE_PLANE 1 -/* GR06 */ -#define VGA_GR_MEM_MODE_MASK 0xc -#define VGA_GR_MEM_MODE_SHIFT 2 -#define VGA_GR_MEM_A0000_AFFFF 0 -#define VGA_GR_MEM_A0000_BFFFF 1 -#define VGA_GR_MEM_B0000_B7FFF 2 -#define VGA_GR_MEM_B0000_BFFFF 3 - -#define VGA_DACMASK 0x3c6 -#define VGA_DACRX 0x3c7 -#define VGA_DACWX 0x3c8 -#define VGA_DACDATA 0x3c9 - -#define VGA_CR_INDEX_MDA 0x3b4 -#define VGA_CR_DATA_MDA 0x3b5 -#define VGA_CR_INDEX_CGA 0x3d4 -#define VGA_CR_DATA_CGA 0x3d5 - -#define MI_PREDICATE_SRC0 _MMIO(0x2400) -#define MI_PREDICATE_SRC0_UDW _MMIO(0x2400 + 4) -#define MI_PREDICATE_SRC1 _MMIO(0x2408) -#define MI_PREDICATE_SRC1_UDW _MMIO(0x2408 + 4) -#define MI_PREDICATE_DATA _MMIO(0x2410) -#define MI_PREDICATE_RESULT _MMIO(0x2418) -#define MI_PREDICATE_RESULT_1 _MMIO(0x241c) -#define MI_PREDICATE_RESULT_2 _MMIO(0x2214) -#define LOWER_SLICE_ENABLED (1 << 0) -#define LOWER_SLICE_DISABLED (0 << 0) - -/* - * Registers used only by the command parser - */ -#define BCS_SWCTRL _MMIO(0x22200) -#define BCS_SRC_Y REG_BIT(0) -#define BCS_DST_Y REG_BIT(1) - -/* There are 16 GPR registers */ -#define BCS_GPR(n) _MMIO(0x22600 + (n) * 8) -#define BCS_GPR_UDW(n) _MMIO(0x22600 + (n) * 8 + 4) - -#define GPGPU_THREADS_DISPATCHED _MMIO(0x2290) -#define GPGPU_THREADS_DISPATCHED_UDW _MMIO(0x2290 + 4) -#define HS_INVOCATION_COUNT _MMIO(0x2300) -#define HS_INVOCATION_COUNT_UDW _MMIO(0x2300 + 4) -#define DS_INVOCATION_COUNT _MMIO(0x2308) -#define DS_INVOCATION_COUNT_UDW _MMIO(0x2308 + 4) -#define IA_VERTICES_COUNT _MMIO(0x2310) -#define IA_VERTICES_COUNT_UDW _MMIO(0x2310 + 4) -#define IA_PRIMITIVES_COUNT _MMIO(0x2318) -#define IA_PRIMITIVES_COUNT_UDW _MMIO(0x2318 + 4) -#define VS_INVOCATION_COUNT _MMIO(0x2320) -#define VS_INVOCATION_COUNT_UDW _MMIO(0x2320 + 4) -#define GS_INVOCATION_COUNT _MMIO(0x2328) -#define GS_INVOCATION_COUNT_UDW _MMIO(0x2328 + 4) -#define GS_PRIMITIVES_COUNT _MMIO(0x2330) -#define GS_PRIMITIVES_COUNT_UDW _MMIO(0x2330 + 4) -#define CL_INVOCATION_COUNT _MMIO(0x2338) -#define CL_INVOCATION_COUNT_UDW _MMIO(0x2338 + 4) -#define CL_PRIMITIVES_COUNT _MMIO(0x2340) -#define CL_PRIMITIVES_COUNT_UDW _MMIO(0x2340 + 4) -#define PS_INVOCATION_COUNT _MMIO(0x2348) -#define PS_INVOCATION_COUNT_UDW _MMIO(0x2348 + 4) -#define PS_DEPTH_COUNT _MMIO(0x2350) -#define PS_DEPTH_COUNT_UDW _MMIO(0x2350 + 4) - -/* There are the 4 64-bit counter registers, one for each stream output */ -#define GEN7_SO_NUM_PRIMS_WRITTEN(n) _MMIO(0x5200 + (n) * 8) -#define GEN7_SO_NUM_PRIMS_WRITTEN_UDW(n) _MMIO(0x5200 + (n) * 8 + 4) - -#define GEN7_SO_PRIM_STORAGE_NEEDED(n) _MMIO(0x5240 + (n) * 8) -#define GEN7_SO_PRIM_STORAGE_NEEDED_UDW(n) _MMIO(0x5240 + (n) * 8 + 4) - -#define GEN7_3DPRIM_END_OFFSET _MMIO(0x2420) -#define GEN7_3DPRIM_START_VERTEX _MMIO(0x2430) -#define GEN7_3DPRIM_VERTEX_COUNT _MMIO(0x2434) -#define GEN7_3DPRIM_INSTANCE_COUNT _MMIO(0x2438) -#define GEN7_3DPRIM_START_INSTANCE _MMIO(0x243C) -#define GEN7_3DPRIM_BASE_VERTEX _MMIO(0x2440) - -#define GEN7_GPGPU_DISPATCHDIMX _MMIO(0x2500) -#define GEN7_GPGPU_DISPATCHDIMY _MMIO(0x2504) -#define GEN7_GPGPU_DISPATCHDIMZ _MMIO(0x2508) - -/* There are the 16 64-bit CS General Purpose Registers */ -#define HSW_CS_GPR(n) _MMIO(0x2600 + (n) * 8) -#define HSW_CS_GPR_UDW(n) _MMIO(0x2600 + (n) * 8 + 4) - -#define GEN7_OACONTROL _MMIO(0x2360) -#define GEN7_OACONTROL_CTX_MASK 0xFFFFF000 -#define GEN7_OACONTROL_TIMER_PERIOD_MASK 0x3F -#define GEN7_OACONTROL_TIMER_PERIOD_SHIFT 6 -#define GEN7_OACONTROL_TIMER_ENABLE (1 << 5) -#define GEN7_OACONTROL_FORMAT_A13 (0 << 2) -#define GEN7_OACONTROL_FORMAT_A29 (1 << 2) -#define GEN7_OACONTROL_FORMAT_A13_B8_C8 (2 << 2) -#define GEN7_OACONTROL_FORMAT_A29_B8_C8 (3 << 2) -#define GEN7_OACONTROL_FORMAT_B4_C8 (4 << 2) -#define GEN7_OACONTROL_FORMAT_A45_B8_C8 (5 << 2) -#define GEN7_OACONTROL_FORMAT_B4_C8_A16 (6 << 2) -#define GEN7_OACONTROL_FORMAT_C4_B8 (7 << 2) -#define GEN7_OACONTROL_FORMAT_SHIFT 2 -#define GEN7_OACONTROL_PER_CTX_ENABLE (1 << 1) -#define GEN7_OACONTROL_ENABLE (1 << 0) - -#define GEN8_OACTXID _MMIO(0x2364) - -#define GEN8_OA_DEBUG _MMIO(0x2B04) -#define GEN9_OA_DEBUG_DISABLE_CLK_RATIO_REPORTS (1 << 5) -#define GEN9_OA_DEBUG_INCLUDE_CLK_RATIO (1 << 6) -#define GEN9_OA_DEBUG_DISABLE_GO_1_0_REPORTS (1 << 2) -#define GEN9_OA_DEBUG_DISABLE_CTX_SWITCH_REPORTS (1 << 1) - -#define GEN8_OACONTROL _MMIO(0x2B00) -#define GEN8_OA_REPORT_FORMAT_A12 (0 << 2) -#define GEN8_OA_REPORT_FORMAT_A12_B8_C8 (2 << 2) -#define GEN8_OA_REPORT_FORMAT_A36_B8_C8 (5 << 2) -#define GEN8_OA_REPORT_FORMAT_C4_B8 (7 << 2) -#define GEN8_OA_REPORT_FORMAT_SHIFT 2 -#define GEN8_OA_SPECIFIC_CONTEXT_ENABLE (1 << 1) -#define GEN8_OA_COUNTER_ENABLE (1 << 0) - -#define GEN8_OACTXCONTROL _MMIO(0x2360) -#define GEN8_OA_TIMER_PERIOD_MASK 0x3F -#define GEN8_OA_TIMER_PERIOD_SHIFT 2 -#define GEN8_OA_TIMER_ENABLE (1 << 1) -#define GEN8_OA_COUNTER_RESUME (1 << 0) - -#define GEN7_OABUFFER _MMIO(0x23B0) /* R/W */ -#define GEN7_OABUFFER_OVERRUN_DISABLE (1 << 3) -#define GEN7_OABUFFER_EDGE_TRIGGER (1 << 2) -#define GEN7_OABUFFER_STOP_RESUME_ENABLE (1 << 1) -#define GEN7_OABUFFER_RESUME (1 << 0) - -#define GEN8_OABUFFER_UDW _MMIO(0x23b4) -#define GEN8_OABUFFER _MMIO(0x2b14) -#define GEN8_OABUFFER_MEM_SELECT_GGTT (1 << 0) /* 0: PPGTT, 1: GGTT */ - -#define GEN7_OASTATUS1 _MMIO(0x2364) -#define GEN7_OASTATUS1_TAIL_MASK 0xffffffc0 -#define GEN7_OASTATUS1_COUNTER_OVERFLOW (1 << 2) -#define GEN7_OASTATUS1_OABUFFER_OVERFLOW (1 << 1) -#define GEN7_OASTATUS1_REPORT_LOST (1 << 0) - -#define GEN7_OASTATUS2 _MMIO(0x2368) -#define GEN7_OASTATUS2_HEAD_MASK 0xffffffc0 -#define GEN7_OASTATUS2_MEM_SELECT_GGTT (1 << 0) /* 0: PPGTT, 1: GGTT */ - -#define GEN8_OASTATUS _MMIO(0x2b08) -#define GEN8_OASTATUS_TAIL_POINTER_WRAP (1 << 17) -#define GEN8_OASTATUS_HEAD_POINTER_WRAP (1 << 16) -#define GEN8_OASTATUS_OVERRUN_STATUS (1 << 3) -#define GEN8_OASTATUS_COUNTER_OVERFLOW (1 << 2) -#define GEN8_OASTATUS_OABUFFER_OVERFLOW (1 << 1) -#define GEN8_OASTATUS_REPORT_LOST (1 << 0) - -#define GEN8_OAHEADPTR _MMIO(0x2B0C) -#define GEN8_OAHEADPTR_MASK 0xffffffc0 -#define GEN8_OATAILPTR _MMIO(0x2B10) -#define GEN8_OATAILPTR_MASK 0xffffffc0 - -#define OABUFFER_SIZE_128K (0 << 3) -#define OABUFFER_SIZE_256K (1 << 3) -#define OABUFFER_SIZE_512K (2 << 3) -#define OABUFFER_SIZE_1M (3 << 3) -#define OABUFFER_SIZE_2M (4 << 3) -#define OABUFFER_SIZE_4M (5 << 3) -#define OABUFFER_SIZE_8M (6 << 3) -#define OABUFFER_SIZE_16M (7 << 3) - -#define GEN12_OA_TLB_INV_CR _MMIO(0xceec) - -#define GEN12_SQCM _MMIO(0x8724) -#define EN_32B_ACCESS REG_BIT(30) - -/* Gen12 OAR unit */ -#define GEN12_OAR_OACONTROL _MMIO(0x2960) -#define GEN12_OAR_OACONTROL_COUNTER_FORMAT_SHIFT 1 -#define GEN12_OAR_OACONTROL_COUNTER_ENABLE (1 << 0) - -#define GEN12_OACTXCONTROL _MMIO(0x2360) -#define GEN12_OAR_OASTATUS _MMIO(0x2968) - -/* Gen12 OAG unit */ -#define GEN12_OAG_OAHEADPTR _MMIO(0xdb00) -#define GEN12_OAG_OAHEADPTR_MASK 0xffffffc0 -#define GEN12_OAG_OATAILPTR _MMIO(0xdb04) -#define GEN12_OAG_OATAILPTR_MASK 0xffffffc0 - -#define GEN12_OAG_OABUFFER _MMIO(0xdb08) -#define GEN12_OAG_OABUFFER_BUFFER_SIZE_MASK (0x7) -#define GEN12_OAG_OABUFFER_BUFFER_SIZE_SHIFT (3) -#define GEN12_OAG_OABUFFER_MEMORY_SELECT (1 << 0) /* 0: PPGTT, 1: GGTT */ - -#define GEN12_OAG_OAGLBCTXCTRL _MMIO(0x2b28) -#define GEN12_OAG_OAGLBCTXCTRL_TIMER_PERIOD_SHIFT 2 -#define GEN12_OAG_OAGLBCTXCTRL_TIMER_ENABLE (1 << 1) -#define GEN12_OAG_OAGLBCTXCTRL_COUNTER_RESUME (1 << 0) - -#define GEN12_OAG_OACONTROL _MMIO(0xdaf4) -#define GEN12_OAG_OACONTROL_OA_COUNTER_FORMAT_SHIFT 2 -#define GEN12_OAG_OACONTROL_OA_COUNTER_ENABLE (1 << 0) - -#define GEN12_OAG_OA_DEBUG _MMIO(0xdaf8) -#define GEN12_OAG_OA_DEBUG_INCLUDE_CLK_RATIO (1 << 6) -#define GEN12_OAG_OA_DEBUG_DISABLE_CLK_RATIO_REPORTS (1 << 5) -#define GEN12_OAG_OA_DEBUG_DISABLE_GO_1_0_REPORTS (1 << 2) -#define GEN12_OAG_OA_DEBUG_DISABLE_CTX_SWITCH_REPORTS (1 << 1) - -#define GEN12_OAG_OASTATUS _MMIO(0xdafc) -#define GEN12_OAG_OASTATUS_COUNTER_OVERFLOW (1 << 2) -#define GEN12_OAG_OASTATUS_BUFFER_OVERFLOW (1 << 1) -#define GEN12_OAG_OASTATUS_REPORT_LOST (1 << 0) - -/* - * Flexible, Aggregate EU Counter Registers. - * Note: these aren't contiguous - */ -#define EU_PERF_CNTL0 _MMIO(0xe458) -#define EU_PERF_CNTL1 _MMIO(0xe558) -#define EU_PERF_CNTL2 _MMIO(0xe658) -#define EU_PERF_CNTL3 _MMIO(0xe758) -#define EU_PERF_CNTL4 _MMIO(0xe45c) -#define EU_PERF_CNTL5 _MMIO(0xe55c) -#define EU_PERF_CNTL6 _MMIO(0xe65c) - -#define RT_CTRL _MMIO(0xe530) -#define DIS_NULL_QUERY REG_BIT(10) - -/* - * OA Boolean state - */ - -#define OASTARTTRIG1 _MMIO(0x2710) -#define OASTARTTRIG1_THRESHOLD_COUNT_MASK_MBZ 0xffff0000 -#define OASTARTTRIG1_THRESHOLD_MASK 0xffff - -#define OASTARTTRIG2 _MMIO(0x2714) -#define OASTARTTRIG2_INVERT_A_0 (1 << 0) -#define OASTARTTRIG2_INVERT_A_1 (1 << 1) -#define OASTARTTRIG2_INVERT_A_2 (1 << 2) -#define OASTARTTRIG2_INVERT_A_3 (1 << 3) -#define OASTARTTRIG2_INVERT_A_4 (1 << 4) -#define OASTARTTRIG2_INVERT_A_5 (1 << 5) -#define OASTARTTRIG2_INVERT_A_6 (1 << 6) -#define OASTARTTRIG2_INVERT_A_7 (1 << 7) -#define OASTARTTRIG2_INVERT_A_8 (1 << 8) -#define OASTARTTRIG2_INVERT_A_9 (1 << 9) -#define OASTARTTRIG2_INVERT_A_10 (1 << 10) -#define OASTARTTRIG2_INVERT_A_11 (1 << 11) -#define OASTARTTRIG2_INVERT_A_12 (1 << 12) -#define OASTARTTRIG2_INVERT_A_13 (1 << 13) -#define OASTARTTRIG2_INVERT_A_14 (1 << 14) -#define OASTARTTRIG2_INVERT_A_15 (1 << 15) -#define OASTARTTRIG2_INVERT_B_0 (1 << 16) -#define OASTARTTRIG2_INVERT_B_1 (1 << 17) -#define OASTARTTRIG2_INVERT_B_2 (1 << 18) -#define OASTARTTRIG2_INVERT_B_3 (1 << 19) -#define OASTARTTRIG2_INVERT_C_0 (1 << 20) -#define OASTARTTRIG2_INVERT_C_1 (1 << 21) -#define OASTARTTRIG2_INVERT_D_0 (1 << 22) -#define OASTARTTRIG2_THRESHOLD_ENABLE (1 << 23) -#define OASTARTTRIG2_START_TRIG_FLAG_MBZ (1 << 24) -#define OASTARTTRIG2_EVENT_SELECT_0 (1 << 28) -#define OASTARTTRIG2_EVENT_SELECT_1 (1 << 29) -#define OASTARTTRIG2_EVENT_SELECT_2 (1 << 30) -#define OASTARTTRIG2_EVENT_SELECT_3 (1 << 31) - -#define OASTARTTRIG3 _MMIO(0x2718) -#define OASTARTTRIG3_NOA_SELECT_MASK 0xf -#define OASTARTTRIG3_NOA_SELECT_8_SHIFT 0 -#define OASTARTTRIG3_NOA_SELECT_9_SHIFT 4 -#define OASTARTTRIG3_NOA_SELECT_10_SHIFT 8 -#define OASTARTTRIG3_NOA_SELECT_11_SHIFT 12 -#define OASTARTTRIG3_NOA_SELECT_12_SHIFT 16 -#define OASTARTTRIG3_NOA_SELECT_13_SHIFT 20 -#define OASTARTTRIG3_NOA_SELECT_14_SHIFT 24 -#define OASTARTTRIG3_NOA_SELECT_15_SHIFT 28 - -#define OASTARTTRIG4 _MMIO(0x271c) -#define OASTARTTRIG4_NOA_SELECT_MASK 0xf -#define OASTARTTRIG4_NOA_SELECT_0_SHIFT 0 -#define OASTARTTRIG4_NOA_SELECT_1_SHIFT 4 -#define OASTARTTRIG4_NOA_SELECT_2_SHIFT 8 -#define OASTARTTRIG4_NOA_SELECT_3_SHIFT 12 -#define OASTARTTRIG4_NOA_SELECT_4_SHIFT 16 -#define OASTARTTRIG4_NOA_SELECT_5_SHIFT 20 -#define OASTARTTRIG4_NOA_SELECT_6_SHIFT 24 -#define OASTARTTRIG4_NOA_SELECT_7_SHIFT 28 - -#define OASTARTTRIG5 _MMIO(0x2720) -#define OASTARTTRIG5_THRESHOLD_COUNT_MASK_MBZ 0xffff0000 -#define OASTARTTRIG5_THRESHOLD_MASK 0xffff - -#define OASTARTTRIG6 _MMIO(0x2724) -#define OASTARTTRIG6_INVERT_A_0 (1 << 0) -#define OASTARTTRIG6_INVERT_A_1 (1 << 1) -#define OASTARTTRIG6_INVERT_A_2 (1 << 2) -#define OASTARTTRIG6_INVERT_A_3 (1 << 3) -#define OASTARTTRIG6_INVERT_A_4 (1 << 4) -#define OASTARTTRIG6_INVERT_A_5 (1 << 5) -#define OASTARTTRIG6_INVERT_A_6 (1 << 6) -#define OASTARTTRIG6_INVERT_A_7 (1 << 7) -#define OASTARTTRIG6_INVERT_A_8 (1 << 8) -#define OASTARTTRIG6_INVERT_A_9 (1 << 9) -#define OASTARTTRIG6_INVERT_A_10 (1 << 10) -#define OASTARTTRIG6_INVERT_A_11 (1 << 11) -#define OASTARTTRIG6_INVERT_A_12 (1 << 12) -#define OASTARTTRIG6_INVERT_A_13 (1 << 13) -#define OASTARTTRIG6_INVERT_A_14 (1 << 14) -#define OASTARTTRIG6_INVERT_A_15 (1 << 15) -#define OASTARTTRIG6_INVERT_B_0 (1 << 16) -#define OASTARTTRIG6_INVERT_B_1 (1 << 17) -#define OASTARTTRIG6_INVERT_B_2 (1 << 18) -#define OASTARTTRIG6_INVERT_B_3 (1 << 19) -#define OASTARTTRIG6_INVERT_C_0 (1 << 20) -#define OASTARTTRIG6_INVERT_C_1 (1 << 21) -#define OASTARTTRIG6_INVERT_D_0 (1 << 22) -#define OASTARTTRIG6_THRESHOLD_ENABLE (1 << 23) -#define OASTARTTRIG6_START_TRIG_FLAG_MBZ (1 << 24) -#define OASTARTTRIG6_EVENT_SELECT_4 (1 << 28) -#define OASTARTTRIG6_EVENT_SELECT_5 (1 << 29) -#define OASTARTTRIG6_EVENT_SELECT_6 (1 << 30) -#define OASTARTTRIG6_EVENT_SELECT_7 (1 << 31) - -#define OASTARTTRIG7 _MMIO(0x2728) -#define OASTARTTRIG7_NOA_SELECT_MASK 0xf -#define OASTARTTRIG7_NOA_SELECT_8_SHIFT 0 -#define OASTARTTRIG7_NOA_SELECT_9_SHIFT 4 -#define OASTARTTRIG7_NOA_SELECT_10_SHIFT 8 -#define OASTARTTRIG7_NOA_SELECT_11_SHIFT 12 -#define OASTARTTRIG7_NOA_SELECT_12_SHIFT 16 -#define OASTARTTRIG7_NOA_SELECT_13_SHIFT 20 -#define OASTARTTRIG7_NOA_SELECT_14_SHIFT 24 -#define OASTARTTRIG7_NOA_SELECT_15_SHIFT 28 - -#define OASTARTTRIG8 _MMIO(0x272c) -#define OASTARTTRIG8_NOA_SELECT_MASK 0xf -#define OASTARTTRIG8_NOA_SELECT_0_SHIFT 0 -#define OASTARTTRIG8_NOA_SELECT_1_SHIFT 4 -#define OASTARTTRIG8_NOA_SELECT_2_SHIFT 8 -#define OASTARTTRIG8_NOA_SELECT_3_SHIFT 12 -#define OASTARTTRIG8_NOA_SELECT_4_SHIFT 16 -#define OASTARTTRIG8_NOA_SELECT_5_SHIFT 20 -#define OASTARTTRIG8_NOA_SELECT_6_SHIFT 24 -#define OASTARTTRIG8_NOA_SELECT_7_SHIFT 28 - -#define OAREPORTTRIG1 _MMIO(0x2740) -#define OAREPORTTRIG1_THRESHOLD_MASK 0xffff -#define OAREPORTTRIG1_EDGE_LEVEL_TRIGGER_SELECT_MASK 0xffff0000 /* 0=level */ - -#define OAREPORTTRIG2 _MMIO(0x2744) -#define OAREPORTTRIG2_INVERT_A_0 (1 << 0) -#define OAREPORTTRIG2_INVERT_A_1 (1 << 1) -#define OAREPORTTRIG2_INVERT_A_2 (1 << 2) -#define OAREPORTTRIG2_INVERT_A_3 (1 << 3) -#define OAREPORTTRIG2_INVERT_A_4 (1 << 4) -#define OAREPORTTRIG2_INVERT_A_5 (1 << 5) -#define OAREPORTTRIG2_INVERT_A_6 (1 << 6) -#define OAREPORTTRIG2_INVERT_A_7 (1 << 7) -#define OAREPORTTRIG2_INVERT_A_8 (1 << 8) -#define OAREPORTTRIG2_INVERT_A_9 (1 << 9) -#define OAREPORTTRIG2_INVERT_A_10 (1 << 10) -#define OAREPORTTRIG2_INVERT_A_11 (1 << 11) -#define OAREPORTTRIG2_INVERT_A_12 (1 << 12) -#define OAREPORTTRIG2_INVERT_A_13 (1 << 13) -#define OAREPORTTRIG2_INVERT_A_14 (1 << 14) -#define OAREPORTTRIG2_INVERT_A_15 (1 << 15) -#define OAREPORTTRIG2_INVERT_B_0 (1 << 16) -#define OAREPORTTRIG2_INVERT_B_1 (1 << 17) -#define OAREPORTTRIG2_INVERT_B_2 (1 << 18) -#define OAREPORTTRIG2_INVERT_B_3 (1 << 19) -#define OAREPORTTRIG2_INVERT_C_0 (1 << 20) -#define OAREPORTTRIG2_INVERT_C_1 (1 << 21) -#define OAREPORTTRIG2_INVERT_D_0 (1 << 22) -#define OAREPORTTRIG2_THRESHOLD_ENABLE (1 << 23) -#define OAREPORTTRIG2_REPORT_TRIGGER_ENABLE (1 << 31) - -#define OAREPORTTRIG3 _MMIO(0x2748) -#define OAREPORTTRIG3_NOA_SELECT_MASK 0xf -#define OAREPORTTRIG3_NOA_SELECT_8_SHIFT 0 -#define OAREPORTTRIG3_NOA_SELECT_9_SHIFT 4 -#define OAREPORTTRIG3_NOA_SELECT_10_SHIFT 8 -#define OAREPORTTRIG3_NOA_SELECT_11_SHIFT 12 -#define OAREPORTTRIG3_NOA_SELECT_12_SHIFT 16 -#define OAREPORTTRIG3_NOA_SELECT_13_SHIFT 20 -#define OAREPORTTRIG3_NOA_SELECT_14_SHIFT 24 -#define OAREPORTTRIG3_NOA_SELECT_15_SHIFT 28 - -#define OAREPORTTRIG4 _MMIO(0x274c) -#define OAREPORTTRIG4_NOA_SELECT_MASK 0xf -#define OAREPORTTRIG4_NOA_SELECT_0_SHIFT 0 -#define OAREPORTTRIG4_NOA_SELECT_1_SHIFT 4 -#define OAREPORTTRIG4_NOA_SELECT_2_SHIFT 8 -#define OAREPORTTRIG4_NOA_SELECT_3_SHIFT 12 -#define OAREPORTTRIG4_NOA_SELECT_4_SHIFT 16 -#define OAREPORTTRIG4_NOA_SELECT_5_SHIFT 20 -#define OAREPORTTRIG4_NOA_SELECT_6_SHIFT 24 -#define OAREPORTTRIG4_NOA_SELECT_7_SHIFT 28 - -#define OAREPORTTRIG5 _MMIO(0x2750) -#define OAREPORTTRIG5_THRESHOLD_MASK 0xffff -#define OAREPORTTRIG5_EDGE_LEVEL_TRIGGER_SELECT_MASK 0xffff0000 /* 0=level */ - -#define OAREPORTTRIG6 _MMIO(0x2754) -#define OAREPORTTRIG6_INVERT_A_0 (1 << 0) -#define OAREPORTTRIG6_INVERT_A_1 (1 << 1) -#define OAREPORTTRIG6_INVERT_A_2 (1 << 2) -#define OAREPORTTRIG6_INVERT_A_3 (1 << 3) -#define OAREPORTTRIG6_INVERT_A_4 (1 << 4) -#define OAREPORTTRIG6_INVERT_A_5 (1 << 5) -#define OAREPORTTRIG6_INVERT_A_6 (1 << 6) -#define OAREPORTTRIG6_INVERT_A_7 (1 << 7) -#define OAREPORTTRIG6_INVERT_A_8 (1 << 8) -#define OAREPORTTRIG6_INVERT_A_9 (1 << 9) -#define OAREPORTTRIG6_INVERT_A_10 (1 << 10) -#define OAREPORTTRIG6_INVERT_A_11 (1 << 11) -#define OAREPORTTRIG6_INVERT_A_12 (1 << 12) -#define OAREPORTTRIG6_INVERT_A_13 (1 << 13) -#define OAREPORTTRIG6_INVERT_A_14 (1 << 14) -#define OAREPORTTRIG6_INVERT_A_15 (1 << 15) -#define OAREPORTTRIG6_INVERT_B_0 (1 << 16) -#define OAREPORTTRIG6_INVERT_B_1 (1 << 17) -#define OAREPORTTRIG6_INVERT_B_2 (1 << 18) -#define OAREPORTTRIG6_INVERT_B_3 (1 << 19) -#define OAREPORTTRIG6_INVERT_C_0 (1 << 20) -#define OAREPORTTRIG6_INVERT_C_1 (1 << 21) -#define OAREPORTTRIG6_INVERT_D_0 (1 << 22) -#define OAREPORTTRIG6_THRESHOLD_ENABLE (1 << 23) -#define OAREPORTTRIG6_REPORT_TRIGGER_ENABLE (1 << 31) - -#define OAREPORTTRIG7 _MMIO(0x2758) -#define OAREPORTTRIG7_NOA_SELECT_MASK 0xf -#define OAREPORTTRIG7_NOA_SELECT_8_SHIFT 0 -#define OAREPORTTRIG7_NOA_SELECT_9_SHIFT 4 -#define OAREPORTTRIG7_NOA_SELECT_10_SHIFT 8 -#define OAREPORTTRIG7_NOA_SELECT_11_SHIFT 12 -#define OAREPORTTRIG7_NOA_SELECT_12_SHIFT 16 -#define OAREPORTTRIG7_NOA_SELECT_13_SHIFT 20 -#define OAREPORTTRIG7_NOA_SELECT_14_SHIFT 24 -#define OAREPORTTRIG7_NOA_SELECT_15_SHIFT 28 - -#define OAREPORTTRIG8 _MMIO(0x275c) -#define OAREPORTTRIG8_NOA_SELECT_MASK 0xf -#define OAREPORTTRIG8_NOA_SELECT_0_SHIFT 0 -#define OAREPORTTRIG8_NOA_SELECT_1_SHIFT 4 -#define OAREPORTTRIG8_NOA_SELECT_2_SHIFT 8 -#define OAREPORTTRIG8_NOA_SELECT_3_SHIFT 12 -#define OAREPORTTRIG8_NOA_SELECT_4_SHIFT 16 -#define OAREPORTTRIG8_NOA_SELECT_5_SHIFT 20 -#define OAREPORTTRIG8_NOA_SELECT_6_SHIFT 24 -#define OAREPORTTRIG8_NOA_SELECT_7_SHIFT 28 - -/* Same layout as OASTARTTRIGX */ -#define GEN12_OAG_OASTARTTRIG1 _MMIO(0xd900) -#define GEN12_OAG_OASTARTTRIG2 _MMIO(0xd904) -#define GEN12_OAG_OASTARTTRIG3 _MMIO(0xd908) -#define GEN12_OAG_OASTARTTRIG4 _MMIO(0xd90c) -#define GEN12_OAG_OASTARTTRIG5 _MMIO(0xd910) -#define GEN12_OAG_OASTARTTRIG6 _MMIO(0xd914) -#define GEN12_OAG_OASTARTTRIG7 _MMIO(0xd918) -#define GEN12_OAG_OASTARTTRIG8 _MMIO(0xd91c) - -/* Same layout as OAREPORTTRIGX */ -#define GEN12_OAG_OAREPORTTRIG1 _MMIO(0xd920) -#define GEN12_OAG_OAREPORTTRIG2 _MMIO(0xd924) -#define GEN12_OAG_OAREPORTTRIG3 _MMIO(0xd928) -#define GEN12_OAG_OAREPORTTRIG4 _MMIO(0xd92c) -#define GEN12_OAG_OAREPORTTRIG5 _MMIO(0xd930) -#define GEN12_OAG_OAREPORTTRIG6 _MMIO(0xd934) -#define GEN12_OAG_OAREPORTTRIG7 _MMIO(0xd938) -#define GEN12_OAG_OAREPORTTRIG8 _MMIO(0xd93c) - -/* CECX_0 */ -#define OACEC_COMPARE_LESS_OR_EQUAL 6 -#define OACEC_COMPARE_NOT_EQUAL 5 -#define OACEC_COMPARE_LESS_THAN 4 -#define OACEC_COMPARE_GREATER_OR_EQUAL 3 -#define OACEC_COMPARE_EQUAL 2 -#define OACEC_COMPARE_GREATER_THAN 1 -#define OACEC_COMPARE_ANY_EQUAL 0 - -#define OACEC_COMPARE_VALUE_MASK 0xffff -#define OACEC_COMPARE_VALUE_SHIFT 3 - -#define OACEC_SELECT_NOA (0 << 19) -#define OACEC_SELECT_PREV (1 << 19) -#define OACEC_SELECT_BOOLEAN (2 << 19) - -/* 11-bit array 0: pass-through, 1: negated */ -#define GEN12_OASCEC_NEGATE_MASK 0x7ff -#define GEN12_OASCEC_NEGATE_SHIFT 21 - -/* CECX_1 */ -#define OACEC_MASK_MASK 0xffff -#define OACEC_CONSIDERATIONS_MASK 0xffff -#define OACEC_CONSIDERATIONS_SHIFT 16 - -#define OACEC0_0 _MMIO(0x2770) -#define OACEC0_1 _MMIO(0x2774) -#define OACEC1_0 _MMIO(0x2778) -#define OACEC1_1 _MMIO(0x277c) -#define OACEC2_0 _MMIO(0x2780) -#define OACEC2_1 _MMIO(0x2784) -#define OACEC3_0 _MMIO(0x2788) -#define OACEC3_1 _MMIO(0x278c) -#define OACEC4_0 _MMIO(0x2790) -#define OACEC4_1 _MMIO(0x2794) -#define OACEC5_0 _MMIO(0x2798) -#define OACEC5_1 _MMIO(0x279c) -#define OACEC6_0 _MMIO(0x27a0) -#define OACEC6_1 _MMIO(0x27a4) -#define OACEC7_0 _MMIO(0x27a8) -#define OACEC7_1 _MMIO(0x27ac) - -/* Same layout as CECX_Y */ -#define GEN12_OAG_CEC0_0 _MMIO(0xd940) -#define GEN12_OAG_CEC0_1 _MMIO(0xd944) -#define GEN12_OAG_CEC1_0 _MMIO(0xd948) -#define GEN12_OAG_CEC1_1 _MMIO(0xd94c) -#define GEN12_OAG_CEC2_0 _MMIO(0xd950) -#define GEN12_OAG_CEC2_1 _MMIO(0xd954) -#define GEN12_OAG_CEC3_0 _MMIO(0xd958) -#define GEN12_OAG_CEC3_1 _MMIO(0xd95c) -#define GEN12_OAG_CEC4_0 _MMIO(0xd960) -#define GEN12_OAG_CEC4_1 _MMIO(0xd964) -#define GEN12_OAG_CEC5_0 _MMIO(0xd968) -#define GEN12_OAG_CEC5_1 _MMIO(0xd96c) -#define GEN12_OAG_CEC6_0 _MMIO(0xd970) -#define GEN12_OAG_CEC6_1 _MMIO(0xd974) -#define GEN12_OAG_CEC7_0 _MMIO(0xd978) -#define GEN12_OAG_CEC7_1 _MMIO(0xd97c) - -/* Same layout as CECX_Y + negate 11-bit array */ -#define GEN12_OAG_SCEC0_0 _MMIO(0xdc00) -#define GEN12_OAG_SCEC0_1 _MMIO(0xdc04) -#define GEN12_OAG_SCEC1_0 _MMIO(0xdc08) -#define GEN12_OAG_SCEC1_1 _MMIO(0xdc0c) -#define GEN12_OAG_SCEC2_0 _MMIO(0xdc10) -#define GEN12_OAG_SCEC2_1 _MMIO(0xdc14) -#define GEN12_OAG_SCEC3_0 _MMIO(0xdc18) -#define GEN12_OAG_SCEC3_1 _MMIO(0xdc1c) -#define GEN12_OAG_SCEC4_0 _MMIO(0xdc20) -#define GEN12_OAG_SCEC4_1 _MMIO(0xdc24) -#define GEN12_OAG_SCEC5_0 _MMIO(0xdc28) -#define GEN12_OAG_SCEC5_1 _MMIO(0xdc2c) -#define GEN12_OAG_SCEC6_0 _MMIO(0xdc30) -#define GEN12_OAG_SCEC6_1 _MMIO(0xdc34) -#define GEN12_OAG_SCEC7_0 _MMIO(0xdc38) -#define GEN12_OAG_SCEC7_1 _MMIO(0xdc3c) - -/* OA perf counters */ -#define OA_PERFCNT1_LO _MMIO(0x91B8) -#define OA_PERFCNT1_HI _MMIO(0x91BC) -#define OA_PERFCNT2_LO _MMIO(0x91C0) -#define OA_PERFCNT2_HI _MMIO(0x91C4) -#define OA_PERFCNT3_LO _MMIO(0x91C8) -#define OA_PERFCNT3_HI _MMIO(0x91CC) -#define OA_PERFCNT4_LO _MMIO(0x91D8) -#define OA_PERFCNT4_HI _MMIO(0x91DC) - -#define OA_PERFMATRIX_LO _MMIO(0x91C8) -#define OA_PERFMATRIX_HI _MMIO(0x91CC) - -/* RPM unit config (Gen8+) */ -#define RPM_CONFIG0 _MMIO(0x0D00) -#define GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_SHIFT 3 -#define GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_MASK (1 << GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_SHIFT) -#define GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_19_2_MHZ 0 -#define GEN9_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_24_MHZ 1 -#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_SHIFT 3 -#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_MASK (0x7 << GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_SHIFT) -#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_24_MHZ 0 -#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_19_2_MHZ 1 -#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_38_4_MHZ 2 -#define GEN11_RPM_CONFIG0_CRYSTAL_CLOCK_FREQ_25_MHZ 3 -#define GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_SHIFT 1 -#define GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_MASK (0x3 << GEN10_RPM_CONFIG0_CTC_SHIFT_PARAMETER_SHIFT) - -#define RPM_CONFIG1 _MMIO(0x0D04) -#define GEN10_GT_NOA_ENABLE (1 << 9) - -/* GPM unit config (Gen9+) */ -#define CTC_MODE _MMIO(0xA26C) -#define CTC_SOURCE_PARAMETER_MASK 1 -#define CTC_SOURCE_CRYSTAL_CLOCK 0 -#define CTC_SOURCE_DIVIDE_LOGIC 1 -#define CTC_SHIFT_PARAMETER_SHIFT 1 -#define CTC_SHIFT_PARAMETER_MASK (0x3 << CTC_SHIFT_PARAMETER_SHIFT) - -/* RCP unit config (Gen8+) */ -#define RCP_CONFIG _MMIO(0x0D08) - -/* NOA (HSW) */ -#define HSW_MBVID2_NOA0 _MMIO(0x9E80) -#define HSW_MBVID2_NOA1 _MMIO(0x9E84) -#define HSW_MBVID2_NOA2 _MMIO(0x9E88) -#define HSW_MBVID2_NOA3 _MMIO(0x9E8C) -#define HSW_MBVID2_NOA4 _MMIO(0x9E90) -#define HSW_MBVID2_NOA5 _MMIO(0x9E94) -#define HSW_MBVID2_NOA6 _MMIO(0x9E98) -#define HSW_MBVID2_NOA7 _MMIO(0x9E9C) -#define HSW_MBVID2_NOA8 _MMIO(0x9EA0) -#define HSW_MBVID2_NOA9 _MMIO(0x9EA4) - -#define HSW_MBVID2_MISR0 _MMIO(0x9EC0) - -/* NOA (Gen8+) */ -#define NOA_CONFIG(i) _MMIO(0x0D0C + (i) * 4) - -#define MICRO_BP0_0 _MMIO(0x9800) -#define MICRO_BP0_2 _MMIO(0x9804) -#define MICRO_BP0_1 _MMIO(0x9808) - -#define MICRO_BP1_0 _MMIO(0x980C) -#define MICRO_BP1_2 _MMIO(0x9810) -#define MICRO_BP1_1 _MMIO(0x9814) - -#define MICRO_BP2_0 _MMIO(0x9818) -#define MICRO_BP2_2 _MMIO(0x981C) -#define MICRO_BP2_1 _MMIO(0x9820) - -#define MICRO_BP3_0 _MMIO(0x9824) -#define MICRO_BP3_2 _MMIO(0x9828) -#define MICRO_BP3_1 _MMIO(0x982C) - -#define MICRO_BP_TRIGGER _MMIO(0x9830) -#define MICRO_BP3_COUNT_STATUS01 _MMIO(0x9834) -#define MICRO_BP3_COUNT_STATUS23 _MMIO(0x9838) -#define MICRO_BP_FIRED_ARMED _MMIO(0x983C) - -#define GEN12_OAA_DBG_REG _MMIO(0xdc44) -#define GEN12_OAG_OA_PESS _MMIO(0x2b2c) -#define GEN12_OAG_SPCTR_CNF _MMIO(0xdc40) - -#define GDT_CHICKEN_BITS _MMIO(0x9840) -#define GT_NOA_ENABLE 0x00000080 - -#define NOA_DATA _MMIO(0x986C) -#define NOA_WRITE _MMIO(0x9888) -#define GEN10_NOA_WRITE_HIGH _MMIO(0x9884) #define _GEN7_PIPEA_DE_LOAD_SL 0x70068 #define _GEN7_PIPEB_DE_LOAD_SL 0x71068 @@ -1244,177 +252,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define VLV_IOSF_DATA _MMIO(VLV_DISPLAY_BASE + 0x2104) #define VLV_IOSF_ADDR _MMIO(VLV_DISPLAY_BASE + 0x2108) -/* See configdb bunit SB addr map */ -#define BUNIT_REG_BISOC 0x11 - -/* PUNIT_REG_*SSPM0 */ -#define _SSPM0_SSC(val) ((val) << 0) -#define SSPM0_SSC_MASK _SSPM0_SSC(0x3) -#define SSPM0_SSC_PWR_ON _SSPM0_SSC(0x0) -#define SSPM0_SSC_CLK_GATE _SSPM0_SSC(0x1) -#define SSPM0_SSC_RESET _SSPM0_SSC(0x2) -#define SSPM0_SSC_PWR_GATE _SSPM0_SSC(0x3) -#define _SSPM0_SSS(val) ((val) << 24) -#define SSPM0_SSS_MASK _SSPM0_SSS(0x3) -#define SSPM0_SSS_PWR_ON _SSPM0_SSS(0x0) -#define SSPM0_SSS_CLK_GATE _SSPM0_SSS(0x1) -#define SSPM0_SSS_RESET _SSPM0_SSS(0x2) -#define SSPM0_SSS_PWR_GATE _SSPM0_SSS(0x3) - -/* PUNIT_REG_*SSPM1 */ -#define SSPM1_FREQSTAT_SHIFT 24 -#define SSPM1_FREQSTAT_MASK (0x1f << SSPM1_FREQSTAT_SHIFT) -#define SSPM1_FREQGUAR_SHIFT 8 -#define SSPM1_FREQGUAR_MASK (0x1f << SSPM1_FREQGUAR_SHIFT) -#define SSPM1_FREQ_SHIFT 0 -#define SSPM1_FREQ_MASK (0x1f << SSPM1_FREQ_SHIFT) - -#define PUNIT_REG_VEDSSPM0 0x32 -#define PUNIT_REG_VEDSSPM1 0x33 - -#define PUNIT_REG_DSPSSPM 0x36 -#define DSPFREQSTAT_SHIFT_CHV 24 -#define DSPFREQSTAT_MASK_CHV (0x1f << DSPFREQSTAT_SHIFT_CHV) -#define DSPFREQGUAR_SHIFT_CHV 8 -#define DSPFREQGUAR_MASK_CHV (0x1f << DSPFREQGUAR_SHIFT_CHV) -#define DSPFREQSTAT_SHIFT 30 -#define DSPFREQSTAT_MASK (0x3 << DSPFREQSTAT_SHIFT) -#define DSPFREQGUAR_SHIFT 14 -#define DSPFREQGUAR_MASK (0x3 << DSPFREQGUAR_SHIFT) -#define DSP_MAXFIFO_PM5_STATUS (1 << 22) /* chv */ -#define DSP_AUTO_CDCLK_GATE_DISABLE (1 << 7) /* chv */ -#define DSP_MAXFIFO_PM5_ENABLE (1 << 6) /* chv */ -#define _DP_SSC(val, pipe) ((val) << (2 * (pipe))) -#define DP_SSC_MASK(pipe) _DP_SSC(0x3, (pipe)) -#define DP_SSC_PWR_ON(pipe) _DP_SSC(0x0, (pipe)) -#define DP_SSC_CLK_GATE(pipe) _DP_SSC(0x1, (pipe)) -#define DP_SSC_RESET(pipe) _DP_SSC(0x2, (pipe)) -#define DP_SSC_PWR_GATE(pipe) _DP_SSC(0x3, (pipe)) -#define _DP_SSS(val, pipe) ((val) << (2 * (pipe) + 16)) -#define DP_SSS_MASK(pipe) _DP_SSS(0x3, (pipe)) -#define DP_SSS_PWR_ON(pipe) _DP_SSS(0x0, (pipe)) -#define DP_SSS_CLK_GATE(pipe) _DP_SSS(0x1, (pipe)) -#define DP_SSS_RESET(pipe) _DP_SSS(0x2, (pipe)) -#define DP_SSS_PWR_GATE(pipe) _DP_SSS(0x3, (pipe)) - -#define PUNIT_REG_ISPSSPM0 0x39 -#define PUNIT_REG_ISPSSPM1 0x3a - -#define PUNIT_REG_PWRGT_CTRL 0x60 -#define PUNIT_REG_PWRGT_STATUS 0x61 -#define PUNIT_PWRGT_MASK(pw_idx) (3 << ((pw_idx) * 2)) -#define PUNIT_PWRGT_PWR_ON(pw_idx) (0 << ((pw_idx) * 2)) -#define PUNIT_PWRGT_CLK_GATE(pw_idx) (1 << ((pw_idx) * 2)) -#define PUNIT_PWRGT_RESET(pw_idx) (2 << ((pw_idx) * 2)) -#define PUNIT_PWRGT_PWR_GATE(pw_idx) (3 << ((pw_idx) * 2)) - -#define PUNIT_PWGT_IDX_RENDER 0 -#define PUNIT_PWGT_IDX_MEDIA 1 -#define PUNIT_PWGT_IDX_DISP2D 3 -#define PUNIT_PWGT_IDX_DPIO_CMN_BC 5 -#define PUNIT_PWGT_IDX_DPIO_TX_B_LANES_01 6 -#define PUNIT_PWGT_IDX_DPIO_TX_B_LANES_23 7 -#define PUNIT_PWGT_IDX_DPIO_TX_C_LANES_01 8 -#define PUNIT_PWGT_IDX_DPIO_TX_C_LANES_23 9 -#define PUNIT_PWGT_IDX_DPIO_RX0 10 -#define PUNIT_PWGT_IDX_DPIO_RX1 11 -#define PUNIT_PWGT_IDX_DPIO_CMN_D 12 - -#define PUNIT_REG_GPU_LFM 0xd3 -#define PUNIT_REG_GPU_FREQ_REQ 0xd4 -#define PUNIT_REG_GPU_FREQ_STS 0xd8 -#define GPLLENABLE (1 << 4) -#define GENFREQSTATUS (1 << 0) -#define PUNIT_REG_MEDIA_TURBO_FREQ_REQ 0xdc -#define PUNIT_REG_CZ_TIMESTAMP 0xce - -#define PUNIT_FUSE_BUS2 0xf6 /* bits 47:40 */ -#define PUNIT_FUSE_BUS1 0xf5 /* bits 55:48 */ - -#define FB_GFX_FMAX_AT_VMAX_FUSE 0x136 -#define FB_GFX_FREQ_FUSE_MASK 0xff -#define FB_GFX_FMAX_AT_VMAX_2SS4EU_FUSE_SHIFT 24 -#define FB_GFX_FMAX_AT_VMAX_2SS6EU_FUSE_SHIFT 16 -#define FB_GFX_FMAX_AT_VMAX_2SS8EU_FUSE_SHIFT 8 - -#define FB_GFX_FMIN_AT_VMIN_FUSE 0x137 -#define FB_GFX_FMIN_AT_VMIN_FUSE_SHIFT 8 - -#define PUNIT_REG_DDR_SETUP2 0x139 -#define FORCE_DDR_FREQ_REQ_ACK (1 << 8) -#define FORCE_DDR_LOW_FREQ (1 << 1) -#define FORCE_DDR_HIGH_FREQ (1 << 0) - -#define PUNIT_GPU_STATUS_REG 0xdb -#define PUNIT_GPU_STATUS_MAX_FREQ_SHIFT 16 -#define PUNIT_GPU_STATUS_MAX_FREQ_MASK 0xff -#define PUNIT_GPU_STATIS_GFX_MIN_FREQ_SHIFT 8 -#define PUNIT_GPU_STATUS_GFX_MIN_FREQ_MASK 0xff - -#define PUNIT_GPU_DUTYCYCLE_REG 0xdf -#define PUNIT_GPU_DUTYCYCLE_RPE_FREQ_SHIFT 8 -#define PUNIT_GPU_DUTYCYCLE_RPE_FREQ_MASK 0xff - -#define IOSF_NC_FB_GFX_FREQ_FUSE 0x1c -#define FB_GFX_MAX_FREQ_FUSE_SHIFT 3 -#define FB_GFX_MAX_FREQ_FUSE_MASK 0x000007f8 -#define FB_GFX_FGUARANTEED_FREQ_FUSE_SHIFT 11 -#define FB_GFX_FGUARANTEED_FREQ_FUSE_MASK 0x0007f800 -#define IOSF_NC_FB_GFX_FMAX_FUSE_HI 0x34 -#define FB_FMAX_VMIN_FREQ_HI_MASK 0x00000007 -#define IOSF_NC_FB_GFX_FMAX_FUSE_LO 0x30 -#define FB_FMAX_VMIN_FREQ_LO_SHIFT 27 -#define FB_FMAX_VMIN_FREQ_LO_MASK 0xf8000000 - -#define VLV_TURBO_SOC_OVERRIDE 0x04 -#define VLV_OVERRIDE_EN 1 -#define VLV_SOC_TDP_EN (1 << 1) -#define VLV_BIAS_CPU_125_SOC_875 (6 << 2) -#define CHV_BIAS_CPU_50_SOC_50 (3 << 2) - -/* vlv2 north clock has */ -#define CCK_FUSE_REG 0x8 -#define CCK_FUSE_HPLL_FREQ_MASK 0x3 -#define CCK_REG_DSI_PLL_FUSE 0x44 -#define CCK_REG_DSI_PLL_CONTROL 0x48 -#define DSI_PLL_VCO_EN (1 << 31) -#define DSI_PLL_LDO_GATE (1 << 30) -#define DSI_PLL_P1_POST_DIV_SHIFT 17 -#define DSI_PLL_P1_POST_DIV_MASK (0x1ff << 17) -#define DSI_PLL_P2_MUX_DSI0_DIV2 (1 << 13) -#define DSI_PLL_P3_MUX_DSI1_DIV2 (1 << 12) -#define DSI_PLL_MUX_MASK (3 << 9) -#define DSI_PLL_MUX_DSI0_DSIPLL (0 << 10) -#define DSI_PLL_MUX_DSI0_CCK (1 << 10) -#define DSI_PLL_MUX_DSI1_DSIPLL (0 << 9) -#define DSI_PLL_MUX_DSI1_CCK (1 << 9) -#define DSI_PLL_CLK_GATE_MASK (0xf << 5) -#define DSI_PLL_CLK_GATE_DSI0_DSIPLL (1 << 8) -#define DSI_PLL_CLK_GATE_DSI1_DSIPLL (1 << 7) -#define DSI_PLL_CLK_GATE_DSI0_CCK (1 << 6) -#define DSI_PLL_CLK_GATE_DSI1_CCK (1 << 5) -#define DSI_PLL_LOCK (1 << 0) -#define CCK_REG_DSI_PLL_DIVIDER 0x4c -#define DSI_PLL_LFSR (1 << 31) -#define DSI_PLL_FRACTION_EN (1 << 30) -#define DSI_PLL_FRAC_COUNTER_SHIFT 27 -#define DSI_PLL_FRAC_COUNTER_MASK (7 << 27) -#define DSI_PLL_USYNC_CNT_SHIFT 18 -#define DSI_PLL_USYNC_CNT_MASK (0x1ff << 18) -#define DSI_PLL_N1_DIV_SHIFT 16 -#define DSI_PLL_N1_DIV_MASK (3 << 16) -#define DSI_PLL_M1_DIV_SHIFT 0 -#define DSI_PLL_M1_DIV_MASK (0x1ff << 0) -#define CCK_CZ_CLOCK_CONTROL 0x62 -#define CCK_GPLL_CLOCK_CONTROL 0x67 -#define CCK_DISPLAY_CLOCK_CONTROL 0x6b -#define CCK_DISPLAY_REF_CLOCK_CONTROL 0x6c -#define CCK_TRUNK_FORCE_ON (1 << 17) -#define CCK_TRUNK_FORCE_OFF (1 << 16) -#define CCK_FREQUENCY_STATUS (0x1f << 8) -#define CCK_FREQUENCY_STATUS_SHIFT 8 -#define CCK_FREQUENCY_VALUES (0x1f << 0) - /* DPIO registers */ #define DPIO_DEVFN 0 @@ -1905,402 +742,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define OCL2_LDOFUSE_PWR_DIS (1 << 6) #define BXT_PORT_CL1CM_DW30(phy) _BXT_PHY((phy), _PORT_CL1CM_DW30_BC) -/* - * ICL Port/COMBO-PHY Registers - */ -#define _ICL_COMBOPHY_A 0x162000 -#define _ICL_COMBOPHY_B 0x6C000 -#define _EHL_COMBOPHY_C 0x160000 -#define _RKL_COMBOPHY_D 0x161000 -#define _ADL_COMBOPHY_E 0x16B000 - -#define _ICL_COMBOPHY(phy) _PICK(phy, _ICL_COMBOPHY_A, \ - _ICL_COMBOPHY_B, \ - _EHL_COMBOPHY_C, \ - _RKL_COMBOPHY_D, \ - _ADL_COMBOPHY_E) - -/* ICL Port CL_DW registers */ -#define _ICL_PORT_CL_DW(dw, phy) (_ICL_COMBOPHY(phy) + \ - 4 * (dw)) - -#define ICL_PORT_CL_DW5(phy) _MMIO(_ICL_PORT_CL_DW(5, phy)) -#define CL_POWER_DOWN_ENABLE (1 << 4) -#define SUS_CLOCK_CONFIG (3 << 0) - -#define ICL_PORT_CL_DW10(phy) _MMIO(_ICL_PORT_CL_DW(10, phy)) -#define PG_SEQ_DELAY_OVERRIDE_MASK (3 << 25) -#define PG_SEQ_DELAY_OVERRIDE_SHIFT 25 -#define PG_SEQ_DELAY_OVERRIDE_ENABLE (1 << 24) -#define PWR_UP_ALL_LANES (0x0 << 4) -#define PWR_DOWN_LN_3_2_1 (0xe << 4) -#define PWR_DOWN_LN_3_2 (0xc << 4) -#define PWR_DOWN_LN_3 (0x8 << 4) -#define PWR_DOWN_LN_2_1_0 (0x7 << 4) -#define PWR_DOWN_LN_1_0 (0x3 << 4) -#define PWR_DOWN_LN_3_1 (0xa << 4) -#define PWR_DOWN_LN_3_1_0 (0xb << 4) -#define PWR_DOWN_LN_MASK (0xf << 4) -#define PWR_DOWN_LN_SHIFT 4 -#define EDP4K2K_MODE_OVRD_EN (1 << 3) -#define EDP4K2K_MODE_OVRD_OPTIMIZED (1 << 2) - -#define ICL_PORT_CL_DW12(phy) _MMIO(_ICL_PORT_CL_DW(12, phy)) -#define ICL_LANE_ENABLE_AUX (1 << 0) - -/* ICL Port COMP_DW registers */ -#define _ICL_PORT_COMP 0x100 -#define _ICL_PORT_COMP_DW(dw, phy) (_ICL_COMBOPHY(phy) + \ - _ICL_PORT_COMP + 4 * (dw)) - -#define ICL_PORT_COMP_DW0(phy) _MMIO(_ICL_PORT_COMP_DW(0, phy)) -#define COMP_INIT (1 << 31) - -#define ICL_PORT_COMP_DW1(phy) _MMIO(_ICL_PORT_COMP_DW(1, phy)) - -#define ICL_PORT_COMP_DW3(phy) _MMIO(_ICL_PORT_COMP_DW(3, phy)) -#define PROCESS_INFO_DOT_0 (0 << 26) -#define PROCESS_INFO_DOT_1 (1 << 26) -#define PROCESS_INFO_DOT_4 (2 << 26) -#define PROCESS_INFO_MASK (7 << 26) -#define PROCESS_INFO_SHIFT 26 -#define VOLTAGE_INFO_0_85V (0 << 24) -#define VOLTAGE_INFO_0_95V (1 << 24) -#define VOLTAGE_INFO_1_05V (2 << 24) -#define VOLTAGE_INFO_MASK (3 << 24) -#define VOLTAGE_INFO_SHIFT 24 - -#define ICL_PORT_COMP_DW8(phy) _MMIO(_ICL_PORT_COMP_DW(8, phy)) -#define IREFGEN (1 << 24) - -#define ICL_PORT_COMP_DW9(phy) _MMIO(_ICL_PORT_COMP_DW(9, phy)) - -#define ICL_PORT_COMP_DW10(phy) _MMIO(_ICL_PORT_COMP_DW(10, phy)) - -/* ICL Port PCS registers */ -#define _ICL_PORT_PCS_AUX 0x300 -#define _ICL_PORT_PCS_GRP 0x600 -#define _ICL_PORT_PCS_LN(ln) (0x800 + (ln) * 0x100) -#define _ICL_PORT_PCS_DW_AUX(dw, phy) (_ICL_COMBOPHY(phy) + \ - _ICL_PORT_PCS_AUX + 4 * (dw)) -#define _ICL_PORT_PCS_DW_GRP(dw, phy) (_ICL_COMBOPHY(phy) + \ - _ICL_PORT_PCS_GRP + 4 * (dw)) -#define _ICL_PORT_PCS_DW_LN(dw, ln, phy) (_ICL_COMBOPHY(phy) + \ - _ICL_PORT_PCS_LN(ln) + 4 * (dw)) -#define ICL_PORT_PCS_DW1_AUX(phy) _MMIO(_ICL_PORT_PCS_DW_AUX(1, phy)) -#define ICL_PORT_PCS_DW1_GRP(phy) _MMIO(_ICL_PORT_PCS_DW_GRP(1, phy)) -#define ICL_PORT_PCS_DW1_LN(ln, phy) _MMIO(_ICL_PORT_PCS_DW_LN(1, ln, phy)) -#define DCC_MODE_SELECT_MASK (0x3 << 20) -#define DCC_MODE_SELECT_CONTINUOSLY (0x3 << 20) -#define COMMON_KEEPER_EN (1 << 26) -#define LATENCY_OPTIM_MASK (0x3 << 2) -#define LATENCY_OPTIM_VAL(x) ((x) << 2) - -/* ICL Port TX registers */ -#define _ICL_PORT_TX_AUX 0x380 -#define _ICL_PORT_TX_GRP 0x680 -#define _ICL_PORT_TX_LN(ln) (0x880 + (ln) * 0x100) - -#define _ICL_PORT_TX_DW_AUX(dw, phy) (_ICL_COMBOPHY(phy) + \ - _ICL_PORT_TX_AUX + 4 * (dw)) -#define _ICL_PORT_TX_DW_GRP(dw, phy) (_ICL_COMBOPHY(phy) + \ - _ICL_PORT_TX_GRP + 4 * (dw)) -#define _ICL_PORT_TX_DW_LN(dw, ln, phy) (_ICL_COMBOPHY(phy) + \ - _ICL_PORT_TX_LN(ln) + 4 * (dw)) - -#define ICL_PORT_TX_DW2_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(2, phy)) -#define ICL_PORT_TX_DW2_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(2, phy)) -#define ICL_PORT_TX_DW2_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(2, ln, phy)) -#define SWING_SEL_UPPER(x) (((x) >> 3) << 15) -#define SWING_SEL_UPPER_MASK (1 << 15) -#define SWING_SEL_LOWER(x) (((x) & 0x7) << 11) -#define SWING_SEL_LOWER_MASK (0x7 << 11) -#define FRC_LATENCY_OPTIM_MASK (0x7 << 8) -#define FRC_LATENCY_OPTIM_VAL(x) ((x) << 8) -#define RCOMP_SCALAR(x) ((x) << 0) -#define RCOMP_SCALAR_MASK (0xFF << 0) - -#define ICL_PORT_TX_DW4_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(4, phy)) -#define ICL_PORT_TX_DW4_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(4, phy)) -#define ICL_PORT_TX_DW4_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(4, ln, phy)) -#define LOADGEN_SELECT (1 << 31) -#define POST_CURSOR_1(x) ((x) << 12) -#define POST_CURSOR_1_MASK (0x3F << 12) -#define POST_CURSOR_2(x) ((x) << 6) -#define POST_CURSOR_2_MASK (0x3F << 6) -#define CURSOR_COEFF(x) ((x) << 0) -#define CURSOR_COEFF_MASK (0x3F << 0) - -#define ICL_PORT_TX_DW5_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(5, phy)) -#define ICL_PORT_TX_DW5_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(5, phy)) -#define ICL_PORT_TX_DW5_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(5, ln, phy)) -#define TX_TRAINING_EN (1 << 31) -#define TAP2_DISABLE (1 << 30) -#define TAP3_DISABLE (1 << 29) -#define SCALING_MODE_SEL(x) ((x) << 18) -#define SCALING_MODE_SEL_MASK (0x7 << 18) -#define RTERM_SELECT(x) ((x) << 3) -#define RTERM_SELECT_MASK (0x7 << 3) - -#define ICL_PORT_TX_DW7_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(7, phy)) -#define ICL_PORT_TX_DW7_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(7, phy)) -#define ICL_PORT_TX_DW7_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(7, ln, phy)) -#define N_SCALAR(x) ((x) << 24) -#define N_SCALAR_MASK (0x7F << 24) - -#define ICL_PORT_TX_DW8_AUX(phy) _MMIO(_ICL_PORT_TX_DW_AUX(8, phy)) -#define ICL_PORT_TX_DW8_GRP(phy) _MMIO(_ICL_PORT_TX_DW_GRP(8, phy)) -#define ICL_PORT_TX_DW8_LN(ln, phy) _MMIO(_ICL_PORT_TX_DW_LN(8, ln, phy)) -#define ICL_PORT_TX_DW8_ODCC_CLK_SEL REG_BIT(31) -#define ICL_PORT_TX_DW8_ODCC_CLK_DIV_SEL_MASK REG_GENMASK(30, 29) -#define ICL_PORT_TX_DW8_ODCC_CLK_DIV_SEL_DIV2 REG_FIELD_PREP(ICL_PORT_TX_DW8_ODCC_CLK_DIV_SEL_MASK, 0x1) - -#define _ICL_DPHY_CHKN_REG 0x194 -#define ICL_DPHY_CHKN(port) _MMIO(_ICL_COMBOPHY(port) + _ICL_DPHY_CHKN_REG) -#define ICL_DPHY_CHKN_AFE_OVER_PPI_STRAP REG_BIT(7) - -#define MG_PHY_PORT_LN(ln, tc_port, ln0p1, ln0p2, ln1p1) \ - _MMIO(_PORT(tc_port, ln0p1, ln0p2) + (ln) * ((ln1p1) - (ln0p1))) - -#define MG_TX_LINK_PARAMS_TX1LN0_PORT1 0x16812C -#define MG_TX_LINK_PARAMS_TX1LN1_PORT1 0x16852C -#define MG_TX_LINK_PARAMS_TX1LN0_PORT2 0x16912C -#define MG_TX_LINK_PARAMS_TX1LN1_PORT2 0x16952C -#define MG_TX_LINK_PARAMS_TX1LN0_PORT3 0x16A12C -#define MG_TX_LINK_PARAMS_TX1LN1_PORT3 0x16A52C -#define MG_TX_LINK_PARAMS_TX1LN0_PORT4 0x16B12C -#define MG_TX_LINK_PARAMS_TX1LN1_PORT4 0x16B52C -#define MG_TX1_LINK_PARAMS(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_LINK_PARAMS_TX1LN0_PORT1, \ - MG_TX_LINK_PARAMS_TX1LN0_PORT2, \ - MG_TX_LINK_PARAMS_TX1LN1_PORT1) - -#define MG_TX_LINK_PARAMS_TX2LN0_PORT1 0x1680AC -#define MG_TX_LINK_PARAMS_TX2LN1_PORT1 0x1684AC -#define MG_TX_LINK_PARAMS_TX2LN0_PORT2 0x1690AC -#define MG_TX_LINK_PARAMS_TX2LN1_PORT2 0x1694AC -#define MG_TX_LINK_PARAMS_TX2LN0_PORT3 0x16A0AC -#define MG_TX_LINK_PARAMS_TX2LN1_PORT3 0x16A4AC -#define MG_TX_LINK_PARAMS_TX2LN0_PORT4 0x16B0AC -#define MG_TX_LINK_PARAMS_TX2LN1_PORT4 0x16B4AC -#define MG_TX2_LINK_PARAMS(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_LINK_PARAMS_TX2LN0_PORT1, \ - MG_TX_LINK_PARAMS_TX2LN0_PORT2, \ - MG_TX_LINK_PARAMS_TX2LN1_PORT1) -#define CRI_USE_FS32 (1 << 5) - -#define MG_TX_PISO_READLOAD_TX1LN0_PORT1 0x16814C -#define MG_TX_PISO_READLOAD_TX1LN1_PORT1 0x16854C -#define MG_TX_PISO_READLOAD_TX1LN0_PORT2 0x16914C -#define MG_TX_PISO_READLOAD_TX1LN1_PORT2 0x16954C -#define MG_TX_PISO_READLOAD_TX1LN0_PORT3 0x16A14C -#define MG_TX_PISO_READLOAD_TX1LN1_PORT3 0x16A54C -#define MG_TX_PISO_READLOAD_TX1LN0_PORT4 0x16B14C -#define MG_TX_PISO_READLOAD_TX1LN1_PORT4 0x16B54C -#define MG_TX1_PISO_READLOAD(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_PISO_READLOAD_TX1LN0_PORT1, \ - MG_TX_PISO_READLOAD_TX1LN0_PORT2, \ - MG_TX_PISO_READLOAD_TX1LN1_PORT1) - -#define MG_TX_PISO_READLOAD_TX2LN0_PORT1 0x1680CC -#define MG_TX_PISO_READLOAD_TX2LN1_PORT1 0x1684CC -#define MG_TX_PISO_READLOAD_TX2LN0_PORT2 0x1690CC -#define MG_TX_PISO_READLOAD_TX2LN1_PORT2 0x1694CC -#define MG_TX_PISO_READLOAD_TX2LN0_PORT3 0x16A0CC -#define MG_TX_PISO_READLOAD_TX2LN1_PORT3 0x16A4CC -#define MG_TX_PISO_READLOAD_TX2LN0_PORT4 0x16B0CC -#define MG_TX_PISO_READLOAD_TX2LN1_PORT4 0x16B4CC -#define MG_TX2_PISO_READLOAD(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_PISO_READLOAD_TX2LN0_PORT1, \ - MG_TX_PISO_READLOAD_TX2LN0_PORT2, \ - MG_TX_PISO_READLOAD_TX2LN1_PORT1) -#define CRI_CALCINIT (1 << 1) - -#define MG_TX_SWINGCTRL_TX1LN0_PORT1 0x168148 -#define MG_TX_SWINGCTRL_TX1LN1_PORT1 0x168548 -#define MG_TX_SWINGCTRL_TX1LN0_PORT2 0x169148 -#define MG_TX_SWINGCTRL_TX1LN1_PORT2 0x169548 -#define MG_TX_SWINGCTRL_TX1LN0_PORT3 0x16A148 -#define MG_TX_SWINGCTRL_TX1LN1_PORT3 0x16A548 -#define MG_TX_SWINGCTRL_TX1LN0_PORT4 0x16B148 -#define MG_TX_SWINGCTRL_TX1LN1_PORT4 0x16B548 -#define MG_TX1_SWINGCTRL(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_SWINGCTRL_TX1LN0_PORT1, \ - MG_TX_SWINGCTRL_TX1LN0_PORT2, \ - MG_TX_SWINGCTRL_TX1LN1_PORT1) - -#define MG_TX_SWINGCTRL_TX2LN0_PORT1 0x1680C8 -#define MG_TX_SWINGCTRL_TX2LN1_PORT1 0x1684C8 -#define MG_TX_SWINGCTRL_TX2LN0_PORT2 0x1690C8 -#define MG_TX_SWINGCTRL_TX2LN1_PORT2 0x1694C8 -#define MG_TX_SWINGCTRL_TX2LN0_PORT3 0x16A0C8 -#define MG_TX_SWINGCTRL_TX2LN1_PORT3 0x16A4C8 -#define MG_TX_SWINGCTRL_TX2LN0_PORT4 0x16B0C8 -#define MG_TX_SWINGCTRL_TX2LN1_PORT4 0x16B4C8 -#define MG_TX2_SWINGCTRL(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_SWINGCTRL_TX2LN0_PORT1, \ - MG_TX_SWINGCTRL_TX2LN0_PORT2, \ - MG_TX_SWINGCTRL_TX2LN1_PORT1) -#define CRI_TXDEEMPH_OVERRIDE_17_12(x) ((x) << 0) -#define CRI_TXDEEMPH_OVERRIDE_17_12_MASK (0x3F << 0) - -#define MG_TX_DRVCTRL_TX1LN0_TXPORT1 0x168144 -#define MG_TX_DRVCTRL_TX1LN1_TXPORT1 0x168544 -#define MG_TX_DRVCTRL_TX1LN0_TXPORT2 0x169144 -#define MG_TX_DRVCTRL_TX1LN1_TXPORT2 0x169544 -#define MG_TX_DRVCTRL_TX1LN0_TXPORT3 0x16A144 -#define MG_TX_DRVCTRL_TX1LN1_TXPORT3 0x16A544 -#define MG_TX_DRVCTRL_TX1LN0_TXPORT4 0x16B144 -#define MG_TX_DRVCTRL_TX1LN1_TXPORT4 0x16B544 -#define MG_TX1_DRVCTRL(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_DRVCTRL_TX1LN0_TXPORT1, \ - MG_TX_DRVCTRL_TX1LN0_TXPORT2, \ - MG_TX_DRVCTRL_TX1LN1_TXPORT1) - -#define MG_TX_DRVCTRL_TX2LN0_PORT1 0x1680C4 -#define MG_TX_DRVCTRL_TX2LN1_PORT1 0x1684C4 -#define MG_TX_DRVCTRL_TX2LN0_PORT2 0x1690C4 -#define MG_TX_DRVCTRL_TX2LN1_PORT2 0x1694C4 -#define MG_TX_DRVCTRL_TX2LN0_PORT3 0x16A0C4 -#define MG_TX_DRVCTRL_TX2LN1_PORT3 0x16A4C4 -#define MG_TX_DRVCTRL_TX2LN0_PORT4 0x16B0C4 -#define MG_TX_DRVCTRL_TX2LN1_PORT4 0x16B4C4 -#define MG_TX2_DRVCTRL(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_DRVCTRL_TX2LN0_PORT1, \ - MG_TX_DRVCTRL_TX2LN0_PORT2, \ - MG_TX_DRVCTRL_TX2LN1_PORT1) -#define CRI_TXDEEMPH_OVERRIDE_11_6(x) ((x) << 24) -#define CRI_TXDEEMPH_OVERRIDE_11_6_MASK (0x3F << 24) -#define CRI_TXDEEMPH_OVERRIDE_EN (1 << 22) -#define CRI_TXDEEMPH_OVERRIDE_5_0(x) ((x) << 16) -#define CRI_TXDEEMPH_OVERRIDE_5_0_MASK (0x3F << 16) -#define CRI_LOADGEN_SEL(x) ((x) << 12) -#define CRI_LOADGEN_SEL_MASK (0x3 << 12) - -#define MG_CLKHUB_LN0_PORT1 0x16839C -#define MG_CLKHUB_LN1_PORT1 0x16879C -#define MG_CLKHUB_LN0_PORT2 0x16939C -#define MG_CLKHUB_LN1_PORT2 0x16979C -#define MG_CLKHUB_LN0_PORT3 0x16A39C -#define MG_CLKHUB_LN1_PORT3 0x16A79C -#define MG_CLKHUB_LN0_PORT4 0x16B39C -#define MG_CLKHUB_LN1_PORT4 0x16B79C -#define MG_CLKHUB(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_CLKHUB_LN0_PORT1, \ - MG_CLKHUB_LN0_PORT2, \ - MG_CLKHUB_LN1_PORT1) -#define CFG_LOW_RATE_LKREN_EN (1 << 11) - -#define MG_TX_DCC_TX1LN0_PORT1 0x168110 -#define MG_TX_DCC_TX1LN1_PORT1 0x168510 -#define MG_TX_DCC_TX1LN0_PORT2 0x169110 -#define MG_TX_DCC_TX1LN1_PORT2 0x169510 -#define MG_TX_DCC_TX1LN0_PORT3 0x16A110 -#define MG_TX_DCC_TX1LN1_PORT3 0x16A510 -#define MG_TX_DCC_TX1LN0_PORT4 0x16B110 -#define MG_TX_DCC_TX1LN1_PORT4 0x16B510 -#define MG_TX1_DCC(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_DCC_TX1LN0_PORT1, \ - MG_TX_DCC_TX1LN0_PORT2, \ - MG_TX_DCC_TX1LN1_PORT1) -#define MG_TX_DCC_TX2LN0_PORT1 0x168090 -#define MG_TX_DCC_TX2LN1_PORT1 0x168490 -#define MG_TX_DCC_TX2LN0_PORT2 0x169090 -#define MG_TX_DCC_TX2LN1_PORT2 0x169490 -#define MG_TX_DCC_TX2LN0_PORT3 0x16A090 -#define MG_TX_DCC_TX2LN1_PORT3 0x16A490 -#define MG_TX_DCC_TX2LN0_PORT4 0x16B090 -#define MG_TX_DCC_TX2LN1_PORT4 0x16B490 -#define MG_TX2_DCC(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_TX_DCC_TX2LN0_PORT1, \ - MG_TX_DCC_TX2LN0_PORT2, \ - MG_TX_DCC_TX2LN1_PORT1) -#define CFG_AMI_CK_DIV_OVERRIDE_VAL(x) ((x) << 25) -#define CFG_AMI_CK_DIV_OVERRIDE_VAL_MASK (0x3 << 25) -#define CFG_AMI_CK_DIV_OVERRIDE_EN (1 << 24) - -#define MG_DP_MODE_LN0_ACU_PORT1 0x1683A0 -#define MG_DP_MODE_LN1_ACU_PORT1 0x1687A0 -#define MG_DP_MODE_LN0_ACU_PORT2 0x1693A0 -#define MG_DP_MODE_LN1_ACU_PORT2 0x1697A0 -#define MG_DP_MODE_LN0_ACU_PORT3 0x16A3A0 -#define MG_DP_MODE_LN1_ACU_PORT3 0x16A7A0 -#define MG_DP_MODE_LN0_ACU_PORT4 0x16B3A0 -#define MG_DP_MODE_LN1_ACU_PORT4 0x16B7A0 -#define MG_DP_MODE(ln, tc_port) \ - MG_PHY_PORT_LN(ln, tc_port, MG_DP_MODE_LN0_ACU_PORT1, \ - MG_DP_MODE_LN0_ACU_PORT2, \ - MG_DP_MODE_LN1_ACU_PORT1) -#define MG_DP_MODE_CFG_DP_X2_MODE (1 << 7) -#define MG_DP_MODE_CFG_DP_X1_MODE (1 << 6) - -/* - * DG2 SNPS PHY registers (TC1 = PHY_E) - */ -#define _SNPS_PHY_A_BASE 0x168000 -#define _SNPS_PHY_B_BASE 0x169000 -#define _SNPS_PHY(phy) _PHY(phy, \ - _SNPS_PHY_A_BASE, \ - _SNPS_PHY_B_BASE) -#define _SNPS2(phy, reg) (_SNPS_PHY(phy) - \ - _SNPS_PHY_A_BASE + (reg)) -#define _MMIO_SNPS(phy, reg) _MMIO(_SNPS2(phy, reg)) -#define _MMIO_SNPS_LN(ln, phy, reg) _MMIO(_SNPS2(phy, \ - (reg) + (ln) * 0x10)) - -#define SNPS_PHY_MPLLB_CP(phy) _MMIO_SNPS(phy, 0x168000) -#define SNPS_PHY_MPLLB_CP_INT REG_GENMASK(31, 25) -#define SNPS_PHY_MPLLB_CP_INT_GS REG_GENMASK(23, 17) -#define SNPS_PHY_MPLLB_CP_PROP REG_GENMASK(15, 9) -#define SNPS_PHY_MPLLB_CP_PROP_GS REG_GENMASK(7, 1) - -#define SNPS_PHY_MPLLB_DIV(phy) _MMIO_SNPS(phy, 0x168004) -#define SNPS_PHY_MPLLB_FORCE_EN REG_BIT(31) -#define SNPS_PHY_MPLLB_DIV_CLK_EN REG_BIT(30) -#define SNPS_PHY_MPLLB_DIV5_CLK_EN REG_BIT(29) -#define SNPS_PHY_MPLLB_V2I REG_GENMASK(27, 26) -#define SNPS_PHY_MPLLB_FREQ_VCO REG_GENMASK(25, 24) -#define SNPS_PHY_MPLLB_DIV_MULTIPLIER REG_GENMASK(23, 16) -#define SNPS_PHY_MPLLB_PMIX_EN REG_BIT(10) -#define SNPS_PHY_MPLLB_DP2_MODE REG_BIT(9) -#define SNPS_PHY_MPLLB_WORD_DIV2_EN REG_BIT(8) -#define SNPS_PHY_MPLLB_TX_CLK_DIV REG_GENMASK(7, 5) -#define SNPS_PHY_MPLLB_SHIM_DIV32_CLK_SEL REG_BIT(0) - -#define SNPS_PHY_MPLLB_FRACN1(phy) _MMIO_SNPS(phy, 0x168008) -#define SNPS_PHY_MPLLB_FRACN_EN REG_BIT(31) -#define SNPS_PHY_MPLLB_FRACN_CGG_UPDATE_EN REG_BIT(30) -#define SNPS_PHY_MPLLB_FRACN_DEN REG_GENMASK(15, 0) - -#define SNPS_PHY_MPLLB_FRACN2(phy) _MMIO_SNPS(phy, 0x16800C) -#define SNPS_PHY_MPLLB_FRACN_REM REG_GENMASK(31, 16) -#define SNPS_PHY_MPLLB_FRACN_QUOT REG_GENMASK(15, 0) - -#define SNPS_PHY_MPLLB_SSCEN(phy) _MMIO_SNPS(phy, 0x168014) -#define SNPS_PHY_MPLLB_SSC_EN REG_BIT(31) -#define SNPS_PHY_MPLLB_SSC_UP_SPREAD REG_BIT(30) -#define SNPS_PHY_MPLLB_SSC_PEAK REG_GENMASK(29, 10) - -#define SNPS_PHY_MPLLB_SSCSTEP(phy) _MMIO_SNPS(phy, 0x168018) -#define SNPS_PHY_MPLLB_SSC_STEPSIZE REG_GENMASK(31, 11) - -#define SNPS_PHY_MPLLB_DIV2(phy) _MMIO_SNPS(phy, 0x16801C) -#define SNPS_PHY_MPLLB_HDMI_PIXEL_CLK_DIV REG_GENMASK(19, 18) -#define SNPS_PHY_MPLLB_HDMI_DIV REG_GENMASK(17, 15) -#define SNPS_PHY_MPLLB_REF_CLK_DIV REG_GENMASK(14, 12) -#define SNPS_PHY_MPLLB_MULTIPLIER REG_GENMASK(11, 0) - -#define SNPS_PHY_REF_CONTROL(phy) _MMIO_SNPS(phy, 0x168188) -#define SNPS_PHY_REF_CONTROL_REF_RANGE REG_GENMASK(31, 27) - -#define SNPS_PHY_TX_REQ(phy) _MMIO_SNPS(phy, 0x168200) -#define SNPS_PHY_TX_REQ_LN_DIS_PWR_STATE_PSR REG_GENMASK(31, 30) - -#define SNPS_PHY_TX_EQ(ln, phy) _MMIO_SNPS_LN(ln, phy, 0x168300) -#define SNPS_PHY_TX_EQ_MAIN REG_GENMASK(23, 18) -#define SNPS_PHY_TX_EQ_POST REG_GENMASK(15, 10) -#define SNPS_PHY_TX_EQ_PRE REG_GENMASK(7, 2) - /* The spec defines this only for BXT PHY0, but lets assume that this * would exist for PHY1 too if it had a second channel. */ @@ -2309,21 +750,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define BXT_PORT_CL2CM_DW6(phy) _BXT_PHY((phy), _PORT_CL2CM_DW6_BC) #define DW6_OLDO_DYN_PWR_DOWN_EN (1 << 28) -#define FIA1_BASE 0x163000 -#define FIA2_BASE 0x16E000 -#define FIA3_BASE 0x16F000 -#define _FIA(fia) _PICK((fia), FIA1_BASE, FIA2_BASE, FIA3_BASE) -#define _MMIO_FIA(fia, off) _MMIO(_FIA(fia) + (off)) - -/* ICL PHY DFLEX registers */ -#define PORT_TX_DFLEXDPMLE1(fia) _MMIO_FIA((fia), 0x008C0) -#define DFLEXDPMLE1_DPMLETC_MASK(idx) (0xf << (4 * (idx))) -#define DFLEXDPMLE1_DPMLETC_ML0(idx) (1 << (4 * (idx))) -#define DFLEXDPMLE1_DPMLETC_ML1_0(idx) (3 << (4 * (idx))) -#define DFLEXDPMLE1_DPMLETC_ML3(idx) (8 << (4 * (idx))) -#define DFLEXDPMLE1_DPMLETC_ML3_2(idx) (12 << (4 * (idx))) -#define DFLEXDPMLE1_DPMLETC_ML3_0(idx) (15 << (4 * (idx))) - /* BXT PHY Ref registers */ #define _PORT_REF_DW3_A 0x16218C #define _PORT_REF_DW3_BC 0x6C18C @@ -2549,64 +975,8 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define XEHP_VEBOX3_RING_BASE 0x1e8000 #define XEHP_VEBOX4_RING_BASE 0x1f8000 #define BLT_RING_BASE 0x22000 -#define RING_TAIL(base) _MMIO((base) + 0x30) -#define RING_HEAD(base) _MMIO((base) + 0x34) -#define RING_START(base) _MMIO((base) + 0x38) -#define RING_CTL(base) _MMIO((base) + 0x3c) -#define RING_CTL_SIZE(size) ((size) - PAGE_SIZE) /* in bytes -> pages */ -#define RING_SYNC_0(base) _MMIO((base) + 0x40) -#define RING_SYNC_1(base) _MMIO((base) + 0x44) -#define RING_SYNC_2(base) _MMIO((base) + 0x48) -#define GEN6_RVSYNC (RING_SYNC_0(RENDER_RING_BASE)) -#define GEN6_RBSYNC (RING_SYNC_1(RENDER_RING_BASE)) -#define GEN6_RVESYNC (RING_SYNC_2(RENDER_RING_BASE)) -#define GEN6_VBSYNC (RING_SYNC_0(GEN6_BSD_RING_BASE)) -#define GEN6_VRSYNC (RING_SYNC_1(GEN6_BSD_RING_BASE)) -#define GEN6_VVESYNC (RING_SYNC_2(GEN6_BSD_RING_BASE)) -#define GEN6_BRSYNC (RING_SYNC_0(BLT_RING_BASE)) -#define GEN6_BVSYNC (RING_SYNC_1(BLT_RING_BASE)) -#define GEN6_BVESYNC (RING_SYNC_2(BLT_RING_BASE)) -#define GEN6_VEBSYNC (RING_SYNC_0(VEBOX_RING_BASE)) -#define GEN6_VERSYNC (RING_SYNC_1(VEBOX_RING_BASE)) -#define GEN6_VEVSYNC (RING_SYNC_2(VEBOX_RING_BASE)) -#define GEN6_NOSYNC INVALID_MMIO_REG -#define RING_PSMI_CTL(base) _MMIO((base) + 0x50) -#define RING_MAX_IDLE(base) _MMIO((base) + 0x54) -#define RING_HWS_PGA(base) _MMIO((base) + 0x80) -#define RING_ID(base) _MMIO((base) + 0x8c) -#define RING_HWS_PGA_GEN6(base) _MMIO((base) + 0x2080) - -#define RING_CMD_CCTL(base) _MMIO((base) + 0xc4) -/* - * CMD_CCTL read/write fields take a MOCS value and _not_ a table index. - * The lsb of each can be considered a separate enabling bit for encryption. - * 6:0 == default MOCS value for reads => 6:1 == table index for reads. - * 13:7 == default MOCS value for writes => 13:8 == table index for writes. - * 15:14 == Reserved => 31:30 are set to 0. - */ -#define CMD_CCTL_WRITE_OVERRIDE_MASK REG_GENMASK(13, 7) -#define CMD_CCTL_READ_OVERRIDE_MASK REG_GENMASK(6, 0) -#define CMD_CCTL_MOCS_MASK (CMD_CCTL_WRITE_OVERRIDE_MASK | \ - CMD_CCTL_READ_OVERRIDE_MASK) -#define CMD_CCTL_MOCS_OVERRIDE(write, read) \ - (REG_FIELD_PREP(CMD_CCTL_WRITE_OVERRIDE_MASK, (write) << 1) | \ - REG_FIELD_PREP(CMD_CCTL_READ_OVERRIDE_MASK, (read) << 1)) - -#define BLIT_CCTL(base) _MMIO((base) + 0x204) -#define BLIT_CCTL_DST_MOCS_MASK REG_GENMASK(14, 8) -#define BLIT_CCTL_SRC_MOCS_MASK REG_GENMASK(6, 0) -#define BLIT_CCTL_MASK (BLIT_CCTL_DST_MOCS_MASK | \ - BLIT_CCTL_SRC_MOCS_MASK) -#define BLIT_CCTL_MOCS(dst, src) \ - (REG_FIELD_PREP(BLIT_CCTL_DST_MOCS_MASK, (dst) << 1) | \ - REG_FIELD_PREP(BLIT_CCTL_SRC_MOCS_MASK, (src) << 1)) - -#define RING_RESET_CTL(base) _MMIO((base) + 0xd0) -#define RESET_CTL_CAT_ERROR REG_BIT(2) -#define RESET_CTL_READY_TO_RESET REG_BIT(1) -#define RESET_CTL_REQUEST_RESET REG_BIT(0) - -#define RING_SEMA_WAIT_POLL(base) _MMIO((base) + 0x24c) + + #define HSW_GTT_CACHE_EN _MMIO(0x4024) #define GTT_CACHE_EN_ALL 0xF0007FFF @@ -2623,190 +993,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define GEN7_MEDIA_MAX_REQ_COUNT _MMIO(0x4070) #define GEN7_GFX_MAX_REQ_COUNT _MMIO(0x4074) -#define GAMTARBMODE _MMIO(0x04a08) -#define ARB_MODE_BWGTLB_DISABLE (1 << 9) -#define ARB_MODE_SWIZZLE_BDW (1 << 1) -#define RENDER_HWS_PGA_GEN7 _MMIO(0x04080) - -#define _RING_FAULT_REG_RCS 0x4094 -#define _RING_FAULT_REG_VCS 0x4194 -#define _RING_FAULT_REG_BCS 0x4294 -#define _RING_FAULT_REG_VECS 0x4394 -#define RING_FAULT_REG(engine) _MMIO(_PICK((engine)->class, \ - _RING_FAULT_REG_RCS, \ - _RING_FAULT_REG_VCS, \ - _RING_FAULT_REG_VECS, \ - _RING_FAULT_REG_BCS)) -#define GEN8_RING_FAULT_REG _MMIO(0x4094) -#define GEN12_RING_FAULT_REG _MMIO(0xcec4) -#define GEN8_RING_FAULT_ENGINE_ID(x) (((x) >> 12) & 0x7) -#define RING_FAULT_GTTSEL_MASK (1 << 11) -#define RING_FAULT_SRCID(x) (((x) >> 3) & 0xff) -#define RING_FAULT_FAULT_TYPE(x) (((x) >> 1) & 0x3) -#define RING_FAULT_VALID (1 << 0) -#define DONE_REG _MMIO(0x40b0) -#define GEN12_GAM_DONE _MMIO(0xcf68) -#define GEN8_PRIVATE_PAT_LO _MMIO(0x40e0) -#define GEN8_PRIVATE_PAT_HI _MMIO(0x40e0 + 4) -#define GEN10_PAT_INDEX(index) _MMIO(0x40e0 + (index) * 4) -#define GEN12_PAT_INDEX(index) _MMIO(0x4800 + (index) * 4) -#define BSD_HWS_PGA_GEN7 _MMIO(0x04180) -#define GEN12_GFX_CCS_AUX_NV _MMIO(0x4208) -#define GEN12_VD0_AUX_NV _MMIO(0x4218) -#define GEN12_VD1_AUX_NV _MMIO(0x4228) -#define GEN12_VD2_AUX_NV _MMIO(0x4298) -#define GEN12_VD3_AUX_NV _MMIO(0x42A8) -#define GEN12_VE0_AUX_NV _MMIO(0x4238) -#define GEN12_VE1_AUX_NV _MMIO(0x42B8) -#define AUX_INV REG_BIT(0) -#define BLT_HWS_PGA_GEN7 _MMIO(0x04280) -#define VEBOX_HWS_PGA_GEN7 _MMIO(0x04380) -#define RING_ACTHD(base) _MMIO((base) + 0x74) -#define RING_ACTHD_UDW(base) _MMIO((base) + 0x5c) -#define RING_NOPID(base) _MMIO((base) + 0x94) -#define RING_IMR(base) _MMIO((base) + 0xa8) -#define RING_HWSTAM(base) _MMIO((base) + 0x98) -#define RING_TIMESTAMP(base) _MMIO((base) + 0x358) -#define RING_TIMESTAMP_UDW(base) _MMIO((base) + 0x358 + 4) -#define TAIL_ADDR 0x001FFFF8 -#define HEAD_WRAP_COUNT 0xFFE00000 -#define HEAD_WRAP_ONE 0x00200000 -#define HEAD_ADDR 0x001FFFFC -#define RING_NR_PAGES 0x001FF000 -#define RING_REPORT_MASK 0x00000006 -#define RING_REPORT_64K 0x00000002 -#define RING_REPORT_128K 0x00000004 -#define RING_NO_REPORT 0x00000000 -#define RING_VALID_MASK 0x00000001 -#define RING_VALID 0x00000001 -#define RING_INVALID 0x00000000 -#define RING_WAIT_I8XX (1 << 0) /* gen2, PRBx_HEAD */ -#define RING_WAIT (1 << 11) /* gen3+, PRBx_CTL */ -#define RING_WAIT_SEMAPHORE (1 << 10) /* gen6+ */ - -#define MISC_STATUS0 _MMIO(0xA500) -#define MISC_STATUS1 _MMIO(0xA504) - -/* There are 16 64-bit CS General Purpose Registers per-engine on Gen8+ */ -#define GEN8_RING_CS_GPR(base, n) _MMIO((base) + 0x600 + (n) * 8) -#define GEN8_RING_CS_GPR_UDW(base, n) _MMIO((base) + 0x600 + (n) * 8 + 4) - -#define RING_FORCE_TO_NONPRIV(base, i) _MMIO(((base) + 0x4D0) + (i) * 4) -#define RING_FORCE_TO_NONPRIV_ADDRESS_MASK REG_GENMASK(25, 2) -#define RING_FORCE_TO_NONPRIV_ACCESS_RW (0 << 28) /* CFL+ & Gen11+ */ -#define RING_FORCE_TO_NONPRIV_ACCESS_RD (1 << 28) -#define RING_FORCE_TO_NONPRIV_ACCESS_WR (2 << 28) -#define RING_FORCE_TO_NONPRIV_ACCESS_INVALID (3 << 28) -#define RING_FORCE_TO_NONPRIV_ACCESS_MASK (3 << 28) -#define RING_FORCE_TO_NONPRIV_RANGE_1 (0 << 0) /* CFL+ & Gen11+ */ -#define RING_FORCE_TO_NONPRIV_RANGE_4 (1 << 0) -#define RING_FORCE_TO_NONPRIV_RANGE_16 (2 << 0) -#define RING_FORCE_TO_NONPRIV_RANGE_64 (3 << 0) -#define RING_FORCE_TO_NONPRIV_RANGE_MASK (3 << 0) -#define RING_FORCE_TO_NONPRIV_MASK_VALID \ - (RING_FORCE_TO_NONPRIV_RANGE_MASK \ - | RING_FORCE_TO_NONPRIV_ACCESS_MASK) -#define RING_MAX_NONPRIV_SLOTS 12 - -#define GEN7_TLB_RD_ADDR _MMIO(0x4700) - -#define GEN9_GAMT_ECO_REG_RW_IA _MMIO(0x4ab0) -#define GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS (1 << 18) - -#define GEN8_GAMW_ECO_DEV_RW_IA _MMIO(0x4080) -#define GAMW_ECO_ENABLE_64K_IPS_FIELD 0xF -#define GAMW_ECO_DEV_CTX_RELOAD_DISABLE (1 << 7) - -#define GAMT_CHKN_BIT_REG _MMIO(0x4ab8) -#define GAMT_CHKN_DISABLE_L3_COH_PIPE (1 << 31) -#define GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING (1 << 28) -#define GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT (1 << 24) - -#define GEN8_RTCR _MMIO(0x4260) -#define GEN8_M1TCR _MMIO(0x4264) -#define GEN8_M2TCR _MMIO(0x4268) -#define GEN8_BTCR _MMIO(0x426c) -#define GEN8_VTCR _MMIO(0x4270) - -#if 0 -#define PRB0_TAIL _MMIO(0x2030) -#define PRB0_HEAD _MMIO(0x2034) -#define PRB0_START _MMIO(0x2038) -#define PRB0_CTL _MMIO(0x203c) -#define PRB1_TAIL _MMIO(0x2040) /* 915+ only */ -#define PRB1_HEAD _MMIO(0x2044) /* 915+ only */ -#define PRB1_START _MMIO(0x2048) /* 915+ only */ -#define PRB1_CTL _MMIO(0x204c) /* 915+ only */ -#endif -#define IPEIR_I965 _MMIO(0x2064) -#define IPEHR_I965 _MMIO(0x2068) -#define GEN7_SC_INSTDONE _MMIO(0x7100) -#define GEN12_SC_INSTDONE_EXTRA _MMIO(0x7104) -#define GEN12_SC_INSTDONE_EXTRA2 _MMIO(0x7108) -#define GEN7_SAMPLER_INSTDONE _MMIO(0xe160) -#define GEN7_ROW_INSTDONE _MMIO(0xe164) -#define XEHPG_INSTDONE_GEOM_SVG _MMIO(0x666c) -#define MCFG_MCR_SELECTOR _MMIO(0xfd0) -#define SF_MCR_SELECTOR _MMIO(0xfd8) -#define GEN8_MCR_SELECTOR _MMIO(0xfdc) -#define GEN8_MCR_SLICE(slice) (((slice) & 3) << 26) -#define GEN8_MCR_SLICE_MASK GEN8_MCR_SLICE(3) -#define GEN8_MCR_SUBSLICE(subslice) (((subslice) & 3) << 24) -#define GEN8_MCR_SUBSLICE_MASK GEN8_MCR_SUBSLICE(3) -#define GEN11_MCR_SLICE(slice) (((slice) & 0xf) << 27) -#define GEN11_MCR_SLICE_MASK GEN11_MCR_SLICE(0xf) -#define GEN11_MCR_SUBSLICE(subslice) (((subslice) & 0x7) << 24) -#define GEN11_MCR_SUBSLICE_MASK GEN11_MCR_SUBSLICE(0x7) -#define RING_IPEIR(base) _MMIO((base) + 0x64) -#define RING_IPEHR(base) _MMIO((base) + 0x68) -#define RING_EIR(base) _MMIO((base) + 0xb0) -#define RING_EMR(base) _MMIO((base) + 0xb4) -#define RING_ESR(base) _MMIO((base) + 0xb8) -/* - * On GEN4, only the render ring INSTDONE exists and has a different - * layout than the GEN7+ version. - * The GEN2 counterpart of this register is GEN2_INSTDONE. - */ -#define RING_INSTDONE(base) _MMIO((base) + 0x6c) -#define RING_INSTPS(base) _MMIO((base) + 0x70) -#define RING_DMA_FADD(base) _MMIO((base) + 0x78) -#define RING_DMA_FADD_UDW(base) _MMIO((base) + 0x60) /* gen8+ */ -#define RING_INSTPM(base) _MMIO((base) + 0xc0) -#define RING_MI_MODE(base) _MMIO((base) + 0x9c) -#define RING_CMD_BUF_CCTL(base) _MMIO((base) + 0x84) -#define INSTPS _MMIO(0x2070) /* 965+ only */ -#define GEN4_INSTDONE1 _MMIO(0x207c) /* 965+ only, aka INSTDONE_2 on SNB */ -#define ACTHD_I965 _MMIO(0x2074) -#define HWS_PGA _MMIO(0x2080) -#define HWS_ADDRESS_MASK 0xfffff000 -#define HWS_START_ADDRESS_SHIFT 4 -#define PWRCTXA _MMIO(0x2088) /* 965GM+ only */ -#define PWRCTX_EN (1 << 0) -#define IPEIR(base) _MMIO((base) + 0x88) -#define IPEHR(base) _MMIO((base) + 0x8c) -#define GEN2_INSTDONE _MMIO(0x2090) -#define NOPID _MMIO(0x2094) -#define HWSTAM _MMIO(0x2098) -#define DMA_FADD_I8XX(base) _MMIO((base) + 0xd0) -#define RING_BBSTATE(base) _MMIO((base) + 0x110) -#define RING_BB_PPGTT (1 << 5) -#define RING_SBBADDR(base) _MMIO((base) + 0x114) /* hsw+ */ -#define RING_SBBSTATE(base) _MMIO((base) + 0x118) /* hsw+ */ -#define RING_SBBADDR_UDW(base) _MMIO((base) + 0x11c) /* gen8+ */ -#define RING_BBADDR(base) _MMIO((base) + 0x140) -#define RING_BBADDR_UDW(base) _MMIO((base) + 0x168) /* gen8+ */ -#define RING_BB_PER_CTX_PTR(base) _MMIO((base) + 0x1c0) /* gen8+ */ -#define RING_INDIRECT_CTX(base) _MMIO((base) + 0x1c4) /* gen8+ */ -#define RING_INDIRECT_CTX_OFFSET(base) _MMIO((base) + 0x1c8) /* gen8+ */ -#define RING_CTX_TIMESTAMP(base) _MMIO((base) + 0x3a8) /* gen8+ */ - -#define VDBOX_CGCTL3F10(base) _MMIO((base) + 0x3f10) -#define IECPUNIT_CLKGATE_DIS REG_BIT(22) - -#define VDBOX_CGCTL3F18(base) _MMIO((base) + 0x3f18) -#define ALNUNIT_CLKGATE_DIS REG_BIT(13) - -#define ERROR_GEN6 _MMIO(0x40a0) #define GEN7_ERR_INT _MMIO(0x44040) #define ERR_INT_POISON (1 << 31) #define ERR_INT_MMIO_UNCLAIMED (1 << 13) @@ -2819,20 +1005,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define ERR_INT_FIFO_UNDERRUN_A (1 << 0) #define ERR_INT_FIFO_UNDERRUN(pipe) (1 << ((pipe) * 3)) -#define GEN8_FAULT_TLB_DATA0 _MMIO(0x4b10) -#define GEN8_FAULT_TLB_DATA1 _MMIO(0x4b14) -#define GEN12_FAULT_TLB_DATA0 _MMIO(0xceb8) -#define GEN12_FAULT_TLB_DATA1 _MMIO(0xcebc) -#define FAULT_VA_HIGH_BITS (0xf << 0) -#define FAULT_GTT_SEL (1 << 4) - -#define GEN12_GFX_TLB_INV_CR _MMIO(0xced8) -#define GEN12_VD_TLB_INV_CR _MMIO(0xcedc) -#define GEN12_VE_TLB_INV_CR _MMIO(0xcee0) -#define GEN12_BLT_TLB_INV_CR _MMIO(0xcee4) - -#define GEN12_AUX_ERR_DBG _MMIO(0x43f4) - #define FPGA_DBG _MMIO(0x42300) #define FPGA_DBG_RM_NOCLAIM REG_BIT(31) @@ -2860,95 +1032,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define DERRMR_PIPEC_VBLANK (1 << 21) #define DERRMR_PIPEC_HBLANK (1 << 22) - -/* GM45+ chicken bits -- debug workaround bits that may be required - * for various sorts of correct behavior. The top 16 bits of each are - * the enables for writing to the corresponding low bit. - */ -#define _3D_CHICKEN _MMIO(0x2084) -#define _3D_CHICKEN_HIZ_PLANE_DISABLE_MSAA_4X_SNB (1 << 10) -#define _3D_CHICKEN2 _MMIO(0x208c) - -#define FF_SLICE_CHICKEN _MMIO(0x2088) -#define FF_SLICE_CHICKEN_CL_PROVOKING_VERTEX_FIX (1 << 1) - -/* Disables pipelining of read flushes past the SF-WIZ interface. - * Required on all Ironlake steppings according to the B-Spec, but the - * particular danger of not doing so is not specified. - */ -# define _3D_CHICKEN2_WM_READ_PIPELINED (1 << 14) -#define _3D_CHICKEN3 _MMIO(0x2090) -#define _3D_CHICKEN_SF_PROVOKING_VERTEX_FIX (1 << 12) -#define _3D_CHICKEN_SF_DISABLE_OBJEND_CULL (1 << 10) -#define _3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE (1 << 5) -#define _3D_CHICKEN3_SF_DISABLE_FASTCLIP_CULL (1 << 5) -#define _3D_CHICKEN_SDE_LIMIT_FIFO_POLY_DEPTH(x) ((x) << 1) /* gen8+ */ -#define _3D_CHICKEN3_SF_DISABLE_PIPELINED_ATTR_FETCH (1 << 1) /* gen6 */ - -#define MI_MODE _MMIO(0x209c) -# define VS_TIMER_DISPATCH (1 << 6) -# define MI_FLUSH_ENABLE (1 << 12) -# define TGL_NESTED_BB_EN (1 << 12) -# define ASYNC_FLIP_PERF_DISABLE (1 << 14) -# define MODE_IDLE (1 << 9) -# define STOP_RING (1 << 8) - -#define GEN6_GT_MODE _MMIO(0x20d0) -#define GEN7_GT_MODE _MMIO(0x7008) -#define GEN6_WIZ_HASHING(hi, lo) (((hi) << 9) | ((lo) << 7)) -#define GEN6_WIZ_HASHING_8x8 GEN6_WIZ_HASHING(0, 0) -#define GEN6_WIZ_HASHING_8x4 GEN6_WIZ_HASHING(0, 1) -#define GEN6_WIZ_HASHING_16x4 GEN6_WIZ_HASHING(1, 0) -#define GEN6_WIZ_HASHING_MASK GEN6_WIZ_HASHING(1, 1) -#define GEN6_TD_FOUR_ROW_DISPATCH_DISABLE (1 << 5) -#define GEN9_IZ_HASHING_MASK(slice) (0x3 << ((slice) * 2)) -#define GEN9_IZ_HASHING(slice, val) ((val) << ((slice) * 2)) - -/* chicken reg for WaConextSwitchWithConcurrentTLBInvalidate */ -#define GEN9_CSFE_CHICKEN1_RCS _MMIO(0x20D4) -#define GEN9_PREEMPT_GPGPU_SYNC_SWITCH_DISABLE (1 << 2) -#define GEN11_ENABLE_32_PLANE_MODE (1 << 7) - -#define SCCGCTL94DC _MMIO(0x94dc) -#define CG3DDISURB REG_BIT(14) - -#define MLTICTXCTL _MMIO(0xb170) -#define TDONRENDER REG_BIT(2) - -#define L3SQCREG1_CCS0 _MMIO(0xb200) -#define FLUSHALLNONCOH REG_BIT(5) - -/* WaClearTdlStateAckDirtyBits */ -#define GEN8_STATE_ACK _MMIO(0x20F0) -#define GEN9_STATE_ACK_SLICE1 _MMIO(0x20F8) -#define GEN9_STATE_ACK_SLICE2 _MMIO(0x2100) -#define GEN9_STATE_ACK_TDL0 (1 << 12) -#define GEN9_STATE_ACK_TDL1 (1 << 13) -#define GEN9_STATE_ACK_TDL2 (1 << 14) -#define GEN9_STATE_ACK_TDL3 (1 << 15) -#define GEN9_SUBSLICE_TDL_ACK_BITS \ - (GEN9_STATE_ACK_TDL3 | GEN9_STATE_ACK_TDL2 | \ - GEN9_STATE_ACK_TDL1 | GEN9_STATE_ACK_TDL0) - -#define GFX_MODE _MMIO(0x2520) -#define GFX_MODE_GEN7 _MMIO(0x229c) -#define RING_MODE_GEN7(base) _MMIO((base) + 0x29c) -#define GFX_RUN_LIST_ENABLE (1 << 15) -#define GFX_INTERRUPT_STEERING (1 << 14) -#define GFX_TLB_INVALIDATE_EXPLICIT (1 << 13) -#define GFX_SURFACE_FAULT_ENABLE (1 << 12) -#define GFX_REPLAY_MODE (1 << 11) -#define GFX_PSMI_GRANULARITY (1 << 10) -#define GFX_PPGTT_ENABLE (1 << 9) -#define GEN8_GFX_PPGTT_48B (1 << 7) - -#define GFX_FORWARD_VBLANK_MASK (3 << 5) -#define GFX_FORWARD_VBLANK_NEVER (0 << 5) -#define GFX_FORWARD_VBLANK_ALWAYS (1 << 5) -#define GFX_FORWARD_VBLANK_COND (2 << 5) - -#define GEN11_GFX_DISABLE_LEGACY_MODE (1 << 3) - #define VLV_GU_CTL0 _MMIO(VLV_DISPLAY_BASE + 0x2030) #define VLV_GU_CTL1 _MMIO(VLV_DISPLAY_BASE + 0x2034) #define SCPD0 _MMIO(0x209c) /* 915+ only */ @@ -2988,7 +1071,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define INSTPM_FORCE_ORDERING (1 << 7) /* GEN6+ */ #define INSTPM_TLB_INVALIDATE (1 << 9) #define INSTPM_SYNC_FLUSH (1 << 5) -#define ACTHD(base) _MMIO((base) + 0xc8) #define MEM_MODE _MMIO(0x20cc) #define MEM_DISPLAY_B_TRICKLE_FEED_DISABLE (1 << 3) /* 830 only */ #define MEM_DISPLAY_A_TRICKLE_FEED_DISABLE (1 << 2) /* 830/845 only */ @@ -3115,132 +1197,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define MI_AGPBUSY_INT_EN (1 << 1) /* 85x only */ #define MI_AGPBUSY_830_MODE (1 << 0) /* 85x only */ -#define CACHE_MODE_0 _MMIO(0x2120) /* 915+ only */ -#define CM0_PIPELINED_RENDER_FLUSH_DISABLE (1 << 8) -#define CM0_IZ_OPT_DISABLE (1 << 6) -#define CM0_ZR_OPT_DISABLE (1 << 5) -#define CM0_STC_EVICT_DISABLE_LRA_SNB (1 << 5) -#define CM0_DEPTH_EVICT_DISABLE (1 << 4) -#define CM0_COLOR_EVICT_DISABLE (1 << 3) -#define CM0_DEPTH_WRITE_DISABLE (1 << 1) -#define CM0_RC_OP_FLUSH_DISABLE (1 << 0) -#define GFX_FLSH_CNTL _MMIO(0x2170) /* 915+ only */ -#define GFX_FLSH_CNTL_GEN6 _MMIO(0x101008) -#define GFX_FLSH_CNTL_EN (1 << 0) -#define ECOSKPD _MMIO(0x21d0) -#define ECO_CONSTANT_BUFFER_SR_DISABLE REG_BIT(4) -#define ECO_GATING_CX_ONLY (1 << 3) -#define ECO_FLIP_DONE (1 << 0) - -#define CACHE_MODE_0_GEN7 _MMIO(0x7000) /* IVB+ */ -#define RC_OP_FLUSH_ENABLE (1 << 0) -#define HIZ_RAW_STALL_OPT_DISABLE (1 << 2) -#define CACHE_MODE_1 _MMIO(0x7004) /* IVB+ */ -#define PIXEL_SUBSPAN_COLLECT_OPT_DISABLE (1 << 6) -#define GEN8_4x4_STC_OPTIMIZATION_DISABLE (1 << 6) -#define GEN9_PARTIAL_RESOLVE_IN_VC_DISABLE (1 << 1) - -#define GEN6_BLITTER_ECOSKPD _MMIO(0x221d0) -#define GEN6_BLITTER_LOCK_SHIFT 16 -#define GEN6_BLITTER_FBC_NOTIFY (1 << 3) - -#define GEN6_RC_SLEEP_PSMI_CONTROL _MMIO(0x2050) -#define GEN6_PSMI_SLEEP_MSG_DISABLE (1 << 0) -#define GEN12_WAIT_FOR_EVENT_POWER_DOWN_DISABLE REG_BIT(7) -#define GEN8_RC_SEMA_IDLE_MSG_DISABLE (1 << 12) -#define GEN8_FF_DOP_CLOCK_GATE_DISABLE (1 << 10) - -#define GEN6_RCS_PWR_FSM _MMIO(0x22ac) -#define GEN9_RCS_FE_FSM2 _MMIO(0x22a4) - -#define GEN10_CACHE_MODE_SS _MMIO(0xe420) -#define ENABLE_PREFETCH_INTO_IC REG_BIT(3) -#define FLOAT_BLEND_OPTIMIZATION_ENABLE REG_BIT(4) - -/* Fuse readout registers for GT */ -#define HSW_PAVP_FUSE1 _MMIO(0x911C) -#define XEHP_SFC_ENABLE_MASK REG_GENMASK(27, 24) -#define HSW_F1_EU_DIS_MASK REG_GENMASK(17, 16) -#define HSW_F1_EU_DIS_10EUS 0 -#define HSW_F1_EU_DIS_8EUS 1 -#define HSW_F1_EU_DIS_6EUS 2 - -#define CHV_FUSE_GT _MMIO(VLV_DISPLAY_BASE + 0x2168) -#define CHV_FGT_DISABLE_SS0 (1 << 10) -#define CHV_FGT_DISABLE_SS1 (1 << 11) -#define CHV_FGT_EU_DIS_SS0_R0_SHIFT 16 -#define CHV_FGT_EU_DIS_SS0_R0_MASK (0xf << CHV_FGT_EU_DIS_SS0_R0_SHIFT) -#define CHV_FGT_EU_DIS_SS0_R1_SHIFT 20 -#define CHV_FGT_EU_DIS_SS0_R1_MASK (0xf << CHV_FGT_EU_DIS_SS0_R1_SHIFT) -#define CHV_FGT_EU_DIS_SS1_R0_SHIFT 24 -#define CHV_FGT_EU_DIS_SS1_R0_MASK (0xf << CHV_FGT_EU_DIS_SS1_R0_SHIFT) -#define CHV_FGT_EU_DIS_SS1_R1_SHIFT 28 -#define CHV_FGT_EU_DIS_SS1_R1_MASK (0xf << CHV_FGT_EU_DIS_SS1_R1_SHIFT) - -#define GEN8_FUSE2 _MMIO(0x9120) -#define GEN8_F2_SS_DIS_SHIFT 21 -#define GEN8_F2_SS_DIS_MASK (0x7 << GEN8_F2_SS_DIS_SHIFT) -#define GEN8_F2_S_ENA_SHIFT 25 -#define GEN8_F2_S_ENA_MASK (0x7 << GEN8_F2_S_ENA_SHIFT) - -#define GEN9_F2_SS_DIS_SHIFT 20 -#define GEN9_F2_SS_DIS_MASK (0xf << GEN9_F2_SS_DIS_SHIFT) - -#define GEN10_F2_S_ENA_SHIFT 22 -#define GEN10_F2_S_ENA_MASK (0x3f << GEN10_F2_S_ENA_SHIFT) -#define GEN10_F2_SS_DIS_SHIFT 18 -#define GEN10_F2_SS_DIS_MASK (0xf << GEN10_F2_SS_DIS_SHIFT) - -#define GEN10_MIRROR_FUSE3 _MMIO(0x9118) -#define GEN10_L3BANK_PAIR_COUNT 4 -#define GEN10_L3BANK_MASK 0x0F -/* on Xe_HP the same fuses indicates mslices instead of L3 banks */ -#define GEN12_MAX_MSLICES 4 -#define GEN12_MEML3_EN_MASK 0x0F - -#define GEN8_EU_DISABLE0 _MMIO(0x9134) -#define GEN8_EU_DIS0_S0_MASK 0xffffff -#define GEN8_EU_DIS0_S1_SHIFT 24 -#define GEN8_EU_DIS0_S1_MASK (0xff << GEN8_EU_DIS0_S1_SHIFT) - -#define GEN8_EU_DISABLE1 _MMIO(0x9138) -#define GEN8_EU_DIS1_S1_MASK 0xffff -#define GEN8_EU_DIS1_S2_SHIFT 16 -#define GEN8_EU_DIS1_S2_MASK (0xffff << GEN8_EU_DIS1_S2_SHIFT) - -#define GEN8_EU_DISABLE2 _MMIO(0x913c) -#define GEN8_EU_DIS2_S2_MASK 0xff - -#define GEN9_EU_DISABLE(slice) _MMIO(0x9134 + (slice) * 0x4) - -#define GEN10_EU_DISABLE3 _MMIO(0x9140) -#define GEN10_EU_DIS_SS_MASK 0xff - -#define GEN11_GT_VEBOX_VDBOX_DISABLE _MMIO(0x9140) -#define GEN11_GT_VDBOX_DISABLE_MASK 0xff -#define GEN11_GT_VEBOX_DISABLE_SHIFT 16 -#define GEN11_GT_VEBOX_DISABLE_MASK (0x0f << GEN11_GT_VEBOX_DISABLE_SHIFT) - -#define GEN11_EU_DISABLE _MMIO(0x9134) -#define GEN11_EU_DIS_MASK 0xFF - -#define GEN11_GT_SLICE_ENABLE _MMIO(0x9138) -#define GEN11_GT_S_ENA_MASK 0xFF - -#define GEN11_GT_SUBSLICE_DISABLE _MMIO(0x913C) - -#define GEN12_GT_GEOMETRY_DSS_ENABLE _MMIO(0x913C) -#define GEN12_GT_COMPUTE_DSS_ENABLE _MMIO(0x9144) - -#define XEHP_EU_ENABLE _MMIO(0x9134) -#define XEHP_EU_ENA_MASK 0xFF - -#define GEN6_BSD_SLEEP_PSMI_CONTROL _MMIO(0x12050) -#define GEN6_BSD_SLEEP_MSG_DISABLE (1 << 0) -#define GEN6_BSD_SLEEP_FLUSH_DISABLE (1 << 2) -#define GEN6_BSD_SLEEP_INDICATOR (1 << 3) -#define GEN6_BSD_GO_INDICATOR (1 << 4) - /* On modern GEN architectures interrupt control consists of two sets * of registers. The first set pertains to the ring generating the * interrupt. The second control is for the functional block generating the @@ -3398,10 +1354,10 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define FBC_LL_SIZE (1536) /* Framebuffer compression for GM45+ */ -#define DPFC_CB_BASE _MMIO(0x3200) -#define ILK_DPFC_CB_BASE _MMIO(0x43200) -#define DPFC_CONTROL _MMIO(0x3208) -#define ILK_DPFC_CONTROL _MMIO(0x43208) +#define DPFC_CB_BASE _MMIO(0x3200) +#define ILK_DPFC_CB_BASE(fbc_id) _MMIO_PIPE((fbc_id), 0x43200, 0x43240) +#define DPFC_CONTROL _MMIO(0x3208) +#define ILK_DPFC_CONTROL(fbc_id) _MMIO_PIPE((fbc_id), 0x43208, 0x43248) #define DPFC_CTL_EN REG_BIT(31) #define DPFC_CTL_PLANE_MASK_G4X REG_BIT(30) /* g4x-snb */ #define DPFC_CTL_PLANE_G4X(i9xx_plane) REG_FIELD_PREP(DPFC_CTL_PLANE_MASK_G4X, (i9xx_plane)) @@ -3419,28 +1375,28 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define DPFC_CTL_LIMIT_4X REG_FIELD_PREP(DPFC_CTL_LIMIT_MASK, 2) #define DPFC_CTL_FENCENO_MASK REG_GENMASK(3, 0) #define DPFC_CTL_FENCENO(fence) REG_FIELD_PREP(DPFC_CTL_FENCENO_MASK, (fence)) -#define DPFC_RECOMP_CTL _MMIO(0x320c) -#define ILK_DPFC_RECOMP_CTL _MMIO(0x4320c) +#define DPFC_RECOMP_CTL _MMIO(0x320c) +#define ILK_DPFC_RECOMP_CTL(fbc_id) _MMIO_PIPE((fbc_id), 0x4320c, 0x4324c) #define DPFC_RECOMP_STALL_EN REG_BIT(27) #define DPFC_RECOMP_STALL_WM_MASK REG_GENMASK(26, 16) #define DPFC_RECOMP_TIMER_COUNT_MASK REG_GENMASK(5, 0) -#define DPFC_STATUS _MMIO(0x3210) -#define ILK_DPFC_STATUS _MMIO(0x43210) +#define DPFC_STATUS _MMIO(0x3210) +#define ILK_DPFC_STATUS(fbc_id) _MMIO_PIPE((fbc_id), 0x43210, 0x43250) #define DPFC_INVAL_SEG_MASK REG_GENMASK(26, 16) #define DPFC_COMP_SEG_MASK REG_GENMASK(10, 0) -#define DPFC_STATUS2 _MMIO(0x3214) -#define ILK_DPFC_STATUS2 _MMIO(0x43214) +#define DPFC_STATUS2 _MMIO(0x3214) +#define ILK_DPFC_STATUS2(fbc_id) _MMIO_PIPE((fbc_id), 0x43214, 0x43254) #define DPFC_COMP_SEG_MASK_IVB REG_GENMASK(11, 0) -#define DPFC_FENCE_YOFF _MMIO(0x3218) -#define ILK_DPFC_FENCE_YOFF _MMIO(0x43218) -#define DPFC_CHICKEN _MMIO(0x3224) -#define ILK_DPFC_CHICKEN _MMIO(0x43224) +#define DPFC_FENCE_YOFF _MMIO(0x3218) +#define ILK_DPFC_FENCE_YOFF(fbc_id) _MMIO_PIPE((fbc_id), 0x43218, 0x43258) +#define DPFC_CHICKEN _MMIO(0x3224) +#define ILK_DPFC_CHICKEN(fbc_id) _MMIO_PIPE((fbc_id), 0x43224, 0x43264) #define DPFC_HT_MODIFY REG_BIT(31) /* pre-ivb */ #define DPFC_NUKE_ON_ANY_MODIFICATION REG_BIT(23) /* bdw+ */ #define DPFC_CHICKEN_COMP_DUMMY_PIXEL REG_BIT(14) /* glk+ */ #define DPFC_DISABLE_DUMMY0 REG_BIT(8) /* ivb+ */ -#define GLK_FBC_STRIDE _MMIO(0x43228) +#define GLK_FBC_STRIDE(fbc_id) _MMIO_PIPE((fbc_id), 0x43228, 0x43268) #define FBC_STRIDE_OVERRIDE REG_BIT(15) #define FBC_STRIDE_MASK REG_GENMASK(14, 0) #define FBC_STRIDE(x) REG_FIELD_PREP(FBC_STRIDE_MASK, (x)) @@ -3483,9 +1439,9 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define IPS_CTL _MMIO(0x43408) #define IPS_ENABLE (1 << 31) -#define MSG_FBC_REND_STATE _MMIO(0x50380) +#define MSG_FBC_REND_STATE(fbc_id) _MMIO_PIPE((fbc_id), 0x50380, 0x50384) #define FBC_REND_NUKE REG_BIT(2) -#define FBC_REND_CACHE_CLEAN REG_BIT(1) +#define FBC_REND_CACHE_CLEAN REG_BIT(1) /* * GPIO regs @@ -3974,227 +1930,10 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define TSFS_SLOPE_SHIFT 8 #define TSFS_INTR_MASK 0x000000ff -#define CRSTANDVID _MMIO(0x11100) -#define PXVFREQ(fstart) _MMIO(0x11110 + (fstart) * 4) /* P[0-15]VIDFREQ (0x1114c) (Ironlake) */ -#define PXVFREQ_PX_MASK 0x7f000000 -#define PXVFREQ_PX_SHIFT 24 -#define VIDFREQ_BASE _MMIO(0x11110) -#define VIDFREQ1 _MMIO(0x11110) /* VIDFREQ1-4 (0x1111c) (Cantiga) */ -#define VIDFREQ2 _MMIO(0x11114) -#define VIDFREQ3 _MMIO(0x11118) -#define VIDFREQ4 _MMIO(0x1111c) -#define VIDFREQ_P0_MASK 0x1f000000 -#define VIDFREQ_P0_SHIFT 24 -#define VIDFREQ_P0_CSCLK_MASK 0x00f00000 -#define VIDFREQ_P0_CSCLK_SHIFT 20 -#define VIDFREQ_P0_CRCLK_MASK 0x000f0000 -#define VIDFREQ_P0_CRCLK_SHIFT 16 -#define VIDFREQ_P1_MASK 0x00001f00 -#define VIDFREQ_P1_SHIFT 8 -#define VIDFREQ_P1_CSCLK_MASK 0x000000f0 -#define VIDFREQ_P1_CSCLK_SHIFT 4 -#define VIDFREQ_P1_CRCLK_MASK 0x0000000f -#define INTTOEXT_BASE_ILK _MMIO(0x11300) -#define INTTOEXT_BASE _MMIO(0x11120) /* INTTOEXT1-8 (0x1113c) */ -#define INTTOEXT_MAP3_SHIFT 24 -#define INTTOEXT_MAP3_MASK (0x1f << INTTOEXT_MAP3_SHIFT) -#define INTTOEXT_MAP2_SHIFT 16 -#define INTTOEXT_MAP2_MASK (0x1f << INTTOEXT_MAP2_SHIFT) -#define INTTOEXT_MAP1_SHIFT 8 -#define INTTOEXT_MAP1_MASK (0x1f << INTTOEXT_MAP1_SHIFT) -#define INTTOEXT_MAP0_SHIFT 0 -#define INTTOEXT_MAP0_MASK (0x1f << INTTOEXT_MAP0_SHIFT) -#define MEMSWCTL _MMIO(0x11170) /* Ironlake only */ -#define MEMCTL_CMD_MASK 0xe000 -#define MEMCTL_CMD_SHIFT 13 -#define MEMCTL_CMD_RCLK_OFF 0 -#define MEMCTL_CMD_RCLK_ON 1 -#define MEMCTL_CMD_CHFREQ 2 -#define MEMCTL_CMD_CHVID 3 -#define MEMCTL_CMD_VMMOFF 4 -#define MEMCTL_CMD_VMMON 5 -#define MEMCTL_CMD_STS (1 << 12) /* write 1 triggers command, clears - when command complete */ -#define MEMCTL_FREQ_MASK 0x0f00 /* jitter, from 0-15 */ -#define MEMCTL_FREQ_SHIFT 8 -#define MEMCTL_SFCAVM (1 << 7) -#define MEMCTL_TGT_VID_MASK 0x007f -#define MEMIHYST _MMIO(0x1117c) -#define MEMINTREN _MMIO(0x11180) /* 16 bits */ -#define MEMINT_RSEXIT_EN (1 << 8) -#define MEMINT_CX_SUPR_EN (1 << 7) -#define MEMINT_CONT_BUSY_EN (1 << 6) -#define MEMINT_AVG_BUSY_EN (1 << 5) -#define MEMINT_EVAL_CHG_EN (1 << 4) -#define MEMINT_MON_IDLE_EN (1 << 3) -#define MEMINT_UP_EVAL_EN (1 << 2) -#define MEMINT_DOWN_EVAL_EN (1 << 1) -#define MEMINT_SW_CMD_EN (1 << 0) -#define MEMINTRSTR _MMIO(0x11182) /* 16 bits */ -#define MEM_RSEXIT_MASK 0xc000 -#define MEM_RSEXIT_SHIFT 14 -#define MEM_CONT_BUSY_MASK 0x3000 -#define MEM_CONT_BUSY_SHIFT 12 -#define MEM_AVG_BUSY_MASK 0x0c00 -#define MEM_AVG_BUSY_SHIFT 10 -#define MEM_EVAL_CHG_MASK 0x0300 -#define MEM_EVAL_BUSY_SHIFT 8 -#define MEM_MON_IDLE_MASK 0x00c0 -#define MEM_MON_IDLE_SHIFT 6 -#define MEM_UP_EVAL_MASK 0x0030 -#define MEM_UP_EVAL_SHIFT 4 -#define MEM_DOWN_EVAL_MASK 0x000c -#define MEM_DOWN_EVAL_SHIFT 2 -#define MEM_SW_CMD_MASK 0x0003 -#define MEM_INT_STEER_GFX 0 -#define MEM_INT_STEER_CMR 1 -#define MEM_INT_STEER_SMI 2 -#define MEM_INT_STEER_SCI 3 -#define MEMINTRSTS _MMIO(0x11184) -#define MEMINT_RSEXIT (1 << 7) -#define MEMINT_CONT_BUSY (1 << 6) -#define MEMINT_AVG_BUSY (1 << 5) -#define MEMINT_EVAL_CHG (1 << 4) -#define MEMINT_MON_IDLE (1 << 3) -#define MEMINT_UP_EVAL (1 << 2) -#define MEMINT_DOWN_EVAL (1 << 1) -#define MEMINT_SW_CMD (1 << 0) -#define MEMMODECTL _MMIO(0x11190) -#define MEMMODE_BOOST_EN (1 << 31) -#define MEMMODE_BOOST_FREQ_MASK 0x0f000000 /* jitter for boost, 0-15 */ -#define MEMMODE_BOOST_FREQ_SHIFT 24 -#define MEMMODE_IDLE_MODE_MASK 0x00030000 -#define MEMMODE_IDLE_MODE_SHIFT 16 -#define MEMMODE_IDLE_MODE_EVAL 0 -#define MEMMODE_IDLE_MODE_CONT 1 -#define MEMMODE_HWIDLE_EN (1 << 15) -#define MEMMODE_SWMODE_EN (1 << 14) -#define MEMMODE_RCLK_GATE (1 << 13) -#define MEMMODE_HW_UPDATE (1 << 12) -#define MEMMODE_FSTART_MASK 0x00000f00 /* starting jitter, 0-15 */ -#define MEMMODE_FSTART_SHIFT 8 -#define MEMMODE_FMAX_MASK 0x000000f0 /* max jitter, 0-15 */ -#define MEMMODE_FMAX_SHIFT 4 -#define MEMMODE_FMIN_MASK 0x0000000f /* min jitter, 0-15 */ -#define RCBMAXAVG _MMIO(0x1119c) -#define MEMSWCTL2 _MMIO(0x1119e) /* Cantiga only */ -#define SWMEMCMD_RENDER_OFF (0 << 13) -#define SWMEMCMD_RENDER_ON (1 << 13) -#define SWMEMCMD_SWFREQ (2 << 13) -#define SWMEMCMD_TARVID (3 << 13) -#define SWMEMCMD_VRM_OFF (4 << 13) -#define SWMEMCMD_VRM_ON (5 << 13) -#define CMDSTS (1 << 12) -#define SFCAVM (1 << 11) -#define SWFREQ_MASK 0x0380 /* P0-7 */ -#define SWFREQ_SHIFT 7 -#define TARVID_MASK 0x001f -#define MEMSTAT_CTG _MMIO(0x111a0) -#define RCBMINAVG _MMIO(0x111a0) -#define RCUPEI _MMIO(0x111b0) -#define RCDNEI _MMIO(0x111b4) -#define RSTDBYCTL _MMIO(0x111b8) -#define RS1EN (1 << 31) -#define RS2EN (1 << 30) -#define RS3EN (1 << 29) -#define D3RS3EN (1 << 28) /* Display D3 imlies RS3 */ -#define SWPROMORSX (1 << 27) /* RSx promotion timers ignored */ -#define RCWAKERW (1 << 26) /* Resetwarn from PCH causes wakeup */ -#define DPRSLPVREN (1 << 25) /* Fast voltage ramp enable */ -#define GFXTGHYST (1 << 24) /* Hysteresis to allow trunk gating */ -#define RCX_SW_EXIT (1 << 23) /* Leave RSx and prevent re-entry */ -#define RSX_STATUS_MASK (7 << 20) -#define RSX_STATUS_ON (0 << 20) -#define RSX_STATUS_RC1 (1 << 20) -#define RSX_STATUS_RC1E (2 << 20) -#define RSX_STATUS_RS1 (3 << 20) -#define RSX_STATUS_RS2 (4 << 20) /* aka rc6 */ -#define RSX_STATUS_RSVD (5 << 20) /* deep rc6 unsupported on ilk */ -#define RSX_STATUS_RS3 (6 << 20) /* rs3 unsupported on ilk */ -#define RSX_STATUS_RSVD2 (7 << 20) -#define UWRCRSXE (1 << 19) /* wake counter limit prevents rsx */ -#define RSCRP (1 << 18) /* rs requests control on rs1/2 reqs */ -#define JRSC (1 << 17) /* rsx coupled to cpu c-state */ -#define RS2INC0 (1 << 16) /* allow rs2 in cpu c0 */ -#define RS1CONTSAV_MASK (3 << 14) -#define RS1CONTSAV_NO_RS1 (0 << 14) /* rs1 doesn't save/restore context */ -#define RS1CONTSAV_RSVD (1 << 14) -#define RS1CONTSAV_SAVE_RS1 (2 << 14) /* rs1 saves context */ -#define RS1CONTSAV_FULL_RS1 (3 << 14) /* rs1 saves and restores context */ -#define NORMSLEXLAT_MASK (3 << 12) -#define SLOW_RS123 (0 << 12) -#define SLOW_RS23 (1 << 12) -#define SLOW_RS3 (2 << 12) -#define NORMAL_RS123 (3 << 12) -#define RCMODE_TIMEOUT (1 << 11) /* 0 is eval interval method */ -#define IMPROMOEN (1 << 10) /* promo is immediate or delayed until next idle interval (only for timeout method above) */ -#define RCENTSYNC (1 << 9) /* rs coupled to cpu c-state (3/6/7) */ -#define STATELOCK (1 << 7) /* locked to rs_cstate if 0 */ -#define RS_CSTATE_MASK (3 << 4) -#define RS_CSTATE_C367_RS1 (0 << 4) -#define RS_CSTATE_C36_RS1_C7_RS2 (1 << 4) -#define RS_CSTATE_RSVD (2 << 4) -#define RS_CSTATE_C367_RS2 (3 << 4) -#define REDSAVES (1 << 3) /* no context save if was idle during rs0 */ -#define REDRESTORES (1 << 2) /* no restore if was idle during rs0 */ -#define VIDCTL _MMIO(0x111c0) -#define VIDSTS _MMIO(0x111c8) -#define VIDSTART _MMIO(0x111cc) /* 8 bits */ -#define MEMSTAT_ILK _MMIO(0x111f8) -#define MEMSTAT_VID_MASK 0x7f00 -#define MEMSTAT_VID_SHIFT 8 -#define MEMSTAT_PSTATE_MASK 0x00f8 -#define MEMSTAT_PSTATE_SHIFT 3 -#define MEMSTAT_MON_ACTV (1 << 2) -#define MEMSTAT_SRC_CTL_MASK 0x0003 -#define MEMSTAT_SRC_CTL_CORE 0 -#define MEMSTAT_SRC_CTL_TRB 1 -#define MEMSTAT_SRC_CTL_THM 2 -#define MEMSTAT_SRC_CTL_STDBY 3 -#define RCPREVBSYTUPAVG _MMIO(0x113b8) -#define RCPREVBSYTDNAVG _MMIO(0x113bc) -#define PMMISC _MMIO(0x11214) -#define MCPPCE_EN (1 << 0) /* enable PM_MSG from PCH->MPC */ -#define SDEW _MMIO(0x1124c) -#define CSIEW0 _MMIO(0x11250) -#define CSIEW1 _MMIO(0x11254) -#define CSIEW2 _MMIO(0x11258) -#define PEW(i) _MMIO(0x1125c + (i) * 4) /* 5 registers */ -#define DEW(i) _MMIO(0x11270 + (i) * 4) /* 3 registers */ -#define MCHAFE _MMIO(0x112c0) -#define CSIEC _MMIO(0x112e0) -#define DMIEC _MMIO(0x112e4) -#define DDREC _MMIO(0x112e8) -#define PEG0EC _MMIO(0x112ec) -#define PEG1EC _MMIO(0x112f0) -#define GFXEC _MMIO(0x112f4) -#define RPPREVBSYTUPAVG _MMIO(0x113b8) -#define RPPREVBSYTDNAVG _MMIO(0x113bc) -#define ECR _MMIO(0x11600) -#define ECR_GPFE (1 << 31) -#define ECR_IMONE (1 << 30) -#define ECR_CAP_MASK 0x0000001f /* Event range, 0-31 */ -#define OGW0 _MMIO(0x11608) -#define OGW1 _MMIO(0x1160c) -#define EG0 _MMIO(0x11610) -#define EG1 _MMIO(0x11614) -#define EG2 _MMIO(0x11618) -#define EG3 _MMIO(0x1161c) -#define EG4 _MMIO(0x11620) -#define EG5 _MMIO(0x11624) -#define EG6 _MMIO(0x11628) -#define EG7 _MMIO(0x1162c) -#define PXW(i) _MMIO(0x11664 + (i) * 4) /* 4 registers */ -#define PXWL(i) _MMIO(0x11680 + (i) * 8) /* 8 registers */ -#define LCFUSE02 _MMIO(0x116c0) -#define LCFUSE_HIV_MASK 0x000000ff #define CSIPLL0 _MMIO(0x12c10) #define DDRMPLL1 _MMIO(0X12c20) #define PEG_BAND_GAP_DATA _MMIO(0x14d68) -#define GEN6_GT_THREAD_STATUS_REG _MMIO(0x13805c) -#define GEN6_GT_THREAD_STATUS_CORE_MASK 0x7 - #define GEN6_GT_PERF_STATUS _MMIO(MCHBAR_MIRROR_BASE_SNB + 0x5948) #define BXT_GT_PERF_STATUS _MMIO(MCHBAR_MIRROR_BASE_SNB + 0x7070) #define GEN6_RP_STATE_LIMITS _MMIO(MCHBAR_MIRROR_BASE_SNB + 0x5994) @@ -4206,81 +1945,6 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg) #define GEN9_RP_STATE_LIMITS _MMIO(0x138148) #define XEHPSDV_RP_STATE_CAP _MMIO(0x250014) -/* - * Logical Context regs - */ -#define CCID(base) _MMIO((base) + 0x180) -#define CCID_EN BIT(0) -#define CCID_EXTENDED_STATE_RESTORE BIT(2) -#define CCID_EXTENDED_STATE_SAVE BIT(3) -/* - * Notes on SNB/IVB/VLV context size: - * - Power context is saved elsewhere (LLC or stolen) - * - Ring/execlist context is saved on SNB, not on IVB - * - Extended context size already includes render context size - * - We always need to follow the extended context size. - * SNB BSpec has comments indicating that we should use the - * render context size instead if execlists are disabled, but - * based on empirical testing that's just nonsense. - * - Pipelined/VF state is saved on SNB/IVB respectively - * - GT1 size just indicates how much of render context - * doesn't need saving on GT1 - */ -#define CXT_SIZE _MMIO(0x21a0) -#define GEN6_CXT_POWER_SIZE(cxt_reg) (((cxt_reg) >> 24) & 0x3f) -#define GEN6_CXT_RING_SIZE(cxt_reg) (((cxt_reg) >> 18) & 0x3f) -#define GEN6_CXT_RENDER_SIZE(cxt_reg) (((cxt_reg) >> 12) & 0x3f) -#define GEN6_CXT_EXTENDED_SIZE(cxt_reg) (((cxt_reg) >> 6) & 0x3f) -#define GEN6_CXT_PIPELINE_SIZE(cxt_reg) (((cxt_reg) >> 0) & 0x3f) -#define GEN6_CXT_TOTAL_SIZE(cxt_reg) (GEN6_CXT_RING_SIZE(cxt_reg) + \ - GEN6_CXT_EXTENDED_SIZE(cxt_reg) + \ - GEN6_CXT_PIPELINE_SIZE(cxt_reg)) -#define GEN7_CXT_SIZE _MMIO(0x21a8) -#define GEN7_CXT_POWER_SIZE(ctx_reg) (((ctx_reg) >> 25) & 0x7f) -#define GEN7_CXT_RING_SIZE(ctx_reg) (((ctx_reg) >> 22) & 0x7) -#define GEN7_CXT_RENDER_SIZE(ctx_reg) (((ctx_reg) >> 16) & 0x3f) -#define GEN7_CXT_EXTENDED_SIZE(ctx_reg) (((ctx_reg) >> 9) & 0x7f) -#define GEN7_CXT_GT1_SIZE(ctx_reg) (((ctx_reg) >> 6) & 0x7) -#define GEN7_CXT_VFSTATE_SIZE(ctx_reg) (((ctx_reg) >> 0) & 0x3f) -#define GEN7_CXT_TOTAL_SIZE(ctx_reg) (GEN7_CXT_EXTENDED_SIZE(ctx_reg) + \ - GEN7_CXT_VFSTATE_SIZE(ctx_reg)) - -enum { - INTEL_ADVANCED_CONTEXT = 0, - INTEL_LEGACY_32B_CONTEXT, - INTEL_ADVANCED_AD_CONTEXT, - INTEL_LEGACY_64B_CONTEXT -}; - -enum { - FAULT_AND_HANG = 0, - FAULT_AND_HALT, /* Debug only */ - FAULT_AND_STREAM, - FAULT_AND_CONTINUE /* Unsupported */ -}; - -#define CTX_GTT_ADDRESS_MASK GENMASK(31, 12) -#define GEN8_CTX_VALID (1 << 0) -#define GEN8_CTX_FORCE_PD_RESTORE (1 << 1) -#define GEN8_CTX_FORCE_RESTORE (1 << 2) -#define GEN8_CTX_L3LLC_COHERENT (1 << 5) -#define GEN8_CTX_PRIVILEGE (1 << 8) -#define GEN8_CTX_ADDRESSING_MODE_SHIFT 3 - -#define GEN8_CTX_ID_SHIFT 32 -#define GEN8_CTX_ID_WIDTH 21 -#define GEN11_SW_CTX_ID_SHIFT 37 -#define GEN11_SW_CTX_ID_WIDTH 11 -#define GEN11_ENGINE_CLASS_SHIFT 61 -#define GEN11_ENGINE_CLASS_WIDTH 3 -#define GEN11_ENGINE_INSTANCE_SHIFT 48 -#define GEN11_ENGINE_INSTANCE_WIDTH 6 - -#define XEHP_SW_CTX_ID_SHIFT 39 -#define XEHP_SW_CTX_ID_WIDTH 16 -#define XEHP_SW_COUNTER_SHIFT 58 -#define XEHP_SW_COUNTER_WIDTH 6 - #define CHV_CLK_CTL1 _MMIO(0x101100) #define VLV_CLK_CTL2 _MMIO(0x101104) #define CLK_CTL2_CZCOUNT_30NS_SHIFT 28 @@ -4332,75 +1996,6 @@ enum { _MMIO_PIPE(pipe, _CLKGATE_DIS_PSL_A, _CLKGATE_DIS_PSL_B) /* - * GEN10 clock gating regs - */ - -#define UNSLCGCTL9440 _MMIO(0x9440) -#define GAMTLBOACS_CLKGATE_DIS REG_BIT(28) -#define GAMTLBVDBOX5_CLKGATE_DIS REG_BIT(27) -#define GAMTLBVDBOX6_CLKGATE_DIS REG_BIT(26) -#define GAMTLBVDBOX3_CLKGATE_DIS REG_BIT(24) -#define GAMTLBVDBOX4_CLKGATE_DIS REG_BIT(23) -#define GAMTLBVDBOX7_CLKGATE_DIS REG_BIT(22) -#define GAMTLBVDBOX2_CLKGATE_DIS REG_BIT(21) -#define GAMTLBVDBOX0_CLKGATE_DIS REG_BIT(17) -#define GAMTLBKCR_CLKGATE_DIS REG_BIT(16) -#define GAMTLBGUC_CLKGATE_DIS REG_BIT(15) -#define GAMTLBBLT_CLKGATE_DIS REG_BIT(14) -#define GAMTLBVDBOX1_CLKGATE_DIS REG_BIT(6) - -#define UNSLCGCTL9444 _MMIO(0x9444) -#define GAMTLBGFXA0_CLKGATE_DIS REG_BIT(30) -#define GAMTLBGFXA1_CLKGATE_DIS REG_BIT(29) -#define GAMTLBCOMPA0_CLKGATE_DIS REG_BIT(28) -#define GAMTLBCOMPA1_CLKGATE_DIS REG_BIT(27) -#define GAMTLBCOMPB0_CLKGATE_DIS REG_BIT(26) -#define GAMTLBCOMPB1_CLKGATE_DIS REG_BIT(25) -#define GAMTLBCOMPC0_CLKGATE_DIS REG_BIT(24) -#define GAMTLBCOMPC1_CLKGATE_DIS REG_BIT(23) -#define GAMTLBCOMPD0_CLKGATE_DIS REG_BIT(22) -#define GAMTLBCOMPD1_CLKGATE_DIS REG_BIT(21) -#define GAMTLBMERT_CLKGATE_DIS REG_BIT(20) -#define GAMTLBVEBOX3_CLKGATE_DIS REG_BIT(19) -#define GAMTLBVEBOX2_CLKGATE_DIS REG_BIT(18) -#define GAMTLBVEBOX1_CLKGATE_DIS REG_BIT(17) -#define GAMTLBVEBOX0_CLKGATE_DIS REG_BIT(16) -#define LTCDD_CLKGATE_DIS REG_BIT(10) - -#define SLICE_UNIT_LEVEL_CLKGATE _MMIO(0x94d4) -#define SARBUNIT_CLKGATE_DIS (1 << 5) -#define RCCUNIT_CLKGATE_DIS (1 << 7) -#define MSCUNIT_CLKGATE_DIS (1 << 10) -#define NODEDSS_CLKGATE_DIS REG_BIT(12) -#define L3_CLKGATE_DIS REG_BIT(16) -#define L3_CR2X_CLKGATE_DIS REG_BIT(17) - -#define SUBSLICE_UNIT_LEVEL_CLKGATE _MMIO(0x9524) -#define DSS_ROUTER_CLKGATE_DIS REG_BIT(28) -#define GWUNIT_CLKGATE_DIS REG_BIT(16) - -#define SUBSLICE_UNIT_LEVEL_CLKGATE2 _MMIO(0x9528) -#define CPSSUNIT_CLKGATE_DIS REG_BIT(9) - -#define SSMCGCTL9530 _MMIO(0x9530) -#define RTFUNIT_CLKGATE_DIS REG_BIT(18) - -#define UNSLICE_UNIT_LEVEL_CLKGATE _MMIO(0x9434) -#define VFUNIT_CLKGATE_DIS REG_BIT(20) -#define TSGUNIT_CLKGATE_DIS REG_BIT(17) /* XEHPSDV */ -#define CG3DDISCFEG_CLKGATE_DIS REG_BIT(17) /* DG2 */ -#define GAMEDIA_CLKGATE_DIS REG_BIT(11) -#define HSUNIT_CLKGATE_DIS REG_BIT(8) -#define VSUNIT_CLKGATE_DIS REG_BIT(3) - -#define UNSLICE_UNIT_LEVEL_CLKGATE2 _MMIO(0x94e4) -#define VSUNIT_CLKGATE_DIS_TGL REG_BIT(19) -#define PSDUNIT_CLKGATE_DIS REG_BIT(5) - -#define INF_UNIT_LEVEL_CLKGATE _MMIO(0x9560) -#define CGPSF_CLKGATE_DIS (1 << 3) - -/* * Display engine regs */ @@ -4494,6 +2089,10 @@ enum { #define _VSYNC_A 0x60014 #define _EXITLINE_A 0x60018 #define _PIPEASRC 0x6001c +#define PIPESRC_WIDTH_MASK REG_GENMASK(31, 16) +#define PIPESRC_WIDTH(w) REG_FIELD_PREP(PIPESRC_WIDTH_MASK, (w)) +#define PIPESRC_HEIGHT_MASK REG_GENMASK(15, 0) +#define PIPESRC_HEIGHT(h) REG_FIELD_PREP(PIPESRC_HEIGHT_MASK, (h)) #define _BCLRPAT_A 0x60020 #define _VSYNCSHIFT_A 0x60028 #define _PIPE_MULT_A 0x6002c @@ -6144,16 +3743,14 @@ enum { #define _PIPEB_DATA_M_G4X 0x71050 /* Transfer unit size for display port - 1, default is 0x3f (for TU size 64) */ -#define TU_SIZE(x) (((x) - 1) << 25) /* default size 64 */ -#define TU_SIZE_SHIFT 25 -#define TU_SIZE_MASK (0x3f << 25) +#define TU_SIZE_MASK REG_GENMASK(30, 25) +#define TU_SIZE(x) REG_FIELD_PREP(TU_SIZE_MASK, (x) - 1) /* default size 64 */ -#define DATA_LINK_M_N_MASK (0xffffff) +#define DATA_LINK_M_N_MASK REG_GENMASK(23, 0) #define DATA_LINK_N_MAX (0x800000) #define _PIPEA_DATA_N_G4X 0x70054 #define _PIPEB_DATA_N_G4X 0x71054 -#define PIPE_GMCH_DATA_N_MASK (0xffffff) /* * Computing Link M and N values for the Display Port link @@ -6168,11 +3765,8 @@ enum { #define _PIPEA_LINK_M_G4X 0x70060 #define _PIPEB_LINK_M_G4X 0x71060 -#define PIPEA_DP_LINK_M_MASK (0xffffff) - #define _PIPEA_LINK_N_G4X 0x70064 #define _PIPEB_LINK_N_G4X 0x71064 -#define PIPEA_DP_LINK_N_MASK (0xffffff) #define PIPE_DATA_M_G4X(pipe) _MMIO_PIPE(pipe, _PIPEA_DATA_M_G4X, _PIPEB_DATA_M_G4X) #define PIPE_DATA_N_G4X(pipe) _MMIO_PIPE(pipe, _PIPEA_DATA_N_G4X, _PIPEB_DATA_N_G4X) @@ -6183,65 +3777,61 @@ enum { /* Pipe A */ #define _PIPEADSL 0x70000 -#define DSL_LINEMASK_GEN2 0x00000fff -#define DSL_LINEMASK_GEN3 0x00001fff +#define PIPEDSL_CURR_FIELD REG_BIT(31) /* ctg+ */ +#define PIPEDSL_LINE_MASK REG_GENMASK(19, 0) #define _PIPEACONF 0x70008 -#define PIPECONF_ENABLE (1 << 31) -#define PIPECONF_DISABLE 0 -#define PIPECONF_DOUBLE_WIDE (1 << 30) -#define I965_PIPECONF_ACTIVE (1 << 30) -#define PIPECONF_DSI_PLL_LOCKED (1 << 29) /* vlv & pipe A only */ -#define PIPECONF_FRAME_START_DELAY_MASK (3 << 27) /* pre-hsw */ -#define PIPECONF_FRAME_START_DELAY(x) ((x) << 27) /* pre-hsw: 0-3 */ -#define PIPECONF_SINGLE_WIDE 0 -#define PIPECONF_PIPE_UNLOCKED 0 -#define PIPECONF_PIPE_LOCKED (1 << 25) -#define PIPECONF_FORCE_BORDER (1 << 25) -#define PIPECONF_GAMMA_MODE_MASK_I9XX (1 << 24) /* gmch */ -#define PIPECONF_GAMMA_MODE_MASK_ILK (3 << 24) /* ilk-ivb */ -#define PIPECONF_GAMMA_MODE_8BIT (0 << 24) /* gmch,ilk-ivb */ -#define PIPECONF_GAMMA_MODE_10BIT (1 << 24) /* gmch,ilk-ivb */ -#define PIPECONF_GAMMA_MODE_12BIT (2 << 24) /* ilk-ivb */ -#define PIPECONF_GAMMA_MODE_SPLIT (3 << 24) /* ivb */ -#define PIPECONF_GAMMA_MODE(x) ((x) << 24) /* pass in GAMMA_MODE_MODE_* */ -#define PIPECONF_GAMMA_MODE_SHIFT 24 -#define PIPECONF_INTERLACE_MASK (7 << 21) -#define PIPECONF_INTERLACE_MASK_HSW (3 << 21) -/* Note that pre-gen3 does not support interlaced display directly. Panel - * fitting must be disabled on pre-ilk for interlaced. */ -#define PIPECONF_PROGRESSIVE (0 << 21) -#define PIPECONF_INTERLACE_W_SYNC_SHIFT_PANEL (4 << 21) /* gen4 only */ -#define PIPECONF_INTERLACE_W_SYNC_SHIFT (5 << 21) /* gen4 only */ -#define PIPECONF_INTERLACE_W_FIELD_INDICATION (6 << 21) -#define PIPECONF_INTERLACE_FIELD_0_ONLY (7 << 21) /* gen3 only */ -/* Ironlake and later have a complete new set of values for interlaced. PFIT - * means panel fitter required, PF means progressive fetch, DBL means power - * saving pixel doubling. */ -#define PIPECONF_PFIT_PF_INTERLACED_ILK (1 << 21) -#define PIPECONF_INTERLACED_ILK (3 << 21) -#define PIPECONF_INTERLACED_DBL_ILK (4 << 21) /* ilk/snb only */ -#define PIPECONF_PFIT_PF_INTERLACED_DBL_ILK (5 << 21) /* ilk/snb only */ -#define PIPECONF_INTERLACE_MODE_MASK (7 << 21) -#define PIPECONF_EDP_RR_MODE_SWITCH (1 << 20) -#define PIPECONF_CXSR_DOWNCLOCK (1 << 16) -#define PIPECONF_EDP_RR_MODE_SWITCH_VLV (1 << 14) -#define PIPECONF_COLOR_RANGE_SELECT (1 << 13) -#define PIPECONF_OUTPUT_COLORSPACE_MASK (3 << 11) /* ilk-ivb */ -#define PIPECONF_OUTPUT_COLORSPACE_RGB (0 << 11) /* ilk-ivb */ -#define PIPECONF_OUTPUT_COLORSPACE_YUV601 (1 << 11) /* ilk-ivb */ -#define PIPECONF_OUTPUT_COLORSPACE_YUV709 (2 << 11) /* ilk-ivb */ -#define PIPECONF_OUTPUT_COLORSPACE_YUV_HSW (1 << 11) /* hsw only */ -#define PIPECONF_BPC_MASK (0x7 << 5) -#define PIPECONF_8BPC (0 << 5) -#define PIPECONF_10BPC (1 << 5) -#define PIPECONF_6BPC (2 << 5) -#define PIPECONF_12BPC (3 << 5) -#define PIPECONF_DITHER_EN (1 << 4) -#define PIPECONF_DITHER_TYPE_MASK (0x0000000c) -#define PIPECONF_DITHER_TYPE_SP (0 << 2) -#define PIPECONF_DITHER_TYPE_ST1 (1 << 2) -#define PIPECONF_DITHER_TYPE_ST2 (2 << 2) -#define PIPECONF_DITHER_TYPE_TEMP (3 << 2) +#define PIPECONF_ENABLE REG_BIT(31) +#define PIPECONF_DOUBLE_WIDE REG_BIT(30) /* pre-i965 */ +#define PIPECONF_STATE_ENABLE REG_BIT(30) /* i965+ */ +#define PIPECONF_DSI_PLL_LOCKED REG_BIT(29) /* vlv & pipe A only */ +#define PIPECONF_FRAME_START_DELAY_MASK REG_GENMASK(28, 27) /* pre-hsw */ +#define PIPECONF_FRAME_START_DELAY(x) REG_FIELD_PREP(PIPECONF_FRAME_START_DELAY_MASK, (x)) /* pre-hsw: 0-3 */ +#define PIPECONF_PIPE_LOCKED REG_BIT(25) +#define PIPECONF_FORCE_BORDER REG_BIT(25) +#define PIPECONF_GAMMA_MODE_MASK_I9XX REG_BIT(24) /* gmch */ +#define PIPECONF_GAMMA_MODE_MASK_ILK REG_GENMASK(25, 24) /* ilk-ivb */ +#define PIPECONF_GAMMA_MODE_8BIT REG_FIELD_PREP(PIPECONF_GAMMA_MODE_MASK, 0) +#define PIPECONF_GAMMA_MODE_10BIT REG_FIELD_PREP(PIPECONF_GAMMA_MODE_MASK, 1) +#define PIPECONF_GAMMA_MODE_12BIT REG_FIELD_PREP(PIPECONF_GAMMA_MODE_MASK_ILK, 2) /* ilk-ivb */ +#define PIPECONF_GAMMA_MODE_SPLIT REG_FIELD_PREP(PIPECONF_GAMMA_MODE_MASK_ILK, 3) /* ivb */ +#define PIPECONF_GAMMA_MODE(x) REG_FIELD_PREP(PIPECONF_GAMMA_MODE_MASK_ILK, (x)) /* pass in GAMMA_MODE_MODE_* */ +#define PIPECONF_INTERLACE_MASK REG_GENMASK(23, 21) /* gen3+ */ +#define PIPECONF_INTERLACE_PROGRESSIVE REG_FIELD_PREP(PIPECONF_INTERLACE_MASK, 0) +#define PIPECONF_INTERLACE_W_SYNC_SHIFT_PANEL REG_FIELD_PREP(PIPECONF_INTERLACE_MASK, 4) /* gen4 only */ +#define PIPECONF_INTERLACE_W_SYNC_SHIFT REG_FIELD_PREP(PIPECONF_INTERLACE_MASK, 5) /* gen4 only */ +#define PIPECONF_INTERLACE_W_FIELD_INDICATION REG_FIELD_PREP(PIPECONF_INTERLACE_MASK, 6) +#define PIPECONF_INTERLACE_FIELD_0_ONLY REG_FIELD_PREP(PIPECONF_INTERLACE_MASK, 7) /* gen3 only */ +/* + * ilk+: PF/D=progressive fetch/display, IF/D=interlaced fetch/display, + * DBL=power saving pixel doubling, PF-ID* requires panel fitter + */ +#define PIPECONF_INTERLACE_MASK_ILK REG_GENMASK(23, 21) /* ilk+ */ +#define PIPECONF_INTERLACE_MASK_HSW REG_GENMASK(22, 21) /* hsw+ */ +#define PIPECONF_INTERLACE_PF_PD_ILK REG_FIELD_PREP(PIPECONF_INTERLACE_MASK_ILK, 0) +#define PIPECONF_INTERLACE_PF_ID_ILK REG_FIELD_PREP(PIPECONF_INTERLACE_MASK_ILK, 1) +#define PIPECONF_INTERLACE_IF_ID_ILK REG_FIELD_PREP(PIPECONF_INTERLACE_MASK_ILK, 3) +#define PIPECONF_INTERLACE_IF_ID_DBL_ILK REG_FIELD_PREP(PIPECONF_INTERLACE_MASK_ILK, 4) /* ilk/snb only */ +#define PIPECONF_INTERLACE_PF_ID_DBL_ILK REG_FIELD_PREP(PIPECONF_INTERLACE_MASK_ILK, 5) /* ilk/snb only */ +#define PIPECONF_EDP_RR_MODE_SWITCH REG_BIT(20) +#define PIPECONF_CXSR_DOWNCLOCK REG_BIT(16) +#define PIPECONF_EDP_RR_MODE_SWITCH_VLV REG_BIT(14) +#define PIPECONF_COLOR_RANGE_SELECT REG_BIT(13) +#define PIPECONF_OUTPUT_COLORSPACE_MASK REG_GENMASK(12, 11) /* ilk-ivb */ +#define PIPECONF_OUTPUT_COLORSPACE_RGB REG_FIELD_PREP(PIPECONF_OUTPUT_COLORSPACE_MASK, 0) /* ilk-ivb */ +#define PIPECONF_OUTPUT_COLORSPACE_YUV601 REG_FIELD_PREP(PIPECONF_OUTPUT_COLORSPACE_MASK, 1) /* ilk-ivb */ +#define PIPECONF_OUTPUT_COLORSPACE_YUV709 REG_FIELD_PREP(PIPECONF_OUTPUT_COLORSPACE_MASK, 2) /* ilk-ivb */ +#define PIPECONF_OUTPUT_COLORSPACE_YUV_HSW REG_BIT(11) /* hsw only */ +#define PIPECONF_BPC_MASK REG_GENMASK(7, 5) /* ctg-ivb */ +#define PIPECONF_BPC_8 REG_FIELD_PREP(PIPECONF_BPC_MASK, 0) +#define PIPECONF_BPC_10 REG_FIELD_PREP(PIPECONF_BPC_MASK, 1) +#define PIPECONF_BPC_6 REG_FIELD_PREP(PIPECONF_BPC_MASK, 2) +#define PIPECONF_BPC_12 REG_FIELD_PREP(PIPECONF_BPC_MASK, 3) +#define PIPECONF_DITHER_EN REG_BIT(4) +#define PIPECONF_DITHER_TYPE_MASK REG_GENMASK(3, 2) +#define PIPECONF_DITHER_TYPE_SP REG_FIELD_PREP(PIPECONF_DITHER_TYPE_MASK, 0) +#define PIPECONF_DITHER_TYPE_ST1 REG_FIELD_PREP(PIPECONF_DITHER_TYPE_MASK, 1) +#define PIPECONF_DITHER_TYPE_ST2 REG_FIELD_PREP(PIPECONF_DITHER_TYPE_MASK, 2) +#define PIPECONF_DITHER_TYPE_TEMP REG_FIELD_PREP(PIPECONF_DITHER_TYPE_MASK, 3) #define _PIPEASTAT 0x70024 #define PIPE_FIFO_UNDERRUN_STATUS (1UL << 31) #define SPRITE1_FLIP_DONE_INT_EN_VLV (1UL << 30) @@ -6326,38 +3916,41 @@ enum { #define _PIPE_MISC_A 0x70030 #define _PIPE_MISC_B 0x71030 -#define PIPEMISC_YUV420_ENABLE (1 << 27) /* glk+ */ -#define PIPEMISC_YUV420_MODE_FULL_BLEND (1 << 26) /* glk+ */ -#define PIPEMISC_HDR_MODE_PRECISION (1 << 23) /* icl+ */ -#define PIPEMISC_OUTPUT_COLORSPACE_YUV (1 << 11) -#define PIPEMISC_PIXEL_ROUNDING_TRUNC REG_BIT(8) /* tgl+ */ +#define PIPEMISC_YUV420_ENABLE REG_BIT(27) /* glk+ */ +#define PIPEMISC_YUV420_MODE_FULL_BLEND REG_BIT(26) /* glk+ */ +#define PIPEMISC_HDR_MODE_PRECISION REG_BIT(23) /* icl+ */ +#define PIPEMISC_OUTPUT_COLORSPACE_YUV REG_BIT(11) +#define PIPEMISC_PIXEL_ROUNDING_TRUNC REG_BIT(8) /* tgl+ */ /* * For Display < 13, Bits 5-7 of PIPE MISC represent DITHER BPC with * valid values of: 6, 8, 10 BPC. * ADLP+, the bits 5-7 represent PORT OUTPUT BPC with valid values of: * 6, 8, 10, 12 BPC. */ -#define PIPEMISC_BPC_MASK (7 << 5) -#define PIPEMISC_8_BPC (0 << 5) -#define PIPEMISC_10_BPC (1 << 5) -#define PIPEMISC_6_BPC (2 << 5) -#define PIPEMISC_12_BPC_ADLP (4 << 5) /* adlp+ */ -#define PIPEMISC_DITHER_ENABLE (1 << 4) -#define PIPEMISC_DITHER_TYPE_MASK (3 << 2) -#define PIPEMISC_DITHER_TYPE_SP (0 << 2) +#define PIPEMISC_BPC_MASK REG_GENMASK(7, 5) +#define PIPEMISC_BPC_8 REG_FIELD_PREP(PIPEMISC_BPC_MASK, 0) +#define PIPEMISC_BPC_10 REG_FIELD_PREP(PIPEMISC_BPC_MASK, 1) +#define PIPEMISC_BPC_6 REG_FIELD_PREP(PIPEMISC_BPC_MASK, 2) +#define PIPEMISC_BPC_12_ADLP REG_FIELD_PREP(PIPEMISC_BPC_MASK, 4) /* adlp+ */ +#define PIPEMISC_DITHER_ENABLE REG_BIT(4) +#define PIPEMISC_DITHER_TYPE_MASK REG_GENMASK(3, 2) +#define PIPEMISC_DITHER_TYPE_SP REG_FIELD_PREP(PIPEMISC_DITHER_TYPE_MASK, 0) +#define PIPEMISC_DITHER_TYPE_ST1 REG_FIELD_PREP(PIPEMISC_DITHER_TYPE_MASK, 1) +#define PIPEMISC_DITHER_TYPE_ST2 REG_FIELD_PREP(PIPEMISC_DITHER_TYPE_MASK, 2) +#define PIPEMISC_DITHER_TYPE_TEMP REG_FIELD_PREP(PIPEMISC_DITHER_TYPE_MASK, 3) #define PIPEMISC(pipe) _MMIO_PIPE2(pipe, _PIPE_MISC_A) #define _PIPE_MISC2_A 0x7002C #define _PIPE_MISC2_B 0x7102C -#define PIPE_MISC2_BUBBLE_COUNTER_SCALER_EN (0x50 << 24) -#define PIPE_MISC2_BUBBLE_COUNTER_SCALER_DIS (0x14 << 24) -#define PIPE_MISC2_UNDERRUN_BUBBLE_COUNTER_MASK (0xff << 24) +#define PIPE_MISC2_BUBBLE_COUNTER_MASK REG_GENMASK(31, 24) +#define PIPE_MISC2_BUBBLE_COUNTER_SCALER_EN REG_FIELD_PREP(PIPE_MISC2_BUBBLE_COUNTER_MASK, 80) +#define PIPE_MISC2_BUBBLE_COUNTER_SCALER_DIS REG_FIELD_PREP(PIPE_MISC2_BUBBLE_COUNTER_MASK, 20) #define PIPE_MISC2(pipe) _MMIO_PIPE2(pipe, _PIPE_MISC2_A) /* Skylake+ pipe bottom (background) color */ #define _SKL_BOTTOM_COLOR_A 0x70034 -#define SKL_BOTTOM_COLOR_GAMMA_ENABLE (1 << 31) -#define SKL_BOTTOM_COLOR_CSC_ENABLE (1 << 30) +#define SKL_BOTTOM_COLOR_GAMMA_ENABLE REG_BIT(31) +#define SKL_BOTTOM_COLOR_CSC_ENABLE REG_BIT(30) #define SKL_BOTTOM_COLOR(pipe) _MMIO_PIPE2(pipe, _SKL_BOTTOM_COLOR_A) #define _ICL_PIPE_A_STATUS 0x70058 @@ -6774,44 +4367,50 @@ enum { /* Cursor A & B regs */ #define _CURACNTR 0x70080 /* Old style CUR*CNTR flags (desktop 8xx) */ -#define CURSOR_ENABLE 0x80000000 -#define CURSOR_GAMMA_ENABLE 0x40000000 -#define CURSOR_STRIDE_SHIFT 28 -#define CURSOR_STRIDE(x) ((ffs(x) - 9) << CURSOR_STRIDE_SHIFT) /* 256,512,1k,2k */ -#define CURSOR_FORMAT_SHIFT 24 -#define CURSOR_FORMAT_MASK (0x07 << CURSOR_FORMAT_SHIFT) -#define CURSOR_FORMAT_2C (0x00 << CURSOR_FORMAT_SHIFT) -#define CURSOR_FORMAT_3C (0x01 << CURSOR_FORMAT_SHIFT) -#define CURSOR_FORMAT_4C (0x02 << CURSOR_FORMAT_SHIFT) -#define CURSOR_FORMAT_ARGB (0x04 << CURSOR_FORMAT_SHIFT) -#define CURSOR_FORMAT_XRGB (0x05 << CURSOR_FORMAT_SHIFT) +#define CURSOR_ENABLE REG_BIT(31) +#define CURSOR_PIPE_GAMMA_ENABLE REG_BIT(30) +#define CURSOR_STRIDE_MASK REG_GENMASK(29, 28) +#define CURSOR_STRIDE(stride) REG_FIELD_PREP(CURSOR_STRIDE_MASK, ffs(stride) - 9) /* 256,512,1k,2k */ +#define CURSOR_FORMAT_MASK REG_GENMASK(26, 24) +#define CURSOR_FORMAT_2C REG_FIELD_PREP(CURSOR_FORMAT_MASK, 0) +#define CURSOR_FORMAT_3C REG_FIELD_PREP(CURSOR_FORMAT_MASK, 1) +#define CURSOR_FORMAT_4C REG_FIELD_PREP(CURSOR_FORMAT_MASK, 2) +#define CURSOR_FORMAT_ARGB REG_FIELD_PREP(CURSOR_FORMAT_MASK, 4) +#define CURSOR_FORMAT_XRGB REG_FIELD_PREP(CURSOR_FORMAT_MASK, 5) /* New style CUR*CNTR flags */ -#define MCURSOR_MODE 0x27 -#define MCURSOR_MODE_DISABLE 0x00 -#define MCURSOR_MODE_128_32B_AX 0x02 -#define MCURSOR_MODE_256_32B_AX 0x03 -#define MCURSOR_MODE_64_32B_AX 0x07 -#define MCURSOR_MODE_128_ARGB_AX ((1 << 5) | MCURSOR_MODE_128_32B_AX) -#define MCURSOR_MODE_256_ARGB_AX ((1 << 5) | MCURSOR_MODE_256_32B_AX) -#define MCURSOR_MODE_64_ARGB_AX ((1 << 5) | MCURSOR_MODE_64_32B_AX) #define MCURSOR_ARB_SLOTS_MASK REG_GENMASK(30, 28) /* icl+ */ #define MCURSOR_ARB_SLOTS(x) REG_FIELD_PREP(MCURSOR_ARB_SLOTS_MASK, (x)) /* icl+ */ -#define MCURSOR_PIPE_SELECT_MASK (0x3 << 28) -#define MCURSOR_PIPE_SELECT_SHIFT 28 -#define MCURSOR_PIPE_SELECT(pipe) ((pipe) << 28) -#define MCURSOR_GAMMA_ENABLE (1 << 26) -#define MCURSOR_PIPE_CSC_ENABLE (1 << 24) /* ilk+ */ -#define MCURSOR_ROTATE_180 (1 << 15) -#define MCURSOR_TRICKLE_FEED_DISABLE (1 << 14) +#define MCURSOR_PIPE_SEL_MASK REG_GENMASK(29, 28) +#define MCURSOR_PIPE_SEL(pipe) REG_FIELD_PREP(MCURSOR_PIPE_SEL_MASK, (pipe)) +#define MCURSOR_PIPE_GAMMA_ENABLE REG_BIT(26) +#define MCURSOR_PIPE_CSC_ENABLE REG_BIT(24) /* ilk+ */ +#define MCURSOR_ROTATE_180 REG_BIT(15) +#define MCURSOR_TRICKLE_FEED_DISABLE REG_BIT(14) +#define MCURSOR_MODE_MASK 0x27 +#define MCURSOR_MODE_DISABLE 0x00 +#define MCURSOR_MODE_128_32B_AX 0x02 +#define MCURSOR_MODE_256_32B_AX 0x03 +#define MCURSOR_MODE_64_32B_AX 0x07 +#define MCURSOR_MODE_128_ARGB_AX (0x20 | MCURSOR_MODE_128_32B_AX) +#define MCURSOR_MODE_256_ARGB_AX (0x20 | MCURSOR_MODE_256_32B_AX) +#define MCURSOR_MODE_64_ARGB_AX (0x20 | MCURSOR_MODE_64_32B_AX) #define _CURABASE 0x70084 #define _CURAPOS 0x70088 -#define CURSOR_POS_MASK 0x007FF -#define CURSOR_POS_SIGN 0x8000 -#define CURSOR_X_SHIFT 0 -#define CURSOR_Y_SHIFT 16 -#define CURSIZE _MMIO(0x700a0) /* 845/865 */ +#define CURSOR_POS_Y_SIGN REG_BIT(31) +#define CURSOR_POS_Y_MASK REG_GENMASK(30, 16) +#define CURSOR_POS_Y(y) REG_FIELD_PREP(CURSOR_POS_Y_MASK, (y)) +#define CURSOR_POS_X_SIGN REG_BIT(15) +#define CURSOR_POS_X_MASK REG_GENMASK(14, 0) +#define CURSOR_POS_X(x) REG_FIELD_PREP(CURSOR_POS_X_MASK, (x)) +#define _CURASIZE 0x700a0 /* 845/865 */ +#define CURSOR_HEIGHT_MASK REG_GENMASK(21, 12) +#define CURSOR_HEIGHT(h) REG_FIELD_PREP(CURSOR_HEIGHT_MASK, (h)) +#define CURSOR_WIDTH_MASK REG_GENMASK(9, 0) +#define CURSOR_WIDTH(w) REG_FIELD_PREP(CURSOR_WIDTH_MASK, (w)) #define _CUR_FBC_CTL_A 0x700a0 /* ivb+ */ -#define CUR_FBC_CTL_EN (1 << 31) +#define CUR_FBC_EN REG_BIT(31) +#define CUR_FBC_HEIGHT_MASK REG_GENMASK(7, 0) +#define CUR_FBC_HEIGHT(h) REG_FIELD_PREP(CUR_FBC_HEIGHT_MASK, (h)) #define _CURASURFLIVE 0x700ac /* g4x+ */ #define _CURBCNTR 0x700c0 #define _CURBBASE 0x700c4 @@ -6824,6 +4423,7 @@ enum { #define CURCNTR(pipe) _CURSOR2(pipe, _CURACNTR) #define CURBASE(pipe) _CURSOR2(pipe, _CURABASE) #define CURPOS(pipe) _CURSOR2(pipe, _CURAPOS) +#define CURSIZE(pipe) _CURSOR2(pipe, _CURASIZE) #define CUR_FBC_CTL(pipe) _CURSOR2(pipe, _CUR_FBC_CTL_A) #define CURSURFLIVE(pipe) _CURSOR2(pipe, _CURASURFLIVE) @@ -6837,49 +4437,54 @@ enum { /* Display A control */ #define _DSPAADDR_VLV 0x7017C /* vlv/chv */ #define _DSPACNTR 0x70180 -#define DISPLAY_PLANE_ENABLE (1 << 31) -#define DISPLAY_PLANE_DISABLE 0 -#define DISPPLANE_GAMMA_ENABLE (1 << 30) -#define DISPPLANE_GAMMA_DISABLE 0 -#define DISPPLANE_PIXFORMAT_MASK (0xf << 26) -#define DISPPLANE_YUV422 (0x0 << 26) -#define DISPPLANE_8BPP (0x2 << 26) -#define DISPPLANE_BGRA555 (0x3 << 26) -#define DISPPLANE_BGRX555 (0x4 << 26) -#define DISPPLANE_BGRX565 (0x5 << 26) -#define DISPPLANE_BGRX888 (0x6 << 26) -#define DISPPLANE_BGRA888 (0x7 << 26) -#define DISPPLANE_RGBX101010 (0x8 << 26) -#define DISPPLANE_RGBA101010 (0x9 << 26) -#define DISPPLANE_BGRX101010 (0xa << 26) -#define DISPPLANE_BGRA101010 (0xb << 26) -#define DISPPLANE_RGBX161616 (0xc << 26) -#define DISPPLANE_RGBX888 (0xe << 26) -#define DISPPLANE_RGBA888 (0xf << 26) -#define DISPPLANE_STEREO_ENABLE (1 << 25) -#define DISPPLANE_STEREO_DISABLE 0 -#define DISPPLANE_PIPE_CSC_ENABLE (1 << 24) /* ilk+ */ -#define DISPPLANE_SEL_PIPE_SHIFT 24 -#define DISPPLANE_SEL_PIPE_MASK (3 << DISPPLANE_SEL_PIPE_SHIFT) -#define DISPPLANE_SEL_PIPE(pipe) ((pipe) << DISPPLANE_SEL_PIPE_SHIFT) -#define DISPPLANE_SRC_KEY_ENABLE (1 << 22) -#define DISPPLANE_SRC_KEY_DISABLE 0 -#define DISPPLANE_LINE_DOUBLE (1 << 20) -#define DISPPLANE_NO_LINE_DOUBLE 0 -#define DISPPLANE_STEREO_POLARITY_FIRST 0 -#define DISPPLANE_STEREO_POLARITY_SECOND (1 << 18) -#define DISPPLANE_ALPHA_PREMULTIPLY (1 << 16) /* CHV pipe B */ -#define DISPPLANE_ROTATE_180 (1 << 15) -#define DISPPLANE_TRICKLE_FEED_DISABLE (1 << 14) /* Ironlake */ -#define DISPPLANE_TILED (1 << 10) -#define DISPPLANE_ASYNC_FLIP (1 << 9) /* g4x+ */ -#define DISPPLANE_MIRROR (1 << 8) /* CHV pipe B */ +#define DISP_ENABLE REG_BIT(31) +#define DISP_PIPE_GAMMA_ENABLE REG_BIT(30) +#define DISP_FORMAT_MASK REG_GENMASK(29, 26) +#define DISP_FORMAT_8BPP REG_FIELD_PREP(DISP_FORMAT_MASK, 2) +#define DISP_FORMAT_BGRA555 REG_FIELD_PREP(DISP_FORMAT_MASK, 3) +#define DISP_FORMAT_BGRX555 REG_FIELD_PREP(DISP_FORMAT_MASK, 4) +#define DISP_FORMAT_BGRX565 REG_FIELD_PREP(DISP_FORMAT_MASK, 5) +#define DISP_FORMAT_BGRX888 REG_FIELD_PREP(DISP_FORMAT_MASK, 6) +#define DISP_FORMAT_BGRA888 REG_FIELD_PREP(DISP_FORMAT_MASK, 7) +#define DISP_FORMAT_RGBX101010 REG_FIELD_PREP(DISP_FORMAT_MASK, 8) +#define DISP_FORMAT_RGBA101010 REG_FIELD_PREP(DISP_FORMAT_MASK, 9) +#define DISP_FORMAT_BGRX101010 REG_FIELD_PREP(DISP_FORMAT_MASK, 10) +#define DISP_FORMAT_BGRA101010 REG_FIELD_PREP(DISP_FORMAT_MASK, 11) +#define DISP_FORMAT_RGBX161616 REG_FIELD_PREP(DISP_FORMAT_MASK, 12) +#define DISP_FORMAT_RGBX888 REG_FIELD_PREP(DISP_FORMAT_MASK, 14) +#define DISP_FORMAT_RGBA888 REG_FIELD_PREP(DISP_FORMAT_MASK, 15) +#define DISP_STEREO_ENABLE REG_BIT(25) +#define DISP_PIPE_CSC_ENABLE REG_BIT(24) /* ilk+ */ +#define DISP_PIPE_SEL_MASK REG_GENMASK(25, 24) +#define DISP_PIPE_SEL(pipe) REG_FIELD_PREP(DISP_PIPE_SEL_MASK, (pipe)) +#define DISP_SRC_KEY_ENABLE REG_BIT(22) +#define DISP_LINE_DOUBLE REG_BIT(20) +#define DISP_STEREO_POLARITY_SECOND REG_BIT(18) +#define DISP_ALPHA_PREMULTIPLY REG_BIT(16) /* CHV pipe B */ +#define DISP_ROTATE_180 REG_BIT(15) +#define DISP_TRICKLE_FEED_DISABLE REG_BIT(14) /* g4x+ */ +#define DISP_TILED REG_BIT(10) +#define DISP_ASYNC_FLIP REG_BIT(9) /* g4x+ */ +#define DISP_MIRROR REG_BIT(8) /* CHV pipe B */ #define _DSPAADDR 0x70184 #define _DSPASTRIDE 0x70188 #define _DSPAPOS 0x7018C /* reserved */ +#define DISP_POS_Y_MASK REG_GENMASK(31, 0) +#define DISP_POS_Y(y) REG_FIELD_PREP(DISP_POS_Y_MASK, (y)) +#define DISP_POS_X_MASK REG_GENMASK(15, 0) +#define DISP_POS_X(x) REG_FIELD_PREP(DISP_POS_X_MASK, (x)) #define _DSPASIZE 0x70190 +#define DISP_HEIGHT_MASK REG_GENMASK(31, 0) +#define DISP_HEIGHT(h) REG_FIELD_PREP(DISP_HEIGHT_MASK, (h)) +#define DISP_WIDTH_MASK REG_GENMASK(15, 0) +#define DISP_WIDTH(w) REG_FIELD_PREP(DISP_WIDTH_MASK, (w)) #define _DSPASURF 0x7019C /* 965+ only */ +#define DISP_ADDR_MASK REG_GENMASK(31, 12) #define _DSPATILEOFF 0x701A4 /* 965+ only */ +#define DISP_OFFSET_Y_MASK REG_GENMASK(31, 16) +#define DISP_OFFSET_Y(y) REG_FIELD_PREP(DISP_OFFSET_Y_MASK, (y)) +#define DISP_OFFSET_X_MASK REG_GENMASK(15, 0) +#define DISP_OFFSET_X(x) REG_FIELD_PREP(DISP_OFFSET_X_MASK, (x)) #define _DSPAOFFSET 0x701A4 /* HSW */ #define _DSPASURFLIVE 0x701AC #define _DSPAGAMC 0x701E0 @@ -6899,15 +4504,28 @@ enum { /* CHV pipe B blender and primary plane */ #define _CHV_BLEND_A 0x60a00 -#define CHV_BLEND_LEGACY (0 << 30) -#define CHV_BLEND_ANDROID (1 << 30) -#define CHV_BLEND_MPO (2 << 30) -#define CHV_BLEND_MASK (3 << 30) +#define CHV_BLEND_MASK REG_GENMASK(31, 30) +#define CHV_BLEND_LEGACY REG_FIELD_PREP(CHV_BLEND_MASK, 0) +#define CHV_BLEND_ANDROID REG_FIELD_PREP(CHV_BLEND_MASK, 1) +#define CHV_BLEND_MPO REG_FIELD_PREP(CHV_BLEND_MASK, 2) #define _CHV_CANVAS_A 0x60a04 +#define CHV_CANVAS_RED_MASK REG_GENMASK(29, 20) +#define CHV_CANVAS_GREEN_MASK REG_GENMASK(19, 10) +#define CHV_CANVAS_BLUE_MASK REG_GENMASK(9, 0) #define _PRIMPOS_A 0x60a08 +#define PRIM_POS_Y_MASK REG_GENMASK(31, 16) +#define PRIM_POS_Y(y) REG_FIELD_PREP(PRIM_POS_Y_MASK, (y)) +#define PRIM_POS_X_MASK REG_GENMASK(15, 0) +#define PRIM_POS_X(x) REG_FIELD_PREP(PRIM_POS_X_MASK, (x)) #define _PRIMSIZE_A 0x60a0c +#define PRIM_HEIGHT_MASK REG_GENMASK(31, 16) +#define PRIM_HEIGHT(h) REG_FIELD_PREP(PRIM_HEIGHT_MASK, (h)) +#define PRIM_WIDTH_MASK REG_GENMASK(15, 0) +#define PRIM_WIDTH(w) REG_FIELD_PREP(PRIM_WIDTH_MASK, (w)) #define _PRIMCNSTALPHA_A 0x60a10 -#define PRIM_CONST_ALPHA_ENABLE (1 << 31) +#define PRIM_CONST_ALPHA_ENABLE REG_BIT(31) +#define PRIM_CONST_ALPHA_MASK REG_GENMASK(7, 0) +#define PRIM_CONST_ALPHA(alpha) REG_FIELD_PREP(PRIM_CONST_ALPHA_MASK, (alpha)) #define CHV_BLEND(pipe) _MMIO_TRANS2(pipe, _CHV_BLEND_A) #define CHV_CANVAS(pipe) _MMIO_TRANS2(pipe, _CHV_CANVAS_A) @@ -6948,10 +4566,8 @@ enum { /* Display B control */ #define _DSPBCNTR (DISPLAY_MMIO_BASE(dev_priv) + 0x71180) -#define DISPPLANE_ALPHA_TRANS_ENABLE (1 << 15) -#define DISPPLANE_ALPHA_TRANS_DISABLE 0 -#define DISPPLANE_SPRITE_ABOVE_DISPLAY 0 -#define DISPPLANE_SPRITE_ABOVE_OVERLAY (1) +#define DISP_ALPHA_TRANS_ENABLE REG_BIT(15) +#define DISP_SPRITE_ABOVE_OVERLAY REG_BIT(0) #define _DSPBADDR (DISPLAY_MMIO_BASE(dev_priv) + 0x71184) #define _DSPBSTRIDE (DISPLAY_MMIO_BASE(dev_priv) + 0x71188) #define _DSPBPOS (DISPLAY_MMIO_BASE(dev_priv) + 0x7118C) @@ -6967,46 +4583,63 @@ enum { /* Sprite A control */ #define _DVSACNTR 0x72180 -#define DVS_ENABLE (1 << 31) -#define DVS_GAMMA_ENABLE (1 << 30) -#define DVS_YUV_RANGE_CORRECTION_DISABLE (1 << 27) -#define DVS_PIXFORMAT_MASK (3 << 25) -#define DVS_FORMAT_YUV422 (0 << 25) -#define DVS_FORMAT_RGBX101010 (1 << 25) -#define DVS_FORMAT_RGBX888 (2 << 25) -#define DVS_FORMAT_RGBX161616 (3 << 25) -#define DVS_PIPE_CSC_ENABLE (1 << 24) -#define DVS_SOURCE_KEY (1 << 22) -#define DVS_RGB_ORDER_XBGR (1 << 20) -#define DVS_YUV_FORMAT_BT709 (1 << 18) -#define DVS_YUV_ORDER_MASK (3 << 16) -#define DVS_YUV_ORDER_YUYV (0 << 16) -#define DVS_YUV_ORDER_UYVY (1 << 16) -#define DVS_YUV_ORDER_YVYU (2 << 16) -#define DVS_YUV_ORDER_VYUY (3 << 16) -#define DVS_ROTATE_180 (1 << 15) -#define DVS_DEST_KEY (1 << 2) -#define DVS_TRICKLE_FEED_DISABLE (1 << 14) -#define DVS_TILED (1 << 10) +#define DVS_ENABLE REG_BIT(31) +#define DVS_PIPE_GAMMA_ENABLE REG_BIT(30) +#define DVS_YUV_RANGE_CORRECTION_DISABLE REG_BIT(27) +#define DVS_FORMAT_MASK REG_GENMASK(26, 25) +#define DVS_FORMAT_YUV422 REG_FIELD_PREP(DVS_FORMAT_MASK, 0) +#define DVS_FORMAT_RGBX101010 REG_FIELD_PREP(DVS_FORMAT_MASK, 1) +#define DVS_FORMAT_RGBX888 REG_FIELD_PREP(DVS_FORMAT_MASK, 2) +#define DVS_FORMAT_RGBX161616 REG_FIELD_PREP(DVS_FORMAT_MASK, 3) +#define DVS_PIPE_CSC_ENABLE REG_BIT(24) +#define DVS_SOURCE_KEY REG_BIT(22) +#define DVS_RGB_ORDER_XBGR REG_BIT(20) +#define DVS_YUV_FORMAT_BT709 REG_BIT(18) +#define DVS_YUV_ORDER_MASK REG_GENMASK(17, 16) +#define DVS_YUV_ORDER_YUYV REG_FIELD_PREP(DVS_YUV_ORDER_MASK, 0) +#define DVS_YUV_ORDER_UYVY REG_FIELD_PREP(DVS_YUV_ORDER_MASK, 1) +#define DVS_YUV_ORDER_YVYU REG_FIELD_PREP(DVS_YUV_ORDER_MASK, 2) +#define DVS_YUV_ORDER_VYUY REG_FIELD_PREP(DVS_YUV_ORDER_MASK, 3) +#define DVS_ROTATE_180 REG_BIT(15) +#define DVS_TRICKLE_FEED_DISABLE REG_BIT(14) +#define DVS_TILED REG_BIT(10) +#define DVS_DEST_KEY REG_BIT(2) #define _DVSALINOFF 0x72184 #define _DVSASTRIDE 0x72188 #define _DVSAPOS 0x7218c +#define DVS_POS_Y_MASK REG_GENMASK(31, 16) +#define DVS_POS_Y(y) REG_FIELD_PREP(DVS_POS_Y_MASK, (y)) +#define DVS_POS_X_MASK REG_GENMASK(15, 0) +#define DVS_POS_X(x) REG_FIELD_PREP(DVS_POS_X_MASK, (x)) #define _DVSASIZE 0x72190 +#define DVS_HEIGHT_MASK REG_GENMASK(31, 16) +#define DVS_HEIGHT(h) REG_FIELD_PREP(DVS_HEIGHT_MASK, (h)) +#define DVS_WIDTH_MASK REG_GENMASK(15, 0) +#define DVS_WIDTH(w) REG_FIELD_PREP(DVS_WIDTH_MASK, (w)) #define _DVSAKEYVAL 0x72194 #define _DVSAKEYMSK 0x72198 #define _DVSASURF 0x7219c +#define DVS_ADDR_MASK REG_GENMASK(31, 12) #define _DVSAKEYMAXVAL 0x721a0 #define _DVSATILEOFF 0x721a4 +#define DVS_OFFSET_Y_MASK REG_GENMASK(31, 16) +#define DVS_OFFSET_Y(y) REG_FIELD_PREP(DVS_OFFSET_Y_MASK, (y)) +#define DVS_OFFSET_X_MASK REG_GENMASK(15, 0) +#define DVS_OFFSET_X(x) REG_FIELD_PREP(DVS_OFFSET_X_MASK, (x)) #define _DVSASURFLIVE 0x721ac #define _DVSAGAMC_G4X 0x721e0 /* g4x */ #define _DVSASCALE 0x72204 -#define DVS_SCALE_ENABLE (1 << 31) -#define DVS_FILTER_MASK (3 << 29) -#define DVS_FILTER_MEDIUM (0 << 29) -#define DVS_FILTER_ENHANCING (1 << 29) -#define DVS_FILTER_SOFTENING (2 << 29) -#define DVS_VERTICAL_OFFSET_HALF (1 << 28) /* must be enabled below */ -#define DVS_VERTICAL_OFFSET_ENABLE (1 << 27) +#define DVS_SCALE_ENABLE REG_BIT(31) +#define DVS_FILTER_MASK REG_GENMASK(30, 29) +#define DVS_FILTER_MEDIUM REG_FIELD_PREP(DVS_FILTER_MASK, 0) +#define DVS_FILTER_ENHANCING REG_FIELD_PREP(DVS_FILTER_MASK, 1) +#define DVS_FILTER_SOFTENING REG_FIELD_PREP(DVS_FILTER_MASK, 2) +#define DVS_VERTICAL_OFFSET_HALF REG_BIT(28) /* must be enabled below */ +#define DVS_VERTICAL_OFFSET_ENABLE REG_BIT(27) +#define DVS_SRC_WIDTH_MASK REG_GENMASK(26, 16) +#define DVS_SRC_WIDTH(w) REG_FIELD_PREP(DVS_SRC_WIDTH_MASK, (w)) +#define DVS_SRC_HEIGHT_MASK REG_GENMASK(10, 0) +#define DVS_SRC_HEIGHT(h) REG_FIELD_PREP(DVS_SRC_HEIGHT_MASK, (h)) #define _DVSAGAMC_ILK 0x72300 /* ilk/snb */ #define _DVSAGAMCMAX_ILK 0x72340 /* ilk/snb */ @@ -7043,50 +4676,67 @@ enum { #define DVSGAMCMAX_ILK(pipe, i) _MMIO(_PIPE(pipe, _DVSAGAMCMAX_ILK, _DVSBGAMCMAX_ILK) + (i) * 4) /* 3 x u1.10 */ #define _SPRA_CTL 0x70280 -#define SPRITE_ENABLE (1 << 31) -#define SPRITE_GAMMA_ENABLE (1 << 30) -#define SPRITE_YUV_RANGE_CORRECTION_DISABLE (1 << 28) -#define SPRITE_PIXFORMAT_MASK (7 << 25) -#define SPRITE_FORMAT_YUV422 (0 << 25) -#define SPRITE_FORMAT_RGBX101010 (1 << 25) -#define SPRITE_FORMAT_RGBX888 (2 << 25) -#define SPRITE_FORMAT_RGBX161616 (3 << 25) -#define SPRITE_FORMAT_YUV444 (4 << 25) -#define SPRITE_FORMAT_XR_BGR101010 (5 << 25) /* Extended range */ -#define SPRITE_PIPE_CSC_ENABLE (1 << 24) -#define SPRITE_SOURCE_KEY (1 << 22) -#define SPRITE_RGB_ORDER_RGBX (1 << 20) /* only for 888 and 161616 */ -#define SPRITE_YUV_TO_RGB_CSC_DISABLE (1 << 19) -#define SPRITE_YUV_TO_RGB_CSC_FORMAT_BT709 (1 << 18) /* 0 is BT601 */ -#define SPRITE_YUV_ORDER_MASK (3 << 16) -#define SPRITE_YUV_ORDER_YUYV (0 << 16) -#define SPRITE_YUV_ORDER_UYVY (1 << 16) -#define SPRITE_YUV_ORDER_YVYU (2 << 16) -#define SPRITE_YUV_ORDER_VYUY (3 << 16) -#define SPRITE_ROTATE_180 (1 << 15) -#define SPRITE_TRICKLE_FEED_DISABLE (1 << 14) -#define SPRITE_INT_GAMMA_DISABLE (1 << 13) -#define SPRITE_TILED (1 << 10) -#define SPRITE_DEST_KEY (1 << 2) +#define SPRITE_ENABLE REG_BIT(31) +#define SPRITE_PIPE_GAMMA_ENABLE REG_BIT(30) +#define SPRITE_YUV_RANGE_CORRECTION_DISABLE REG_BIT(28) +#define SPRITE_FORMAT_MASK REG_GENMASK(27, 25) +#define SPRITE_FORMAT_YUV422 REG_FIELD_PREP(SPRITE_FORMAT_MASK, 0) +#define SPRITE_FORMAT_RGBX101010 REG_FIELD_PREP(SPRITE_FORMAT_MASK, 1) +#define SPRITE_FORMAT_RGBX888 REG_FIELD_PREP(SPRITE_FORMAT_MASK, 2) +#define SPRITE_FORMAT_RGBX161616 REG_FIELD_PREP(SPRITE_FORMAT_MASK, 3) +#define SPRITE_FORMAT_YUV444 REG_FIELD_PREP(SPRITE_FORMAT_MASK, 4) +#define SPRITE_FORMAT_XR_BGR101010 REG_FIELD_PREP(SPRITE_FORMAT_MASK, 5) /* Extended range */ +#define SPRITE_PIPE_CSC_ENABLE REG_BIT(24) +#define SPRITE_SOURCE_KEY REG_BIT(22) +#define SPRITE_RGB_ORDER_RGBX REG_BIT(20) /* only for 888 and 161616 */ +#define SPRITE_YUV_TO_RGB_CSC_DISABLE REG_BIT(19) +#define SPRITE_YUV_TO_RGB_CSC_FORMAT_BT709 REG_BIT(18) /* 0 is BT601 */ +#define SPRITE_YUV_ORDER_MASK REG_GENMASK(17, 16) +#define SPRITE_YUV_ORDER_YUYV REG_FIELD_PREP(SPRITE_YUV_ORDER_MASK, 0) +#define SPRITE_YUV_ORDER_UYVY REG_FIELD_PREP(SPRITE_YUV_ORDER_MASK, 1) +#define SPRITE_YUV_ORDER_YVYU REG_FIELD_PREP(SPRITE_YUV_ORDER_MASK, 2) +#define SPRITE_YUV_ORDER_VYUY REG_FIELD_PREP(SPRITE_YUV_ORDER_MASK, 3) +#define SPRITE_ROTATE_180 REG_BIT(15) +#define SPRITE_TRICKLE_FEED_DISABLE REG_BIT(14) +#define SPRITE_PLANE_GAMMA_DISABLE REG_BIT(13) +#define SPRITE_TILED REG_BIT(10) +#define SPRITE_DEST_KEY REG_BIT(2) #define _SPRA_LINOFF 0x70284 #define _SPRA_STRIDE 0x70288 #define _SPRA_POS 0x7028c +#define SPRITE_POS_Y_MASK REG_GENMASK(31, 16) +#define SPRITE_POS_Y(y) REG_FIELD_PREP(SPRITE_POS_Y_MASK, (y)) +#define SPRITE_POS_X_MASK REG_GENMASK(15, 0) +#define SPRITE_POS_X(x) REG_FIELD_PREP(SPRITE_POS_X_MASK, (x)) #define _SPRA_SIZE 0x70290 +#define SPRITE_HEIGHT_MASK REG_GENMASK(31, 16) +#define SPRITE_HEIGHT(h) REG_FIELD_PREP(SPRITE_HEIGHT_MASK, (h)) +#define SPRITE_WIDTH_MASK REG_GENMASK(15, 0) +#define SPRITE_WIDTH(w) REG_FIELD_PREP(SPRITE_WIDTH_MASK, (w)) #define _SPRA_KEYVAL 0x70294 #define _SPRA_KEYMSK 0x70298 #define _SPRA_SURF 0x7029c +#define SPRITE_ADDR_MASK REG_GENMASK(31, 12) #define _SPRA_KEYMAX 0x702a0 #define _SPRA_TILEOFF 0x702a4 +#define SPRITE_OFFSET_Y_MASK REG_GENMASK(31, 16) +#define SPRITE_OFFSET_Y(y) REG_FIELD_PREP(SPRITE_OFFSET_Y_MASK, (y)) +#define SPRITE_OFFSET_X_MASK REG_GENMASK(15, 0) +#define SPRITE_OFFSET_X(x) REG_FIELD_PREP(SPRITE_OFFSET_X_MASK, (x)) #define _SPRA_OFFSET 0x702a4 #define _SPRA_SURFLIVE 0x702ac #define _SPRA_SCALE 0x70304 -#define SPRITE_SCALE_ENABLE (1 << 31) -#define SPRITE_FILTER_MASK (3 << 29) -#define SPRITE_FILTER_MEDIUM (0 << 29) -#define SPRITE_FILTER_ENHANCING (1 << 29) -#define SPRITE_FILTER_SOFTENING (2 << 29) -#define SPRITE_VERTICAL_OFFSET_HALF (1 << 28) /* must be enabled below */ -#define SPRITE_VERTICAL_OFFSET_ENABLE (1 << 27) +#define SPRITE_SCALE_ENABLE REG_BIT(31) +#define SPRITE_FILTER_MASK REG_GENMASK(30, 29) +#define SPRITE_FILTER_MEDIUM REG_FIELD_PREP(SPRITE_FILTER_MASK, 0) +#define SPRITE_FILTER_ENHANCING REG_FIELD_PREP(SPRITE_FILTER_MASK, 1) +#define SPRITE_FILTER_SOFTENING REG_FIELD_PREP(SPRITE_FILTER_MASK, 2) +#define SPRITE_VERTICAL_OFFSET_HALF REG_BIT(28) /* must be enabled below */ +#define SPRITE_VERTICAL_OFFSET_ENABLE REG_BIT(27) +#define SPRITE_SRC_WIDTH_MASK REG_GENMASK(26, 16) +#define SPRITE_SRC_WIDTH(w) REG_FIELD_PREP(SPRITE_SRC_WIDTH_MASK, (w)) +#define SPRITE_SRC_HEIGHT_MASK REG_GENMASK(10, 0) +#define SPRITE_SRC_HEIGHT(h) REG_FIELD_PREP(SPRITE_SRC_HEIGHT_MASK, (h)) #define _SPRA_GAMC 0x70400 #define _SPRA_GAMC16 0x70440 #define _SPRA_GAMC17 0x7044c @@ -7126,48 +4776,67 @@ enum { #define SPRSURFLIVE(pipe) _MMIO_PIPE(pipe, _SPRA_SURFLIVE, _SPRB_SURFLIVE) #define _SPACNTR (VLV_DISPLAY_BASE + 0x72180) -#define SP_ENABLE (1 << 31) -#define SP_GAMMA_ENABLE (1 << 30) -#define SP_PIXFORMAT_MASK (0xf << 26) -#define SP_FORMAT_YUV422 (0x0 << 26) -#define SP_FORMAT_8BPP (0x2 << 26) -#define SP_FORMAT_BGR565 (0x5 << 26) -#define SP_FORMAT_BGRX8888 (0x6 << 26) -#define SP_FORMAT_BGRA8888 (0x7 << 26) -#define SP_FORMAT_RGBX1010102 (0x8 << 26) -#define SP_FORMAT_RGBA1010102 (0x9 << 26) -#define SP_FORMAT_BGRX1010102 (0xa << 26) /* CHV pipe B */ -#define SP_FORMAT_BGRA1010102 (0xb << 26) /* CHV pipe B */ -#define SP_FORMAT_RGBX8888 (0xe << 26) -#define SP_FORMAT_RGBA8888 (0xf << 26) -#define SP_ALPHA_PREMULTIPLY (1 << 23) /* CHV pipe B */ -#define SP_SOURCE_KEY (1 << 22) -#define SP_YUV_FORMAT_BT709 (1 << 18) -#define SP_YUV_ORDER_MASK (3 << 16) -#define SP_YUV_ORDER_YUYV (0 << 16) -#define SP_YUV_ORDER_UYVY (1 << 16) -#define SP_YUV_ORDER_YVYU (2 << 16) -#define SP_YUV_ORDER_VYUY (3 << 16) -#define SP_ROTATE_180 (1 << 15) -#define SP_TILED (1 << 10) -#define SP_MIRROR (1 << 8) /* CHV pipe B */ +#define SP_ENABLE REG_BIT(31) +#define SP_PIPE_GAMMA_ENABLE REG_BIT(30) +#define SP_FORMAT_MASK REG_GENMASK(29, 26) +#define SP_FORMAT_YUV422 REG_FIELD_PREP(SP_FORMAT_MASK, 0) +#define SP_FORMAT_8BPP REG_FIELD_PREP(SP_FORMAT_MASK, 2) +#define SP_FORMAT_BGR565 REG_FIELD_PREP(SP_FORMAT_MASK, 5) +#define SP_FORMAT_BGRX8888 REG_FIELD_PREP(SP_FORMAT_MASK, 6) +#define SP_FORMAT_BGRA8888 REG_FIELD_PREP(SP_FORMAT_MASK, 7) +#define SP_FORMAT_RGBX1010102 REG_FIELD_PREP(SP_FORMAT_MASK, 8) +#define SP_FORMAT_RGBA1010102 REG_FIELD_PREP(SP_FORMAT_MASK, 9) +#define SP_FORMAT_BGRX1010102 REG_FIELD_PREP(SP_FORMAT_MASK, 10) /* CHV pipe B */ +#define SP_FORMAT_BGRA1010102 REG_FIELD_PREP(SP_FORMAT_MASK, 11) /* CHV pipe B */ +#define SP_FORMAT_RGBX8888 REG_FIELD_PREP(SP_FORMAT_MASK, 14) +#define SP_FORMAT_RGBA8888 REG_FIELD_PREP(SP_FORMAT_MASK, 15) +#define SP_ALPHA_PREMULTIPLY REG_BIT(23) /* CHV pipe B */ +#define SP_SOURCE_KEY REG_BIT(22) +#define SP_YUV_FORMAT_BT709 REG_BIT(18) +#define SP_YUV_ORDER_MASK REG_GENMASK(17, 16) +#define SP_YUV_ORDER_YUYV REG_FIELD_PREP(SP_YUV_ORDER_MASK, 0) +#define SP_YUV_ORDER_UYVY REG_FIELD_PREP(SP_YUV_ORDER_MASK, 1) +#define SP_YUV_ORDER_YVYU REG_FIELD_PREP(SP_YUV_ORDER_MASK, 2) +#define SP_YUV_ORDER_VYUY REG_FIELD_PREP(SP_YUV_ORDER_MASK, 3) +#define SP_ROTATE_180 REG_BIT(15) +#define SP_TILED REG_BIT(10) +#define SP_MIRROR REG_BIT(8) /* CHV pipe B */ #define _SPALINOFF (VLV_DISPLAY_BASE + 0x72184) #define _SPASTRIDE (VLV_DISPLAY_BASE + 0x72188) #define _SPAPOS (VLV_DISPLAY_BASE + 0x7218c) +#define SP_POS_Y_MASK REG_GENMASK(31, 16) +#define SP_POS_Y(y) REG_FIELD_PREP(SP_POS_Y_MASK, (y)) +#define SP_POS_X_MASK REG_GENMASK(15, 0) +#define SP_POS_X(x) REG_FIELD_PREP(SP_POS_X_MASK, (x)) #define _SPASIZE (VLV_DISPLAY_BASE + 0x72190) +#define SP_HEIGHT_MASK REG_GENMASK(31, 16) +#define SP_HEIGHT(h) REG_FIELD_PREP(SP_HEIGHT_MASK, (h)) +#define SP_WIDTH_MASK REG_GENMASK(15, 0) +#define SP_WIDTH(w) REG_FIELD_PREP(SP_WIDTH_MASK, (w)) #define _SPAKEYMINVAL (VLV_DISPLAY_BASE + 0x72194) #define _SPAKEYMSK (VLV_DISPLAY_BASE + 0x72198) #define _SPASURF (VLV_DISPLAY_BASE + 0x7219c) +#define SP_ADDR_MASK REG_GENMASK(31, 12) #define _SPAKEYMAXVAL (VLV_DISPLAY_BASE + 0x721a0) #define _SPATILEOFF (VLV_DISPLAY_BASE + 0x721a4) +#define SP_OFFSET_Y_MASK REG_GENMASK(31, 16) +#define SP_OFFSET_Y(y) REG_FIELD_PREP(SP_OFFSET_Y_MASK, (y)) +#define SP_OFFSET_X_MASK REG_GENMASK(15, 0) +#define SP_OFFSET_X(x) REG_FIELD_PREP(SP_OFFSET_X_MASK, (x)) #define _SPACONSTALPHA (VLV_DISPLAY_BASE + 0x721a8) -#define SP_CONST_ALPHA_ENABLE (1 << 31) +#define SP_CONST_ALPHA_ENABLE REG_BIT(31) +#define SP_CONST_ALPHA_MASK REG_GENMASK(7, 0) +#define SP_CONST_ALPHA(alpha) REG_FIELD_PREP(SP_CONST_ALPHA_MASK, (alpha)) #define _SPACLRC0 (VLV_DISPLAY_BASE + 0x721d0) -#define SP_CONTRAST(x) ((x) << 18) /* u3.6 */ -#define SP_BRIGHTNESS(x) ((x) & 0xff) /* s8 */ +#define SP_CONTRAST_MASK REG_GENMASK(26, 18) +#define SP_CONTRAST(x) REG_FIELD_PREP(SP_CONTRAST_MASK, (x)) /* u3.6 */ +#define SP_BRIGHTNESS_MASK REG_GENMASK(7, 0) +#define SP_BRIGHTNESS(x) REG_FIELD_PREP(SP_BRIGHTNESS_MASK, (x)) /* s8 */ #define _SPACLRC1 (VLV_DISPLAY_BASE + 0x721d4) -#define SP_SH_SIN(x) (((x) & 0x7ff) << 16) /* s4.7 */ -#define SP_SH_COS(x) (x) /* u3.7 */ +#define SP_SH_SIN_MASK REG_GENMASK(26, 16) +#define SP_SH_SIN(x) REG_FIELD_PREP(SP_SH_SIN_MASK, (x)) /* s4.7 */ +#define SP_SH_COS_MASK REG_GENMASK(9, 0) +#define SP_SH_COS(x) REG_FIELD_PREP(SP_SH_COS_MASK, (x)) /* u3.7 */ #define _SPAGAMC (VLV_DISPLAY_BASE + 0x721e0) #define _SPBCNTR (VLV_DISPLAY_BASE + 0x72280) @@ -7218,112 +4887,135 @@ enum { #define SPCSCYGOFF(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d900) #define SPCSCCBOFF(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d904) #define SPCSCCROFF(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d908) -#define SPCSC_OOFF(x) (((x) & 0x7ff) << 16) /* s11 */ -#define SPCSC_IOFF(x) (((x) & 0x7ff) << 0) /* s11 */ +#define SPCSC_OOFF_MASK REG_GENMASK(26, 16) +#define SPCSC_OOFF(x) REG_FIELD_PREP(SPCSC_OOFF_MASK, (x) & 0x7ff) /* s11 */ +#define SPCSC_IOFF_MASK REG_GENMASK(10, 0) +#define SPCSC_IOFF(x) REG_FIELD_PREP(SPCSC_IOFF_MASK, (x) & 0x7ff) /* s11 */ #define SPCSCC01(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d90c) #define SPCSCC23(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d910) #define SPCSCC45(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d914) #define SPCSCC67(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d918) #define SPCSCC8(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d91c) -#define SPCSC_C1(x) (((x) & 0x7fff) << 16) /* s3.12 */ -#define SPCSC_C0(x) (((x) & 0x7fff) << 0) /* s3.12 */ +#define SPCSC_C1_MASK REG_GENMASK(30, 16) +#define SPCSC_C1(x) REG_FIELD_PREP(SPCSC_C1_MASK, (x) & 0x7fff) /* s3.12 */ +#define SPCSC_C0_MASK REG_GENMASK(14, 0) +#define SPCSC_C0(x) REG_FIELD_PREP(SPCSC_C0_MASK, (x) & 0x7fff) /* s3.12 */ #define SPCSCYGICLAMP(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d920) #define SPCSCCBICLAMP(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d924) #define SPCSCCRICLAMP(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d928) -#define SPCSC_IMAX(x) (((x) & 0x7ff) << 16) /* s11 */ -#define SPCSC_IMIN(x) (((x) & 0x7ff) << 0) /* s11 */ +#define SPCSC_IMAX_MASK REG_GENMASK(26, 16) +#define SPCSC_IMAX(x) REG_FIELD_PREP(SPCSC_IMAX_MASK, (x) & 0x7ff) /* s11 */ +#define SPCSC_IMIN_MASK REG_GENMASK(10, 0) +#define SPCSC_IMIN(x) REG_FIELD_PREP(SPCSC_IMIN_MASK, (x) & 0x7ff) /* s11 */ #define SPCSCYGOCLAMP(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d92c) #define SPCSCCBOCLAMP(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d930) #define SPCSCCROCLAMP(plane_id) _MMIO_CHV_SPCSC(plane_id, 0x6d934) -#define SPCSC_OMAX(x) ((x) << 16) /* u10 */ -#define SPCSC_OMIN(x) ((x) << 0) /* u10 */ +#define SPCSC_OMAX_MASK REG_GENMASK(25, 16) +#define SPCSC_OMAX(x) REG_FIELD_PREP(SPCSC_OMAX_MASK, (x)) /* u10 */ +#define SPCSC_OMIN_MASK REG_GENMASK(9, 0) +#define SPCSC_OMIN(x) REG_FIELD_PREP(SPCSC_OMIN_MASK, (x)) /* u10 */ /* Skylake plane registers */ #define _PLANE_CTL_1_A 0x70180 #define _PLANE_CTL_2_A 0x70280 #define _PLANE_CTL_3_A 0x70380 -#define PLANE_CTL_ENABLE (1 << 31) +#define PLANE_CTL_ENABLE REG_BIT(31) #define PLANE_CTL_ARB_SLOTS_MASK REG_GENMASK(30, 28) /* icl+ */ #define PLANE_CTL_ARB_SLOTS(x) REG_FIELD_PREP(PLANE_CTL_ARB_SLOTS_MASK, (x)) /* icl+ */ -#define PLANE_CTL_PIPE_GAMMA_ENABLE (1 << 30) /* Pre-GLK */ -#define PLANE_CTL_YUV_RANGE_CORRECTION_DISABLE (1 << 28) +#define PLANE_CTL_PIPE_GAMMA_ENABLE REG_BIT(30) /* Pre-GLK */ +#define PLANE_CTL_YUV_RANGE_CORRECTION_DISABLE REG_BIT(28) /* * ICL+ uses the same PLANE_CTL_FORMAT bits, but the field definition * expanded to include bit 23 as well. However, the shift-24 based values * correctly map to the same formats in ICL, as long as bit 23 is set to 0 */ -#define PLANE_CTL_FORMAT_MASK (0xf << 24) -#define PLANE_CTL_FORMAT_YUV422 (0 << 24) -#define PLANE_CTL_FORMAT_NV12 (1 << 24) -#define PLANE_CTL_FORMAT_XRGB_2101010 (2 << 24) -#define PLANE_CTL_FORMAT_P010 (3 << 24) -#define PLANE_CTL_FORMAT_XRGB_8888 (4 << 24) -#define PLANE_CTL_FORMAT_P012 (5 << 24) -#define PLANE_CTL_FORMAT_XRGB_16161616F (6 << 24) -#define PLANE_CTL_FORMAT_P016 (7 << 24) -#define PLANE_CTL_FORMAT_XYUV (8 << 24) -#define PLANE_CTL_FORMAT_INDEXED (12 << 24) -#define PLANE_CTL_FORMAT_RGB_565 (14 << 24) -#define ICL_PLANE_CTL_FORMAT_MASK (0x1f << 23) -#define PLANE_CTL_PIPE_CSC_ENABLE (1 << 23) /* Pre-GLK */ -#define PLANE_CTL_FORMAT_Y210 (1 << 23) -#define PLANE_CTL_FORMAT_Y212 (3 << 23) -#define PLANE_CTL_FORMAT_Y216 (5 << 23) -#define PLANE_CTL_FORMAT_Y410 (7 << 23) -#define PLANE_CTL_FORMAT_Y412 (9 << 23) -#define PLANE_CTL_FORMAT_Y416 (0xb << 23) -#define PLANE_CTL_KEY_ENABLE_MASK (0x3 << 21) -#define PLANE_CTL_KEY_ENABLE_SOURCE (1 << 21) -#define PLANE_CTL_KEY_ENABLE_DESTINATION (2 << 21) -#define PLANE_CTL_ORDER_BGRX (0 << 20) -#define PLANE_CTL_ORDER_RGBX (1 << 20) -#define PLANE_CTL_YUV420_Y_PLANE (1 << 19) -#define PLANE_CTL_YUV_TO_RGB_CSC_FORMAT_BT709 (1 << 18) -#define PLANE_CTL_YUV422_ORDER_MASK (0x3 << 16) -#define PLANE_CTL_YUV422_ORDER_YUYV (0 << 16) -#define PLANE_CTL_YUV422_ORDER_UYVY (1 << 16) -#define PLANE_CTL_YUV422_ORDER_YVYU (2 << 16) -#define PLANE_CTL_YUV422_ORDER_VYUY (3 << 16) -#define PLANE_CTL_RENDER_DECOMPRESSION_ENABLE (1 << 15) -#define PLANE_CTL_TRICKLE_FEED_DISABLE (1 << 14) -#define PLANE_CTL_CLEAR_COLOR_DISABLE (1 << 13) /* TGL+ */ -#define PLANE_CTL_PLANE_GAMMA_DISABLE (1 << 13) /* Pre-GLK */ -#define PLANE_CTL_TILED_MASK (0x7 << 10) -#define PLANE_CTL_TILED_LINEAR (0 << 10) -#define PLANE_CTL_TILED_X (1 << 10) -#define PLANE_CTL_TILED_Y (4 << 10) -#define PLANE_CTL_TILED_YF (5 << 10) -#define PLANE_CTL_ASYNC_FLIP (1 << 9) -#define PLANE_CTL_FLIP_HORIZONTAL (1 << 8) -#define PLANE_CTL_MEDIA_DECOMPRESSION_ENABLE (1 << 4) /* TGL+ */ -#define PLANE_CTL_ALPHA_MASK (0x3 << 4) /* Pre-GLK */ -#define PLANE_CTL_ALPHA_DISABLE (0 << 4) -#define PLANE_CTL_ALPHA_SW_PREMULTIPLY (2 << 4) -#define PLANE_CTL_ALPHA_HW_PREMULTIPLY (3 << 4) -#define PLANE_CTL_ROTATE_MASK 0x3 -#define PLANE_CTL_ROTATE_0 0x0 -#define PLANE_CTL_ROTATE_90 0x1 -#define PLANE_CTL_ROTATE_180 0x2 -#define PLANE_CTL_ROTATE_270 0x3 +#define PLANE_CTL_FORMAT_MASK_SKL REG_GENMASK(27, 24) /* pre-icl */ +#define PLANE_CTL_FORMAT_MASK_ICL REG_GENMASK(27, 23) /* icl+ */ +#define PLANE_CTL_FORMAT_YUV422 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 0) +#define PLANE_CTL_FORMAT_NV12 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 1) +#define PLANE_CTL_FORMAT_XRGB_2101010 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 2) +#define PLANE_CTL_FORMAT_P010 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 3) +#define PLANE_CTL_FORMAT_XRGB_8888 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 4) +#define PLANE_CTL_FORMAT_P012 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 5) +#define PLANE_CTL_FORMAT_XRGB_16161616F REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 6) +#define PLANE_CTL_FORMAT_P016 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 7) +#define PLANE_CTL_FORMAT_XYUV REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 8) +#define PLANE_CTL_FORMAT_INDEXED REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 12) +#define PLANE_CTL_FORMAT_RGB_565 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_SKL, 14) +#define PLANE_CTL_FORMAT_Y210 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_ICL, 1) +#define PLANE_CTL_FORMAT_Y212 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_ICL, 3) +#define PLANE_CTL_FORMAT_Y216 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_ICL, 5) +#define PLANE_CTL_FORMAT_Y410 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_ICL, 7) +#define PLANE_CTL_FORMAT_Y412 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_ICL, 9) +#define PLANE_CTL_FORMAT_Y416 REG_FIELD_PREP(PLANE_CTL_FORMAT_MASK_ICL, 11) +#define PLANE_CTL_PIPE_CSC_ENABLE REG_BIT(23) /* Pre-GLK */ +#define PLANE_CTL_KEY_ENABLE_MASK REG_GENMASK(22, 21) +#define PLANE_CTL_KEY_ENABLE_SOURCE REG_FIELD_PREP(PLANE_CTL_KEY_ENABLE_MASK, 1) +#define PLANE_CTL_KEY_ENABLE_DESTINATION REG_FIELD_PREP(PLANE_CTL_KEY_ENABLE_MASK, 2) +#define PLANE_CTL_ORDER_RGBX REG_BIT(20) +#define PLANE_CTL_YUV420_Y_PLANE REG_BIT(19) +#define PLANE_CTL_YUV_TO_RGB_CSC_FORMAT_BT709 REG_BIT(18) +#define PLANE_CTL_YUV422_ORDER_MASK REG_GENMASK(17, 16) +#define PLANE_CTL_YUV422_ORDER_YUYV REG_FIELD_PREP(PLANE_CTL_YUV422_ORDER_MASK, 0) +#define PLANE_CTL_YUV422_ORDER_UYVY REG_FIELD_PREP(PLANE_CTL_YUV422_ORDER_MASK, 1) +#define PLANE_CTL_YUV422_ORDER_YVYU REG_FIELD_PREP(PLANE_CTL_YUV422_ORDER_MASK, 2) +#define PLANE_CTL_YUV422_ORDER_VYUY REG_FIELD_PREP(PLANE_CTL_YUV422_ORDER_MASK, 3) +#define PLANE_CTL_RENDER_DECOMPRESSION_ENABLE REG_BIT(15) +#define PLANE_CTL_TRICKLE_FEED_DISABLE REG_BIT(14) +#define PLANE_CTL_CLEAR_COLOR_DISABLE REG_BIT(13) /* TGL+ */ +#define PLANE_CTL_PLANE_GAMMA_DISABLE REG_BIT(13) /* Pre-GLK */ +#define PLANE_CTL_TILED_MASK REG_GENMASK(12, 10) +#define PLANE_CTL_TILED_LINEAR REG_FIELD_PREP(PLANE_CTL_TILED_MASK, 0) +#define PLANE_CTL_TILED_X REG_FIELD_PREP(PLANE_CTL_TILED_MASK, 1) +#define PLANE_CTL_TILED_Y REG_FIELD_PREP(PLANE_CTL_TILED_MASK, 4) +#define PLANE_CTL_TILED_YF REG_FIELD_PREP(PLANE_CTL_TILED_MASK, 5) +#define PLANE_CTL_ASYNC_FLIP REG_BIT(9) +#define PLANE_CTL_FLIP_HORIZONTAL REG_BIT(8) +#define PLANE_CTL_MEDIA_DECOMPRESSION_ENABLE REG_BIT(4) /* TGL+ */ +#define PLANE_CTL_ALPHA_MASK REG_GENMASK(5, 4) /* Pre-GLK */ +#define PLANE_CTL_ALPHA_DISABLE REG_FIELD_PREP(PLANE_CTL_ALPHA_MASK, 0) +#define PLANE_CTL_ALPHA_SW_PREMULTIPLY REG_FIELD_PREP(PLANE_CTL_ALPHA_MASK, 2) +#define PLANE_CTL_ALPHA_HW_PREMULTIPLY REG_FIELD_PREP(PLANE_CTL_ALPHA_MASK, 3) +#define PLANE_CTL_ROTATE_MASK REG_GENMASK(1, 0) +#define PLANE_CTL_ROTATE_0 REG_FIELD_PREP(PLANE_CTL_ROTATE_MASK, 0) +#define PLANE_CTL_ROTATE_90 REG_FIELD_PREP(PLANE_CTL_ROTATE_MASK, 1) +#define PLANE_CTL_ROTATE_180 REG_FIELD_PREP(PLANE_CTL_ROTATE_MASK, 2) +#define PLANE_CTL_ROTATE_270 REG_FIELD_PREP(PLANE_CTL_ROTATE_MASK, 3) #define _PLANE_STRIDE_1_A 0x70188 #define _PLANE_STRIDE_2_A 0x70288 #define _PLANE_STRIDE_3_A 0x70388 +#define PLANE_STRIDE__MASK REG_GENMASK(11, 0) +#define PLANE_STRIDE_(stride) REG_FIELD_PREP(PLANE_STRIDE__MASK, (stride)) #define _PLANE_POS_1_A 0x7018c #define _PLANE_POS_2_A 0x7028c #define _PLANE_POS_3_A 0x7038c +#define PLANE_POS_Y_MASK REG_GENMASK(31, 16) +#define PLANE_POS_Y(y) REG_FIELD_PREP(PLANE_POS_Y_MASK, (y)) +#define PLANE_POS_X_MASK REG_GENMASK(15, 0) +#define PLANE_POS_X(x) REG_FIELD_PREP(PLANE_POS_X_MASK, (x)) #define _PLANE_SIZE_1_A 0x70190 #define _PLANE_SIZE_2_A 0x70290 #define _PLANE_SIZE_3_A 0x70390 +#define PLANE_HEIGHT_MASK REG_GENMASK(31, 16) +#define PLANE_HEIGHT(h) REG_FIELD_PREP(PLANE_HEIGHT_MASK, (h)) +#define PLANE_WIDTH_MASK REG_GENMASK(15, 0) +#define PLANE_WIDTH(w) REG_FIELD_PREP(PLANE_WIDTH_MASK, (w)) #define _PLANE_SURF_1_A 0x7019c #define _PLANE_SURF_2_A 0x7029c #define _PLANE_SURF_3_A 0x7039c +#define PLANE_SURF_ADDR_MASK REG_GENMASK(31, 12) +#define PLANE_SURF_DECRYPT REG_BIT(2) #define _PLANE_OFFSET_1_A 0x701a4 #define _PLANE_OFFSET_2_A 0x702a4 #define _PLANE_OFFSET_3_A 0x703a4 +#define PLANE_OFFSET_Y_MASK REG_GENMASK(31, 16) +#define PLANE_OFFSET_Y(y) REG_FIELD_PREP(PLANE_OFFSET_Y_MASK, (y)) +#define PLANE_OFFSET_X_MASK REG_GENMASK(15, 0) +#define PLANE_OFFSET_X(x) REG_FIELD_PREP(PLANE_OFFSET_X_MASK, (x)) #define _PLANE_KEYVAL_1_A 0x70194 #define _PLANE_KEYVAL_2_A 0x70294 #define _PLANE_KEYMSK_1_A 0x70198 @@ -7335,42 +5027,49 @@ enum { #define _PLANE_CC_VAL_1_A 0x701b4 #define _PLANE_CC_VAL_2_A 0x702b4 #define _PLANE_AUX_DIST_1_A 0x701c0 +#define PLANE_AUX_DISTANCE_MASK REG_GENMASK(31, 12) +#define PLANE_AUX_STRIDE_MASK REG_GENMASK(11, 0) +#define PLANE_AUX_STRIDE(stride) REG_FIELD_PREP(PLANE_AUX_STRIDE_MASK, (stride)) #define _PLANE_AUX_DIST_2_A 0x702c0 #define _PLANE_AUX_OFFSET_1_A 0x701c4 #define _PLANE_AUX_OFFSET_2_A 0x702c4 #define _PLANE_CUS_CTL_1_A 0x701c8 #define _PLANE_CUS_CTL_2_A 0x702c8 -#define PLANE_CUS_ENABLE (1 << 31) -#define PLANE_CUS_Y_PLANE_4_RKL (0 << 30) -#define PLANE_CUS_Y_PLANE_5_RKL (1 << 30) -#define PLANE_CUS_Y_PLANE_6_ICL (0 << 30) -#define PLANE_CUS_Y_PLANE_7_ICL (1 << 30) -#define PLANE_CUS_HPHASE_SIGN_NEGATIVE (1 << 19) -#define PLANE_CUS_HPHASE_0 (0 << 16) -#define PLANE_CUS_HPHASE_0_25 (1 << 16) -#define PLANE_CUS_HPHASE_0_5 (2 << 16) -#define PLANE_CUS_VPHASE_SIGN_NEGATIVE (1 << 15) -#define PLANE_CUS_VPHASE_0 (0 << 12) -#define PLANE_CUS_VPHASE_0_25 (1 << 12) -#define PLANE_CUS_VPHASE_0_5 (2 << 12) +#define PLANE_CUS_ENABLE REG_BIT(31) +#define PLANE_CUS_Y_PLANE_MASK REG_BIT(30) +#define PLANE_CUS_Y_PLANE_4_RKL REG_FIELD_PREP(PLANE_CUS_Y_PLANE_MASK, 0) +#define PLANE_CUS_Y_PLANE_5_RKL REG_FIELD_PREP(PLANE_CUS_Y_PLANE_MASK, 1) +#define PLANE_CUS_Y_PLANE_6_ICL REG_FIELD_PREP(PLANE_CUS_Y_PLANE_MASK, 0) +#define PLANE_CUS_Y_PLANE_7_ICL REG_FIELD_PREP(PLANE_CUS_Y_PLANE_MASK, 1) +#define PLANE_CUS_HPHASE_SIGN_NEGATIVE REG_BIT(19) +#define PLANE_CUS_HPHASE_MASK REG_GENMASK(17, 16) +#define PLANE_CUS_HPHASE_0 REG_FIELD_PREP(PLANE_CUS_HPHASE_MASK, 0) +#define PLANE_CUS_HPHASE_0_25 REG_FIELD_PREP(PLANE_CUS_HPHASE_MASK, 1) +#define PLANE_CUS_HPHASE_0_5 REG_FIELD_PREP(PLANE_CUS_HPHASE_MASK, 2) +#define PLANE_CUS_VPHASE_SIGN_NEGATIVE REG_BIT(15) +#define PLANE_CUS_VPHASE_MASK REG_GENMASK(13, 12) +#define PLANE_CUS_VPHASE_0 REG_FIELD_PREP(PLANE_CUS_VPHASE_MASK, 0) +#define PLANE_CUS_VPHASE_0_25 REG_FIELD_PREP(PLANE_CUS_VPHASE_MASK, 1) +#define PLANE_CUS_VPHASE_0_5 REG_FIELD_PREP(PLANE_CUS_VPHASE_MASK, 2) #define _PLANE_COLOR_CTL_1_A 0x701CC /* GLK+ */ #define _PLANE_COLOR_CTL_2_A 0x702CC /* GLK+ */ #define _PLANE_COLOR_CTL_3_A 0x703CC /* GLK+ */ -#define PLANE_COLOR_PIPE_GAMMA_ENABLE (1 << 30) /* Pre-ICL */ -#define PLANE_COLOR_YUV_RANGE_CORRECTION_DISABLE (1 << 28) +#define PLANE_COLOR_PIPE_GAMMA_ENABLE REG_BIT(30) /* Pre-ICL */ +#define PLANE_COLOR_YUV_RANGE_CORRECTION_DISABLE REG_BIT(28) +#define PLANE_COLOR_PIPE_CSC_ENABLE REG_BIT(23) /* Pre-ICL */ #define PLANE_COLOR_PLANE_CSC_ENABLE REG_BIT(21) /* ICL+ */ -#define PLANE_COLOR_INPUT_CSC_ENABLE (1 << 20) /* ICL+ */ -#define PLANE_COLOR_PIPE_CSC_ENABLE (1 << 23) /* Pre-ICL */ -#define PLANE_COLOR_CSC_MODE_BYPASS (0 << 17) -#define PLANE_COLOR_CSC_MODE_YUV601_TO_RGB601 (1 << 17) -#define PLANE_COLOR_CSC_MODE_YUV709_TO_RGB709 (2 << 17) -#define PLANE_COLOR_CSC_MODE_YUV2020_TO_RGB2020 (3 << 17) -#define PLANE_COLOR_CSC_MODE_RGB709_TO_RGB2020 (4 << 17) -#define PLANE_COLOR_PLANE_GAMMA_DISABLE (1 << 13) -#define PLANE_COLOR_ALPHA_MASK (0x3 << 4) -#define PLANE_COLOR_ALPHA_DISABLE (0 << 4) -#define PLANE_COLOR_ALPHA_SW_PREMULTIPLY (2 << 4) -#define PLANE_COLOR_ALPHA_HW_PREMULTIPLY (3 << 4) +#define PLANE_COLOR_INPUT_CSC_ENABLE REG_BIT(20) /* ICL+ */ +#define PLANE_COLOR_CSC_MODE_MASK REG_GENMASK(19, 17) +#define PLANE_COLOR_CSC_MODE_BYPASS REG_FIELD_PREP(PLANE_COLOR_CSC_MODE_MASK, 0) +#define PLANE_COLOR_CSC_MODE_YUV601_TO_RGB601 REG_FIELD_PREP(PLANE_COLOR_CSC_MODE_MASK, 1) +#define PLANE_COLOR_CSC_MODE_YUV709_TO_RGB709 REG_FIELD_PREP(PLANE_COLOR_CSC_MODE_MASK, 2) +#define PLANE_COLOR_CSC_MODE_YUV2020_TO_RGB2020 REG_FIELD_PREP(PLANE_COLOR_CSC_MODE_MASK, 3) +#define PLANE_COLOR_CSC_MODE_RGB709_TO_RGB2020 REG_FIELD_PREP(PLANE_COLOR_CSC_MODE_MASK, 4) +#define PLANE_COLOR_PLANE_GAMMA_DISABLE REG_BIT(13) +#define PLANE_COLOR_ALPHA_MASK REG_GENMASK(5, 4) +#define PLANE_COLOR_ALPHA_DISABLE REG_FIELD_PREP(PLANE_COLOR_ALPHA_MASK, 0) +#define PLANE_COLOR_ALPHA_SW_PREMULTIPLY REG_FIELD_PREP(PLANE_COLOR_ALPHA_MASK, 2) +#define PLANE_COLOR_ALPHA_HW_PREMULTIPLY REG_FIELD_PREP(PLANE_COLOR_ALPHA_MASK, 3) #define _PLANE_BUF_CFG_1_A 0x7027c #define _PLANE_BUF_CFG_2_A 0x7037c #define _PLANE_NV12_BUF_CFG_1_A 0x70278 @@ -7453,8 +5152,6 @@ enum { _PIPE(pipe, _PLANE_STRIDE_3_A, _PLANE_STRIDE_3_B) #define PLANE_STRIDE(pipe, plane) \ _MMIO_PLANE(plane, _PLANE_STRIDE_1(pipe), _PLANE_STRIDE_2(pipe)) -#define PLANE_STRIDE_MASK REG_GENMASK(10, 0) -#define PLANE_STRIDE_MASK_XELPD REG_GENMASK(11, 0) #define _PLANE_POS_1_B 0x7118c #define _PLANE_POS_2_B 0x7128c @@ -7482,7 +5179,6 @@ enum { #define _PLANE_SURF_3(pipe) _PIPE(pipe, _PLANE_SURF_3_A, _PLANE_SURF_3_B) #define PLANE_SURF(pipe, plane) \ _MMIO_PLANE(plane, _PLANE_SURF_1(pipe), _PLANE_SURF_2(pipe)) -#define PLANE_SURF_DECRYPT REG_BIT(2) #define _PLANE_OFFSET_1_B 0x711a4 #define _PLANE_OFFSET_2_B 0x712a4 @@ -7514,8 +5210,11 @@ enum { #define _PLANE_BUF_CFG_1_B 0x7127c #define _PLANE_BUF_CFG_2_B 0x7137c -#define DDB_ENTRY_MASK 0xFFF /* skl+: 10 bits, icl+ 11 bits, adlp+ 12 bits */ -#define DDB_ENTRY_END_SHIFT 16 +/* skl+: 10 bits, icl+ 11 bits, adlp+ 12 bits */ +#define PLANE_BUF_END_MASK REG_GENMASK(27, 16) +#define PLANE_BUF_END(end) REG_FIELD_PREP(PLANE_BUF_END_MASK, (end)) +#define PLANE_BUF_START_MASK REG_GENMASK(11, 0) +#define PLANE_BUF_START(start) REG_FIELD_PREP(PLANE_BUF_START_MASK, (start)) #define _PLANE_BUF_CFG_1(pipe) \ _PIPE(pipe, _PLANE_BUF_CFG_1_A, _PLANE_BUF_CFG_1_B) #define _PLANE_BUF_CFG_2(pipe) \ @@ -7670,24 +5369,13 @@ enum { #define _PIPEA_DATA_M1 0x60030 -#define PIPE_DATA_M1_OFFSET 0 #define _PIPEA_DATA_N1 0x60034 -#define PIPE_DATA_N1_OFFSET 0 - #define _PIPEA_DATA_M2 0x60038 -#define PIPE_DATA_M2_OFFSET 0 #define _PIPEA_DATA_N2 0x6003c -#define PIPE_DATA_N2_OFFSET 0 - #define _PIPEA_LINK_M1 0x60040 -#define PIPE_LINK_M1_OFFSET 0 #define _PIPEA_LINK_N1 0x60044 -#define PIPE_LINK_N1_OFFSET 0 - #define _PIPEA_LINK_M2 0x60048 -#define PIPE_LINK_M2_OFFSET 0 #define _PIPEA_LINK_N2 0x6004c -#define PIPE_LINK_N2_OFFSET 0 /* PIPEB timing regs are same start from 0x61000 */ @@ -8199,63 +5887,6 @@ enum { #define GEN11_HOTPLUG_CTL_SHORT_DETECT(hpd_pin) (1 << (_HPD_PIN_TC(hpd_pin) * 4)) #define GEN11_HOTPLUG_CTL_NO_DETECT(hpd_pin) (0 << (_HPD_PIN_TC(hpd_pin) * 4)) -#define GEN11_GT_INTR_DW0 _MMIO(0x190018) -#define GEN11_CSME (31) -#define GEN11_GUNIT (28) -#define GEN11_GUC (25) -#define GEN11_WDPERF (20) -#define GEN11_KCR (19) -#define GEN11_GTPM (16) -#define GEN11_BCS (15) -#define GEN11_RCS0 (0) - -#define GEN11_GT_INTR_DW1 _MMIO(0x19001c) -#define GEN11_VECS(x) (31 - (x)) -#define GEN11_VCS(x) (x) - -#define GEN11_GT_INTR_DW(x) _MMIO(0x190018 + ((x) * 4)) - -#define GEN11_INTR_IDENTITY_REG0 _MMIO(0x190060) -#define GEN11_INTR_IDENTITY_REG1 _MMIO(0x190064) -#define GEN11_INTR_DATA_VALID (1 << 31) -#define GEN11_INTR_ENGINE_CLASS(x) (((x) & GENMASK(18, 16)) >> 16) -#define GEN11_INTR_ENGINE_INSTANCE(x) (((x) & GENMASK(25, 20)) >> 20) -#define GEN11_INTR_ENGINE_INTR(x) ((x) & 0xffff) -/* irq instances for OTHER_CLASS */ -#define OTHER_GUC_INSTANCE 0 -#define OTHER_GTPM_INSTANCE 1 -#define OTHER_KCR_INSTANCE 4 - -#define GEN11_INTR_IDENTITY_REG(x) _MMIO(0x190060 + ((x) * 4)) - -#define GEN11_IIR_REG0_SELECTOR _MMIO(0x190070) -#define GEN11_IIR_REG1_SELECTOR _MMIO(0x190074) - -#define GEN11_IIR_REG_SELECTOR(x) _MMIO(0x190070 + ((x) * 4)) - -#define GEN11_RENDER_COPY_INTR_ENABLE _MMIO(0x190030) -#define GEN11_VCS_VECS_INTR_ENABLE _MMIO(0x190034) -#define GEN11_GUC_SG_INTR_ENABLE _MMIO(0x190038) -#define GEN11_GPM_WGBOXPERF_INTR_ENABLE _MMIO(0x19003c) -#define GEN11_CRYPTO_RSVD_INTR_ENABLE _MMIO(0x190040) -#define GEN11_GUNIT_CSME_INTR_ENABLE _MMIO(0x190044) - -#define GEN11_RCS0_RSVD_INTR_MASK _MMIO(0x190090) -#define GEN11_BCS_RSVD_INTR_MASK _MMIO(0x1900a0) -#define GEN11_VCS0_VCS1_INTR_MASK _MMIO(0x1900a8) -#define GEN11_VCS2_VCS3_INTR_MASK _MMIO(0x1900ac) -#define GEN12_VCS4_VCS5_INTR_MASK _MMIO(0x1900b0) -#define GEN12_VCS6_VCS7_INTR_MASK _MMIO(0x1900b4) -#define GEN11_VECS0_VECS1_INTR_MASK _MMIO(0x1900d0) -#define GEN12_VECS2_VECS3_INTR_MASK _MMIO(0x1900d4) -#define GEN11_GUC_SG_INTR_MASK _MMIO(0x1900e8) -#define GEN11_GPM_WGBOXPERF_INTR_MASK _MMIO(0x1900ec) -#define GEN11_CRYPTO_RSVD_INTR_MASK _MMIO(0x1900f0) -#define GEN11_GUNIT_CSME_INTR_MASK _MMIO(0x1900f4) - -#define ENGINE1_MASK REG_GENMASK(31, 16) -#define ENGINE0_MASK REG_GENMASK(15, 0) - #define ILK_DISPLAY_CHICKEN2 _MMIO(0x42004) /* Required on all Ironlake and Sandybridge according to the B-Spec. */ #define ILK_ELPIN_409_SELECT (1 << 25) @@ -8442,142 +6073,6 @@ enum { #define ICL_DSSM_CDCLK_PLL_REFCLK_19_2MHz (1 << 29) #define ICL_DSSM_CDCLK_PLL_REFCLK_38_4MHz (2 << 29) -#define GEN7_FF_SLICE_CS_CHICKEN1 _MMIO(0x20e0) -#define GEN9_FFSC_PERCTX_PREEMPT_CTRL (1 << 14) - -#define FF_SLICE_CS_CHICKEN2 _MMIO(0x20e4) -#define GEN9_TSG_BARRIER_ACK_DISABLE (1 << 8) -#define GEN9_POOLED_EU_LOAD_BALANCING_FIX_DISABLE (1 << 10) - -#define GEN9_CS_DEBUG_MODE1 _MMIO(0x20ec) -#define FF_DOP_CLOCK_GATE_DISABLE REG_BIT(1) -#define GEN9_CTX_PREEMPT_REG _MMIO(0x2248) -#define GEN12_DISABLE_POSH_BUSY_FF_DOP_CG REG_BIT(11) - -#define GEN12_CS_DEBUG_MODE1_CCCSUNIT_BE_COMMON _MMIO(0x20EC) -#define GEN12_REPLAY_MODE_GRANULARITY REG_BIT(0) - -#define GEN8_CS_CHICKEN1 _MMIO(0x2580) -#define GEN9_PREEMPT_3D_OBJECT_LEVEL (1 << 0) -#define GEN9_PREEMPT_GPGPU_LEVEL(hi, lo) (((hi) << 2) | ((lo) << 1)) -#define GEN9_PREEMPT_GPGPU_MID_THREAD_LEVEL GEN9_PREEMPT_GPGPU_LEVEL(0, 0) -#define GEN9_PREEMPT_GPGPU_THREAD_GROUP_LEVEL GEN9_PREEMPT_GPGPU_LEVEL(0, 1) -#define GEN9_PREEMPT_GPGPU_COMMAND_LEVEL GEN9_PREEMPT_GPGPU_LEVEL(1, 0) -#define GEN9_PREEMPT_GPGPU_LEVEL_MASK GEN9_PREEMPT_GPGPU_LEVEL(1, 1) - -/* GEN7 chicken */ -#define GEN7_COMMON_SLICE_CHICKEN1 _MMIO(0x7010) - #define GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC (1 << 10) - #define GEN9_RHWO_OPTIMIZATION_DISABLE (1 << 14) - -#define COMMON_SLICE_CHICKEN2 _MMIO(0x7014) - #define GEN9_PBE_COMPRESSED_HASH_SELECTION (1 << 13) - #define GEN9_DISABLE_GATHER_AT_SET_SHADER_COMMON_SLICE (1 << 12) - #define GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION (1 << 8) - #define GEN8_CSC2_SBE_VUE_CACHE_CONSERVATIVE (1 << 0) - -#define GEN8_L3CNTLREG _MMIO(0x7034) - #define GEN8_ERRDETBCTRL (1 << 9) - -#define GEN11_COMMON_SLICE_CHICKEN3 _MMIO(0x7304) -#define DG1_FLOAT_POINT_BLEND_OPT_STRICT_MODE_EN REG_BIT(12) -#define XEHP_DUAL_SIMD8_SEQ_MERGE_DISABLE REG_BIT(12) -#define GEN11_BLEND_EMB_FIX_DISABLE_IN_RCC REG_BIT(11) -#define GEN12_DISABLE_CPS_AWARE_COLOR_PIPE REG_BIT(9) - -#define HIZ_CHICKEN _MMIO(0x7018) -# define CHV_HZ_8X8_MODE_IN_1X REG_BIT(15) -# define DG1_HZ_READ_SUPPRESSION_OPTIMIZATION_DISABLE REG_BIT(14) -# define BDW_HIZ_POWER_COMPILER_CLOCK_GATING_DISABLE REG_BIT(3) - -#define GEN9_SLICE_COMMON_ECO_CHICKEN0 _MMIO(0x7308) -#define DISABLE_PIXEL_MASK_CAMMING (1 << 14) - -#define GEN9_SLICE_COMMON_ECO_CHICKEN1 _MMIO(0x731c) -#define GEN11_STATE_CACHE_REDIRECT_TO_CS (1 << 11) - -#define GEN7_SARCHKMD _MMIO(0xB000) -#define GEN7_DISABLE_DEMAND_PREFETCH (1 << 31) -#define GEN7_DISABLE_SAMPLER_PREFETCH (1 << 30) - -#define GEN7_L3SQCREG1 _MMIO(0xB010) -#define VLV_B0_WA_L3SQCREG1_VALUE 0x00D30000 - -#define GEN8_L3SQCREG1 _MMIO(0xB100) -/* - * Note that on CHV the following has an off-by-one error wrt. to BSpec. - * Using the formula in BSpec leads to a hang, while the formula here works - * fine and matches the formulas for all other platforms. A BSpec change - * request has been filed to clarify this. - */ -#define L3_GENERAL_PRIO_CREDITS(x) (((x) >> 1) << 19) -#define L3_HIGH_PRIO_CREDITS(x) (((x) >> 1) << 14) -#define L3_PRIO_CREDITS_MASK ((0x1f << 19) | (0x1f << 14)) - -#define GEN7_L3CNTLREG1 _MMIO(0xB01C) -#define GEN7_WA_FOR_GEN7_L3_CONTROL 0x3C47FF8C -#define GEN7_L3AGDIS (1 << 19) -#define GEN7_L3CNTLREG2 _MMIO(0xB020) -#define GEN7_L3CNTLREG3 _MMIO(0xB024) - -#define GEN7_L3_CHICKEN_MODE_REGISTER _MMIO(0xB030) -#define GEN7_WA_L3_CHICKEN_MODE 0x20000000 -#define GEN10_L3_CHICKEN_MODE_REGISTER _MMIO(0xB114) -#define GEN11_I2M_WRITE_DISABLE (1 << 28) - -#define GEN7_L3SQCREG4 _MMIO(0xb034) -#define L3SQ_URB_READ_CAM_MATCH_DISABLE (1 << 27) - -#define GEN11_SCRATCH2 _MMIO(0xb140) -#define GEN11_COHERENT_PARTIAL_WRITE_MERGE_ENABLE (1 << 19) - -#define GEN8_L3SQCREG4 _MMIO(0xb118) -#define GEN11_LQSC_CLEAN_EVICT_DISABLE (1 << 6) -#define GEN8_LQSC_RO_PERF_DIS (1 << 27) -#define GEN8_LQSC_FLUSH_COHERENT_LINES (1 << 21) -#define GEN8_LQSQ_NONIA_COHERENT_ATOMICS_ENABLE REG_BIT(22) - -#define GEN11_L3SQCREG5 _MMIO(0xb158) -#define L3_PWM_TIMER_INIT_VAL_MASK REG_GENMASK(9, 0) - -#define XEHP_L3SCQREG7 _MMIO(0xb188) -#define BLEND_FILL_CACHING_OPT_DIS REG_BIT(3) - -/* GEN8 chicken */ -#define HDC_CHICKEN0 _MMIO(0x7300) -#define ICL_HDC_MODE _MMIO(0xE5F4) -#define HDC_FORCE_CSR_NON_COHERENT_OVR_DISABLE (1 << 15) -#define HDC_FENCE_DEST_SLM_DISABLE (1 << 14) -#define HDC_DONOT_FETCH_MEM_WHEN_MASKED (1 << 11) -#define HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT (1 << 5) -#define HDC_FORCE_NON_COHERENT (1 << 4) -#define HDC_BARRIER_PERFORMANCE_DISABLE (1 << 10) - -#define GEN12_HDC_CHICKEN0 _MMIO(0xE5F0) -#define LSC_L1_FLUSH_CTL_3D_DATAPORT_FLUSH_EVENTS_MASK REG_GENMASK(13, 11) - -#define SARB_CHICKEN1 _MMIO(0xe90c) -#define COMP_CKN_IN REG_GENMASK(30, 29) - -#define GEN8_HDC_CHICKEN1 _MMIO(0x7304) - -/* GEN9 chicken */ -#define SLICE_ECO_CHICKEN0 _MMIO(0x7308) -#define PIXEL_MASK_CAMMING_DISABLE (1 << 14) - -#define GEN9_WM_CHICKEN3 _MMIO(0x5588) -#define GEN9_FACTOR_IN_CLR_VAL_HIZ (1 << 9) - -/* WaCatErrorRejectionIssue */ -#define GEN7_SQ_CHICKEN_MBCUNIT_CONFIG _MMIO(0x9030) -#define GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB (1 << 11) - -#define HSW_SCRATCH1 _MMIO(0xb038) -#define HSW_SCRATCH1_L3_DATA_ATOMICS_DISABLE (1 << 27) - -#define BDW_SCRATCH1 _MMIO(0xb11c) -#define GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE (1 << 2) - /*GEN11 chicken */ #define _PIPEA_CHICKEN 0x70038 #define _PIPEB_CHICKEN 0x71038 @@ -8590,16 +6085,6 @@ enum { #define DG2_RENDER_CCSTAG_4_3_EN REG_BIT(12) #define PER_PIXEL_ALPHA_BYPASS_EN REG_BIT(7) -#define VFLSKPD _MMIO(0x62a8) -#define DIS_OVER_FETCH_CACHE REG_BIT(1) -#define DIS_MULT_MISS_RD_SQUASH REG_BIT(0) - -#define FF_MODE2 _MMIO(0x6604) -#define FF_MODE2_GS_TIMER_MASK REG_GENMASK(31, 24) -#define FF_MODE2_GS_TIMER_224 REG_FIELD_PREP(FF_MODE2_GS_TIMER_MASK, 224) -#define FF_MODE2_TDS_TIMER_MASK REG_GENMASK(23, 16) -#define FF_MODE2_TDS_TIMER_128 REG_FIELD_PREP(FF_MODE2_TDS_TIMER_MASK, 4) - /* PCH */ #define PCH_DISPLAY_BASE 0xc0000u @@ -9014,22 +6499,19 @@ enum { #define _PCH_TRANSBCONF 0xf1008 #define PCH_TRANSCONF(pipe) _MMIO_PIPE(pipe, _PCH_TRANSACONF, _PCH_TRANSBCONF) #define LPT_TRANSCONF PCH_TRANSCONF(PIPE_A) /* lpt has only one transcoder */ -#define TRANS_DISABLE (0 << 31) -#define TRANS_ENABLE (1 << 31) -#define TRANS_STATE_MASK (1 << 30) -#define TRANS_STATE_DISABLE (0 << 30) -#define TRANS_STATE_ENABLE (1 << 30) -#define TRANS_FRAME_START_DELAY_MASK (3 << 27) /* ibx */ -#define TRANS_FRAME_START_DELAY(x) ((x) << 27) /* ibx: 0-3 */ -#define TRANS_INTERLACE_MASK (7 << 21) -#define TRANS_PROGRESSIVE (0 << 21) -#define TRANS_INTERLACED (3 << 21) -#define TRANS_LEGACY_INTERLACED_ILK (2 << 21) -#define TRANS_8BPC (0 << 5) -#define TRANS_10BPC (1 << 5) -#define TRANS_6BPC (2 << 5) -#define TRANS_12BPC (3 << 5) - +#define TRANS_ENABLE REG_BIT(31) +#define TRANS_STATE_ENABLE REG_BIT(30) +#define TRANS_FRAME_START_DELAY_MASK REG_GENMASK(28, 27) /* ibx */ +#define TRANS_FRAME_START_DELAY(x) REG_FIELD_PREP(TRANS_FRAME_START_DELAY_MASK, (x)) /* ibx: 0-3 */ +#define TRANS_INTERLACE_MASK REG_GENMASK(23, 21) +#define TRANS_INTERLACE_PROGRESSIVE REG_FIELD_PREP(TRANS_INTERLACE_MASK, 0) +#define TRANS_INTERLACE_LEGACY_VSYNC_IBX REG_FIELD_PREP(TRANS_INTERLACE_MASK, 2) /* ibx */ +#define TRANS_INTERLACE_INTERLACED REG_FIELD_PREP(TRANS_INTERLACE_MASK, 3) +#define TRANS_BPC_MASK REG_GENMASK(7, 5) /* ibx */ +#define TRANS_BPC_8 REG_FIELD_PREP(TRANS_BPC_MASK, 0) +#define TRANS_BPC_10 REG_FIELD_PREP(TRANS_BPC_MASK, 1) +#define TRANS_BPC_6 REG_FIELD_PREP(TRANS_BPC_MASK, 2) +#define TRANS_BPC_12 REG_FIELD_PREP(TRANS_BPC_MASK, 3) #define _TRANSA_CHICKEN1 0xf0060 #define _TRANSB_CHICKEN1 0xf1060 #define TRANS_CHICKEN1(pipe) _MMIO_PIPE(pipe, _TRANSA_CHICKEN1, _TRANSB_CHICKEN1) @@ -9239,22 +6721,19 @@ enum { #define _TRANS_DP_CTL_B 0xe1300 #define _TRANS_DP_CTL_C 0xe2300 #define TRANS_DP_CTL(pipe) _MMIO_PIPE(pipe, _TRANS_DP_CTL_A, _TRANS_DP_CTL_B) -#define TRANS_DP_OUTPUT_ENABLE (1 << 31) -#define TRANS_DP_PORT_SEL_MASK (3 << 29) -#define TRANS_DP_PORT_SEL_NONE (3 << 29) -#define TRANS_DP_PORT_SEL(port) (((port) - PORT_B) << 29) -#define TRANS_DP_AUDIO_ONLY (1 << 26) -#define TRANS_DP_ENH_FRAMING (1 << 18) -#define TRANS_DP_8BPC (0 << 9) -#define TRANS_DP_10BPC (1 << 9) -#define TRANS_DP_6BPC (2 << 9) -#define TRANS_DP_12BPC (3 << 9) -#define TRANS_DP_BPC_MASK (3 << 9) -#define TRANS_DP_VSYNC_ACTIVE_HIGH (1 << 4) -#define TRANS_DP_VSYNC_ACTIVE_LOW 0 -#define TRANS_DP_HSYNC_ACTIVE_HIGH (1 << 3) -#define TRANS_DP_HSYNC_ACTIVE_LOW 0 -#define TRANS_DP_SYNC_MASK (3 << 3) +#define TRANS_DP_OUTPUT_ENABLE REG_BIT(31) +#define TRANS_DP_PORT_SEL_MASK REG_GENMASK(30, 29) +#define TRANS_DP_PORT_SEL_NONE REG_FIELD_PREP(TRANS_DP_PORT_SEL_MASK, 3) +#define TRANS_DP_PORT_SEL(port) REG_FIELD_PREP(TRANS_DP_PORT_SEL_MASK, (port) - PORT_B) +#define TRANS_DP_AUDIO_ONLY REG_BIT(26) +#define TRANS_DP_ENH_FRAMING REG_BIT(18) +#define TRANS_DP_BPC_MASK REG_GENMASK(10, 9) +#define TRANS_DP_BPC_8 REG_FIELD_PREP(TRANS_DP_BPC_MASK, 0) +#define TRANS_DP_BPC_10 REG_FIELD_PREP(TRANS_DP_BPC_MASK, 1) +#define TRANS_DP_BPC_6 REG_FIELD_PREP(TRANS_DP_BPC_MASK, 2) +#define TRANS_DP_BPC_12 REG_FIELD_PREP(TRANS_DP_BPC_MASK, 3) +#define TRANS_DP_VSYNC_ACTIVE_HIGH REG_BIT(4) +#define TRANS_DP_HSYNC_ACTIVE_HIGH REG_BIT(3) #define _TRANS_DP2_CTL_A 0x600a0 #define _TRANS_DP2_CTL_B 0x610a0 @@ -9313,261 +6792,16 @@ enum { #define VLV_PMWGICZ _MMIO(0x1300a4) -#define RC6_LOCATION _MMIO(0xD40) -#define RC6_CTX_IN_DRAM (1 << 0) -#define RC6_CTX_BASE _MMIO(0xD48) -#define RC6_CTX_BASE_MASK 0xFFFFFFF0 -#define PWRCTX_MAXCNT_RCSUNIT _MMIO(0x2054) -#define PWRCTX_MAXCNT_VCSUNIT0 _MMIO(0x12054) -#define PWRCTX_MAXCNT_BCSUNIT _MMIO(0x22054) -#define PWRCTX_MAXCNT_VECSUNIT _MMIO(0x1A054) -#define PWRCTX_MAXCNT_VCSUNIT1 _MMIO(0x1C054) -#define IDLE_TIME_MASK 0xFFFFF -#define FORCEWAKE _MMIO(0xA18C) -#define FORCEWAKE_VLV _MMIO(0x1300b0) -#define FORCEWAKE_ACK_VLV _MMIO(0x1300b4) -#define FORCEWAKE_MEDIA_VLV _MMIO(0x1300b8) -#define FORCEWAKE_ACK_MEDIA_VLV _MMIO(0x1300bc) -#define FORCEWAKE_ACK_HSW _MMIO(0x130044) -#define FORCEWAKE_ACK _MMIO(0x130090) -#define VLV_GTLC_WAKE_CTRL _MMIO(0x130090) -#define VLV_GTLC_RENDER_CTX_EXISTS (1 << 25) -#define VLV_GTLC_MEDIA_CTX_EXISTS (1 << 24) -#define VLV_GTLC_ALLOWWAKEREQ (1 << 0) - -#define VLV_GTLC_PW_STATUS _MMIO(0x130094) -#define VLV_GTLC_ALLOWWAKEACK (1 << 0) -#define VLV_GTLC_ALLOWWAKEERR (1 << 1) -#define VLV_GTLC_PW_MEDIA_STATUS_MASK (1 << 5) -#define VLV_GTLC_PW_RENDER_STATUS_MASK (1 << 7) -#define FORCEWAKE_MT _MMIO(0xa188) /* multi-threaded */ -#define FORCEWAKE_MEDIA_GEN9 _MMIO(0xa270) -#define FORCEWAKE_MEDIA_VDBOX_GEN11(n) _MMIO(0xa540 + (n) * 4) -#define FORCEWAKE_MEDIA_VEBOX_GEN11(n) _MMIO(0xa560 + (n) * 4) -#define FORCEWAKE_RENDER_GEN9 _MMIO(0xa278) -#define FORCEWAKE_GT_GEN9 _MMIO(0xa188) -#define FORCEWAKE_ACK_MEDIA_GEN9 _MMIO(0x0D88) -#define FORCEWAKE_ACK_MEDIA_VDBOX_GEN11(n) _MMIO(0x0D50 + (n) * 4) -#define FORCEWAKE_ACK_MEDIA_VEBOX_GEN11(n) _MMIO(0x0D70 + (n) * 4) -#define FORCEWAKE_ACK_RENDER_GEN9 _MMIO(0x0D84) -#define FORCEWAKE_ACK_GT_GEN9 _MMIO(0x130044) -#define FORCEWAKE_KERNEL BIT(0) -#define FORCEWAKE_USER BIT(1) -#define FORCEWAKE_KERNEL_FALLBACK BIT(15) -#define FORCEWAKE_MT_ACK _MMIO(0x130040) -#define ECOBUS _MMIO(0xa180) -#define FORCEWAKE_MT_ENABLE (1 << 5) -#define VLV_SPAREG2H _MMIO(0xA194) -#define GEN9_PWRGT_DOMAIN_STATUS _MMIO(0xA2A0) -#define GEN9_PWRGT_MEDIA_STATUS_MASK (1 << 0) -#define GEN9_PWRGT_RENDER_STATUS_MASK (1 << 1) - -#define GTFIFODBG _MMIO(0x120000) -#define GT_FIFO_SBDEDICATE_FREE_ENTRY_CHV (0x1f << 20) -#define GT_FIFO_FREE_ENTRIES_CHV (0x7f << 13) -#define GT_FIFO_SBDROPERR (1 << 6) -#define GT_FIFO_BLOBDROPERR (1 << 5) -#define GT_FIFO_SB_READ_ABORTERR (1 << 4) -#define GT_FIFO_DROPERR (1 << 3) -#define GT_FIFO_OVFERR (1 << 2) -#define GT_FIFO_IAWRERR (1 << 1) -#define GT_FIFO_IARDERR (1 << 0) - -#define GTFIFOCTL _MMIO(0x120008) -#define GT_FIFO_FREE_ENTRIES_MASK 0x7f -#define GT_FIFO_NUM_RESERVED_ENTRIES 20 -#define GT_FIFO_CTL_BLOCK_ALL_POLICY_STALL (1 << 12) -#define GT_FIFO_CTL_RC6_POLICY_STALL (1 << 11) - -#define HSW_IDICR _MMIO(0x9008) -#define IDIHASHMSK(x) (((x) & 0x3f) << 16) #define HSW_EDRAM_CAP _MMIO(0x120010) #define EDRAM_ENABLED 0x1 #define EDRAM_NUM_BANKS(cap) (((cap) >> 1) & 0xf) #define EDRAM_WAYS_IDX(cap) (((cap) >> 5) & 0x7) #define EDRAM_SETS_IDX(cap) (((cap) >> 8) & 0x3) -#define GEN6_UCGCTL1 _MMIO(0x9400) -# define GEN6_GAMUNIT_CLOCK_GATE_DISABLE (1 << 22) -# define GEN6_EU_TCUNIT_CLOCK_GATE_DISABLE (1 << 16) -# define GEN6_BLBUNIT_CLOCK_GATE_DISABLE (1 << 5) -# define GEN6_CSUNIT_CLOCK_GATE_DISABLE (1 << 7) - -#define GEN6_UCGCTL2 _MMIO(0x9404) -# define GEN6_VFUNIT_CLOCK_GATE_DISABLE (1 << 31) -# define GEN7_VDSUNIT_CLOCK_GATE_DISABLE (1 << 30) -# define GEN7_TDLUNIT_CLOCK_GATE_DISABLE (1 << 22) -# define GEN6_RCZUNIT_CLOCK_GATE_DISABLE (1 << 13) -# define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE (1 << 12) -# define GEN6_RCCUNIT_CLOCK_GATE_DISABLE (1 << 11) - -#define GEN6_UCGCTL3 _MMIO(0x9408) -# define GEN6_OACSUNIT_CLOCK_GATE_DISABLE (1 << 20) - -#define GEN7_UCGCTL4 _MMIO(0x940c) -#define GEN7_L3BANK2X_CLOCK_GATE_DISABLE (1 << 25) -#define GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE (1 << 14) - -#define GEN6_RCGCTL1 _MMIO(0x9410) -#define GEN6_RCGCTL2 _MMIO(0x9414) -#define GEN6_RSTCTL _MMIO(0x9420) - -#define GEN8_UCGCTL6 _MMIO(0x9430) -#define GEN8_GAPSUNIT_CLOCK_GATE_DISABLE (1 << 24) -#define GEN8_SDEUNIT_CLOCK_GATE_DISABLE (1 << 14) -#define GEN8_HDCUNIT_CLOCK_GATE_DISABLE_HDCREQ (1 << 28) - -#define UNSLCGCTL9430 _MMIO(0x9430) -#define MSQDUNIT_CLKGATE_DIS REG_BIT(3) - -#define GEN6_GFXPAUSE _MMIO(0xA000) -#define GEN6_RPNSWREQ _MMIO(0xA008) -#define GEN6_TURBO_DISABLE (1 << 31) -#define GEN6_FREQUENCY(x) ((x) << 25) -#define HSW_FREQUENCY(x) ((x) << 24) -#define GEN9_FREQUENCY(x) ((x) << 23) -#define GEN6_OFFSET(x) ((x) << 19) -#define GEN6_AGGRESSIVE_TURBO (0 << 15) -#define GEN9_SW_REQ_UNSLICE_RATIO_SHIFT 23 -#define GEN9_IGNORE_SLICE_RATIO (0 << 0) - -#define GEN6_RC_VIDEO_FREQ _MMIO(0xA00C) -#define GEN6_RC_CONTROL _MMIO(0xA090) -#define GEN6_RC_CTL_RC6pp_ENABLE (1 << 16) -#define GEN6_RC_CTL_RC6p_ENABLE (1 << 17) -#define GEN6_RC_CTL_RC6_ENABLE (1 << 18) -#define GEN6_RC_CTL_RC1e_ENABLE (1 << 20) -#define GEN6_RC_CTL_RC7_ENABLE (1 << 22) -#define VLV_RC_CTL_CTX_RST_PARALLEL (1 << 24) -#define GEN7_RC_CTL_TO_MODE (1 << 28) -#define GEN6_RC_CTL_EI_MODE(x) ((x) << 27) -#define GEN6_RC_CTL_HW_ENABLE (1 << 31) -#define GEN6_RP_DOWN_TIMEOUT _MMIO(0xA010) -#define GEN6_RP_INTERRUPT_LIMITS _MMIO(0xA014) -#define GEN6_RPSTAT1 _MMIO(0xA01C) -#define GEN6_CAGF_SHIFT 8 -#define HSW_CAGF_SHIFT 7 -#define GEN9_CAGF_SHIFT 23 -#define GEN6_CAGF_MASK (0x7f << GEN6_CAGF_SHIFT) -#define HSW_CAGF_MASK (0x7f << HSW_CAGF_SHIFT) -#define GEN9_CAGF_MASK (0x1ff << GEN9_CAGF_SHIFT) -#define GEN6_RP_CONTROL _MMIO(0xA024) -#define GEN6_RP_MEDIA_TURBO (1 << 11) -#define GEN6_RP_MEDIA_MODE_MASK (3 << 9) -#define GEN6_RP_MEDIA_HW_TURBO_MODE (3 << 9) -#define GEN6_RP_MEDIA_HW_NORMAL_MODE (2 << 9) -#define GEN6_RP_MEDIA_HW_MODE (1 << 9) -#define GEN6_RP_MEDIA_SW_MODE (0 << 9) -#define GEN6_RP_MEDIA_IS_GFX (1 << 8) -#define GEN6_RP_ENABLE (1 << 7) -#define GEN6_RP_UP_IDLE_MIN (0x1 << 3) -#define GEN6_RP_UP_BUSY_AVG (0x2 << 3) -#define GEN6_RP_UP_BUSY_CONT (0x4 << 3) -#define GEN6_RP_DOWN_IDLE_AVG (0x2 << 0) -#define GEN6_RP_DOWN_IDLE_CONT (0x1 << 0) -#define GEN6_RPSWCTL_SHIFT 9 -#define GEN9_RPSWCTL_ENABLE (0x2 << GEN6_RPSWCTL_SHIFT) -#define GEN9_RPSWCTL_DISABLE (0x0 << GEN6_RPSWCTL_SHIFT) -#define GEN6_RP_UP_THRESHOLD _MMIO(0xA02C) -#define GEN6_RP_DOWN_THRESHOLD _MMIO(0xA030) -#define GEN6_RP_CUR_UP_EI _MMIO(0xA050) -#define GEN6_RP_EI_MASK 0xffffff -#define GEN6_CURICONT_MASK GEN6_RP_EI_MASK -#define GEN6_RP_CUR_UP _MMIO(0xA054) -#define GEN6_CURBSYTAVG_MASK GEN6_RP_EI_MASK -#define GEN6_RP_PREV_UP _MMIO(0xA058) -#define GEN6_RP_CUR_DOWN_EI _MMIO(0xA05C) -#define GEN6_CURIAVG_MASK GEN6_RP_EI_MASK -#define GEN6_RP_CUR_DOWN _MMIO(0xA060) -#define GEN6_RP_PREV_DOWN _MMIO(0xA064) -#define GEN6_RP_UP_EI _MMIO(0xA068) -#define GEN6_RP_DOWN_EI _MMIO(0xA06C) -#define GEN6_RP_IDLE_HYSTERSIS _MMIO(0xA070) -#define GEN6_RPDEUHWTC _MMIO(0xA080) -#define GEN6_RPDEUC _MMIO(0xA084) -#define GEN6_RPDEUCSW _MMIO(0xA088) -#define GEN6_RC_STATE _MMIO(0xA094) -#define RC_SW_TARGET_STATE_SHIFT 16 -#define RC_SW_TARGET_STATE_MASK (7 << RC_SW_TARGET_STATE_SHIFT) -#define GEN6_RC1_WAKE_RATE_LIMIT _MMIO(0xA098) -#define GEN6_RC6_WAKE_RATE_LIMIT _MMIO(0xA09C) -#define GEN6_RC6pp_WAKE_RATE_LIMIT _MMIO(0xA0A0) -#define GEN10_MEDIA_WAKE_RATE_LIMIT _MMIO(0xA0A0) -#define GEN6_RC_EVALUATION_INTERVAL _MMIO(0xA0A8) -#define GEN6_RC_IDLE_HYSTERSIS _MMIO(0xA0AC) -#define GEN6_RC_SLEEP _MMIO(0xA0B0) -#define GEN6_RCUBMABDTMR _MMIO(0xA0B0) -#define GEN6_RC1e_THRESHOLD _MMIO(0xA0B4) -#define GEN6_RC6_THRESHOLD _MMIO(0xA0B8) -#define GEN6_RC6p_THRESHOLD _MMIO(0xA0BC) -#define VLV_RCEDATA _MMIO(0xA0BC) -#define GEN6_RC6pp_THRESHOLD _MMIO(0xA0C0) -#define GEN6_PMINTRMSK _MMIO(0xA168) -#define GEN8_PMINTR_DISABLE_REDIRECT_TO_GUC (1 << 31) -#define ARAT_EXPIRED_INTRMSK (1 << 9) -#define GEN8_MISC_CTRL0 _MMIO(0xA180) -#define VLV_PWRDWNUPCTL _MMIO(0xA294) -#define GEN9_MEDIA_PG_IDLE_HYSTERESIS _MMIO(0xA0C4) -#define GEN9_RENDER_PG_IDLE_HYSTERESIS _MMIO(0xA0C8) -#define GEN9_PG_ENABLE _MMIO(0xA210) -#define GEN9_RENDER_PG_ENABLE REG_BIT(0) -#define GEN9_MEDIA_PG_ENABLE REG_BIT(1) -#define GEN11_MEDIA_SAMPLER_PG_ENABLE REG_BIT(2) -#define VDN_HCP_POWERGATE_ENABLE(n) REG_BIT(3 + 2 * (n)) -#define VDN_MFX_POWERGATE_ENABLE(n) REG_BIT(4 + 2 * (n)) -#define GEN8_PUSHBUS_CONTROL _MMIO(0xA248) -#define GEN8_PUSHBUS_ENABLE _MMIO(0xA250) -#define GEN8_PUSHBUS_SHIFT _MMIO(0xA25C) - #define VLV_CHICKEN_3 _MMIO(VLV_DISPLAY_BASE + 0x7040C) #define PIXEL_OVERLAP_CNT_MASK (3 << 30) #define PIXEL_OVERLAP_CNT_SHIFT 30 -#define GEN6_PMISR _MMIO(0x44020) -#define GEN6_PMIMR _MMIO(0x44024) /* rps_lock */ -#define GEN6_PMIIR _MMIO(0x44028) -#define GEN6_PMIER _MMIO(0x4402C) -#define GEN6_PM_MBOX_EVENT (1 << 25) -#define GEN6_PM_THERMAL_EVENT (1 << 24) - -/* - * For Gen11 these are in the upper word of the GPM_WGBOXPERF - * registers. Shifting is handled on accessing the imr and ier. - */ -#define GEN6_PM_RP_DOWN_TIMEOUT (1 << 6) -#define GEN6_PM_RP_UP_THRESHOLD (1 << 5) -#define GEN6_PM_RP_DOWN_THRESHOLD (1 << 4) -#define GEN6_PM_RP_UP_EI_EXPIRED (1 << 2) -#define GEN6_PM_RP_DOWN_EI_EXPIRED (1 << 1) -#define GEN6_PM_RPS_EVENTS (GEN6_PM_RP_UP_EI_EXPIRED | \ - GEN6_PM_RP_UP_THRESHOLD | \ - GEN6_PM_RP_DOWN_EI_EXPIRED | \ - GEN6_PM_RP_DOWN_THRESHOLD | \ - GEN6_PM_RP_DOWN_TIMEOUT) - -#define GEN7_GT_SCRATCH(i) _MMIO(0x4F100 + (i) * 4) -#define GEN7_GT_SCRATCH_REG_NUM 8 - -#define VLV_GTLC_SURVIVABILITY_REG _MMIO(0x130098) -#define VLV_GFX_CLK_STATUS_BIT (1 << 3) -#define VLV_GFX_CLK_FORCE_ON_BIT (1 << 2) - -#define GEN6_GT_GFX_RC6_LOCKED _MMIO(0x138104) -#define VLV_COUNTER_CONTROL _MMIO(0x138104) -#define VLV_COUNT_RANGE_HIGH (1 << 15) -#define VLV_MEDIA_RC0_COUNT_EN (1 << 5) -#define VLV_RENDER_RC0_COUNT_EN (1 << 4) -#define VLV_MEDIA_RC6_COUNT_EN (1 << 1) -#define VLV_RENDER_RC6_COUNT_EN (1 << 0) -#define GEN6_GT_GFX_RC6 _MMIO(0x138108) -#define VLV_GT_RENDER_RC6 _MMIO(0x138108) -#define VLV_GT_MEDIA_RC6 _MMIO(0x13810C) - -#define GEN6_GT_GFX_RC6p _MMIO(0x13810C) -#define GEN6_GT_GFX_RC6pp _MMIO(0x138110) -#define VLV_RENDER_C0_COUNT _MMIO(0x138118) -#define VLV_MEDIA_C0_COUNT _MMIO(0x13811C) - #define GEN6_PCODE_MAILBOX _MMIO(0x138124) #define GEN6_PCODE_READY (1 << 31) #define GEN6_PCODE_ERROR_MASK 0xFF @@ -9634,82 +6868,6 @@ enum { #define GEN6_PCODE_FREQ_RING_RATIO_SHIFT 16 #define GEN6_PCODE_DATA1 _MMIO(0x13812C) -#define GEN6_GT_CORE_STATUS _MMIO(0x138060) -#define GEN6_CORE_CPD_STATE_MASK (7 << 4) -#define GEN6_RCn_MASK 7 -#define GEN6_RC0 0 -#define GEN6_RC3 2 -#define GEN6_RC6 3 -#define GEN6_RC7 4 - -#define GEN8_GT_SLICE_INFO _MMIO(0x138064) -#define GEN8_LSLICESTAT_MASK 0x7 - -#define CHV_POWER_SS0_SIG1 _MMIO(0xa720) -#define CHV_POWER_SS1_SIG1 _MMIO(0xa728) -#define CHV_SS_PG_ENABLE (1 << 1) -#define CHV_EU08_PG_ENABLE (1 << 9) -#define CHV_EU19_PG_ENABLE (1 << 17) -#define CHV_EU210_PG_ENABLE (1 << 25) - -#define CHV_POWER_SS0_SIG2 _MMIO(0xa724) -#define CHV_POWER_SS1_SIG2 _MMIO(0xa72c) -#define CHV_EU311_PG_ENABLE (1 << 1) - -#define GEN9_SLICE_PGCTL_ACK(slice) _MMIO(0x804c + (slice) * 0x4) -#define GEN10_SLICE_PGCTL_ACK(slice) _MMIO(0x804c + ((slice) / 3) * 0x34 + \ - ((slice) % 3) * 0x4) -#define GEN9_PGCTL_SLICE_ACK (1 << 0) -#define GEN9_PGCTL_SS_ACK(subslice) (1 << (2 + (subslice) * 2)) -#define GEN10_PGCTL_VALID_SS_MASK(slice) ((slice) == 0 ? 0x7F : 0x1F) - -#define GEN9_SS01_EU_PGCTL_ACK(slice) _MMIO(0x805c + (slice) * 0x8) -#define GEN10_SS01_EU_PGCTL_ACK(slice) _MMIO(0x805c + ((slice) / 3) * 0x30 + \ - ((slice) % 3) * 0x8) -#define GEN9_SS23_EU_PGCTL_ACK(slice) _MMIO(0x8060 + (slice) * 0x8) -#define GEN10_SS23_EU_PGCTL_ACK(slice) _MMIO(0x8060 + ((slice) / 3) * 0x30 + \ - ((slice) % 3) * 0x8) -#define GEN9_PGCTL_SSA_EU08_ACK (1 << 0) -#define GEN9_PGCTL_SSA_EU19_ACK (1 << 2) -#define GEN9_PGCTL_SSA_EU210_ACK (1 << 4) -#define GEN9_PGCTL_SSA_EU311_ACK (1 << 6) -#define GEN9_PGCTL_SSB_EU08_ACK (1 << 8) -#define GEN9_PGCTL_SSB_EU19_ACK (1 << 10) -#define GEN9_PGCTL_SSB_EU210_ACK (1 << 12) -#define GEN9_PGCTL_SSB_EU311_ACK (1 << 14) - -#define GEN7_MISCCPCTL _MMIO(0x9424) -#define GEN7_DOP_CLOCK_GATE_ENABLE (1 << 0) -#define GEN8_DOP_CLOCK_GATE_CFCLK_ENABLE (1 << 2) -#define GEN8_DOP_CLOCK_GATE_GUC_ENABLE (1 << 4) -#define GEN8_DOP_CLOCK_GATE_MEDIA_ENABLE (1 << 6) - -#define GEN8_GARBCNTL _MMIO(0xB004) -#define GEN9_GAPS_TSV_CREDIT_DISABLE (1 << 7) -#define GEN11_ARBITRATION_PRIO_ORDER_MASK (0x3f << 22) -#define GEN11_HASH_CTRL_EXCL_MASK (0x7f << 0) -#define GEN11_HASH_CTRL_EXCL_BIT0 (1 << 0) - -#define GEN11_GLBLINVL _MMIO(0xB404) -#define GEN11_BANK_HASH_ADDR_EXCL_MASK (0x7f << 5) -#define GEN11_BANK_HASH_ADDR_EXCL_BIT0 (1 << 5) - -#define GEN10_DFR_RATIO_EN_AND_CHICKEN _MMIO(0x9550) -#define DFR_DISABLE (1 << 9) - -#define GEN11_GACB_PERF_CTRL _MMIO(0x4B80) -#define GEN11_HASH_CTRL_MASK (0x3 << 12 | 0xf << 0) -#define GEN11_HASH_CTRL_BIT0 (1 << 0) -#define GEN11_HASH_CTRL_BIT4 (1 << 12) - -#define GEN11_LSN_UNSLCVC _MMIO(0xB43C) -#define GEN11_LSN_UNSLCVC_GAFS_HALF_CL2_MAXALLOC (1 << 9) -#define GEN11_LSN_UNSLCVC_GAFS_HALF_SF_MAXALLOC (1 << 7) - -#define GEN10_SAMPLER_MODE _MMIO(0xE18C) -#define ENABLE_SMALLPL REG_BIT(15) -#define GEN11_SAMPLER_ENABLE_HEADLESS_MSG REG_BIT(5) - /* IVYBRIDGE DPF */ #define GEN7_L3CDERRST1(slice) _MMIO(0xB008 + (slice) * 0x200) /* L3CD Error Status 1 */ #define GEN7_L3CDERRST1_ROW_MASK (0x7ff << 14) @@ -9724,73 +6882,6 @@ enum { (((reg) & GEN7_L3CDERRST1_SUBBANK_MASK) >> 8) #define GEN7_L3CDERRST1_ENABLE (1 << 7) -#define GEN7_L3LOG(slice, i) _MMIO(0xB070 + (slice) * 0x200 + (i) * 4) -#define GEN7_L3LOG_SIZE 0x80 - -#define GEN7_HALF_SLICE_CHICKEN1 _MMIO(0xe100) /* IVB GT1 + VLV */ -#define GEN7_HALF_SLICE_CHICKEN1_GT2 _MMIO(0xf100) -#define GEN7_MAX_PS_THREAD_DEP (8 << 12) -#define GEN7_SINGLE_SUBSCAN_DISPATCH_ENABLE (1 << 10) -#define GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE (1 << 4) -#define GEN7_PSD_SINGLE_PORT_DISPATCH_ENABLE (1 << 3) - -#define GEN9_HALF_SLICE_CHICKEN5 _MMIO(0xe188) -#define GEN9_DG_MIRROR_FIX_ENABLE (1 << 5) -#define GEN9_CCS_TLB_PREFETCH_ENABLE (1 << 3) - -#define GEN8_ROW_CHICKEN _MMIO(0xe4f0) -#define FLOW_CONTROL_ENABLE REG_BIT(15) -#define UGM_BACKUP_MODE REG_BIT(13) -#define MDQ_ARBITRATION_MODE REG_BIT(12) -#define PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE REG_BIT(8) -#define STALL_DOP_GATING_DISABLE REG_BIT(5) -#define THROTTLE_12_5 REG_GENMASK(4, 2) -#define DISABLE_EARLY_EOT REG_BIT(1) - -#define GEN7_ROW_CHICKEN2 _MMIO(0xe4f4) -#define GEN12_DISABLE_READ_SUPPRESSION REG_BIT(15) -#define GEN12_DISABLE_EARLY_READ REG_BIT(14) -#define GEN12_ENABLE_LARGE_GRF_MODE REG_BIT(12) -#define GEN12_PUSH_CONST_DEREF_HOLD_DIS REG_BIT(8) - -#define LSC_CHICKEN_BIT_0 _MMIO(0xe7c8) -#define FORCE_1_SUB_MESSAGE_PER_FRAGMENT REG_BIT(15) -#define LSC_CHICKEN_BIT_0_UDW _MMIO(0xe7c8 + 4) -#define DIS_CHAIN_2XSIMD8 REG_BIT(55 - 32) -#define FORCE_SLM_FENCE_SCOPE_TO_TILE REG_BIT(42 - 32) -#define FORCE_UGM_FENCE_SCOPE_TO_TILE REG_BIT(41 - 32) -#define MAXREQS_PER_BANK REG_GENMASK(39 - 32, 37 - 32) -#define DISABLE_128B_EVICTION_COMMAND_UDW REG_BIT(36 - 32) - -#define GEN7_ROW_CHICKEN2_GT2 _MMIO(0xf4f4) -#define DOP_CLOCK_GATING_DISABLE (1 << 0) -#define PUSH_CONSTANT_DEREF_DISABLE (1 << 8) -#define GEN11_TDL_CLOCK_GATING_FIX_DISABLE (1 << 1) - -#define GEN9_ROW_CHICKEN4 _MMIO(0xe48c) -#define GEN12_DISABLE_GRF_CLEAR REG_BIT(13) -#define GEN12_DISABLE_TDL_PUSH REG_BIT(9) -#define GEN11_DIS_PICK_2ND_EU REG_BIT(7) -#define GEN12_DISABLE_HDR_PAST_PAYLOAD_HOLD_FIX REG_BIT(4) - -#define HSW_ROW_CHICKEN3 _MMIO(0xe49c) -#define HSW_ROW_CHICKEN3_L3_GLOBAL_ATOMICS_DISABLE (1 << 6) - -#define HALF_SLICE_CHICKEN2 _MMIO(0xe180) -#define GEN8_ST_PO_DISABLE (1 << 13) - -#define HALF_SLICE_CHICKEN3 _MMIO(0xe184) -#define HSW_SAMPLE_C_PERFORMANCE (1 << 9) -#define GEN8_CENTROID_PIXEL_OPT_DIS (1 << 8) -#define GEN9_DISABLE_OCL_OOB_SUPPRESS_LOGIC (1 << 5) -#define GEN8_SAMPLER_POWER_BYPASS_DIS (1 << 1) - -#define GEN9_HALF_SLICE_CHICKEN7 _MMIO(0xe194) -#define DG2_DISABLE_ROUND_ENABLE_ALLOW_FOR_SSLA REG_BIT(15) -#define GEN9_SAMPLER_HASH_COMPRESSED_READ_ADDR REG_BIT(8) -#define GEN9_ENABLE_YV12_BUGFIX REG_BIT(4) -#define GEN9_ENABLE_GPGPU_PREEMPTION REG_BIT(2) - /* Audio */ #define G4X_AUD_VID_DID _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x62020) #define INTEL_AUDIO_DEVCL 0x808629FB @@ -10839,149 +7930,6 @@ enum skl_power_gate { PORTTC1_PLL_ENABLE, \ PORTTC2_PLL_ENABLE) -#define _MG_REFCLKIN_CTL_PORT1 0x16892C -#define _MG_REFCLKIN_CTL_PORT2 0x16992C -#define _MG_REFCLKIN_CTL_PORT3 0x16A92C -#define _MG_REFCLKIN_CTL_PORT4 0x16B92C -#define MG_REFCLKIN_CTL_OD_2_MUX(x) ((x) << 8) -#define MG_REFCLKIN_CTL_OD_2_MUX_MASK (0x7 << 8) -#define MG_REFCLKIN_CTL(tc_port) _MMIO_PORT((tc_port), \ - _MG_REFCLKIN_CTL_PORT1, \ - _MG_REFCLKIN_CTL_PORT2) - -#define _MG_CLKTOP2_CORECLKCTL1_PORT1 0x1688D8 -#define _MG_CLKTOP2_CORECLKCTL1_PORT2 0x1698D8 -#define _MG_CLKTOP2_CORECLKCTL1_PORT3 0x16A8D8 -#define _MG_CLKTOP2_CORECLKCTL1_PORT4 0x16B8D8 -#define MG_CLKTOP2_CORECLKCTL1_B_DIVRATIO(x) ((x) << 16) -#define MG_CLKTOP2_CORECLKCTL1_B_DIVRATIO_MASK (0xff << 16) -#define MG_CLKTOP2_CORECLKCTL1_A_DIVRATIO(x) ((x) << 8) -#define MG_CLKTOP2_CORECLKCTL1_A_DIVRATIO_MASK (0xff << 8) -#define MG_CLKTOP2_CORECLKCTL1(tc_port) _MMIO_PORT((tc_port), \ - _MG_CLKTOP2_CORECLKCTL1_PORT1, \ - _MG_CLKTOP2_CORECLKCTL1_PORT2) - -#define _MG_CLKTOP2_HSCLKCTL_PORT1 0x1688D4 -#define _MG_CLKTOP2_HSCLKCTL_PORT2 0x1698D4 -#define _MG_CLKTOP2_HSCLKCTL_PORT3 0x16A8D4 -#define _MG_CLKTOP2_HSCLKCTL_PORT4 0x16B8D4 -#define MG_CLKTOP2_HSCLKCTL_CORE_INPUTSEL(x) ((x) << 16) -#define MG_CLKTOP2_HSCLKCTL_CORE_INPUTSEL_MASK (0x1 << 16) -#define MG_CLKTOP2_HSCLKCTL_TLINEDRV_CLKSEL(x) ((x) << 14) -#define MG_CLKTOP2_HSCLKCTL_TLINEDRV_CLKSEL_MASK (0x3 << 14) -#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_MASK (0x3 << 12) -#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_2 (0 << 12) -#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_3 (1 << 12) -#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_5 (2 << 12) -#define MG_CLKTOP2_HSCLKCTL_HSDIV_RATIO_7 (3 << 12) -#define MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO(x) ((x) << 8) -#define MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_SHIFT 8 -#define MG_CLKTOP2_HSCLKCTL_DSDIV_RATIO_MASK (0xf << 8) -#define MG_CLKTOP2_HSCLKCTL(tc_port) _MMIO_PORT((tc_port), \ - _MG_CLKTOP2_HSCLKCTL_PORT1, \ - _MG_CLKTOP2_HSCLKCTL_PORT2) - -#define _MG_PLL_DIV0_PORT1 0x168A00 -#define _MG_PLL_DIV0_PORT2 0x169A00 -#define _MG_PLL_DIV0_PORT3 0x16AA00 -#define _MG_PLL_DIV0_PORT4 0x16BA00 -#define MG_PLL_DIV0_FRACNEN_H (1 << 30) -#define MG_PLL_DIV0_FBDIV_FRAC_MASK (0x3fffff << 8) -#define MG_PLL_DIV0_FBDIV_FRAC_SHIFT 8 -#define MG_PLL_DIV0_FBDIV_FRAC(x) ((x) << 8) -#define MG_PLL_DIV0_FBDIV_INT_MASK (0xff << 0) -#define MG_PLL_DIV0_FBDIV_INT(x) ((x) << 0) -#define MG_PLL_DIV0(tc_port) _MMIO_PORT((tc_port), _MG_PLL_DIV0_PORT1, \ - _MG_PLL_DIV0_PORT2) - -#define _MG_PLL_DIV1_PORT1 0x168A04 -#define _MG_PLL_DIV1_PORT2 0x169A04 -#define _MG_PLL_DIV1_PORT3 0x16AA04 -#define _MG_PLL_DIV1_PORT4 0x16BA04 -#define MG_PLL_DIV1_IREF_NDIVRATIO(x) ((x) << 16) -#define MG_PLL_DIV1_DITHER_DIV_1 (0 << 12) -#define MG_PLL_DIV1_DITHER_DIV_2 (1 << 12) -#define MG_PLL_DIV1_DITHER_DIV_4 (2 << 12) -#define MG_PLL_DIV1_DITHER_DIV_8 (3 << 12) -#define MG_PLL_DIV1_NDIVRATIO(x) ((x) << 4) -#define MG_PLL_DIV1_FBPREDIV_MASK (0xf << 0) -#define MG_PLL_DIV1_FBPREDIV(x) ((x) << 0) -#define MG_PLL_DIV1(tc_port) _MMIO_PORT((tc_port), _MG_PLL_DIV1_PORT1, \ - _MG_PLL_DIV1_PORT2) - -#define _MG_PLL_LF_PORT1 0x168A08 -#define _MG_PLL_LF_PORT2 0x169A08 -#define _MG_PLL_LF_PORT3 0x16AA08 -#define _MG_PLL_LF_PORT4 0x16BA08 -#define MG_PLL_LF_TDCTARGETCNT(x) ((x) << 24) -#define MG_PLL_LF_AFCCNTSEL_256 (0 << 20) -#define MG_PLL_LF_AFCCNTSEL_512 (1 << 20) -#define MG_PLL_LF_GAINCTRL(x) ((x) << 16) -#define MG_PLL_LF_INT_COEFF(x) ((x) << 8) -#define MG_PLL_LF_PROP_COEFF(x) ((x) << 0) -#define MG_PLL_LF(tc_port) _MMIO_PORT((tc_port), _MG_PLL_LF_PORT1, \ - _MG_PLL_LF_PORT2) - -#define _MG_PLL_FRAC_LOCK_PORT1 0x168A0C -#define _MG_PLL_FRAC_LOCK_PORT2 0x169A0C -#define _MG_PLL_FRAC_LOCK_PORT3 0x16AA0C -#define _MG_PLL_FRAC_LOCK_PORT4 0x16BA0C -#define MG_PLL_FRAC_LOCK_TRUELOCK_CRIT_32 (1 << 18) -#define MG_PLL_FRAC_LOCK_EARLYLOCK_CRIT_32 (1 << 16) -#define MG_PLL_FRAC_LOCK_LOCKTHRESH(x) ((x) << 11) -#define MG_PLL_FRAC_LOCK_DCODITHEREN (1 << 10) -#define MG_PLL_FRAC_LOCK_FEEDFWRDCAL_EN (1 << 8) -#define MG_PLL_FRAC_LOCK_FEEDFWRDGAIN(x) ((x) << 0) -#define MG_PLL_FRAC_LOCK(tc_port) _MMIO_PORT((tc_port), \ - _MG_PLL_FRAC_LOCK_PORT1, \ - _MG_PLL_FRAC_LOCK_PORT2) - -#define _MG_PLL_SSC_PORT1 0x168A10 -#define _MG_PLL_SSC_PORT2 0x169A10 -#define _MG_PLL_SSC_PORT3 0x16AA10 -#define _MG_PLL_SSC_PORT4 0x16BA10 -#define MG_PLL_SSC_EN (1 << 28) -#define MG_PLL_SSC_TYPE(x) ((x) << 26) -#define MG_PLL_SSC_STEPLENGTH(x) ((x) << 16) -#define MG_PLL_SSC_STEPNUM(x) ((x) << 10) -#define MG_PLL_SSC_FLLEN (1 << 9) -#define MG_PLL_SSC_STEPSIZE(x) ((x) << 0) -#define MG_PLL_SSC(tc_port) _MMIO_PORT((tc_port), _MG_PLL_SSC_PORT1, \ - _MG_PLL_SSC_PORT2) - -#define _MG_PLL_BIAS_PORT1 0x168A14 -#define _MG_PLL_BIAS_PORT2 0x169A14 -#define _MG_PLL_BIAS_PORT3 0x16AA14 -#define _MG_PLL_BIAS_PORT4 0x16BA14 -#define MG_PLL_BIAS_BIAS_GB_SEL(x) ((x) << 30) -#define MG_PLL_BIAS_BIAS_GB_SEL_MASK (0x3 << 30) -#define MG_PLL_BIAS_INIT_DCOAMP(x) ((x) << 24) -#define MG_PLL_BIAS_INIT_DCOAMP_MASK (0x3f << 24) -#define MG_PLL_BIAS_BIAS_BONUS(x) ((x) << 16) -#define MG_PLL_BIAS_BIAS_BONUS_MASK (0xff << 16) -#define MG_PLL_BIAS_BIASCAL_EN (1 << 15) -#define MG_PLL_BIAS_CTRIM(x) ((x) << 8) -#define MG_PLL_BIAS_CTRIM_MASK (0x1f << 8) -#define MG_PLL_BIAS_VREF_RDAC(x) ((x) << 5) -#define MG_PLL_BIAS_VREF_RDAC_MASK (0x7 << 5) -#define MG_PLL_BIAS_IREFTRIM(x) ((x) << 0) -#define MG_PLL_BIAS_IREFTRIM_MASK (0x1f << 0) -#define MG_PLL_BIAS(tc_port) _MMIO_PORT((tc_port), _MG_PLL_BIAS_PORT1, \ - _MG_PLL_BIAS_PORT2) - -#define _MG_PLL_TDC_COLDST_BIAS_PORT1 0x168A18 -#define _MG_PLL_TDC_COLDST_BIAS_PORT2 0x169A18 -#define _MG_PLL_TDC_COLDST_BIAS_PORT3 0x16AA18 -#define _MG_PLL_TDC_COLDST_BIAS_PORT4 0x16BA18 -#define MG_PLL_TDC_COLDST_IREFINT_EN (1 << 27) -#define MG_PLL_TDC_COLDST_REFBIAS_START_PULSE_W(x) ((x) << 17) -#define MG_PLL_TDC_COLDST_COLDSTART (1 << 16) -#define MG_PLL_TDC_TDCOVCCORR_EN (1 << 2) -#define MG_PLL_TDC_TDCSEL(x) ((x) << 0) -#define MG_PLL_TDC_COLDST_BIAS(tc_port) _MMIO_PORT((tc_port), \ - _MG_PLL_TDC_COLDST_BIAS_PORT1, \ - _MG_PLL_TDC_COLDST_BIAS_PORT2) - #define _ICL_DPLL0_CFGCR0 0x164000 #define _ICL_DPLL1_CFGCR0 0x164080 #define ICL_DPLL_CFGCR0(pll) _MMIO_PLL(pll, _ICL_DPLL0_CFGCR0, \ @@ -12599,39 +9547,6 @@ enum skl_power_gate { #define MIPI_READ_DATA_VALID(port) _MMIO_MIPI(port, _MIPIA_READ_DATA_VALID, _MIPIC_READ_DATA_VALID) #define READ_DATA_VALID(n) (1 << (n)) -/* MOCS (Memory Object Control State) registers */ -#define GEN9_LNCFCMOCS(i) _MMIO(0xb020 + (i) * 4) /* L3 Cache Control */ -#define GEN9_LNCFCMOCS_REG_COUNT 32 - -#define __GEN9_RCS0_MOCS0 0xc800 -#define GEN9_GFX_MOCS(i) _MMIO(__GEN9_RCS0_MOCS0 + (i) * 4) -#define __GEN9_VCS0_MOCS0 0xc900 -#define GEN9_MFX0_MOCS(i) _MMIO(__GEN9_VCS0_MOCS0 + (i) * 4) -#define __GEN9_VCS1_MOCS0 0xca00 -#define GEN9_MFX1_MOCS(i) _MMIO(__GEN9_VCS1_MOCS0 + (i) * 4) -#define __GEN9_VECS0_MOCS0 0xcb00 -#define GEN9_VEBOX_MOCS(i) _MMIO(__GEN9_VECS0_MOCS0 + (i) * 4) -#define __GEN9_BCS0_MOCS0 0xcc00 -#define GEN9_BLT_MOCS(i) _MMIO(__GEN9_BCS0_MOCS0 + (i) * 4) -#define __GEN11_VCS2_MOCS0 0x10000 -#define GEN11_MFX2_MOCS(i) _MMIO(__GEN11_VCS2_MOCS0 + (i) * 4) - -#define GEN9_SCRATCH_LNCF1 _MMIO(0xb008) -#define GEN9_LNCF_NONIA_COHERENT_ATOMICS_ENABLE REG_BIT(0) - -#define GEN9_SCRATCH1 _MMIO(0xb11c) -#define EVICTION_PERF_FIX_ENABLE REG_BIT(8) - -#define GEN10_SCRATCH_LNCF2 _MMIO(0xb0a0) -#define PMFLUSHDONE_LNICRSDROP (1 << 20) -#define PMFLUSH_GAPL3UNBLOCK (1 << 21) -#define PMFLUSHDONE_LNEBLK (1 << 22) - -#define XEHP_L3NODEARBCFG _MMIO(0xb0b4) -#define XEHP_LNESPARE REG_BIT(19) - -#define GEN12_GLOBAL_MOCS(i) _MMIO(0x4000 + (i) * 4) /* Global MOCS regs */ - #define GEN12_GSMBASE _MMIO(0x108100) #define GEN12_DSMBASE _MMIO(0x1080C0) @@ -12640,17 +9555,6 @@ enum skl_power_gate { #define SGGI_DIS REG_BIT(15) #define SGR_DIS REG_BIT(13) -/* gamt regs */ -#define GEN8_L3_LRA_1_GPGPU _MMIO(0x4dd4) -#define GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_BDW 0x67F1427F /* max/min for LRA1/2 */ -#define GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_CHV 0x5FF101FF /* max/min for LRA1/2 */ -#define GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_SKL 0x67F1427F /* " " */ -#define GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_BXT 0x5FF101FF /* " " */ - -#define MMCD_MISC_CTRL _MMIO(0x4ddc) /* skl+ */ -#define MMCD_PCLA (1 << 31) -#define MMCD_HOTSPOT_EN (1 << 27) - #define _ICL_PHY_MISC_A 0x64C00 #define _ICL_PHY_MISC_B 0x64C04 #define ICL_PHY_MISC(port) _MMIO_PORT(port, _ICL_PHY_MISC_A, \ @@ -12994,6 +9898,14 @@ enum skl_power_gate { #define TCSS_DDI_STATUS_HPD_LIVE_STATUS_TBT REG_BIT(1) #define TCSS_DDI_STATUS_HPD_LIVE_STATUS_ALT REG_BIT(0) +#define PRIMARY_SPI_TRIGGER _MMIO(0x102040) +#define PRIMARY_SPI_ADDRESS _MMIO(0x102080) +#define PRIMARY_SPI_REGIONID _MMIO(0x102084) +#define SPI_STATIC_REGIONS _MMIO(0x102090) +#define OPTIONROM_SPI_REGIONID_MASK REG_GENMASK(7, 0) +#define OROM_OFFSET _MMIO(0x1020c0) +#define OROM_OFFSET_MASK REG_GENMASK(20, 16) + /* This register controls the Display State Buffer (DSB) engines. */ #define _DSBSL_INSTANCE_BASE 0x70B00 #define DSBSL_INSTANCE(pipe, id) (_DSBSL_INSTANCE_BASE + \ @@ -13016,7 +9928,4 @@ enum skl_power_gate { #define CLKGATE_DIS_MISC _MMIO(0x46534) #define CLKGATE_DIS_MISC_DMASC_GATING_DIS REG_BIT(21) -#define SLICE_COMMON_ECO_CHICKEN1 _MMIO(0x731C) -#define MSC_MSAA_REODER_BUF_BYPASS_DISABLE REG_BIT(14) - #endif /* _I915_REG_H_ */ diff --git a/drivers/gpu/drm/i915/i915_reg_defs.h b/drivers/gpu/drm/i915/i915_reg_defs.h new file mode 100644 index 000000000000..34d37bbf08cd --- /dev/null +++ b/drivers/gpu/drm/i915/i915_reg_defs.h @@ -0,0 +1,101 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __I915_REG_DEFS__ +#define __I915_REG_DEFS__ + +#include <linux/bitfield.h> +#include <linux/bits.h> + +/** + * REG_BIT() - Prepare a u32 bit value + * @__n: 0-based bit number + * + * Local wrapper for BIT() to force u32, with compile time checks. + * + * @return: Value with bit @__n set. + */ +#define REG_BIT(__n) \ + ((u32)(BIT(__n) + \ + BUILD_BUG_ON_ZERO(__is_constexpr(__n) && \ + ((__n) < 0 || (__n) > 31)))) + +/** + * REG_GENMASK() - Prepare a continuous u32 bitmask + * @__high: 0-based high bit + * @__low: 0-based low bit + * + * Local wrapper for GENMASK() to force u32, with compile time checks. + * + * @return: Continuous bitmask from @__high to @__low, inclusive. + */ +#define REG_GENMASK(__high, __low) \ + ((u32)(GENMASK(__high, __low) + \ + BUILD_BUG_ON_ZERO(__is_constexpr(__high) && \ + __is_constexpr(__low) && \ + ((__low) < 0 || (__high) > 31 || (__low) > (__high))))) + +/* + * Local integer constant expression version of is_power_of_2(). + */ +#define IS_POWER_OF_2(__x) ((__x) && (((__x) & ((__x) - 1)) == 0)) + +/** + * REG_FIELD_PREP() - Prepare a u32 bitfield value + * @__mask: shifted mask defining the field's length and position + * @__val: value to put in the field + * + * Local copy of FIELD_PREP() to generate an integer constant expression, force + * u32 and for consistency with REG_FIELD_GET(), REG_BIT() and REG_GENMASK(). + * + * @return: @__val masked and shifted into the field defined by @__mask. + */ +#define REG_FIELD_PREP(__mask, __val) \ + ((u32)((((typeof(__mask))(__val) << __bf_shf(__mask)) & (__mask)) + \ + BUILD_BUG_ON_ZERO(!__is_constexpr(__mask)) + \ + BUILD_BUG_ON_ZERO((__mask) == 0 || (__mask) > U32_MAX) + \ + BUILD_BUG_ON_ZERO(!IS_POWER_OF_2((__mask) + (1ULL << __bf_shf(__mask)))) + \ + BUILD_BUG_ON_ZERO(__builtin_choose_expr(__is_constexpr(__val), (~((__mask) >> __bf_shf(__mask)) & (__val)), 0)))) + +/** + * REG_FIELD_GET() - Extract a u32 bitfield value + * @__mask: shifted mask defining the field's length and position + * @__val: value to extract the bitfield value from + * + * Local wrapper for FIELD_GET() to force u32 and for consistency with + * REG_FIELD_PREP(), REG_BIT() and REG_GENMASK(). + * + * @return: Masked and shifted value of the field defined by @__mask in @__val. + */ +#define REG_FIELD_GET(__mask, __val) ((u32)FIELD_GET(__mask, __val)) + +typedef struct { + u32 reg; +} i915_reg_t; + +#define _MMIO(r) ((const i915_reg_t){ .reg = (r) }) + +#define INVALID_MMIO_REG _MMIO(0) + +static __always_inline u32 i915_mmio_reg_offset(i915_reg_t reg) +{ + return reg.reg; +} + +static inline bool i915_mmio_reg_equal(i915_reg_t a, i915_reg_t b) +{ + return i915_mmio_reg_offset(a) == i915_mmio_reg_offset(b); +} + +static inline bool i915_mmio_reg_valid(i915_reg_t reg) +{ + return !i915_mmio_reg_equal(reg, INVALID_MMIO_REG); +} + +#define VLV_DISPLAY_BASE 0x180000 + +#define GEN12_SFC_DONE_MAX 4 + +#endif /* __I915_REG_DEFS__ */ diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index 76cf5ac91e94..5d94f86940f7 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -36,6 +36,7 @@ #include "gt/intel_context.h" #include "gt/intel_engine.h" #include "gt/intel_engine_heartbeat.h" +#include "gt/intel_engine_regs.h" #include "gt/intel_gpu_commands.h" #include "gt/intel_reset.h" #include "gt/intel_ring.h" diff --git a/drivers/gpu/drm/i915/i915_scatterlist.c b/drivers/gpu/drm/i915/i915_scatterlist.c index 41f2adb6a583..159571b9bd24 100644 --- a/drivers/gpu/drm/i915/i915_scatterlist.c +++ b/drivers/gpu/drm/i915/i915_scatterlist.c @@ -5,10 +5,9 @@ */ #include "i915_scatterlist.h" - -#include "i915_buddy.h" #include "i915_ttm_buddy_manager.h" +#include <drm/drm_buddy.h> #include <drm/drm_mm.h> #include <linux/slab.h> @@ -153,9 +152,9 @@ struct i915_refct_sgt *i915_rsgt_from_buddy_resource(struct ttm_resource *res, struct i915_ttm_buddy_resource *bman_res = to_ttm_buddy_resource(res); const u64 size = res->num_pages << PAGE_SHIFT; const u64 max_segment = rounddown(UINT_MAX, PAGE_SIZE); - struct i915_buddy_mm *mm = bman_res->mm; + struct drm_buddy *mm = bman_res->mm; struct list_head *blocks = &bman_res->blocks; - struct i915_buddy_block *block; + struct drm_buddy_block *block; struct i915_refct_sgt *rsgt; struct scatterlist *sg; struct sg_table *st; @@ -181,8 +180,8 @@ struct i915_refct_sgt *i915_rsgt_from_buddy_resource(struct ttm_resource *res, list_for_each_entry(block, blocks, link) { u64 block_size, offset; - block_size = min_t(u64, size, i915_buddy_block_size(mm, block)); - offset = i915_buddy_block_offset(block); + block_size = min_t(u64, size, drm_buddy_block_size(mm, block)); + offset = drm_buddy_block_offset(block); while (block_size) { u64 len; diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c index f7b55f34dba8..889f5b7dc78e 100644 --- a/drivers/gpu/drm/i915/i915_suspend.c +++ b/drivers/gpu/drm/i915/i915_suspend.c @@ -32,6 +32,7 @@ #include "i915_drv.h" #include "i915_reg.h" #include "i915_suspend.h" +#include "intel_pci_config.h" static void intel_save_swf(struct drm_i915_private *dev_priv) { diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c index fae4d1f4f275..a4d1759375b9 100644 --- a/drivers/gpu/drm/i915/i915_sysfs.c +++ b/drivers/gpu/drm/i915/i915_sysfs.c @@ -30,6 +30,7 @@ #include <linux/stat.h> #include <linux/sysfs.h> +#include "gt/intel_gt_regs.h" #include "gt/intel_rc6.h" #include "gt/intel_rps.h" #include "gt/sysfs_engines.h" diff --git a/drivers/gpu/drm/i915/i915_ttm_buddy_manager.c b/drivers/gpu/drm/i915/i915_ttm_buddy_manager.c index d59fbb019032..247714bab044 100644 --- a/drivers/gpu/drm/i915/i915_ttm_buddy_manager.c +++ b/drivers/gpu/drm/i915/i915_ttm_buddy_manager.c @@ -8,14 +8,15 @@ #include <drm/ttm/ttm_bo_driver.h> #include <drm/ttm/ttm_placement.h> +#include <drm/drm_buddy.h> + #include "i915_ttm_buddy_manager.h" -#include "i915_buddy.h" #include "i915_gem.h" struct i915_ttm_buddy_manager { struct ttm_resource_manager manager; - struct i915_buddy_mm mm; + struct drm_buddy mm; struct list_head reserved; struct mutex lock; u64 default_page_size; @@ -34,7 +35,7 @@ static int i915_ttm_buddy_man_alloc(struct ttm_resource_manager *man, { struct i915_ttm_buddy_manager *bman = to_buddy_manager(man); struct i915_ttm_buddy_resource *bman_res; - struct i915_buddy_mm *mm = &bman->mm; + struct drm_buddy *mm = &bman->mm; unsigned long n_pages; unsigned int min_order; u64 min_page_size; @@ -73,7 +74,7 @@ static int i915_ttm_buddy_man_alloc(struct ttm_resource_manager *man, n_pages = size >> ilog2(mm->chunk_size); do { - struct i915_buddy_block *block; + struct drm_buddy_block *block; unsigned int order; order = fls(n_pages) - 1; @@ -82,7 +83,7 @@ static int i915_ttm_buddy_man_alloc(struct ttm_resource_manager *man, do { mutex_lock(&bman->lock); - block = i915_buddy_alloc(mm, order); + block = drm_buddy_alloc_blocks(mm, order); mutex_unlock(&bman->lock); if (!IS_ERR(block)) break; @@ -106,9 +107,10 @@ static int i915_ttm_buddy_man_alloc(struct ttm_resource_manager *man, err_free_blocks: mutex_lock(&bman->lock); - i915_buddy_free_list(mm, &bman_res->blocks); + drm_buddy_free_list(mm, &bman_res->blocks); mutex_unlock(&bman->lock); err_free_res: + ttm_resource_fini(man, &bman_res->base); kfree(bman_res); return err; } @@ -120,9 +122,10 @@ static void i915_ttm_buddy_man_free(struct ttm_resource_manager *man, struct i915_ttm_buddy_manager *bman = to_buddy_manager(man); mutex_lock(&bman->lock); - i915_buddy_free_list(&bman->mm, &bman_res->blocks); + drm_buddy_free_list(&bman->mm, &bman_res->blocks); mutex_unlock(&bman->lock); + ttm_resource_fini(man, res); kfree(bman_res); } @@ -130,17 +133,17 @@ static void i915_ttm_buddy_man_debug(struct ttm_resource_manager *man, struct drm_printer *printer) { struct i915_ttm_buddy_manager *bman = to_buddy_manager(man); - struct i915_buddy_block *block; + struct drm_buddy_block *block; mutex_lock(&bman->lock); drm_printf(printer, "default_page_size: %lluKiB\n", bman->default_page_size >> 10); - i915_buddy_print(&bman->mm, printer); + drm_buddy_print(&bman->mm, printer); drm_printf(printer, "reserved:\n"); list_for_each_entry(block, &bman->reserved, link) - i915_buddy_block_print(&bman->mm, block, printer); + drm_buddy_block_print(&bman->mm, block, printer); mutex_unlock(&bman->lock); } @@ -190,7 +193,7 @@ int i915_ttm_buddy_man_init(struct ttm_device *bdev, if (!bman) return -ENOMEM; - err = i915_buddy_init(&bman->mm, size, chunk_size); + err = drm_buddy_init(&bman->mm, size, chunk_size); if (err) goto err_free_bman; @@ -202,7 +205,7 @@ int i915_ttm_buddy_man_init(struct ttm_device *bdev, man = &bman->manager; man->use_tt = use_tt; man->func = &i915_ttm_buddy_manager_func; - ttm_resource_manager_init(man, bman->mm.size >> PAGE_SHIFT); + ttm_resource_manager_init(man, bdev, bman->mm.size >> PAGE_SHIFT); ttm_resource_manager_set_used(man, true); ttm_set_driver_manager(bdev, type, man); @@ -228,7 +231,7 @@ int i915_ttm_buddy_man_fini(struct ttm_device *bdev, unsigned int type) { struct ttm_resource_manager *man = ttm_manager_type(bdev, type); struct i915_ttm_buddy_manager *bman = to_buddy_manager(man); - struct i915_buddy_mm *mm = &bman->mm; + struct drm_buddy *mm = &bman->mm; int ret; ttm_resource_manager_set_used(man, false); @@ -240,8 +243,8 @@ int i915_ttm_buddy_man_fini(struct ttm_device *bdev, unsigned int type) ttm_set_driver_manager(bdev, type, NULL); mutex_lock(&bman->lock); - i915_buddy_free_list(mm, &bman->reserved); - i915_buddy_fini(mm); + drm_buddy_free_list(mm, &bman->reserved); + drm_buddy_fini(mm); mutex_unlock(&bman->lock); ttm_resource_manager_cleanup(man); @@ -264,11 +267,11 @@ int i915_ttm_buddy_man_reserve(struct ttm_resource_manager *man, u64 start, u64 size) { struct i915_ttm_buddy_manager *bman = to_buddy_manager(man); - struct i915_buddy_mm *mm = &bman->mm; + struct drm_buddy *mm = &bman->mm; int ret; mutex_lock(&bman->lock); - ret = i915_buddy_alloc_range(mm, &bman->reserved, start, size); + ret = drm_buddy_alloc_range(mm, &bman->reserved, start, size); mutex_unlock(&bman->lock); return ret; diff --git a/drivers/gpu/drm/i915/i915_ttm_buddy_manager.h b/drivers/gpu/drm/i915/i915_ttm_buddy_manager.h index 0722d33f3e14..312077941411 100644 --- a/drivers/gpu/drm/i915/i915_ttm_buddy_manager.h +++ b/drivers/gpu/drm/i915/i915_ttm_buddy_manager.h @@ -13,7 +13,7 @@ struct ttm_device; struct ttm_resource_manager; -struct i915_buddy_mm; +struct drm_buddy; /** * struct i915_ttm_buddy_resource @@ -28,7 +28,7 @@ struct i915_buddy_mm; struct i915_ttm_buddy_resource { struct ttm_resource base; struct list_head blocks; - struct i915_buddy_mm *mm; + struct drm_buddy *mm; }; /** diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h index 7a5925072466..bfafd0afd117 100644 --- a/drivers/gpu/drm/i915/i915_utils.h +++ b/drivers/gpu/drm/i915/i915_utils.h @@ -37,21 +37,6 @@ struct timer_list; #define FDO_BUG_URL "https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs" -#undef WARN_ON -/* Many gcc seem to no see through this and fall over :( */ -#if 0 -#define WARN_ON(x) ({ \ - bool __i915_warn_cond = (x); \ - if (__builtin_constant_p(__i915_warn_cond)) \ - BUILD_BUG_ON(__i915_warn_cond); \ - WARN(__i915_warn_cond, "WARN_ON(" #x ")"); }) -#else -#define WARN_ON(x) WARN((x), "%s", "WARN_ON(" __stringify(x) ")") -#endif - -#undef WARN_ON_ONCE -#define WARN_ON_ONCE(x) WARN_ONCE((x), "%s", "WARN_ON_ONCE(" __stringify(x) ")") - #define MISSING_CASE(x) WARN(1, "Missing case (%s == %ld)\n", \ __stringify(x), (long)(x)) diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index c0d6d5526abe..68cf1d392250 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -26,14 +26,15 @@ #include <drm/drm_gem.h> #include "display/intel_frontbuffer.h" - #include "gem/i915_gem_lmem.h" +#include "gem/i915_gem_tiling.h" #include "gt/intel_engine.h" #include "gt/intel_engine_heartbeat.h" #include "gt/intel_gt.h" #include "gt/intel_gt_requests.h" #include "i915_drv.h" +#include "i915_gem_evict.h" #include "i915_sw_fence_work.h" #include "i915_trace.h" #include "i915_vma.h" diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c index 04fd266d70e2..94da5aa37391 100644 --- a/drivers/gpu/drm/i915/intel_device_info.c +++ b/drivers/gpu/drm/i915/intel_device_info.c @@ -114,7 +114,7 @@ void intel_device_info_print_static(const struct intel_device_info *info, DEV_INFO_FOR_EACH_FLAG(PRINT_FLAG); #undef PRINT_FLAG -#define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, yesno(info->display.name)); +#define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, yesno(info->display.name)) DEV_INFO_DISPLAY_FOR_EACH_FLAG(PRINT_FLAG); #undef PRINT_FLAG } @@ -170,6 +170,10 @@ static const u16 subplatform_portf_ids[] = { INTEL_ICL_PORT_F_IDS(0), }; +static const u16 subplatform_n_ids[] = { + INTEL_ADLN_IDS(0), +}; + static const u16 subplatform_rpls_ids[] = { INTEL_RPLS_IDS(0), }; @@ -210,6 +214,9 @@ void intel_device_info_subplatform_init(struct drm_i915_private *i915) } else if (find_devid(devid, subplatform_portf_ids, ARRAY_SIZE(subplatform_portf_ids))) { mask = BIT(INTEL_SUBPLATFORM_PORTF); + } else if (find_devid(devid, subplatform_n_ids, + ARRAY_SIZE(subplatform_n_ids))) { + mask = BIT(INTEL_SUBPLATFORM_N); } else if (find_devid(devid, subplatform_rpls_ids, ARRAY_SIZE(subplatform_rpls_ids))) { mask = BIT(INTEL_SUBPLATFORM_RPL_S); @@ -328,6 +335,7 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv) "Display fused off, disabling\n"); info->display.pipe_mask = 0; info->display.cpu_transcoder_mask = 0; + info->display.fbc_mask = 0; } else if (fuse_strap & IVB_PIPE_C_DISABLE) { drm_info(&dev_priv->drm, "PipeC fused off\n"); info->display.pipe_mask &= ~BIT(PIPE_C); @@ -339,6 +347,7 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv) if (dfsm & SKL_DFSM_PIPE_A_DISABLE) { info->display.pipe_mask &= ~BIT(PIPE_A); info->display.cpu_transcoder_mask &= ~BIT(TRANSCODER_A); + info->display.fbc_mask &= ~BIT(INTEL_FBC_A); } if (dfsm & SKL_DFSM_PIPE_B_DISABLE) { info->display.pipe_mask &= ~BIT(PIPE_B); @@ -359,7 +368,7 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv) info->display.has_hdcp = 0; if (dfsm & SKL_DFSM_DISPLAY_PM_DISABLE) - info->display.has_fbc = 0; + info->display.fbc_mask = 0; if (DISPLAY_VER(dev_priv) >= 11 && (dfsm & ICL_DFSM_DMC_DISABLE)) info->display.has_dmc = 0; diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h index 78597d382445..3699b1c539ea 100644 --- a/drivers/gpu/drm/i915/intel_device_info.h +++ b/drivers/gpu/drm/i915/intel_device_info.h @@ -113,6 +113,9 @@ enum intel_platform { /* ADL-S */ #define INTEL_SUBPLATFORM_RPL_S 0 +/* ADL-P */ +#define INTEL_SUBPLATFORM_N 0 + enum intel_ppgtt_type { INTEL_PPGTT_NONE = I915_GEM_PPGTT_NONE, INTEL_PPGTT_ALIASING = I915_GEM_PPGTT_ALIASING, @@ -156,7 +159,6 @@ enum intel_ppgtt_type { func(has_dp_mst); \ func(has_dsb); \ func(has_dsc); \ - func(has_fbc); \ func(has_fpga_dbg); \ func(has_gmch); \ func(has_hdcp); \ @@ -206,6 +208,7 @@ struct intel_device_info { u8 pipe_mask; u8 cpu_transcoder_mask; + u8 fbc_mask; u8 abox_mask; #define DEFINE_FLAG(name) u8 name:1 diff --git a/drivers/gpu/drm/i915/intel_dram.c b/drivers/gpu/drm/i915/intel_dram.c index 84bb212bae4b..723bd0411a0e 100644 --- a/drivers/gpu/drm/i915/intel_dram.c +++ b/drivers/gpu/drm/i915/intel_dram.c @@ -4,6 +4,7 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "intel_dram.h" #include "intel_pcode.h" @@ -389,10 +390,8 @@ static int icl_pcode_read_mem_global_info(struct drm_i915_private *dev_priv) u32 val = 0; int ret; - ret = sandybridge_pcode_read(dev_priv, - ICL_PCODE_MEM_SUBSYSYSTEM_INFO | - ICL_PCODE_MEM_SS_READ_GLOBAL_INFO, - &val, NULL); + ret = snb_pcode_read(dev_priv, ICL_PCODE_MEM_SUBSYSYSTEM_INFO | + ICL_PCODE_MEM_SS_READ_GLOBAL_INFO, &val, NULL); if (ret) return ret; diff --git a/drivers/gpu/drm/i915/intel_pci_config.h b/drivers/gpu/drm/i915/intel_pci_config.h new file mode 100644 index 000000000000..12cd9d4f23de --- /dev/null +++ b/drivers/gpu/drm/i915/intel_pci_config.h @@ -0,0 +1,85 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef __INTEL_PCI_CONFIG_H__ +#define __INTEL_PCI_CONFIG_H__ + +/* BSM in include/drm/i915_drm.h */ + +#define MCHBAR_I915 0x44 +#define MCHBAR_I965 0x48 +#define MCHBAR_SIZE (4 * 4096) + +#define DEVEN 0x54 +#define DEVEN_MCHBAR_EN (1 << 28) + +#define HPLLCC 0xc0 /* 85x only */ +#define GC_CLOCK_CONTROL_MASK (0x7 << 0) +#define GC_CLOCK_133_200 (0 << 0) +#define GC_CLOCK_100_200 (1 << 0) +#define GC_CLOCK_100_133 (2 << 0) +#define GC_CLOCK_133_266 (3 << 0) +#define GC_CLOCK_133_200_2 (4 << 0) +#define GC_CLOCK_133_266_2 (5 << 0) +#define GC_CLOCK_166_266 (6 << 0) +#define GC_CLOCK_166_250 (7 << 0) + +#define I915_GDRST 0xc0 +#define GRDOM_FULL (0 << 2) +#define GRDOM_RENDER (1 << 2) +#define GRDOM_MEDIA (3 << 2) +#define GRDOM_MASK (3 << 2) +#define GRDOM_RESET_STATUS (1 << 1) +#define GRDOM_RESET_ENABLE (1 << 0) + +/* BSpec only has register offset, PCI device and bit found empirically */ +#define I830_CLOCK_GATE 0xc8 /* device 0 */ +#define I830_L2_CACHE_CLOCK_GATE_DISABLE (1 << 2) + +#define GCDGMBUS 0xcc + +#define GCFGC2 0xda +#define GCFGC 0xf0 /* 915+ only */ +#define GC_LOW_FREQUENCY_ENABLE (1 << 7) +#define GC_DISPLAY_CLOCK_190_200_MHZ (0 << 4) +#define GC_DISPLAY_CLOCK_333_320_MHZ (4 << 4) +#define GC_DISPLAY_CLOCK_267_MHZ_PNV (0 << 4) +#define GC_DISPLAY_CLOCK_333_MHZ_PNV (1 << 4) +#define GC_DISPLAY_CLOCK_444_MHZ_PNV (2 << 4) +#define GC_DISPLAY_CLOCK_200_MHZ_PNV (5 << 4) +#define GC_DISPLAY_CLOCK_133_MHZ_PNV (6 << 4) +#define GC_DISPLAY_CLOCK_167_MHZ_PNV (7 << 4) +#define GC_DISPLAY_CLOCK_MASK (7 << 4) +#define GM45_GC_RENDER_CLOCK_MASK (0xf << 0) +#define GM45_GC_RENDER_CLOCK_266_MHZ (8 << 0) +#define GM45_GC_RENDER_CLOCK_320_MHZ (9 << 0) +#define GM45_GC_RENDER_CLOCK_400_MHZ (0xb << 0) +#define GM45_GC_RENDER_CLOCK_533_MHZ (0xc << 0) +#define I965_GC_RENDER_CLOCK_MASK (0xf << 0) +#define I965_GC_RENDER_CLOCK_267_MHZ (2 << 0) +#define I965_GC_RENDER_CLOCK_333_MHZ (3 << 0) +#define I965_GC_RENDER_CLOCK_444_MHZ (4 << 0) +#define I965_GC_RENDER_CLOCK_533_MHZ (5 << 0) +#define I945_GC_RENDER_CLOCK_MASK (7 << 0) +#define I945_GC_RENDER_CLOCK_166_MHZ (0 << 0) +#define I945_GC_RENDER_CLOCK_200_MHZ (1 << 0) +#define I945_GC_RENDER_CLOCK_250_MHZ (3 << 0) +#define I945_GC_RENDER_CLOCK_400_MHZ (5 << 0) +#define I915_GC_RENDER_CLOCK_MASK (7 << 0) +#define I915_GC_RENDER_CLOCK_166_MHZ (0 << 0) +#define I915_GC_RENDER_CLOCK_200_MHZ (1 << 0) +#define I915_GC_RENDER_CLOCK_333_MHZ (4 << 0) + +#define ASLE 0xe4 +#define ASLS 0xfc + +#define SWSCI 0xe8 +#define SWSCI_SCISEL (1 << 15) +#define SWSCI_GSSCIE (1 << 0) + +/* legacy/combination backlight modes, also called LBB */ +#define LBPC 0xf4 + +#endif /* __INTEL_PCI_CONFIG_H__ */ diff --git a/drivers/gpu/drm/i915/intel_pcode.c b/drivers/gpu/drm/i915/intel_pcode.c index e8c886e4e78d..391a37492ce5 100644 --- a/drivers/gpu/drm/i915/intel_pcode.c +++ b/drivers/gpu/drm/i915/intel_pcode.c @@ -4,6 +4,7 @@ */ #include "i915_drv.h" +#include "i915_reg.h" #include "intel_pcode.h" static int gen6_check_mailbox_status(u32 mbox) @@ -51,11 +52,10 @@ static int gen7_check_mailbox_status(u32 mbox) } } -static int __sandybridge_pcode_rw(struct drm_i915_private *i915, - u32 mbox, u32 *val, u32 *val1, - int fast_timeout_us, - int slow_timeout_ms, - bool is_read) +static int __snb_pcode_rw(struct drm_i915_private *i915, u32 mbox, + u32 *val, u32 *val1, + int fast_timeout_us, int slow_timeout_ms, + bool is_read) { struct intel_uncore *uncore = &i915->uncore; @@ -94,15 +94,12 @@ static int __sandybridge_pcode_rw(struct drm_i915_private *i915, return gen6_check_mailbox_status(mbox); } -int sandybridge_pcode_read(struct drm_i915_private *i915, u32 mbox, - u32 *val, u32 *val1) +int snb_pcode_read(struct drm_i915_private *i915, u32 mbox, u32 *val, u32 *val1) { int err; mutex_lock(&i915->sb_lock); - err = __sandybridge_pcode_rw(i915, mbox, val, val1, - 500, 20, - true); + err = __snb_pcode_rw(i915, mbox, val, val1, 500, 20, true); mutex_unlock(&i915->sb_lock); if (err) { @@ -114,17 +111,14 @@ int sandybridge_pcode_read(struct drm_i915_private *i915, u32 mbox, return err; } -int sandybridge_pcode_write_timeout(struct drm_i915_private *i915, - u32 mbox, u32 val, - int fast_timeout_us, - int slow_timeout_ms) +int snb_pcode_write_timeout(struct drm_i915_private *i915, u32 mbox, u32 val, + int fast_timeout_us, int slow_timeout_ms) { int err; mutex_lock(&i915->sb_lock); - err = __sandybridge_pcode_rw(i915, mbox, &val, NULL, - fast_timeout_us, slow_timeout_ms, - false); + err = __snb_pcode_rw(i915, mbox, &val, NULL, + fast_timeout_us, slow_timeout_ms, false); mutex_unlock(&i915->sb_lock); if (err) { @@ -140,9 +134,7 @@ static bool skl_pcode_try_request(struct drm_i915_private *i915, u32 mbox, u32 request, u32 reply_mask, u32 reply, u32 *status) { - *status = __sandybridge_pcode_rw(i915, mbox, &request, NULL, - 500, 0, - true); + *status = __snb_pcode_rw(i915, mbox, &request, NULL, 500, 0, true); return *status || ((request & reply_mask) == reply); } diff --git a/drivers/gpu/drm/i915/intel_pcode.h b/drivers/gpu/drm/i915/intel_pcode.h index 50806649d4b6..0962a17fac48 100644 --- a/drivers/gpu/drm/i915/intel_pcode.h +++ b/drivers/gpu/drm/i915/intel_pcode.h @@ -10,13 +10,11 @@ struct drm_i915_private; -int sandybridge_pcode_read(struct drm_i915_private *i915, u32 mbox, - u32 *val, u32 *val1); -int sandybridge_pcode_write_timeout(struct drm_i915_private *i915, u32 mbox, - u32 val, int fast_timeout_us, - int slow_timeout_ms); -#define sandybridge_pcode_write(i915, mbox, val) \ - sandybridge_pcode_write_timeout(i915, mbox, val, 500, 0) +int snb_pcode_read(struct drm_i915_private *i915, u32 mbox, u32 *val, u32 *val1); +int snb_pcode_write_timeout(struct drm_i915_private *i915, u32 mbox, u32 val, + int fast_timeout_us, int slow_timeout_ms); +#define snb_pcode_write(i915, mbox, val) \ + snb_pcode_write_timeout(i915, mbox, val, 500, 0) int skl_pcode_request(struct drm_i915_private *i915, u32 mbox, u32 request, u32 reply_mask, u32 reply, int timeout_base_ms); diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 32bc155f5dc0..23d4bb011fc8 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -43,6 +43,8 @@ #include "display/intel_sprite.h" #include "display/skl_universal_plane.h" +#include "gt/intel_engine_regs.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_llc.h" #include "i915_drv.h" @@ -78,8 +80,6 @@ struct intel_wm_config { static void gen9_init_clock_gating(struct drm_i915_private *dev_priv) { - enum pipe pipe; - if (HAS_LLC(dev_priv)) { /* * WaCompressedResourceDisplayNewHashMode:skl,kbl @@ -93,16 +93,6 @@ static void gen9_init_clock_gating(struct drm_i915_private *dev_priv) SKL_DE_COMPRESSED_HASH_MODE); } - for_each_pipe(dev_priv, pipe) { - /* - * "Plane N strech max must be programmed to 11b (x1) - * when Async flips are enabled on that plane." - */ - if (!IS_GEMINILAKE(dev_priv) && intel_vtd_active(dev_priv)) - intel_uncore_rmw(&dev_priv->uncore, CHICKEN_PIPESL_1(pipe), - SKL_PLANE1_STRETCH_MAX_MASK, SKL_PLANE1_STRETCH_MAX_X1); - } - /* See Bspec note for PSR2_CTL bit 31, Wa#828:skl,bxt,kbl,cfl */ intel_uncore_write(&dev_priv->uncore, CHICKEN_PAR1_1, intel_uncore_read(&dev_priv->uncore, CHICKEN_PAR1_1) | SKL_EDP_PSR_FIX_RDWRAP); @@ -160,8 +150,9 @@ static void bxt_init_clock_gating(struct drm_i915_private *dev_priv) * WaFbcHighMemBwCorruptionAvoidance:bxt * Display WA #0883: bxt */ - intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN, intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN) | - DPFC_DISABLE_DUMMY0); + intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A), + intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A)) | + DPFC_DISABLE_DUMMY0); } static void glk_init_clock_gating(struct drm_i915_private *dev_priv) @@ -876,7 +867,7 @@ static bool intel_crtc_active(struct intel_crtc *crtc) * crtc->state->active once we have proper CRTC states wired up * for atomic. */ - return crtc->active && crtc->base.primary->state->fb && + return crtc && crtc->active && crtc->base.primary->state->fb && crtc->config->hw.adjusted_mode.crtc_clock; } @@ -915,15 +906,13 @@ static void pnv_update_wm(struct drm_i915_private *dev_priv) crtc = single_enabled_crtc(dev_priv); if (crtc) { - const struct drm_display_mode *pipe_mode = - &crtc->config->hw.pipe_mode; const struct drm_framebuffer *fb = crtc->base.primary->state->fb; + int pixel_rate = crtc->config->pixel_rate; int cpp = fb->format->cpp[0]; - int clock = pipe_mode->crtc_clock; /* Display SR */ - wm = intel_calculate_wm(clock, &pnv_display_wm, + wm = intel_calculate_wm(pixel_rate, &pnv_display_wm, pnv_display_wm.fifo_size, cpp, latency->display_sr); reg = intel_uncore_read(&dev_priv->uncore, DSPFW1); @@ -933,7 +922,7 @@ static void pnv_update_wm(struct drm_i915_private *dev_priv) drm_dbg_kms(&dev_priv->drm, "DSPFW1 register is %x\n", reg); /* cursor SR */ - wm = intel_calculate_wm(clock, &pnv_cursor_wm, + wm = intel_calculate_wm(pixel_rate, &pnv_cursor_wm, pnv_display_wm.fifo_size, 4, latency->cursor_sr); reg = intel_uncore_read(&dev_priv->uncore, DSPFW3); @@ -942,7 +931,7 @@ static void pnv_update_wm(struct drm_i915_private *dev_priv) intel_uncore_write(&dev_priv->uncore, DSPFW3, reg); /* Display HPLL off SR */ - wm = intel_calculate_wm(clock, &pnv_display_hplloff_wm, + wm = intel_calculate_wm(pixel_rate, &pnv_display_hplloff_wm, pnv_display_hplloff_wm.fifo_size, cpp, latency->display_hpll_disable); reg = intel_uncore_read(&dev_priv->uncore, DSPFW3); @@ -951,7 +940,7 @@ static void pnv_update_wm(struct drm_i915_private *dev_priv) intel_uncore_write(&dev_priv->uncore, DSPFW3, reg); /* cursor HPLL off SR */ - wm = intel_calculate_wm(clock, &pnv_cursor_hplloff_wm, + wm = intel_calculate_wm(pixel_rate, &pnv_cursor_hplloff_wm, pnv_display_hplloff_wm.fifo_size, 4, latency->cursor_hpll_disable); reg = intel_uncore_read(&dev_priv->uncore, DSPFW3); @@ -1154,7 +1143,7 @@ static u16 g4x_compute_wm(const struct intel_crtc_state *crtc_state, const struct drm_display_mode *pipe_mode = &crtc_state->hw.pipe_mode; unsigned int latency = dev_priv->wm.pri_latency[level] * 10; - unsigned int clock, htotal, cpp, width, wm; + unsigned int pixel_rate, htotal, cpp, width, wm; if (latency == 0) return USHRT_MAX; @@ -1175,21 +1164,20 @@ static u16 g4x_compute_wm(const struct intel_crtc_state *crtc_state, level != G4X_WM_LEVEL_NORMAL) cpp = max(cpp, 4u); - clock = pipe_mode->crtc_clock; + pixel_rate = crtc_state->pixel_rate; htotal = pipe_mode->crtc_htotal; - - width = drm_rect_width(&plane_state->uapi.dst); + width = drm_rect_width(&plane_state->uapi.src) >> 16; if (plane->id == PLANE_CURSOR) { - wm = intel_wm_method2(clock, htotal, width, cpp, latency); + wm = intel_wm_method2(pixel_rate, htotal, width, cpp, latency); } else if (plane->id == PLANE_PRIMARY && level == G4X_WM_LEVEL_NORMAL) { - wm = intel_wm_method1(clock, cpp, latency); + wm = intel_wm_method1(pixel_rate, cpp, latency); } else { unsigned int small, large; - small = intel_wm_method1(clock, cpp, latency); - large = intel_wm_method2(clock, htotal, width, cpp, latency); + small = intel_wm_method1(pixel_rate, cpp, latency); + large = intel_wm_method2(pixel_rate, htotal, width, cpp, latency); wm = min(small, large); } @@ -1674,7 +1662,7 @@ static u16 vlv_compute_wm_level(const struct intel_crtc_state *crtc_state, struct drm_i915_private *dev_priv = to_i915(plane->base.dev); const struct drm_display_mode *pipe_mode = &crtc_state->hw.pipe_mode; - unsigned int clock, htotal, cpp, width, wm; + unsigned int pixel_rate, htotal, cpp, width, wm; if (dev_priv->wm.pri_latency[level] == 0) return USHRT_MAX; @@ -1683,9 +1671,9 @@ static u16 vlv_compute_wm_level(const struct intel_crtc_state *crtc_state, return 0; cpp = plane_state->hw.fb->format->cpp[0]; - clock = pipe_mode->crtc_clock; + pixel_rate = crtc_state->pixel_rate; htotal = pipe_mode->crtc_htotal; - width = crtc_state->pipe_src_w; + width = drm_rect_width(&plane_state->uapi.src) >> 16; if (plane->id == PLANE_CURSOR) { /* @@ -1696,7 +1684,7 @@ static u16 vlv_compute_wm_level(const struct intel_crtc_state *crtc_state, */ wm = 63; } else { - wm = vlv_wm_method2(clock, htotal, width, cpp, + wm = vlv_wm_method2(pixel_rate, htotal, width, cpp, dev_priv->wm.pri_latency[level] * 10); } @@ -2277,14 +2265,14 @@ static void i965_update_wm(struct drm_i915_private *dev_priv) &crtc->config->hw.pipe_mode; const struct drm_framebuffer *fb = crtc->base.primary->state->fb; - int clock = pipe_mode->crtc_clock; + int pixel_rate = crtc->config->pixel_rate; int htotal = pipe_mode->crtc_htotal; - int hdisplay = crtc->config->pipe_src_w; + int width = drm_rect_width(&crtc->base.primary->state->src) >> 16; int cpp = fb->format->cpp[0]; int entries; - entries = intel_wm_method2(clock, htotal, - hdisplay, cpp, sr_latency_ns / 100); + entries = intel_wm_method2(pixel_rate, htotal, + width, cpp, sr_latency_ns / 100); entries = DIV_ROUND_UP(entries, I915_FIFO_LINE_SIZE); srwm = I965_FIFO_SIZE - entries; if (srwm < 0) @@ -2294,7 +2282,7 @@ static void i965_update_wm(struct drm_i915_private *dev_priv) "self-refresh entries: %d, wm: %d\n", entries, srwm); - entries = intel_wm_method2(clock, htotal, + entries = intel_wm_method2(pixel_rate, htotal, crtc->base.cursor->state->crtc_w, 4, sr_latency_ns / 100); entries = DIV_ROUND_UP(entries, @@ -2358,7 +2346,7 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv) int cwm, srwm = 1; int fifo_size; int planea_wm, planeb_wm; - struct intel_crtc *crtc, *enabled = NULL; + struct intel_crtc *crtc; if (IS_I945GM(dev_priv)) wm_info = &i945_wm_info; @@ -2373,8 +2361,6 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv) fifo_size = i9xx_get_fifo_size(dev_priv, PLANE_A); crtc = intel_crtc_for_plane(dev_priv, PLANE_A); if (intel_crtc_active(crtc)) { - const struct drm_display_mode *pipe_mode = - &crtc->config->hw.pipe_mode; const struct drm_framebuffer *fb = crtc->base.primary->state->fb; int cpp; @@ -2384,10 +2370,9 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv) else cpp = fb->format->cpp[0]; - planea_wm = intel_calculate_wm(pipe_mode->crtc_clock, + planea_wm = intel_calculate_wm(crtc->config->pixel_rate, wm_info, fifo_size, cpp, pessimal_latency_ns); - enabled = crtc; } else { planea_wm = fifo_size - wm_info->guard_size; if (planea_wm > (long)wm_info->max_wm) @@ -2403,8 +2388,6 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv) fifo_size = i9xx_get_fifo_size(dev_priv, PLANE_B); crtc = intel_crtc_for_plane(dev_priv, PLANE_B); if (intel_crtc_active(crtc)) { - const struct drm_display_mode *pipe_mode = - &crtc->config->hw.pipe_mode; const struct drm_framebuffer *fb = crtc->base.primary->state->fb; int cpp; @@ -2414,13 +2397,9 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv) else cpp = fb->format->cpp[0]; - planeb_wm = intel_calculate_wm(pipe_mode->crtc_clock, + planeb_wm = intel_calculate_wm(crtc->config->pixel_rate, wm_info, fifo_size, cpp, pessimal_latency_ns); - if (enabled == NULL) - enabled = crtc; - else - enabled = NULL; } else { planeb_wm = fifo_size - wm_info->guard_size; if (planeb_wm > (long)wm_info->max_wm) @@ -2430,14 +2409,15 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv) drm_dbg_kms(&dev_priv->drm, "FIFO watermarks - A: %d, B: %d\n", planea_wm, planeb_wm); - if (IS_I915GM(dev_priv) && enabled) { + crtc = single_enabled_crtc(dev_priv); + if (IS_I915GM(dev_priv) && crtc) { struct drm_i915_gem_object *obj; - obj = intel_fb_obj(enabled->base.primary->state->fb); + obj = intel_fb_obj(crtc->base.primary->state->fb); /* self-refresh seems busted with untiled */ if (!i915_gem_object_is_tiled(obj)) - enabled = NULL; + crtc = NULL; } /* @@ -2449,16 +2429,16 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv) intel_set_memory_cxsr(dev_priv, false); /* Calc sr entries for one plane configs */ - if (HAS_FW_BLC(dev_priv) && enabled) { + if (HAS_FW_BLC(dev_priv) && crtc) { /* self-refresh has much higher latency */ static const int sr_latency_ns = 6000; const struct drm_display_mode *pipe_mode = - &enabled->config->hw.pipe_mode; + &crtc->config->hw.pipe_mode; const struct drm_framebuffer *fb = - enabled->base.primary->state->fb; - int clock = pipe_mode->crtc_clock; + crtc->base.primary->state->fb; + int pixel_rate = crtc->config->pixel_rate; int htotal = pipe_mode->crtc_htotal; - int hdisplay = enabled->config->pipe_src_w; + int width = drm_rect_width(&crtc->base.primary->state->src) >> 16; int cpp; int entries; @@ -2467,7 +2447,7 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv) else cpp = fb->format->cpp[0]; - entries = intel_wm_method2(clock, htotal, hdisplay, cpp, + entries = intel_wm_method2(pixel_rate, htotal, width, cpp, sr_latency_ns / 100); entries = DIV_ROUND_UP(entries, wm_info->cacheline_size); drm_dbg_kms(&dev_priv->drm, @@ -2497,14 +2477,13 @@ static void i9xx_update_wm(struct drm_i915_private *dev_priv) intel_uncore_write(&dev_priv->uncore, FW_BLC, fwater_lo); intel_uncore_write(&dev_priv->uncore, FW_BLC2, fwater_hi); - if (enabled) + if (crtc) intel_set_memory_cxsr(dev_priv, true); } static void i845_update_wm(struct drm_i915_private *dev_priv) { struct intel_crtc *crtc; - const struct drm_display_mode *pipe_mode; u32 fwater_lo; int planea_wm; @@ -2512,8 +2491,7 @@ static void i845_update_wm(struct drm_i915_private *dev_priv) if (crtc == NULL) return; - pipe_mode = &crtc->config->hw.pipe_mode; - planea_wm = intel_calculate_wm(pipe_mode->crtc_clock, + planea_wm = intel_calculate_wm(crtc->config->pixel_rate, &i845_wm_info, i845_get_fifo_size(dev_priv, PLANE_A), 4, pessimal_latency_ns); @@ -2604,7 +2582,7 @@ static u32 ilk_compute_pri_wm(const struct intel_crtc_state *crtc_state, method2 = ilk_wm_method2(crtc_state->pixel_rate, crtc_state->hw.pipe_mode.crtc_htotal, - drm_rect_width(&plane_state->uapi.dst), + drm_rect_width(&plane_state->uapi.src) >> 16, cpp, mem_value); return min(method1, method2); @@ -2632,7 +2610,7 @@ static u32 ilk_compute_spr_wm(const struct intel_crtc_state *crtc_state, method1 = ilk_wm_method1(crtc_state->pixel_rate, cpp, mem_value); method2 = ilk_wm_method2(crtc_state->pixel_rate, crtc_state->hw.pipe_mode.crtc_htotal, - drm_rect_width(&plane_state->uapi.dst), + drm_rect_width(&plane_state->uapi.src) >> 16, cpp, mem_value); return min(method1, method2); } @@ -2657,7 +2635,7 @@ static u32 ilk_compute_cur_wm(const struct intel_crtc_state *crtc_state, return ilk_wm_method2(crtc_state->pixel_rate, crtc_state->hw.pipe_mode.crtc_htotal, - drm_rect_width(&plane_state->uapi.dst), + drm_rect_width(&plane_state->uapi.src) >> 16, cpp, mem_value); } @@ -2673,7 +2651,7 @@ static u32 ilk_compute_fbc_wm(const struct intel_crtc_state *crtc_state, cpp = plane_state->hw.fb->format->cpp[0]; - return ilk_wm_fbc(pri_val, drm_rect_width(&plane_state->uapi.dst), + return ilk_wm_fbc(pri_val, drm_rect_width(&plane_state->uapi.src) >> 16, cpp); } @@ -2888,9 +2866,8 @@ static void intel_read_wm_latency(struct drm_i915_private *dev_priv, /* read the first set of memory latencies[0:3] */ val = 0; /* data0 to be programmed to 0 for first set */ - ret = sandybridge_pcode_read(dev_priv, - GEN9_PCODE_READ_MEM_LATENCY, - &val, NULL); + ret = snb_pcode_read(dev_priv, GEN9_PCODE_READ_MEM_LATENCY, + &val, NULL); if (ret) { drm_err(&dev_priv->drm, @@ -2908,9 +2885,8 @@ static void intel_read_wm_latency(struct drm_i915_private *dev_priv, /* read the second set of memory latencies[4:7] */ val = 1; /* data0 to be programmed to 1 for second set */ - ret = sandybridge_pcode_read(dev_priv, - GEN9_PCODE_READ_MEM_LATENCY, - &val, NULL); + ret = snb_pcode_read(dev_priv, GEN9_PCODE_READ_MEM_LATENCY, + &val, NULL); if (ret) { drm_err(&dev_priv->drm, "SKL Mailbox read error = %d\n", ret); @@ -3700,9 +3676,9 @@ skl_setup_sagv_block_time(struct drm_i915_private *dev_priv) u32 val = 0; int ret; - ret = sandybridge_pcode_read(dev_priv, - GEN12_PCODE_READ_SAGV_BLOCK_TIME_US, - &val, NULL); + ret = snb_pcode_read(dev_priv, + GEN12_PCODE_READ_SAGV_BLOCK_TIME_US, + &val, NULL); if (!ret) { dev_priv->sagv_block_time_us = val; return; @@ -3749,8 +3725,8 @@ intel_enable_sagv(struct drm_i915_private *dev_priv) return 0; drm_dbg_kms(&dev_priv->drm, "Enabling SAGV\n"); - ret = sandybridge_pcode_write(dev_priv, GEN9_PCODE_SAGV_CONTROL, - GEN9_SAGV_ENABLE); + ret = snb_pcode_write(dev_priv, GEN9_PCODE_SAGV_CONTROL, + GEN9_SAGV_ENABLE); /* We don't need to wait for SAGV when enabling */ @@ -4252,7 +4228,9 @@ static int skl_compute_wm_params(const struct intel_crtc_state *crtc_state, u64 modifier, unsigned int rotation, u32 plane_pixel_rate, struct skl_wm_params *wp, int color_plane); + static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state, + struct intel_plane *plane, int level, unsigned int latency, const struct skl_wm_params *wp, @@ -4263,6 +4241,7 @@ static unsigned int skl_cursor_allocation(const struct intel_crtc_state *crtc_state, int num_active) { + struct intel_plane *plane = to_intel_plane(crtc_state->uapi.crtc->cursor); struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); int level, max_level = ilk_wm_max_level(dev_priv); struct skl_wm_level wm = {}; @@ -4279,7 +4258,7 @@ skl_cursor_allocation(const struct intel_crtc_state *crtc_state, for (level = 0; level <= max_level; level++) { unsigned int latency = dev_priv->wm.skl_latency[level]; - skl_compute_plane_wm(crtc_state, level, latency, &wp, &wm, &wm); + skl_compute_plane_wm(crtc_state, plane, level, latency, &wp, &wm, &wm); if (wm.min_ddb_alloc == U16_MAX) break; @@ -4292,11 +4271,10 @@ skl_cursor_allocation(const struct intel_crtc_state *crtc_state, static void skl_ddb_entry_init_from_hw(struct drm_i915_private *dev_priv, struct skl_ddb_entry *entry, u32 reg) { - entry->start = reg & DDB_ENTRY_MASK; - entry->end = (reg >> DDB_ENTRY_END_SHIFT) & DDB_ENTRY_MASK; - + entry->start = REG_FIELD_GET(PLANE_BUF_START_MASK, reg); + entry->end = REG_FIELD_GET(PLANE_BUF_END_MASK, reg); if (entry->end) - entry->end += 1; + entry->end++; } static void @@ -4320,7 +4298,7 @@ skl_ddb_get_hw_plane_state(struct drm_i915_private *dev_priv, /* No DDB allocated for disabled planes */ if (val & PLANE_CTL_ENABLE) - fourcc = skl_format_to_fourcc(val & PLANE_CTL_FORMAT_MASK, + fourcc = skl_format_to_fourcc(val & PLANE_CTL_FORMAT_MASK_SKL, val & PLANE_CTL_ORDER_RGBX, val & PLANE_CTL_ALPHA_MASK); @@ -4939,6 +4917,28 @@ static u8 skl_compute_dbuf_slices(struct intel_crtc *crtc, u8 active_pipes, bool return active_pipes & BIT(pipe) ? BIT(DBUF_S1) : 0; } +static bool +use_min_ddb(const struct intel_crtc_state *crtc_state, + struct intel_plane *plane) +{ + struct drm_i915_private *i915 = to_i915(plane->base.dev); + + return DISPLAY_VER(i915) >= 13 && + crtc_state->uapi.async_flip && + plane->async_flip; +} + +static bool +use_minimal_wm0_only(const struct intel_crtc_state *crtc_state, + struct intel_plane *plane) +{ + struct drm_i915_private *i915 = to_i915(plane->base.dev); + + return DISPLAY_VER(i915) >= 13 && + crtc_state->uapi.async_flip && + plane->async_flip; +} + static u64 skl_plane_relative_data_rate(const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state, @@ -4957,6 +4957,14 @@ skl_plane_relative_data_rate(const struct intel_crtc_state *crtc_state, if (plane->id == PLANE_CURSOR) return 0; + /* + * We calculate extra ddb based on ratio plane rate/total data rate + * in case, in some cases we should not allocate extra ddb for the plane, + * so do not count its data rate, if this is the case. + */ + if (use_min_ddb(crtc_state, plane)) + return 0; + if (color_plane == 1 && !intel_format_info_is_yuv_semiplanar(fb->format, fb->modifier)) return 0; @@ -5531,6 +5539,7 @@ static int skl_wm_max_lines(struct drm_i915_private *dev_priv) } static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state, + struct intel_plane *plane, int level, unsigned int latency, const struct skl_wm_params *wp, @@ -5542,7 +5551,8 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state, uint_fixed_16_16_t selected_result; u32 blocks, lines, min_ddb_alloc = 0; - if (latency == 0) { + if (latency == 0 || + (use_minimal_wm0_only(crtc_state, plane) && level > 0)) { /* reject it */ result->min_ddb_alloc = U16_MAX; return; @@ -5658,6 +5668,7 @@ static void skl_compute_plane_wm(const struct intel_crtc_state *crtc_state, static void skl_compute_wm_levels(const struct intel_crtc_state *crtc_state, + struct intel_plane *plane, const struct skl_wm_params *wm_params, struct skl_wm_level *levels) { @@ -5669,7 +5680,7 @@ skl_compute_wm_levels(const struct intel_crtc_state *crtc_state, struct skl_wm_level *result = &levels[level]; unsigned int latency = dev_priv->wm.skl_latency[level]; - skl_compute_plane_wm(crtc_state, level, latency, + skl_compute_plane_wm(crtc_state, plane, level, latency, wm_params, result_prev, result); result_prev = result; @@ -5677,6 +5688,7 @@ skl_compute_wm_levels(const struct intel_crtc_state *crtc_state, } static void tgl_compute_sagv_wm(const struct intel_crtc_state *crtc_state, + struct intel_plane *plane, const struct skl_wm_params *wm_params, struct skl_plane_wm *plane_wm) { @@ -5685,7 +5697,7 @@ static void tgl_compute_sagv_wm(const struct intel_crtc_state *crtc_state, struct skl_wm_level *levels = plane_wm->wm; unsigned int latency = dev_priv->wm.skl_latency[0] + dev_priv->sagv_block_time_us; - skl_compute_plane_wm(crtc_state, 0, latency, + skl_compute_plane_wm(crtc_state, plane, 0, latency, wm_params, &levels[0], sagv_wm); } @@ -5755,11 +5767,11 @@ static void skl_compute_transition_wm(struct drm_i915_private *dev_priv, static int skl_build_plane_wm_single(struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state, - enum plane_id plane_id, int color_plane) + struct intel_plane *plane, int color_plane) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - struct skl_plane_wm *wm = &crtc_state->wm.skl.raw.planes[plane_id]; + struct skl_plane_wm *wm = &crtc_state->wm.skl.raw.planes[plane->id]; struct skl_wm_params wm_params; int ret; @@ -5768,13 +5780,13 @@ static int skl_build_plane_wm_single(struct intel_crtc_state *crtc_state, if (ret) return ret; - skl_compute_wm_levels(crtc_state, &wm_params, wm->wm); + skl_compute_wm_levels(crtc_state, plane, &wm_params, wm->wm); skl_compute_transition_wm(dev_priv, &wm->trans_wm, &wm->wm[0], &wm_params); if (DISPLAY_VER(dev_priv) >= 12) { - tgl_compute_sagv_wm(crtc_state, &wm_params, wm); + tgl_compute_sagv_wm(crtc_state, plane, &wm_params, wm); skl_compute_transition_wm(dev_priv, &wm->sagv.trans_wm, &wm->sagv.wm0, &wm_params); @@ -5785,9 +5797,9 @@ static int skl_build_plane_wm_single(struct intel_crtc_state *crtc_state, static int skl_build_plane_wm_uv(struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state, - enum plane_id plane_id) + struct intel_plane *plane) { - struct skl_plane_wm *wm = &crtc_state->wm.skl.raw.planes[plane_id]; + struct skl_plane_wm *wm = &crtc_state->wm.skl.raw.planes[plane->id]; struct skl_wm_params wm_params; int ret; @@ -5799,7 +5811,7 @@ static int skl_build_plane_wm_uv(struct intel_crtc_state *crtc_state, if (ret) return ret; - skl_compute_wm_levels(crtc_state, &wm_params, wm->uv_wm); + skl_compute_wm_levels(crtc_state, plane, &wm_params, wm->uv_wm); return 0; } @@ -5819,13 +5831,13 @@ static int skl_build_plane_wm(struct intel_crtc_state *crtc_state, return 0; ret = skl_build_plane_wm_single(crtc_state, plane_state, - plane_id, 0); + plane, 0); if (ret) return ret; if (fb->format->is_yuv && fb->format->num_planes > 1) { ret = skl_build_plane_wm_uv(crtc_state, plane_state, - plane_id); + plane); if (ret) return ret; } @@ -5850,7 +5862,6 @@ static int icl_build_plane_wm(struct intel_crtc_state *crtc_state, if (plane_state->planar_linked_plane) { const struct drm_framebuffer *fb = plane_state->hw.fb; - enum plane_id y_plane_id = plane_state->planar_linked_plane->id; drm_WARN_ON(&dev_priv->drm, !intel_wm_plane_visible(crtc_state, plane_state)); @@ -5858,17 +5869,17 @@ static int icl_build_plane_wm(struct intel_crtc_state *crtc_state, fb->format->num_planes == 1); ret = skl_build_plane_wm_single(crtc_state, plane_state, - y_plane_id, 0); + plane_state->planar_linked_plane, 0); if (ret) return ret; ret = skl_build_plane_wm_single(crtc_state, plane_state, - plane_id, 1); + plane, 1); if (ret) return ret; } else if (intel_wm_plane_visible(crtc_state, plane_state)) { ret = skl_build_plane_wm_single(crtc_state, plane_state, - plane_id, 0); + plane, 0); if (ret) return ret; } @@ -5914,7 +5925,8 @@ static void skl_ddb_entry_write(struct drm_i915_private *dev_priv, { if (entry->end) intel_de_write_fw(dev_priv, reg, - (entry->end - 1) << 16 | entry->start); + PLANE_BUF_END(entry->end - 1) | + PLANE_BUF_START(entry->start)); else intel_de_write_fw(dev_priv, reg, 0); } @@ -7317,7 +7329,7 @@ static void g4x_disable_trickle_feed(struct drm_i915_private *dev_priv) for_each_pipe(dev_priv, pipe) { intel_uncore_write(&dev_priv->uncore, DSPCNTR(pipe), intel_uncore_read(&dev_priv->uncore, DSPCNTR(pipe)) | - DISPPLANE_TRICKLE_FEED_DISABLE); + DISP_TRICKLE_FEED_DISABLE); intel_uncore_write(&dev_priv->uncore, DSPSURF(pipe), intel_uncore_read(&dev_priv->uncore, DSPSURF(pipe))); intel_uncore_posting_read(&dev_priv->uncore, DSPSURF(pipe)); @@ -7548,8 +7560,8 @@ static void gen8_set_l3sqc_credits(struct drm_i915_private *dev_priv, static void icl_init_clock_gating(struct drm_i915_private *dev_priv) { /* Wa_1409120013:icl,ehl */ - intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN, - DPFC_CHICKEN_COMP_DUMMY_PIXEL); + intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A), + DPFC_CHICKEN_COMP_DUMMY_PIXEL); /*Wa_14010594013:icl, ehl */ intel_uncore_rmw(&dev_priv->uncore, GEN8_CHICKEN_DCPR_1, @@ -7561,7 +7573,7 @@ static void gen12lp_init_clock_gating(struct drm_i915_private *dev_priv) /* Wa_1409120013:tgl,rkl,adl-s,dg1,dg2 */ if (IS_TIGERLAKE(dev_priv) || IS_ROCKETLAKE(dev_priv) || IS_ALDERLAKE_S(dev_priv) || IS_DG1(dev_priv) || IS_DG2(dev_priv)) - intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN, + intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A), DPFC_CHICKEN_COMP_DUMMY_PIXEL); /* Wa_1409825376:tgl (pre-prod)*/ @@ -7646,8 +7658,9 @@ static void cfl_init_clock_gating(struct drm_i915_private *dev_priv) * WaFbcNukeOnHostModify:cfl * Display WA #0873: cfl */ - intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN, intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN) | - DPFC_NUKE_ON_ANY_MODIFICATION); + intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A), + intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A)) | + DPFC_NUKE_ON_ANY_MODIFICATION); } static void kbl_init_clock_gating(struct drm_i915_private *dev_priv) @@ -7679,8 +7692,9 @@ static void kbl_init_clock_gating(struct drm_i915_private *dev_priv) * WaFbcNukeOnHostModify:kbl * Display WA #0873: kbl */ - intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN, intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN) | - DPFC_NUKE_ON_ANY_MODIFICATION); + intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A), + intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A)) | + DPFC_NUKE_ON_ANY_MODIFICATION); } static void skl_init_clock_gating(struct drm_i915_private *dev_priv) @@ -7706,15 +7720,17 @@ static void skl_init_clock_gating(struct drm_i915_private *dev_priv) * WaFbcNukeOnHostModify:skl * Display WA #0873: skl */ - intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN, intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN) | - DPFC_NUKE_ON_ANY_MODIFICATION); + intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A), + intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A)) | + DPFC_NUKE_ON_ANY_MODIFICATION); /* * WaFbcHighMemBwCorruptionAvoidance:skl * Display WA #0883: skl */ - intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN, intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN) | - DPFC_DISABLE_DUMMY0); + intel_uncore_write(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A), + intel_uncore_read(&dev_priv->uncore, ILK_DPFC_CHICKEN(INTEL_FBC_A)) | + DPFC_DISABLE_DUMMY0); } static void bdw_init_clock_gating(struct drm_i915_private *dev_priv) @@ -7746,7 +7762,7 @@ static void bdw_init_clock_gating(struct drm_i915_private *dev_priv) intel_uncore_read(&dev_priv->uncore, GEN7_FF_THREAD_MODE) & ~(GEN8_FF_DS_REF_CNT_FFME | GEN7_FF_VS_REF_CNT_FFME)); - intel_uncore_write(&dev_priv->uncore, GEN6_RC_SLEEP_PSMI_CONTROL, + intel_uncore_write(&dev_priv->uncore, RING_PSMI_CTL(RENDER_RING_BASE), _MASKED_BIT_ENABLE(GEN8_RC_SEMA_IDLE_MSG_DISABLE)); /* WaDisableSDEUnitClockGating:bdw */ @@ -7887,7 +7903,7 @@ static void chv_init_clock_gating(struct drm_i915_private *dev_priv) ~(GEN8_FF_DS_REF_CNT_FFME | GEN7_FF_VS_REF_CNT_FFME)); /* WaDisableSemaphoreAndSyncFlipWait:chv */ - intel_uncore_write(&dev_priv->uncore, GEN6_RC_SLEEP_PSMI_CONTROL, + intel_uncore_write(&dev_priv->uncore, RING_PSMI_CTL(RENDER_RING_BASE), _MASKED_BIT_ENABLE(GEN8_RC_SEMA_IDLE_MSG_DISABLE)); /* WaDisableCSUnitClockGating:chv */ @@ -7960,10 +7976,12 @@ static void gen3_init_clock_gating(struct drm_i915_private *dev_priv) intel_uncore_write(&dev_priv->uncore, D_STATE, dstate); if (IS_PINEVIEW(dev_priv)) - intel_uncore_write(&dev_priv->uncore, ECOSKPD, _MASKED_BIT_ENABLE(ECO_GATING_CX_ONLY)); + intel_uncore_write(&dev_priv->uncore, ECOSKPD(RENDER_RING_BASE), + _MASKED_BIT_ENABLE(ECO_GATING_CX_ONLY)); /* IIR "flip pending" means done if this bit is set */ - intel_uncore_write(&dev_priv->uncore, ECOSKPD, _MASKED_BIT_DISABLE(ECO_FLIP_DONE)); + intel_uncore_write(&dev_priv->uncore, ECOSKPD(RENDER_RING_BASE), + _MASKED_BIT_DISABLE(ECO_FLIP_DONE)); /* interrupts should cause a wake up from C3 */ intel_uncore_write(&dev_priv->uncore, INSTPM, _MASKED_BIT_ENABLE(INSTPM_AGPBUSY_INT_EN)); diff --git a/drivers/gpu/drm/i915/intel_pm.h b/drivers/gpu/drm/i915/intel_pm.h index d2243653a893..51705151b842 100644 --- a/drivers/gpu/drm/i915/intel_pm.h +++ b/drivers/gpu/drm/i915/intel_pm.h @@ -12,7 +12,6 @@ #include "display/intel_global_state.h" #include "i915_drv.h" -#include "i915_reg.h" struct drm_device; struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index 64c2708efc9e..3293ac71bcf8 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -77,9 +77,6 @@ track_intel_runtime_pm_wakeref(struct intel_runtime_pm *rpm) depot_stack_handle_t stack, *stacks; unsigned long flags; - if (!rpm->available) - return -1; - stack = __save_depot_stack(); if (!stack) return -1; diff --git a/drivers/gpu/drm/i915/intel_sbi.c b/drivers/gpu/drm/i915/intel_sbi.c index 5ba8490a31e6..5c6e517c73f4 100644 --- a/drivers/gpu/drm/i915/intel_sbi.c +++ b/drivers/gpu/drm/i915/intel_sbi.c @@ -7,6 +7,7 @@ #include "i915_drv.h" #include "intel_sbi.h" +#include "i915_reg.h" /* SBI access */ static int intel_sbi_rw(struct drm_i915_private *i915, u16 reg, diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 778da3179b3c..850ebfae31af 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -23,7 +23,8 @@ #include <linux/pm_runtime.h> -#include "gt/intel_lrc_reg.h" /* for shadow reg list */ +#include "gt/intel_engine_regs.h" +#include "gt/intel_gt_regs.h" #include "i915_drv.h" #include "i915_iosf_mbi.h" @@ -2273,76 +2274,6 @@ void intel_uncore_fini_mmio(struct intel_uncore *uncore) } } -static const struct reg_whitelist { - i915_reg_t offset_ldw; - i915_reg_t offset_udw; - u8 min_graphics_ver; - u8 max_graphics_ver; - u8 size; -} reg_read_whitelist[] = { { - .offset_ldw = RING_TIMESTAMP(RENDER_RING_BASE), - .offset_udw = RING_TIMESTAMP_UDW(RENDER_RING_BASE), - .min_graphics_ver = 4, - .max_graphics_ver = 12, - .size = 8 -} }; - -int i915_reg_read_ioctl(struct drm_device *dev, - void *data, struct drm_file *file) -{ - struct drm_i915_private *i915 = to_i915(dev); - struct intel_uncore *uncore = &i915->uncore; - struct drm_i915_reg_read *reg = data; - struct reg_whitelist const *entry; - intel_wakeref_t wakeref; - unsigned int flags; - int remain; - int ret = 0; - - entry = reg_read_whitelist; - remain = ARRAY_SIZE(reg_read_whitelist); - while (remain) { - u32 entry_offset = i915_mmio_reg_offset(entry->offset_ldw); - - GEM_BUG_ON(!is_power_of_2(entry->size)); - GEM_BUG_ON(entry->size > 8); - GEM_BUG_ON(entry_offset & (entry->size - 1)); - - if (IS_GRAPHICS_VER(i915, entry->min_graphics_ver, entry->max_graphics_ver) && - entry_offset == (reg->offset & -entry->size)) - break; - entry++; - remain--; - } - - if (!remain) - return -EINVAL; - - flags = reg->offset & (entry->size - 1); - - with_intel_runtime_pm(&i915->runtime_pm, wakeref) { - if (entry->size == 8 && flags == I915_REG_READ_8B_WA) - reg->val = intel_uncore_read64_2x32(uncore, - entry->offset_ldw, - entry->offset_udw); - else if (entry->size == 8 && flags == 0) - reg->val = intel_uncore_read64(uncore, - entry->offset_ldw); - else if (entry->size == 4 && flags == 0) - reg->val = intel_uncore_read(uncore, entry->offset_ldw); - else if (entry->size == 2 && flags == 0) - reg->val = intel_uncore_read16(uncore, - entry->offset_ldw); - else if (entry->size == 1 && flags == 0) - reg->val = intel_uncore_read8(uncore, - entry->offset_ldw); - else - ret = -EINVAL; - } - - return ret; -} - /** * __intel_wait_for_register_fw - wait until register matches expected state * @uncore: the struct intel_uncore diff --git a/drivers/gpu/drm/i915/intel_uncore.h b/drivers/gpu/drm/i915/intel_uncore.h index 2a15b2b2e2fc..6ff56d673e2b 100644 --- a/drivers/gpu/drm/i915/intel_uncore.h +++ b/drivers/gpu/drm/i915/intel_uncore.h @@ -30,7 +30,7 @@ #include <linux/hrtimer.h> #include <linux/io-64-nonatomic-lo-hi.h> -#include "i915_reg.h" +#include "i915_reg_defs.h" struct drm_i915_private; struct intel_runtime_pm; diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_irq.c b/drivers/gpu/drm/i915/pxp/intel_pxp_irq.c index 8d5553772ded..04745f914407 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_irq.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_irq.c @@ -7,6 +7,7 @@ #include "intel_pxp_irq.h" #include "intel_pxp_session.h" #include "gt/intel_gt_irq.h" +#include "gt/intel_gt_regs.h" #include "gt/intel_gt_types.h" #include "i915_irq.h" #include "i915_reg.h" diff --git a/drivers/gpu/drm/i915/selftests/i915_buddy.c b/drivers/gpu/drm/i915/selftests/i915_buddy.c deleted file mode 100644 index d61ec9c951bf..000000000000 --- a/drivers/gpu/drm/i915/selftests/i915_buddy.c +++ /dev/null @@ -1,787 +0,0 @@ -// SPDX-License-Identifier: MIT -/* - * Copyright © 2019 Intel Corporation - */ - -#include <linux/prime_numbers.h> - -#include "../i915_selftest.h" -#include "i915_random.h" - -static void __igt_dump_block(struct i915_buddy_mm *mm, - struct i915_buddy_block *block, - bool buddy) -{ - pr_err("block info: header=%llx, state=%u, order=%d, offset=%llx size=%llx root=%s buddy=%s\n", - block->header, - i915_buddy_block_state(block), - i915_buddy_block_order(block), - i915_buddy_block_offset(block), - i915_buddy_block_size(mm, block), - yesno(!block->parent), - yesno(buddy)); -} - -static void igt_dump_block(struct i915_buddy_mm *mm, - struct i915_buddy_block *block) -{ - struct i915_buddy_block *buddy; - - __igt_dump_block(mm, block, false); - - buddy = get_buddy(block); - if (buddy) - __igt_dump_block(mm, buddy, true); -} - -static int igt_check_block(struct i915_buddy_mm *mm, - struct i915_buddy_block *block) -{ - struct i915_buddy_block *buddy; - unsigned int block_state; - u64 block_size; - u64 offset; - int err = 0; - - block_state = i915_buddy_block_state(block); - - if (block_state != I915_BUDDY_ALLOCATED && - block_state != I915_BUDDY_FREE && - block_state != I915_BUDDY_SPLIT) { - pr_err("block state mismatch\n"); - err = -EINVAL; - } - - block_size = i915_buddy_block_size(mm, block); - offset = i915_buddy_block_offset(block); - - if (block_size < mm->chunk_size) { - pr_err("block size smaller than min size\n"); - err = -EINVAL; - } - - if (!is_power_of_2(block_size)) { - pr_err("block size not power of two\n"); - err = -EINVAL; - } - - if (!IS_ALIGNED(block_size, mm->chunk_size)) { - pr_err("block size not aligned to min size\n"); - err = -EINVAL; - } - - if (!IS_ALIGNED(offset, mm->chunk_size)) { - pr_err("block offset not aligned to min size\n"); - err = -EINVAL; - } - - if (!IS_ALIGNED(offset, block_size)) { - pr_err("block offset not aligned to block size\n"); - err = -EINVAL; - } - - buddy = get_buddy(block); - - if (!buddy && block->parent) { - pr_err("buddy has gone fishing\n"); - err = -EINVAL; - } - - if (buddy) { - if (i915_buddy_block_offset(buddy) != (offset ^ block_size)) { - pr_err("buddy has wrong offset\n"); - err = -EINVAL; - } - - if (i915_buddy_block_size(mm, buddy) != block_size) { - pr_err("buddy size mismatch\n"); - err = -EINVAL; - } - - if (i915_buddy_block_state(buddy) == block_state && - block_state == I915_BUDDY_FREE) { - pr_err("block and its buddy are free\n"); - err = -EINVAL; - } - } - - return err; -} - -static int igt_check_blocks(struct i915_buddy_mm *mm, - struct list_head *blocks, - u64 expected_size, - bool is_contiguous) -{ - struct i915_buddy_block *block; - struct i915_buddy_block *prev; - u64 total; - int err = 0; - - block = NULL; - prev = NULL; - total = 0; - - list_for_each_entry(block, blocks, link) { - err = igt_check_block(mm, block); - - if (!i915_buddy_block_is_allocated(block)) { - pr_err("block not allocated\n"), - err = -EINVAL; - } - - if (is_contiguous && prev) { - u64 prev_block_size; - u64 prev_offset; - u64 offset; - - prev_offset = i915_buddy_block_offset(prev); - prev_block_size = i915_buddy_block_size(mm, prev); - offset = i915_buddy_block_offset(block); - - if (offset != (prev_offset + prev_block_size)) { - pr_err("block offset mismatch\n"); - err = -EINVAL; - } - } - - if (err) - break; - - total += i915_buddy_block_size(mm, block); - prev = block; - } - - if (!err) { - if (total != expected_size) { - pr_err("size mismatch, expected=%llx, found=%llx\n", - expected_size, total); - err = -EINVAL; - } - return err; - } - - if (prev) { - pr_err("prev block, dump:\n"); - igt_dump_block(mm, prev); - } - - pr_err("bad block, dump:\n"); - igt_dump_block(mm, block); - - return err; -} - -static int igt_check_mm(struct i915_buddy_mm *mm) -{ - struct i915_buddy_block *root; - struct i915_buddy_block *prev; - unsigned int i; - u64 total; - int err = 0; - - if (!mm->n_roots) { - pr_err("n_roots is zero\n"); - return -EINVAL; - } - - if (mm->n_roots != hweight64(mm->size)) { - pr_err("n_roots mismatch, n_roots=%u, expected=%lu\n", - mm->n_roots, hweight64(mm->size)); - return -EINVAL; - } - - root = NULL; - prev = NULL; - total = 0; - - for (i = 0; i < mm->n_roots; ++i) { - struct i915_buddy_block *block; - unsigned int order; - - root = mm->roots[i]; - if (!root) { - pr_err("root(%u) is NULL\n", i); - err = -EINVAL; - break; - } - - err = igt_check_block(mm, root); - - if (!i915_buddy_block_is_free(root)) { - pr_err("root not free\n"); - err = -EINVAL; - } - - order = i915_buddy_block_order(root); - - if (!i) { - if (order != mm->max_order) { - pr_err("max order root missing\n"); - err = -EINVAL; - } - } - - if (prev) { - u64 prev_block_size; - u64 prev_offset; - u64 offset; - - prev_offset = i915_buddy_block_offset(prev); - prev_block_size = i915_buddy_block_size(mm, prev); - offset = i915_buddy_block_offset(root); - - if (offset != (prev_offset + prev_block_size)) { - pr_err("root offset mismatch\n"); - err = -EINVAL; - } - } - - block = list_first_entry_or_null(&mm->free_list[order], - struct i915_buddy_block, - link); - if (block != root) { - pr_err("root mismatch at order=%u\n", order); - err = -EINVAL; - } - - if (err) - break; - - prev = root; - total += i915_buddy_block_size(mm, root); - } - - if (!err) { - if (total != mm->size) { - pr_err("expected mm size=%llx, found=%llx\n", mm->size, - total); - err = -EINVAL; - } - return err; - } - - if (prev) { - pr_err("prev root(%u), dump:\n", i - 1); - igt_dump_block(mm, prev); - } - - if (root) { - pr_err("bad root(%u), dump:\n", i); - igt_dump_block(mm, root); - } - - return err; -} - -static void igt_mm_config(u64 *size, u64 *chunk_size) -{ - I915_RND_STATE(prng); - u32 s, ms; - - /* Nothing fancy, just try to get an interesting bit pattern */ - - prandom_seed_state(&prng, i915_selftest.random_seed); - - /* Let size be a random number of pages up to 8 GB (2M pages) */ - s = 1 + i915_prandom_u32_max_state((BIT(33 - 12)) - 1, &prng); - /* Let the chunk size be a random power of 2 less than size */ - ms = BIT(i915_prandom_u32_max_state(ilog2(s), &prng)); - /* Round size down to the chunk size */ - s &= -ms; - - /* Convert from pages to bytes */ - *chunk_size = (u64)ms << 12; - *size = (u64)s << 12; -} - -static int igt_buddy_alloc_smoke(void *arg) -{ - struct i915_buddy_mm mm; - IGT_TIMEOUT(end_time); - I915_RND_STATE(prng); - u64 chunk_size; - u64 mm_size; - int *order; - int err, i; - - igt_mm_config(&mm_size, &chunk_size); - - pr_info("buddy_init with size=%llx, chunk_size=%llx\n", mm_size, chunk_size); - - err = i915_buddy_init(&mm, mm_size, chunk_size); - if (err) { - pr_err("buddy_init failed(%d)\n", err); - return err; - } - - order = i915_random_order(mm.max_order + 1, &prng); - if (!order) - goto out_fini; - - for (i = 0; i <= mm.max_order; ++i) { - struct i915_buddy_block *block; - int max_order = order[i]; - bool timeout = false; - LIST_HEAD(blocks); - int order; - u64 total; - - err = igt_check_mm(&mm); - if (err) { - pr_err("pre-mm check failed, abort\n"); - break; - } - - pr_info("filling from max_order=%u\n", max_order); - - order = max_order; - total = 0; - - do { -retry: - block = i915_buddy_alloc(&mm, order); - if (IS_ERR(block)) { - err = PTR_ERR(block); - if (err == -ENOMEM) { - pr_info("buddy_alloc hit -ENOMEM with order=%d\n", - order); - } else { - if (order--) { - err = 0; - goto retry; - } - - pr_err("buddy_alloc with order=%d failed(%d)\n", - order, err); - } - - break; - } - - list_add_tail(&block->link, &blocks); - - if (i915_buddy_block_order(block) != order) { - pr_err("buddy_alloc order mismatch\n"); - err = -EINVAL; - break; - } - - total += i915_buddy_block_size(&mm, block); - - if (__igt_timeout(end_time, NULL)) { - timeout = true; - break; - } - } while (total < mm.size); - - if (!err) - err = igt_check_blocks(&mm, &blocks, total, false); - - i915_buddy_free_list(&mm, &blocks); - - if (!err) { - err = igt_check_mm(&mm); - if (err) - pr_err("post-mm check failed\n"); - } - - if (err || timeout) - break; - - cond_resched(); - } - - if (err == -ENOMEM) - err = 0; - - kfree(order); -out_fini: - i915_buddy_fini(&mm); - - return err; -} - -static int igt_buddy_alloc_pessimistic(void *arg) -{ - const unsigned int max_order = 16; - struct i915_buddy_block *block, *bn; - struct i915_buddy_mm mm; - unsigned int order; - LIST_HEAD(blocks); - int err; - - /* - * Create a pot-sized mm, then allocate one of each possible - * order within. This should leave the mm with exactly one - * page left. - */ - - err = i915_buddy_init(&mm, PAGE_SIZE << max_order, PAGE_SIZE); - if (err) { - pr_err("buddy_init failed(%d)\n", err); - return err; - } - GEM_BUG_ON(mm.max_order != max_order); - - for (order = 0; order < max_order; order++) { - block = i915_buddy_alloc(&mm, order); - if (IS_ERR(block)) { - pr_info("buddy_alloc hit -ENOMEM with order=%d\n", - order); - err = PTR_ERR(block); - goto err; - } - - list_add_tail(&block->link, &blocks); - } - - /* And now the last remaining block available */ - block = i915_buddy_alloc(&mm, 0); - if (IS_ERR(block)) { - pr_info("buddy_alloc hit -ENOMEM on final alloc\n"); - err = PTR_ERR(block); - goto err; - } - list_add_tail(&block->link, &blocks); - - /* Should be completely full! */ - for (order = max_order; order--; ) { - block = i915_buddy_alloc(&mm, order); - if (!IS_ERR(block)) { - pr_info("buddy_alloc unexpectedly succeeded at order %d, it should be full!", - order); - list_add_tail(&block->link, &blocks); - err = -EINVAL; - goto err; - } - } - - block = list_last_entry(&blocks, typeof(*block), link); - list_del(&block->link); - i915_buddy_free(&mm, block); - - /* As we free in increasing size, we make available larger blocks */ - order = 1; - list_for_each_entry_safe(block, bn, &blocks, link) { - list_del(&block->link); - i915_buddy_free(&mm, block); - - block = i915_buddy_alloc(&mm, order); - if (IS_ERR(block)) { - pr_info("buddy_alloc (realloc) hit -ENOMEM with order=%d\n", - order); - err = PTR_ERR(block); - goto err; - } - i915_buddy_free(&mm, block); - order++; - } - - /* To confirm, now the whole mm should be available */ - block = i915_buddy_alloc(&mm, max_order); - if (IS_ERR(block)) { - pr_info("buddy_alloc (realloc) hit -ENOMEM with order=%d\n", - max_order); - err = PTR_ERR(block); - goto err; - } - i915_buddy_free(&mm, block); - -err: - i915_buddy_free_list(&mm, &blocks); - i915_buddy_fini(&mm); - return err; -} - -static int igt_buddy_alloc_optimistic(void *arg) -{ - const int max_order = 16; - struct i915_buddy_block *block; - struct i915_buddy_mm mm; - LIST_HEAD(blocks); - int order; - int err; - - /* - * Create a mm with one block of each order available, and - * try to allocate them all. - */ - - err = i915_buddy_init(&mm, - PAGE_SIZE * ((1 << (max_order + 1)) - 1), - PAGE_SIZE); - if (err) { - pr_err("buddy_init failed(%d)\n", err); - return err; - } - GEM_BUG_ON(mm.max_order != max_order); - - for (order = 0; order <= max_order; order++) { - block = i915_buddy_alloc(&mm, order); - if (IS_ERR(block)) { - pr_info("buddy_alloc hit -ENOMEM with order=%d\n", - order); - err = PTR_ERR(block); - goto err; - } - - list_add_tail(&block->link, &blocks); - } - - /* Should be completely full! */ - block = i915_buddy_alloc(&mm, 0); - if (!IS_ERR(block)) { - pr_info("buddy_alloc unexpectedly succeeded, it should be full!"); - list_add_tail(&block->link, &blocks); - err = -EINVAL; - goto err; - } - -err: - i915_buddy_free_list(&mm, &blocks); - i915_buddy_fini(&mm); - return err; -} - -static int igt_buddy_alloc_pathological(void *arg) -{ - const int max_order = 16; - struct i915_buddy_block *block; - struct i915_buddy_mm mm; - LIST_HEAD(blocks); - LIST_HEAD(holes); - int order, top; - int err; - - /* - * Create a pot-sized mm, then allocate one of each possible - * order within. This should leave the mm with exactly one - * page left. Free the largest block, then whittle down again. - * Eventually we will have a fully 50% fragmented mm. - */ - - err = i915_buddy_init(&mm, PAGE_SIZE << max_order, PAGE_SIZE); - if (err) { - pr_err("buddy_init failed(%d)\n", err); - return err; - } - GEM_BUG_ON(mm.max_order != max_order); - - for (top = max_order; top; top--) { - /* Make room by freeing the largest allocated block */ - block = list_first_entry_or_null(&blocks, typeof(*block), link); - if (block) { - list_del(&block->link); - i915_buddy_free(&mm, block); - } - - for (order = top; order--; ) { - block = i915_buddy_alloc(&mm, order); - if (IS_ERR(block)) { - pr_info("buddy_alloc hit -ENOMEM with order=%d, top=%d\n", - order, top); - err = PTR_ERR(block); - goto err; - } - list_add_tail(&block->link, &blocks); - } - - /* There should be one final page for this sub-allocation */ - block = i915_buddy_alloc(&mm, 0); - if (IS_ERR(block)) { - pr_info("buddy_alloc hit -ENOMEM for hole\n"); - err = PTR_ERR(block); - goto err; - } - list_add_tail(&block->link, &holes); - - block = i915_buddy_alloc(&mm, top); - if (!IS_ERR(block)) { - pr_info("buddy_alloc unexpectedly succeeded at top-order %d/%d, it should be full!", - top, max_order); - list_add_tail(&block->link, &blocks); - err = -EINVAL; - goto err; - } - } - - i915_buddy_free_list(&mm, &holes); - - /* Nothing larger than blocks of chunk_size now available */ - for (order = 1; order <= max_order; order++) { - block = i915_buddy_alloc(&mm, order); - if (!IS_ERR(block)) { - pr_info("buddy_alloc unexpectedly succeeded at order %d, it should be full!", - order); - list_add_tail(&block->link, &blocks); - err = -EINVAL; - goto err; - } - } - -err: - list_splice_tail(&holes, &blocks); - i915_buddy_free_list(&mm, &blocks); - i915_buddy_fini(&mm); - return err; -} - -static int igt_buddy_alloc_range(void *arg) -{ - struct i915_buddy_mm mm; - unsigned long page_num; - LIST_HEAD(blocks); - u64 chunk_size; - u64 offset; - u64 size; - u64 rem; - int err; - - igt_mm_config(&size, &chunk_size); - - pr_info("buddy_init with size=%llx, chunk_size=%llx\n", size, chunk_size); - - err = i915_buddy_init(&mm, size, chunk_size); - if (err) { - pr_err("buddy_init failed(%d)\n", err); - return err; - } - - err = igt_check_mm(&mm); - if (err) { - pr_err("pre-mm check failed, abort, abort, abort!\n"); - goto err_fini; - } - - rem = mm.size; - offset = 0; - - for_each_prime_number_from(page_num, 1, ULONG_MAX - 1) { - struct i915_buddy_block *block; - LIST_HEAD(tmp); - - size = min(page_num * mm.chunk_size, rem); - - err = i915_buddy_alloc_range(&mm, &tmp, offset, size); - if (err) { - if (err == -ENOMEM) { - pr_info("alloc_range hit -ENOMEM with size=%llx\n", - size); - } else { - pr_err("alloc_range with offset=%llx, size=%llx failed(%d)\n", - offset, size, err); - } - - break; - } - - block = list_first_entry_or_null(&tmp, - struct i915_buddy_block, - link); - if (!block) { - pr_err("alloc_range has no blocks\n"); - err = -EINVAL; - break; - } - - if (i915_buddy_block_offset(block) != offset) { - pr_err("alloc_range start offset mismatch, found=%llx, expected=%llx\n", - i915_buddy_block_offset(block), offset); - err = -EINVAL; - } - - if (!err) - err = igt_check_blocks(&mm, &tmp, size, true); - - list_splice_tail(&tmp, &blocks); - - if (err) - break; - - offset += size; - - rem -= size; - if (!rem) - break; - - cond_resched(); - } - - if (err == -ENOMEM) - err = 0; - - i915_buddy_free_list(&mm, &blocks); - - if (!err) { - err = igt_check_mm(&mm); - if (err) - pr_err("post-mm check failed\n"); - } - -err_fini: - i915_buddy_fini(&mm); - - return err; -} - -static int igt_buddy_alloc_limit(void *arg) -{ - struct i915_buddy_block *block; - struct i915_buddy_mm mm; - const u64 size = U64_MAX; - int err; - - err = i915_buddy_init(&mm, size, PAGE_SIZE); - if (err) - return err; - - if (mm.max_order != I915_BUDDY_MAX_ORDER) { - pr_err("mm.max_order(%d) != %d\n", - mm.max_order, I915_BUDDY_MAX_ORDER); - err = -EINVAL; - goto out_fini; - } - - block = i915_buddy_alloc(&mm, mm.max_order); - if (IS_ERR(block)) { - err = PTR_ERR(block); - goto out_fini; - } - - if (i915_buddy_block_order(block) != mm.max_order) { - pr_err("block order(%d) != %d\n", - i915_buddy_block_order(block), mm.max_order); - err = -EINVAL; - goto out_free; - } - - if (i915_buddy_block_size(&mm, block) != - BIT_ULL(mm.max_order) * PAGE_SIZE) { - pr_err("block size(%llu) != %llu\n", - i915_buddy_block_size(&mm, block), - BIT_ULL(mm.max_order) * PAGE_SIZE); - err = -EINVAL; - goto out_free; - } - -out_free: - i915_buddy_free(&mm, block); -out_fini: - i915_buddy_fini(&mm); - return err; -} - -int i915_buddy_mock_selftests(void) -{ - static const struct i915_subtest tests[] = { - SUBTEST(igt_buddy_alloc_pessimistic), - SUBTEST(igt_buddy_alloc_optimistic), - SUBTEST(igt_buddy_alloc_pathological), - SUBTEST(igt_buddy_alloc_smoke), - SUBTEST(igt_buddy_alloc_range), - SUBTEST(igt_buddy_alloc_limit), - }; - - return i915_subtests(tests, NULL); -} diff --git a/drivers/gpu/drm/i915/selftests/i915_mock_selftests.h b/drivers/gpu/drm/i915/selftests/i915_mock_selftests.h index 793fb28a770d..0c22e0fc9059 100644 --- a/drivers/gpu/drm/i915/selftests/i915_mock_selftests.h +++ b/drivers/gpu/drm/i915/selftests/i915_mock_selftests.h @@ -33,4 +33,3 @@ selftest(evict, i915_gem_evict_mock_selftests) selftest(gtt, i915_gem_gtt_mock_selftests) selftest(hugepages, i915_gem_huge_page_mock_selftests) selftest(memory_region, intel_memory_region_mock_selftests) -selftest(buddy, i915_buddy_mock_selftests) diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c index 8255561ff853..7acba1d2135e 100644 --- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c +++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c @@ -6,6 +6,8 @@ #include <linux/prime_numbers.h> #include <linux/sort.h> +#include <drm/drm_buddy.h> + #include "../i915_selftest.h" #include "mock_drm.h" @@ -20,7 +22,6 @@ #include "gt/intel_engine_pm.h" #include "gt/intel_engine_user.h" #include "gt/intel_gt.h" -#include "i915_buddy.h" #include "gt/intel_migrate.h" #include "i915_memcpy.h" #include "i915_ttm_buddy_manager.h" @@ -369,7 +370,7 @@ static int igt_mock_splintered_region(void *arg) struct drm_i915_private *i915 = mem->i915; struct i915_ttm_buddy_resource *res; struct drm_i915_gem_object *obj; - struct i915_buddy_mm *mm; + struct drm_buddy *mm; unsigned int expected_order; LIST_HEAD(objects); u64 size; @@ -454,8 +455,8 @@ static int igt_mock_max_segment(void *arg) struct drm_i915_private *i915 = mem->i915; struct i915_ttm_buddy_resource *res; struct drm_i915_gem_object *obj; - struct i915_buddy_block *block; - struct i915_buddy_mm *mm; + struct drm_buddy_block *block; + struct drm_buddy *mm; struct list_head *blocks; struct scatterlist *sg; LIST_HEAD(objects); @@ -485,8 +486,8 @@ static int igt_mock_max_segment(void *arg) mm = res->mm; size = 0; list_for_each_entry(block, blocks, link) { - if (i915_buddy_block_size(mm, block) > size) - size = i915_buddy_block_size(mm, block); + if (drm_buddy_block_size(mm, block) > size) + size = drm_buddy_block_size(mm, block); } if (size < max_segment) { pr_err("%s: Failed to create a huge contiguous block [> %u], largest block %lld\n", diff --git a/drivers/gpu/drm/i915/vlv_sideband.c b/drivers/gpu/drm/i915/vlv_sideband.c index ed2ac5752ac4..c26001300ebd 100644 --- a/drivers/gpu/drm/i915/vlv_sideband.c +++ b/drivers/gpu/drm/i915/vlv_sideband.c @@ -5,6 +5,7 @@ #include "i915_drv.h" #include "i915_iosf_mbi.h" +#include "i915_reg.h" #include "vlv_sideband.h" /* diff --git a/drivers/gpu/drm/i915/vlv_sideband.h b/drivers/gpu/drm/i915/vlv_sideband.h index d7732f612e7f..9ce283d96b80 100644 --- a/drivers/gpu/drm/i915/vlv_sideband.h +++ b/drivers/gpu/drm/i915/vlv_sideband.h @@ -9,6 +9,8 @@ #include <linux/bitops.h> #include <linux/types.h> +#include "vlv_sideband_reg.h" + enum pipe; struct drm_i915_private; diff --git a/drivers/gpu/drm/i915/vlv_sideband_reg.h b/drivers/gpu/drm/i915/vlv_sideband_reg.h new file mode 100644 index 000000000000..b7fbff3d0409 --- /dev/null +++ b/drivers/gpu/drm/i915/vlv_sideband_reg.h @@ -0,0 +1,180 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2022 Intel Corporation + */ + +#ifndef _VLV_SIDEBAND_REG_H_ +#define _VLV_SIDEBAND_REG_H_ + +/* See configdb bunit SB addr map */ +#define BUNIT_REG_BISOC 0x11 + +/* PUNIT_REG_*SSPM0 */ +#define _SSPM0_SSC(val) ((val) << 0) +#define SSPM0_SSC_MASK _SSPM0_SSC(0x3) +#define SSPM0_SSC_PWR_ON _SSPM0_SSC(0x0) +#define SSPM0_SSC_CLK_GATE _SSPM0_SSC(0x1) +#define SSPM0_SSC_RESET _SSPM0_SSC(0x2) +#define SSPM0_SSC_PWR_GATE _SSPM0_SSC(0x3) +#define _SSPM0_SSS(val) ((val) << 24) +#define SSPM0_SSS_MASK _SSPM0_SSS(0x3) +#define SSPM0_SSS_PWR_ON _SSPM0_SSS(0x0) +#define SSPM0_SSS_CLK_GATE _SSPM0_SSS(0x1) +#define SSPM0_SSS_RESET _SSPM0_SSS(0x2) +#define SSPM0_SSS_PWR_GATE _SSPM0_SSS(0x3) + +/* PUNIT_REG_*SSPM1 */ +#define SSPM1_FREQSTAT_SHIFT 24 +#define SSPM1_FREQSTAT_MASK (0x1f << SSPM1_FREQSTAT_SHIFT) +#define SSPM1_FREQGUAR_SHIFT 8 +#define SSPM1_FREQGUAR_MASK (0x1f << SSPM1_FREQGUAR_SHIFT) +#define SSPM1_FREQ_SHIFT 0 +#define SSPM1_FREQ_MASK (0x1f << SSPM1_FREQ_SHIFT) + +#define PUNIT_REG_VEDSSPM0 0x32 +#define PUNIT_REG_VEDSSPM1 0x33 + +#define PUNIT_REG_DSPSSPM 0x36 +#define DSPFREQSTAT_SHIFT_CHV 24 +#define DSPFREQSTAT_MASK_CHV (0x1f << DSPFREQSTAT_SHIFT_CHV) +#define DSPFREQGUAR_SHIFT_CHV 8 +#define DSPFREQGUAR_MASK_CHV (0x1f << DSPFREQGUAR_SHIFT_CHV) +#define DSPFREQSTAT_SHIFT 30 +#define DSPFREQSTAT_MASK (0x3 << DSPFREQSTAT_SHIFT) +#define DSPFREQGUAR_SHIFT 14 +#define DSPFREQGUAR_MASK (0x3 << DSPFREQGUAR_SHIFT) +#define DSP_MAXFIFO_PM5_STATUS (1 << 22) /* chv */ +#define DSP_AUTO_CDCLK_GATE_DISABLE (1 << 7) /* chv */ +#define DSP_MAXFIFO_PM5_ENABLE (1 << 6) /* chv */ +#define _DP_SSC(val, pipe) ((val) << (2 * (pipe))) +#define DP_SSC_MASK(pipe) _DP_SSC(0x3, (pipe)) +#define DP_SSC_PWR_ON(pipe) _DP_SSC(0x0, (pipe)) +#define DP_SSC_CLK_GATE(pipe) _DP_SSC(0x1, (pipe)) +#define DP_SSC_RESET(pipe) _DP_SSC(0x2, (pipe)) +#define DP_SSC_PWR_GATE(pipe) _DP_SSC(0x3, (pipe)) +#define _DP_SSS(val, pipe) ((val) << (2 * (pipe) + 16)) +#define DP_SSS_MASK(pipe) _DP_SSS(0x3, (pipe)) +#define DP_SSS_PWR_ON(pipe) _DP_SSS(0x0, (pipe)) +#define DP_SSS_CLK_GATE(pipe) _DP_SSS(0x1, (pipe)) +#define DP_SSS_RESET(pipe) _DP_SSS(0x2, (pipe)) +#define DP_SSS_PWR_GATE(pipe) _DP_SSS(0x3, (pipe)) + +#define PUNIT_REG_ISPSSPM0 0x39 +#define PUNIT_REG_ISPSSPM1 0x3a + +#define PUNIT_REG_PWRGT_CTRL 0x60 +#define PUNIT_REG_PWRGT_STATUS 0x61 +#define PUNIT_PWRGT_MASK(pw_idx) (3 << ((pw_idx) * 2)) +#define PUNIT_PWRGT_PWR_ON(pw_idx) (0 << ((pw_idx) * 2)) +#define PUNIT_PWRGT_CLK_GATE(pw_idx) (1 << ((pw_idx) * 2)) +#define PUNIT_PWRGT_RESET(pw_idx) (2 << ((pw_idx) * 2)) +#define PUNIT_PWRGT_PWR_GATE(pw_idx) (3 << ((pw_idx) * 2)) + +#define PUNIT_PWGT_IDX_RENDER 0 +#define PUNIT_PWGT_IDX_MEDIA 1 +#define PUNIT_PWGT_IDX_DISP2D 3 +#define PUNIT_PWGT_IDX_DPIO_CMN_BC 5 +#define PUNIT_PWGT_IDX_DPIO_TX_B_LANES_01 6 +#define PUNIT_PWGT_IDX_DPIO_TX_B_LANES_23 7 +#define PUNIT_PWGT_IDX_DPIO_TX_C_LANES_01 8 +#define PUNIT_PWGT_IDX_DPIO_TX_C_LANES_23 9 +#define PUNIT_PWGT_IDX_DPIO_RX0 10 +#define PUNIT_PWGT_IDX_DPIO_RX1 11 +#define PUNIT_PWGT_IDX_DPIO_CMN_D 12 + +#define PUNIT_REG_GPU_LFM 0xd3 +#define PUNIT_REG_GPU_FREQ_REQ 0xd4 +#define PUNIT_REG_GPU_FREQ_STS 0xd8 +#define GPLLENABLE (1 << 4) +#define GENFREQSTATUS (1 << 0) +#define PUNIT_REG_MEDIA_TURBO_FREQ_REQ 0xdc +#define PUNIT_REG_CZ_TIMESTAMP 0xce + +#define PUNIT_FUSE_BUS2 0xf6 /* bits 47:40 */ +#define PUNIT_FUSE_BUS1 0xf5 /* bits 55:48 */ + +#define FB_GFX_FMAX_AT_VMAX_FUSE 0x136 +#define FB_GFX_FREQ_FUSE_MASK 0xff +#define FB_GFX_FMAX_AT_VMAX_2SS4EU_FUSE_SHIFT 24 +#define FB_GFX_FMAX_AT_VMAX_2SS6EU_FUSE_SHIFT 16 +#define FB_GFX_FMAX_AT_VMAX_2SS8EU_FUSE_SHIFT 8 + +#define FB_GFX_FMIN_AT_VMIN_FUSE 0x137 +#define FB_GFX_FMIN_AT_VMIN_FUSE_SHIFT 8 + +#define PUNIT_REG_DDR_SETUP2 0x139 +#define FORCE_DDR_FREQ_REQ_ACK (1 << 8) +#define FORCE_DDR_LOW_FREQ (1 << 1) +#define FORCE_DDR_HIGH_FREQ (1 << 0) + +#define PUNIT_GPU_STATUS_REG 0xdb +#define PUNIT_GPU_STATUS_MAX_FREQ_SHIFT 16 +#define PUNIT_GPU_STATUS_MAX_FREQ_MASK 0xff +#define PUNIT_GPU_STATIS_GFX_MIN_FREQ_SHIFT 8 +#define PUNIT_GPU_STATUS_GFX_MIN_FREQ_MASK 0xff + +#define PUNIT_GPU_DUTYCYCLE_REG 0xdf +#define PUNIT_GPU_DUTYCYCLE_RPE_FREQ_SHIFT 8 +#define PUNIT_GPU_DUTYCYCLE_RPE_FREQ_MASK 0xff + +#define IOSF_NC_FB_GFX_FREQ_FUSE 0x1c +#define FB_GFX_MAX_FREQ_FUSE_SHIFT 3 +#define FB_GFX_MAX_FREQ_FUSE_MASK 0x000007f8 +#define FB_GFX_FGUARANTEED_FREQ_FUSE_SHIFT 11 +#define FB_GFX_FGUARANTEED_FREQ_FUSE_MASK 0x0007f800 +#define IOSF_NC_FB_GFX_FMAX_FUSE_HI 0x34 +#define FB_FMAX_VMIN_FREQ_HI_MASK 0x00000007 +#define IOSF_NC_FB_GFX_FMAX_FUSE_LO 0x30 +#define FB_FMAX_VMIN_FREQ_LO_SHIFT 27 +#define FB_FMAX_VMIN_FREQ_LO_MASK 0xf8000000 + +#define VLV_TURBO_SOC_OVERRIDE 0x04 +#define VLV_OVERRIDE_EN 1 +#define VLV_SOC_TDP_EN (1 << 1) +#define VLV_BIAS_CPU_125_SOC_875 (6 << 2) +#define CHV_BIAS_CPU_50_SOC_50 (3 << 2) + +/* vlv2 north clock has */ +#define CCK_FUSE_REG 0x8 +#define CCK_FUSE_HPLL_FREQ_MASK 0x3 +#define CCK_REG_DSI_PLL_FUSE 0x44 +#define CCK_REG_DSI_PLL_CONTROL 0x48 +#define DSI_PLL_VCO_EN (1 << 31) +#define DSI_PLL_LDO_GATE (1 << 30) +#define DSI_PLL_P1_POST_DIV_SHIFT 17 +#define DSI_PLL_P1_POST_DIV_MASK (0x1ff << 17) +#define DSI_PLL_P2_MUX_DSI0_DIV2 (1 << 13) +#define DSI_PLL_P3_MUX_DSI1_DIV2 (1 << 12) +#define DSI_PLL_MUX_MASK (3 << 9) +#define DSI_PLL_MUX_DSI0_DSIPLL (0 << 10) +#define DSI_PLL_MUX_DSI0_CCK (1 << 10) +#define DSI_PLL_MUX_DSI1_DSIPLL (0 << 9) +#define DSI_PLL_MUX_DSI1_CCK (1 << 9) +#define DSI_PLL_CLK_GATE_MASK (0xf << 5) +#define DSI_PLL_CLK_GATE_DSI0_DSIPLL (1 << 8) +#define DSI_PLL_CLK_GATE_DSI1_DSIPLL (1 << 7) +#define DSI_PLL_CLK_GATE_DSI0_CCK (1 << 6) +#define DSI_PLL_CLK_GATE_DSI1_CCK (1 << 5) +#define DSI_PLL_LOCK (1 << 0) +#define CCK_REG_DSI_PLL_DIVIDER 0x4c +#define DSI_PLL_LFSR (1 << 31) +#define DSI_PLL_FRACTION_EN (1 << 30) +#define DSI_PLL_FRAC_COUNTER_SHIFT 27 +#define DSI_PLL_FRAC_COUNTER_MASK (7 << 27) +#define DSI_PLL_USYNC_CNT_SHIFT 18 +#define DSI_PLL_USYNC_CNT_MASK (0x1ff << 18) +#define DSI_PLL_N1_DIV_SHIFT 16 +#define DSI_PLL_N1_DIV_MASK (3 << 16) +#define DSI_PLL_M1_DIV_SHIFT 0 +#define DSI_PLL_M1_DIV_MASK (0x1ff << 0) +#define CCK_CZ_CLOCK_CONTROL 0x62 +#define CCK_GPLL_CLOCK_CONTROL 0x67 +#define CCK_DISPLAY_CLOCK_CONTROL 0x6b +#define CCK_DISPLAY_REF_CLOCK_CONTROL 0x6c +#define CCK_TRUNK_FORCE_ON (1 << 17) +#define CCK_TRUNK_FORCE_OFF (1 << 16) +#define CCK_FREQUENCY_STATUS (0x1f << 8) +#define CCK_FREQUENCY_STATUS_SHIFT 8 +#define CCK_FREQUENCY_VALUES (0x1f << 0) + +#endif /* _VLV_SIDEBAND_REG_H_ */ diff --git a/drivers/gpu/drm/i915/vlv_suspend.c b/drivers/gpu/drm/i915/vlv_suspend.c index 23adb64d640a..1d9da32195c2 100644 --- a/drivers/gpu/drm/i915/vlv_suspend.c +++ b/drivers/gpu/drm/i915/vlv_suspend.c @@ -14,6 +14,8 @@ #include "intel_pm.h" #include "vlv_suspend.h" +#include "gt/intel_gt_regs.h" + struct vlv_s0ix_state { /* GAM */ u32 wr_watermark; diff --git a/drivers/gpu/drm/imx/dcss/dcss-drv.c b/drivers/gpu/drm/imx/dcss/dcss-drv.c index 8dc2f85c514b..24147ee7080e 100644 --- a/drivers/gpu/drm/imx/dcss/dcss-drv.c +++ b/drivers/gpu/drm/imx/dcss/dcss-drv.c @@ -6,6 +6,7 @@ #include <linux/module.h> #include <linux/kernel.h> #include <linux/platform_device.h> +#include <drm/drm_module.h> #include <drm/drm_of.h> #include "dcss-dev.h" @@ -131,7 +132,7 @@ static struct platform_driver dcss_platform_driver = { }, }; -module_platform_driver(dcss_platform_driver); +drm_module_platform_driver(dcss_platform_driver); MODULE_AUTHOR("Laurentiu Palcu <laurentiu.palcu@nxp.com>"); MODULE_DESCRIPTION("DCSS driver for i.MX8MQ"); diff --git a/drivers/gpu/drm/ingenic/ingenic-drm-drv.c b/drivers/gpu/drm/ingenic/ingenic-drm-drv.c index b4943a56be09..542c4af70661 100644 --- a/drivers/gpu/drm/ingenic/ingenic-drm-drv.c +++ b/drivers/gpu/drm/ingenic/ingenic-drm-drv.c @@ -6,6 +6,7 @@ #include "ingenic-drm.h" +#include <linux/bitfield.h> #include <linux/component.h> #include <linux/clk.h> #include <linux/dma-mapping.h> @@ -49,6 +50,11 @@ struct ingenic_dma_hwdesc { u32 addr; u32 id; u32 cmd; + /* extended hw descriptor for jz4780 */ + u32 offsize; + u32 pagewidth; + u32 cpos; + u32 dessize; } __aligned(16); struct ingenic_dma_hwdescs { @@ -60,6 +66,7 @@ struct jz_soc_info { bool needs_dev_clk; bool has_osd; bool map_noncoherent; + bool use_extended_hwdesc; unsigned int max_width, max_height; const u32 *formats_f0, *formats_f1; unsigned int num_formats_f0, num_formats_f1; @@ -173,7 +180,6 @@ static const struct regmap_config ingenic_drm_regmap_config = { .val_bits = 32, .reg_stride = 4, - .max_register = JZ_REG_LCD_SIZE1, .writeable_reg = ingenic_drm_writeable_reg, }; @@ -447,6 +453,9 @@ static int ingenic_drm_plane_atomic_check(struct drm_plane *plane, if (!crtc) return 0; + if (plane == &priv->f0) + return -EINVAL; + crtc_state = drm_atomic_get_existing_crtc_state(state, crtc); if (WARN_ON(!crtc_state)) @@ -663,6 +672,33 @@ static void ingenic_drm_plane_atomic_update(struct drm_plane *plane, hwdesc->cmd = JZ_LCD_CMD_EOF_IRQ | (width * height * cpp / 4); hwdesc->next = dma_hwdesc_addr(priv, next_id); + if (priv->soc_info->use_extended_hwdesc) { + hwdesc->cmd |= JZ_LCD_CMD_FRM_ENABLE; + + /* Extended 8-byte descriptor */ + hwdesc->cpos = 0; + hwdesc->offsize = 0; + hwdesc->pagewidth = 0; + + switch (newstate->fb->format->format) { + case DRM_FORMAT_XRGB1555: + hwdesc->cpos |= JZ_LCD_CPOS_RGB555; + fallthrough; + case DRM_FORMAT_RGB565: + hwdesc->cpos |= JZ_LCD_CPOS_BPP_15_16; + break; + case DRM_FORMAT_XRGB8888: + hwdesc->cpos |= JZ_LCD_CPOS_BPP_18_24; + break; + } + hwdesc->cpos |= (JZ_LCD_CPOS_COEFFICIENT_1 << + JZ_LCD_CPOS_COEFFICIENT_OFFSET); + hwdesc->dessize = + (0xff << JZ_LCD_DESSIZE_ALPHA_OFFSET) | + FIELD_PREP(JZ_LCD_DESSIZE_HEIGHT_MASK, height - 1) | + FIELD_PREP(JZ_LCD_DESSIZE_WIDTH_MASK, width - 1); + } + if (drm_atomic_crtc_needs_modeset(crtc_state)) { fourcc = newstate->fb->format->format; @@ -694,6 +730,9 @@ static void ingenic_drm_encoder_atomic_mode_set(struct drm_encoder *encoder, | JZ_LCD_CFG_SPL_DISABLE | JZ_LCD_CFG_REV_DISABLE; } + if (priv->soc_info->use_extended_hwdesc) + cfg |= JZ_LCD_CFG_DESCRIPTOR_8; + if (mode->flags & DRM_MODE_FLAG_NHSYNC) cfg |= JZ_LCD_CFG_HSYNC_ACTIVE_LOW; if (mode->flags & DRM_MODE_FLAG_NVSYNC) @@ -1011,6 +1050,8 @@ static int ingenic_drm_bind(struct device *dev, bool has_components) struct ingenic_drm_bridge *ib; struct drm_device *drm; void __iomem *base; + struct resource *res; + struct regmap_config regmap_config; long parent_rate; unsigned int i, clone_mask = 0; int ret, irq; @@ -1056,14 +1097,16 @@ static int ingenic_drm_bind(struct device *dev, bool has_components) drm->mode_config.funcs = &ingenic_drm_mode_config_funcs; drm->mode_config.helper_private = &ingenic_drm_mode_config_helpers; - base = devm_platform_ioremap_resource(pdev, 0); + base = devm_platform_get_and_ioremap_resource(pdev, 0, &res); if (IS_ERR(base)) { dev_err(dev, "Failed to get memory resource\n"); return PTR_ERR(base); } + regmap_config = ingenic_drm_regmap_config; + regmap_config.max_register = res->end - res->start; priv->map = devm_regmap_init_mmio(dev, base, - &ingenic_drm_regmap_config); + ®map_config); if (IS_ERR(priv->map)) { dev_err(dev, "Failed to create regmap\n"); return PTR_ERR(priv->map); @@ -1465,10 +1508,23 @@ static const struct jz_soc_info jz4770_soc_info = { .num_formats_f0 = ARRAY_SIZE(jz4770_formats_f0), }; +static const struct jz_soc_info jz4780_soc_info = { + .needs_dev_clk = true, + .has_osd = true, + .use_extended_hwdesc = true, + .max_width = 4096, + .max_height = 2048, + .formats_f1 = jz4770_formats_f1, + .num_formats_f1 = ARRAY_SIZE(jz4770_formats_f1), + .formats_f0 = jz4770_formats_f0, + .num_formats_f0 = ARRAY_SIZE(jz4770_formats_f0), +}; + static const struct of_device_id ingenic_drm_of_match[] = { { .compatible = "ingenic,jz4740-lcd", .data = &jz4740_soc_info }, { .compatible = "ingenic,jz4725b-lcd", .data = &jz4725b_soc_info }, { .compatible = "ingenic,jz4770-lcd", .data = &jz4770_soc_info }, + { .compatible = "ingenic,jz4780-lcd", .data = &jz4780_soc_info }, { /* sentinel */ }, }; MODULE_DEVICE_TABLE(of, ingenic_drm_of_match); diff --git a/drivers/gpu/drm/ingenic/ingenic-drm.h b/drivers/gpu/drm/ingenic/ingenic-drm.h index 22654ac1dde1..cb1d09b62588 100644 --- a/drivers/gpu/drm/ingenic/ingenic-drm.h +++ b/drivers/gpu/drm/ingenic/ingenic-drm.h @@ -44,8 +44,11 @@ #define JZ_REG_LCD_XYP1 0x124 #define JZ_REG_LCD_SIZE0 0x128 #define JZ_REG_LCD_SIZE1 0x12c +#define JZ_REG_LCD_PCFG 0x2c0 #define JZ_LCD_CFG_SLCD BIT(31) +#define JZ_LCD_CFG_DESCRIPTOR_8 BIT(28) +#define JZ_LCD_CFG_RECOVER_FIFO_UNDERRUN BIT(25) #define JZ_LCD_CFG_PS_DISABLE BIT(23) #define JZ_LCD_CFG_CLS_DISABLE BIT(22) #define JZ_LCD_CFG_SPL_DISABLE BIT(21) @@ -63,6 +66,7 @@ #define JZ_LCD_CFG_DE_ACTIVE_LOW BIT(9) #define JZ_LCD_CFG_VSYNC_ACTIVE_LOW BIT(8) #define JZ_LCD_CFG_18_BIT BIT(7) +#define JZ_LCD_CFG_24_BIT BIT(6) #define JZ_LCD_CFG_PDW (BIT(5) | BIT(4)) #define JZ_LCD_CFG_MODE_GENERIC_16BIT 0 @@ -132,6 +136,7 @@ #define JZ_LCD_CMD_SOF_IRQ BIT(31) #define JZ_LCD_CMD_EOF_IRQ BIT(30) #define JZ_LCD_CMD_ENABLE_PAL BIT(28) +#define JZ_LCD_CMD_FRM_ENABLE BIT(26) #define JZ_LCD_SYNC_MASK 0x3ff @@ -153,6 +158,7 @@ #define JZ_LCD_RGBC_EVEN_BGR (0x5 << 0) #define JZ_LCD_OSDC_OSDEN BIT(0) +#define JZ_LCD_OSDC_ALPHAEN BIT(2) #define JZ_LCD_OSDC_F0EN BIT(3) #define JZ_LCD_OSDC_F1EN BIT(4) @@ -176,6 +182,38 @@ #define JZ_LCD_SIZE01_WIDTH_LSB 0 #define JZ_LCD_SIZE01_HEIGHT_LSB 16 +#define JZ_LCD_DESSIZE_ALPHA_OFFSET 24 +#define JZ_LCD_DESSIZE_HEIGHT_MASK GENMASK(23, 12) +#define JZ_LCD_DESSIZE_WIDTH_MASK GENMASK(11, 0) + +#define JZ_LCD_CPOS_BPP_15_16 (4 << 27) +#define JZ_LCD_CPOS_BPP_18_24 (5 << 27) +#define JZ_LCD_CPOS_BPP_30 (7 << 27) +#define JZ_LCD_CPOS_RGB555 BIT(30) +#define JZ_LCD_CPOS_PREMULTIPLY_LCD BIT(26) +#define JZ_LCD_CPOS_COEFFICIENT_OFFSET 24 +#define JZ_LCD_CPOS_COEFFICIENT_0 0 +#define JZ_LCD_CPOS_COEFFICIENT_1 1 +#define JZ_LCD_CPOS_COEFFICIENT_ALPHA1 2 +#define JZ_LCD_CPOS_COEFFICIENT_1_ALPHA1 3 + +#define JZ_LCD_RGBC_RGB_PADDING BIT(15) +#define JZ_LCD_RGBC_RGB_PADDING_FIRST BIT(14) +#define JZ_LCD_RGBC_422 BIT(8) +#define JZ_LCD_RGBC_RGB_FORMAT_ENABLE BIT(7) + +#define JZ_LCD_PCFG_PRI_MODE BIT(31) +#define JZ_LCD_PCFG_HP_BST_4 (0 << 28) +#define JZ_LCD_PCFG_HP_BST_8 (1 << 28) +#define JZ_LCD_PCFG_HP_BST_16 (2 << 28) +#define JZ_LCD_PCFG_HP_BST_32 (3 << 28) +#define JZ_LCD_PCFG_HP_BST_64 (4 << 28) +#define JZ_LCD_PCFG_HP_BST_16_CONT (5 << 28) +#define JZ_LCD_PCFG_HP_BST_DISABLE (7 << 28) +#define JZ_LCD_PCFG_THRESHOLD2_OFFSET 18 +#define JZ_LCD_PCFG_THRESHOLD1_OFFSET 9 +#define JZ_LCD_PCFG_THRESHOLD0_OFFSET 0 + struct device; struct drm_plane; struct drm_plane_state; diff --git a/drivers/gpu/drm/lima/lima_gem.c b/drivers/gpu/drm/lima/lima_gem.c index f9a9198ef198..d0c2b1422b3b 100644 --- a/drivers/gpu/drm/lima/lima_gem.c +++ b/drivers/gpu/drm/lima/lima_gem.c @@ -2,6 +2,7 @@ /* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */ #include <linux/mm.h> +#include <linux/iosys-map.h> #include <linux/sync_file.h> #include <linux/pagemap.h> #include <linux/shmem_fs.h> @@ -182,7 +183,7 @@ static int lima_gem_pin(struct drm_gem_object *obj) return drm_gem_shmem_pin(&bo->base); } -static int lima_gem_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) +static int lima_gem_vmap(struct drm_gem_object *obj, struct iosys_map *map) { struct lima_bo *bo = to_lima_bo(obj); diff --git a/drivers/gpu/drm/lima/lima_sched.c b/drivers/gpu/drm/lima/lima_sched.c index 5612d73f238f..390c969f74ad 100644 --- a/drivers/gpu/drm/lima/lima_sched.c +++ b/drivers/gpu/drm/lima/lima_sched.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 OR MIT /* Copyright 2017-2019 Qiang Yu <yuq825@gmail.com> */ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/kthread.h> #include <linux/slab.h> #include <linux/vmalloc.h> @@ -284,7 +284,7 @@ static void lima_sched_build_error_task_list(struct lima_sched_task *task) struct lima_dump_chunk_buffer *buffer_chunk; u32 size, task_size, mem_size; int i; - struct dma_buf_map map; + struct iosys_map map; int ret; mutex_lock(&dev->error_task_list_lock); diff --git a/drivers/gpu/drm/mediatek/mtk_drm_gem.c b/drivers/gpu/drm/mediatek/mtk_drm_gem.c index d0544962cfc1..139d7724c6d0 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_gem.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_gem.c @@ -220,7 +220,7 @@ struct drm_gem_object *mtk_gem_prime_import_sg_table(struct drm_device *dev, return &mtk_gem->base; } -int mtk_drm_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) +int mtk_drm_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map) { struct mtk_drm_gem_obj *mtk_gem = to_mtk_gem_obj(obj); struct sg_table *sgt = NULL; @@ -247,12 +247,13 @@ int mtk_drm_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) out: kfree(sgt); - dma_buf_map_set_vaddr(map, mtk_gem->kvaddr); + iosys_map_set_vaddr(map, mtk_gem->kvaddr); return 0; } -void mtk_drm_gem_prime_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map) +void mtk_drm_gem_prime_vunmap(struct drm_gem_object *obj, + struct iosys_map *map) { struct mtk_drm_gem_obj *mtk_gem = to_mtk_gem_obj(obj); void *vaddr = map->vaddr; diff --git a/drivers/gpu/drm/mediatek/mtk_drm_gem.h b/drivers/gpu/drm/mediatek/mtk_drm_gem.h index 9a359a06cb73..78f23b07a02e 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_gem.h +++ b/drivers/gpu/drm/mediatek/mtk_drm_gem.h @@ -42,7 +42,8 @@ int mtk_drm_gem_dumb_create(struct drm_file *file_priv, struct drm_device *dev, struct sg_table *mtk_gem_prime_get_sg_table(struct drm_gem_object *obj); struct drm_gem_object *mtk_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sg); -int mtk_drm_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map); -void mtk_drm_gem_prime_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map); +int mtk_drm_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map); +void mtk_drm_gem_prime_vunmap(struct drm_gem_object *obj, + struct iosys_map *map); #endif diff --git a/drivers/gpu/drm/meson/meson_drv.c b/drivers/gpu/drm/meson/meson_drv.c index 80f1d439841a..26aeaf0ab86e 100644 --- a/drivers/gpu/drm/meson/meson_drv.c +++ b/drivers/gpu/drm/meson/meson_drv.c @@ -302,42 +302,42 @@ static int meson_drv_bind_master(struct device *dev, bool has_components) if (priv->afbcd.ops) { ret = priv->afbcd.ops->init(priv); if (ret) - return ret; + goto free_drm; } /* Encoder Initialization */ ret = meson_encoder_cvbs_init(priv); if (ret) - goto free_drm; + goto exit_afbcd; if (has_components) { ret = component_bind_all(drm->dev, drm); if (ret) { dev_err(drm->dev, "Couldn't bind all components\n"); - goto free_drm; + goto exit_afbcd; } } ret = meson_encoder_hdmi_init(priv); if (ret) - goto free_drm; + goto exit_afbcd; ret = meson_plane_create(priv); if (ret) - goto free_drm; + goto exit_afbcd; ret = meson_overlay_create(priv); if (ret) - goto free_drm; + goto exit_afbcd; ret = meson_crtc_create(priv); if (ret) - goto free_drm; + goto exit_afbcd; ret = request_irq(priv->vsync_irq, meson_irq, 0, drm->driver->name, drm); if (ret) - goto free_drm; + goto exit_afbcd; drm_mode_config_reset(drm); @@ -355,6 +355,9 @@ static int meson_drv_bind_master(struct device *dev, bool has_components) uninstall_irq: free_irq(priv->vsync_irq, drm); +exit_afbcd: + if (priv->afbcd.ops) + priv->afbcd.ops->exit(priv); free_drm: drm_dev_put(drm); @@ -385,10 +388,8 @@ static void meson_drv_unbind(struct device *dev) free_irq(priv->vsync_irq, drm); drm_dev_put(drm); - if (priv->afbcd.ops) { - priv->afbcd.ops->reset(priv); - meson_rdma_free(priv); - } + if (priv->afbcd.ops) + priv->afbcd.ops->exit(priv); } static const struct component_master_ops meson_drv_master_ops = { diff --git a/drivers/gpu/drm/meson/meson_osd_afbcd.c b/drivers/gpu/drm/meson/meson_osd_afbcd.c index ffc6b584dbf8..0cdbe899402f 100644 --- a/drivers/gpu/drm/meson/meson_osd_afbcd.c +++ b/drivers/gpu/drm/meson/meson_osd_afbcd.c @@ -79,11 +79,6 @@ static bool meson_gxm_afbcd_supported_fmt(u64 modifier, uint32_t format) return meson_gxm_afbcd_pixel_fmt(modifier, format) >= 0; } -static int meson_gxm_afbcd_init(struct meson_drm *priv) -{ - return 0; -} - static int meson_gxm_afbcd_reset(struct meson_drm *priv) { writel_relaxed(VIU_SW_RESET_OSD1_AFBCD, @@ -93,6 +88,16 @@ static int meson_gxm_afbcd_reset(struct meson_drm *priv) return 0; } +static int meson_gxm_afbcd_init(struct meson_drm *priv) +{ + return 0; +} + +static void meson_gxm_afbcd_exit(struct meson_drm *priv) +{ + meson_gxm_afbcd_reset(priv); +} + static int meson_gxm_afbcd_enable(struct meson_drm *priv) { writel_relaxed(FIELD_PREP(OSD1_AFBCD_ID_FIFO_THRD, 0x40) | @@ -172,6 +177,7 @@ static int meson_gxm_afbcd_setup(struct meson_drm *priv) struct meson_afbcd_ops meson_afbcd_gxm_ops = { .init = meson_gxm_afbcd_init, + .exit = meson_gxm_afbcd_exit, .reset = meson_gxm_afbcd_reset, .enable = meson_gxm_afbcd_enable, .disable = meson_gxm_afbcd_disable, @@ -269,6 +275,18 @@ static bool meson_g12a_afbcd_supported_fmt(u64 modifier, uint32_t format) return meson_g12a_afbcd_pixel_fmt(modifier, format) >= 0; } +static int meson_g12a_afbcd_reset(struct meson_drm *priv) +{ + meson_rdma_reset(priv); + + meson_rdma_writel_sync(priv, VIU_SW_RESET_G12A_AFBC_ARB | + VIU_SW_RESET_G12A_OSD1_AFBCD, + VIU_SW_RESET); + meson_rdma_writel_sync(priv, 0, VIU_SW_RESET); + + return 0; +} + static int meson_g12a_afbcd_init(struct meson_drm *priv) { int ret; @@ -286,16 +304,10 @@ static int meson_g12a_afbcd_init(struct meson_drm *priv) return 0; } -static int meson_g12a_afbcd_reset(struct meson_drm *priv) +static void meson_g12a_afbcd_exit(struct meson_drm *priv) { - meson_rdma_reset(priv); - - meson_rdma_writel_sync(priv, VIU_SW_RESET_G12A_AFBC_ARB | - VIU_SW_RESET_G12A_OSD1_AFBCD, - VIU_SW_RESET); - meson_rdma_writel_sync(priv, 0, VIU_SW_RESET); - - return 0; + meson_g12a_afbcd_reset(priv); + meson_rdma_free(priv); } static int meson_g12a_afbcd_enable(struct meson_drm *priv) @@ -380,6 +392,7 @@ static int meson_g12a_afbcd_setup(struct meson_drm *priv) struct meson_afbcd_ops meson_afbcd_g12a_ops = { .init = meson_g12a_afbcd_init, + .exit = meson_g12a_afbcd_exit, .reset = meson_g12a_afbcd_reset, .enable = meson_g12a_afbcd_enable, .disable = meson_g12a_afbcd_disable, diff --git a/drivers/gpu/drm/meson/meson_osd_afbcd.h b/drivers/gpu/drm/meson/meson_osd_afbcd.h index 5e5523304f42..e77ddeb6416f 100644 --- a/drivers/gpu/drm/meson/meson_osd_afbcd.h +++ b/drivers/gpu/drm/meson/meson_osd_afbcd.h @@ -14,6 +14,7 @@ struct meson_afbcd_ops { int (*init)(struct meson_drm *priv); + void (*exit)(struct meson_drm *priv); int (*reset)(struct meson_drm *priv); int (*enable)(struct meson_drm *priv); int (*disable)(struct meson_drm *priv); diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c index b983541a4c53..6e18d3bbd720 100644 --- a/drivers/gpu/drm/mgag200/mgag200_mode.c +++ b/drivers/gpu/drm/mgag200/mgag200_mode.c @@ -9,7 +9,7 @@ */ #include <linux/delay.h> -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_atomic_state_helper.h> @@ -529,7 +529,10 @@ static void mgag200_set_format_regs(struct mga_device *mdev, WREG_GFX(3, 0x00); WREG_GFX(4, 0x00); WREG_GFX(5, 0x40); - WREG_GFX(6, 0x05); + /* GCTL6 should be 0x05, but we configure memmapsl to 0xb8000 (text mode), + * so that it doesn't hang when running kexec/kdump on G200_SE rev42. + */ + WREG_GFX(6, 0x0d); WREG_GFX(7, 0x0f); WREG_GFX(8, 0x0f); @@ -845,7 +848,7 @@ mgag200_simple_display_pipe_mode_valid(struct drm_simple_display_pipe *pipe, static void mgag200_handle_damage(struct mga_device *mdev, struct drm_framebuffer *fb, - struct drm_rect *clip, const struct dma_buf_map *map) + struct drm_rect *clip, const struct iosys_map *map) { void __iomem *dst = mdev->vram; void *vmap = map->vaddr; /* TODO: Use mapping abstraction properly */ diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig index 1eae5a9645f4..4fd931413705 100644 --- a/drivers/gpu/drm/msm/Kconfig +++ b/drivers/gpu/drm/msm/Kconfig @@ -12,6 +12,7 @@ config DRM_MSM select IOMMU_IO_PGTABLE select QCOM_MDT_LOADER if ARCH_QCOM select REGULATOR + select DRM_DP_HELPER select DRM_KMS_HELPER select DRM_PANEL select DRM_BRIDGE diff --git a/drivers/gpu/drm/msm/dp/dp_audio.c b/drivers/gpu/drm/msm/dp/dp_audio.c index d7e4a39a904e..4553f4985434 100644 --- a/drivers/gpu/drm/msm/dp/dp_audio.c +++ b/drivers/gpu/drm/msm/dp/dp_audio.c @@ -8,7 +8,7 @@ #include <linux/of_platform.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include "dp_catalog.h" diff --git a/drivers/gpu/drm/msm/dp/dp_aux.h b/drivers/gpu/drm/msm/dp/dp_aux.h index 0728cc09c9ec..82afc8d5210f 100644 --- a/drivers/gpu/drm/msm/dp/dp_aux.h +++ b/drivers/gpu/drm/msm/dp/dp_aux.h @@ -7,7 +7,7 @@ #define _DP_AUX_H_ #include "dp_catalog.h" -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> int dp_aux_register(struct drm_dp_aux *dp_aux); void dp_aux_unregister(struct drm_dp_aux *dp_aux); diff --git a/drivers/gpu/drm/msm/dp/dp_catalog.c b/drivers/gpu/drm/msm/dp/dp_catalog.c index 6ae9b29044b6..8a6d3ead3440 100644 --- a/drivers/gpu/drm/msm/dp/dp_catalog.c +++ b/drivers/gpu/drm/msm/dp/dp_catalog.c @@ -10,7 +10,7 @@ #include <linux/phy/phy.h> #include <linux/phy/phy-dp.h> #include <linux/rational.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_print.h> #include "dp_catalog.h" diff --git a/drivers/gpu/drm/msm/dp/dp_ctrl.c b/drivers/gpu/drm/msm/dp/dp_ctrl.c index c724cb0bde9d..88ca6c3aedd3 100644 --- a/drivers/gpu/drm/msm/dp/dp_ctrl.c +++ b/drivers/gpu/drm/msm/dp/dp_ctrl.c @@ -12,7 +12,7 @@ #include <linux/phy/phy-dp.h> #include <linux/pm_opp.h> #include <drm/drm_fixed.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_print.h> #include "dp_reg.h" diff --git a/drivers/gpu/drm/msm/edp/edp.h b/drivers/gpu/drm/msm/edp/edp.h new file mode 100644 index 000000000000..1a82d7a4af9f --- /dev/null +++ b/drivers/gpu/drm/msm/edp/edp.h @@ -0,0 +1,77 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2014-2015, The Linux Foundation. All rights reserved. + */ + +#ifndef __EDP_CONNECTOR_H__ +#define __EDP_CONNECTOR_H__ + +#include <linux/i2c.h> +#include <linux/interrupt.h> +#include <linux/kernel.h> +#include <linux/platform_device.h> +#include <drm/dp/drm_dp_helper.h> +#include <drm/drm_bridge.h> +#include <drm/drm_crtc.h> + +#include "msm_drv.h" + +#define edp_read(offset) msm_readl((offset)) +#define edp_write(offset, data) msm_writel((data), (offset)) + +struct edp_ctrl; +struct edp_aux; +struct edp_phy; + +struct msm_edp { + struct drm_device *dev; + struct platform_device *pdev; + + struct drm_connector *connector; + struct drm_bridge *bridge; + + /* the encoder we are hooked to (outside of eDP block) */ + struct drm_encoder *encoder; + + struct edp_ctrl *ctrl; + + int irq; +}; + +/* eDP bridge */ +struct drm_bridge *msm_edp_bridge_init(struct msm_edp *edp); +void edp_bridge_destroy(struct drm_bridge *bridge); + +/* eDP connector */ +struct drm_connector *msm_edp_connector_init(struct msm_edp *edp); + +/* AUX */ +void *msm_edp_aux_init(struct msm_edp *edp, void __iomem *regbase, struct drm_dp_aux **drm_aux); +void msm_edp_aux_destroy(struct device *dev, struct edp_aux *aux); +irqreturn_t msm_edp_aux_irq(struct edp_aux *aux, u32 isr); +void msm_edp_aux_ctrl(struct edp_aux *aux, int enable); + +/* Phy */ +bool msm_edp_phy_ready(struct edp_phy *phy); +void msm_edp_phy_ctrl(struct edp_phy *phy, int enable); +void msm_edp_phy_vm_pe_init(struct edp_phy *phy); +void msm_edp_phy_vm_pe_cfg(struct edp_phy *phy, u32 v0, u32 v1); +void msm_edp_phy_lane_power_ctrl(struct edp_phy *phy, bool up, u32 max_lane); +void *msm_edp_phy_init(struct device *dev, void __iomem *regbase); + +/* Ctrl */ +irqreturn_t msm_edp_ctrl_irq(struct edp_ctrl *ctrl); +void msm_edp_ctrl_power(struct edp_ctrl *ctrl, bool on); +int msm_edp_ctrl_init(struct msm_edp *edp); +void msm_edp_ctrl_destroy(struct edp_ctrl *ctrl); +bool msm_edp_ctrl_panel_connected(struct edp_ctrl *ctrl); +int msm_edp_ctrl_get_panel_info(struct edp_ctrl *ctrl, + struct drm_connector *connector, struct edid **edid); +int msm_edp_ctrl_timing_cfg(struct edp_ctrl *ctrl, + const struct drm_display_mode *mode, + const struct drm_display_info *info); +/* @pixel_rate is in kHz */ +bool msm_edp_ctrl_pixel_clock_valid(struct edp_ctrl *ctrl, + u32 pixel_rate, u32 *pm, u32 *pn); + +#endif /* __EDP_CONNECTOR_H__ */ diff --git a/drivers/gpu/drm/msm/edp/edp_ctrl.c b/drivers/gpu/drm/msm/edp/edp_ctrl.c new file mode 100644 index 000000000000..9f537b1fd849 --- /dev/null +++ b/drivers/gpu/drm/msm/edp/edp_ctrl.c @@ -0,0 +1,1373 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2014-2015, The Linux Foundation. All rights reserved. + */ + +#include <linux/clk.h> +#include <linux/gpio/consumer.h> +#include <linux/regulator/consumer.h> +#include <drm/dp/drm_dp_helper.h> +#include <drm/drm_crtc.h> +#include <drm/drm_edid.h> + +#include "edp.h" +#include "edp.xml.h" + +#define VDDA_UA_ON_LOAD 100000 /* uA units */ +#define VDDA_UA_OFF_LOAD 100 /* uA units */ + +#define DPCD_LINK_VOLTAGE_MAX 4 +#define DPCD_LINK_PRE_EMPHASIS_MAX 4 + +#define EDP_LINK_BW_MAX DP_LINK_BW_2_7 + +/* Link training return value */ +#define EDP_TRAIN_FAIL -1 +#define EDP_TRAIN_SUCCESS 0 +#define EDP_TRAIN_RECONFIG 1 + +#define EDP_CLK_MASK_AHB BIT(0) +#define EDP_CLK_MASK_AUX BIT(1) +#define EDP_CLK_MASK_LINK BIT(2) +#define EDP_CLK_MASK_PIXEL BIT(3) +#define EDP_CLK_MASK_MDP_CORE BIT(4) +#define EDP_CLK_MASK_LINK_CHAN (EDP_CLK_MASK_LINK | EDP_CLK_MASK_PIXEL) +#define EDP_CLK_MASK_AUX_CHAN \ + (EDP_CLK_MASK_AHB | EDP_CLK_MASK_AUX | EDP_CLK_MASK_MDP_CORE) +#define EDP_CLK_MASK_ALL (EDP_CLK_MASK_AUX_CHAN | EDP_CLK_MASK_LINK_CHAN) + +#define EDP_BACKLIGHT_MAX 255 + +#define EDP_INTR_STATUS1 \ + (EDP_INTERRUPT_REG_1_HPD | EDP_INTERRUPT_REG_1_AUX_I2C_DONE | \ + EDP_INTERRUPT_REG_1_WRONG_ADDR | EDP_INTERRUPT_REG_1_TIMEOUT | \ + EDP_INTERRUPT_REG_1_NACK_DEFER | EDP_INTERRUPT_REG_1_WRONG_DATA_CNT | \ + EDP_INTERRUPT_REG_1_I2C_NACK | EDP_INTERRUPT_REG_1_I2C_DEFER | \ + EDP_INTERRUPT_REG_1_PLL_UNLOCK | EDP_INTERRUPT_REG_1_AUX_ERROR) +#define EDP_INTR_MASK1 (EDP_INTR_STATUS1 << 2) +#define EDP_INTR_STATUS2 \ + (EDP_INTERRUPT_REG_2_READY_FOR_VIDEO | \ + EDP_INTERRUPT_REG_2_IDLE_PATTERNs_SENT | \ + EDP_INTERRUPT_REG_2_FRAME_END | EDP_INTERRUPT_REG_2_CRC_UPDATED) +#define EDP_INTR_MASK2 (EDP_INTR_STATUS2 << 2) + +struct edp_ctrl { + struct platform_device *pdev; + + void __iomem *base; + + /* regulators */ + struct regulator *vdda_vreg; /* 1.8 V */ + struct regulator *lvl_vreg; + + /* clocks */ + struct clk *aux_clk; + struct clk *pixel_clk; + struct clk *ahb_clk; + struct clk *link_clk; + struct clk *mdp_core_clk; + + /* gpios */ + struct gpio_desc *panel_en_gpio; + struct gpio_desc *panel_hpd_gpio; + + /* completion and mutex */ + struct completion idle_comp; + struct mutex dev_mutex; /* To protect device power status */ + + /* work queue */ + struct work_struct on_work; + struct work_struct off_work; + struct workqueue_struct *workqueue; + + /* Interrupt register lock */ + spinlock_t irq_lock; + + bool edp_connected; + bool power_on; + + /* edid raw data */ + struct edid *edid; + + struct drm_dp_aux *drm_aux; + + /* dpcd raw data */ + u8 dpcd[DP_RECEIVER_CAP_SIZE]; + + /* Link status */ + u8 link_rate; + u8 lane_cnt; + u8 v_level; + u8 p_level; + + /* Timing status */ + u8 interlaced; + u32 pixel_rate; /* in kHz */ + u32 color_depth; + + struct edp_aux *aux; + struct edp_phy *phy; +}; + +struct edp_pixel_clk_div { + u32 rate; /* in kHz */ + u32 m; + u32 n; +}; + +#define EDP_PIXEL_CLK_NUM 8 +static const struct edp_pixel_clk_div clk_divs[2][EDP_PIXEL_CLK_NUM] = { + { /* Link clock = 162MHz, source clock = 810MHz */ + {119000, 31, 211}, /* WSXGA+ 1680x1050@60Hz CVT */ + {130250, 32, 199}, /* UXGA 1600x1200@60Hz CVT */ + {148500, 11, 60}, /* FHD 1920x1080@60Hz */ + {154000, 50, 263}, /* WUXGA 1920x1200@60Hz CVT */ + {209250, 31, 120}, /* QXGA 2048x1536@60Hz CVT */ + {268500, 119, 359}, /* WQXGA 2560x1600@60Hz CVT */ + {138530, 33, 193}, /* AUO B116HAN03.0 Panel */ + {141400, 48, 275}, /* AUO B133HTN01.2 Panel */ + }, + { /* Link clock = 270MHz, source clock = 675MHz */ + {119000, 52, 295}, /* WSXGA+ 1680x1050@60Hz CVT */ + {130250, 11, 57}, /* UXGA 1600x1200@60Hz CVT */ + {148500, 11, 50}, /* FHD 1920x1080@60Hz */ + {154000, 47, 206}, /* WUXGA 1920x1200@60Hz CVT */ + {209250, 31, 100}, /* QXGA 2048x1536@60Hz CVT */ + {268500, 107, 269}, /* WQXGA 2560x1600@60Hz CVT */ + {138530, 63, 307}, /* AUO B116HAN03.0 Panel */ + {141400, 53, 253}, /* AUO B133HTN01.2 Panel */ + }, +}; + +static int edp_clk_init(struct edp_ctrl *ctrl) +{ + struct platform_device *pdev = ctrl->pdev; + int ret; + + ctrl->aux_clk = msm_clk_get(pdev, "core"); + if (IS_ERR(ctrl->aux_clk)) { + ret = PTR_ERR(ctrl->aux_clk); + pr_err("%s: Can't find core clock, %d\n", __func__, ret); + ctrl->aux_clk = NULL; + return ret; + } + + ctrl->pixel_clk = msm_clk_get(pdev, "pixel"); + if (IS_ERR(ctrl->pixel_clk)) { + ret = PTR_ERR(ctrl->pixel_clk); + pr_err("%s: Can't find pixel clock, %d\n", __func__, ret); + ctrl->pixel_clk = NULL; + return ret; + } + + ctrl->ahb_clk = msm_clk_get(pdev, "iface"); + if (IS_ERR(ctrl->ahb_clk)) { + ret = PTR_ERR(ctrl->ahb_clk); + pr_err("%s: Can't find iface clock, %d\n", __func__, ret); + ctrl->ahb_clk = NULL; + return ret; + } + + ctrl->link_clk = msm_clk_get(pdev, "link"); + if (IS_ERR(ctrl->link_clk)) { + ret = PTR_ERR(ctrl->link_clk); + pr_err("%s: Can't find link clock, %d\n", __func__, ret); + ctrl->link_clk = NULL; + return ret; + } + + /* need mdp core clock to receive irq */ + ctrl->mdp_core_clk = msm_clk_get(pdev, "mdp_core"); + if (IS_ERR(ctrl->mdp_core_clk)) { + ret = PTR_ERR(ctrl->mdp_core_clk); + pr_err("%s: Can't find mdp_core clock, %d\n", __func__, ret); + ctrl->mdp_core_clk = NULL; + return ret; + } + + return 0; +} + +static int edp_clk_enable(struct edp_ctrl *ctrl, u32 clk_mask) +{ + int ret; + + DBG("mask=%x", clk_mask); + /* ahb_clk should be enabled first */ + if (clk_mask & EDP_CLK_MASK_AHB) { + ret = clk_prepare_enable(ctrl->ahb_clk); + if (ret) { + pr_err("%s: Failed to enable ahb clk\n", __func__); + goto f0; + } + } + if (clk_mask & EDP_CLK_MASK_AUX) { + ret = clk_set_rate(ctrl->aux_clk, 19200000); + if (ret) { + pr_err("%s: Failed to set rate aux clk\n", __func__); + goto f1; + } + ret = clk_prepare_enable(ctrl->aux_clk); + if (ret) { + pr_err("%s: Failed to enable aux clk\n", __func__); + goto f1; + } + } + /* Need to set rate and enable link_clk prior to pixel_clk */ + if (clk_mask & EDP_CLK_MASK_LINK) { + DBG("edp->link_clk, set_rate %ld", + (unsigned long)ctrl->link_rate * 27000000); + ret = clk_set_rate(ctrl->link_clk, + (unsigned long)ctrl->link_rate * 27000000); + if (ret) { + pr_err("%s: Failed to set rate to link clk\n", + __func__); + goto f2; + } + + ret = clk_prepare_enable(ctrl->link_clk); + if (ret) { + pr_err("%s: Failed to enable link clk\n", __func__); + goto f2; + } + } + if (clk_mask & EDP_CLK_MASK_PIXEL) { + DBG("edp->pixel_clk, set_rate %ld", + (unsigned long)ctrl->pixel_rate * 1000); + ret = clk_set_rate(ctrl->pixel_clk, + (unsigned long)ctrl->pixel_rate * 1000); + if (ret) { + pr_err("%s: Failed to set rate to pixel clk\n", + __func__); + goto f3; + } + + ret = clk_prepare_enable(ctrl->pixel_clk); + if (ret) { + pr_err("%s: Failed to enable pixel clk\n", __func__); + goto f3; + } + } + if (clk_mask & EDP_CLK_MASK_MDP_CORE) { + ret = clk_prepare_enable(ctrl->mdp_core_clk); + if (ret) { + pr_err("%s: Failed to enable mdp core clk\n", __func__); + goto f4; + } + } + + return 0; + +f4: + if (clk_mask & EDP_CLK_MASK_PIXEL) + clk_disable_unprepare(ctrl->pixel_clk); +f3: + if (clk_mask & EDP_CLK_MASK_LINK) + clk_disable_unprepare(ctrl->link_clk); +f2: + if (clk_mask & EDP_CLK_MASK_AUX) + clk_disable_unprepare(ctrl->aux_clk); +f1: + if (clk_mask & EDP_CLK_MASK_AHB) + clk_disable_unprepare(ctrl->ahb_clk); +f0: + return ret; +} + +static void edp_clk_disable(struct edp_ctrl *ctrl, u32 clk_mask) +{ + if (clk_mask & EDP_CLK_MASK_MDP_CORE) + clk_disable_unprepare(ctrl->mdp_core_clk); + if (clk_mask & EDP_CLK_MASK_PIXEL) + clk_disable_unprepare(ctrl->pixel_clk); + if (clk_mask & EDP_CLK_MASK_LINK) + clk_disable_unprepare(ctrl->link_clk); + if (clk_mask & EDP_CLK_MASK_AUX) + clk_disable_unprepare(ctrl->aux_clk); + if (clk_mask & EDP_CLK_MASK_AHB) + clk_disable_unprepare(ctrl->ahb_clk); +} + +static int edp_regulator_init(struct edp_ctrl *ctrl) +{ + struct device *dev = &ctrl->pdev->dev; + int ret; + + DBG(""); + ctrl->vdda_vreg = devm_regulator_get(dev, "vdda"); + ret = PTR_ERR_OR_ZERO(ctrl->vdda_vreg); + if (ret) { + pr_err("%s: Could not get vdda reg, ret = %d\n", __func__, + ret); + ctrl->vdda_vreg = NULL; + return ret; + } + ctrl->lvl_vreg = devm_regulator_get(dev, "lvl-vdd"); + ret = PTR_ERR_OR_ZERO(ctrl->lvl_vreg); + if (ret) { + pr_err("%s: Could not get lvl-vdd reg, ret = %d\n", __func__, + ret); + ctrl->lvl_vreg = NULL; + return ret; + } + + return 0; +} + +static int edp_regulator_enable(struct edp_ctrl *ctrl) +{ + int ret; + + ret = regulator_set_load(ctrl->vdda_vreg, VDDA_UA_ON_LOAD); + if (ret < 0) { + pr_err("%s: vdda_vreg set regulator mode failed.\n", __func__); + goto vdda_set_fail; + } + + ret = regulator_enable(ctrl->vdda_vreg); + if (ret) { + pr_err("%s: Failed to enable vdda_vreg regulator.\n", __func__); + goto vdda_enable_fail; + } + + ret = regulator_enable(ctrl->lvl_vreg); + if (ret) { + pr_err("Failed to enable lvl-vdd reg regulator, %d", ret); + goto lvl_enable_fail; + } + + DBG("exit"); + return 0; + +lvl_enable_fail: + regulator_disable(ctrl->vdda_vreg); +vdda_enable_fail: + regulator_set_load(ctrl->vdda_vreg, VDDA_UA_OFF_LOAD); +vdda_set_fail: + return ret; +} + +static void edp_regulator_disable(struct edp_ctrl *ctrl) +{ + regulator_disable(ctrl->lvl_vreg); + regulator_disable(ctrl->vdda_vreg); + regulator_set_load(ctrl->vdda_vreg, VDDA_UA_OFF_LOAD); +} + +static int edp_gpio_config(struct edp_ctrl *ctrl) +{ + struct device *dev = &ctrl->pdev->dev; + int ret; + + ctrl->panel_hpd_gpio = devm_gpiod_get(dev, "panel-hpd", GPIOD_IN); + if (IS_ERR(ctrl->panel_hpd_gpio)) { + ret = PTR_ERR(ctrl->panel_hpd_gpio); + ctrl->panel_hpd_gpio = NULL; + pr_err("%s: cannot get panel-hpd-gpios, %d\n", __func__, ret); + return ret; + } + + ctrl->panel_en_gpio = devm_gpiod_get(dev, "panel-en", GPIOD_OUT_LOW); + if (IS_ERR(ctrl->panel_en_gpio)) { + ret = PTR_ERR(ctrl->panel_en_gpio); + ctrl->panel_en_gpio = NULL; + pr_err("%s: cannot get panel-en-gpios, %d\n", __func__, ret); + return ret; + } + + DBG("gpio on"); + + return 0; +} + +static void edp_ctrl_irq_enable(struct edp_ctrl *ctrl, int enable) +{ + unsigned long flags; + + DBG("%d", enable); + spin_lock_irqsave(&ctrl->irq_lock, flags); + if (enable) { + edp_write(ctrl->base + REG_EDP_INTERRUPT_REG_1, EDP_INTR_MASK1); + edp_write(ctrl->base + REG_EDP_INTERRUPT_REG_2, EDP_INTR_MASK2); + } else { + edp_write(ctrl->base + REG_EDP_INTERRUPT_REG_1, 0x0); + edp_write(ctrl->base + REG_EDP_INTERRUPT_REG_2, 0x0); + } + spin_unlock_irqrestore(&ctrl->irq_lock, flags); + DBG("exit"); +} + +static void edp_fill_link_cfg(struct edp_ctrl *ctrl) +{ + u32 prate; + u32 lrate; + u32 bpp; + u8 max_lane = drm_dp_max_lane_count(ctrl->dpcd); + u8 lane; + + prate = ctrl->pixel_rate; + bpp = ctrl->color_depth * 3; + + /* + * By default, use the maximum link rate and minimum lane count, + * so that we can do rate down shift during link training. + */ + ctrl->link_rate = ctrl->dpcd[DP_MAX_LINK_RATE]; + + prate *= bpp; + prate /= 8; /* in kByte */ + + lrate = 270000; /* in kHz */ + lrate *= ctrl->link_rate; + lrate /= 10; /* in kByte, 10 bits --> 8 bits */ + + for (lane = 1; lane <= max_lane; lane <<= 1) { + if (lrate >= prate) + break; + lrate <<= 1; + } + + ctrl->lane_cnt = lane; + DBG("rate=%d lane=%d", ctrl->link_rate, ctrl->lane_cnt); +} + +static void edp_config_ctrl(struct edp_ctrl *ctrl) +{ + u32 data; + enum edp_color_depth depth; + + data = EDP_CONFIGURATION_CTRL_LANES(ctrl->lane_cnt - 1); + + if (drm_dp_enhanced_frame_cap(ctrl->dpcd)) + data |= EDP_CONFIGURATION_CTRL_ENHANCED_FRAMING; + + depth = EDP_6BIT; + if (ctrl->color_depth == 8) + depth = EDP_8BIT; + + data |= EDP_CONFIGURATION_CTRL_COLOR(depth); + + if (!ctrl->interlaced) /* progressive */ + data |= EDP_CONFIGURATION_CTRL_PROGRESSIVE; + + data |= (EDP_CONFIGURATION_CTRL_SYNC_CLK | + EDP_CONFIGURATION_CTRL_STATIC_MVID); + + edp_write(ctrl->base + REG_EDP_CONFIGURATION_CTRL, data); +} + +static void edp_state_ctrl(struct edp_ctrl *ctrl, u32 state) +{ + edp_write(ctrl->base + REG_EDP_STATE_CTRL, state); + /* Make sure H/W status is set */ + wmb(); +} + +static int edp_lane_set_write(struct edp_ctrl *ctrl, + u8 voltage_level, u8 pre_emphasis_level) +{ + int i; + u8 buf[4]; + + if (voltage_level >= DPCD_LINK_VOLTAGE_MAX) + voltage_level |= 0x04; + + if (pre_emphasis_level >= DPCD_LINK_PRE_EMPHASIS_MAX) + pre_emphasis_level |= 0x04; + + pre_emphasis_level <<= 3; + + for (i = 0; i < 4; i++) + buf[i] = voltage_level | pre_emphasis_level; + + DBG("%s: p|v=0x%x", __func__, voltage_level | pre_emphasis_level); + if (drm_dp_dpcd_write(ctrl->drm_aux, 0x103, buf, 4) < 4) { + pr_err("%s: Set sw/pe to panel failed\n", __func__); + return -ENOLINK; + } + + return 0; +} + +static int edp_train_pattern_set_write(struct edp_ctrl *ctrl, u8 pattern) +{ + u8 p = pattern; + + DBG("pattern=%x", p); + if (drm_dp_dpcd_write(ctrl->drm_aux, + DP_TRAINING_PATTERN_SET, &p, 1) < 1) { + pr_err("%s: Set training pattern to panel failed\n", __func__); + return -ENOLINK; + } + + return 0; +} + +static void edp_sink_train_set_adjust(struct edp_ctrl *ctrl, + const u8 *link_status) +{ + int i; + u8 max = 0; + u8 data; + + /* use the max level across lanes */ + for (i = 0; i < ctrl->lane_cnt; i++) { + data = drm_dp_get_adjust_request_voltage(link_status, i); + DBG("lane=%d req_voltage_swing=0x%x", i, data); + if (max < data) + max = data; + } + + ctrl->v_level = max >> DP_TRAIN_VOLTAGE_SWING_SHIFT; + + /* use the max level across lanes */ + max = 0; + for (i = 0; i < ctrl->lane_cnt; i++) { + data = drm_dp_get_adjust_request_pre_emphasis(link_status, i); + DBG("lane=%d req_pre_emphasis=0x%x", i, data); + if (max < data) + max = data; + } + + ctrl->p_level = max >> DP_TRAIN_PRE_EMPHASIS_SHIFT; + DBG("v_level=%d, p_level=%d", ctrl->v_level, ctrl->p_level); +} + +static void edp_host_train_set(struct edp_ctrl *ctrl, u32 train) +{ + int cnt = 10; + u32 data; + u32 shift = train - 1; + + DBG("train=%d", train); + + edp_state_ctrl(ctrl, EDP_STATE_CTRL_TRAIN_PATTERN_1 << shift); + while (--cnt) { + data = edp_read(ctrl->base + REG_EDP_MAINLINK_READY); + if (data & (EDP_MAINLINK_READY_TRAIN_PATTERN_1_READY << shift)) + break; + } + + if (cnt == 0) + pr_err("%s: set link_train=%d failed\n", __func__, train); +} + +static const u8 vm_pre_emphasis[4][4] = { + {0x03, 0x06, 0x09, 0x0C}, /* pe0, 0 db */ + {0x03, 0x06, 0x09, 0xFF}, /* pe1, 3.5 db */ + {0x03, 0x06, 0xFF, 0xFF}, /* pe2, 6.0 db */ + {0x03, 0xFF, 0xFF, 0xFF} /* pe3, 9.5 db */ +}; + +/* voltage swing, 0.2v and 1.0v are not support */ +static const u8 vm_voltage_swing[4][4] = { + {0x14, 0x18, 0x1A, 0x1E}, /* sw0, 0.4v */ + {0x18, 0x1A, 0x1E, 0xFF}, /* sw1, 0.6 v */ + {0x1A, 0x1E, 0xFF, 0xFF}, /* sw1, 0.8 v */ + {0x1E, 0xFF, 0xFF, 0xFF} /* sw1, 1.2 v, optional */ +}; + +static int edp_voltage_pre_emphasise_set(struct edp_ctrl *ctrl) +{ + u32 value0; + u32 value1; + + DBG("v=%d p=%d", ctrl->v_level, ctrl->p_level); + + value0 = vm_pre_emphasis[(int)(ctrl->v_level)][(int)(ctrl->p_level)]; + value1 = vm_voltage_swing[(int)(ctrl->v_level)][(int)(ctrl->p_level)]; + + /* Configure host and panel only if both values are allowed */ + if (value0 != 0xFF && value1 != 0xFF) { + msm_edp_phy_vm_pe_cfg(ctrl->phy, value0, value1); + return edp_lane_set_write(ctrl, ctrl->v_level, ctrl->p_level); + } + + return -EINVAL; +} + +static int edp_start_link_train_1(struct edp_ctrl *ctrl) +{ + u8 link_status[DP_LINK_STATUS_SIZE]; + u8 old_v_level; + int tries; + int ret; + int rlen; + + DBG(""); + + edp_host_train_set(ctrl, DP_TRAINING_PATTERN_1); + ret = edp_voltage_pre_emphasise_set(ctrl); + if (ret) + return ret; + ret = edp_train_pattern_set_write(ctrl, + DP_TRAINING_PATTERN_1 | DP_RECOVERED_CLOCK_OUT_EN); + if (ret) + return ret; + + tries = 0; + old_v_level = ctrl->v_level; + while (1) { + drm_dp_link_train_clock_recovery_delay(ctrl->drm_aux, ctrl->dpcd); + + rlen = drm_dp_dpcd_read_link_status(ctrl->drm_aux, link_status); + if (rlen < DP_LINK_STATUS_SIZE) { + pr_err("%s: read link status failed\n", __func__); + return -ENOLINK; + } + if (drm_dp_clock_recovery_ok(link_status, ctrl->lane_cnt)) { + ret = 0; + break; + } + + if (ctrl->v_level == DPCD_LINK_VOLTAGE_MAX) { + ret = -1; + break; + } + + if (old_v_level == ctrl->v_level) { + tries++; + if (tries >= 5) { + ret = -1; + break; + } + } else { + tries = 0; + old_v_level = ctrl->v_level; + } + + edp_sink_train_set_adjust(ctrl, link_status); + ret = edp_voltage_pre_emphasise_set(ctrl); + if (ret) + return ret; + } + + return ret; +} + +static int edp_start_link_train_2(struct edp_ctrl *ctrl) +{ + u8 link_status[DP_LINK_STATUS_SIZE]; + int tries = 0; + int ret; + int rlen; + + DBG(""); + + edp_host_train_set(ctrl, DP_TRAINING_PATTERN_2); + ret = edp_voltage_pre_emphasise_set(ctrl); + if (ret) + return ret; + + ret = edp_train_pattern_set_write(ctrl, + DP_TRAINING_PATTERN_2 | DP_RECOVERED_CLOCK_OUT_EN); + if (ret) + return ret; + + while (1) { + drm_dp_link_train_channel_eq_delay(ctrl->drm_aux, ctrl->dpcd); + + rlen = drm_dp_dpcd_read_link_status(ctrl->drm_aux, link_status); + if (rlen < DP_LINK_STATUS_SIZE) { + pr_err("%s: read link status failed\n", __func__); + return -ENOLINK; + } + if (drm_dp_channel_eq_ok(link_status, ctrl->lane_cnt)) { + ret = 0; + break; + } + + tries++; + if (tries > 10) { + ret = -1; + break; + } + + edp_sink_train_set_adjust(ctrl, link_status); + ret = edp_voltage_pre_emphasise_set(ctrl); + if (ret) + return ret; + } + + return ret; +} + +static int edp_link_rate_down_shift(struct edp_ctrl *ctrl) +{ + u32 prate, lrate, bpp; + u8 rate, lane, max_lane; + int changed = 0; + + rate = ctrl->link_rate; + lane = ctrl->lane_cnt; + max_lane = drm_dp_max_lane_count(ctrl->dpcd); + + bpp = ctrl->color_depth * 3; + prate = ctrl->pixel_rate; + prate *= bpp; + prate /= 8; /* in kByte */ + + if (rate > DP_LINK_BW_1_62 && rate <= EDP_LINK_BW_MAX) { + rate -= 4; /* reduce rate */ + changed++; + } + + if (changed) { + if (lane >= 1 && lane < max_lane) + lane <<= 1; /* increase lane */ + + lrate = 270000; /* in kHz */ + lrate *= rate; + lrate /= 10; /* kByte, 10 bits --> 8 bits */ + lrate *= lane; + + DBG("new lrate=%u prate=%u(kHz) rate=%d lane=%d p=%u b=%d", + lrate, prate, rate, lane, + ctrl->pixel_rate, + bpp); + + if (lrate > prate) { + ctrl->link_rate = rate; + ctrl->lane_cnt = lane; + DBG("new rate=%d %d", rate, lane); + return 0; + } + } + + return -EINVAL; +} + +static int edp_clear_training_pattern(struct edp_ctrl *ctrl) +{ + int ret; + + ret = edp_train_pattern_set_write(ctrl, 0); + + drm_dp_link_train_channel_eq_delay(ctrl->drm_aux, ctrl->dpcd); + + return ret; +} + +static int edp_do_link_train(struct edp_ctrl *ctrl) +{ + u8 values[2]; + int ret; + + DBG(""); + /* + * Set the current link rate and lane cnt to panel. They may have been + * adjusted and the values are different from them in DPCD CAP + */ + values[0] = ctrl->lane_cnt; + values[1] = ctrl->link_rate; + + if (drm_dp_enhanced_frame_cap(ctrl->dpcd)) + values[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN; + + if (drm_dp_dpcd_write(ctrl->drm_aux, DP_LINK_BW_SET, values, + sizeof(values)) < 0) + return EDP_TRAIN_FAIL; + + ctrl->v_level = 0; /* start from default level */ + ctrl->p_level = 0; + + edp_state_ctrl(ctrl, 0); + if (edp_clear_training_pattern(ctrl)) + return EDP_TRAIN_FAIL; + + ret = edp_start_link_train_1(ctrl); + if (ret < 0) { + if (edp_link_rate_down_shift(ctrl) == 0) { + DBG("link reconfig"); + ret = EDP_TRAIN_RECONFIG; + goto clear; + } else { + pr_err("%s: Training 1 failed", __func__); + ret = EDP_TRAIN_FAIL; + goto clear; + } + } + DBG("Training 1 completed successfully"); + + edp_state_ctrl(ctrl, 0); + if (edp_clear_training_pattern(ctrl)) + return EDP_TRAIN_FAIL; + + ret = edp_start_link_train_2(ctrl); + if (ret < 0) { + if (edp_link_rate_down_shift(ctrl) == 0) { + DBG("link reconfig"); + ret = EDP_TRAIN_RECONFIG; + goto clear; + } else { + pr_err("%s: Training 2 failed", __func__); + ret = EDP_TRAIN_FAIL; + goto clear; + } + } + DBG("Training 2 completed successfully"); + + edp_state_ctrl(ctrl, EDP_STATE_CTRL_SEND_VIDEO); +clear: + edp_clear_training_pattern(ctrl); + + return ret; +} + +static void edp_clock_synchrous(struct edp_ctrl *ctrl, int sync) +{ + u32 data; + enum edp_color_depth depth; + + data = edp_read(ctrl->base + REG_EDP_MISC1_MISC0); + + if (sync) + data |= EDP_MISC1_MISC0_SYNC; + else + data &= ~EDP_MISC1_MISC0_SYNC; + + /* only legacy rgb mode supported */ + depth = EDP_6BIT; /* Default */ + if (ctrl->color_depth == 8) + depth = EDP_8BIT; + else if (ctrl->color_depth == 10) + depth = EDP_10BIT; + else if (ctrl->color_depth == 12) + depth = EDP_12BIT; + else if (ctrl->color_depth == 16) + depth = EDP_16BIT; + + data |= EDP_MISC1_MISC0_COLOR(depth); + + edp_write(ctrl->base + REG_EDP_MISC1_MISC0, data); +} + +static int edp_sw_mvid_nvid(struct edp_ctrl *ctrl, u32 m, u32 n) +{ + u32 n_multi, m_multi = 5; + + if (ctrl->link_rate == DP_LINK_BW_1_62) { + n_multi = 1; + } else if (ctrl->link_rate == DP_LINK_BW_2_7) { + n_multi = 2; + } else { + pr_err("%s: Invalid link rate, %d\n", __func__, + ctrl->link_rate); + return -EINVAL; + } + + edp_write(ctrl->base + REG_EDP_SOFTWARE_MVID, m * m_multi); + edp_write(ctrl->base + REG_EDP_SOFTWARE_NVID, n * n_multi); + + return 0; +} + +static void edp_mainlink_ctrl(struct edp_ctrl *ctrl, int enable) +{ + u32 data = 0; + + edp_write(ctrl->base + REG_EDP_MAINLINK_CTRL, EDP_MAINLINK_CTRL_RESET); + /* Make sure fully reset */ + wmb(); + usleep_range(500, 1000); + + if (enable) + data |= EDP_MAINLINK_CTRL_ENABLE; + + edp_write(ctrl->base + REG_EDP_MAINLINK_CTRL, data); +} + +static void edp_ctrl_phy_aux_enable(struct edp_ctrl *ctrl, int enable) +{ + if (enable) { + edp_regulator_enable(ctrl); + edp_clk_enable(ctrl, EDP_CLK_MASK_AUX_CHAN); + msm_edp_phy_ctrl(ctrl->phy, 1); + msm_edp_aux_ctrl(ctrl->aux, 1); + gpiod_set_value(ctrl->panel_en_gpio, 1); + } else { + gpiod_set_value(ctrl->panel_en_gpio, 0); + msm_edp_aux_ctrl(ctrl->aux, 0); + msm_edp_phy_ctrl(ctrl->phy, 0); + edp_clk_disable(ctrl, EDP_CLK_MASK_AUX_CHAN); + edp_regulator_disable(ctrl); + } +} + +static void edp_ctrl_link_enable(struct edp_ctrl *ctrl, int enable) +{ + u32 m, n; + + if (enable) { + /* Enable link channel clocks */ + edp_clk_enable(ctrl, EDP_CLK_MASK_LINK_CHAN); + + msm_edp_phy_lane_power_ctrl(ctrl->phy, true, ctrl->lane_cnt); + + msm_edp_phy_vm_pe_init(ctrl->phy); + + /* Make sure phy is programed */ + wmb(); + msm_edp_phy_ready(ctrl->phy); + + edp_config_ctrl(ctrl); + msm_edp_ctrl_pixel_clock_valid(ctrl, ctrl->pixel_rate, &m, &n); + edp_sw_mvid_nvid(ctrl, m, n); + edp_mainlink_ctrl(ctrl, 1); + } else { + edp_mainlink_ctrl(ctrl, 0); + + msm_edp_phy_lane_power_ctrl(ctrl->phy, false, 0); + edp_clk_disable(ctrl, EDP_CLK_MASK_LINK_CHAN); + } +} + +static int edp_ctrl_training(struct edp_ctrl *ctrl) +{ + int ret; + + /* Do link training only when power is on */ + if (!ctrl->power_on) + return -EINVAL; + +train_start: + ret = edp_do_link_train(ctrl); + if (ret == EDP_TRAIN_RECONFIG) { + /* Re-configure main link */ + edp_ctrl_irq_enable(ctrl, 0); + edp_ctrl_link_enable(ctrl, 0); + msm_edp_phy_ctrl(ctrl->phy, 0); + + /* Make sure link is fully disabled */ + wmb(); + usleep_range(500, 1000); + + msm_edp_phy_ctrl(ctrl->phy, 1); + edp_ctrl_link_enable(ctrl, 1); + edp_ctrl_irq_enable(ctrl, 1); + goto train_start; + } + + return ret; +} + +static void edp_ctrl_on_worker(struct work_struct *work) +{ + struct edp_ctrl *ctrl = container_of( + work, struct edp_ctrl, on_work); + u8 value; + int ret; + + mutex_lock(&ctrl->dev_mutex); + + if (ctrl->power_on) { + DBG("already on"); + goto unlock_ret; + } + + edp_ctrl_phy_aux_enable(ctrl, 1); + edp_ctrl_link_enable(ctrl, 1); + + edp_ctrl_irq_enable(ctrl, 1); + + /* DP_SET_POWER register is only available on DPCD v1.1 and later */ + if (ctrl->dpcd[DP_DPCD_REV] >= 0x11) { + ret = drm_dp_dpcd_readb(ctrl->drm_aux, DP_SET_POWER, &value); + if (ret < 0) + goto fail; + + value &= ~DP_SET_POWER_MASK; + value |= DP_SET_POWER_D0; + + ret = drm_dp_dpcd_writeb(ctrl->drm_aux, DP_SET_POWER, value); + if (ret < 0) + goto fail; + + /* + * According to the DP 1.1 specification, a "Sink Device must + * exit the power saving state within 1 ms" (Section 2.5.3.1, + * Table 5-52, "Sink Control Field" (register 0x600). + */ + usleep_range(1000, 2000); + } + + ctrl->power_on = true; + + /* Start link training */ + ret = edp_ctrl_training(ctrl); + if (ret != EDP_TRAIN_SUCCESS) + goto fail; + + DBG("DONE"); + goto unlock_ret; + +fail: + edp_ctrl_irq_enable(ctrl, 0); + edp_ctrl_link_enable(ctrl, 0); + edp_ctrl_phy_aux_enable(ctrl, 0); + ctrl->power_on = false; +unlock_ret: + mutex_unlock(&ctrl->dev_mutex); +} + +static void edp_ctrl_off_worker(struct work_struct *work) +{ + struct edp_ctrl *ctrl = container_of( + work, struct edp_ctrl, off_work); + unsigned long time_left; + + mutex_lock(&ctrl->dev_mutex); + + if (!ctrl->power_on) { + DBG("already off"); + goto unlock_ret; + } + + reinit_completion(&ctrl->idle_comp); + edp_state_ctrl(ctrl, EDP_STATE_CTRL_PUSH_IDLE); + + time_left = wait_for_completion_timeout(&ctrl->idle_comp, + msecs_to_jiffies(500)); + if (!time_left) + DBG("%s: idle pattern timedout\n", __func__); + + edp_state_ctrl(ctrl, 0); + + /* DP_SET_POWER register is only available on DPCD v1.1 and later */ + if (ctrl->dpcd[DP_DPCD_REV] >= 0x11) { + u8 value; + int ret; + + ret = drm_dp_dpcd_readb(ctrl->drm_aux, DP_SET_POWER, &value); + if (ret > 0) { + value &= ~DP_SET_POWER_MASK; + value |= DP_SET_POWER_D3; + + drm_dp_dpcd_writeb(ctrl->drm_aux, DP_SET_POWER, value); + } + } + + edp_ctrl_irq_enable(ctrl, 0); + + edp_ctrl_link_enable(ctrl, 0); + + edp_ctrl_phy_aux_enable(ctrl, 0); + + ctrl->power_on = false; + +unlock_ret: + mutex_unlock(&ctrl->dev_mutex); +} + +irqreturn_t msm_edp_ctrl_irq(struct edp_ctrl *ctrl) +{ + u32 isr1, isr2, mask1, mask2; + u32 ack; + + DBG(""); + spin_lock(&ctrl->irq_lock); + isr1 = edp_read(ctrl->base + REG_EDP_INTERRUPT_REG_1); + isr2 = edp_read(ctrl->base + REG_EDP_INTERRUPT_REG_2); + + mask1 = isr1 & EDP_INTR_MASK1; + mask2 = isr2 & EDP_INTR_MASK2; + + isr1 &= ~mask1; /* remove masks bit */ + isr2 &= ~mask2; + + DBG("isr=%x mask=%x isr2=%x mask2=%x", + isr1, mask1, isr2, mask2); + + ack = isr1 & EDP_INTR_STATUS1; + ack <<= 1; /* ack bits */ + ack |= mask1; + edp_write(ctrl->base + REG_EDP_INTERRUPT_REG_1, ack); + + ack = isr2 & EDP_INTR_STATUS2; + ack <<= 1; /* ack bits */ + ack |= mask2; + edp_write(ctrl->base + REG_EDP_INTERRUPT_REG_2, ack); + spin_unlock(&ctrl->irq_lock); + + if (isr1 & EDP_INTERRUPT_REG_1_HPD) + DBG("edp_hpd"); + + if (isr2 & EDP_INTERRUPT_REG_2_READY_FOR_VIDEO) + DBG("edp_video_ready"); + + if (isr2 & EDP_INTERRUPT_REG_2_IDLE_PATTERNs_SENT) { + DBG("idle_patterns_sent"); + complete(&ctrl->idle_comp); + } + + msm_edp_aux_irq(ctrl->aux, isr1); + + return IRQ_HANDLED; +} + +void msm_edp_ctrl_power(struct edp_ctrl *ctrl, bool on) +{ + if (on) + queue_work(ctrl->workqueue, &ctrl->on_work); + else + queue_work(ctrl->workqueue, &ctrl->off_work); +} + +int msm_edp_ctrl_init(struct msm_edp *edp) +{ + struct edp_ctrl *ctrl = NULL; + struct device *dev; + int ret; + + if (!edp) { + pr_err("%s: edp is NULL!\n", __func__); + return -EINVAL; + } + + dev = &edp->pdev->dev; + ctrl = devm_kzalloc(dev, sizeof(*ctrl), GFP_KERNEL); + if (!ctrl) + return -ENOMEM; + + edp->ctrl = ctrl; + ctrl->pdev = edp->pdev; + + ctrl->base = msm_ioremap(ctrl->pdev, "edp", "eDP"); + if (IS_ERR(ctrl->base)) + return PTR_ERR(ctrl->base); + + /* Get regulator, clock, gpio, pwm */ + ret = edp_regulator_init(ctrl); + if (ret) { + pr_err("%s:regulator init fail\n", __func__); + return ret; + } + ret = edp_clk_init(ctrl); + if (ret) { + pr_err("%s:clk init fail\n", __func__); + return ret; + } + ret = edp_gpio_config(ctrl); + if (ret) { + pr_err("%s:failed to configure GPIOs: %d", __func__, ret); + return ret; + } + + /* Init aux and phy */ + ctrl->aux = msm_edp_aux_init(edp, ctrl->base, &ctrl->drm_aux); + if (!ctrl->aux || !ctrl->drm_aux) { + pr_err("%s:failed to init aux\n", __func__); + return -ENOMEM; + } + + ctrl->phy = msm_edp_phy_init(dev, ctrl->base); + if (!ctrl->phy) { + pr_err("%s:failed to init phy\n", __func__); + ret = -ENOMEM; + goto err_destory_aux; + } + + spin_lock_init(&ctrl->irq_lock); + mutex_init(&ctrl->dev_mutex); + init_completion(&ctrl->idle_comp); + + /* setup workqueue */ + ctrl->workqueue = alloc_ordered_workqueue("edp_drm_work", 0); + INIT_WORK(&ctrl->on_work, edp_ctrl_on_worker); + INIT_WORK(&ctrl->off_work, edp_ctrl_off_worker); + + return 0; + +err_destory_aux: + msm_edp_aux_destroy(dev, ctrl->aux); + ctrl->aux = NULL; + return ret; +} + +void msm_edp_ctrl_destroy(struct edp_ctrl *ctrl) +{ + if (!ctrl) + return; + + if (ctrl->workqueue) { + destroy_workqueue(ctrl->workqueue); + ctrl->workqueue = NULL; + } + + if (ctrl->aux) { + msm_edp_aux_destroy(&ctrl->pdev->dev, ctrl->aux); + ctrl->aux = NULL; + } + + kfree(ctrl->edid); + ctrl->edid = NULL; + + mutex_destroy(&ctrl->dev_mutex); +} + +bool msm_edp_ctrl_panel_connected(struct edp_ctrl *ctrl) +{ + mutex_lock(&ctrl->dev_mutex); + DBG("connect status = %d", ctrl->edp_connected); + if (ctrl->edp_connected) { + mutex_unlock(&ctrl->dev_mutex); + return true; + } + + if (!ctrl->power_on) { + edp_ctrl_phy_aux_enable(ctrl, 1); + edp_ctrl_irq_enable(ctrl, 1); + } + + if (drm_dp_dpcd_read(ctrl->drm_aux, DP_DPCD_REV, ctrl->dpcd, + DP_RECEIVER_CAP_SIZE) < DP_RECEIVER_CAP_SIZE) { + pr_err("%s: AUX channel is NOT ready\n", __func__); + memset(ctrl->dpcd, 0, DP_RECEIVER_CAP_SIZE); + } else { + ctrl->edp_connected = true; + } + + if (!ctrl->power_on) { + edp_ctrl_irq_enable(ctrl, 0); + edp_ctrl_phy_aux_enable(ctrl, 0); + } + + DBG("exit: connect status=%d", ctrl->edp_connected); + + mutex_unlock(&ctrl->dev_mutex); + + return ctrl->edp_connected; +} + +int msm_edp_ctrl_get_panel_info(struct edp_ctrl *ctrl, + struct drm_connector *connector, struct edid **edid) +{ + mutex_lock(&ctrl->dev_mutex); + + if (ctrl->edid) { + if (edid) { + DBG("Just return edid buffer"); + *edid = ctrl->edid; + } + goto unlock_ret; + } + + if (!ctrl->power_on) { + edp_ctrl_phy_aux_enable(ctrl, 1); + edp_ctrl_irq_enable(ctrl, 1); + } + + /* Initialize link rate as panel max link rate */ + ctrl->link_rate = ctrl->dpcd[DP_MAX_LINK_RATE]; + + ctrl->edid = drm_get_edid(connector, &ctrl->drm_aux->ddc); + if (!ctrl->edid) { + pr_err("%s: edid read fail\n", __func__); + goto disable_ret; + } + + if (edid) + *edid = ctrl->edid; + +disable_ret: + if (!ctrl->power_on) { + edp_ctrl_irq_enable(ctrl, 0); + edp_ctrl_phy_aux_enable(ctrl, 0); + } +unlock_ret: + mutex_unlock(&ctrl->dev_mutex); + return 0; +} + +int msm_edp_ctrl_timing_cfg(struct edp_ctrl *ctrl, + const struct drm_display_mode *mode, + const struct drm_display_info *info) +{ + u32 hstart_from_sync, vstart_from_sync; + u32 data; + int ret = 0; + + mutex_lock(&ctrl->dev_mutex); + /* + * Need to keep color depth, pixel rate and + * interlaced information in ctrl context + */ + ctrl->color_depth = info->bpc; + ctrl->pixel_rate = mode->clock; + ctrl->interlaced = !!(mode->flags & DRM_MODE_FLAG_INTERLACE); + + /* Fill initial link config based on passed in timing */ + edp_fill_link_cfg(ctrl); + + if (edp_clk_enable(ctrl, EDP_CLK_MASK_AHB)) { + pr_err("%s, fail to prepare enable ahb clk\n", __func__); + ret = -EINVAL; + goto unlock_ret; + } + edp_clock_synchrous(ctrl, 1); + + /* Configure eDP timing to HW */ + edp_write(ctrl->base + REG_EDP_TOTAL_HOR_VER, + EDP_TOTAL_HOR_VER_HORIZ(mode->htotal) | + EDP_TOTAL_HOR_VER_VERT(mode->vtotal)); + + vstart_from_sync = mode->vtotal - mode->vsync_start; + hstart_from_sync = mode->htotal - mode->hsync_start; + edp_write(ctrl->base + REG_EDP_START_HOR_VER_FROM_SYNC, + EDP_START_HOR_VER_FROM_SYNC_HORIZ(hstart_from_sync) | + EDP_START_HOR_VER_FROM_SYNC_VERT(vstart_from_sync)); + + data = EDP_HSYNC_VSYNC_WIDTH_POLARITY_VERT( + mode->vsync_end - mode->vsync_start); + data |= EDP_HSYNC_VSYNC_WIDTH_POLARITY_HORIZ( + mode->hsync_end - mode->hsync_start); + if (mode->flags & DRM_MODE_FLAG_NVSYNC) + data |= EDP_HSYNC_VSYNC_WIDTH_POLARITY_NVSYNC; + if (mode->flags & DRM_MODE_FLAG_NHSYNC) + data |= EDP_HSYNC_VSYNC_WIDTH_POLARITY_NHSYNC; + edp_write(ctrl->base + REG_EDP_HSYNC_VSYNC_WIDTH_POLARITY, data); + + edp_write(ctrl->base + REG_EDP_ACTIVE_HOR_VER, + EDP_ACTIVE_HOR_VER_HORIZ(mode->hdisplay) | + EDP_ACTIVE_HOR_VER_VERT(mode->vdisplay)); + + edp_clk_disable(ctrl, EDP_CLK_MASK_AHB); + +unlock_ret: + mutex_unlock(&ctrl->dev_mutex); + return ret; +} + +bool msm_edp_ctrl_pixel_clock_valid(struct edp_ctrl *ctrl, + u32 pixel_rate, u32 *pm, u32 *pn) +{ + const struct edp_pixel_clk_div *divs; + u32 err = 1; /* 1% error tolerance */ + u32 clk_err; + int i; + + if (ctrl->link_rate == DP_LINK_BW_1_62) { + divs = clk_divs[0]; + } else if (ctrl->link_rate == DP_LINK_BW_2_7) { + divs = clk_divs[1]; + } else { + pr_err("%s: Invalid link rate,%d\n", __func__, ctrl->link_rate); + return false; + } + + for (i = 0; i < EDP_PIXEL_CLK_NUM; i++) { + clk_err = abs(divs[i].rate - pixel_rate); + if ((divs[i].rate * err / 100) >= clk_err) { + if (pm) + *pm = divs[i].m; + if (pn) + *pn = divs[i].n; + return true; + } + } + + DBG("pixel clock %d(kHz) not supported", pixel_rate); + + return false; +} + diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index d7574e6bd4e4..ae52412d529a 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -309,8 +309,8 @@ void msm_gem_shrinker_init(struct drm_device *dev); void msm_gem_shrinker_cleanup(struct drm_device *dev); struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj); -int msm_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map); -void msm_gem_prime_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map); +int msm_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map); +void msm_gem_prime_vunmap(struct drm_gem_object *obj, struct iosys_map *map); struct drm_gem_object *msm_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sg); int msm_gem_prime_pin(struct drm_gem_object *obj); diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c index fc94e061d6a7..e8f1b7a2ca9c 100644 --- a/drivers/gpu/drm/msm/msm_gem_prime.c +++ b/drivers/gpu/drm/msm/msm_gem_prime.c @@ -22,19 +22,19 @@ struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj) return drm_prime_pages_to_sg(obj->dev, msm_obj->pages, npages); } -int msm_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) +int msm_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map) { void *vaddr; vaddr = msm_gem_get_vaddr(obj); if (IS_ERR(vaddr)) return PTR_ERR(vaddr); - dma_buf_map_set_vaddr(map, vaddr); + iosys_map_set_vaddr(map, vaddr); return 0; } -void msm_gem_prime_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map) +void msm_gem_prime_vunmap(struct drm_gem_object *obj, struct iosys_map *map) { msm_gem_put_vaddr(obj); } diff --git a/drivers/gpu/drm/mxsfb/mxsfb_drv.c b/drivers/gpu/drm/mxsfb/mxsfb_drv.c index 86d78634a979..375f26d4a417 100644 --- a/drivers/gpu/drm/mxsfb/mxsfb_drv.c +++ b/drivers/gpu/drm/mxsfb/mxsfb_drv.c @@ -374,12 +374,20 @@ static int mxsfb_remove(struct platform_device *pdev) struct drm_device *drm = platform_get_drvdata(pdev); drm_dev_unregister(drm); + drm_atomic_helper_shutdown(drm); mxsfb_unload(drm); drm_dev_put(drm); return 0; } +static void mxsfb_shutdown(struct platform_device *pdev) +{ + struct drm_device *drm = platform_get_drvdata(pdev); + + drm_atomic_helper_shutdown(drm); +} + #ifdef CONFIG_PM_SLEEP static int mxsfb_suspend(struct device *dev) { @@ -403,6 +411,7 @@ static const struct dev_pm_ops mxsfb_pm_ops = { static struct platform_driver mxsfb_platform_driver = { .probe = mxsfb_probe, .remove = mxsfb_remove, + .shutdown = mxsfb_shutdown, .driver = { .name = "mxsfb", .of_match_table = mxsfb_dt_ids, diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig index 9436310d0854..3ec690b6f0b4 100644 --- a/drivers/gpu/drm/nouveau/Kconfig +++ b/drivers/gpu/drm/nouveau/Kconfig @@ -4,6 +4,7 @@ config DRM_NOUVEAU depends on DRM && PCI && MMU select IOMMU_API select FW_LOADER + select DRM_DP_HELPER select DRM_KMS_HELPER select DRM_TTM select DRM_TTM_HELPER diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index ae1f41205520..df58c6445c51 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -35,7 +35,7 @@ #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_fb_helper.h> #include <drm/drm_plane_helper.h> diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h index 40f90e353540..1b173191cc41 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.h +++ b/drivers/gpu/drm/nouveau/nouveau_connector.h @@ -36,7 +36,7 @@ #include <drm/drm_crtc.h> #include <drm/drm_edid.h> #include <drm/drm_encoder.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_util.h> #include "nouveau_crtc.h" diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c index 040ed88d362d..724d40ddd452 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dp.c +++ b/drivers/gpu/drm/nouveau/nouveau_dp.c @@ -22,7 +22,7 @@ * Authors: Ben Skeggs */ -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include "nouveau_drv.h" #include "nouveau_connector.h" @@ -147,6 +147,21 @@ nouveau_dp_detect(struct nouveau_connector *nv_connector, nv_encoder->dp.link_nr = dpcd[DP_MAX_LANE_COUNT] & DP_MAX_LANE_COUNT_MASK; + if (connector->connector_type == DRM_MODE_CONNECTOR_eDP && dpcd[DP_DPCD_REV] >= 0x13) { + struct drm_dp_aux *aux = &nv_connector->aux; + int ret, i; + u8 sink_rates[16]; + + ret = drm_dp_dpcd_read(aux, DP_SUPPORTED_LINK_RATES, sink_rates, sizeof(sink_rates)); + if (ret == sizeof(sink_rates)) { + for (i = 0; i < ARRAY_SIZE(sink_rates); i += 2) { + int val = ((sink_rates[i + 1] << 8) | sink_rates[i]) * 200 / 10; + if (val && (i == 0 || val > nv_encoder->dp.link_bw)) + nv_encoder->dp.link_bw = val; + } + } + } + NV_DEBUG(drm, "display: %dx%d dpcd 0x%02x\n", nv_encoder->dp.link_nr, nv_encoder->dp.link_bw, dpcd[DP_DPCD_REV]); diff --git a/drivers/gpu/drm/nouveau/nouveau_encoder.h b/drivers/gpu/drm/nouveau/nouveau_encoder.h index 77c2fed76e8b..65ed84f88cca 100644 --- a/drivers/gpu/drm/nouveau/nouveau_encoder.h +++ b/drivers/gpu/drm/nouveau/nouveau_encoder.h @@ -30,8 +30,8 @@ #include <subdev/bios/dcb.h> #include <drm/drm_encoder_slave.h> -#include <drm/drm_dp_helper.h> -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include "dispnv04/disp.h" struct nv50_head_atom; struct nouveau_connector; diff --git a/drivers/gpu/drm/nouveau/nouveau_mem.c b/drivers/gpu/drm/nouveau/nouveau_mem.c index 2ca3207c13fc..2e517cdc24c9 100644 --- a/drivers/gpu/drm/nouveau/nouveau_mem.c +++ b/drivers/gpu/drm/nouveau/nouveau_mem.c @@ -162,11 +162,12 @@ nouveau_mem_vram(struct ttm_resource *reg, bool contig, u8 page) } void -nouveau_mem_del(struct ttm_resource *reg) +nouveau_mem_del(struct ttm_resource_manager *man, struct ttm_resource *reg) { struct nouveau_mem *mem = nouveau_mem(reg); nouveau_mem_fini(mem); + ttm_resource_fini(man, reg); kfree(mem); } diff --git a/drivers/gpu/drm/nouveau/nouveau_mem.h b/drivers/gpu/drm/nouveau/nouveau_mem.h index 2c01166a90f2..325551eba5cd 100644 --- a/drivers/gpu/drm/nouveau/nouveau_mem.h +++ b/drivers/gpu/drm/nouveau/nouveau_mem.h @@ -23,7 +23,8 @@ nouveau_mem(struct ttm_resource *reg) int nouveau_mem_new(struct nouveau_cli *, u8 kind, u8 comp, struct ttm_resource **); -void nouveau_mem_del(struct ttm_resource *); +void nouveau_mem_del(struct ttm_resource_manager *man, + struct ttm_resource *); int nouveau_mem_vram(struct ttm_resource *, bool contig, u8 page); int nouveau_mem_host(struct ttm_resource *, struct ttm_tt *); void nouveau_mem_fini(struct nouveau_mem *); diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c index 2ca9d9a9e5d5..85f1f5a0fe5d 100644 --- a/drivers/gpu/drm/nouveau/nouveau_ttm.c +++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c @@ -36,9 +36,10 @@ #include <core/tegra.h> static void -nouveau_manager_del(struct ttm_resource_manager *man, struct ttm_resource *reg) +nouveau_manager_del(struct ttm_resource_manager *man, + struct ttm_resource *reg) { - nouveau_mem_del(reg); + nouveau_mem_del(man, reg); } static int @@ -62,7 +63,7 @@ nouveau_vram_manager_new(struct ttm_resource_manager *man, ret = nouveau_mem_vram(*res, nvbo->contig, nvbo->page); if (ret) { - nouveau_mem_del(*res); + nouveau_mem_del(man, *res); return ret; } @@ -118,7 +119,7 @@ nv04_gart_manager_new(struct ttm_resource_manager *man, ret = nvif_vmm_get(&mem->cli->vmm.vmm, PTES, false, 12, 0, (long)(*res)->num_pages << PAGE_SHIFT, &mem->vma[0]); if (ret) { - nouveau_mem_del(*res); + nouveau_mem_del(man, *res); return ret; } @@ -163,7 +164,7 @@ nouveau_ttm_init_vram(struct nouveau_drm *drm) man->func = &nouveau_vram_manager; - ttm_resource_manager_init(man, + ttm_resource_manager_init(man, &drm->ttm.bdev, drm->gem.vram_available >> PAGE_SHIFT); ttm_set_driver_manager(&drm->ttm.bdev, TTM_PL_VRAM, man); ttm_resource_manager_set_used(man, true); @@ -210,7 +211,7 @@ nouveau_ttm_init_gtt(struct nouveau_drm *drm) man->func = func; man->use_tt = true; - ttm_resource_manager_init(man, size_pages); + ttm_resource_manager_init(man, &drm->ttm.bdev, size_pages); ttm_set_driver_manager(&drm->ttm.bdev, TTM_PL_TT, man); ttm_resource_manager_set_used(man, true); return 0; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dp.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dp.c index 9669472a2749..8e09315b8fb3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dp.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dp.c @@ -41,6 +41,10 @@ struct lt_state { struct nvkm_dp *dp; + + int repeaters; + int repeater; + u8 stat[6]; u8 conf[4]; bool pc2; @@ -52,14 +56,26 @@ static int nvkm_dp_train_sense(struct lt_state *lt, bool pc, u32 delay) { struct nvkm_dp *dp = lt->dp; + u32 addr; int ret; - if (dp->dpcd[DPCD_RC0E_AUX_RD_INTERVAL]) - mdelay(dp->dpcd[DPCD_RC0E_AUX_RD_INTERVAL] * 4); + usleep_range(delay, delay * 2); + + if (lt->repeater) + addr = DPCD_LTTPR_LANE0_1_STATUS(lt->repeater); + else + addr = DPCD_LS02; + + ret = nvkm_rdaux(dp->aux, addr, <->stat[0], 3); + if (ret) + return ret; + + if (lt->repeater) + addr = DPCD_LTTPR_LANE0_1_ADJUST(lt->repeater); else - udelay(delay); + addr = DPCD_LS06; - ret = nvkm_rdaux(dp->aux, DPCD_LS02, lt->stat, 6); + ret = nvkm_rdaux(dp->aux, addr, <->stat[4], 2); if (ret) return ret; @@ -85,6 +101,7 @@ nvkm_dp_train_drive(struct lt_state *lt, bool pc) struct nvbios_dpout info; struct nvbios_dpcfg ocfg; u8 ver, hdr, cnt, len; + u32 addr; u32 data; int ret, i; @@ -113,6 +130,9 @@ nvkm_dp_train_drive(struct lt_state *lt, bool pc) OUTP_TRACE(&dp->outp, "config lane %d %02x %02x", i, lt->conf[i], lpc2); + if (lt->repeater != lt->repeaters) + continue; + data = nvbios_dpout_match(bios, dp->outp.info.hasht, dp->outp.info.hashm, &ver, &hdr, &cnt, &len, &info); @@ -129,7 +149,12 @@ nvkm_dp_train_drive(struct lt_state *lt, bool pc) ocfg.pe, ocfg.tx_pu); } - ret = nvkm_wraux(dp->aux, DPCD_LC03(0), lt->conf, 4); + if (lt->repeater) + addr = DPCD_LTTPR_LANE0_SET(lt->repeater); + else + addr = DPCD_LC03(0); + + ret = nvkm_wraux(dp->aux, addr, lt->conf, 4); if (ret) return ret; @@ -146,32 +171,59 @@ static void nvkm_dp_train_pattern(struct lt_state *lt, u8 pattern) { struct nvkm_dp *dp = lt->dp; + u32 addr; u8 sink_tp; OUTP_TRACE(&dp->outp, "training pattern %d", pattern); dp->outp.ior->func->dp.pattern(dp->outp.ior, pattern); - nvkm_rdaux(dp->aux, DPCD_LC02, &sink_tp, 1); + if (lt->repeater) + addr = DPCD_LTTPR_PATTERN_SET(lt->repeater); + else + addr = DPCD_LC02; + + nvkm_rdaux(dp->aux, addr, &sink_tp, 1); sink_tp &= ~DPCD_LC02_TRAINING_PATTERN_SET; - sink_tp |= pattern; - nvkm_wraux(dp->aux, DPCD_LC02, &sink_tp, 1); + sink_tp |= (pattern != 4) ? pattern : 7; + + if (pattern != 0) + sink_tp |= DPCD_LC02_SCRAMBLING_DISABLE; + else + sink_tp &= ~DPCD_LC02_SCRAMBLING_DISABLE; + nvkm_wraux(dp->aux, addr, &sink_tp, 1); } static int nvkm_dp_train_eq(struct lt_state *lt) { + struct nvkm_i2c_aux *aux = lt->dp->aux; bool eq_done = false, cr_done = true; - int tries = 0, i; + int tries = 0, usec = 0, i; + u8 data; - if (lt->dp->dpcd[DPCD_RC02] & DPCD_RC02_TPS3_SUPPORTED) - nvkm_dp_train_pattern(lt, 3); - else - nvkm_dp_train_pattern(lt, 2); + if (lt->repeater) { + if (!nvkm_rdaux(aux, DPCD_LTTPR_AUX_RD_INTERVAL(lt->repeater), &data, sizeof(data))) + usec = (data & DPCD_RC0E_AUX_RD_INTERVAL) * 4000; + + nvkm_dp_train_pattern(lt, 4); + } else { + if (lt->dp->dpcd[DPCD_RC00_DPCD_REV] >= 0x14 && + lt->dp->dpcd[DPCD_RC03] & DPCD_RC03_TPS4_SUPPORTED) + nvkm_dp_train_pattern(lt, 4); + else + if (lt->dp->dpcd[DPCD_RC00_DPCD_REV] >= 0x12 && + lt->dp->dpcd[DPCD_RC02] & DPCD_RC02_TPS3_SUPPORTED) + nvkm_dp_train_pattern(lt, 3); + else + nvkm_dp_train_pattern(lt, 2); + + usec = (lt->dp->dpcd[DPCD_RC0E] & DPCD_RC0E_AUX_RD_INTERVAL) * 4000; + } do { if ((tries && nvkm_dp_train_drive(lt, lt->pc2)) || - nvkm_dp_train_sense(lt, lt->pc2, 400)) + nvkm_dp_train_sense(lt, lt->pc2, usec ? usec : 400)) break; eq_done = !!(lt->stat[2] & DPCD_LS04_INTERLANE_ALIGN_DONE); @@ -193,13 +245,16 @@ nvkm_dp_train_cr(struct lt_state *lt) { bool cr_done = false, abort = false; int voltage = lt->conf[0] & DPCD_LC03_VOLTAGE_SWING_SET; - int tries = 0, i; + int tries = 0, usec = 0, i; nvkm_dp_train_pattern(lt, 1); + if (lt->dp->dpcd[DPCD_RC00_DPCD_REV] < 0x14 && !lt->repeater) + usec = (lt->dp->dpcd[DPCD_RC0E] & DPCD_RC0E_AUX_RD_INTERVAL) * 4000; + do { if (nvkm_dp_train_drive(lt, false) || - nvkm_dp_train_sense(lt, false, 100)) + nvkm_dp_train_sense(lt, false, usec ? usec : 100)) break; cr_done = true; @@ -223,7 +278,7 @@ nvkm_dp_train_cr(struct lt_state *lt) } static int -nvkm_dp_train_links(struct nvkm_dp *dp) +nvkm_dp_train_links(struct nvkm_dp *dp, int rate) { struct nvkm_ior *ior = dp->outp.ior; struct nvkm_disp *disp = dp->outp.disp; @@ -233,13 +288,15 @@ nvkm_dp_train_links(struct nvkm_dp *dp) .dp = dp, }; u32 lnkcmp; - u8 sink[2]; + u8 sink[2], data; int ret; OUTP_DBG(&dp->outp, "training %d x %d MB/s", ior->dp.nr, ior->dp.bw * 27); /* Intersect misc. capabilities of the OR and sink. */ + if (disp->engine.subdev.device->chipset < 0x110) + dp->dpcd[DPCD_RC03] &= ~DPCD_RC03_TPS4_SUPPORTED; if (disp->engine.subdev.device->chipset < 0xd0) dp->dpcd[DPCD_RC02] &= ~DPCD_RC02_TPS3_SUPPORTED; lt.pc2 = dp->dpcd[DPCD_RC02] & DPCD_RC02_TPS3_SUPPORTED; @@ -287,8 +344,22 @@ nvkm_dp_train_links(struct nvkm_dp *dp) ior->func->dp.power(ior, ior->dp.nr); + /* Select LTTPR non-transparent mode if we have a valid configuration, + * use transparent mode otherwise. + */ + if (dp->lttpr[0] >= 0x14) { + data = DPCD_LTTPR_MODE_TRANSPARENT; + nvkm_wraux(dp->aux, DPCD_LTTPR_MODE, &data, sizeof(data)); + + if (dp->lttprs) { + data = DPCD_LTTPR_MODE_NON_TRANSPARENT; + nvkm_wraux(dp->aux, DPCD_LTTPR_MODE, &data, sizeof(data)); + lt.repeaters = dp->lttprs; + } + } + /* Set desired link configuration on the sink. */ - sink[0] = ior->dp.bw; + sink[0] = (dp->rate[rate].dpcd < 0) ? ior->dp.bw : 0; sink[1] = ior->dp.nr; if (ior->dp.ef) sink[1] |= DPCD_LC01_ENHANCED_FRAME_EN; @@ -297,12 +368,33 @@ nvkm_dp_train_links(struct nvkm_dp *dp) if (ret) return ret; + if (dp->rate[rate].dpcd >= 0) { + ret = nvkm_rdaux(dp->aux, DPCD_LC15_LINK_RATE_SET, &sink[0], sizeof(sink[0])); + if (ret) + return ret; + + sink[0] &= ~DPCD_LC15_LINK_RATE_SET_MASK; + sink[0] |= dp->rate[rate].dpcd; + + ret = nvkm_wraux(dp->aux, DPCD_LC15_LINK_RATE_SET, &sink[0], sizeof(sink[0])); + if (ret) + return ret; + } + /* Attempt to train the link in this configuration. */ - memset(lt.stat, 0x00, sizeof(lt.stat)); - ret = nvkm_dp_train_cr(<); - if (ret == 0) - ret = nvkm_dp_train_eq(<); - nvkm_dp_train_pattern(<, 0); + for (lt.repeater = lt.repeaters; lt.repeater >= 0; lt.repeater--) { + if (lt.repeater) + OUTP_DBG(&dp->outp, "training LTTPR%d", lt.repeater); + else + OUTP_DBG(&dp->outp, "training sink"); + + memset(lt.stat, 0x00, sizeof(lt.stat)); + ret = nvkm_dp_train_cr(<); + if (ret == 0) + ret = nvkm_dp_train_eq(<); + nvkm_dp_train_pattern(<, 0); + } + return ret; } @@ -345,63 +437,13 @@ nvkm_dp_train_init(struct nvkm_dp *dp) } } -static const struct dp_rates { - u32 rate; - u8 bw; - u8 nr; -} nvkm_dp_rates[] = { - { 2160000, 0x14, 4 }, - { 1080000, 0x0a, 4 }, - { 1080000, 0x14, 2 }, - { 648000, 0x06, 4 }, - { 540000, 0x0a, 2 }, - { 540000, 0x14, 1 }, - { 324000, 0x06, 2 }, - { 270000, 0x0a, 1 }, - { 162000, 0x06, 1 }, - {} -}; - static int nvkm_dp_train(struct nvkm_dp *dp, u32 dataKBps) { struct nvkm_ior *ior = dp->outp.ior; - const u8 sink_nr = dp->dpcd[DPCD_RC02] & DPCD_RC02_MAX_LANE_COUNT; - const u8 sink_bw = dp->dpcd[DPCD_RC01_MAX_LINK_RATE]; - const u8 outp_nr = dp->outp.info.dpconf.link_nr; - const u8 outp_bw = dp->outp.info.dpconf.link_bw; - const struct dp_rates *failsafe = NULL, *cfg; - int ret = -EINVAL; + int ret = -EINVAL, nr, rate; u8 pwr; - /* Find the lowest configuration of the OR that can support - * the required link rate. - * - * We will refuse to program the OR to lower rates, even if - * link training fails at higher rates (or even if the sink - * can't support the rate at all, though the DD is supposed - * to prevent such situations from happening). - * - * Attempting to do so can cause the entire display to hang, - * and it's better to have a failed modeset than that. - */ - for (cfg = nvkm_dp_rates; cfg->rate; cfg++) { - if (cfg->nr <= outp_nr && cfg->bw <= outp_bw) { - /* Try to respect sink limits too when selecting - * lowest link configuration. - */ - if (!failsafe || - (cfg->nr <= sink_nr && cfg->bw <= sink_bw)) - failsafe = cfg; - } - - if (failsafe && cfg[1].rate < dataKBps) - break; - } - - if (WARN_ON(!failsafe)) - return ret; - /* Ensure sink is not in a low-power state. */ if (!nvkm_rdaux(dp->aux, DPCD_SC00, &pwr, 1)) { if ((pwr & DPCD_SC00_SET_POWER) != DPCD_SC00_SET_POWER_D0) { @@ -411,25 +453,22 @@ nvkm_dp_train(struct nvkm_dp *dp, u32 dataKBps) } } + ior->dp.mst = dp->lt.mst; + ior->dp.ef = dp->dpcd[DPCD_RC02] & DPCD_RC02_ENHANCED_FRAME_CAP; + ior->dp.nr = 0; + /* Link training. */ - OUTP_DBG(&dp->outp, "training (min: %d x %d MB/s)", - failsafe->nr, failsafe->bw * 27); + OUTP_DBG(&dp->outp, "training"); nvkm_dp_train_init(dp); - for (cfg = nvkm_dp_rates; ret < 0 && cfg <= failsafe; cfg++) { - /* Skip configurations not supported by both OR and sink. */ - if ((cfg->nr > outp_nr || cfg->bw > outp_bw || - cfg->nr > sink_nr || cfg->bw > sink_bw)) { - if (cfg != failsafe) - continue; - OUTP_ERR(&dp->outp, "link rate unsupported by sink"); + for (nr = dp->links; ret < 0 && nr; nr >>= 1) { + for (rate = 0; ret < 0 && rate < dp->rates; rate++) { + if (dp->rate[rate].rate * nr >= dataKBps || WARN_ON(!ior->dp.nr)) { + /* Program selected link configuration. */ + ior->dp.bw = dp->rate[rate].rate / 27000; + ior->dp.nr = nr; + ret = nvkm_dp_train_links(dp, rate); + } } - ior->dp.mst = dp->lt.mst; - ior->dp.ef = dp->dpcd[DPCD_RC02] & DPCD_RC02_ENHANCED_FRAME_CAP; - ior->dp.bw = cfg->bw; - ior->dp.nr = cfg->nr; - - /* Program selected link configuration. */ - ret = nvkm_dp_train_links(dp); } nvkm_dp_train_fini(dp); if (ret < 0) @@ -527,6 +566,47 @@ done: } static bool +nvkm_dp_enable_supported_link_rates(struct nvkm_dp *dp) +{ + u8 sink_rates[DPCD_RC10_SUPPORTED_LINK_RATES__SIZE]; + int i, j, k; + + if (dp->outp.conn->info.type != DCB_CONNECTOR_eDP || + dp->dpcd[DPCD_RC00_DPCD_REV] < 0x13 || + nvkm_rdaux(dp->aux, DPCD_RC10_SUPPORTED_LINK_RATES(0), sink_rates, sizeof(sink_rates))) + return false; + + for (i = 0; i < ARRAY_SIZE(sink_rates); i += 2) { + const u32 rate = ((sink_rates[i + 1] << 8) | sink_rates[i]) * 200 / 10; + + if (!rate || WARN_ON(dp->rates == ARRAY_SIZE(dp->rate))) + break; + + if (rate > dp->outp.info.dpconf.link_bw * 27000) { + OUTP_DBG(&dp->outp, "rate %d !outp", rate); + continue; + } + + for (j = 0; j < dp->rates; j++) { + if (rate > dp->rate[j].rate) { + for (k = dp->rates; k > j; k--) + dp->rate[k] = dp->rate[k - 1]; + break; + } + } + + dp->rate[j].dpcd = i / 2; + dp->rate[j].rate = rate; + dp->rates++; + } + + for (i = 0; i < dp->rates; i++) + OUTP_DBG(&dp->outp, "link_rate[%d] = %d", dp->rate[i].dpcd, dp->rate[i].rate); + + return dp->rates != 0; +} + +static bool nvkm_dp_enable(struct nvkm_dp *dp, bool enable) { struct nvkm_i2c_aux *aux = dp->aux; @@ -538,9 +618,60 @@ nvkm_dp_enable(struct nvkm_dp *dp, bool enable) dp->present = true; } - if (!nvkm_rdaux(aux, DPCD_RC00_DPCD_REV, dp->dpcd, - sizeof(dp->dpcd))) + /* Detect any LTTPRs before reading DPCD receiver caps. */ + if (!nvkm_rdaux(aux, DPCD_LTTPR_REV, dp->lttpr, sizeof(dp->lttpr)) && + dp->lttpr[0] >= 0x14 && dp->lttpr[2]) { + switch (dp->lttpr[2]) { + case 0x80: dp->lttprs = 1; break; + case 0x40: dp->lttprs = 2; break; + case 0x20: dp->lttprs = 3; break; + case 0x10: dp->lttprs = 4; break; + case 0x08: dp->lttprs = 5; break; + case 0x04: dp->lttprs = 6; break; + case 0x02: dp->lttprs = 7; break; + case 0x01: dp->lttprs = 8; break; + default: + /* Unknown LTTPR count, we'll switch to transparent mode. */ + WARN_ON(1); + dp->lttprs = 0; + break; + } + } else { + /* No LTTPR support, or zero LTTPR count - don't touch it at all. */ + memset(dp->lttpr, 0x00, sizeof(dp->lttpr)); + } + + if (!nvkm_rdaux(aux, DPCD_RC00_DPCD_REV, dp->dpcd, sizeof(dp->dpcd))) { + const u8 rates[] = { 0x1e, 0x14, 0x0a, 0x06, 0 }; + const u8 *rate; + int rate_max; + + dp->rates = 0; + dp->links = dp->dpcd[DPCD_RC02] & DPCD_RC02_MAX_LANE_COUNT; + dp->links = min(dp->links, dp->outp.info.dpconf.link_nr); + if (dp->lttprs && dp->lttpr[4]) + dp->links = min_t(int, dp->links, dp->lttpr[4]); + + rate_max = dp->dpcd[DPCD_RC01_MAX_LINK_RATE]; + rate_max = min(rate_max, dp->outp.info.dpconf.link_bw); + if (dp->lttprs && dp->lttpr[1]) + rate_max = min_t(int, rate_max, dp->lttpr[1]); + + if (!nvkm_dp_enable_supported_link_rates(dp)) { + for (rate = rates; *rate; rate++) { + if (*rate <= rate_max) { + if (WARN_ON(dp->rates == ARRAY_SIZE(dp->rate))) + break; + + dp->rate[dp->rates].dpcd = -1; + dp->rate[dp->rates].rate = *rate * 27000; + dp->rates++; + } + } + } + return true; + } } if (dp->present) { diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dp.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dp.h index e484d0c3b0d4..8e59dd469da6 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dp.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dp.h @@ -9,10 +9,7 @@ #include <subdev/bios/dp.h> struct nvkm_dp { - union { - struct nvkm_outp base; - struct nvkm_outp outp; - }; + struct nvkm_outp outp; struct nvbios_dpout info; u8 version; @@ -21,8 +18,17 @@ struct nvkm_dp { struct nvkm_notify hpd; bool present; + u8 lttpr[6]; + u8 lttprs; u8 dpcd[16]; + struct { + int dpcd; /* -1, or index into SUPPORTED_LINK_RATES table */ + u32 rate; + } rate[8]; + int rates; + int links; + struct mutex mutex; struct { atomic_t done; @@ -42,8 +48,12 @@ void nvkm_dp_disable(struct nvkm_outp *, struct nvkm_ior *); #define DPCD_RC02_TPS3_SUPPORTED 0x40 #define DPCD_RC02_MAX_LANE_COUNT 0x1f #define DPCD_RC03 0x00003 +#define DPCD_RC03_TPS4_SUPPORTED 0x80 #define DPCD_RC03_MAX_DOWNSPREAD 0x01 -#define DPCD_RC0E_AUX_RD_INTERVAL 0x0000e +#define DPCD_RC0E 0x0000e +#define DPCD_RC0E_AUX_RD_INTERVAL 0x7f +#define DPCD_RC10_SUPPORTED_LINK_RATES(i) 0x00010 +#define DPCD_RC10_SUPPORTED_LINK_RATES__SIZE 16 /* DPCD Link Configuration */ #define DPCD_LC00_LINK_BW_SET 0x00100 @@ -51,7 +61,8 @@ void nvkm_dp_disable(struct nvkm_outp *, struct nvkm_ior *); #define DPCD_LC01_ENHANCED_FRAME_EN 0x80 #define DPCD_LC01_LANE_COUNT_SET 0x1f #define DPCD_LC02 0x00102 -#define DPCD_LC02_TRAINING_PATTERN_SET 0x03 +#define DPCD_LC02_TRAINING_PATTERN_SET 0x0f +#define DPCD_LC02_SCRAMBLING_DISABLE 0x20 #define DPCD_LC03(l) ((l) + 0x00103) #define DPCD_LC03_MAX_PRE_EMPHASIS_REACHED 0x20 #define DPCD_LC03_PRE_EMPHASIS_SET 0x18 @@ -67,6 +78,8 @@ void nvkm_dp_disable(struct nvkm_outp *, struct nvkm_ior *); #define DPCD_LC10_LANE3_POST_CURSOR2_SET 0x30 #define DPCD_LC10_LANE2_MAX_POST_CURSOR2_REACHED 0x04 #define DPCD_LC10_LANE2_POST_CURSOR2_SET 0x03 +#define DPCD_LC15_LINK_RATE_SET 0x00115 +#define DPCD_LC15_LINK_RATE_SET_MASK 0x07 /* DPCD Link/Sink Status */ #define DPCD_LS02 0x00202 @@ -108,4 +121,14 @@ void nvkm_dp_disable(struct nvkm_outp *, struct nvkm_ior *); #define DPCD_SC00_SET_POWER 0x03 #define DPCD_SC00_SET_POWER_D0 0x01 #define DPCD_SC00_SET_POWER_D3 0x03 + +#define DPCD_LTTPR_REV 0xf0000 +#define DPCD_LTTPR_MODE 0xf0003 +#define DPCD_LTTPR_MODE_TRANSPARENT 0x55 +#define DPCD_LTTPR_MODE_NON_TRANSPARENT 0xaa +#define DPCD_LTTPR_PATTERN_SET(i) ((i - 1) * 0x50 + 0xf0010) +#define DPCD_LTTPR_LANE0_SET(i) ((i - 1) * 0x50 + 0xf0011) +#define DPCD_LTTPR_AUX_RD_INTERVAL(i) ((i - 1) * 0x50 + 0xf0020) +#define DPCD_LTTPR_LANE0_1_STATUS(i) ((i - 1) * 0x50 + 0xf0030) +#define DPCD_LTTPR_LANE0_1_ADJUST(i) ((i - 1) * 0x50 + 0xf0033) #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg94.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg94.c index 4d59d02525d9..56b8f4411988 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg94.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorg94.c @@ -77,7 +77,18 @@ g94_sor_dp_pattern(struct nvkm_ior *sor, int pattern) { struct nvkm_device *device = sor->disp->engine.subdev.device; const u32 loff = nv50_sor_link(sor); - nvkm_mask(device, 0x61c10c + loff, 0x0f000000, pattern << 24); + u32 data; + + switch (pattern) { + case 0: data = 0x00001000; break; + case 1: data = 0x01000000; break; + case 2: data = 0x02000000; break; + default: + WARN_ON(1); + return; + } + + nvkm_mask(device, 0x61c10c + loff, 0x0f001000, data); } void diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorga102.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorga102.c index 033827de9116..d2c05f5c4aa0 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorga102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorga102.c @@ -37,6 +37,10 @@ ga102_sor_dp_links(struct nvkm_ior *sor, struct nvkm_i2c_aux *aux) case 0x0a: clksor |= 0x00040000; break; case 0x14: clksor |= 0x00080000; break; case 0x1e: clksor |= 0x000c0000; break; + case 0x08: clksor |= 0x00100000; break; + case 0x09: clksor |= 0x00140000; break; + case 0x0c: clksor |= 0x00180000; break; + case 0x10: clksor |= 0x001c0000; break; default: WARN_ON(1); return -EINVAL; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c index 3b3643fb1019..c431e0b9fc11 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgf119.c @@ -92,7 +92,19 @@ gf119_sor_dp_pattern(struct nvkm_ior *sor, int pattern) { struct nvkm_device *device = sor->disp->engine.subdev.device; const u32 soff = nv50_ior_base(sor); - nvkm_mask(device, 0x61c110 + soff, 0x0f0f0f0f, 0x01010101 * pattern); + u32 data; + + switch (pattern) { + case 0: data = 0x10101010; break; + case 1: data = 0x01010101; break; + case 2: data = 0x02020202; break; + case 3: data = 0x03030303; break; + default: + WARN_ON(1); + return; + } + + nvkm_mask(device, 0x61c110 + soff, 0x1f1f1f1f, data); } int diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm107.c index 38045c92197f..3696bfd3bfd7 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm107.c @@ -28,11 +28,23 @@ gm107_sor_dp_pattern(struct nvkm_ior *sor, int pattern) { struct nvkm_device *device = sor->disp->engine.subdev.device; const u32 soff = nv50_ior_base(sor); - const u32 data = 0x01010101 * pattern; + u32 mask = 0x1f1f1f1f, data; + + switch (pattern) { + case 0: data = 0x10101010; break; + case 1: data = 0x01010101; break; + case 2: data = 0x02020202; break; + case 3: data = 0x03030303; break; + case 4: data = 0x1b1b1b1b; break; + default: + WARN_ON(1); + return; + } + if (sor->asy.link & 1) - nvkm_mask(device, 0x61c110 + soff, 0x0f0f0f0f, data); + nvkm_mask(device, 0x61c110 + soff, mask, data); else - nvkm_mask(device, 0x61c12c + soff, 0x0f0f0f0f, data); + nvkm_mask(device, 0x61c12c + soff, mask, data); } static const struct nvkm_ior_func diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/acr/hsfw.c b/drivers/gpu/drm/nouveau/nvkm/subdev/acr/hsfw.c index 667fa016496e..a6ea89a5d51a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/acr/hsfw.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/acr/hsfw.c @@ -142,11 +142,12 @@ nvkm_acr_hsfw_load_bl(struct nvkm_acr *acr, const char *name, int ver, hsfw->imem_size = desc->code_size; hsfw->imem_tag = desc->start_tag; - hsfw->imem = kmalloc(desc->code_size, GFP_KERNEL); - memcpy(hsfw->imem, data + desc->code_off, desc->code_size); - + hsfw->imem = kmemdup(data + desc->code_off, desc->code_size, GFP_KERNEL); nvkm_firmware_put(fw); - return 0; + if (!hsfw->imem) + return -ENOMEM; + else + return 0; } int diff --git a/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c index 57af3d97be77..ab3fc86e7256 100644 --- a/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c +++ b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c @@ -93,6 +93,7 @@ struct dma_buf *omap_gem_prime_export(struct drm_gem_object *obj, int flags) exp_info.size = omap_gem_mmap_size(obj); exp_info.flags = flags; exp_info.priv = obj; + exp_info.resv = obj->resv; return drm_gem_dmabuf_export(obj->dev, &exp_info); } diff --git a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c index 5fcbde789ddb..1be150ac758f 100644 --- a/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c +++ b/drivers/gpu/drm/panel/panel-boe-tv101wum-nl6.c @@ -86,7 +86,7 @@ static const struct panel_init_cmd boe_tv110c9m_init_cmd[] = { _INIT_DCS_CMD(0x0F, 0x73), _INIT_DCS_CMD(0x95, 0xE6), _INIT_DCS_CMD(0x96, 0xF0), - _INIT_DCS_CMD(0x30, 0x11), + _INIT_DCS_CMD(0x30, 0x00), _INIT_DCS_CMD(0x6D, 0x66), _INIT_DCS_CMD(0x75, 0xA2), _INIT_DCS_CMD(0x77, 0x3B), @@ -112,17 +112,17 @@ static const struct panel_init_cmd boe_tv110c9m_init_cmd[] = { _INIT_DCS_CMD(0xB1, 0x00, 0xD2, 0x01, 0x0B, 0x01, 0x34, 0x01, 0x76, 0x01, 0xA3, 0x01, 0xEF, 0x02, 0x27, 0x02, 0x29), _INIT_DCS_CMD(0xB2, 0x02, 0x5F, 0x02, 0x9E, 0x02, 0xC9, 0x03, 0x00, 0x03, 0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73), - _INIT_DCS_CMD(0xB3, 0x03, 0x86, 0x03, 0x9A, 0x03, 0xA7, 0x03, 0xCF, 0x03, 0xDE, 0x03, 0xE0), + _INIT_DCS_CMD(0xB3, 0x03, 0x86, 0x03, 0x9A, 0x03, 0xAF, 0x03, 0xDF, 0x03, 0xF5, 0x03, 0xE0), _INIT_DCS_CMD(0xB4, 0x00, 0x00, 0x00, 0x1B, 0x00, 0x45, 0x00, 0x65, 0x00, 0x81, 0x00, 0x99, 0x00, 0xAE, 0x00, 0xC1), _INIT_DCS_CMD(0xB5, 0x00, 0xD2, 0x01, 0x0B, 0x01, 0x34, 0x01, 0x76, 0x01, 0xA3, 0x01, 0xEF, 0x02, 0x27, 0x02, 0x29), _INIT_DCS_CMD(0xB6, 0x02, 0x5F, 0x02, 0x9E, 0x02, 0xC9, 0x03, 0x00, 0x03, 0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73), - _INIT_DCS_CMD(0xB7, 0x03, 0x86, 0x03, 0x9A, 0x03, 0xA7, 0x03, 0xCF, 0x03, 0xDE, 0x03, 0xE0), + _INIT_DCS_CMD(0xB7, 0x03, 0x86, 0x03, 0x9A, 0x03, 0xAF, 0x03, 0xDF, 0x03, 0xF5, 0x03, 0xE0), _INIT_DCS_CMD(0xB8, 0x00, 0x00, 0x00, 0x1B, 0x00, 0x45, 0x00, 0x65, 0x00, 0x81, 0x00, 0x99, 0x00, 0xAE, 0x00, 0xC1), _INIT_DCS_CMD(0xB9, 0x00, 0xD2, 0x01, 0x0B, 0x01, 0x34, 0x01, 0x76, 0x01, 0xA3, 0x01, 0xEF, 0x02, 0x27, 0x02, 0x29), _INIT_DCS_CMD(0xBA, 0x02, 0x5F, 0x02, 0x9E, 0x02, 0xC9, 0x03, 0x00, 0x03, 0x26, 0x03, 0x53, 0x03, 0x63, 0x03, 0x73), - _INIT_DCS_CMD(0xBB, 0x03, 0x86, 0x03, 0x9A, 0x03, 0xA7, 0x03, 0xCF, 0x03, 0xDE, 0x03, 0xE0), + _INIT_DCS_CMD(0xBB, 0x03, 0x86, 0x03, 0x9A, 0x03, 0xAF, 0x03, 0xDF, 0x03, 0xF5, 0x03, 0xE0), _INIT_DCS_CMD(0xFF, 0x24), _INIT_DCS_CMD(0xFB, 0x01), diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c index 176ef0c3cc1d..a394a15dc3fb 100644 --- a/drivers/gpu/drm/panel/panel-edp.c +++ b/drivers/gpu/drm/panel/panel-edp.c @@ -36,8 +36,8 @@ #include <drm/drm_crtc.h> #include <drm/drm_device.h> -#include <drm/drm_dp_aux_bus.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_aux_bus.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_panel.h> /** @@ -1745,6 +1745,19 @@ static const struct panel_delay delay_200_500_e50 = { .enable = 50, }; +static const struct panel_delay delay_200_500_e80_d50 = { + .hpd_absent = 200, + .unprepare = 500, + .enable = 80, + .disable = 50, +}; + +static const struct panel_delay delay_100_500_e200 = { + .hpd_absent = 100, + .unprepare = 500, + .enable = 200, +}; + #define EDP_PANEL_ENTRY(vend_chr_0, vend_chr_1, vend_chr_2, product_id, _delay, _name) \ { \ .name = _name, \ @@ -1768,13 +1781,17 @@ static const struct edp_panel_entry edp_panels[] = { EDP_PANEL_ENTRY('B', 'O', 'E', 0x07d1, &boe_nv133fhm_n61.delay, "NV133FHM-N61"), EDP_PANEL_ENTRY('B', 'O', 'E', 0x082d, &boe_nv133fhm_n61.delay, "NV133FHM-N62"), EDP_PANEL_ENTRY('B', 'O', 'E', 0x098d, &boe_nv110wtm_n61.delay, "NV110WTM-N61"), + EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a5d, &delay_200_500_e50, "NV116WHM-N45"), EDP_PANEL_ENTRY('C', 'M', 'N', 0x114c, &innolux_n116bca_ea1.delay, "N116BCA-EA1"), EDP_PANEL_ENTRY('K', 'D', 'B', 0x0624, &kingdisplay_kd116n21_30nv_a010.delay, "116N21-30NV-A010"), + EDP_PANEL_ENTRY('K', 'D', 'B', 0x1120, &delay_200_500_e80_d50, "116N29-30NK-C007"), EDP_PANEL_ENTRY('S', 'H', 'P', 0x154c, &delay_200_500_p2e100, "LQ116M1JW10"), + EDP_PANEL_ENTRY('S', 'T', 'A', 0x0100, &delay_100_500_e200, "2081116HHD028001-51D"), + { /* sentinal */ } }; diff --git a/drivers/gpu/drm/panel/panel-samsung-atna33xc20.c b/drivers/gpu/drm/panel/panel-samsung-atna33xc20.c index 221db6512859..20666b6217e7 100644 --- a/drivers/gpu/drm/panel/panel-samsung-atna33xc20.c +++ b/drivers/gpu/drm/panel/panel-samsung-atna33xc20.c @@ -14,8 +14,8 @@ #include <linux/pm_runtime.h> #include <linux/regulator/consumer.h> -#include <drm/drm_dp_aux_bus.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_aux_bus.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_panel.h> diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c index 3c08f9827acf..c324a3ed34b9 100644 --- a/drivers/gpu/drm/panel/panel-simple.c +++ b/drivers/gpu/drm/panel/panel-simple.c @@ -2526,6 +2526,36 @@ static const struct panel_desc mitsubishi_aa070mc01 = { .bus_flags = DRM_BUS_FLAG_DE_HIGH, }; +static const struct display_timing multi_inno_mi0700s4t_6_timing = { + .pixelclock = { 29000000, 33000000, 38000000 }, + .hactive = { 800, 800, 800 }, + .hfront_porch = { 180, 210, 240 }, + .hback_porch = { 16, 16, 16 }, + .hsync_len = { 30, 30, 30 }, + .vactive = { 480, 480, 480 }, + .vfront_porch = { 12, 22, 32 }, + .vback_porch = { 10, 10, 10 }, + .vsync_len = { 13, 13, 13 }, + .flags = DISPLAY_FLAGS_HSYNC_LOW | DISPLAY_FLAGS_VSYNC_LOW | + DISPLAY_FLAGS_DE_HIGH | DISPLAY_FLAGS_PIXDATA_POSEDGE | + DISPLAY_FLAGS_SYNC_POSEDGE, +}; + +static const struct panel_desc multi_inno_mi0700s4t_6 = { + .timings = &multi_inno_mi0700s4t_6_timing, + .num_timings = 1, + .bpc = 8, + .size = { + .width = 154, + .height = 86, + }, + .bus_format = MEDIA_BUS_FMT_RGB888_1X24, + .bus_flags = DRM_BUS_FLAG_DE_HIGH | + DRM_BUS_FLAG_PIXDATA_SAMPLE_NEGEDGE | + DRM_BUS_FLAG_SYNC_SAMPLE_NEGEDGE, + .connector_type = DRM_MODE_CONNECTOR_DPI, +}; + static const struct display_timing multi_inno_mi1010ait_1cp_timing = { .pixelclock = { 68900000, 70000000, 73400000 }, .hactive = { 1280, 1280, 1280 }, @@ -3873,6 +3903,9 @@ static const struct of_device_id platform_of_match[] = { .compatible = "mitsubishi,aa070mc01-ca1", .data = &mitsubishi_aa070mc01, }, { + .compatible = "multi-inno,mi0700s4t-6", + .data = &multi_inno_mi0700s4t_6, + }, { .compatible = "multi-inno,mi1010ait-1cp", .data = &multi_inno_mi1010ait_1cp, }, { diff --git a/drivers/gpu/drm/panfrost/panfrost_features.h b/drivers/gpu/drm/panfrost/panfrost_features.h index 5056777c7744..34f2bae1ec8c 100644 --- a/drivers/gpu/drm/panfrost/panfrost_features.h +++ b/drivers/gpu/drm/panfrost/panfrost_features.h @@ -12,24 +12,6 @@ enum panfrost_hw_feature { HW_FEATURE_JOBCHAIN_DISAMBIGUATION, HW_FEATURE_PWRON_DURING_PWROFF_TRANS, HW_FEATURE_XAFFINITY, - HW_FEATURE_OUT_OF_ORDER_EXEC, - HW_FEATURE_MRT, - HW_FEATURE_BRNDOUT_CC, - HW_FEATURE_INTERPIPE_REG_ALIASING, - HW_FEATURE_LD_ST_TILEBUFFER, - HW_FEATURE_MSAA_16X, - HW_FEATURE_32_BIT_UNIFORM_ADDRESS, - HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL, - HW_FEATURE_OPTIMIZED_COVERAGE_MASK, - HW_FEATURE_T7XX_PAIRING_RULES, - HW_FEATURE_LD_ST_LEA_TEX, - HW_FEATURE_LINEAR_FILTER_FLOAT, - HW_FEATURE_WORKGROUP_ROUND_MULTIPLE_OF_4, - HW_FEATURE_IMAGES_IN_FRAGMENT_SHADERS, - HW_FEATURE_TEST4_DATUM_MODE, - HW_FEATURE_NEXT_INSTRUCTION_TYPE, - HW_FEATURE_BRNDOUT_KILL, - HW_FEATURE_WARPING, HW_FEATURE_V4, HW_FEATURE_FLUSH_REDUCTION, HW_FEATURE_PROTECTED_MODE, @@ -42,128 +24,31 @@ enum panfrost_hw_feature { }; #define hw_features_t600 (\ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ BIT_ULL(HW_FEATURE_V4)) -#define hw_features_t620 (\ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ - BIT_ULL(HW_FEATURE_V4)) - -#define hw_features_t720 (\ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ - BIT_ULL(HW_FEATURE_OPTIMIZED_COVERAGE_MASK) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ - BIT_ULL(HW_FEATURE_WORKGROUP_ROUND_MULTIPLE_OF_4) | \ - BIT_ULL(HW_FEATURE_WARPING) | \ - BIT_ULL(HW_FEATURE_V4)) +#define hw_features_t620 hw_features_t600 +#define hw_features_t720 hw_features_t600 #define hw_features_t760 (\ BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_MSAA_16X) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT)) -// T860 -#define hw_features_t860 (\ - BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ - BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ - BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_MSAA_16X) | \ - BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ - BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT)) +#define hw_features_t860 hw_features_t760 -#define hw_features_t880 hw_features_t860 +#define hw_features_t880 hw_features_t760 -#define hw_features_t830 (\ - BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ - BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ - BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_WARPING) | \ - BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ - BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT)) +#define hw_features_t830 hw_features_t760 -#define hw_features_t820 (\ - BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ - BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ - BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_WARPING) | \ - BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ - BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT)) +#define hw_features_t820 hw_features_t760 #define hw_features_g71 (\ BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_WARPING) | \ - BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_MSAA_16X) | \ - BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ @@ -173,71 +58,18 @@ enum panfrost_hw_feature { BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_WARPING) | \ - BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_MSAA_16X) | \ - BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \ BIT_ULL(HW_FEATURE_COHERENCY_REG)) -#define hw_features_g51 (\ - BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ - BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ - BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_WARPING) | \ - BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_MSAA_16X) | \ - BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ - BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ - BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ - BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ - BIT_ULL(HW_FEATURE_PROTECTED_DEBUG_MODE) | \ - BIT_ULL(HW_FEATURE_COHERENCY_REG)) +#define hw_features_g51 hw_features_g72 #define hw_features_g52 (\ BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_WARPING) | \ - BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_MSAA_16X) | \ - BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ @@ -248,21 +80,6 @@ enum panfrost_hw_feature { BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_WARPING) | \ - BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_MSAA_16X) | \ - BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ @@ -276,21 +93,6 @@ enum panfrost_hw_feature { BIT_ULL(HW_FEATURE_JOBCHAIN_DISAMBIGUATION) | \ BIT_ULL(HW_FEATURE_PWRON_DURING_PWROFF_TRANS) | \ BIT_ULL(HW_FEATURE_XAFFINITY) | \ - BIT_ULL(HW_FEATURE_WARPING) | \ - BIT_ULL(HW_FEATURE_INTERPIPE_REG_ALIASING) | \ - BIT_ULL(HW_FEATURE_32_BIT_UNIFORM_ADDRESS) | \ - BIT_ULL(HW_FEATURE_ATTR_AUTO_TYPE_INFERRAL) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_CC) | \ - BIT_ULL(HW_FEATURE_BRNDOUT_KILL) | \ - BIT_ULL(HW_FEATURE_LD_ST_LEA_TEX) | \ - BIT_ULL(HW_FEATURE_LD_ST_TILEBUFFER) | \ - BIT_ULL(HW_FEATURE_LINEAR_FILTER_FLOAT) | \ - BIT_ULL(HW_FEATURE_MRT) | \ - BIT_ULL(HW_FEATURE_MSAA_16X) | \ - BIT_ULL(HW_FEATURE_NEXT_INSTRUCTION_TYPE) | \ - BIT_ULL(HW_FEATURE_OUT_OF_ORDER_EXEC) | \ - BIT_ULL(HW_FEATURE_T7XX_PAIRING_RULES) | \ - BIT_ULL(HW_FEATURE_TEST4_DATUM_MODE) | \ BIT_ULL(HW_FEATURE_THREAD_GROUP_SPLIT) | \ BIT_ULL(HW_FEATURE_FLUSH_REDUCTION) | \ BIT_ULL(HW_FEATURE_PROTECTED_MODE) | \ diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.c b/drivers/gpu/drm/panfrost/panfrost_gpu.c index bbe628b306ee..15cec831a99a 100644 --- a/drivers/gpu/drm/panfrost/panfrost_gpu.c +++ b/drivers/gpu/drm/panfrost/panfrost_gpu.c @@ -320,19 +320,36 @@ void panfrost_gpu_power_on(struct panfrost_device *pfdev) { int ret; u32 val; + u64 core_mask = U64_MAX; panfrost_gpu_init_quirks(pfdev); - /* Just turn on everything for now */ - gpu_write(pfdev, L2_PWRON_LO, pfdev->features.l2_present); + if (pfdev->features.l2_present != 1) { + /* + * Only support one core group now. + * ~(l2_present - 1) unsets all bits in l2_present except + * the bottom bit. (l2_present - 2) has all the bits in + * the first core group set. AND them together to generate + * a mask of cores in the first core group. + */ + core_mask = ~(pfdev->features.l2_present - 1) & + (pfdev->features.l2_present - 2); + dev_info_once(pfdev->dev, "using only 1st core group (%lu cores from %lu)\n", + hweight64(core_mask), + hweight64(pfdev->features.shader_present)); + } + gpu_write(pfdev, L2_PWRON_LO, pfdev->features.l2_present & core_mask); ret = readl_relaxed_poll_timeout(pfdev->iomem + L2_READY_LO, - val, val == pfdev->features.l2_present, 100, 20000); + val, val == (pfdev->features.l2_present & core_mask), + 100, 20000); if (ret) dev_err(pfdev->dev, "error powering up gpu L2"); - gpu_write(pfdev, SHADER_PWRON_LO, pfdev->features.shader_present); + gpu_write(pfdev, SHADER_PWRON_LO, + pfdev->features.shader_present & core_mask); ret = readl_relaxed_poll_timeout(pfdev->iomem + SHADER_READY_LO, - val, val == pfdev->features.shader_present, 100, 20000); + val, val == (pfdev->features.shader_present & core_mask), + 100, 20000); if (ret) dev_err(pfdev->dev, "error powering up gpu shader"); @@ -360,8 +377,11 @@ int panfrost_gpu_init(struct panfrost_device *pfdev) panfrost_gpu_init_features(pfdev); - dma_set_mask_and_coherent(pfdev->dev, + err = dma_set_mask_and_coherent(pfdev->dev, DMA_BIT_MASK(FIELD_GET(0xff00, pfdev->features.mmu_features))); + if (err) + return err; + dma_set_max_seg_size(pfdev->dev, UINT_MAX); irq = platform_get_irq_byname(to_platform_device(pfdev->dev), "gpu"); diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c index 1d36df5af98d..bc0df93f7f21 100644 --- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c +++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c @@ -1,16 +1,17 @@ // SPDX-License-Identifier: GPL-2.0 /* Copyright 2019 Collabora Ltd */ -#include <drm/drm_file.h> -#include <drm/drm_gem_shmem_helper.h> -#include <drm/panfrost_drm.h> #include <linux/completion.h> -#include <linux/dma-buf-map.h> #include <linux/iopoll.h> +#include <linux/iosys-map.h> #include <linux/pm_runtime.h> #include <linux/slab.h> #include <linux/uaccess.h> +#include <drm/drm_file.h> +#include <drm/drm_gem_shmem_helper.h> +#include <drm/panfrost_drm.h> + #include "panfrost_device.h" #include "panfrost_features.h" #include "panfrost_gem.h" @@ -73,7 +74,7 @@ static int panfrost_perfcnt_enable_locked(struct panfrost_device *pfdev, { struct panfrost_file_priv *user = file_priv->driver_priv; struct panfrost_perfcnt *perfcnt = pfdev->perfcnt; - struct dma_buf_map map; + struct iosys_map map; struct drm_gem_shmem_object *bo; u32 cfg, as; int ret; @@ -181,7 +182,7 @@ static int panfrost_perfcnt_disable_locked(struct panfrost_device *pfdev, { struct panfrost_file_priv *user = file_priv->driver_priv; struct panfrost_perfcnt *perfcnt = pfdev->perfcnt; - struct dma_buf_map map = DMA_BUF_MAP_INIT_VADDR(perfcnt->buf); + struct iosys_map map = IOSYS_MAP_INIT_VADDR(perfcnt->buf); if (user != perfcnt->user) return -EINVAL; diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index 9e0a1e836011..9a9c29b1d3e1 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -25,7 +25,7 @@ #include <linux/crc32.h> #include <linux/delay.h> -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <drm/drm_drv.h> #include <drm/drm_atomic.h> @@ -566,8 +566,8 @@ static struct qxl_bo *qxl_create_cursor(struct qxl_device *qdev, { static const u32 size = 64 * 64 * 4; struct qxl_bo *cursor_bo; - struct dma_buf_map cursor_map; - struct dma_buf_map user_map; + struct iosys_map cursor_map; + struct iosys_map user_map; struct qxl_cursor cursor; int ret; @@ -1183,7 +1183,7 @@ int qxl_create_monitors_object(struct qxl_device *qdev) { int ret; struct drm_gem_object *gobj; - struct dma_buf_map map; + struct iosys_map map; int monitors_config_size = sizeof(struct qxl_monitors_config) + qxl_num_crtc * sizeof(struct qxl_head); diff --git a/drivers/gpu/drm/qxl/qxl_draw.c b/drivers/gpu/drm/qxl/qxl_draw.c index 7d27891e87fa..a93de9e1977a 100644 --- a/drivers/gpu/drm/qxl/qxl_draw.c +++ b/drivers/gpu/drm/qxl/qxl_draw.c @@ -20,7 +20,7 @@ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <drm/drm_fourcc.h> @@ -44,7 +44,7 @@ static struct qxl_rect *drawable_set_clipping(struct qxl_device *qdev, unsigned int num_clips, struct qxl_bo *clips_bo) { - struct dma_buf_map map; + struct iosys_map map; struct qxl_clip_rects *dev_clips; int ret; @@ -146,7 +146,7 @@ void qxl_draw_dirty_fb(struct qxl_device *qdev, int stride = fb->pitches[0]; /* depth is not actually interesting, we don't mask with it */ int depth = fb->format->cpp[0] * 8; - struct dma_buf_map surface_map; + struct iosys_map surface_map; uint8_t *surface_base; struct qxl_release *release; struct qxl_bo *clips_bo; diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h index 359266d9e860..9796099ff18f 100644 --- a/drivers/gpu/drm/qxl/qxl_drv.h +++ b/drivers/gpu/drm/qxl/qxl_drv.h @@ -30,7 +30,7 @@ * Definitions taken from spice-protocol, plus kernel driver specific bits. */ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/dma-fence.h> #include <linux/firmware.h> #include <linux/platform_device.h> @@ -50,7 +50,7 @@ #include "qxl_dev.h" -struct dma_buf_map; +struct iosys_map; #define DRIVER_AUTHOR "Dave Airlie" @@ -81,7 +81,7 @@ struct qxl_bo { /* Protected by tbo.reserved */ struct ttm_place placements[3]; struct ttm_placement placement; - struct dma_buf_map map; + struct iosys_map map; void *kptr; unsigned int map_count; int type; @@ -431,9 +431,9 @@ struct sg_table *qxl_gem_prime_get_sg_table(struct drm_gem_object *obj); struct drm_gem_object *qxl_gem_prime_import_sg_table( struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sgt); -int qxl_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map); +int qxl_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map); void qxl_gem_prime_vunmap(struct drm_gem_object *obj, - struct dma_buf_map *map); + struct iosys_map *map); /* qxl_irq.c */ int qxl_irq_init(struct qxl_device *qdev); diff --git a/drivers/gpu/drm/qxl/qxl_object.c b/drivers/gpu/drm/qxl/qxl_object.c index fbb36e3e8564..b42a657e4c2f 100644 --- a/drivers/gpu/drm/qxl/qxl_object.c +++ b/drivers/gpu/drm/qxl/qxl_object.c @@ -23,7 +23,7 @@ * Alon Levy */ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/io-mapping.h> #include "qxl_drv.h" @@ -158,7 +158,7 @@ int qxl_bo_create(struct qxl_device *qdev, unsigned long size, return 0; } -int qxl_bo_vmap_locked(struct qxl_bo *bo, struct dma_buf_map *map) +int qxl_bo_vmap_locked(struct qxl_bo *bo, struct iosys_map *map) { int r; @@ -184,7 +184,7 @@ out: return 0; } -int qxl_bo_vmap(struct qxl_bo *bo, struct dma_buf_map *map) +int qxl_bo_vmap(struct qxl_bo *bo, struct iosys_map *map) { int r; @@ -210,7 +210,7 @@ void *qxl_bo_kmap_atomic_page(struct qxl_device *qdev, void *rptr; int ret; struct io_mapping *map; - struct dma_buf_map bo_map; + struct iosys_map bo_map; if (bo->tbo.resource->mem_type == TTM_PL_VRAM) map = qdev->vram_mapping; diff --git a/drivers/gpu/drm/qxl/qxl_object.h b/drivers/gpu/drm/qxl/qxl_object.h index cee4b52b75dd..53392cb90eec 100644 --- a/drivers/gpu/drm/qxl/qxl_object.h +++ b/drivers/gpu/drm/qxl/qxl_object.h @@ -59,8 +59,8 @@ extern int qxl_bo_create(struct qxl_device *qdev, u32 priority, struct qxl_surface *surf, struct qxl_bo **bo_ptr); -int qxl_bo_vmap(struct qxl_bo *bo, struct dma_buf_map *map); -int qxl_bo_vmap_locked(struct qxl_bo *bo, struct dma_buf_map *map); +int qxl_bo_vmap(struct qxl_bo *bo, struct iosys_map *map); +int qxl_bo_vmap_locked(struct qxl_bo *bo, struct iosys_map *map); int qxl_bo_vunmap(struct qxl_bo *bo); void qxl_bo_vunmap_locked(struct qxl_bo *bo); void *qxl_bo_kmap_atomic_page(struct qxl_device *qdev, struct qxl_bo *bo, int page_offset); diff --git a/drivers/gpu/drm/qxl/qxl_prime.c b/drivers/gpu/drm/qxl/qxl_prime.c index 4a10cb0a413b..142d01415acb 100644 --- a/drivers/gpu/drm/qxl/qxl_prime.c +++ b/drivers/gpu/drm/qxl/qxl_prime.c @@ -54,7 +54,7 @@ struct drm_gem_object *qxl_gem_prime_import_sg_table( return ERR_PTR(-ENOSYS); } -int qxl_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) +int qxl_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map) { struct qxl_bo *bo = gem_to_qxl_bo(obj); int ret; @@ -67,7 +67,7 @@ int qxl_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) } void qxl_gem_prime_vunmap(struct drm_gem_object *obj, - struct dma_buf_map *map) + struct iosys_map *map) { struct qxl_bo *bo = gem_to_qxl_bo(obj); diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c index 4c1e551d9714..4798cf23d251 100644 --- a/drivers/gpu/drm/radeon/atombios_dp.c +++ b/drivers/gpu/drm/radeon/atombios_dp.c @@ -30,7 +30,7 @@ #include "atom.h" #include "atom-bits.h" -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> /* move these to drm_dp_helper.c/h */ #define DP_LINK_CONFIGURATION_SIZE 9 diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index 81b4de7be9f2..5819737c21c6 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -8517,8 +8517,8 @@ int cik_suspend(struct radeon_device *rdev) cik_cp_enable(rdev, false); cik_sdma_enable(rdev, false); if (rdev->has_uvd) { - uvd_v1_0_fini(rdev); radeon_uvd_suspend(rdev); + uvd_v1_0_fini(rdev); } if (rdev->has_vce) radeon_vce_suspend(rdev); diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index eeb590d2dec2..455f8036aa54 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -5156,8 +5156,8 @@ int evergreen_suspend(struct radeon_device *rdev) radeon_pm_suspend(rdev); radeon_audio_fini(rdev); if (rdev->has_uvd) { - uvd_v1_0_fini(rdev); radeon_uvd_suspend(rdev); + uvd_v1_0_fini(rdev); } r700_cp_stop(rdev); r600_dma_stop(rdev); diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index 4a364ca7a1be..927e5f42e97d 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -2323,8 +2323,8 @@ int cayman_suspend(struct radeon_device *rdev) cayman_cp_enable(rdev, false); cayman_dma_stop(rdev); if (rdev->has_uvd) { - uvd_v1_0_fini(rdev); radeon_uvd_suspend(rdev); + uvd_v1_0_fini(rdev); } evergreen_irq_suspend(rdev); radeon_wb_disable(rdev); diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index ca3fcae2adb5..dd78fc499402 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -3232,8 +3232,8 @@ int r600_suspend(struct radeon_device *rdev) radeon_audio_fini(rdev); r600_cp_stop(rdev); if (rdev->has_uvd) { - uvd_v1_0_fini(rdev); radeon_uvd_suspend(rdev); + uvd_v1_0_fini(rdev); } r600_irq_suspend(rdev); radeon_wb_disable(rdev); diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 607ad5620bd9..a7925a8290b2 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -27,7 +27,7 @@ #include <drm/drm_edid.h> #include <drm/drm_crtc_helper.h> #include <drm/drm_fb_helper.h> -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_probe_helper.h> #include <drm/radeon_drm.h> #include "radeon.h" @@ -204,7 +204,7 @@ int radeon_get_monitor_bpc(struct drm_connector *connector) /* Check if bpc is within clock limit. Try to degrade gracefully otherwise */ if ((bpc == 12) && (mode_clock * 3/2 > max_tmds_clock)) { - if ((connector->display_info.edid_hdmi_dc_modes & DRM_EDID_HDMI_DC_30) && + if ((connector->display_info.edid_hdmi_rgb444_dc_modes & DRM_EDID_HDMI_DC_30) && (mode_clock * 5/4 <= max_tmds_clock)) bpc = 10; else diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 4f0fbf667431..15692cb241fc 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -1086,19 +1086,6 @@ static unsigned int radeon_vga_set_decode(struct pci_dev *pdev, bool state) } /** - * radeon_check_pot_argument - check that argument is a power of two - * - * @arg: value to check - * - * Validates that a certain argument is a power of two (all asics). - * Returns true if argument is valid. - */ -static bool radeon_check_pot_argument(int arg) -{ - return (arg & (arg - 1)) == 0; -} - -/** * radeon_gart_size_auto - Determine a sensible default GART size * according to ASIC family. * @@ -1126,7 +1113,7 @@ static int radeon_gart_size_auto(enum radeon_family family) static void radeon_check_arguments(struct radeon_device *rdev) { /* vramlimit must be a power of two */ - if (!radeon_check_pot_argument(radeon_vram_limit)) { + if (!is_power_of_2(radeon_vram_limit)) { dev_warn(rdev->dev, "vram limit (%d) must be a power of 2\n", radeon_vram_limit); radeon_vram_limit = 0; @@ -1140,7 +1127,7 @@ static void radeon_check_arguments(struct radeon_device *rdev) dev_warn(rdev->dev, "gart size (%d) too small\n", radeon_gart_size); radeon_gart_size = radeon_gart_size_auto(rdev->family); - } else if (!radeon_check_pot_argument(radeon_gart_size)) { + } else if (!is_power_of_2(radeon_gart_size)) { dev_warn(rdev->dev, "gart size (%d) must be a power of 2\n", radeon_gart_size); radeon_gart_size = radeon_gart_size_auto(rdev->family); @@ -1163,7 +1150,7 @@ static void radeon_check_arguments(struct radeon_device *rdev) break; } - if (!radeon_check_pot_argument(radeon_vm_size)) { + if (!is_power_of_2(radeon_vm_size)) { dev_warn(rdev->dev, "VM size (%d) must be a power of 2\n", radeon_vm_size); radeon_vm_size = 4; diff --git a/drivers/gpu/drm/radeon/radeon_dp_mst.c b/drivers/gpu/drm/radeon/radeon_dp_mst.c index 751c2c075e09..9f26baf7adb0 100644 --- a/drivers/gpu/drm/radeon/radeon_dp_mst.c +++ b/drivers/gpu/drm/radeon/radeon_dp_mst.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: MIT -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_fb_helper.h> #include <drm/drm_file.h> #include <drm/drm_probe_helper.h> diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index a36a4f2c76b0..f563284a7fac 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -26,6 +26,7 @@ * Jerome Glisse */ +#include <linux/iosys-map.h> #include <linux/pci.h> #include <drm/drm_device.h> diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index fe16f140a6b4..5288dc7a4897 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -33,8 +33,8 @@ #include <drm/drm_crtc.h> #include <drm/drm_edid.h> #include <drm/drm_encoder.h> -#include <drm/drm_dp_helper.h> -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_fixed.h> #include <drm/drm_crtc_helper.h> #include <linux/i2c.h> diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c index 56ede9d63b12..87536d205593 100644 --- a/drivers/gpu/drm/radeon/radeon_object.c +++ b/drivers/gpu/drm/radeon/radeon_object.c @@ -567,7 +567,6 @@ int radeon_bo_get_surface_reg(struct radeon_bo *bo) return 0; if (bo->surface_reg >= 0) { - reg = &rdev->surface_regs[bo->surface_reg]; i = bo->surface_reg; goto out; } diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index 11b21d605584..0d1283cdc8fb 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -802,7 +802,7 @@ static int radeon_mm_vram_dump_table_show(struct seq_file *m, void *unused) TTM_PL_VRAM); struct drm_printer p = drm_seq_file_printer(m); - man->func->debug(man, &p); + ttm_resource_manager_debug(man, &p); return 0; } @@ -820,7 +820,7 @@ static int radeon_mm_gtt_dump_table_show(struct seq_file *m, void *unused) TTM_PL_TT); struct drm_printer p = drm_seq_file_printer(m); - man->func->debug(man, &p); + ttm_resource_manager_debug(man, &p); return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c index 377f9cdb5b53..0558d928d98d 100644 --- a/drivers/gpu/drm/radeon/radeon_uvd.c +++ b/drivers/gpu/drm/radeon/radeon_uvd.c @@ -497,6 +497,7 @@ static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo, handle = msg[2]; if (handle == 0) { + radeon_bo_kunmap(bo); DRM_ERROR("Invalid UVD handle!\n"); return -EINVAL; } @@ -559,12 +560,10 @@ static int radeon_uvd_cs_msg(struct radeon_cs_parser *p, struct radeon_bo *bo, return 0; default: - DRM_ERROR("Illegal UVD message type (%d)!\n", msg_type); - return -EINVAL; } - BUG(); + radeon_bo_kunmap(bo); return -EINVAL; } diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c index e592e57be1bb..38796af4fadd 100644 --- a/drivers/gpu/drm/radeon/rv770.c +++ b/drivers/gpu/drm/radeon/rv770.c @@ -1894,8 +1894,8 @@ int rv770_suspend(struct radeon_device *rdev) radeon_pm_suspend(rdev); radeon_audio_fini(rdev); if (rdev->has_uvd) { - uvd_v1_0_fini(rdev); radeon_uvd_suspend(rdev); + uvd_v1_0_fini(rdev); } r700_cp_stop(rdev); r600_dma_stop(rdev); diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 013e44ed0f39..8d5e4b25609d 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -6800,8 +6800,8 @@ int si_suspend(struct radeon_device *rdev) si_cp_enable(rdev, false); cayman_dma_stop(rdev); if (rdev->has_uvd) { - uvd_v1_0_fini(rdev); radeon_uvd_suspend(rdev); + uvd_v1_0_fini(rdev); } if (rdev->has_vce) radeon_vce_suspend(rdev); diff --git a/drivers/gpu/drm/rockchip/Kconfig b/drivers/gpu/drm/rockchip/Kconfig index 9f1ecefc3933..fa5cfda4e90e 100644 --- a/drivers/gpu/drm/rockchip/Kconfig +++ b/drivers/gpu/drm/rockchip/Kconfig @@ -2,11 +2,13 @@ config DRM_ROCKCHIP tristate "DRM Support for Rockchip" depends on DRM && ROCKCHIP_IOMMU + select DRM_DP_HELPER select DRM_GEM_CMA_HELPER select DRM_KMS_HELPER select DRM_PANEL select VIDEOMODE_HELPERS select DRM_ANALOGIX_DP if ROCKCHIP_ANALOGIX_DP + select DRM_DP_HELPER if ROCKCHIP_ANALOGIX_DP select DRM_DW_HDMI if ROCKCHIP_DW_HDMI select DRM_DW_MIPI_DSI if ROCKCHIP_DW_MIPI_DSI select GENERIC_PHY if ROCKCHIP_DW_MIPI_DSI diff --git a/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c b/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c index 8abb5ac26807..c82901d9a9cc 100644 --- a/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c +++ b/drivers/gpu/drm/rockchip/analogix_dp-rockchip.c @@ -22,7 +22,7 @@ #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> #include <drm/bridge/analogix_dp.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> #include <drm/drm_probe_helper.h> @@ -117,7 +117,7 @@ static int rockchip_dp_get_modes(struct analogix_dp_plat_data *plat_data, { struct drm_display_info *di = &connector->display_info; /* VOP couldn't output YUV video format for eDP rightly */ - u32 mask = DRM_COLOR_FORMAT_YCRCB444 | DRM_COLOR_FORMAT_YCRCB422; + u32 mask = DRM_COLOR_FORMAT_YCBCR444 | DRM_COLOR_FORMAT_YCBCR422; if ((di->color_formats & mask)) { DRM_DEBUG_KMS("Swapping display color format from YUV to RGB\n"); diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.c b/drivers/gpu/drm/rockchip/cdn-dp-core.c index 16497c31d9f9..4740cc14beb8 100644 --- a/drivers/gpu/drm/rockchip/cdn-dp-core.c +++ b/drivers/gpu/drm/rockchip/cdn-dp-core.c @@ -16,7 +16,7 @@ #include <sound/hdmi-codec.h> #include <drm/drm_atomic_helper.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_of.h> #include <drm/drm_probe_helper.h> diff --git a/drivers/gpu/drm/rockchip/cdn-dp-core.h b/drivers/gpu/drm/rockchip/cdn-dp-core.h index 81ac9b658a70..0d044146f4e9 100644 --- a/drivers/gpu/drm/rockchip/cdn-dp-core.h +++ b/drivers/gpu/drm/rockchip/cdn-dp-core.h @@ -7,7 +7,7 @@ #ifndef _CDN_DP_CORE_H #define _CDN_DP_CORE_H -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_panel.h> #include <drm/drm_probe_helper.h> diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c index 63eb73b624aa..985584147da1 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.c @@ -510,7 +510,7 @@ err_free_rk_obj: return ERR_PTR(ret); } -int rockchip_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) +int rockchip_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map) { struct rockchip_gem_object *rk_obj = to_rockchip_obj(obj); @@ -519,18 +519,19 @@ int rockchip_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) pgprot_writecombine(PAGE_KERNEL)); if (!vaddr) return -ENOMEM; - dma_buf_map_set_vaddr(map, vaddr); + iosys_map_set_vaddr(map, vaddr); return 0; } if (rk_obj->dma_attrs & DMA_ATTR_NO_KERNEL_MAPPING) return -ENOMEM; - dma_buf_map_set_vaddr(map, rk_obj->kvaddr); + iosys_map_set_vaddr(map, rk_obj->kvaddr); return 0; } -void rockchip_gem_prime_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map) +void rockchip_gem_prime_vunmap(struct drm_gem_object *obj, + struct iosys_map *map) { struct rockchip_gem_object *rk_obj = to_rockchip_obj(obj); diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_gem.h b/drivers/gpu/drm/rockchip/rockchip_drm_gem.h index 47c1861eece0..72f59ac6d258 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_gem.h +++ b/drivers/gpu/drm/rockchip/rockchip_drm_gem.h @@ -31,8 +31,9 @@ struct drm_gem_object * rockchip_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sg); -int rockchip_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map); -void rockchip_gem_prime_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map); +int rockchip_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map); +void rockchip_gem_prime_vunmap(struct drm_gem_object *obj, + struct iosys_map *map); struct rockchip_gem_object * rockchip_gem_create_object(struct drm_device *drm, unsigned int size, diff --git a/drivers/gpu/drm/rockchip/rockchip_lvds.c b/drivers/gpu/drm/rockchip/rockchip_lvds.c index be74c87a8be4..0b972418067e 100644 --- a/drivers/gpu/drm/rockchip/rockchip_lvds.c +++ b/drivers/gpu/drm/rockchip/rockchip_lvds.c @@ -20,7 +20,7 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> #include <drm/drm_bridge_connector.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> #include <drm/drm_probe_helper.h> diff --git a/drivers/gpu/drm/rockchip/rockchip_rgb.c b/drivers/gpu/drm/rockchip/rockchip_rgb.c index 09be9678f2bd..2494b079489d 100644 --- a/drivers/gpu/drm/rockchip/rockchip_rgb.c +++ b/drivers/gpu/drm/rockchip/rockchip_rgb.c @@ -11,7 +11,7 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_bridge.h> #include <drm/drm_bridge_connector.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_of.h> #include <drm/drm_panel.h> #include <drm/drm_probe_helper.h> diff --git a/drivers/gpu/drm/selftests/test-drm_dp_mst_helper.c b/drivers/gpu/drm/selftests/test-drm_dp_mst_helper.c index 6b4759ed6bfd..fc1deb1231a2 100644 --- a/drivers/gpu/drm/selftests/test-drm_dp_mst_helper.c +++ b/drivers/gpu/drm/selftests/test-drm_dp_mst_helper.c @@ -7,10 +7,10 @@ #include <linux/random.h> -#include <drm/drm_dp_mst_helper.h> +#include <drm/dp/drm_dp_mst_helper.h> #include <drm/drm_print.h> -#include "../drm_dp_mst_topology_internal.h" +#include "../dp/drm_dp_mst_topology_internal.h" #include "test-drm_modeset_common.h" int igt_dp_mst_calc_pbn_mode(void *ignored) @@ -131,8 +131,10 @@ sideband_msg_req_encode_decode(struct drm_dp_sideband_msg_req_body *in) return false; txmsg = kzalloc(sizeof(*txmsg), GFP_KERNEL); - if (!txmsg) + if (!txmsg) { + kfree(out); return false; + } drm_dp_encode_sideband_req(in, txmsg); ret = drm_dp_decode_sideband_req(txmsg, out); diff --git a/drivers/gpu/drm/selftests/test-drm_plane_helper.c b/drivers/gpu/drm/selftests/test-drm_plane_helper.c index 0a9553f51796..ceebeede55ea 100644 --- a/drivers/gpu/drm/selftests/test-drm_plane_helper.c +++ b/drivers/gpu/drm/selftests/test-drm_plane_helper.c @@ -87,11 +87,15 @@ int igt_check_plane_state(void *ignored) DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, }; + struct drm_plane plane = { + .dev = NULL + }; struct drm_framebuffer fb = { .width = 2048, .height = 2048 }; struct drm_plane_state plane_state = { + .plane = &plane, .crtc = ZERO_SIZE_PTR, .fb = &fb, .rotation = DRM_MODE_ROTATE_0 diff --git a/drivers/gpu/drm/stm/drv.c b/drivers/gpu/drm/stm/drv.c index 222869b232ae..9f441aadf2d5 100644 --- a/drivers/gpu/drm/stm/drv.c +++ b/drivers/gpu/drm/stm/drv.c @@ -14,6 +14,7 @@ #include <linux/of_platform.h> #include <linux/pm_runtime.h> +#include <drm/drm_aperture.h> #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_drv.h> @@ -183,6 +184,10 @@ static int stm_drm_platform_probe(struct platform_device *pdev) DRM_DEBUG("%s\n", __func__); + ret = drm_aperture_remove_framebuffers(false, &drv_driver); + if (ret) + return ret; + dma_set_coherent_mask(dev, DMA_BIT_MASK(32)); ddev = drm_dev_alloc(&drv_driver, dev); diff --git a/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c b/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c index 32cb41b2202f..89897d5f5c72 100644 --- a/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c +++ b/drivers/gpu/drm/stm/dw_mipi_dsi-stm.c @@ -247,14 +247,6 @@ dw_mipi_dsi_get_lane_mbps(void *priv_data, const struct drm_display_mode *mode, int ret, bpp; u32 val; - /* Update lane capabilities according to hw version */ - dsi->lane_min_kbps = LANE_MIN_KBPS; - dsi->lane_max_kbps = LANE_MAX_KBPS; - if (dsi->hw_version == HWVER_131) { - dsi->lane_min_kbps *= 2; - dsi->lane_max_kbps *= 2; - } - pll_in_khz = (unsigned int)(clk_get_rate(dsi->pllref_clk) / 1000); /* Compute requested pll out */ @@ -330,6 +322,103 @@ dw_mipi_dsi_phy_get_timing(void *priv_data, unsigned int lane_mbps, return 0; } +#define CLK_TOLERANCE_HZ 50 + +static enum drm_mode_status +dw_mipi_dsi_stm_mode_valid(void *priv_data, + const struct drm_display_mode *mode, + unsigned long mode_flags, u32 lanes, u32 format) +{ + struct dw_mipi_dsi_stm *dsi = priv_data; + unsigned int idf, ndiv, odf, pll_in_khz, pll_out_khz; + int ret, bpp; + + bpp = mipi_dsi_pixel_format_to_bpp(format); + if (bpp < 0) + return MODE_BAD; + + /* Compute requested pll out */ + pll_out_khz = mode->clock * bpp / lanes; + + if (pll_out_khz > dsi->lane_max_kbps) + return MODE_CLOCK_HIGH; + + if (mode_flags & MIPI_DSI_MODE_VIDEO_BURST) { + /* Add 20% to pll out to be higher than pixel bw */ + pll_out_khz = (pll_out_khz * 12) / 10; + } else { + if (pll_out_khz < dsi->lane_min_kbps) + return MODE_CLOCK_LOW; + } + + /* Compute best pll parameters */ + idf = 0; + ndiv = 0; + odf = 0; + pll_in_khz = clk_get_rate(dsi->pllref_clk) / 1000; + ret = dsi_pll_get_params(dsi, pll_in_khz, pll_out_khz, &idf, &ndiv, &odf); + if (ret) { + DRM_WARN("Warning dsi_pll_get_params(): bad params\n"); + return MODE_ERROR; + } + + if (!(mode_flags & MIPI_DSI_MODE_VIDEO_BURST)) { + unsigned int px_clock_hz, target_px_clock_hz, lane_mbps; + int dsi_short_packet_size_px, hfp, hsync, hbp, delay_to_lp; + struct dw_mipi_dsi_dphy_timing dphy_timing; + + /* Get the adjusted pll out value */ + pll_out_khz = dsi_pll_get_clkout_khz(pll_in_khz, idf, ndiv, odf); + + px_clock_hz = DIV_ROUND_CLOSEST_ULL(1000ULL * pll_out_khz * lanes, bpp); + target_px_clock_hz = mode->clock * 1000; + /* + * Filter modes according to the clock value, particularly useful for + * hdmi modes that require precise pixel clocks. + */ + if (px_clock_hz < target_px_clock_hz - CLK_TOLERANCE_HZ || + px_clock_hz > target_px_clock_hz + CLK_TOLERANCE_HZ) + return MODE_CLOCK_RANGE; + + /* sync packets are codes as DSI short packets (4 bytes) */ + dsi_short_packet_size_px = DIV_ROUND_UP(4 * BITS_PER_BYTE, bpp); + + hfp = mode->hsync_start - mode->hdisplay; + hsync = mode->hsync_end - mode->hsync_start; + hbp = mode->htotal - mode->hsync_end; + + /* hsync must be longer than 4 bytes HSS packets */ + if (hsync < dsi_short_packet_size_px) + return MODE_HSYNC_NARROW; + + if (mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE) { + /* HBP must be longer than 4 bytes HSE packets */ + if (hbp < dsi_short_packet_size_px) + return MODE_HSYNC_NARROW; + hbp -= dsi_short_packet_size_px; + } else { + /* With sync events HBP extends in the hsync */ + hbp += hsync - dsi_short_packet_size_px; + } + + lane_mbps = pll_out_khz / 1000; + ret = dw_mipi_dsi_phy_get_timing(priv_data, lane_mbps, &dphy_timing); + if (ret) + return MODE_ERROR; + /* + * In non-burst mode DSI has to enter in LP during HFP + * (horizontal front porch) or HBP (horizontal back porch) to + * resync with LTDC pixel clock. + */ + delay_to_lp = DIV_ROUND_UP((dphy_timing.data_hs2lp + dphy_timing.data_lp2hs) * + lanes * BITS_PER_BYTE, bpp); + if (hfp < delay_to_lp && hbp < delay_to_lp) + return MODE_HSYNC; + } + + return MODE_OK; +} + static const struct dw_mipi_dsi_phy_ops dw_mipi_dsi_stm_phy_ops = { .init = dw_mipi_dsi_phy_init, .power_on = dw_mipi_dsi_phy_power_on, @@ -340,6 +429,7 @@ static const struct dw_mipi_dsi_phy_ops dw_mipi_dsi_stm_phy_ops = { static struct dw_mipi_dsi_plat_data dw_mipi_dsi_stm_plat_data = { .max_data_lanes = 2, + .mode_valid = dw_mipi_dsi_stm_mode_valid, .phy_ops = &dw_mipi_dsi_stm_phy_ops, }; @@ -417,6 +507,14 @@ static int dw_mipi_dsi_stm_probe(struct platform_device *pdev) goto err_dsi_probe; } + /* set lane capabilities according to hw version */ + dsi->lane_min_kbps = LANE_MIN_KBPS; + dsi->lane_max_kbps = LANE_MAX_KBPS; + if (dsi->hw_version == HWVER_131) { + dsi->lane_min_kbps *= 2; + dsi->lane_max_kbps *= 2; + } + dw_mipi_dsi_stm_plat_data.base = dsi->base; dw_mipi_dsi_stm_plat_data.priv_data = dsi; diff --git a/drivers/gpu/drm/stm/ltdc.c b/drivers/gpu/drm/stm/ltdc.c index dbdee954692a..5eeb32c9c9ce 100644 --- a/drivers/gpu/drm/stm/ltdc.c +++ b/drivers/gpu/drm/stm/ltdc.c @@ -18,6 +18,7 @@ #include <linux/pinctrl/consumer.h> #include <linux/platform_device.h> #include <linux/pm_runtime.h> +#include <linux/regmap.h> #include <linux/reset.h> #include <drm/drm_atomic.h> @@ -46,15 +47,15 @@ #define HWVER_10200 0x010200 #define HWVER_10300 0x010300 #define HWVER_20101 0x020101 +#define HWVER_40100 0x040100 /* * The address of some registers depends on the HW version: such registers have - * an extra offset specified with reg_ofs. + * an extra offset specified with layer_ofs. */ -#define REG_OFS_NONE 0 -#define REG_OFS_4 4 /* Insertion of "Layer Conf. 2" reg */ -#define REG_OFS (ldev->caps.reg_ofs) -#define LAY_OFS 0x80 /* Register Offset between 2 layers */ +#define LAY_OFS_0 0x80 +#define LAY_OFS_1 0x100 +#define LAY_OFS (ldev->caps.layer_ofs) /* Global register offsets */ #define LTDC_IDR 0x0000 /* IDentification */ @@ -75,29 +76,35 @@ #define LTDC_LIPCR 0x0040 /* Line Interrupt Position Conf. */ #define LTDC_CPSR 0x0044 /* Current Position Status */ #define LTDC_CDSR 0x0048 /* Current Display Status */ +#define LTDC_EDCR 0x0060 /* External Display Control */ +#define LTDC_FUT 0x0090 /* Fifo underrun Threshold */ /* Layer register offsets */ -#define LTDC_L1LC1R (0x80) /* L1 Layer Configuration 1 */ -#define LTDC_L1LC2R (0x84) /* L1 Layer Configuration 2 */ -#define LTDC_L1CR (0x84 + REG_OFS)/* L1 Control */ -#define LTDC_L1WHPCR (0x88 + REG_OFS)/* L1 Window Hor Position Config */ -#define LTDC_L1WVPCR (0x8C + REG_OFS)/* L1 Window Vert Position Config */ -#define LTDC_L1CKCR (0x90 + REG_OFS)/* L1 Color Keying Configuration */ -#define LTDC_L1PFCR (0x94 + REG_OFS)/* L1 Pixel Format Configuration */ -#define LTDC_L1CACR (0x98 + REG_OFS)/* L1 Constant Alpha Config */ -#define LTDC_L1DCCR (0x9C + REG_OFS)/* L1 Default Color Configuration */ -#define LTDC_L1BFCR (0xA0 + REG_OFS)/* L1 Blend Factors Configuration */ -#define LTDC_L1FBBCR (0xA4 + REG_OFS)/* L1 FrameBuffer Bus Control */ -#define LTDC_L1AFBCR (0xA8 + REG_OFS)/* L1 AuxFB Control */ -#define LTDC_L1CFBAR (0xAC + REG_OFS)/* L1 Color FrameBuffer Address */ -#define LTDC_L1CFBLR (0xB0 + REG_OFS)/* L1 Color FrameBuffer Length */ -#define LTDC_L1CFBLNR (0xB4 + REG_OFS)/* L1 Color FrameBuffer Line Nb */ -#define LTDC_L1AFBAR (0xB8 + REG_OFS)/* L1 AuxFB Address */ -#define LTDC_L1AFBLR (0xBC + REG_OFS)/* L1 AuxFB Length */ -#define LTDC_L1AFBLNR (0xC0 + REG_OFS)/* L1 AuxFB Line Number */ -#define LTDC_L1CLUTWR (0xC4 + REG_OFS)/* L1 CLUT Write */ -#define LTDC_L1YS1R (0xE0 + REG_OFS)/* L1 YCbCr Scale 1 */ -#define LTDC_L1YS2R (0xE4 + REG_OFS)/* L1 YCbCr Scale 2 */ +#define LTDC_L1C0R (ldev->caps.layer_regs[0]) /* L1 configuration 0 */ +#define LTDC_L1C1R (ldev->caps.layer_regs[1]) /* L1 configuration 1 */ +#define LTDC_L1RCR (ldev->caps.layer_regs[2]) /* L1 reload control */ +#define LTDC_L1CR (ldev->caps.layer_regs[3]) /* L1 control register */ +#define LTDC_L1WHPCR (ldev->caps.layer_regs[4]) /* L1 window horizontal position configuration */ +#define LTDC_L1WVPCR (ldev->caps.layer_regs[5]) /* L1 window vertical position configuration */ +#define LTDC_L1CKCR (ldev->caps.layer_regs[6]) /* L1 color keying configuration */ +#define LTDC_L1PFCR (ldev->caps.layer_regs[7]) /* L1 pixel format configuration */ +#define LTDC_L1CACR (ldev->caps.layer_regs[8]) /* L1 constant alpha configuration */ +#define LTDC_L1DCCR (ldev->caps.layer_regs[9]) /* L1 default color configuration */ +#define LTDC_L1BFCR (ldev->caps.layer_regs[10]) /* L1 blending factors configuration */ +#define LTDC_L1BLCR (ldev->caps.layer_regs[11]) /* L1 burst length configuration */ +#define LTDC_L1PCR (ldev->caps.layer_regs[12]) /* L1 planar configuration */ +#define LTDC_L1CFBAR (ldev->caps.layer_regs[13]) /* L1 color frame buffer address */ +#define LTDC_L1CFBLR (ldev->caps.layer_regs[14]) /* L1 color frame buffer length */ +#define LTDC_L1CFBLNR (ldev->caps.layer_regs[15]) /* L1 color frame buffer line number */ +#define LTDC_L1AFBA0R (ldev->caps.layer_regs[16]) /* L1 auxiliary frame buffer address 0 */ +#define LTDC_L1AFBA1R (ldev->caps.layer_regs[17]) /* L1 auxiliary frame buffer address 1 */ +#define LTDC_L1AFBLR (ldev->caps.layer_regs[18]) /* L1 auxiliary frame buffer length */ +#define LTDC_L1AFBLNR (ldev->caps.layer_regs[19]) /* L1 auxiliary frame buffer line number */ +#define LTDC_L1CLUTWR (ldev->caps.layer_regs[20]) /* L1 CLUT write */ +#define LTDC_L1CYR0R (ldev->caps.layer_regs[21]) /* L1 Conversion YCbCr RGB 0 */ +#define LTDC_L1CYR1R (ldev->caps.layer_regs[22]) /* L1 Conversion YCbCr RGB 1 */ +#define LTDC_L1FPF0R (ldev->caps.layer_regs[23]) /* L1 Flexible Pixel Format 0 */ +#define LTDC_L1FPF1R (ldev->caps.layer_regs[24]) /* L1 Flexible Pixel Format 1 */ /* Bit definitions */ #define SSCR_VSH GENMASK(10, 0) /* Vertical Synchronization Height */ @@ -164,6 +171,10 @@ #define ISR_TERRIF BIT(2) /* Transfer ERRor Interrupt Flag */ #define ISR_RRIF BIT(3) /* Register Reload Interrupt Flag */ +#define EDCR_OCYEN BIT(25) /* Output Conversion to YCbCr 422: ENable */ +#define EDCR_OCYSEL BIT(26) /* Output Conversion to YCbCr 422: SELection of the CCIR */ +#define EDCR_OCYCO BIT(27) /* Output Conversion to YCbCr 422: Chrominance Order */ + #define LXCR_LEN BIT(0) /* Layer ENable */ #define LXCR_COLKEN BIT(1) /* Color Keying Enable */ #define LXCR_CLUTEN BIT(4) /* Color Look-Up Table ENable */ @@ -175,6 +186,7 @@ #define LXWVPCR_WVSPPOS GENMASK(26, 16) /* Window Vertical StoP POSition */ #define LXPFCR_PF GENMASK(2, 0) /* Pixel Format */ +#define PF_FLEXIBLE 0x7 /* Flexible Pixel Format selected */ #define LXCACR_CONSTA GENMASK(7, 0) /* CONSTant Alpha */ @@ -186,6 +198,25 @@ #define LXCFBLNR_CFBLN GENMASK(10, 0) /* Color Frame Buffer Line Number */ +#define LXCR_C1R_YIA BIT(0) /* Ycbcr 422 Interleaved Ability */ +#define LXCR_C1R_YSPA BIT(1) /* Ycbcr 420 Semi-Planar Ability */ +#define LXCR_C1R_YFPA BIT(2) /* Ycbcr 420 Full-Planar Ability */ +#define LXCR_C1R_SCA BIT(31) /* SCaling Ability*/ + +#define LxPCR_YREN BIT(9) /* Y Rescale Enable for the color dynamic range */ +#define LxPCR_OF BIT(8) /* Odd pixel First */ +#define LxPCR_CBF BIT(7) /* CB component First */ +#define LxPCR_YF BIT(6) /* Y component First */ +#define LxPCR_YCM GENMASK(5, 4) /* Ycbcr Conversion Mode */ +#define YCM_I 0x0 /* Interleaved 422 */ +#define YCM_SP 0x1 /* Semi-Planar 420 */ +#define YCM_FP 0x2 /* Full-Planar 420 */ +#define LxPCR_YCEN BIT(3) /* YCbCr-to-RGB Conversion Enable */ + +#define LXRCR_IMR BIT(0) /* IMmediate Reload */ +#define LXRCR_VBR BIT(1) /* Vertical Blanking Reload */ +#define LXRCR_GRMSK BIT(2) /* Global (centralized) Reload MaSKed */ + #define CLUT_SIZE 256 #define CONSTA_MAX 0xFF /* CONSTant Alpha MAX= 1.0 */ @@ -201,8 +232,12 @@ enum ltdc_pix_fmt { /* RGB formats */ PF_ARGB8888, /* ARGB [32 bits] */ PF_RGBA8888, /* RGBA [32 bits] */ + PF_ABGR8888, /* ABGR [32 bits] */ + PF_BGRA8888, /* BGRA [32 bits] */ PF_RGB888, /* RGB [24 bits] */ + PF_BGR888, /* BGR [24 bits] */ PF_RGB565, /* RGB [16 bits] */ + PF_BGR565, /* BGR [16 bits] */ PF_ARGB1555, /* ARGB A:1 bit RGB:15 bits [16 bits] */ PF_ARGB4444, /* ARGB A:4 bits R/G/B: 4 bits each [16 bits] */ /* Indexed formats */ @@ -234,36 +269,198 @@ static const enum ltdc_pix_fmt ltdc_pix_fmt_a1[NB_PF] = { PF_ARGB4444 /* 0x07 */ }; -static const u64 ltdc_format_modifiers[] = { - DRM_FORMAT_MOD_LINEAR, - DRM_FORMAT_MOD_INVALID +static const enum ltdc_pix_fmt ltdc_pix_fmt_a2[NB_PF] = { + PF_ARGB8888, /* 0x00 */ + PF_ABGR8888, /* 0x01 */ + PF_RGBA8888, /* 0x02 */ + PF_BGRA8888, /* 0x03 */ + PF_RGB565, /* 0x04 */ + PF_BGR565, /* 0x05 */ + PF_RGB888, /* 0x06 */ + PF_NONE /* 0x07 */ }; -static inline u32 reg_read(void __iomem *base, u32 reg) -{ - return readl_relaxed(base + reg); -} +static const u32 ltdc_drm_fmt_a0[] = { + DRM_FORMAT_ARGB8888, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_RGB888, + DRM_FORMAT_RGB565, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_ARGB4444, + DRM_FORMAT_XRGB4444, + DRM_FORMAT_C8 +}; -static inline void reg_write(void __iomem *base, u32 reg, u32 val) -{ - writel_relaxed(val, base + reg); -} +static const u32 ltdc_drm_fmt_a1[] = { + DRM_FORMAT_ARGB8888, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_RGB888, + DRM_FORMAT_RGB565, + DRM_FORMAT_RGBA8888, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_ARGB4444, + DRM_FORMAT_XRGB4444, + DRM_FORMAT_C8 +}; -static inline void reg_set(void __iomem *base, u32 reg, u32 mask) -{ - reg_write(base, reg, reg_read(base, reg) | mask); -} +static const u32 ltdc_drm_fmt_a2[] = { + DRM_FORMAT_ARGB8888, + DRM_FORMAT_XRGB8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_XBGR8888, + DRM_FORMAT_RGBA8888, + DRM_FORMAT_RGBX8888, + DRM_FORMAT_BGRA8888, + DRM_FORMAT_BGRX8888, + DRM_FORMAT_RGB565, + DRM_FORMAT_BGR565, + DRM_FORMAT_RGB888, + DRM_FORMAT_BGR888, + DRM_FORMAT_ARGB1555, + DRM_FORMAT_XRGB1555, + DRM_FORMAT_ARGB4444, + DRM_FORMAT_XRGB4444, + DRM_FORMAT_C8 +}; -static inline void reg_clear(void __iomem *base, u32 reg, u32 mask) -{ - reg_write(base, reg, reg_read(base, reg) & ~mask); -} +static const u32 ltdc_drm_fmt_ycbcr_cp[] = { + DRM_FORMAT_YUYV, + DRM_FORMAT_YVYU, + DRM_FORMAT_UYVY, + DRM_FORMAT_VYUY +}; -static inline void reg_update_bits(void __iomem *base, u32 reg, u32 mask, - u32 val) -{ - reg_write(base, reg, (reg_read(base, reg) & ~mask) | val); -} +static const u32 ltdc_drm_fmt_ycbcr_sp[] = { + DRM_FORMAT_NV12, + DRM_FORMAT_NV21 +}; + +static const u32 ltdc_drm_fmt_ycbcr_fp[] = { + DRM_FORMAT_YUV420, + DRM_FORMAT_YVU420 +}; + +/* Layer register offsets */ +static const u32 ltdc_layer_regs_a0[] = { + 0x80, /* L1 configuration 0 */ + 0x00, /* not available */ + 0x00, /* not available */ + 0x84, /* L1 control register */ + 0x88, /* L1 window horizontal position configuration */ + 0x8c, /* L1 window vertical position configuration */ + 0x90, /* L1 color keying configuration */ + 0x94, /* L1 pixel format configuration */ + 0x98, /* L1 constant alpha configuration */ + 0x9c, /* L1 default color configuration */ + 0xa0, /* L1 blending factors configuration */ + 0x00, /* not available */ + 0x00, /* not available */ + 0xac, /* L1 color frame buffer address */ + 0xb0, /* L1 color frame buffer length */ + 0xb4, /* L1 color frame buffer line number */ + 0x00, /* not available */ + 0x00, /* not available */ + 0x00, /* not available */ + 0x00, /* not available */ + 0xc4, /* L1 CLUT write */ + 0x00, /* not available */ + 0x00, /* not available */ + 0x00, /* not available */ + 0x00 /* not available */ +}; + +static const u32 ltdc_layer_regs_a1[] = { + 0x80, /* L1 configuration 0 */ + 0x84, /* L1 configuration 1 */ + 0x00, /* L1 reload control */ + 0x88, /* L1 control register */ + 0x8c, /* L1 window horizontal position configuration */ + 0x90, /* L1 window vertical position configuration */ + 0x94, /* L1 color keying configuration */ + 0x98, /* L1 pixel format configuration */ + 0x9c, /* L1 constant alpha configuration */ + 0xa0, /* L1 default color configuration */ + 0xa4, /* L1 blending factors configuration */ + 0xa8, /* L1 burst length configuration */ + 0x00, /* not available */ + 0xac, /* L1 color frame buffer address */ + 0xb0, /* L1 color frame buffer length */ + 0xb4, /* L1 color frame buffer line number */ + 0xb8, /* L1 auxiliary frame buffer address 0 */ + 0xbc, /* L1 auxiliary frame buffer address 1 */ + 0xc0, /* L1 auxiliary frame buffer length */ + 0xc4, /* L1 auxiliary frame buffer line number */ + 0xc8, /* L1 CLUT write */ + 0x00, /* not available */ + 0x00, /* not available */ + 0x00, /* not available */ + 0x00 /* not available */ +}; + +static const u32 ltdc_layer_regs_a2[] = { + 0x100, /* L1 configuration 0 */ + 0x104, /* L1 configuration 1 */ + 0x108, /* L1 reload control */ + 0x10c, /* L1 control register */ + 0x110, /* L1 window horizontal position configuration */ + 0x114, /* L1 window vertical position configuration */ + 0x118, /* L1 color keying configuration */ + 0x11c, /* L1 pixel format configuration */ + 0x120, /* L1 constant alpha configuration */ + 0x124, /* L1 default color configuration */ + 0x128, /* L1 blending factors configuration */ + 0x12c, /* L1 burst length configuration */ + 0x130, /* L1 planar configuration */ + 0x134, /* L1 color frame buffer address */ + 0x138, /* L1 color frame buffer length */ + 0x13c, /* L1 color frame buffer line number */ + 0x140, /* L1 auxiliary frame buffer address 0 */ + 0x144, /* L1 auxiliary frame buffer address 1 */ + 0x148, /* L1 auxiliary frame buffer length */ + 0x14c, /* L1 auxiliary frame buffer line number */ + 0x150, /* L1 CLUT write */ + 0x16c, /* L1 Conversion YCbCr RGB 0 */ + 0x170, /* L1 Conversion YCbCr RGB 1 */ + 0x174, /* L1 Flexible Pixel Format 0 */ + 0x178 /* L1 Flexible Pixel Format 1 */ +}; + +static const u64 ltdc_format_modifiers[] = { + DRM_FORMAT_MOD_LINEAR, + DRM_FORMAT_MOD_INVALID +}; + +static const struct regmap_config stm32_ltdc_regmap_cfg = { + .reg_bits = 32, + .val_bits = 32, + .reg_stride = sizeof(u32), + .max_register = 0x400, + .use_relaxed_mmio = true, + .cache_type = REGCACHE_NONE, +}; + +static const u32 ltdc_ycbcr2rgb_coeffs[DRM_COLOR_ENCODING_MAX][DRM_COLOR_RANGE_MAX][2] = { + [DRM_COLOR_YCBCR_BT601][DRM_COLOR_YCBCR_LIMITED_RANGE] = { + 0x02040199, /* (b_cb = 516 / r_cr = 409) */ + 0x006400D0 /* (g_cb = 100 / g_cr = 208) */ + }, + [DRM_COLOR_YCBCR_BT601][DRM_COLOR_YCBCR_FULL_RANGE] = { + 0x01C60167, /* (b_cb = 454 / r_cr = 359) */ + 0x005800B7 /* (g_cb = 88 / g_cr = 183) */ + }, + [DRM_COLOR_YCBCR_BT709][DRM_COLOR_YCBCR_LIMITED_RANGE] = { + 0x021D01CB, /* (b_cb = 541 / r_cr = 459) */ + 0x00370089 /* (g_cb = 55 / g_cr = 137) */ + }, + [DRM_COLOR_YCBCR_BT709][DRM_COLOR_YCBCR_FULL_RANGE] = { + 0x01DB0193, /* (b_cb = 475 / r_cr = 403) */ + 0x00300078 /* (g_cb = 48 / g_cr = 120) */ + } + /* BT2020 not supported */ +}; static inline struct ltdc_device *crtc_to_ltdc(struct drm_crtc *crtc) { @@ -289,16 +486,30 @@ static inline enum ltdc_pix_fmt to_ltdc_pixelformat(u32 drm_fmt) case DRM_FORMAT_XRGB8888: pf = PF_ARGB8888; break; + case DRM_FORMAT_ABGR8888: + case DRM_FORMAT_XBGR8888: + pf = PF_ABGR8888; + break; case DRM_FORMAT_RGBA8888: case DRM_FORMAT_RGBX8888: pf = PF_RGBA8888; break; + case DRM_FORMAT_BGRA8888: + case DRM_FORMAT_BGRX8888: + pf = PF_BGRA8888; + break; case DRM_FORMAT_RGB888: pf = PF_RGB888; break; + case DRM_FORMAT_BGR888: + pf = PF_BGR888; + break; case DRM_FORMAT_RGB565: pf = PF_RGB565; break; + case DRM_FORMAT_BGR565: + pf = PF_BGR565; + break; case DRM_FORMAT_ARGB1555: case DRM_FORMAT_XRGB1555: pf = PF_ARGB1555; @@ -319,49 +530,138 @@ static inline enum ltdc_pix_fmt to_ltdc_pixelformat(u32 drm_fmt) return pf; } -static inline u32 to_drm_pixelformat(enum ltdc_pix_fmt pf) +static inline u32 ltdc_set_flexible_pixel_format(struct drm_plane *plane, enum ltdc_pix_fmt pix_fmt) { - switch (pf) { - case PF_ARGB8888: - return DRM_FORMAT_ARGB8888; - case PF_RGBA8888: - return DRM_FORMAT_RGBA8888; - case PF_RGB888: - return DRM_FORMAT_RGB888; - case PF_RGB565: - return DRM_FORMAT_RGB565; + struct ltdc_device *ldev = plane_to_ltdc(plane); + u32 lofs = plane->index * LAY_OFS, ret = PF_FLEXIBLE; + int psize, alen, apos, rlen, rpos, glen, gpos, blen, bpos; + + switch (pix_fmt) { + case PF_BGR888: + psize = 3; + alen = 0; apos = 0; rlen = 8; rpos = 0; + glen = 8; gpos = 8; blen = 8; bpos = 16; + break; case PF_ARGB1555: - return DRM_FORMAT_ARGB1555; + psize = 2; + alen = 1; apos = 15; rlen = 5; rpos = 10; + glen = 5; gpos = 5; blen = 5; bpos = 0; + break; case PF_ARGB4444: - return DRM_FORMAT_ARGB4444; + psize = 2; + alen = 4; apos = 12; rlen = 4; rpos = 8; + glen = 4; gpos = 4; blen = 4; bpos = 0; + break; case PF_L8: - return DRM_FORMAT_C8; - case PF_AL44: /* No DRM support */ - case PF_AL88: /* No DRM support */ - case PF_NONE: + psize = 1; + alen = 0; apos = 0; rlen = 8; rpos = 0; + glen = 8; gpos = 0; blen = 8; bpos = 0; + break; + case PF_AL44: + psize = 1; + alen = 4; apos = 4; rlen = 4; rpos = 0; + glen = 4; gpos = 0; blen = 4; bpos = 0; + break; + case PF_AL88: + psize = 2; + alen = 8; apos = 8; rlen = 8; rpos = 0; + glen = 8; gpos = 0; blen = 8; bpos = 0; + break; default: - return 0; + ret = NB_PF; /* error case, trace msg is handled by the caller */ + break; } + + if (ret == PF_FLEXIBLE) { + regmap_write(ldev->regmap, LTDC_L1FPF0R + lofs, + (rlen << 14) + (rpos << 9) + (alen << 5) + apos); + + regmap_write(ldev->regmap, LTDC_L1FPF1R + lofs, + (psize << 18) + (blen << 14) + (bpos << 9) + (glen << 5) + gpos); + } + + return ret; } -static inline u32 get_pixelformat_without_alpha(u32 drm) +/* + * All non-alpha color formats derived from native alpha color formats are + * either characterized by a FourCC format code + */ +static inline u32 is_xrgb(u32 drm) { - switch (drm) { - case DRM_FORMAT_ARGB4444: - return DRM_FORMAT_XRGB4444; - case DRM_FORMAT_RGBA4444: - return DRM_FORMAT_RGBX4444; - case DRM_FORMAT_ARGB1555: - return DRM_FORMAT_XRGB1555; - case DRM_FORMAT_RGBA5551: - return DRM_FORMAT_RGBX5551; - case DRM_FORMAT_ARGB8888: - return DRM_FORMAT_XRGB8888; - case DRM_FORMAT_RGBA8888: - return DRM_FORMAT_RGBX8888; + return ((drm & 0xFF) == 'X' || ((drm >> 8) & 0xFF) == 'X'); +} + +static inline void ltdc_set_ycbcr_config(struct drm_plane *plane, u32 drm_pix_fmt) +{ + struct ltdc_device *ldev = plane_to_ltdc(plane); + struct drm_plane_state *state = plane->state; + u32 lofs = plane->index * LAY_OFS; + u32 val; + + switch (drm_pix_fmt) { + case DRM_FORMAT_YUYV: + val = (YCM_I << 4) | LxPCR_YF | LxPCR_CBF; + break; + case DRM_FORMAT_YVYU: + val = (YCM_I << 4) | LxPCR_YF; + break; + case DRM_FORMAT_UYVY: + val = (YCM_I << 4) | LxPCR_CBF; + break; + case DRM_FORMAT_VYUY: + val = (YCM_I << 4); + break; + case DRM_FORMAT_NV12: + val = (YCM_SP << 4) | LxPCR_CBF; + break; + case DRM_FORMAT_NV21: + val = (YCM_SP << 4); + break; + case DRM_FORMAT_YUV420: + case DRM_FORMAT_YVU420: + val = (YCM_FP << 4); + break; default: - return 0; + /* RGB or not a YCbCr supported format */ + break; + } + + /* Enable limited range */ + if (state->color_range == DRM_COLOR_YCBCR_LIMITED_RANGE) + val |= LxPCR_YREN; + + /* enable ycbcr conversion */ + val |= LxPCR_YCEN; + + regmap_write(ldev->regmap, LTDC_L1PCR + lofs, val); +} + +static inline void ltdc_set_ycbcr_coeffs(struct drm_plane *plane) +{ + struct ltdc_device *ldev = plane_to_ltdc(plane); + struct drm_plane_state *state = plane->state; + enum drm_color_encoding enc = state->color_encoding; + enum drm_color_range ran = state->color_range; + u32 lofs = plane->index * LAY_OFS; + + if (enc != DRM_COLOR_YCBCR_BT601 && enc != DRM_COLOR_YCBCR_BT709) { + DRM_ERROR("color encoding %d not supported, use bt601 by default\n", enc); + /* set by default color encoding to DRM_COLOR_YCBCR_BT601 */ + enc = DRM_COLOR_YCBCR_BT601; + } + + if (ran != DRM_COLOR_YCBCR_LIMITED_RANGE && ran != DRM_COLOR_YCBCR_FULL_RANGE) { + DRM_ERROR("color range %d not supported, use limited range by default\n", ran); + /* set by default color range to DRM_COLOR_YCBCR_LIMITED_RANGE */ + ran = DRM_COLOR_YCBCR_LIMITED_RANGE; } + + DRM_DEBUG_DRIVER("Color encoding=%d, range=%d\n", enc, ran); + regmap_write(ldev->regmap, LTDC_L1CYR0R + lofs, + ltdc_ycbcr2rgb_coeffs[enc][ran][0]); + regmap_write(ldev->regmap, LTDC_L1CYR1R + lofs, + ltdc_ycbcr2rgb_coeffs[enc][ran][1]); } static irqreturn_t ltdc_irq_thread(int irq, void *arg) @@ -390,9 +690,13 @@ static irqreturn_t ltdc_irq(int irq, void *arg) struct drm_device *ddev = arg; struct ltdc_device *ldev = ddev->dev_private; - /* Read & Clear the interrupt status */ - ldev->irq_status = reg_read(ldev->regs, LTDC_ISR); - reg_write(ldev->regs, LTDC_ICR, ldev->irq_status); + /* + * Read & Clear the interrupt status + * In order to write / read registers in this critical section + * very quickly, the regmap functions are not used. + */ + ldev->irq_status = readl_relaxed(ldev->regs + LTDC_ISR); + writel_relaxed(ldev->irq_status, ldev->regs + LTDC_ICR); return IRQ_WAKE_THREAD; } @@ -416,7 +720,7 @@ static void ltdc_crtc_update_clut(struct drm_crtc *crtc) for (i = 0; i < CLUT_SIZE; i++, lut++) { val = ((lut->red << 8) & 0xff0000) | (lut->green & 0xff00) | (lut->blue >> 8) | (i << 24); - reg_write(ldev->regs, LTDC_L1CLUTWR, val); + regmap_write(ldev->regmap, LTDC_L1CLUTWR, val); } } @@ -431,13 +735,14 @@ static void ltdc_crtc_atomic_enable(struct drm_crtc *crtc, pm_runtime_get_sync(ddev->dev); /* Sets the background color value */ - reg_write(ldev->regs, LTDC_BCCR, BCCR_BCBLACK); + regmap_write(ldev->regmap, LTDC_BCCR, BCCR_BCBLACK); /* Enable IRQ */ - reg_set(ldev->regs, LTDC_IER, IER_RRIE | IER_FUIE | IER_TERRIE); + regmap_set_bits(ldev->regmap, LTDC_IER, IER_RRIE | IER_FUIE | IER_TERRIE); /* Commit shadow registers = update planes at next vblank */ - reg_set(ldev->regs, LTDC_SRCR, SRCR_VBR); + if (!ldev->caps.plane_reg_shadow) + regmap_set_bits(ldev->regmap, LTDC_SRCR, SRCR_VBR); drm_crtc_vblank_on(crtc); } @@ -453,10 +758,11 @@ static void ltdc_crtc_atomic_disable(struct drm_crtc *crtc, drm_crtc_vblank_off(crtc); /* disable IRQ */ - reg_clear(ldev->regs, LTDC_IER, IER_RRIE | IER_FUIE | IER_TERRIE); + regmap_clear_bits(ldev->regmap, LTDC_IER, IER_RRIE | IER_FUIE | IER_TERRIE); /* immediately commit disable of layers before switching off LTDC */ - reg_set(ldev->regs, LTDC_SRCR, SRCR_IMR); + if (!ldev->caps.plane_reg_shadow) + regmap_set_bits(ldev->regmap, LTDC_SRCR, SRCR_IMR); pm_runtime_put_sync(ddev->dev); } @@ -533,6 +839,7 @@ static void ltdc_crtc_mode_set_nofb(struct drm_crtc *crtc) struct drm_display_mode *mode = &crtc->state->adjusted_mode; u32 hsync, vsync, accum_hbp, accum_vbp, accum_act_w, accum_act_h; u32 total_width, total_height; + u32 bus_formats = MEDIA_BUS_FMT_RGB888_1X24; u32 bus_flags = 0; u32 val; int ret; @@ -558,8 +865,11 @@ static void ltdc_crtc_mode_set_nofb(struct drm_crtc *crtc) if (bridge && bridge->timings) bus_flags = bridge->timings->input_bus_flags; - else if (connector) + else if (connector) { bus_flags = connector->display_info.bus_flags; + if (connector->display_info.num_bus_formats) + bus_formats = connector->display_info.bus_formats[0]; + } if (!pm_runtime_active(ddev->dev)) { ret = pm_runtime_get_sync(ddev->dev); @@ -604,26 +914,56 @@ static void ltdc_crtc_mode_set_nofb(struct drm_crtc *crtc) if (bus_flags & DRM_BUS_FLAG_PIXDATA_DRIVE_NEGEDGE) val |= GCR_PCPOL; - reg_update_bits(ldev->regs, LTDC_GCR, - GCR_HSPOL | GCR_VSPOL | GCR_DEPOL | GCR_PCPOL, val); + regmap_update_bits(ldev->regmap, LTDC_GCR, + GCR_HSPOL | GCR_VSPOL | GCR_DEPOL | GCR_PCPOL, val); /* Set Synchronization size */ val = (hsync << 16) | vsync; - reg_update_bits(ldev->regs, LTDC_SSCR, SSCR_VSH | SSCR_HSW, val); + regmap_update_bits(ldev->regmap, LTDC_SSCR, SSCR_VSH | SSCR_HSW, val); /* Set Accumulated Back porch */ val = (accum_hbp << 16) | accum_vbp; - reg_update_bits(ldev->regs, LTDC_BPCR, BPCR_AVBP | BPCR_AHBP, val); + regmap_update_bits(ldev->regmap, LTDC_BPCR, BPCR_AVBP | BPCR_AHBP, val); /* Set Accumulated Active Width */ val = (accum_act_w << 16) | accum_act_h; - reg_update_bits(ldev->regs, LTDC_AWCR, AWCR_AAW | AWCR_AAH, val); + regmap_update_bits(ldev->regmap, LTDC_AWCR, AWCR_AAW | AWCR_AAH, val); /* Set total width & height */ val = (total_width << 16) | total_height; - reg_update_bits(ldev->regs, LTDC_TWCR, TWCR_TOTALH | TWCR_TOTALW, val); + regmap_update_bits(ldev->regmap, LTDC_TWCR, TWCR_TOTALH | TWCR_TOTALW, val); - reg_write(ldev->regs, LTDC_LIPCR, (accum_act_h + 1)); + regmap_write(ldev->regmap, LTDC_LIPCR, (accum_act_h + 1)); + + /* Configure the output format (hw version dependent) */ + if (ldev->caps.ycbcr_output) { + /* Input video dynamic_range & colorimetry */ + int vic = drm_match_cea_mode(mode); + u32 val; + + if (vic == 6 || vic == 7 || vic == 21 || vic == 22 || + vic == 2 || vic == 3 || vic == 17 || vic == 18) + /* ITU-R BT.601 */ + val = 0; + else + /* ITU-R BT.709 */ + val = EDCR_OCYSEL; + + switch (bus_formats) { + case MEDIA_BUS_FMT_YUYV8_1X16: + /* enable ycbcr output converter */ + regmap_write(ldev->regmap, LTDC_EDCR, EDCR_OCYEN | val); + break; + case MEDIA_BUS_FMT_YVYU8_1X16: + /* enable ycbcr output converter & invert chrominance order */ + regmap_write(ldev->regmap, LTDC_EDCR, EDCR_OCYEN | EDCR_OCYCO | val); + break; + default: + /* disable ycbcr output converter */ + regmap_write(ldev->regmap, LTDC_EDCR, 0); + break; + } + } } static void ltdc_crtc_atomic_flush(struct drm_crtc *crtc, @@ -638,7 +978,8 @@ static void ltdc_crtc_atomic_flush(struct drm_crtc *crtc, ltdc_crtc_update_clut(crtc); /* Commit shadow registers = update planes at next vblank */ - reg_set(ldev->regs, LTDC_SRCR, SRCR_VBR); + if (!ldev->caps.plane_reg_shadow) + regmap_set_bits(ldev->regmap, LTDC_SRCR, SRCR_VBR); if (event) { crtc->state->event = NULL; @@ -680,10 +1021,14 @@ static bool ltdc_crtc_get_scanout_position(struct drm_crtc *crtc, * simplify the code and only test if line > vactive_end */ if (pm_runtime_active(ddev->dev)) { - line = reg_read(ldev->regs, LTDC_CPSR) & CPSR_CYPOS; - vactive_start = reg_read(ldev->regs, LTDC_BPCR) & BPCR_AVBP; - vactive_end = reg_read(ldev->regs, LTDC_AWCR) & AWCR_AAH; - vtotal = reg_read(ldev->regs, LTDC_TWCR) & TWCR_TOTALH; + regmap_read(ldev->regmap, LTDC_CPSR, &line); + line &= CPSR_CYPOS; + regmap_read(ldev->regmap, LTDC_BPCR, &vactive_start); + vactive_start &= BPCR_AVBP; + regmap_read(ldev->regmap, LTDC_AWCR, &vactive_end); + vactive_end &= AWCR_AAH; + regmap_read(ldev->regmap, LTDC_TWCR, &vtotal); + vtotal &= TWCR_TOTALH; if (line > vactive_end) *vpos = line - vtotal - vactive_start; @@ -719,7 +1064,7 @@ static int ltdc_crtc_enable_vblank(struct drm_crtc *crtc) DRM_DEBUG_DRIVER("\n"); if (state->enable) - reg_set(ldev->regs, LTDC_IER, IER_LIE); + regmap_set_bits(ldev->regmap, LTDC_IER, IER_LIE); else return -EPERM; @@ -731,7 +1076,7 @@ static void ltdc_crtc_disable_vblank(struct drm_crtc *crtc) struct ltdc_device *ldev = crtc_to_ltdc(crtc); DRM_DEBUG_DRIVER("\n"); - reg_clear(ldev->regs, LTDC_IER, IER_LIE); + regmap_clear_bits(ldev->regmap, LTDC_IER, IER_LIE); } static const struct drm_crtc_funcs ltdc_crtc_funcs = { @@ -789,7 +1134,7 @@ static void ltdc_plane_atomic_update(struct drm_plane *plane, u32 y0 = newstate->crtc_y; u32 y1 = newstate->crtc_y + newstate->crtc_h - 1; u32 src_x, src_y, src_w, src_h; - u32 val, pitch_in_bytes, line_length, paddr, ahbp, avbp, bpcr; + u32 val, pitch_in_bytes, line_length, line_number, paddr, ahbp, avbp, bpcr; enum ltdc_pix_fmt pf; if (!newstate->crtc || !fb) { @@ -809,19 +1154,20 @@ static void ltdc_plane_atomic_update(struct drm_plane *plane, newstate->crtc_w, newstate->crtc_h, newstate->crtc_x, newstate->crtc_y); - bpcr = reg_read(ldev->regs, LTDC_BPCR); + regmap_read(ldev->regmap, LTDC_BPCR, &bpcr); + ahbp = (bpcr & BPCR_AHBP) >> 16; avbp = bpcr & BPCR_AVBP; /* Configures the horizontal start and stop position */ val = ((x1 + 1 + ahbp) << 16) + (x0 + 1 + ahbp); - reg_update_bits(ldev->regs, LTDC_L1WHPCR + lofs, - LXWHPCR_WHSTPOS | LXWHPCR_WHSPPOS, val); + regmap_write_bits(ldev->regmap, LTDC_L1WHPCR + lofs, + LXWHPCR_WHSTPOS | LXWHPCR_WHSPPOS, val); /* Configures the vertical start and stop position */ val = ((y1 + 1 + avbp) << 16) + (y0 + 1 + avbp); - reg_update_bits(ldev->regs, LTDC_L1WVPCR + lofs, - LXWVPCR_WVSTPOS | LXWVPCR_WVSPPOS, val); + regmap_write_bits(ldev->regmap, LTDC_L1WVPCR + lofs, + LXWVPCR_WVSTPOS | LXWVPCR_WVSPPOS, val); /* Specifies the pixel format */ pf = to_ltdc_pixelformat(fb->format->format); @@ -829,24 +1175,27 @@ static void ltdc_plane_atomic_update(struct drm_plane *plane, if (ldev->caps.pix_fmt_hw[val] == pf) break; + /* Use the flexible color format feature if necessary and available */ + if (ldev->caps.pix_fmt_flex && val == NB_PF) + val = ltdc_set_flexible_pixel_format(plane, pf); + if (val == NB_PF) { DRM_ERROR("Pixel format %.4s not supported\n", (char *)&fb->format->format); val = 0; /* set by default ARGB 32 bits */ } - reg_update_bits(ldev->regs, LTDC_L1PFCR + lofs, LXPFCR_PF, val); + regmap_write_bits(ldev->regmap, LTDC_L1PFCR + lofs, LXPFCR_PF, val); /* Configures the color frame buffer pitch in bytes & line length */ pitch_in_bytes = fb->pitches[0]; line_length = fb->format->cpp[0] * (x1 - x0 + 1) + (ldev->caps.bus_width >> 3) - 1; val = ((pitch_in_bytes << 16) | line_length); - reg_update_bits(ldev->regs, LTDC_L1CFBLR + lofs, - LXCFBLR_CFBLL | LXCFBLR_CFBP, val); + regmap_write_bits(ldev->regmap, LTDC_L1CFBLR + lofs, LXCFBLR_CFBLL | LXCFBLR_CFBP, val); /* Specifies the constant alpha value */ val = newstate->alpha >> 8; - reg_update_bits(ldev->regs, LTDC_L1CACR + lofs, LXCACR_CONSTA, val); + regmap_write_bits(ldev->regmap, LTDC_L1CACR + lofs, LXCACR_CONSTA, val); /* Specifies the blending factors */ val = BF1_PAXCA | BF2_1PAXCA; @@ -858,24 +1207,98 @@ static void ltdc_plane_atomic_update(struct drm_plane *plane, plane->type != DRM_PLANE_TYPE_PRIMARY) val = BF1_PAXCA | BF2_1PAXCA; - reg_update_bits(ldev->regs, LTDC_L1BFCR + lofs, - LXBFCR_BF2 | LXBFCR_BF1, val); + regmap_write_bits(ldev->regmap, LTDC_L1BFCR + lofs, LXBFCR_BF2 | LXBFCR_BF1, val); /* Configures the frame buffer line number */ - val = y1 - y0 + 1; - reg_update_bits(ldev->regs, LTDC_L1CFBLNR + lofs, LXCFBLNR_CFBLN, val); + line_number = y1 - y0 + 1; + regmap_write_bits(ldev->regmap, LTDC_L1CFBLNR + lofs, LXCFBLNR_CFBLN, line_number); /* Sets the FB address */ paddr = (u32)drm_fb_cma_get_gem_addr(fb, newstate, 0); DRM_DEBUG_DRIVER("fb: phys 0x%08x", paddr); - reg_write(ldev->regs, LTDC_L1CFBAR + lofs, paddr); + regmap_write(ldev->regmap, LTDC_L1CFBAR + lofs, paddr); + + if (ldev->caps.ycbcr_input) { + if (fb->format->is_yuv) { + switch (fb->format->format) { + case DRM_FORMAT_NV12: + case DRM_FORMAT_NV21: + /* Configure the auxiliary frame buffer address 0 & 1 */ + paddr = (u32)drm_fb_cma_get_gem_addr(fb, newstate, 1); + regmap_write(ldev->regmap, LTDC_L1AFBA0R + lofs, paddr); + regmap_write(ldev->regmap, LTDC_L1AFBA1R + lofs, paddr + 1); + + /* Configure the buffer length */ + val = ((pitch_in_bytes << 16) | line_length); + regmap_write(ldev->regmap, LTDC_L1AFBLR + lofs, val); + + /* Configure the frame buffer line number */ + val = (line_number >> 1); + regmap_write(ldev->regmap, LTDC_L1AFBLNR + lofs, val); + break; + case DRM_FORMAT_YUV420: + /* Configure the auxiliary frame buffer address 0 */ + paddr = (u32)drm_fb_cma_get_gem_addr(fb, newstate, 1); + regmap_write(ldev->regmap, LTDC_L1AFBA0R + lofs, paddr); + + /* Configure the auxiliary frame buffer address 1 */ + paddr = (u32)drm_fb_cma_get_gem_addr(fb, newstate, 2); + regmap_write(ldev->regmap, LTDC_L1AFBA1R + lofs, paddr); + + line_length = ((fb->format->cpp[0] * (x1 - x0 + 1)) >> 1) + + (ldev->caps.bus_width >> 3) - 1; + + /* Configure the buffer length */ + val = (((pitch_in_bytes >> 1) << 16) | line_length); + regmap_write(ldev->regmap, LTDC_L1AFBLR + lofs, val); + + /* Configure the frame buffer line number */ + val = (line_number >> 1); + regmap_write(ldev->regmap, LTDC_L1AFBLNR + lofs, val); + break; + case DRM_FORMAT_YVU420: + /* Configure the auxiliary frame buffer address 0 */ + paddr = (u32)drm_fb_cma_get_gem_addr(fb, newstate, 2); + regmap_write(ldev->regmap, LTDC_L1AFBA0R + lofs, paddr); + + /* Configure the auxiliary frame buffer address 1 */ + paddr = (u32)drm_fb_cma_get_gem_addr(fb, newstate, 1); + regmap_write(ldev->regmap, LTDC_L1AFBA1R + lofs, paddr); + + line_length = ((fb->format->cpp[0] * (x1 - x0 + 1)) >> 1) + + (ldev->caps.bus_width >> 3) - 1; + + /* Configure the buffer length */ + val = (((pitch_in_bytes >> 1) << 16) | line_length); + regmap_write(ldev->regmap, LTDC_L1AFBLR + lofs, val); + + /* Configure the frame buffer line number */ + val = (line_number >> 1); + regmap_write(ldev->regmap, LTDC_L1AFBLNR + lofs, val); + break; + } + + /* Configure YCbC conversion coefficient */ + ltdc_set_ycbcr_coeffs(plane); + + /* Configure YCbCr format and enable/disable conversion */ + ltdc_set_ycbcr_config(plane, fb->format->format); + } else { + /* disable ycbcr conversion */ + regmap_write(ldev->regmap, LTDC_L1PCR + lofs, 0); + } + } /* Enable layer and CLUT if needed */ val = fb->format->format == DRM_FORMAT_C8 ? LXCR_CLUTEN : 0; val |= LXCR_LEN; - reg_update_bits(ldev->regs, LTDC_L1CR + lofs, - LXCR_LEN | LXCR_CLUTEN, val); + regmap_write_bits(ldev->regmap, LTDC_L1CR + lofs, LXCR_LEN | LXCR_CLUTEN, val); + + /* Commit shadow registers = update plane at next vblank */ + if (ldev->caps.plane_reg_shadow) + regmap_write_bits(ldev->regmap, LTDC_L1RCR + lofs, + LXRCR_IMR | LXRCR_VBR | LXRCR_GRMSK, LXRCR_VBR); ldev->plane_fpsi[plane->index].counter++; @@ -900,7 +1323,12 @@ static void ltdc_plane_atomic_disable(struct drm_plane *plane, u32 lofs = plane->index * LAY_OFS; /* disable layer */ - reg_clear(ldev->regs, LTDC_L1CR + lofs, LXCR_LEN); + regmap_write_bits(ldev->regmap, LTDC_L1CR + lofs, LXCR_LEN, 0); + + /* Commit shadow registers = update plane at next vblank */ + if (ldev->caps.plane_reg_shadow) + regmap_write_bits(ldev->regmap, LTDC_L1RCR + lofs, + LXRCR_IMR | LXRCR_VBR | LXRCR_GRMSK, LXRCR_VBR); DRM_DEBUG_DRIVER("CRTC:%d plane:%d\n", oldstate->crtc->base.id, plane->base.id); @@ -925,16 +1353,6 @@ static void ltdc_plane_atomic_print_state(struct drm_printer *p, fpsi->counter = 0; } -static bool ltdc_plane_format_mod_supported(struct drm_plane *plane, - u32 format, - u64 modifier) -{ - if (modifier == DRM_FORMAT_MOD_LINEAR) - return true; - - return false; -} - static const struct drm_plane_funcs ltdc_plane_funcs = { .update_plane = drm_atomic_helper_update_plane, .disable_plane = drm_atomic_helper_disable_plane, @@ -943,7 +1361,6 @@ static const struct drm_plane_funcs ltdc_plane_funcs = { .atomic_duplicate_state = drm_atomic_helper_plane_duplicate_state, .atomic_destroy_state = drm_atomic_helper_plane_destroy_state, .atomic_print_state = ltdc_plane_atomic_print_state, - .format_mod_supported = ltdc_plane_format_mod_supported, }; static const struct drm_plane_helper_funcs ltdc_plane_helper_funcs = { @@ -953,36 +1370,58 @@ static const struct drm_plane_helper_funcs ltdc_plane_helper_funcs = { }; static struct drm_plane *ltdc_plane_create(struct drm_device *ddev, - enum drm_plane_type type) + enum drm_plane_type type, + int index) { unsigned long possible_crtcs = CRTC_MASK; struct ltdc_device *ldev = ddev->dev_private; struct device *dev = ddev->dev; struct drm_plane *plane; unsigned int i, nb_fmt = 0; - u32 formats[NB_PF * 2]; - u32 drm_fmt, drm_fmt_no_alpha; + u32 *formats; + u32 drm_fmt; const u64 *modifiers = ltdc_format_modifiers; + u32 lofs = index * LAY_OFS; + u32 val; int ret; - /* Get supported pixel formats */ - for (i = 0; i < NB_PF; i++) { - drm_fmt = to_drm_pixelformat(ldev->caps.pix_fmt_hw[i]); - if (!drm_fmt) - continue; - formats[nb_fmt++] = drm_fmt; + /* Allocate the biggest size according to supported color formats */ + formats = devm_kzalloc(dev, (ldev->caps.pix_fmt_nb + + ARRAY_SIZE(ltdc_drm_fmt_ycbcr_cp) + + ARRAY_SIZE(ltdc_drm_fmt_ycbcr_sp) + + ARRAY_SIZE(ltdc_drm_fmt_ycbcr_fp)) * + sizeof(*formats), GFP_KERNEL); - /* Add the no-alpha related format if any & supported */ - drm_fmt_no_alpha = get_pixelformat_without_alpha(drm_fmt); - if (!drm_fmt_no_alpha) - continue; + for (i = 0; i < ldev->caps.pix_fmt_nb; i++) { + drm_fmt = ldev->caps.pix_fmt_drm[i]; /* Manage hw-specific capabilities */ - if (ldev->caps.non_alpha_only_l1 && - type != DRM_PLANE_TYPE_PRIMARY) - continue; + if (ldev->caps.non_alpha_only_l1) + /* XR24 & RX24 like formats supported only on primary layer */ + if (type != DRM_PLANE_TYPE_PRIMARY && is_xrgb(drm_fmt)) + continue; - formats[nb_fmt++] = drm_fmt_no_alpha; + formats[nb_fmt++] = drm_fmt; + } + + /* Add YCbCr supported pixel formats */ + if (ldev->caps.ycbcr_input) { + regmap_read(ldev->regmap, LTDC_L1C1R + lofs, &val); + if (val & LXCR_C1R_YIA) { + memcpy(&formats[nb_fmt], ltdc_drm_fmt_ycbcr_cp, + ARRAY_SIZE(ltdc_drm_fmt_ycbcr_cp) * sizeof(*formats)); + nb_fmt += ARRAY_SIZE(ltdc_drm_fmt_ycbcr_cp); + } + if (val & LXCR_C1R_YSPA) { + memcpy(&formats[nb_fmt], ltdc_drm_fmt_ycbcr_sp, + ARRAY_SIZE(ltdc_drm_fmt_ycbcr_sp) * sizeof(*formats)); + nb_fmt += ARRAY_SIZE(ltdc_drm_fmt_ycbcr_sp); + } + if (val & LXCR_C1R_YFPA) { + memcpy(&formats[nb_fmt], ltdc_drm_fmt_ycbcr_fp, + ARRAY_SIZE(ltdc_drm_fmt_ycbcr_fp) * sizeof(*formats)); + nb_fmt += ARRAY_SIZE(ltdc_drm_fmt_ycbcr_fp); + } } plane = devm_kzalloc(dev, sizeof(*plane), GFP_KERNEL); @@ -995,6 +1434,17 @@ static struct drm_plane *ltdc_plane_create(struct drm_device *ddev, if (ret < 0) return NULL; + if (ldev->caps.ycbcr_input) { + if (val & (LXCR_C1R_YIA | LXCR_C1R_YSPA | LXCR_C1R_YFPA)) + drm_plane_create_color_properties(plane, + BIT(DRM_COLOR_YCBCR_BT601) | + BIT(DRM_COLOR_YCBCR_BT709), + BIT(DRM_COLOR_YCBCR_LIMITED_RANGE) | + BIT(DRM_COLOR_YCBCR_FULL_RANGE), + DRM_COLOR_YCBCR_BT601, + DRM_COLOR_YCBCR_LIMITED_RANGE); + } + drm_plane_helper_add(plane, <dc_plane_helper_funcs); drm_plane_create_alpha_property(plane); @@ -1020,7 +1470,7 @@ static int ltdc_crtc_init(struct drm_device *ddev, struct drm_crtc *crtc) unsigned int i; int ret; - primary = ltdc_plane_create(ddev, DRM_PLANE_TYPE_PRIMARY); + primary = ltdc_plane_create(ddev, DRM_PLANE_TYPE_PRIMARY, 0); if (!primary) { DRM_ERROR("Can not create primary plane\n"); return -EINVAL; @@ -1044,7 +1494,7 @@ static int ltdc_crtc_init(struct drm_device *ddev, struct drm_crtc *crtc) /* Add planes. Note : the first layer is used by primary plane */ for (i = 1; i < ldev->caps.nb_layers; i++) { - overlay = ltdc_plane_create(ddev, DRM_PLANE_TYPE_OVERLAY); + overlay = ltdc_plane_create(ddev, DRM_PLANE_TYPE_OVERLAY, i); if (!overlay) { ret = -ENOMEM; DRM_ERROR("Can not create overlay plane %d\n", i); @@ -1068,7 +1518,7 @@ static void ltdc_encoder_disable(struct drm_encoder *encoder) DRM_DEBUG_DRIVER("\n"); /* Disable LTDC */ - reg_clear(ldev->regs, LTDC_GCR, GCR_LTDCEN); + regmap_clear_bits(ldev->regmap, LTDC_GCR, GCR_LTDCEN); /* Set to sleep state the pinctrl whatever type of encoder */ pinctrl_pm_select_sleep_state(ddev->dev); @@ -1082,7 +1532,7 @@ static void ltdc_encoder_enable(struct drm_encoder *encoder) DRM_DEBUG_DRIVER("\n"); /* Enable LTDC */ - reg_set(ldev->regs, LTDC_GCR, GCR_LTDCEN); + regmap_set_bits(ldev->regmap, LTDC_GCR, GCR_LTDCEN); } static void ltdc_encoder_mode_set(struct drm_encoder *encoder, @@ -1145,21 +1595,25 @@ static int ltdc_get_caps(struct drm_device *ddev) * at least 1 layer must be managed & the number of layers * must not exceed LTDC_MAX_LAYER */ - lcr = reg_read(ldev->regs, LTDC_LCR); + regmap_read(ldev->regmap, LTDC_LCR, &lcr); ldev->caps.nb_layers = clamp((int)lcr, 1, LTDC_MAX_LAYER); /* set data bus width */ - gc2r = reg_read(ldev->regs, LTDC_GC2R); + regmap_read(ldev->regmap, LTDC_GC2R, &gc2r); bus_width_log2 = (gc2r & GC2R_BW) >> 4; ldev->caps.bus_width = 8 << bus_width_log2; - ldev->caps.hw_version = reg_read(ldev->regs, LTDC_IDR); + regmap_read(ldev->regmap, LTDC_IDR, &ldev->caps.hw_version); switch (ldev->caps.hw_version) { case HWVER_10200: case HWVER_10300: - ldev->caps.reg_ofs = REG_OFS_NONE; + ldev->caps.layer_ofs = LAY_OFS_0; + ldev->caps.layer_regs = ltdc_layer_regs_a0; ldev->caps.pix_fmt_hw = ltdc_pix_fmt_a0; + ldev->caps.pix_fmt_drm = ltdc_drm_fmt_a0; + ldev->caps.pix_fmt_nb = ARRAY_SIZE(ltdc_drm_fmt_a0); + ldev->caps.pix_fmt_flex = false; /* * Hw older versions support non-alpha color formats derived * from native alpha color formats only on the primary layer. @@ -1172,13 +1626,37 @@ static int ltdc_get_caps(struct drm_device *ddev) if (ldev->caps.hw_version == HWVER_10200) ldev->caps.pad_max_freq_hz = 65000000; ldev->caps.nb_irq = 2; + ldev->caps.ycbcr_input = false; + ldev->caps.ycbcr_output = false; + ldev->caps.plane_reg_shadow = false; break; case HWVER_20101: - ldev->caps.reg_ofs = REG_OFS_4; + ldev->caps.layer_ofs = LAY_OFS_0; + ldev->caps.layer_regs = ltdc_layer_regs_a1; ldev->caps.pix_fmt_hw = ltdc_pix_fmt_a1; + ldev->caps.pix_fmt_drm = ltdc_drm_fmt_a1; + ldev->caps.pix_fmt_nb = ARRAY_SIZE(ltdc_drm_fmt_a1); + ldev->caps.pix_fmt_flex = false; ldev->caps.non_alpha_only_l1 = false; ldev->caps.pad_max_freq_hz = 150000000; ldev->caps.nb_irq = 4; + ldev->caps.ycbcr_input = false; + ldev->caps.ycbcr_output = false; + ldev->caps.plane_reg_shadow = false; + break; + case HWVER_40100: + ldev->caps.layer_ofs = LAY_OFS_1; + ldev->caps.layer_regs = ltdc_layer_regs_a2; + ldev->caps.pix_fmt_hw = ltdc_pix_fmt_a2; + ldev->caps.pix_fmt_drm = ltdc_drm_fmt_a2; + ldev->caps.pix_fmt_nb = ARRAY_SIZE(ltdc_drm_fmt_a2); + ldev->caps.pix_fmt_flex = true; + ldev->caps.non_alpha_only_l1 = false; + ldev->caps.pad_max_freq_hz = 90000000; + ldev->caps.nb_irq = 2; + ldev->caps.ycbcr_input = true; + ldev->caps.ycbcr_output = true; + ldev->caps.plane_reg_shadow = true; break; default: return -ENODEV; @@ -1296,9 +1774,15 @@ int ltdc_load(struct drm_device *ddev) goto err; } + ldev->regmap = devm_regmap_init_mmio(&pdev->dev, ldev->regs, &stm32_ltdc_regmap_cfg); + if (IS_ERR(ldev->regmap)) { + DRM_ERROR("Unable to regmap ltdc registers\n"); + ret = PTR_ERR(ldev->regmap); + goto err; + } + /* Disable interrupts */ - reg_clear(ldev->regs, LTDC_IER, - IER_LIE | IER_RRIE | IER_FUIE | IER_TERRIE); + regmap_clear_bits(ldev->regmap, LTDC_IER, IER_LIE | IER_RRIE | IER_FUIE | IER_TERRIE); ret = ltdc_get_caps(ddev); if (ret) { diff --git a/drivers/gpu/drm/stm/ltdc.h b/drivers/gpu/drm/stm/ltdc.h index f153b908c70e..6968d1ca5149 100644 --- a/drivers/gpu/drm/stm/ltdc.h +++ b/drivers/gpu/drm/stm/ltdc.h @@ -14,12 +14,19 @@ struct ltdc_caps { u32 hw_version; /* hardware version */ u32 nb_layers; /* number of supported layers */ - u32 reg_ofs; /* register offset for applicable regs */ + u32 layer_ofs; /* layer offset for applicable regs */ + const u32 *layer_regs; /* layer register offset */ u32 bus_width; /* bus width (32 or 64 bits) */ - const u32 *pix_fmt_hw; /* supported pixel formats */ + const u32 *pix_fmt_hw; /* supported hw pixel formats */ + const u32 *pix_fmt_drm; /* supported drm pixel formats */ + int pix_fmt_nb; /* number of pixel format */ + bool pix_fmt_flex; /* pixel format flexibility supported */ bool non_alpha_only_l1; /* non-native no-alpha formats on layer 1 */ int pad_max_freq_hz; /* max frequency supported by pad */ int nb_irq; /* number of hardware interrupts */ + bool ycbcr_input; /* ycbcr input converter supported */ + bool ycbcr_output; /* ycbcr output converter supported */ + bool plane_reg_shadow; /* plane shadow registers ability */ }; #define LTDC_MAX_LAYER 4 @@ -31,6 +38,7 @@ struct fps_info { struct ltdc_device { void __iomem *regs; + struct regmap *regmap; struct clk *pixel_clk; /* lcd pixel clock */ struct mutex err_lock; /* protecting error_status */ struct ltdc_caps caps; diff --git a/drivers/gpu/drm/tegra/Kconfig b/drivers/gpu/drm/tegra/Kconfig index 8cf5aeb9db6c..18c319b804c0 100644 --- a/drivers/gpu/drm/tegra/Kconfig +++ b/drivers/gpu/drm/tegra/Kconfig @@ -5,6 +5,7 @@ config DRM_TEGRA depends on COMMON_CLK depends on DRM depends on OF + select DRM_DP_HELPER select DRM_KMS_HELPER select DRM_MIPI_DSI select DRM_PANEL diff --git a/drivers/gpu/drm/tegra/dp.c b/drivers/gpu/drm/tegra/dp.c index 70dfb7d1dec5..e4369e5b2943 100644 --- a/drivers/gpu/drm/tegra/dp.c +++ b/drivers/gpu/drm/tegra/dp.c @@ -5,7 +5,7 @@ */ #include <drm/drm_crtc.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_print.h> #include "dp.h" diff --git a/drivers/gpu/drm/tegra/dpaux.c b/drivers/gpu/drm/tegra/dpaux.c index 1f96e416fa08..8ca500977a46 100644 --- a/drivers/gpu/drm/tegra/dpaux.c +++ b/drivers/gpu/drm/tegra/dpaux.c @@ -18,7 +18,7 @@ #include <linux/reset.h> #include <linux/workqueue.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_panel.h> #include "dp.h" diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c index fce0e52973c2..0063403ab5e1 100644 --- a/drivers/gpu/drm/tegra/gem.c +++ b/drivers/gpu/drm/tegra/gem.c @@ -174,7 +174,7 @@ static void tegra_bo_unpin(struct host1x_bo_mapping *map) static void *tegra_bo_mmap(struct host1x_bo *bo) { struct tegra_bo *obj = host1x_to_tegra_bo(bo); - struct dma_buf_map map; + struct iosys_map map; int ret; if (obj->vaddr) { @@ -191,7 +191,7 @@ static void *tegra_bo_mmap(struct host1x_bo *bo) static void tegra_bo_munmap(struct host1x_bo *bo, void *addr) { struct tegra_bo *obj = host1x_to_tegra_bo(bo); - struct dma_buf_map map = DMA_BUF_MAP_INIT_VADDR(addr); + struct iosys_map map = IOSYS_MAP_INIT_VADDR(addr); if (obj->vaddr) return; @@ -699,17 +699,17 @@ static int tegra_gem_prime_mmap(struct dma_buf *buf, struct vm_area_struct *vma) return __tegra_gem_mmap(gem, vma); } -static int tegra_gem_prime_vmap(struct dma_buf *buf, struct dma_buf_map *map) +static int tegra_gem_prime_vmap(struct dma_buf *buf, struct iosys_map *map) { struct drm_gem_object *gem = buf->priv; struct tegra_bo *bo = to_tegra_bo(gem); - dma_buf_map_set_vaddr(map, bo->vaddr); + iosys_map_set_vaddr(map, bo->vaddr); return 0; } -static void tegra_gem_prime_vunmap(struct dma_buf *buf, struct dma_buf_map *map) +static void tegra_gem_prime_vunmap(struct dma_buf *buf, struct iosys_map *map) { } diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c index 0ea320c1092b..b125572feb84 100644 --- a/drivers/gpu/drm/tegra/sor.c +++ b/drivers/gpu/drm/tegra/sor.c @@ -18,7 +18,7 @@ #include <drm/drm_atomic_helper.h> #include <drm/drm_debugfs.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_file.h> #include <drm/drm_panel.h> #include <drm/drm_scdc_helper.h> diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c index 3ddb7c710a3d..cc567c87057d 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c @@ -60,8 +60,6 @@ void tilcdc_module_cleanup(struct tilcdc_module *mod) list_del(&mod->list); } -static struct of_device_id tilcdc_of_match[]; - static int tilcdc_atomic_check(struct drm_device *dev, struct drm_atomic_state *state) { @@ -587,7 +585,7 @@ static int tilcdc_pdev_remove(struct platform_device *pdev) return 0; } -static struct of_device_id tilcdc_of_match[] = { +static const struct of_device_id tilcdc_of_match[] = { { .compatible = "ti,am33xx-tilcdc", }, { .compatible = "ti,da850-tilcdc", }, { }, diff --git a/drivers/gpu/drm/tiny/bochs.c b/drivers/gpu/drm/tiny/bochs.c index fc26a1ce11ee..ed971c8bb446 100644 --- a/drivers/gpu/drm/tiny/bochs.c +++ b/drivers/gpu/drm/tiny/bochs.c @@ -10,6 +10,7 @@ #include <drm/drm_gem_framebuffer_helper.h> #include <drm/drm_gem_vram_helper.h> #include <drm/drm_managed.h> +#include <drm/drm_module.h> #include <drm/drm_probe_helper.h> #include <drm/drm_simple_kms_helper.h> @@ -716,24 +717,7 @@ static struct pci_driver bochs_pci_driver = { /* ---------------------------------------------------------------------- */ /* module init/exit */ -static int __init bochs_init(void) -{ - if (drm_firmware_drivers_only() && bochs_modeset == -1) - return -EINVAL; - - if (bochs_modeset == 0) - return -EINVAL; - - return pci_register_driver(&bochs_pci_driver); -} - -static void __exit bochs_exit(void) -{ - pci_unregister_driver(&bochs_pci_driver); -} - -module_init(bochs_init); -module_exit(bochs_exit); +drm_module_pci_driver_if_modeset(bochs_pci_driver, bochs_modeset); MODULE_DEVICE_TABLE(pci, bochs_pci_tbl); MODULE_AUTHOR("Gerd Hoffmann <kraxel@redhat.com>"); diff --git a/drivers/gpu/drm/tiny/cirrus.c b/drivers/gpu/drm/tiny/cirrus.c index c95d9ff7d600..c8e791840862 100644 --- a/drivers/gpu/drm/tiny/cirrus.c +++ b/drivers/gpu/drm/tiny/cirrus.c @@ -16,7 +16,7 @@ * Copyright 1999-2001 Jeff Garzik <jgarzik@pobox.com> */ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/module.h> #include <linux/pci.h> @@ -39,6 +39,7 @@ #include <drm/drm_ioctl.h> #include <drm/drm_managed.h> #include <drm/drm_modeset_helper_vtables.h> +#include <drm/drm_module.h> #include <drm/drm_probe_helper.h> #include <drm/drm_simple_kms_helper.h> @@ -312,7 +313,8 @@ static int cirrus_mode_set(struct cirrus_device *cirrus, return 0; } -static int cirrus_fb_blit_rect(struct drm_framebuffer *fb, const struct dma_buf_map *map, +static int cirrus_fb_blit_rect(struct drm_framebuffer *fb, + const struct iosys_map *map, struct drm_rect *rect) { struct cirrus_device *cirrus = to_cirrus(fb->dev); @@ -344,7 +346,8 @@ static int cirrus_fb_blit_rect(struct drm_framebuffer *fb, const struct dma_buf_ return 0; } -static int cirrus_fb_blit_fullscreen(struct drm_framebuffer *fb, const struct dma_buf_map *map) +static int cirrus_fb_blit_fullscreen(struct drm_framebuffer *fb, + const struct iosys_map *map) { struct drm_rect fullscreen = { .x1 = 0, @@ -633,21 +636,7 @@ static struct pci_driver cirrus_pci_driver = { .remove = cirrus_pci_remove, }; -static int __init cirrus_init(void) -{ - if (drm_firmware_drivers_only()) - return -EINVAL; - - return pci_register_driver(&cirrus_pci_driver); -} - -static void __exit cirrus_exit(void) -{ - pci_unregister_driver(&cirrus_pci_driver); -} - -module_init(cirrus_init); -module_exit(cirrus_exit); +drm_module_pci_driver(cirrus_pci_driver) MODULE_DEVICE_TABLE(pci, pciidlist); MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/tiny/gm12u320.c b/drivers/gpu/drm/tiny/gm12u320.c index 6bc0c298739c..648e585d40a8 100644 --- a/drivers/gpu/drm/tiny/gm12u320.c +++ b/drivers/gpu/drm/tiny/gm12u320.c @@ -95,7 +95,7 @@ struct gm12u320_device { struct drm_rect rect; int frame; int draw_status_timeout; - struct dma_buf_map src_map; + struct iosys_map src_map; } fb_update; }; @@ -395,7 +395,8 @@ err: GM12U320_ERR("Frame update error: %d\n", ret); } -static void gm12u320_fb_mark_dirty(struct drm_framebuffer *fb, const struct dma_buf_map *map, +static void gm12u320_fb_mark_dirty(struct drm_framebuffer *fb, + const struct iosys_map *map, struct drm_rect *dirty) { struct gm12u320_device *gm12u320 = to_gm12u320(fb->dev); @@ -438,7 +439,7 @@ static void gm12u320_stop_fb_update(struct gm12u320_device *gm12u320) mutex_lock(&gm12u320->fb_update.lock); old_fb = gm12u320->fb_update.fb; gm12u320->fb_update.fb = NULL; - dma_buf_map_clear(&gm12u320->fb_update.src_map); + iosys_map_clear(&gm12u320->fb_update.src_map); mutex_unlock(&gm12u320->fb_update.lock); drm_framebuffer_put(old_fb); diff --git a/drivers/gpu/drm/tiny/simpledrm.c b/drivers/gpu/drm/tiny/simpledrm.c index 04146da2d1d8..d191e87a37b5 100644 --- a/drivers/gpu/drm/tiny/simpledrm.c +++ b/drivers/gpu/drm/tiny/simpledrm.c @@ -526,21 +526,33 @@ static int simpledrm_device_init_mm(struct simpledrm_device *sdev) { struct drm_device *dev = &sdev->dev; struct platform_device *pdev = sdev->pdev; - struct resource *mem; + struct resource *res, *mem; void __iomem *screen_base; int ret; - mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!mem) + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) return -EINVAL; - ret = devm_aperture_acquire_from_firmware(dev, mem->start, resource_size(mem)); + ret = devm_aperture_acquire_from_firmware(dev, res->start, resource_size(res)); if (ret) { drm_err(dev, "could not acquire memory range %pr: error %d\n", - mem, ret); + res, ret); return ret; } + mem = devm_request_mem_region(&pdev->dev, res->start, resource_size(res), + sdev->dev.driver->name); + if (!mem) { + /* + * We cannot make this fatal. Sometimes this comes from magic + * spaces our resource handlers simply don't know about. Use + * the I/O-memory resource as-is and try to map that instead. + */ + drm_warn(dev, "could not acquire memory region %pr\n", res); + mem = res; + } + screen_base = devm_ioremap_wc(&pdev->dev, mem->start, resource_size(mem)); if (!screen_base) diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 72a94301bc95..2b8caa1efaa3 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -33,7 +33,7 @@ #include <drm/ttm/ttm_placement.h> #include <drm/drm_cache.h> #include <drm/drm_vma_manager.h> -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/io.h> #include <linux/highmem.h> #include <linux/wait.h> @@ -93,7 +93,7 @@ void ttm_move_memcpy(bool clear, { const struct ttm_kmap_iter_ops *dst_ops = dst_iter->ops; const struct ttm_kmap_iter_ops *src_ops = src_iter->ops; - struct dma_buf_map src_map, dst_map; + struct iosys_map src_map, dst_map; pgoff_t i; /* Single TTM move. NOP */ @@ -241,6 +241,11 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, if (bo->type != ttm_bo_type_sg) fbo->base.base.resv = &fbo->base.base._resv; + if (fbo->base.resource) { + ttm_resource_set_bo(fbo->base.resource, &fbo->base); + bo->resource = NULL; + } + dma_resv_init(&fbo->base.base._resv); fbo->base.base.dev = NULL; ret = dma_resv_trylock(&fbo->base.base._resv); @@ -385,7 +390,7 @@ void ttm_bo_kunmap(struct ttm_bo_kmap_obj *map) } EXPORT_SYMBOL(ttm_bo_kunmap); -int ttm_bo_vmap(struct ttm_buffer_object *bo, struct dma_buf_map *map) +int ttm_bo_vmap(struct ttm_buffer_object *bo, struct iosys_map *map) { struct ttm_resource *mem = bo->resource; int ret; @@ -413,7 +418,7 @@ int ttm_bo_vmap(struct ttm_buffer_object *bo, struct dma_buf_map *map) if (!vaddr_iomem) return -ENOMEM; - dma_buf_map_set_vaddr_iomem(map, vaddr_iomem); + iosys_map_set_vaddr_iomem(map, vaddr_iomem); } else { struct ttm_operation_ctx ctx = { @@ -437,25 +442,25 @@ int ttm_bo_vmap(struct ttm_buffer_object *bo, struct dma_buf_map *map) if (!vaddr) return -ENOMEM; - dma_buf_map_set_vaddr(map, vaddr); + iosys_map_set_vaddr(map, vaddr); } return 0; } EXPORT_SYMBOL(ttm_bo_vmap); -void ttm_bo_vunmap(struct ttm_buffer_object *bo, struct dma_buf_map *map) +void ttm_bo_vunmap(struct ttm_buffer_object *bo, struct iosys_map *map) { struct ttm_resource *mem = bo->resource; - if (dma_buf_map_is_null(map)) + if (iosys_map_is_null(map)) return; if (!map->is_iomem) vunmap(map->vaddr); else if (!mem->bus.addr) iounmap(map->vaddr_iomem); - dma_buf_map_clear(map); + iosys_map_clear(map); ttm_mem_io_free(bo->bdev, bo->resource); } @@ -509,7 +514,6 @@ static int ttm_bo_move_to_ghost(struct ttm_buffer_object *bo, ghost_obj->ttm = NULL; else bo->ttm = NULL; - bo->resource = NULL; dma_resv_unlock(&ghost_obj->base._resv); ttm_bo_put(ghost_obj); @@ -637,7 +641,6 @@ int ttm_bo_pipeline_gutting(struct ttm_buffer_object *bo) dma_resv_unlock(&ghost->base._resv); ttm_bo_put(ghost); bo->ttm = ttm; - bo->resource = NULL; ttm_bo_assign_mem(bo, sys_res); return 0; diff --git a/drivers/gpu/drm/ttm/ttm_range_manager.c b/drivers/gpu/drm/ttm/ttm_range_manager.c index 072e0baf2ab4..8cd4f3fb9f79 100644 --- a/drivers/gpu/drm/ttm/ttm_range_manager.c +++ b/drivers/gpu/drm/ttm/ttm_range_manager.c @@ -89,6 +89,7 @@ static int ttm_range_man_alloc(struct ttm_resource_manager *man, spin_unlock(&rman->lock); if (unlikely(ret)) { + ttm_resource_fini(man, *res); kfree(node); return ret; } @@ -108,6 +109,7 @@ static void ttm_range_man_free(struct ttm_resource_manager *man, drm_mm_remove_node(&node->mm_nodes[0]); spin_unlock(&rman->lock); + ttm_resource_fini(man, res); kfree(node); } @@ -156,7 +158,7 @@ int ttm_range_man_init_nocheck(struct ttm_device *bdev, man->func = &ttm_range_manager_func; - ttm_resource_manager_init(man, p_size); + ttm_resource_manager_init(man, bdev, p_size); drm_mm_init(&rman->mm, 0, p_size); spin_lock_init(&rman->lock); diff --git a/drivers/gpu/drm/ttm/ttm_resource.c b/drivers/gpu/drm/ttm/ttm_resource.c index 035d71332d18..29ee2eda7b7d 100644 --- a/drivers/gpu/drm/ttm/ttm_resource.c +++ b/drivers/gpu/drm/ttm/ttm_resource.c @@ -22,13 +22,21 @@ * Authors: Christian König */ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/io-mapping.h> #include <linux/scatterlist.h> #include <drm/ttm/ttm_resource.h> #include <drm/ttm/ttm_bo_driver.h> +/** + * ttm_resource_init - resource object constructure + * @bo: buffer object this resources is allocated for + * @place: placement of the resource + * @res: the resource object to inistilize + * + * Initialize a new resource object. Counterpart of &ttm_resource_fini. + */ void ttm_resource_init(struct ttm_buffer_object *bo, const struct ttm_place *place, struct ttm_resource *res) @@ -41,9 +49,25 @@ void ttm_resource_init(struct ttm_buffer_object *bo, res->bus.offset = 0; res->bus.is_iomem = false; res->bus.caching = ttm_cached; + res->bo = bo; } EXPORT_SYMBOL(ttm_resource_init); +/** + * ttm_resource_fini - resource destructor + * @man: the resource manager this resource belongs to + * @res: the resource to clean up + * + * Should be used by resource manager backends to clean up the TTM resource + * objects before freeing the underlying structure. Counterpart of + * &ttm_resource_init + */ +void ttm_resource_fini(struct ttm_resource_manager *man, + struct ttm_resource *res) +{ +} +EXPORT_SYMBOL(ttm_resource_fini); + int ttm_resource_alloc(struct ttm_buffer_object *bo, const struct ttm_place *place, struct ttm_resource **res_ptr) @@ -116,20 +140,31 @@ bool ttm_resource_compat(struct ttm_resource *res, } EXPORT_SYMBOL(ttm_resource_compat); +void ttm_resource_set_bo(struct ttm_resource *res, + struct ttm_buffer_object *bo) +{ + spin_lock(&bo->bdev->lru_lock); + res->bo = bo; + spin_unlock(&bo->bdev->lru_lock); +} + /** * ttm_resource_manager_init * * @man: memory manager object to init + * @bdev: ttm device this manager belongs to * @p_size: size managed area in pages. * * Initialise core parts of a manager object. */ void ttm_resource_manager_init(struct ttm_resource_manager *man, + struct ttm_device *bdev, unsigned long p_size) { unsigned i; spin_lock_init(&man->move_lock); + man->bdev = bdev; man->size = p_size; for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i) @@ -209,7 +244,7 @@ void ttm_resource_manager_debug(struct ttm_resource_manager *man, EXPORT_SYMBOL(ttm_resource_manager_debug); static void ttm_kmap_iter_iomap_map_local(struct ttm_kmap_iter *iter, - struct dma_buf_map *dmap, + struct iosys_map *dmap, pgoff_t i) { struct ttm_kmap_iter_iomap *iter_io = @@ -236,11 +271,11 @@ retry: addr = io_mapping_map_local_wc(iter_io->iomap, iter_io->cache.offs + (((resource_size_t)i - iter_io->cache.i) << PAGE_SHIFT)); - dma_buf_map_set_vaddr_iomem(dmap, addr); + iosys_map_set_vaddr_iomem(dmap, addr); } static void ttm_kmap_iter_iomap_unmap_local(struct ttm_kmap_iter *iter, - struct dma_buf_map *map) + struct iosys_map *map) { io_mapping_unmap_local(map->vaddr_iomem); } @@ -291,14 +326,14 @@ EXPORT_SYMBOL(ttm_kmap_iter_iomap_init); */ static void ttm_kmap_iter_linear_io_map_local(struct ttm_kmap_iter *iter, - struct dma_buf_map *dmap, + struct iosys_map *dmap, pgoff_t i) { struct ttm_kmap_iter_linear_io *iter_io = container_of(iter, typeof(*iter_io), base); *dmap = iter_io->dmap; - dma_buf_map_incr(dmap, i * PAGE_SIZE); + iosys_map_incr(dmap, i * PAGE_SIZE); } static const struct ttm_kmap_iter_ops ttm_kmap_iter_linear_io_ops = { @@ -334,7 +369,7 @@ ttm_kmap_iter_linear_io_init(struct ttm_kmap_iter_linear_io *iter_io, } if (mem->bus.addr) { - dma_buf_map_set_vaddr(&iter_io->dmap, mem->bus.addr); + iosys_map_set_vaddr(&iter_io->dmap, mem->bus.addr); iter_io->needs_unmap = false; } else { size_t bus_size = (size_t)mem->num_pages << PAGE_SHIFT; @@ -342,23 +377,23 @@ ttm_kmap_iter_linear_io_init(struct ttm_kmap_iter_linear_io *iter_io, iter_io->needs_unmap = true; memset(&iter_io->dmap, 0, sizeof(iter_io->dmap)); if (mem->bus.caching == ttm_write_combined) - dma_buf_map_set_vaddr_iomem(&iter_io->dmap, - ioremap_wc(mem->bus.offset, - bus_size)); + iosys_map_set_vaddr_iomem(&iter_io->dmap, + ioremap_wc(mem->bus.offset, + bus_size)); else if (mem->bus.caching == ttm_cached) - dma_buf_map_set_vaddr(&iter_io->dmap, - memremap(mem->bus.offset, bus_size, - MEMREMAP_WB | - MEMREMAP_WT | - MEMREMAP_WC)); + iosys_map_set_vaddr(&iter_io->dmap, + memremap(mem->bus.offset, bus_size, + MEMREMAP_WB | + MEMREMAP_WT | + MEMREMAP_WC)); /* If uncached requested or if mapping cached or wc failed */ - if (dma_buf_map_is_null(&iter_io->dmap)) - dma_buf_map_set_vaddr_iomem(&iter_io->dmap, - ioremap(mem->bus.offset, - bus_size)); + if (iosys_map_is_null(&iter_io->dmap)) + iosys_map_set_vaddr_iomem(&iter_io->dmap, + ioremap(mem->bus.offset, + bus_size)); - if (dma_buf_map_is_null(&iter_io->dmap)) { + if (iosys_map_is_null(&iter_io->dmap)) { ret = -ENOMEM; goto out_io_free; } @@ -387,7 +422,7 @@ ttm_kmap_iter_linear_io_fini(struct ttm_kmap_iter_linear_io *iter_io, struct ttm_device *bdev, struct ttm_resource *mem) { - if (iter_io->needs_unmap && dma_buf_map_is_set(&iter_io->dmap)) { + if (iter_io->needs_unmap && iosys_map_is_set(&iter_io->dmap)) { if (iter_io->dmap.is_iomem) iounmap(iter_io->dmap.vaddr_iomem); else diff --git a/drivers/gpu/drm/ttm/ttm_sys_manager.c b/drivers/gpu/drm/ttm/ttm_sys_manager.c index 63aca52f75e1..2ced169513cb 100644 --- a/drivers/gpu/drm/ttm/ttm_sys_manager.c +++ b/drivers/gpu/drm/ttm/ttm_sys_manager.c @@ -23,6 +23,7 @@ static int ttm_sys_man_alloc(struct ttm_resource_manager *man, static void ttm_sys_man_free(struct ttm_resource_manager *man, struct ttm_resource *res) { + ttm_resource_fini(man, res); kfree(res); } @@ -42,7 +43,7 @@ void ttm_sys_man_init(struct ttm_device *bdev) man->use_tt = true; man->func = &ttm_sys_manager_func; - ttm_resource_manager_init(man, 0); + ttm_resource_manager_init(man, bdev, 0); ttm_set_driver_manager(bdev, TTM_PL_SYSTEM, man); ttm_resource_manager_set_used(man, true); } diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c index 79c870a3bef8..d234aab800a0 100644 --- a/drivers/gpu/drm/ttm/ttm_tt.c +++ b/drivers/gpu/drm/ttm/ttm_tt.c @@ -406,18 +406,18 @@ void ttm_tt_mgr_init(unsigned long num_pages, unsigned long num_dma32_pages) } static void ttm_kmap_iter_tt_map_local(struct ttm_kmap_iter *iter, - struct dma_buf_map *dmap, + struct iosys_map *dmap, pgoff_t i) { struct ttm_kmap_iter_tt *iter_tt = container_of(iter, typeof(*iter_tt), base); - dma_buf_map_set_vaddr(dmap, kmap_local_page_prot(iter_tt->tt->pages[i], - iter_tt->prot)); + iosys_map_set_vaddr(dmap, kmap_local_page_prot(iter_tt->tt->pages[i], + iter_tt->prot)); } static void ttm_kmap_iter_tt_unmap_local(struct ttm_kmap_iter *iter, - struct dma_buf_map *map) + struct iosys_map *map) { kunmap_local(map->vaddr); } diff --git a/drivers/gpu/drm/udl/udl_modeset.c b/drivers/gpu/drm/udl/udl_modeset.c index 32232228dae9..e67c40a48fb4 100644 --- a/drivers/gpu/drm/udl/udl_modeset.c +++ b/drivers/gpu/drm/udl/udl_modeset.c @@ -264,7 +264,8 @@ static int udl_aligned_damage_clip(struct drm_rect *clip, int x, int y, return 0; } -static int udl_handle_damage(struct drm_framebuffer *fb, const struct dma_buf_map *map, +static int udl_handle_damage(struct drm_framebuffer *fb, + const struct iosys_map *map, int x, int y, int width, int height) { struct drm_device *dev = fb->dev; diff --git a/drivers/gpu/drm/v3d/v3d_drv.c b/drivers/gpu/drm/v3d/v3d_drv.c index bd46396a1ae0..1afcd54fbbd5 100644 --- a/drivers/gpu/drm/v3d/v3d_drv.c +++ b/drivers/gpu/drm/v3d/v3d_drv.c @@ -219,6 +219,7 @@ static int v3d_platform_drm_probe(struct platform_device *pdev) int ret; u32 mmu_debug; u32 ident1; + u64 mask; v3d = devm_drm_dev_alloc(dev, &v3d_drm_driver, struct v3d_dev, drm); if (IS_ERR(v3d)) @@ -237,8 +238,11 @@ static int v3d_platform_drm_probe(struct platform_device *pdev) return ret; mmu_debug = V3D_READ(V3D_MMU_DEBUG_INFO); - dma_set_mask_and_coherent(dev, - DMA_BIT_MASK(30 + V3D_GET_FIELD(mmu_debug, V3D_MMU_PA_WIDTH))); + mask = DMA_BIT_MASK(30 + V3D_GET_FIELD(mmu_debug, V3D_MMU_PA_WIDTH)); + ret = dma_set_mask_and_coherent(dev, mask); + if (ret) + return ret; + v3d->va_width = 30 + V3D_GET_FIELD(mmu_debug, V3D_MMU_VA_WIDTH); ident1 = V3D_READ(V3D_HUB_IDENT1); diff --git a/drivers/gpu/drm/vboxvideo/vbox_mode.c b/drivers/gpu/drm/vboxvideo/vbox_mode.c index 4227a915b06a..4017b0a621fc 100644 --- a/drivers/gpu/drm/vboxvideo/vbox_mode.c +++ b/drivers/gpu/drm/vboxvideo/vbox_mode.c @@ -10,7 +10,7 @@ * Hans de Goede <hdegoede@redhat.com> */ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/export.h> #include <drm/drm_atomic.h> @@ -398,7 +398,7 @@ static void vbox_cursor_atomic_update(struct drm_plane *plane, u32 height = new_state->crtc_h; struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(new_state); - struct dma_buf_map map = shadow_plane_state->data[0]; + struct iosys_map map = shadow_plane_state->data[0]; u8 *src = map.vaddr; /* TODO: Use mapping abstraction properly */ size_t data_size, mask_size; u32 flags; diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c index 6d1281a343e9..e451cc5bcfac 100644 --- a/drivers/gpu/drm/vc4/vc4_bo.c +++ b/drivers/gpu/drm/vc4/vc4_bo.c @@ -355,8 +355,6 @@ static struct vc4_bo *vc4_bo_get_from_cache(struct drm_device *dev, uint32_t page_index = bo_page_index(size); struct vc4_bo *bo = NULL; - size = roundup(size, PAGE_SIZE); - mutex_lock(&vc4->bo_lock); if (page_index >= vc4->bo_cache.size_list_size) goto out; diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index 16abc3a3d601..a03053c8e22c 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -37,6 +37,8 @@ #include <drm/drm_fb_helper.h> #include <drm/drm_vblank.h> +#include <soc/bcm2835/raspberrypi-firmware.h> + #include "uapi/drm/vc4_drm.h" #include "vc4_drv.h" @@ -215,6 +217,7 @@ static void vc4_match_add_drivers(struct device *dev, static int vc4_drm_bind(struct device *dev) { struct platform_device *pdev = to_platform_device(dev); + struct rpi_firmware *firmware = NULL; struct drm_device *drm; struct vc4_dev *vc4; struct device_node *node; @@ -251,15 +254,34 @@ static int vc4_drm_bind(struct device *dev) if (ret) return ret; - ret = component_bind_all(dev, drm); + node = of_find_compatible_node(NULL, NULL, "raspberrypi,bcm2835-firmware"); + if (node) { + firmware = rpi_firmware_get(node); + of_node_put(node); + + if (!firmware) + return -EPROBE_DEFER; + } + + ret = drm_aperture_remove_framebuffers(false, &vc4_drm_driver); if (ret) return ret; - ret = vc4_plane_create_additional_planes(drm); + if (firmware) { + ret = rpi_firmware_property(firmware, + RPI_FIRMWARE_NOTIFY_DISPLAY_DONE, + NULL, 0); + if (ret) + drm_warn(drm, "Couldn't stop firmware display driver: %d\n", ret); + + rpi_firmware_put(firmware); + } + + ret = component_bind_all(dev, drm); if (ret) - goto unbind_all; + return ret; - ret = drm_aperture_remove_framebuffers(false, &vc4_drm_driver); + ret = vc4_plane_create_additional_planes(drm); if (ret) goto unbind_all; @@ -357,6 +379,9 @@ static int __init vc4_drm_register(void) { int ret; + if (drm_firmware_drivers_only()) + return -ENODEV; + ret = platform_register_drivers(component_drivers, ARRAY_SIZE(component_drivers)); if (ret) diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index 3a1626f261e5..a43e50c9775a 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -104,6 +104,15 @@ static bool vc4_hdmi_mode_needs_scrambling(const struct drm_display_mode *mode) return (mode->clock * 1000) > HDMI_14_MAX_TMDS_CLK; } +static bool vc4_hdmi_is_full_range_rgb(struct vc4_hdmi *vc4_hdmi, + const struct drm_display_mode *mode) +{ + struct vc4_hdmi_encoder *vc4_encoder = &vc4_hdmi->encoder; + + return !vc4_encoder->hdmi_monitor || + drm_default_rgb_quant_range(mode) == HDMI_QUANTIZATION_RANGE_FULL; +} + static int vc4_hdmi_debugfs_regs(struct seq_file *m, void *unused) { struct drm_info_node *node = (struct drm_info_node *)m->private; @@ -475,7 +484,6 @@ static void vc4_hdmi_write_infoframe(struct drm_encoder *encoder, static void vc4_hdmi_set_avi_infoframe(struct drm_encoder *encoder) { struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder); - struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder); struct drm_connector *connector = &vc4_hdmi->connector; struct drm_connector_state *cstate = connector->state; const struct drm_display_mode *mode = &vc4_hdmi->saved_adjusted_mode; @@ -493,10 +501,10 @@ static void vc4_hdmi_set_avi_infoframe(struct drm_encoder *encoder) drm_hdmi_avi_infoframe_quant_range(&frame.avi, connector, mode, - vc4_encoder->limited_rgb_range ? - HDMI_QUANTIZATION_RANGE_LIMITED : - HDMI_QUANTIZATION_RANGE_FULL); - drm_hdmi_avi_infoframe_colorspace(&frame.avi, cstate); + vc4_hdmi_is_full_range_rgb(vc4_hdmi, mode) ? + HDMI_QUANTIZATION_RANGE_FULL : + HDMI_QUANTIZATION_RANGE_LIMITED); + drm_hdmi_avi_infoframe_colorimetry(&frame.avi, cstate); drm_hdmi_avi_infoframe_bars(&frame.avi, cstate); vc4_hdmi_write_infoframe(encoder, &frame); @@ -720,7 +728,9 @@ static void vc4_hdmi_encoder_disable(struct drm_encoder *encoder) mutex_unlock(&vc4_hdmi->mutex); } -static void vc4_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi, bool enable) +static void vc4_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi, + struct drm_connector_state *state, + const struct drm_display_mode *mode) { unsigned long flags; u32 csc_ctl; @@ -730,7 +740,7 @@ static void vc4_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi, bool enable) csc_ctl = VC4_SET_FIELD(VC4_HD_CSC_CTL_ORDER_BGR, VC4_HD_CSC_CTL_ORDER); - if (enable) { + if (!vc4_hdmi_is_full_range_rgb(vc4_hdmi, mode)) { /* CEA VICs other than #1 requre limited range RGB * output unless overridden by an AVI infoframe. * Apply a colorspace conversion to squash 0-255 down @@ -760,44 +770,67 @@ static void vc4_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi, bool enable) spin_unlock_irqrestore(&vc4_hdmi->hw_lock, flags); } -static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi, bool enable) +/* + * If we need to output Full Range RGB, then use the unity matrix + * + * [ 1 0 0 0] + * [ 0 1 0 0] + * [ 0 0 1 0] + * + * Matrix is signed 2p13 fixed point, with signed 9p6 offsets + */ +static const u16 vc5_hdmi_csc_full_rgb_unity[3][4] = { + { 0x2000, 0x0000, 0x0000, 0x0000 }, + { 0x0000, 0x2000, 0x0000, 0x0000 }, + { 0x0000, 0x0000, 0x2000, 0x0000 }, +}; + +/* + * CEA VICs other than #1 require limited range RGB output unless + * overridden by an AVI infoframe. Apply a colorspace conversion to + * squash 0-255 down to 16-235. The matrix here is: + * + * [ 0.8594 0 0 16] + * [ 0 0.8594 0 16] + * [ 0 0 0.8594 16] + * + * Matrix is signed 2p13 fixed point, with signed 9p6 offsets + */ +static const u16 vc5_hdmi_csc_full_rgb_to_limited_rgb[3][4] = { + { 0x1b80, 0x0000, 0x0000, 0x0400 }, + { 0x0000, 0x1b80, 0x0000, 0x0400 }, + { 0x0000, 0x0000, 0x1b80, 0x0400 }, +}; + +static void vc5_hdmi_set_csc_coeffs(struct vc4_hdmi *vc4_hdmi, + const u16 coeffs[3][4]) { - unsigned long flags; - u32 csc_ctl; + lockdep_assert_held(&vc4_hdmi->hw_lock); + + HDMI_WRITE(HDMI_CSC_12_11, (coeffs[0][1] << 16) | coeffs[0][0]); + HDMI_WRITE(HDMI_CSC_14_13, (coeffs[0][3] << 16) | coeffs[0][2]); + HDMI_WRITE(HDMI_CSC_22_21, (coeffs[1][1] << 16) | coeffs[1][0]); + HDMI_WRITE(HDMI_CSC_24_23, (coeffs[1][3] << 16) | coeffs[1][2]); + HDMI_WRITE(HDMI_CSC_32_31, (coeffs[2][1] << 16) | coeffs[2][0]); + HDMI_WRITE(HDMI_CSC_34_33, (coeffs[2][3] << 16) | coeffs[2][2]); +} - csc_ctl = 0x07; /* RGB_CONVERT_MODE = custom matrix, || USE_RGB_TO_YCBCR */ +static void vc5_hdmi_csc_setup(struct vc4_hdmi *vc4_hdmi, + struct drm_connector_state *state, + const struct drm_display_mode *mode) +{ + unsigned long flags; + u32 csc_ctl = VC5_MT_CP_CSC_CTL_ENABLE | VC4_SET_FIELD(VC4_HD_CSC_CTL_MODE_CUSTOM, + VC5_MT_CP_CSC_CTL_MODE); spin_lock_irqsave(&vc4_hdmi->hw_lock, flags); - if (enable) { - /* CEA VICs other than #1 requre limited range RGB - * output unless overridden by an AVI infoframe. - * Apply a colorspace conversion to squash 0-255 down - * to 16-235. The matrix here is: - * - * [ 0.8594 0 0 16] - * [ 0 0.8594 0 16] - * [ 0 0 0.8594 16] - * [ 0 0 0 1] - * Matrix is signed 2p13 fixed point, with signed 9p6 offsets - */ - HDMI_WRITE(HDMI_CSC_12_11, (0x0000 << 16) | 0x1b80); - HDMI_WRITE(HDMI_CSC_14_13, (0x0400 << 16) | 0x0000); - HDMI_WRITE(HDMI_CSC_22_21, (0x1b80 << 16) | 0x0000); - HDMI_WRITE(HDMI_CSC_24_23, (0x0400 << 16) | 0x0000); - HDMI_WRITE(HDMI_CSC_32_31, (0x0000 << 16) | 0x0000); - HDMI_WRITE(HDMI_CSC_34_33, (0x0400 << 16) | 0x1b80); - } else { - /* Still use the matrix for full range, but make it unity. - * Matrix is signed 2p13 fixed point, with signed 9p6 offsets - */ - HDMI_WRITE(HDMI_CSC_12_11, (0x0000 << 16) | 0x2000); - HDMI_WRITE(HDMI_CSC_14_13, (0x0000 << 16) | 0x0000); - HDMI_WRITE(HDMI_CSC_22_21, (0x2000 << 16) | 0x0000); - HDMI_WRITE(HDMI_CSC_24_23, (0x0000 << 16) | 0x0000); - HDMI_WRITE(HDMI_CSC_32_31, (0x0000 << 16) | 0x0000); - HDMI_WRITE(HDMI_CSC_34_33, (0x0000 << 16) | 0x2000); - } + HDMI_WRITE(HDMI_VEC_INTERFACE_XBAR, 0x354021); + + if (!vc4_hdmi_is_full_range_rgb(vc4_hdmi, mode)) + vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_to_limited_rgb); + else + vc5_hdmi_set_csc_coeffs(vc4_hdmi, vc5_hdmi_csc_full_rgb_unity); HDMI_WRITE(HDMI_CSC_CTL, csc_ctl); @@ -883,7 +916,6 @@ static void vc5_hdmi_set_timings(struct vc4_hdmi *vc4_hdmi, spin_lock_irqsave(&vc4_hdmi->hw_lock, flags); - HDMI_WRITE(HDMI_VEC_INTERFACE_XBAR, 0x354021); HDMI_WRITE(HDMI_HORZA, (vsync_pos ? VC5_HDMI_HORZA_VPOS : 0) | (hsync_pos ? VC5_HDMI_HORZA_HPOS : 0) | @@ -1107,24 +1139,16 @@ static void vc4_hdmi_encoder_pre_crtc_enable(struct drm_encoder *encoder, struct drm_atomic_state *state) { struct vc4_hdmi *vc4_hdmi = encoder_to_vc4_hdmi(encoder); + struct drm_connector *connector = &vc4_hdmi->connector; struct drm_display_mode *mode = &vc4_hdmi->saved_adjusted_mode; - struct vc4_hdmi_encoder *vc4_encoder = to_vc4_hdmi_encoder(encoder); + struct drm_connector_state *conn_state = + drm_atomic_get_new_connector_state(state, connector); unsigned long flags; mutex_lock(&vc4_hdmi->mutex); - if (vc4_encoder->hdmi_monitor && - drm_default_rgb_quant_range(mode) == HDMI_QUANTIZATION_RANGE_LIMITED) { - if (vc4_hdmi->variant->csc_setup) - vc4_hdmi->variant->csc_setup(vc4_hdmi, true); - - vc4_encoder->limited_rgb_range = true; - } else { - if (vc4_hdmi->variant->csc_setup) - vc4_hdmi->variant->csc_setup(vc4_hdmi, false); - - vc4_encoder->limited_rgb_range = false; - } + if (vc4_hdmi->variant->csc_setup) + vc4_hdmi->variant->csc_setup(vc4_hdmi, conn_state, mode); spin_lock_irqsave(&vc4_hdmi->hw_lock, flags); HDMI_WRITE(HDMI_FIFO_CTL, VC4_HDMI_FIFO_CTL_MASTER_SLAVE_N); diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.h b/drivers/gpu/drm/vc4/vc4_hdmi.h index 6ffdd4ec5fb6..1076faeab616 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.h +++ b/drivers/gpu/drm/vc4/vc4_hdmi.h @@ -12,7 +12,6 @@ struct vc4_hdmi_encoder { struct vc4_encoder base; bool hdmi_monitor; - bool limited_rgb_range; }; static inline struct vc4_hdmi_encoder * @@ -77,7 +76,9 @@ struct vc4_hdmi_variant { void (*reset)(struct vc4_hdmi *vc4_hdmi); /* Callback to enable / disable the CSC */ - void (*csc_setup)(struct vc4_hdmi *vc4_hdmi, bool enable); + void (*csc_setup)(struct vc4_hdmi *vc4_hdmi, + struct drm_connector_state *state, + const struct drm_display_mode *mode); /* Callback to configure the video timings in the HDMI block */ void (*set_timings)(struct vc4_hdmi *vc4_hdmi, diff --git a/drivers/gpu/drm/vc4/vc4_regs.h b/drivers/gpu/drm/vc4/vc4_regs.h index 7538b84a6dca..33410718089e 100644 --- a/drivers/gpu/drm/vc4/vc4_regs.h +++ b/drivers/gpu/drm/vc4/vc4_regs.h @@ -774,6 +774,9 @@ enum { # define VC4_HD_CSC_CTL_RGB2YCC BIT(1) # define VC4_HD_CSC_CTL_ENABLE BIT(0) +# define VC5_MT_CP_CSC_CTL_ENABLE BIT(2) +# define VC5_MT_CP_CSC_CTL_MODE_MASK VC4_MASK(1, 0) + # define VC4_DVP_HT_CLOCK_STOP_PIXEL BIT(1) /* HVS display list information. */ diff --git a/drivers/gpu/drm/virtio/virtgpu_gem.c b/drivers/gpu/drm/virtio/virtgpu_gem.c index 2de61b63ef91..48d3c9955f0d 100644 --- a/drivers/gpu/drm/virtio/virtgpu_gem.c +++ b/drivers/gpu/drm/virtio/virtgpu_gem.c @@ -248,6 +248,9 @@ void virtio_gpu_array_put_free(struct virtio_gpu_object_array *objs) { u32 i; + if (!objs) + return; + for (i = 0; i < objs->nents; i++) drm_gem_object_put(objs->objs[i]); virtio_gpu_array_free(objs); diff --git a/drivers/gpu/drm/vkms/vkms_composer.c b/drivers/gpu/drm/vkms/vkms_composer.c index 9e8204be9a14..c6a1036bf2ea 100644 --- a/drivers/gpu/drm/vkms/vkms_composer.c +++ b/drivers/gpu/drm/vkms/vkms_composer.c @@ -157,7 +157,7 @@ static void compose_plane(struct vkms_composer *primary_composer, void *vaddr; void (*pixel_blend)(const u8 *p_src, u8 *p_dst); - if (WARN_ON(dma_buf_map_is_null(&primary_composer->map[0]))) + if (WARN_ON(iosys_map_is_null(&primary_composer->map[0]))) return; vaddr = plane_composer->map[0].vaddr; @@ -187,7 +187,7 @@ static int compose_active_planes(void **vaddr_out, } } - if (WARN_ON(dma_buf_map_is_null(&primary_composer->map[0]))) + if (WARN_ON(iosys_map_is_null(&primary_composer->map[0]))) return -EINVAL; vaddr = primary_composer->map[0].vaddr; diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h index d48c23d40ce5..91e63b12f60f 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.h +++ b/drivers/gpu/drm/vkms/vkms_drv.h @@ -20,15 +20,17 @@ #define XRES_MAX 8192 #define YRES_MAX 8192 +#define NUM_OVERLAY_PLANES 8 + struct vkms_writeback_job { - struct dma_buf_map map[DRM_FORMAT_MAX_PLANES]; - struct dma_buf_map data[DRM_FORMAT_MAX_PLANES]; + struct iosys_map map[DRM_FORMAT_MAX_PLANES]; + struct iosys_map data[DRM_FORMAT_MAX_PLANES]; }; struct vkms_composer { struct drm_framebuffer fb; struct drm_rect src, dst; - struct dma_buf_map map[4]; + struct iosys_map map[4]; unsigned int offset; unsigned int pitch; unsigned int cpp; diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c index 04406bd3ff02..ba0e82ae549a 100644 --- a/drivers/gpu/drm/vkms/vkms_output.c +++ b/drivers/gpu/drm/vkms/vkms_output.c @@ -32,6 +32,21 @@ static const struct drm_connector_helper_funcs vkms_conn_helper_funcs = { .get_modes = vkms_conn_get_modes, }; +static int vkms_add_overlay_plane(struct vkms_device *vkmsdev, int index, + struct drm_crtc *crtc) +{ + struct vkms_plane *overlay; + + overlay = vkms_plane_init(vkmsdev, DRM_PLANE_TYPE_OVERLAY, index); + if (IS_ERR(overlay)) + return PTR_ERR(overlay); + + if (!overlay->base.possible_crtcs) + overlay->base.possible_crtcs = drm_crtc_mask(crtc); + + return 0; +} + int vkms_output_init(struct vkms_device *vkmsdev, int index) { struct vkms_output *output = &vkmsdev->output; @@ -39,21 +54,21 @@ int vkms_output_init(struct vkms_device *vkmsdev, int index) struct drm_connector *connector = &output->connector; struct drm_encoder *encoder = &output->encoder; struct drm_crtc *crtc = &output->crtc; - struct vkms_plane *primary, *cursor = NULL, *overlay = NULL; + struct vkms_plane *primary, *cursor = NULL; int ret; int writeback; + unsigned int n; primary = vkms_plane_init(vkmsdev, DRM_PLANE_TYPE_PRIMARY, index); if (IS_ERR(primary)) return PTR_ERR(primary); if (vkmsdev->config->overlay) { - overlay = vkms_plane_init(vkmsdev, DRM_PLANE_TYPE_OVERLAY, index); - if (IS_ERR(overlay)) - return PTR_ERR(overlay); - - if (!overlay->base.possible_crtcs) - overlay->base.possible_crtcs = drm_crtc_mask(crtc); + for (n = 0; n < NUM_OVERLAY_PLANES; n++) { + ret = vkms_add_overlay_plane(vkmsdev, index, crtc); + if (ret) + return ret; + } } if (vkmsdev->config->cursor) { diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c index 32409e15244b..d8eb674b49a6 100644 --- a/drivers/gpu/drm/vkms/vkms_plane.c +++ b/drivers/gpu/drm/vkms/vkms_plane.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0+ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> diff --git a/drivers/gpu/drm/vkms/vkms_writeback.c b/drivers/gpu/drm/vkms/vkms_writeback.c index 8694227f555f..af1604dfbbaf 100644 --- a/drivers/gpu/drm/vkms/vkms_writeback.c +++ b/drivers/gpu/drm/vkms/vkms_writeback.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0+ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <drm/drm_atomic.h> #include <drm/drm_fourcc.h> diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c b/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c index ebb4505a31a3..60e3cc537f36 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_gmrid_manager.c @@ -117,6 +117,7 @@ nospace: gman->used_gmr_pages -= (*res)->num_pages; spin_unlock(&gman->lock); ida_free(&gman->gmr_ida, id); + ttm_resource_fini(man, *res); kfree(*res); return -ENOSPC; } @@ -130,6 +131,7 @@ static void vmw_gmrid_man_put_node(struct ttm_resource_manager *man, spin_lock(&gman->lock); gman->used_gmr_pages -= res->num_pages; spin_unlock(&gman->lock); + ttm_resource_fini(man, res); kfree(res); } @@ -160,7 +162,7 @@ int vmw_gmrid_man_init(struct vmw_private *dev_priv, int type) man->func = &vmw_gmrid_manager_func; man->use_tt = true; - ttm_resource_manager_init(man, 0); + ttm_resource_manager_init(man, &dev_priv->bdev, 0); spin_lock_init(&gman->lock); gman->used_gmr_pages = 0; ida_init(&gman->gmr_ida); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_system_manager.c b/drivers/gpu/drm/vmwgfx/vmwgfx_system_manager.c index b0005b03a617..d3007bf1b8f5 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_system_manager.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_system_manager.c @@ -49,6 +49,7 @@ static int vmw_sys_man_alloc(struct ttm_resource_manager *man, static void vmw_sys_man_free(struct ttm_resource_manager *man, struct ttm_resource *res) { + ttm_resource_fini(man, res); kfree(res); } @@ -69,7 +70,7 @@ int vmw_sys_man_init(struct vmw_private *dev_priv) man->use_tt = true; man->func = &vmw_sys_manager_func; - ttm_resource_manager_init(man, 0); + ttm_resource_manager_init(man, bdev, 0); ttm_set_driver_manager(bdev, VMW_PL_SYSTEM, man); ttm_resource_manager_set_used(man, true); return 0; diff --git a/drivers/gpu/drm/xen/xen_drm_front_gem.c b/drivers/gpu/drm/xen/xen_drm_front_gem.c index dd358ba2bf8e..5a5bf4e5b717 100644 --- a/drivers/gpu/drm/xen/xen_drm_front_gem.c +++ b/drivers/gpu/drm/xen/xen_drm_front_gem.c @@ -280,7 +280,8 @@ xen_drm_front_gem_import_sg_table(struct drm_device *dev, return &xen_obj->base; } -int xen_drm_front_gem_prime_vmap(struct drm_gem_object *gem_obj, struct dma_buf_map *map) +int xen_drm_front_gem_prime_vmap(struct drm_gem_object *gem_obj, + struct iosys_map *map) { struct xen_gem_object *xen_obj = to_xen_gem_obj(gem_obj); void *vaddr; @@ -293,13 +294,13 @@ int xen_drm_front_gem_prime_vmap(struct drm_gem_object *gem_obj, struct dma_buf_ VM_MAP, PAGE_KERNEL); if (!vaddr) return -ENOMEM; - dma_buf_map_set_vaddr(map, vaddr); + iosys_map_set_vaddr(map, vaddr); return 0; } void xen_drm_front_gem_prime_vunmap(struct drm_gem_object *gem_obj, - struct dma_buf_map *map) + struct iosys_map *map) { vunmap(map->vaddr); } diff --git a/drivers/gpu/drm/xen/xen_drm_front_gem.h b/drivers/gpu/drm/xen/xen_drm_front_gem.h index eaea470f7001..a718a1f382a3 100644 --- a/drivers/gpu/drm/xen/xen_drm_front_gem.h +++ b/drivers/gpu/drm/xen/xen_drm_front_gem.h @@ -12,7 +12,7 @@ #define __XEN_DRM_FRONT_GEM_H struct dma_buf_attachment; -struct dma_buf_map; +struct iosys_map; struct drm_device; struct drm_gem_object; struct sg_table; @@ -32,9 +32,9 @@ struct page **xen_drm_front_gem_get_pages(struct drm_gem_object *obj); void xen_drm_front_gem_free_object_unlocked(struct drm_gem_object *gem_obj); int xen_drm_front_gem_prime_vmap(struct drm_gem_object *gem_obj, - struct dma_buf_map *map); + struct iosys_map *map); void xen_drm_front_gem_prime_vunmap(struct drm_gem_object *gem_obj, - struct dma_buf_map *map); + struct iosys_map *map); #endif /* __XEN_DRM_FRONT_GEM_H */ diff --git a/drivers/gpu/drm/xlnx/Kconfig b/drivers/gpu/drm/xlnx/Kconfig index d8d38d86d5c6..06cf477dbcdd 100644 --- a/drivers/gpu/drm/xlnx/Kconfig +++ b/drivers/gpu/drm/xlnx/Kconfig @@ -6,6 +6,7 @@ config DRM_ZYNQMP_DPSUB depends on PHY_XILINX_ZYNQMP depends on XILINX_ZYNQMP_DPDMA select DMA_ENGINE + select DRM_DP_HELPER select DRM_GEM_CMA_HELPER select DRM_KMS_HELPER select GENERIC_PHY diff --git a/drivers/gpu/drm/xlnx/zynqmp_dp.c b/drivers/gpu/drm/xlnx/zynqmp_dp.c index 6f588dc09ba6..b1bbbb1d0a54 100644 --- a/drivers/gpu/drm/xlnx/zynqmp_dp.c +++ b/drivers/gpu/drm/xlnx/zynqmp_dp.c @@ -13,7 +13,7 @@ #include <drm/drm_connector.h> #include <drm/drm_crtc.h> #include <drm/drm_device.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_edid.h> #include <drm/drm_encoder.h> #include <drm/drm_managed.h> diff --git a/drivers/media/common/videobuf2/videobuf2-dma-contig.c b/drivers/media/common/videobuf2/videobuf2-dma-contig.c index 7c4096e62173..ecf065cd4a67 100644 --- a/drivers/media/common/videobuf2/videobuf2-dma-contig.c +++ b/drivers/media/common/videobuf2/videobuf2-dma-contig.c @@ -99,7 +99,7 @@ static void *vb2_dc_vaddr(struct vb2_buffer *vb, void *buf_priv) return buf->vaddr; if (buf->db_attach) { - struct dma_buf_map map; + struct iosys_map map; if (!dma_buf_vmap(buf->db_attach->dmabuf, &map)) buf->vaddr = map.vaddr; @@ -446,7 +446,7 @@ vb2_dc_dmabuf_ops_end_cpu_access(struct dma_buf *dbuf, return 0; } -static int vb2_dc_dmabuf_ops_vmap(struct dma_buf *dbuf, struct dma_buf_map *map) +static int vb2_dc_dmabuf_ops_vmap(struct dma_buf *dbuf, struct iosys_map *map) { struct vb2_dc_buf *buf; void *vaddr; @@ -456,7 +456,7 @@ static int vb2_dc_dmabuf_ops_vmap(struct dma_buf *dbuf, struct dma_buf_map *map) if (!vaddr) return -EINVAL; - dma_buf_map_set_vaddr(map, vaddr); + iosys_map_set_vaddr(map, vaddr); return 0; } @@ -737,7 +737,7 @@ static void vb2_dc_unmap_dmabuf(void *mem_priv) { struct vb2_dc_buf *buf = mem_priv; struct sg_table *sgt = buf->dma_sgt; - struct dma_buf_map map = DMA_BUF_MAP_INIT_VADDR(buf->vaddr); + struct iosys_map map = IOSYS_MAP_INIT_VADDR(buf->vaddr); if (WARN_ON(!buf->db_attach)) { pr_err("trying to unpin a not attached buffer\n"); diff --git a/drivers/media/common/videobuf2/videobuf2-dma-sg.c b/drivers/media/common/videobuf2/videobuf2-dma-sg.c index 90acafd9a290..f8a21c560ad2 100644 --- a/drivers/media/common/videobuf2/videobuf2-dma-sg.c +++ b/drivers/media/common/videobuf2/videobuf2-dma-sg.c @@ -303,7 +303,7 @@ static void vb2_dma_sg_put_userptr(void *buf_priv) static void *vb2_dma_sg_vaddr(struct vb2_buffer *vb, void *buf_priv) { struct vb2_dma_sg_buf *buf = buf_priv; - struct dma_buf_map map; + struct iosys_map map; int ret; BUG_ON(!buf); @@ -492,11 +492,12 @@ vb2_dma_sg_dmabuf_ops_end_cpu_access(struct dma_buf *dbuf, return 0; } -static int vb2_dma_sg_dmabuf_ops_vmap(struct dma_buf *dbuf, struct dma_buf_map *map) +static int vb2_dma_sg_dmabuf_ops_vmap(struct dma_buf *dbuf, + struct iosys_map *map) { struct vb2_dma_sg_buf *buf = dbuf->priv; - dma_buf_map_set_vaddr(map, buf->vaddr); + iosys_map_set_vaddr(map, buf->vaddr); return 0; } @@ -581,7 +582,7 @@ static void vb2_dma_sg_unmap_dmabuf(void *mem_priv) { struct vb2_dma_sg_buf *buf = mem_priv; struct sg_table *sgt = buf->dma_sgt; - struct dma_buf_map map = DMA_BUF_MAP_INIT_VADDR(buf->vaddr); + struct iosys_map map = IOSYS_MAP_INIT_VADDR(buf->vaddr); if (WARN_ON(!buf->db_attach)) { pr_err("trying to unpin a not attached buffer\n"); diff --git a/drivers/media/common/videobuf2/videobuf2-vmalloc.c b/drivers/media/common/videobuf2/videobuf2-vmalloc.c index 0bbfea66554f..948152f1596b 100644 --- a/drivers/media/common/videobuf2/videobuf2-vmalloc.c +++ b/drivers/media/common/videobuf2/videobuf2-vmalloc.c @@ -312,11 +312,12 @@ static void vb2_vmalloc_dmabuf_ops_release(struct dma_buf *dbuf) vb2_vmalloc_put(dbuf->priv); } -static int vb2_vmalloc_dmabuf_ops_vmap(struct dma_buf *dbuf, struct dma_buf_map *map) +static int vb2_vmalloc_dmabuf_ops_vmap(struct dma_buf *dbuf, + struct iosys_map *map) { struct vb2_vmalloc_buf *buf = dbuf->priv; - dma_buf_map_set_vaddr(map, buf->vaddr); + iosys_map_set_vaddr(map, buf->vaddr); return 0; } @@ -372,7 +373,7 @@ static struct dma_buf *vb2_vmalloc_get_dmabuf(struct vb2_buffer *vb, static int vb2_vmalloc_map_dmabuf(void *mem_priv) { struct vb2_vmalloc_buf *buf = mem_priv; - struct dma_buf_map map; + struct iosys_map map; int ret; ret = dma_buf_vmap(buf->dbuf, &map); @@ -386,7 +387,7 @@ static int vb2_vmalloc_map_dmabuf(void *mem_priv) static void vb2_vmalloc_unmap_dmabuf(void *mem_priv) { struct vb2_vmalloc_buf *buf = mem_priv; - struct dma_buf_map map = DMA_BUF_MAP_INIT_VADDR(buf->vaddr); + struct iosys_map map = IOSYS_MAP_INIT_VADDR(buf->vaddr); dma_buf_vunmap(buf->dbuf, &map); buf->vaddr = NULL; @@ -395,7 +396,7 @@ static void vb2_vmalloc_unmap_dmabuf(void *mem_priv) static void vb2_vmalloc_detach_dmabuf(void *mem_priv) { struct vb2_vmalloc_buf *buf = mem_priv; - struct dma_buf_map map = DMA_BUF_MAP_INIT_VADDR(buf->vaddr); + struct iosys_map map = IOSYS_MAP_INIT_VADDR(buf->vaddr); if (buf->vaddr) dma_buf_vunmap(buf->dbuf, &map); diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c index aa1682b94a23..d60b176ffa95 100644 --- a/drivers/misc/fastrpc.c +++ b/drivers/misc/fastrpc.c @@ -587,11 +587,11 @@ static void fastrpc_dma_buf_detatch(struct dma_buf *dmabuf, kfree(a); } -static int fastrpc_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map) +static int fastrpc_vmap(struct dma_buf *dmabuf, struct iosys_map *map) { struct fastrpc_buf *buf = dmabuf->priv; - dma_buf_map_set_vaddr(map, buf->virt); + iosys_map_set_vaddr(map, buf->virt); return 0; } diff --git a/drivers/platform/chrome/Kconfig b/drivers/platform/chrome/Kconfig index ccc23d8686e8..75e93efd669f 100644 --- a/drivers/platform/chrome/Kconfig +++ b/drivers/platform/chrome/Kconfig @@ -243,6 +243,17 @@ config CROS_USBPD_NOTIFY To compile this driver as a module, choose M here: the module will be called cros_usbpd_notify. +config CHROMEOS_PRIVACY_SCREEN + tristate "ChromeOS Privacy Screen support" + depends on ACPI + depends on DRM + select DRM_PRIVACY_SCREEN + help + This driver provides the support needed for the in-built electronic + privacy screen that is present on some ChromeOS devices. When enabled, + this should probably always be built into the kernel to avoid or + minimize drm probe deferral. + source "drivers/platform/chrome/wilco_ec/Kconfig" endif # CHROMEOS_PLATFORMS diff --git a/drivers/platform/chrome/Makefile b/drivers/platform/chrome/Makefile index f901d2e43166..5d4be9735d9d 100644 --- a/drivers/platform/chrome/Makefile +++ b/drivers/platform/chrome/Makefile @@ -4,6 +4,7 @@ CFLAGS_cros_ec_trace.o:= -I$(src) obj-$(CONFIG_CHROMEOS_LAPTOP) += chromeos_laptop.o +obj-$(CONFIG_CHROMEOS_PRIVACY_SCREEN) += chromeos_privacy_screen.o obj-$(CONFIG_CHROMEOS_PSTORE) += chromeos_pstore.o obj-$(CONFIG_CHROMEOS_TBMC) += chromeos_tbmc.o obj-$(CONFIG_CROS_EC) += cros_ec.o diff --git a/drivers/platform/chrome/chromeos_privacy_screen.c b/drivers/platform/chrome/chromeos_privacy_screen.c new file mode 100644 index 000000000000..77e9f5ee8e33 --- /dev/null +++ b/drivers/platform/chrome/chromeos_privacy_screen.c @@ -0,0 +1,153 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* + * ChromeOS Privacy Screen support + * + * Copyright (C) 2022 Google LLC + * + * This is the Chromeos privacy screen provider, present on certain chromebooks, + * represented by a GOOG0010 device in the ACPI. This ACPI device, if present, + * will cause the i915 drm driver to probe defer until this driver registers + * the privacy-screen. + */ + +#include <linux/acpi.h> +#include <drm/drm_privacy_screen_driver.h> + +/* + * The DSM (Device Specific Method) constants below are the agreed API with + * the firmware team, on how to control privacy screen using ACPI methods. + */ +#define PRIV_SCRN_DSM_REVID 1 /* DSM version */ +#define PRIV_SCRN_DSM_FN_GET_STATUS 1 /* Get privacy screen status */ +#define PRIV_SCRN_DSM_FN_ENABLE 2 /* Enable privacy screen */ +#define PRIV_SCRN_DSM_FN_DISABLE 3 /* Disable privacy screen */ + +static const guid_t chromeos_privacy_screen_dsm_guid = + GUID_INIT(0xc7033113, 0x8720, 0x4ceb, + 0x90, 0x90, 0x9d, 0x52, 0xb3, 0xe5, 0x2d, 0x73); + +static void +chromeos_privacy_screen_get_hw_state(struct drm_privacy_screen + *drm_privacy_screen) +{ + union acpi_object *obj; + acpi_handle handle; + struct device *privacy_screen = + drm_privacy_screen_get_drvdata(drm_privacy_screen); + + handle = acpi_device_handle(to_acpi_device(privacy_screen)); + obj = acpi_evaluate_dsm(handle, &chromeos_privacy_screen_dsm_guid, + PRIV_SCRN_DSM_REVID, + PRIV_SCRN_DSM_FN_GET_STATUS, NULL); + if (!obj) { + dev_err(privacy_screen, + "_DSM failed to get privacy-screen state\n"); + return; + } + + if (obj->type != ACPI_TYPE_INTEGER) + dev_err(privacy_screen, + "Bad _DSM to get privacy-screen state\n"); + else if (obj->integer.value == 1) + drm_privacy_screen->hw_state = drm_privacy_screen->sw_state = + PRIVACY_SCREEN_ENABLED; + else + drm_privacy_screen->hw_state = drm_privacy_screen->sw_state = + PRIVACY_SCREEN_DISABLED; + + ACPI_FREE(obj); +} + +static int +chromeos_privacy_screen_set_sw_state(struct drm_privacy_screen + *drm_privacy_screen, + enum drm_privacy_screen_status state) +{ + union acpi_object *obj = NULL; + acpi_handle handle; + struct device *privacy_screen = + drm_privacy_screen_get_drvdata(drm_privacy_screen); + + handle = acpi_device_handle(to_acpi_device(privacy_screen)); + + if (state == PRIVACY_SCREEN_DISABLED) { + obj = acpi_evaluate_dsm(handle, + &chromeos_privacy_screen_dsm_guid, + PRIV_SCRN_DSM_REVID, + PRIV_SCRN_DSM_FN_DISABLE, NULL); + } else if (state == PRIVACY_SCREEN_ENABLED) { + obj = acpi_evaluate_dsm(handle, + &chromeos_privacy_screen_dsm_guid, + PRIV_SCRN_DSM_REVID, + PRIV_SCRN_DSM_FN_ENABLE, NULL); + } else { + dev_err(privacy_screen, + "Bad attempt to set privacy-screen status to %u\n", + state); + return -EINVAL; + } + + if (!obj) { + dev_err(privacy_screen, + "_DSM failed to set privacy-screen state\n"); + return -EIO; + } + + drm_privacy_screen->hw_state = drm_privacy_screen->sw_state = state; + ACPI_FREE(obj); + return 0; +} + +static const struct drm_privacy_screen_ops chromeos_privacy_screen_ops = { + .get_hw_state = chromeos_privacy_screen_get_hw_state, + .set_sw_state = chromeos_privacy_screen_set_sw_state, +}; + +static int chromeos_privacy_screen_add(struct acpi_device *adev) +{ + struct drm_privacy_screen *drm_privacy_screen = + drm_privacy_screen_register(&adev->dev, + &chromeos_privacy_screen_ops, + &adev->dev); + + if (IS_ERR(drm_privacy_screen)) { + dev_err(&adev->dev, "Error registering privacy-screen\n"); + return PTR_ERR(drm_privacy_screen); + } + + adev->driver_data = drm_privacy_screen; + dev_info(&adev->dev, "registered privacy-screen '%s'\n", + dev_name(&drm_privacy_screen->dev)); + + return 0; +} + +static int chromeos_privacy_screen_remove(struct acpi_device *adev) +{ + struct drm_privacy_screen *drm_privacy_screen = acpi_driver_data(adev); + + drm_privacy_screen_unregister(drm_privacy_screen); + return 0; +} + +static const struct acpi_device_id chromeos_privacy_screen_device_ids[] = { + {"GOOG0010", 0}, /* Google's electronic privacy screen for eDP-1 */ + {} +}; +MODULE_DEVICE_TABLE(acpi, chromeos_privacy_screen_device_ids); + +static struct acpi_driver chromeos_privacy_screen_driver = { + .name = "chromeos_privacy_screen_driver", + .class = "ChromeOS", + .ids = chromeos_privacy_screen_device_ids, + .ops = { + .add = chromeos_privacy_screen_add, + .remove = chromeos_privacy_screen_remove, + }, +}; + +module_acpi_driver(chromeos_privacy_screen_driver); +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("ChromeOS ACPI Privacy Screen driver"); +MODULE_AUTHOR("Rajat Jain <rajatja@google.com>"); diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c index bd045486b933..e1626cda6b67 100644 --- a/drivers/platform/x86/thinkpad_acpi.c +++ b/drivers/platform/x86/thinkpad_acpi.c @@ -9876,7 +9876,7 @@ static int tpacpi_lcdshadow_init(struct ibm_init_struct *iibm) return 0; lcdshadow_dev = drm_privacy_screen_register(&tpacpi_pdev->dev, - &lcdshadow_ops); + &lcdshadow_ops, NULL); if (IS_ERR(lcdshadow_dev)) return PTR_ERR(lcdshadow_dev); diff --git a/drivers/video/fbdev/asiliantfb.c b/drivers/video/fbdev/asiliantfb.c index 84c56f525889..f8ef62542f7f 100644 --- a/drivers/video/fbdev/asiliantfb.c +++ b/drivers/video/fbdev/asiliantfb.c @@ -110,7 +110,7 @@ static const struct fb_ops asiliantfb_ops = { static void asiliant_calc_dclk2(u32 *ppixclock, u8 *dclk2_m, u8 *dclk2_n, u8 *dclk2_div) { unsigned pixclock = *ppixclock; - unsigned Ftarget = 1000000 * (1000000 / pixclock); + unsigned Ftarget; unsigned n; unsigned best_error = 0xffffffff; unsigned best_m = 0xffffffff, diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c index 13083ad8d751..ad9aac06427a 100644 --- a/drivers/video/fbdev/core/fbmem.c +++ b/drivers/video/fbdev/core/fbmem.c @@ -25,6 +25,7 @@ #include <linux/init.h> #include <linux/linux_logo.h> #include <linux/proc_fs.h> +#include <linux/platform_device.h> #include <linux/seq_file.h> #include <linux/console.h> #include <linux/kmod.h> @@ -1559,18 +1560,36 @@ static void do_remove_conflicting_framebuffers(struct apertures_struct *a, /* check all firmware fbs and kick off if the base addr overlaps */ for_each_registered_fb(i) { struct apertures_struct *gen_aper; + struct device *device; if (!(registered_fb[i]->flags & FBINFO_MISC_FIRMWARE)) continue; gen_aper = registered_fb[i]->apertures; + device = registered_fb[i]->device; if (fb_do_apertures_overlap(gen_aper, a) || (primary && gen_aper && gen_aper->count && gen_aper->ranges[0].base == VGA_FB_PHYS)) { printk(KERN_INFO "fb%d: switching to %s from %s\n", i, name, registered_fb[i]->fix.id); - do_unregister_framebuffer(registered_fb[i]); + + /* + * If we kick-out a firmware driver, we also want to remove + * the underlying platform device, such as simple-framebuffer, + * VESA, EFI, etc. A native driver will then be able to + * allocate the memory range. + * + * If it's not a platform device, at least print a warning. A + * fix would add code to remove the device from the system. + */ + if (dev_is_platform(device)) { + registered_fb[i]->forced_out = true; + platform_device_unregister(to_platform_device(device)); + } else { + pr_warn("fb%d: cannot remove device\n", i); + do_unregister_framebuffer(registered_fb[i]); + } } } } @@ -1900,9 +1919,13 @@ EXPORT_SYMBOL(register_framebuffer); void unregister_framebuffer(struct fb_info *fb_info) { - mutex_lock(®istration_lock); + bool forced_out = fb_info->forced_out; + + if (!forced_out) + mutex_lock(®istration_lock); do_unregister_framebuffer(fb_info); - mutex_unlock(®istration_lock); + if (!forced_out) + mutex_unlock(®istration_lock); } EXPORT_SYMBOL(unregister_framebuffer); diff --git a/drivers/video/fbdev/s3c-fb.c b/drivers/video/fbdev/s3c-fb.c index 3b134e1bbc38..68408c499627 100644 --- a/drivers/video/fbdev/s3c-fb.c +++ b/drivers/video/fbdev/s3c-fb.c @@ -489,7 +489,7 @@ static int s3c_fb_set_par(struct fb_info *info) struct s3c_fb_win *win = info->par; struct s3c_fb *sfb = win->parent; void __iomem *regs = sfb->regs; - void __iomem *buf = regs; + void __iomem *buf; int win_no = win->index; u32 alpha = 0; u32 data; diff --git a/drivers/video/fbdev/simplefb.c b/drivers/video/fbdev/simplefb.c index 57541887188b..94fc9c6d0411 100644 --- a/drivers/video/fbdev/simplefb.c +++ b/drivers/video/fbdev/simplefb.c @@ -66,16 +66,36 @@ static int simplefb_setcolreg(u_int regno, u_int red, u_int green, u_int blue, return 0; } -struct simplefb_par; +struct simplefb_par { + u32 palette[PSEUDO_PALETTE_SIZE]; + struct resource *mem; +#if defined CONFIG_OF && defined CONFIG_COMMON_CLK + bool clks_enabled; + unsigned int clk_count; + struct clk **clks; +#endif +#if defined CONFIG_OF && defined CONFIG_REGULATOR + bool regulators_enabled; + u32 regulator_count; + struct regulator **regulators; +#endif +}; + static void simplefb_clocks_destroy(struct simplefb_par *par); static void simplefb_regulators_destroy(struct simplefb_par *par); static void simplefb_destroy(struct fb_info *info) { + struct simplefb_par *par = info->par; + struct resource *mem = par->mem; + simplefb_regulators_destroy(info->par); simplefb_clocks_destroy(info->par); if (info->screen_base) iounmap(info->screen_base); + + if (mem) + release_mem_region(mem->start, resource_size(mem)); } static const struct fb_ops simplefb_ops = { @@ -169,20 +189,6 @@ static int simplefb_parse_pd(struct platform_device *pdev, return 0; } -struct simplefb_par { - u32 palette[PSEUDO_PALETTE_SIZE]; -#if defined CONFIG_OF && defined CONFIG_COMMON_CLK - bool clks_enabled; - unsigned int clk_count; - struct clk **clks; -#endif -#if defined CONFIG_OF && defined CONFIG_REGULATOR - bool regulators_enabled; - u32 regulator_count; - struct regulator **regulators; -#endif -}; - #if defined CONFIG_OF && defined CONFIG_COMMON_CLK /* * Clock handling code. @@ -405,7 +411,7 @@ static int simplefb_probe(struct platform_device *pdev) struct simplefb_params params; struct fb_info *info; struct simplefb_par *par; - struct resource *mem; + struct resource *res, *mem; /* * Generic drivers must not be registered if a framebuffer exists. @@ -430,15 +436,28 @@ static int simplefb_probe(struct platform_device *pdev) if (ret) return ret; - mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!mem) { + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) { dev_err(&pdev->dev, "No memory resource\n"); return -EINVAL; } + mem = request_mem_region(res->start, resource_size(res), "simplefb"); + if (!mem) { + /* + * We cannot make this fatal. Sometimes this comes from magic + * spaces our resource handlers simply don't know about. Use + * the I/O-memory resource as-is and try to map that instead. + */ + dev_warn(&pdev->dev, "simplefb: cannot reserve video memory at %pR\n", res); + mem = res; + } + info = framebuffer_alloc(sizeof(struct simplefb_par), &pdev->dev); - if (!info) - return -ENOMEM; + if (!info) { + ret = -ENOMEM; + goto error_release_mem_region; + } platform_set_drvdata(pdev, info); par = info->par; @@ -495,6 +514,9 @@ static int simplefb_probe(struct platform_device *pdev) info->var.xres, info->var.yres, info->var.bits_per_pixel, info->fix.line_length); + if (mem != res) + par->mem = mem; /* release in clean-up handler */ + ret = register_framebuffer(info); if (ret < 0) { dev_err(&pdev->dev, "Unable to register simplefb: %d\n", ret); @@ -513,6 +535,9 @@ error_unmap: iounmap(info->screen_base); error_fb_release: framebuffer_release(info); +error_release_mem_region: + if (mem != res) + release_mem_region(mem->start, resource_size(mem)); return ret; } diff --git a/drivers/video/fbdev/vga16fb.c b/drivers/video/fbdev/vga16fb.c index 96e312a3eac7..d21f68f3ee44 100644 --- a/drivers/video/fbdev/vga16fb.c +++ b/drivers/video/fbdev/vga16fb.c @@ -1351,7 +1351,12 @@ static int vga16fb_probe(struct platform_device *dev) printk(KERN_INFO "vga16fb: mapped to 0x%p\n", info->screen_base); par = info->par; +#if defined(CONFIG_X86) + par->isVGA = screen_info.orig_video_isVGA == VIDEO_TYPE_VGAC; +#else + /* non-x86 architectures treat orig_video_isVGA as a boolean flag */ par->isVGA = screen_info.orig_video_isVGA; +#endif par->palette_blanked = 0; par->vesa_blanked = 0; diff --git a/include/drm/bridge/dw_mipi_dsi.h b/include/drm/bridge/dw_mipi_dsi.h index bda8aa7c2280..5286a53a1875 100644 --- a/include/drm/bridge/dw_mipi_dsi.h +++ b/include/drm/bridge/dw_mipi_dsi.h @@ -51,7 +51,9 @@ struct dw_mipi_dsi_plat_data { unsigned int max_data_lanes; enum drm_mode_status (*mode_valid)(void *priv_data, - const struct drm_display_mode *mode); + const struct drm_display_mode *mode, + unsigned long mode_flags, + u32 lanes, u32 format); const struct dw_mipi_dsi_phy_ops *phy_ops; const struct dw_mipi_dsi_host_ops *host_ops; diff --git a/include/drm/drm_dp_aux_bus.h b/include/drm/dp/drm_dp_aux_bus.h index 4f19b20b1dd6..4f19b20b1dd6 100644 --- a/include/drm/drm_dp_aux_bus.h +++ b/include/drm/dp/drm_dp_aux_bus.h diff --git a/include/drm/drm_dp_dual_mode_helper.h b/include/drm/dp/drm_dp_dual_mode_helper.h index 7ee482265087..7ee482265087 100644 --- a/include/drm/drm_dp_dual_mode_helper.h +++ b/include/drm/dp/drm_dp_dual_mode_helper.h diff --git a/include/drm/drm_dp_helper.h b/include/drm/dp/drm_dp_helper.h index 30359e434c3f..98d020835b49 100644 --- a/include/drm/drm_dp_helper.h +++ b/include/drm/dp/drm_dp_helper.h @@ -1038,11 +1038,8 @@ struct drm_panel; #define DP_SIDEBAND_MSG_UP_REQ_BASE 0x1600 /* 1.2 MST */ /* DPRX Event Status Indicator */ -#define DP_SINK_COUNT_ESI 0x2002 /* 1.2 */ -/* 0-5 sink count */ -# define DP_SINK_COUNT_CP_READY (1 << 6) - -#define DP_DEVICE_SERVICE_IRQ_VECTOR_ESI0 0x2003 /* 1.2 */ +#define DP_SINK_COUNT_ESI 0x2002 /* same as 0x200 */ +#define DP_DEVICE_SERVICE_IRQ_VECTOR_ESI0 0x2003 /* same as 0x201 */ #define DP_DEVICE_SERVICE_IRQ_VECTOR_ESI1 0x2004 /* 1.2 */ # define DP_RX_GTC_MSTR_REQ_STATUS_CHANGE (1 << 0) diff --git a/include/drm/drm_dp_mst_helper.h b/include/drm/dp/drm_dp_mst_helper.h index 78044ac5b59b..08276eb8c187 100644 --- a/include/drm/drm_dp_mst_helper.h +++ b/include/drm/dp/drm_dp_mst_helper.h @@ -23,7 +23,7 @@ #define _DRM_DP_MST_HELPER_H_ #include <linux/types.h> -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> #include <drm/drm_atomic.h> #if IS_ENABLED(CONFIG_DRM_DEBUG_DP_MST_TOPOLOGY_REFS) diff --git a/include/drm/drm_buddy.h b/include/drm/drm_buddy.h new file mode 100644 index 000000000000..f524db152413 --- /dev/null +++ b/include/drm/drm_buddy.h @@ -0,0 +1,150 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2021 Intel Corporation + */ + +#ifndef __DRM_BUDDY_H__ +#define __DRM_BUDDY_H__ + +#include <linux/bitops.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/sched.h> + +#include <drm/drm_print.h> + +#define range_overflows(start, size, max) ({ \ + typeof(start) start__ = (start); \ + typeof(size) size__ = (size); \ + typeof(max) max__ = (max); \ + (void)(&start__ == &size__); \ + (void)(&start__ == &max__); \ + start__ >= max__ || size__ > max__ - start__; \ +}) + +struct drm_buddy_block { +#define DRM_BUDDY_HEADER_OFFSET GENMASK_ULL(63, 12) +#define DRM_BUDDY_HEADER_STATE GENMASK_ULL(11, 10) +#define DRM_BUDDY_ALLOCATED (1 << 10) +#define DRM_BUDDY_FREE (2 << 10) +#define DRM_BUDDY_SPLIT (3 << 10) +/* Free to be used, if needed in the future */ +#define DRM_BUDDY_HEADER_UNUSED GENMASK_ULL(9, 6) +#define DRM_BUDDY_HEADER_ORDER GENMASK_ULL(5, 0) + u64 header; + + struct drm_buddy_block *left; + struct drm_buddy_block *right; + struct drm_buddy_block *parent; + + void *private; /* owned by creator */ + + /* + * While the block is allocated by the user through drm_buddy_alloc*, + * the user has ownership of the link, for example to maintain within + * a list, if so desired. As soon as the block is freed with + * drm_buddy_free* ownership is given back to the mm. + */ + struct list_head link; + struct list_head tmp_link; +}; + +/* Order-zero must be at least PAGE_SIZE */ +#define DRM_BUDDY_MAX_ORDER (63 - PAGE_SHIFT) + +/* + * Binary Buddy System. + * + * Locking should be handled by the user, a simple mutex around + * drm_buddy_alloc* and drm_buddy_free* should suffice. + */ +struct drm_buddy { + /* Maintain a free list for each order. */ + struct list_head *free_list; + + /* + * Maintain explicit binary tree(s) to track the allocation of the + * address space. This gives us a simple way of finding a buddy block + * and performing the potentially recursive merge step when freeing a + * block. Nodes are either allocated or free, in which case they will + * also exist on the respective free list. + */ + struct drm_buddy_block **roots; + + /* + * Anything from here is public, and remains static for the lifetime of + * the mm. Everything above is considered do-not-touch. + */ + unsigned int n_roots; + unsigned int max_order; + + /* Must be at least PAGE_SIZE */ + u64 chunk_size; + u64 size; + u64 avail; +}; + +static inline u64 +drm_buddy_block_offset(struct drm_buddy_block *block) +{ + return block->header & DRM_BUDDY_HEADER_OFFSET; +} + +static inline unsigned int +drm_buddy_block_order(struct drm_buddy_block *block) +{ + return block->header & DRM_BUDDY_HEADER_ORDER; +} + +static inline unsigned int +drm_buddy_block_state(struct drm_buddy_block *block) +{ + return block->header & DRM_BUDDY_HEADER_STATE; +} + +static inline bool +drm_buddy_block_is_allocated(struct drm_buddy_block *block) +{ + return drm_buddy_block_state(block) == DRM_BUDDY_ALLOCATED; +} + +static inline bool +drm_buddy_block_is_free(struct drm_buddy_block *block) +{ + return drm_buddy_block_state(block) == DRM_BUDDY_FREE; +} + +static inline bool +drm_buddy_block_is_split(struct drm_buddy_block *block) +{ + return drm_buddy_block_state(block) == DRM_BUDDY_SPLIT; +} + +static inline u64 +drm_buddy_block_size(struct drm_buddy *mm, + struct drm_buddy_block *block) +{ + return mm->chunk_size << drm_buddy_block_order(block); +} + +int drm_buddy_init(struct drm_buddy *mm, u64 size, u64 chunk_size); + +void drm_buddy_fini(struct drm_buddy *mm); + +struct drm_buddy_block * +drm_buddy_alloc_blocks(struct drm_buddy *mm, unsigned int order); + +int drm_buddy_alloc_range(struct drm_buddy *mm, + struct list_head *blocks, + u64 start, u64 size); + +void drm_buddy_free_block(struct drm_buddy *mm, struct drm_buddy_block *block); + +void drm_buddy_free_list(struct drm_buddy *mm, struct list_head *objects); + +void drm_buddy_print(struct drm_buddy *mm, struct drm_printer *p); +void drm_buddy_block_print(struct drm_buddy *mm, + struct drm_buddy_block *block, + struct drm_printer *p); + +#endif diff --git a/include/drm/drm_cache.h b/include/drm/drm_cache.h index cc9de1632dd3..22deb216b59c 100644 --- a/include/drm/drm_cache.h +++ b/include/drm/drm_cache.h @@ -35,7 +35,7 @@ #include <linux/scatterlist.h> -struct dma_buf_map; +struct iosys_map; void drm_clflush_pages(struct page *pages[], unsigned long num_pages); void drm_clflush_sg(struct sg_table *st); @@ -74,7 +74,7 @@ static inline bool drm_arch_can_wc_memory(void) void drm_memcpy_init_early(void); -void drm_memcpy_from_wc(struct dma_buf_map *dst, - const struct dma_buf_map *src, +void drm_memcpy_from_wc(struct iosys_map *dst, + const struct iosys_map *src, unsigned long len); #endif diff --git a/include/drm/drm_client.h b/include/drm/drm_client.h index f07f2fb02e75..4fc8018eddda 100644 --- a/include/drm/drm_client.h +++ b/include/drm/drm_client.h @@ -3,7 +3,7 @@ #ifndef _DRM_CLIENT_H_ #define _DRM_CLIENT_H_ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/lockdep.h> #include <linux/mutex.h> #include <linux/types.h> @@ -144,7 +144,7 @@ struct drm_client_buffer { /** * @map: Virtual address for the buffer */ - struct dma_buf_map map; + struct iosys_map map; /** * @fb: DRM framebuffer @@ -156,7 +156,8 @@ struct drm_client_buffer * drm_client_framebuffer_create(struct drm_client_dev *client, u32 width, u32 height, u32 format); void drm_client_framebuffer_delete(struct drm_client_buffer *buffer); int drm_client_framebuffer_flush(struct drm_client_buffer *buffer, struct drm_rect *rect); -int drm_client_buffer_vmap(struct drm_client_buffer *buffer, struct dma_buf_map *map); +int drm_client_buffer_vmap(struct drm_client_buffer *buffer, + struct iosys_map *map); void drm_client_buffer_vunmap(struct drm_client_buffer *buffer); int drm_client_modeset_create(struct drm_client_dev *client); diff --git a/include/drm/drm_connector.h b/include/drm/drm_connector.h index b501d0badaea..64cf5f88c05b 100644 --- a/include/drm/drm_connector.h +++ b/include/drm/drm_connector.h @@ -522,9 +522,9 @@ struct drm_display_info { enum subpixel_order subpixel_order; #define DRM_COLOR_FORMAT_RGB444 (1<<0) -#define DRM_COLOR_FORMAT_YCRCB444 (1<<1) -#define DRM_COLOR_FORMAT_YCRCB422 (1<<2) -#define DRM_COLOR_FORMAT_YCRCB420 (1<<3) +#define DRM_COLOR_FORMAT_YCBCR444 (1<<1) +#define DRM_COLOR_FORMAT_YCBCR422 (1<<2) +#define DRM_COLOR_FORMAT_YCBCR420 (1<<3) /** * @panel_orientation: Read only connector property for built-in panels, @@ -592,10 +592,16 @@ struct drm_display_info { bool rgb_quant_range_selectable; /** - * @edid_hdmi_dc_modes: Mask of supported hdmi deep color modes. Even - * more stuff redundant with @bus_formats. + * @edid_hdmi_dc_rgb444_modes: Mask of supported hdmi deep color modes + * in RGB 4:4:4. Even more stuff redundant with @bus_formats. */ - u8 edid_hdmi_dc_modes; + u8 edid_hdmi_rgb444_dc_modes; + + /** + * @edid_hdmi_dc_ycbcr444_modes: Mask of supported hdmi deep color + * modes in YCbCr 4:4:4. Even more stuff redundant with @bus_formats. + */ + u8 edid_hdmi_ycbcr444_dc_modes; /** * @cea_rev: CEA revision of the HDMI sink. diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index 13eeba2a750a..a70baea0636c 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -285,6 +285,10 @@ struct drm_crtc_state { * Lookup table for converting pixel data after the color conversion * matrix @ctm. See drm_crtc_enable_color_mgmt(). The blob (if not * NULL) is an array of &struct drm_color_lut. + * + * Note that for mostly historical reasons stemming from Xorg heritage, + * this is also used to store the color map (also sometimes color lut, + * CLUT or color palette) for indexed formats like DRM_FORMAT_C8. */ struct drm_property_blob *gamma_lut; @@ -1075,12 +1079,18 @@ struct drm_crtc { /** * @gamma_size: Size of legacy gamma ramp reported to userspace. Set up * by calling drm_mode_crtc_set_gamma_size(). + * + * Note that atomic drivers need to instead use + * &drm_crtc_state.gamma_lut. See drm_crtc_enable_color_mgmt(). */ uint32_t gamma_size; /** * @gamma_store: Gamma ramp values used by the legacy SETGAMMA and * GETGAMMA IOCTls. Set up by calling drm_mode_crtc_set_gamma_size(). + * + * Note that atomic drivers need to instead use + * &drm_crtc_state.gamma_lut. See drm_crtc_enable_color_mgmt(). */ uint16_t *gamma_store; @@ -1135,14 +1145,12 @@ struct drm_crtc { */ spinlock_t commit_lock; -#ifdef CONFIG_DEBUG_FS /** * @debugfs_entry: * * Debugfs directory for this CRTC. */ struct dentry *debugfs_entry; -#endif /** * @crc: diff --git a/include/drm/drm_dsc.h b/include/drm/drm_dsc.h index cf43561e60fa..ca022e960dcc 100644 --- a/include/drm/drm_dsc.h +++ b/include/drm/drm_dsc.h @@ -8,7 +8,7 @@ #ifndef DRM_DSC_H_ #define DRM_DSC_H_ -#include <drm/drm_dp_helper.h> +#include <drm/dp/drm_dp_helper.h> /* VESA Display Stream Compression DSC 1.2 constants */ #define DSC_NUM_BUF_RANGES 15 diff --git a/include/drm/drm_edid.h b/include/drm/drm_edid.h index 18f6c700f6d0..144c495b99c4 100644 --- a/include/drm/drm_edid.h +++ b/include/drm/drm_edid.h @@ -401,8 +401,8 @@ drm_hdmi_vendor_infoframe_from_display_mode(struct hdmi_vendor_infoframe *frame, const struct drm_display_mode *mode); void -drm_hdmi_avi_infoframe_colorspace(struct hdmi_avi_infoframe *frame, - const struct drm_connector_state *conn_state); +drm_hdmi_avi_infoframe_colorimetry(struct hdmi_avi_infoframe *frame, + const struct drm_connector_state *conn_state); void drm_hdmi_avi_infoframe_bars(struct hdmi_avi_infoframe *frame, diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h index 35e7f44c2a75..e2941cee14b6 100644 --- a/include/drm/drm_gem.h +++ b/include/drm/drm_gem.h @@ -39,7 +39,7 @@ #include <drm/drm_vma_manager.h> -struct dma_buf_map; +struct iosys_map; struct drm_gem_object; /** @@ -139,7 +139,7 @@ struct drm_gem_object_funcs { * * This callback is optional. */ - int (*vmap)(struct drm_gem_object *obj, struct dma_buf_map *map); + int (*vmap)(struct drm_gem_object *obj, struct iosys_map *map); /** * @vunmap: @@ -149,7 +149,7 @@ struct drm_gem_object_funcs { * * This callback is optional. */ - void (*vunmap)(struct drm_gem_object *obj, struct dma_buf_map *map); + void (*vunmap)(struct drm_gem_object *obj, struct iosys_map *map); /** * @mmap: diff --git a/include/drm/drm_gem_atomic_helper.h b/include/drm/drm_gem_atomic_helper.h index 0b1e2dd2ac3f..6e3319e9001a 100644 --- a/include/drm/drm_gem_atomic_helper.h +++ b/include/drm/drm_gem_atomic_helper.h @@ -3,7 +3,7 @@ #ifndef __DRM_GEM_ATOMIC_HELPER_H__ #define __DRM_GEM_ATOMIC_HELPER_H__ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <drm/drm_fourcc.h> #include <drm/drm_plane.h> @@ -59,7 +59,7 @@ struct drm_shadow_plane_state { * The memory mappings stored in map should be established in the plane's * prepare_fb callback and removed in the cleanup_fb callback. */ - struct dma_buf_map map[DRM_FORMAT_MAX_PLANES]; + struct iosys_map map[DRM_FORMAT_MAX_PLANES]; /** * @data: Address of each framebuffer BO's data @@ -67,7 +67,7 @@ struct drm_shadow_plane_state { * The address of the data stored in each mapping. This is different * for framebuffers with non-zero offset fields. */ - struct dma_buf_map data[DRM_FORMAT_MAX_PLANES]; + struct iosys_map data[DRM_FORMAT_MAX_PLANES]; }; /** diff --git a/include/drm/drm_gem_cma_helper.h b/include/drm/drm_gem_cma_helper.h index adb507a9dbf0..fbda4ce5d5fb 100644 --- a/include/drm/drm_gem_cma_helper.h +++ b/include/drm/drm_gem_cma_helper.h @@ -38,7 +38,8 @@ void drm_gem_cma_free(struct drm_gem_cma_object *cma_obj); void drm_gem_cma_print_info(const struct drm_gem_cma_object *cma_obj, struct drm_printer *p, unsigned int indent); struct sg_table *drm_gem_cma_get_sg_table(struct drm_gem_cma_object *cma_obj); -int drm_gem_cma_vmap(struct drm_gem_cma_object *cma_obj, struct dma_buf_map *map); +int drm_gem_cma_vmap(struct drm_gem_cma_object *cma_obj, + struct iosys_map *map); int drm_gem_cma_mmap(struct drm_gem_cma_object *cma_obj, struct vm_area_struct *vma); extern const struct vm_operations_struct drm_gem_cma_vm_ops; @@ -106,7 +107,8 @@ static inline struct sg_table *drm_gem_cma_object_get_sg_table(struct drm_gem_ob * Returns: * 0 on success or a negative error code on failure. */ -static inline int drm_gem_cma_object_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) +static inline int drm_gem_cma_object_vmap(struct drm_gem_object *obj, + struct iosys_map *map) { struct drm_gem_cma_object *cma_obj = to_drm_gem_cma_obj(obj); diff --git a/include/drm/drm_gem_framebuffer_helper.h b/include/drm/drm_gem_framebuffer_helper.h index 905727719ead..1091e4fa08cb 100644 --- a/include/drm/drm_gem_framebuffer_helper.h +++ b/include/drm/drm_gem_framebuffer_helper.h @@ -2,7 +2,7 @@ #define __DRM_GEM_FB_HELPER_H__ #include <linux/dma-buf.h> -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <drm/drm_fourcc.h> @@ -40,10 +40,10 @@ drm_gem_fb_create_with_dirty(struct drm_device *dev, struct drm_file *file, const struct drm_mode_fb_cmd2 *mode_cmd); int drm_gem_fb_vmap(struct drm_framebuffer *fb, - struct dma_buf_map map[static DRM_FORMAT_MAX_PLANES], - struct dma_buf_map data[DRM_FORMAT_MAX_PLANES]); + struct iosys_map map[static DRM_FORMAT_MAX_PLANES], + struct iosys_map data[DRM_FORMAT_MAX_PLANES]); void drm_gem_fb_vunmap(struct drm_framebuffer *fb, - struct dma_buf_map map[static DRM_FORMAT_MAX_PLANES]); + struct iosys_map map[static DRM_FORMAT_MAX_PLANES]); int drm_gem_fb_begin_cpu_access(struct drm_framebuffer *fb, enum dma_data_direction dir); void drm_gem_fb_end_cpu_access(struct drm_framebuffer *fb, enum dma_data_direction dir); diff --git a/include/drm/drm_gem_shmem_helper.h b/include/drm/drm_gem_shmem_helper.h index 311d66c9cf4b..68347b63fc71 100644 --- a/include/drm/drm_gem_shmem_helper.h +++ b/include/drm/drm_gem_shmem_helper.h @@ -113,8 +113,10 @@ int drm_gem_shmem_get_pages(struct drm_gem_shmem_object *shmem); void drm_gem_shmem_put_pages(struct drm_gem_shmem_object *shmem); int drm_gem_shmem_pin(struct drm_gem_shmem_object *shmem); void drm_gem_shmem_unpin(struct drm_gem_shmem_object *shmem); -int drm_gem_shmem_vmap(struct drm_gem_shmem_object *shmem, struct dma_buf_map *map); -void drm_gem_shmem_vunmap(struct drm_gem_shmem_object *shmem, struct dma_buf_map *map); +int drm_gem_shmem_vmap(struct drm_gem_shmem_object *shmem, + struct iosys_map *map); +void drm_gem_shmem_vunmap(struct drm_gem_shmem_object *shmem, + struct iosys_map *map); int drm_gem_shmem_mmap(struct drm_gem_shmem_object *shmem, struct vm_area_struct *vma); int drm_gem_shmem_madvise(struct drm_gem_shmem_object *shmem, int madv); @@ -226,7 +228,8 @@ static inline struct sg_table *drm_gem_shmem_object_get_sg_table(struct drm_gem_ * Returns: * 0 on success or a negative error code on failure. */ -static inline int drm_gem_shmem_object_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) +static inline int drm_gem_shmem_object_vmap(struct drm_gem_object *obj, + struct iosys_map *map) { struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj); @@ -241,7 +244,8 @@ static inline int drm_gem_shmem_object_vmap(struct drm_gem_object *obj, struct d * This function wraps drm_gem_shmem_vunmap(). Drivers that employ the shmem helpers should * use it as their &drm_gem_object_funcs.vunmap handler. */ -static inline void drm_gem_shmem_object_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map) +static inline void drm_gem_shmem_object_vunmap(struct drm_gem_object *obj, + struct iosys_map *map) { struct drm_gem_shmem_object *shmem = to_drm_gem_shmem_obj(obj); diff --git a/include/drm/drm_gem_ttm_helper.h b/include/drm/drm_gem_ttm_helper.h index 78040f6cc6f3..4c003b4f173e 100644 --- a/include/drm/drm_gem_ttm_helper.h +++ b/include/drm/drm_gem_ttm_helper.h @@ -10,7 +10,7 @@ #include <drm/ttm/ttm_bo_api.h> #include <drm/ttm/ttm_bo_driver.h> -struct dma_buf_map; +struct iosys_map; #define drm_gem_ttm_of_gem(gem_obj) \ container_of(gem_obj, struct ttm_buffer_object, base) @@ -18,9 +18,9 @@ struct dma_buf_map; void drm_gem_ttm_print_info(struct drm_printer *p, unsigned int indent, const struct drm_gem_object *gem); int drm_gem_ttm_vmap(struct drm_gem_object *gem, - struct dma_buf_map *map); + struct iosys_map *map); void drm_gem_ttm_vunmap(struct drm_gem_object *gem, - struct dma_buf_map *map); + struct iosys_map *map); int drm_gem_ttm_mmap(struct drm_gem_object *gem, struct vm_area_struct *vma); diff --git a/include/drm/drm_gem_vram_helper.h b/include/drm/drm_gem_vram_helper.h index b4ce27a72773..c083a1d71cf4 100644 --- a/include/drm/drm_gem_vram_helper.h +++ b/include/drm/drm_gem_vram_helper.h @@ -12,7 +12,7 @@ #include <drm/ttm/ttm_bo_driver.h> #include <linux/container_of.h> -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> struct drm_mode_create_dumb; struct drm_plane; @@ -51,7 +51,7 @@ struct vm_area_struct; */ struct drm_gem_vram_object { struct ttm_buffer_object bo; - struct dma_buf_map map; + struct iosys_map map; /** * @vmap_use_count: @@ -97,8 +97,9 @@ void drm_gem_vram_put(struct drm_gem_vram_object *gbo); s64 drm_gem_vram_offset(struct drm_gem_vram_object *gbo); int drm_gem_vram_pin(struct drm_gem_vram_object *gbo, unsigned long pl_flag); int drm_gem_vram_unpin(struct drm_gem_vram_object *gbo); -int drm_gem_vram_vmap(struct drm_gem_vram_object *gbo, struct dma_buf_map *map); -void drm_gem_vram_vunmap(struct drm_gem_vram_object *gbo, struct dma_buf_map *map); +int drm_gem_vram_vmap(struct drm_gem_vram_object *gbo, struct iosys_map *map); +void drm_gem_vram_vunmap(struct drm_gem_vram_object *gbo, + struct iosys_map *map); int drm_gem_vram_fill_create_dumb(struct drm_file *file, struct drm_device *dev, diff --git a/include/drm/drm_mipi_dbi.h b/include/drm/drm_mipi_dbi.h index 05e194958265..6fe13cce2670 100644 --- a/include/drm/drm_mipi_dbi.h +++ b/include/drm/drm_mipi_dbi.h @@ -194,7 +194,7 @@ int mipi_dbi_buf_copy(void *dst, struct drm_framebuffer *fb, #ifdef CONFIG_DEBUG_FS void mipi_dbi_debugfs_init(struct drm_minor *minor); #else -#define mipi_dbi_debugfs_init NULL +static inline void mipi_dbi_debugfs_init(struct drm_minor *minor) {} #endif #endif /* __LINUX_MIPI_DBI_H */ diff --git a/include/drm/drm_modeset_lock.h b/include/drm/drm_modeset_lock.h index b84693fbd2b5..ec4f543c3d95 100644 --- a/include/drm/drm_modeset_lock.h +++ b/include/drm/drm_modeset_lock.h @@ -34,6 +34,7 @@ struct drm_modeset_lock; * struct drm_modeset_acquire_ctx - locking context (see ww_acquire_ctx) * @ww_ctx: base acquire ctx * @contended: used internally for -EDEADLK handling + * @stack_depot: used internally for contention debugging * @locked: list of held locks * @trylock_only: trylock mode used in atomic contexts/panic notifiers * @interruptible: whether interruptible locking should be used. diff --git a/include/drm/drm_module.h b/include/drm/drm_module.h new file mode 100644 index 000000000000..4db1ae03d9a5 --- /dev/null +++ b/include/drm/drm_module.h @@ -0,0 +1,125 @@ +/* SPDX-License-Identifier: MIT */ + +#ifndef DRM_MODULE_H +#define DRM_MODULE_H + +#include <linux/pci.h> +#include <linux/platform_device.h> + +#include <drm/drm_drv.h> + +/** + * DOC: overview + * + * This library provides helpers registering DRM drivers during module + * initialization and shutdown. The provided helpers act like bus-specific + * module helpers, such as module_pci_driver(), but respect additional + * parameters that control DRM driver registration. + * + * Below is an example of initializing a DRM driver for a device on the + * PCI bus. + * + * .. code-block:: c + * + * struct pci_driver my_pci_drv = { + * }; + * + * drm_module_pci_driver(my_pci_drv); + * + * The generated code will test if DRM drivers are enabled and register + * the PCI driver my_pci_drv. For more complex module initialization, you + * can still use module_init() and module_exit() in your driver. + */ + +/* + * PCI drivers + */ + +static inline int __init drm_pci_register_driver(struct pci_driver *pci_drv) +{ + if (drm_firmware_drivers_only()) + return -ENODEV; + + return pci_register_driver(pci_drv); +} + +/** + * drm_module_pci_driver - Register a DRM driver for PCI-based devices + * @__pci_drv: the PCI driver structure + * + * Registers a DRM driver for devices on the PCI bus. The helper + * macro behaves like module_pci_driver() but tests the state of + * drm_firmware_drivers_only(). For more complex module initialization, + * use module_init() and module_exit() directly. + * + * Each module may only use this macro once. Calling it replaces + * module_init() and module_exit(). + */ +#define drm_module_pci_driver(__pci_drv) \ + module_driver(__pci_drv, drm_pci_register_driver, pci_unregister_driver) + +static inline int __init +drm_pci_register_driver_if_modeset(struct pci_driver *pci_drv, int modeset) +{ + if (drm_firmware_drivers_only() && modeset == -1) + return -ENODEV; + if (modeset == 0) + return -ENODEV; + + return pci_register_driver(pci_drv); +} + +static inline void __exit +drm_pci_unregister_driver_if_modeset(struct pci_driver *pci_drv, int modeset) +{ + pci_unregister_driver(pci_drv); +} + +/** + * drm_module_pci_driver_if_modeset - Register a DRM driver for PCI-based devices + * @__pci_drv: the PCI driver structure + * @__modeset: an additional parameter that disables the driver + * + * This macro is deprecated and only provided for existing drivers. For + * new drivers, use drm_module_pci_driver(). + * + * Registers a DRM driver for devices on the PCI bus. The helper macro + * behaves like drm_module_pci_driver() with an additional driver-specific + * flag. If __modeset is 0, the driver has been disabled, if __modeset is + * -1 the driver state depends on the global DRM state. For all other + * values, the PCI driver has been enabled. The default should be -1. + */ +#define drm_module_pci_driver_if_modeset(__pci_drv, __modeset) \ + module_driver(__pci_drv, drm_pci_register_driver_if_modeset, \ + drm_pci_unregister_driver_if_modeset, __modeset) + +/* + * Platform drivers + */ + +static inline int __init +drm_platform_driver_register(struct platform_driver *platform_drv) +{ + if (drm_firmware_drivers_only()) + return -ENODEV; + + return platform_driver_register(platform_drv); +} + +/** + * drm_module_platform_driver - Register a DRM driver for platform devices + * @__platform_drv: the platform driver structure + * + * Registers a DRM driver for devices on the platform bus. The helper + * macro behaves like module_platform_driver() but tests the state of + * drm_firmware_drivers_only(). For more complex module initialization, + * use module_init() and module_exit() directly. + * + * Each module may only use this macro once. Calling it replaces + * module_init() and module_exit(). + */ +#define drm_module_platform_driver(__platform_drv) \ + module_driver(__platform_drv, drm_platform_driver_register, \ + platform_driver_unregister) + +#endif diff --git a/include/drm/drm_plane.h b/include/drm/drm_plane.h index 0c1102dc4d88..06759badf99f 100644 --- a/include/drm/drm_plane.h +++ b/include/drm/drm_plane.h @@ -516,7 +516,7 @@ struct drm_plane_funcs { * This optional hook is used for the DRM to determine if the given * format/modifier combination is valid for the plane. This allows the * DRM to generate the correct format bitmask (which formats apply to - * which modifier), and to valdiate modifiers at atomic_check time. + * which modifier), and to validate modifiers at atomic_check time. * * If not present, then any modifier in the plane's modifier * list is allowed with any of the plane's formats. diff --git a/include/drm/drm_prime.h b/include/drm/drm_prime.h index 54f2c58305d2..2a1d01e5b56b 100644 --- a/include/drm/drm_prime.h +++ b/include/drm/drm_prime.h @@ -54,7 +54,7 @@ struct device; struct dma_buf_export_info; struct dma_buf; struct dma_buf_attachment; -struct dma_buf_map; +struct iosys_map; enum dma_data_direction; @@ -83,8 +83,8 @@ struct sg_table *drm_gem_map_dma_buf(struct dma_buf_attachment *attach, void drm_gem_unmap_dma_buf(struct dma_buf_attachment *attach, struct sg_table *sgt, enum dma_data_direction dir); -int drm_gem_dmabuf_vmap(struct dma_buf *dma_buf, struct dma_buf_map *map); -void drm_gem_dmabuf_vunmap(struct dma_buf *dma_buf, struct dma_buf_map *map); +int drm_gem_dmabuf_vmap(struct dma_buf *dma_buf, struct iosys_map *map); +void drm_gem_dmabuf_vunmap(struct dma_buf *dma_buf, struct iosys_map *map); int drm_gem_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma); int drm_gem_dmabuf_mmap(struct dma_buf *dma_buf, struct vm_area_struct *vma); diff --git a/include/drm/drm_privacy_screen_driver.h b/include/drm/drm_privacy_screen_driver.h index 24591b607675..4ef246d5706f 100644 --- a/include/drm/drm_privacy_screen_driver.h +++ b/include/drm/drm_privacy_screen_driver.h @@ -73,10 +73,21 @@ struct drm_privacy_screen { * for more info. */ enum drm_privacy_screen_status hw_state; + /** + * @drvdata: Private data owned by the privacy screen provider + */ + void *drvdata; }; +static inline +void *drm_privacy_screen_get_drvdata(struct drm_privacy_screen *priv) +{ + return priv->drvdata; +} + struct drm_privacy_screen *drm_privacy_screen_register( - struct device *parent, const struct drm_privacy_screen_ops *ops); + struct device *parent, const struct drm_privacy_screen_ops *ops, + void *data); void drm_privacy_screen_unregister(struct drm_privacy_screen *priv); void drm_privacy_screen_call_notifier_chain(struct drm_privacy_screen *priv); diff --git a/include/drm/i915_pciids.h b/include/drm/i915_pciids.h index baf3d1d3d566..533890dc9da1 100644 --- a/include/drm/i915_pciids.h +++ b/include/drm/i915_pciids.h @@ -666,6 +666,12 @@ INTEL_VGA_DEVICE(0x46C2, info), \ INTEL_VGA_DEVICE(0x46C3, info) +/* ADL-N */ +#define INTEL_ADLN_IDS(info) \ + INTEL_VGA_DEVICE(0x46D0, info), \ + INTEL_VGA_DEVICE(0x46D1, info), \ + INTEL_VGA_DEVICE(0x46D2, info) + /* RPL-S */ #define INTEL_RPLS_IDS(info) \ INTEL_VGA_DEVICE(0xA780, info), \ diff --git a/include/drm/ttm/ttm_bo_api.h b/include/drm/ttm/ttm_bo_api.h index c17b2df9178b..155b19ee12fb 100644 --- a/include/drm/ttm/ttm_bo_api.h +++ b/include/drm/ttm/ttm_bo_api.h @@ -47,7 +47,7 @@ struct ttm_global; struct ttm_device; -struct dma_buf_map; +struct iosys_map; struct drm_mm_node; @@ -481,17 +481,17 @@ void ttm_bo_kunmap(struct ttm_bo_kmap_obj *map); * ttm_bo_vmap * * @bo: The buffer object. - * @map: pointer to a struct dma_buf_map representing the map. + * @map: pointer to a struct iosys_map representing the map. * * Sets up a kernel virtual mapping, using ioremap or vmap to the * data in the buffer object. The parameter @map returns the virtual - * address as struct dma_buf_map. Unmap the buffer with ttm_bo_vunmap(). + * address as struct iosys_map. Unmap the buffer with ttm_bo_vunmap(). * * Returns * -ENOMEM: Out of memory. * -EINVAL: Invalid range. */ -int ttm_bo_vmap(struct ttm_buffer_object *bo, struct dma_buf_map *map); +int ttm_bo_vmap(struct ttm_buffer_object *bo, struct iosys_map *map); /** * ttm_bo_vunmap @@ -501,7 +501,7 @@ int ttm_bo_vmap(struct ttm_buffer_object *bo, struct dma_buf_map *map); * * Unmaps a kernel map set up by ttm_bo_vmap(). */ -void ttm_bo_vunmap(struct ttm_buffer_object *bo, struct dma_buf_map *map); +void ttm_bo_vunmap(struct ttm_buffer_object *bo, struct iosys_map *map); /** * ttm_bo_mmap_obj - mmap memory backed by a ttm buffer object. diff --git a/include/drm/ttm/ttm_kmap_iter.h b/include/drm/ttm/ttm_kmap_iter.h index 8bb00fd39d6c..cc5c09a211b4 100644 --- a/include/drm/ttm/ttm_kmap_iter.h +++ b/include/drm/ttm/ttm_kmap_iter.h @@ -8,7 +8,7 @@ #include <linux/types.h> struct ttm_kmap_iter; -struct dma_buf_map; +struct iosys_map; /** * struct ttm_kmap_iter_ops - Ops structure for a struct @@ -24,22 +24,22 @@ struct ttm_kmap_iter_ops { * kmap_local semantics. * @res_iter: Pointer to the struct ttm_kmap_iter representing * the resource. - * @dmap: The struct dma_buf_map holding the virtual address after + * @dmap: The struct iosys_map holding the virtual address after * the operation. * @i: The location within the resource to map. PAGE_SIZE granularity. */ void (*map_local)(struct ttm_kmap_iter *res_iter, - struct dma_buf_map *dmap, pgoff_t i); + struct iosys_map *dmap, pgoff_t i); /** * unmap_local() - Unmap a PAGE_SIZE part of the resource previously * mapped using kmap_local. * @res_iter: Pointer to the struct ttm_kmap_iter representing * the resource. - * @dmap: The struct dma_buf_map holding the virtual address after + * @dmap: The struct iosys_map holding the virtual address after * the operation. */ void (*unmap_local)(struct ttm_kmap_iter *res_iter, - struct dma_buf_map *dmap); + struct iosys_map *dmap); bool maps_tt; }; diff --git a/include/drm/ttm/ttm_resource.h b/include/drm/ttm/ttm_resource.h index 5952051091cd..4fd727b52da1 100644 --- a/include/drm/ttm/ttm_resource.h +++ b/include/drm/ttm/ttm_resource.h @@ -27,7 +27,7 @@ #include <linux/types.h> #include <linux/mutex.h> -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/dma-fence.h> #include <drm/drm_print.h> #include <drm/ttm/ttm_caching.h> @@ -41,7 +41,7 @@ struct ttm_resource; struct ttm_place; struct ttm_buffer_object; struct ttm_placement; -struct dma_buf_map; +struct iosys_map; struct io_mapping; struct sg_table; struct scatterlist; @@ -105,11 +105,11 @@ struct ttm_resource_manager_func { * @use_type: The memory type is enabled. * @use_tt: If a TT object should be used for the backing store. * @size: Size of the managed region. + * @bdev: ttm device this manager belongs to * @func: structure pointer implementing the range manager. See above * @move_lock: lock for move fence - * static information. bdev::driver::io_mem_free is never used. - * @lru: The lru list for this memory type. * @move: The fence of the last pipelined move operation. + * @lru: The lru list for this memory type. * * This structure is used to identify and manage memory types for a device. */ @@ -119,20 +119,21 @@ struct ttm_resource_manager { */ bool use_type; bool use_tt; + struct ttm_device *bdev; uint64_t size; const struct ttm_resource_manager_func *func; spinlock_t move_lock; /* - * Protected by the global->lru_lock. + * Protected by @move_lock. */ - - struct list_head lru[TTM_MAX_BO_PRIORITY]; + struct dma_fence *move; /* - * Protected by @move_lock. + * Protected by the global->lru_lock. */ - struct dma_fence *move; + + struct list_head lru[TTM_MAX_BO_PRIORITY]; }; /** @@ -160,6 +161,7 @@ struct ttm_bus_placement { * @mem_type: Resource type of the allocation. * @placement: Placement flags. * @bus: Placement on io bus accessible to the CPU + * @bo: weak reference to the BO, protected by ttm_device::lru_lock * * Structure indicating the placement and space resources used by a * buffer object. @@ -170,6 +172,7 @@ struct ttm_resource { uint32_t mem_type; uint32_t placement; struct ttm_bus_placement bus; + struct ttm_buffer_object *bo; }; /** @@ -207,7 +210,7 @@ struct ttm_kmap_iter_iomap { */ struct ttm_kmap_iter_linear_io { struct ttm_kmap_iter base; - struct dma_buf_map dmap; + struct iosys_map dmap; bool needs_unmap; }; @@ -261,14 +264,20 @@ ttm_resource_manager_cleanup(struct ttm_resource_manager *man) void ttm_resource_init(struct ttm_buffer_object *bo, const struct ttm_place *place, struct ttm_resource *res); +void ttm_resource_fini(struct ttm_resource_manager *man, + struct ttm_resource *res); + int ttm_resource_alloc(struct ttm_buffer_object *bo, const struct ttm_place *place, struct ttm_resource **res); void ttm_resource_free(struct ttm_buffer_object *bo, struct ttm_resource **res); bool ttm_resource_compat(struct ttm_resource *res, struct ttm_placement *placement); +void ttm_resource_set_bo(struct ttm_resource *res, + struct ttm_buffer_object *bo); void ttm_resource_manager_init(struct ttm_resource_manager *man, + struct ttm_device *bdev, unsigned long p_size); int ttm_resource_manager_evict_all(struct ttm_device *bdev, diff --git a/include/linux/dma-buf-map.h b/include/linux/dma-buf-map.h index 278d489e4bdd..19fa0b5ae5ec 100644 --- a/include/linux/dma-buf-map.h +++ b/include/linux/dma-buf-map.h @@ -52,13 +52,13 @@ * * struct dma_buf_map map = DMA_BUF_MAP_INIT_VADDR(0xdeadbeaf); * - * dma_buf_map_set_vaddr(&map. 0xdeadbeaf); + * dma_buf_map_set_vaddr(&map, 0xdeadbeaf); * * To set an address in I/O memory, use dma_buf_map_set_vaddr_iomem(). * * .. code-block:: c * - * dma_buf_map_set_vaddr_iomem(&map. 0xdeadbeaf); + * dma_buf_map_set_vaddr_iomem(&map, 0xdeadbeaf); * * Instances of struct dma_buf_map do not have to be cleaned up, but * can be cleared to NULL with dma_buf_map_clear(). Cleared mappings diff --git a/include/linux/dma-buf.h b/include/linux/dma-buf.h index 7ab50076e7a6..2097760e8e95 100644 --- a/include/linux/dma-buf.h +++ b/include/linux/dma-buf.h @@ -13,7 +13,7 @@ #ifndef __DMA_BUF_H__ #define __DMA_BUF_H__ -#include <linux/dma-buf-map.h> +#include <linux/iosys-map.h> #include <linux/file.h> #include <linux/err.h> #include <linux/scatterlist.h> @@ -283,8 +283,8 @@ struct dma_buf_ops { */ int (*mmap)(struct dma_buf *, struct vm_area_struct *vma); - int (*vmap)(struct dma_buf *dmabuf, struct dma_buf_map *map); - void (*vunmap)(struct dma_buf *dmabuf, struct dma_buf_map *map); + int (*vmap)(struct dma_buf *dmabuf, struct iosys_map *map); + void (*vunmap)(struct dma_buf *dmabuf, struct iosys_map *map); }; /** @@ -347,7 +347,7 @@ struct dma_buf { * @vmap_ptr: * The current vmap ptr if @vmapping_counter > 0. Protected by @lock. */ - struct dma_buf_map vmap_ptr; + struct iosys_map vmap_ptr; /** * @exp_name: @@ -628,6 +628,6 @@ int dma_buf_end_cpu_access(struct dma_buf *dma_buf, int dma_buf_mmap(struct dma_buf *, struct vm_area_struct *, unsigned long); -int dma_buf_vmap(struct dma_buf *dmabuf, struct dma_buf_map *map); -void dma_buf_vunmap(struct dma_buf *dmabuf, struct dma_buf_map *map); +int dma_buf_vmap(struct dma_buf *dmabuf, struct iosys_map *map); +void dma_buf_vunmap(struct dma_buf *dmabuf, struct iosys_map *map); #endif /* __DMA_BUF_H__ */ diff --git a/include/linux/dma-resv.h b/include/linux/dma-resv.h index eebf04325b34..a715df97b31a 100644 --- a/include/linux/dma-resv.h +++ b/include/linux/dma-resv.h @@ -458,8 +458,8 @@ void dma_resv_fini(struct dma_resv *obj); int dma_resv_reserve_shared(struct dma_resv *obj, unsigned int num_fences); void dma_resv_add_shared_fence(struct dma_resv *obj, struct dma_fence *fence); void dma_resv_add_excl_fence(struct dma_resv *obj, struct dma_fence *fence); -int dma_resv_get_fences(struct dma_resv *obj, struct dma_fence **pfence_excl, - unsigned *pshared_count, struct dma_fence ***pshared); +int dma_resv_get_fences(struct dma_resv *obj, bool write, + unsigned int *num_fences, struct dma_fence ***fences); int dma_resv_copy_fences(struct dma_resv *dst, struct dma_resv *src); long dma_resv_wait_timeout(struct dma_resv *obj, bool wait_all, bool intr, unsigned long timeout); diff --git a/include/linux/fb.h b/include/linux/fb.h index 02f362c661c8..3d7306c9a706 100644 --- a/include/linux/fb.h +++ b/include/linux/fb.h @@ -502,6 +502,7 @@ struct fb_info { } *apertures; bool skip_vt_switch; /* no VT switch on suspend/resume required */ + bool forced_out; /* set when being removed by another driver */ }; static inline struct apertures_struct *alloc_apertures(unsigned int max_num) { diff --git a/include/linux/iosys-map.h b/include/linux/iosys-map.h new file mode 100644 index 000000000000..f4186f91caa6 --- /dev/null +++ b/include/linux/iosys-map.h @@ -0,0 +1,257 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Pointer abstraction for IO/system memory + */ + +#ifndef __IOSYS_MAP_H__ +#define __IOSYS_MAP_H__ + +#include <linux/io.h> +#include <linux/string.h> + +/** + * DOC: overview + * + * When accessing a memory region, depending on its location, users may have to + * access it with I/O operations or memory load/store operations. For example, + * copying to system memory could be done with memcpy(), copying to I/O memory + * would be done with memcpy_toio(). + * + * .. code-block:: c + * + * void *vaddr = ...; // pointer to system memory + * memcpy(vaddr, src, len); + * + * void *vaddr_iomem = ...; // pointer to I/O memory + * memcpy_toio(vaddr, _iomem, src, len); + * + * The user of such pointer may not have information about the mapping of that + * region or may want to have a single code path to handle operations on that + * buffer, regardless if it's located in system or IO memory. The type + * :c:type:`struct iosys_map <iosys_map>` and its helpers abstract that so the + * buffer can be passed around to other drivers or have separate duties inside + * the same driver for allocation, read and write operations. + * + * Open-coding access to :c:type:`struct iosys_map <iosys_map>` is considered + * bad style. Rather then accessing its fields directly, use one of the provided + * helper functions, or implement your own. For example, instances of + * :c:type:`struct iosys_map <iosys_map>` can be initialized statically with + * IOSYS_MAP_INIT_VADDR(), or at runtime with iosys_map_set_vaddr(). These + * helpers will set an address in system memory. + * + * .. code-block:: c + * + * struct iosys_map map = IOSYS_MAP_INIT_VADDR(0xdeadbeaf); + * + * iosys_map_set_vaddr(&map, 0xdeadbeaf); + * + * To set an address in I/O memory, use iosys_map_set_vaddr_iomem(). + * + * .. code-block:: c + * + * iosys_map_set_vaddr_iomem(&map, 0xdeadbeaf); + * + * Instances of struct iosys_map do not have to be cleaned up, but + * can be cleared to NULL with iosys_map_clear(). Cleared mappings + * always refer to system memory. + * + * .. code-block:: c + * + * iosys_map_clear(&map); + * + * Test if a mapping is valid with either iosys_map_is_set() or + * iosys_map_is_null(). + * + * .. code-block:: c + * + * if (iosys_map_is_set(&map) != iosys_map_is_null(&map)) + * // always true + * + * Instances of :c:type:`struct iosys_map <iosys_map>` can be compared for + * equality with iosys_map_is_equal(). Mappings that point to different memory + * spaces, system or I/O, are never equal. That's even true if both spaces are + * located in the same address space, both mappings contain the same address + * value, or both mappings refer to NULL. + * + * .. code-block:: c + * + * struct iosys_map sys_map; // refers to system memory + * struct iosys_map io_map; // refers to I/O memory + * + * if (iosys_map_is_equal(&sys_map, &io_map)) + * // always false + * + * A set up instance of struct iosys_map can be used to access or manipulate the + * buffer memory. Depending on the location of the memory, the provided helpers + * will pick the correct operations. Data can be copied into the memory with + * iosys_map_memcpy_to(). The address can be manipulated with iosys_map_incr(). + * + * .. code-block:: c + * + * const void *src = ...; // source buffer + * size_t len = ...; // length of src + * + * iosys_map_memcpy_to(&map, src, len); + * iosys_map_incr(&map, len); // go to first byte after the memcpy + */ + +/** + * struct iosys_map - Pointer to IO/system memory + * @vaddr_iomem: The buffer's address if in I/O memory + * @vaddr: The buffer's address if in system memory + * @is_iomem: True if the buffer is located in I/O memory, or false + * otherwise. + */ +struct iosys_map { + union { + void __iomem *vaddr_iomem; + void *vaddr; + }; + bool is_iomem; +}; + +/** + * IOSYS_MAP_INIT_VADDR - Initializes struct iosys_map to an address in system memory + * @vaddr_: A system-memory address + */ +#define IOSYS_MAP_INIT_VADDR(vaddr_) \ + { \ + .vaddr = (vaddr_), \ + .is_iomem = false, \ + } + +/** + * iosys_map_set_vaddr - Sets a iosys mapping structure to an address in system memory + * @map: The iosys_map structure + * @vaddr: A system-memory address + * + * Sets the address and clears the I/O-memory flag. + */ +static inline void iosys_map_set_vaddr(struct iosys_map *map, void *vaddr) +{ + map->vaddr = vaddr; + map->is_iomem = false; +} + +/** + * iosys_map_set_vaddr_iomem - Sets a iosys mapping structure to an address in I/O memory + * @map: The iosys_map structure + * @vaddr_iomem: An I/O-memory address + * + * Sets the address and the I/O-memory flag. + */ +static inline void iosys_map_set_vaddr_iomem(struct iosys_map *map, + void __iomem *vaddr_iomem) +{ + map->vaddr_iomem = vaddr_iomem; + map->is_iomem = true; +} + +/** + * iosys_map_is_equal - Compares two iosys mapping structures for equality + * @lhs: The iosys_map structure + * @rhs: A iosys_map structure to compare with + * + * Two iosys mapping structures are equal if they both refer to the same type of memory + * and to the same address within that memory. + * + * Returns: + * True is both structures are equal, or false otherwise. + */ +static inline bool iosys_map_is_equal(const struct iosys_map *lhs, + const struct iosys_map *rhs) +{ + if (lhs->is_iomem != rhs->is_iomem) + return false; + else if (lhs->is_iomem) + return lhs->vaddr_iomem == rhs->vaddr_iomem; + else + return lhs->vaddr == rhs->vaddr; +} + +/** + * iosys_map_is_null - Tests for a iosys mapping to be NULL + * @map: The iosys_map structure + * + * Depending on the state of struct iosys_map.is_iomem, tests if the + * mapping is NULL. + * + * Returns: + * True if the mapping is NULL, or false otherwise. + */ +static inline bool iosys_map_is_null(const struct iosys_map *map) +{ + if (map->is_iomem) + return !map->vaddr_iomem; + return !map->vaddr; +} + +/** + * iosys_map_is_set - Tests if the iosys mapping has been set + * @map: The iosys_map structure + * + * Depending on the state of struct iosys_map.is_iomem, tests if the + * mapping has been set. + * + * Returns: + * True if the mapping is been set, or false otherwise. + */ +static inline bool iosys_map_is_set(const struct iosys_map *map) +{ + return !iosys_map_is_null(map); +} + +/** + * iosys_map_clear - Clears a iosys mapping structure + * @map: The iosys_map structure + * + * Clears all fields to zero, including struct iosys_map.is_iomem, so + * mapping structures that were set to point to I/O memory are reset for + * system memory. Pointers are cleared to NULL. This is the default. + */ +static inline void iosys_map_clear(struct iosys_map *map) +{ + if (map->is_iomem) { + map->vaddr_iomem = NULL; + map->is_iomem = false; + } else { + map->vaddr = NULL; + } +} + +/** + * iosys_map_memcpy_to - Memcpy into iosys mapping + * @dst: The iosys_map structure + * @src: The source buffer + * @len: The number of byte in src + * + * Copies data into a iosys mapping. The source buffer is in system + * memory. Depending on the buffer's location, the helper picks the correct + * method of accessing the memory. + */ +static inline void iosys_map_memcpy_to(struct iosys_map *dst, const void *src, + size_t len) +{ + if (dst->is_iomem) + memcpy_toio(dst->vaddr_iomem, src, len); + else + memcpy(dst->vaddr, src, len); +} + +/** + * iosys_map_incr - Increments the address stored in a iosys mapping + * @map: The iosys_map structure + * @incr: The number of bytes to increment + * + * Increments the address stored in a iosys mapping. Depending on the + * buffer's location, the correct value will be updated. + */ +static inline void iosys_map_incr(struct iosys_map *map, size_t incr) +{ + if (map->is_iomem) + map->vaddr_iomem += incr; + else + map->vaddr += incr; +} + +#endif /* __IOSYS_MAP_H__ */ diff --git a/include/linux/rwsem.h b/include/linux/rwsem.h index f9348769e558..efa5c324369a 100644 --- a/include/linux/rwsem.h +++ b/include/linux/rwsem.h @@ -230,7 +230,7 @@ extern void _down_write_nest_lock(struct rw_semaphore *sem, struct lockdep_map * do { \ typecheck(struct lockdep_map *, &(nest_lock)->dep_map); \ _down_write_nest_lock(sem, &(nest_lock)->dep_map); \ -} while (0); +} while (0) /* * Take/release a lock when not the owner will release it. diff --git a/include/soc/bcm2835/raspberrypi-firmware.h b/include/soc/bcm2835/raspberrypi-firmware.h index 73ad784fca96..811ea668c4a1 100644 --- a/include/soc/bcm2835/raspberrypi-firmware.h +++ b/include/soc/bcm2835/raspberrypi-firmware.h @@ -91,6 +91,7 @@ enum rpi_firmware_property_tag { RPI_FIRMWARE_GET_POE_HAT_VAL = 0x00030049, RPI_FIRMWARE_SET_POE_HAT_VAL = 0x00030050, RPI_FIRMWARE_NOTIFY_XHCI_RESET = 0x00030058, + RPI_FIRMWARE_NOTIFY_DISPLAY_DONE = 0x00030066, /* Dispmanx TAGS */ RPI_FIRMWARE_FRAMEBUFFER_ALLOCATE = 0x00040001, diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index 0b94ec7b73e7..76b580d10a52 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -206,6 +206,8 @@ union drm_amdgpu_bo_list { #define AMDGPU_CTX_OP_FREE_CTX 2 #define AMDGPU_CTX_OP_QUERY_STATE 3 #define AMDGPU_CTX_OP_QUERY_STATE2 4 +#define AMDGPU_CTX_OP_GET_STABLE_PSTATE 5 +#define AMDGPU_CTX_OP_SET_STABLE_PSTATE 6 /* GPU reset status */ #define AMDGPU_CTX_NO_RESET 0 @@ -238,10 +240,18 @@ union drm_amdgpu_bo_list { #define AMDGPU_CTX_PRIORITY_HIGH 512 #define AMDGPU_CTX_PRIORITY_VERY_HIGH 1023 +/* select a stable profiling pstate for perfmon tools */ +#define AMDGPU_CTX_STABLE_PSTATE_FLAGS_MASK 0xf +#define AMDGPU_CTX_STABLE_PSTATE_NONE 0 +#define AMDGPU_CTX_STABLE_PSTATE_STANDARD 1 +#define AMDGPU_CTX_STABLE_PSTATE_MIN_SCLK 2 +#define AMDGPU_CTX_STABLE_PSTATE_MIN_MCLK 3 +#define AMDGPU_CTX_STABLE_PSTATE_PEAK 4 + struct drm_amdgpu_ctx_in { /** AMDGPU_CTX_OP_* */ __u32 op; - /** For future use, no flags defined so far */ + /** Flags */ __u32 flags; __u32 ctx_id; /** AMDGPU_CTX_PRIORITY_* */ @@ -262,6 +272,11 @@ union drm_amdgpu_ctx_out { /** Reset status since the last call of the ioctl. */ __u32 reset_status; } state; + + struct { + __u32 flags; + __u32 _pad; + } pstate; }; union drm_amdgpu_ctx { @@ -728,6 +743,8 @@ struct drm_amdgpu_cs_chunk_data { #define AMDGPU_INFO_FW_DMCUB 0x14 /* Subquery id: Query TOC firmware version */ #define AMDGPU_INFO_FW_TOC 0x15 + /* Subquery id: Query CAP firmware version */ + #define AMDGPU_INFO_FW_CAP 0x16 /* number of bytes moved for TTM migration */ #define AMDGPU_INFO_NUM_BYTES_MOVED 0x0f diff --git a/include/uapi/drm/panfrost_drm.h b/include/uapi/drm/panfrost_drm.h index 061e700dd06c..9e40277d8185 100644 --- a/include/uapi/drm/panfrost_drm.h +++ b/include/uapi/drm/panfrost_drm.h @@ -84,14 +84,14 @@ struct drm_panfrost_wait_bo { __s64 timeout_ns; /* absolute */ }; +/* Valid flags to pass to drm_panfrost_create_bo */ #define PANFROST_BO_NOEXEC 1 #define PANFROST_BO_HEAP 2 /** * struct drm_panfrost_create_bo - ioctl argument for creating Panfrost BOs. * - * There are currently no values for the flags argument, but it may be - * used in a future extension. + * The flags argument is a bit mask of PANFROST_BO_* flags. */ struct drm_panfrost_create_bo { __u32 size; diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h index af96af174dc4..6e4268f5e482 100644 --- a/include/uapi/linux/kfd_ioctl.h +++ b/include/uapi/linux/kfd_ioctl.h @@ -32,9 +32,10 @@ * - 1.4 - Indicate new SRAM EDC bit in device properties * - 1.5 - Add SVM API * - 1.6 - Query clear flags in SVM get_attr API + * - 1.7 - Checkpoint Restore (CRIU) API */ #define KFD_IOCTL_MAJOR_VERSION 1 -#define KFD_IOCTL_MINOR_VERSION 6 +#define KFD_IOCTL_MINOR_VERSION 7 struct kfd_ioctl_get_version_args { __u32 major_version; /* from KFD */ @@ -468,6 +469,82 @@ struct kfd_ioctl_smi_events_args { __u32 anon_fd; /* from KFD */ }; +/************************************************************************************************** + * CRIU IOCTLs (Checkpoint Restore In Userspace) + * + * When checkpointing a process, the userspace application will perform: + * 1. PROCESS_INFO op to determine current process information. This pauses execution and evicts + * all the queues. + * 2. CHECKPOINT op to checkpoint process contents (BOs, queues, events, svm-ranges) + * 3. UNPAUSE op to un-evict all the queues + * + * When restoring a process, the CRIU userspace application will perform: + * + * 1. RESTORE op to restore process contents + * 2. RESUME op to start the process + * + * Note: Queues are forced into an evicted state after a successful PROCESS_INFO. User + * application needs to perform an UNPAUSE operation after calling PROCESS_INFO. + */ + +enum kfd_criu_op { + KFD_CRIU_OP_PROCESS_INFO, + KFD_CRIU_OP_CHECKPOINT, + KFD_CRIU_OP_UNPAUSE, + KFD_CRIU_OP_RESTORE, + KFD_CRIU_OP_RESUME, +}; + +/** + * kfd_ioctl_criu_args - Arguments perform CRIU operation + * @devices: [in/out] User pointer to memory location for devices information. + * This is an array of type kfd_criu_device_bucket. + * @bos: [in/out] User pointer to memory location for BOs information + * This is an array of type kfd_criu_bo_bucket. + * @priv_data: [in/out] User pointer to memory location for private data + * @priv_data_size: [in/out] Size of priv_data in bytes + * @num_devices: [in/out] Number of GPUs used by process. Size of @devices array. + * @num_bos [in/out] Number of BOs used by process. Size of @bos array. + * @num_objects: [in/out] Number of objects used by process. Objects are opaque to + * user application. + * @pid: [in/out] PID of the process being checkpointed + * @op [in] Type of operation (kfd_criu_op) + * + * Return: 0 on success, -errno on failure + */ +struct kfd_ioctl_criu_args { + __u64 devices; /* Used during ops: CHECKPOINT, RESTORE */ + __u64 bos; /* Used during ops: CHECKPOINT, RESTORE */ + __u64 priv_data; /* Used during ops: CHECKPOINT, RESTORE */ + __u64 priv_data_size; /* Used during ops: PROCESS_INFO, RESTORE */ + __u32 num_devices; /* Used during ops: PROCESS_INFO, RESTORE */ + __u32 num_bos; /* Used during ops: PROCESS_INFO, RESTORE */ + __u32 num_objects; /* Used during ops: PROCESS_INFO, RESTORE */ + __u32 pid; /* Used during ops: PROCESS_INFO, RESUME */ + __u32 op; +}; + +struct kfd_criu_device_bucket { + __u32 user_gpu_id; + __u32 actual_gpu_id; + __u32 drm_fd; + __u32 pad; +}; + +struct kfd_criu_bo_bucket { + __u64 addr; + __u64 size; + __u64 offset; + __u64 restored_offset; /* During restore, updated offset for BO */ + __u32 gpu_id; /* This is the user_gpu_id */ + __u32 alloc_flags; + __u32 dmabuf_fd; + __u32 pad; +}; + +/* CRIU IOCTLs - END */ +/**************************************************************************************************/ + /* Register offset inside the remapped mmio page */ enum kfd_mmio_remap { @@ -679,16 +756,16 @@ struct kfd_ioctl_set_xnack_mode_args { #define AMDKFD_IOC_WAIT_EVENTS \ AMDKFD_IOWR(0x0C, struct kfd_ioctl_wait_events_args) -#define AMDKFD_IOC_DBG_REGISTER \ +#define AMDKFD_IOC_DBG_REGISTER_DEPRECATED \ AMDKFD_IOW(0x0D, struct kfd_ioctl_dbg_register_args) -#define AMDKFD_IOC_DBG_UNREGISTER \ +#define AMDKFD_IOC_DBG_UNREGISTER_DEPRECATED \ AMDKFD_IOW(0x0E, struct kfd_ioctl_dbg_unregister_args) -#define AMDKFD_IOC_DBG_ADDRESS_WATCH \ +#define AMDKFD_IOC_DBG_ADDRESS_WATCH_DEPRECATED \ AMDKFD_IOW(0x0F, struct kfd_ioctl_dbg_address_watch_args) -#define AMDKFD_IOC_DBG_WAVE_CONTROL \ +#define AMDKFD_IOC_DBG_WAVE_CONTROL_DEPRECATED \ AMDKFD_IOW(0x10, struct kfd_ioctl_dbg_wave_control_args) #define AMDKFD_IOC_SET_SCRATCH_BACKING_VA \ @@ -742,7 +819,10 @@ struct kfd_ioctl_set_xnack_mode_args { #define AMDKFD_IOC_SET_XNACK_MODE \ AMDKFD_IOWR(0x21, struct kfd_ioctl_set_xnack_mode_args) +#define AMDKFD_IOC_CRIU_OP \ + AMDKFD_IOWR(0x22, struct kfd_ioctl_criu_args) + #define AMDKFD_COMMAND_START 0x01 -#define AMDKFD_COMMAND_END 0x22 +#define AMDKFD_COMMAND_END 0x23 #endif |