summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2010-03-17libdrm_radeon: Optimize cs_gem_reloc to do less looping.radeon_reloc_optimizationPauli Nieminen8-41/+138
bo->referenced_in_cs is checked if bo is already in cs. Adding and removing reference in bo is done with atomic operations to allow parallel access to a bo from multiple contexts. cs->id generation code quarentees there is not duplicated ids which limits number of cs->ids to 32. If there is more cs objects rest will get id 0. V2: - Fix configure to check for atomics operations if libdrm_radeon is only selected. - Make atomic operations private to libdrm. This optimization decreases cs_write_reloc share of torcs profiling from 4.3% to 2.6%. Tested-by: Michel Dänzer <michel@daenzer.net> Signed-off-by: Pauli Nieminen <suokkos@gmail.com>
2010-03-17libdrm: Move intel_atomic.h to libdrm core for sharing.Pauli Nieminen5-8/+8
intel_atomic.h includes very usefull atomic operations for lock free parrallel access of variables. Moving these to core libdrm for code sharing with radeon. Signed-off-by: Pauli Nieminen <suokkos@gmail.com>
2010-03-13nouveau: Fix up the stride of NV20TCL_LIGHT_BACK_*.Francisco Jerez1-6/+6
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
2010-03-07nouveau: Small lighting related addition to nouveau_class.h.Francisco Jerez1-0/+1
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
2010-03-07intel: Repeat execbuffer if interrupted by signalChris Wilson1-1/+1
Repeat while EINTR, not EAGAIN! One more source of corruption erradicated, hurray! Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2010-03-06nouveau: Update nouveau_class.h.Francisco Jerez1-71/+33
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
2010-03-04intel: Only align Y-tiling pitch to the Y tile width.Eric Anholt1-1/+6
Fixes piglit depth-tex-modes on gen4.
2010-03-04intel: Propagate some more error returnsChris Wilson1-12/+20
Ensure that errors from the kernel are propagated back to the caller, and not masked with return 0; Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2010-03-03Increment version to 2.4.19 for release.Eric Anholt1-1/+1
The primary motivation of this release is to expose the new execbuf2 Intel API.
2010-03-03intel: Update the needs_fence flag of buffers on the validate list.Eric Anholt1-1/+5
Fixes fbo-copyteximage on i915 with texture tiling and execbuf2 fenced relocs.
2010-03-02intel: Don't enable execbuf2 fenced relocs unless we have execbuf2.Eric Anholt1-2/+3
2010-03-02intel: Don't tile-align pitch for untiled buffers.Eric Anholt1-1/+1
This allows Mesa to use drm_intel_bo_alloc_tiled() for its tiled buffers, since it makes its decision about pitch before telling libdrm. They happen to be the same choices for the tiled case.
2010-03-02intel: Fix typo in conversion from IS_GEN to bufmgr_gem->gen.Eric Anholt1-1/+1
Luckily I caught the bug with the first consumer of the interface.
2010-03-02intel: add a comment about tiled buffer alloc height alignment from Mesa.Eric Anholt1-0/+11
2010-03-02nouveau: make sure initial kalloc for user bo ends up in the right placeMaarten Maathuis1-0/+6
- Currently reloc'ing a user bo to gart will first cause an allocation in vram, which is then written to by cpu, then the bo gets moved to gart. Acked-by: Francisco Jerez <currojerez@riseup.net> Signed-off-by: Maarten Maathuis <madman2003@gmail.com>
2010-03-02intel: Use an integer for chipset generation instead of many conditionals.Eric Anholt1-8/+17
Saves a bunch of comparisons in hot paths.
2010-03-02libdrm/intel: execbuf2 supportJesse Barnes6-35/+331
This patch to libdrm adds support for the new execbuf2 ioctl. If detected, it will be used instead of the old ioctl. By using the new drm_intel_bufmgr_gem_enable_fenced_relocs(), you can indicate that any time a fence register is actually required for a relocation target you will call drm_intel_bo_emit_reloc_fence instead of drm_intel_bo_emit_reloc, which will reduce fence register pressure. Signed-off-by: Eric Anholt <eric@anholt.net>
2010-02-28tests/modetest: Don't get_props on a connector after freeing it.Marcin Kościelnicki1-2/+2
2010-02-28tests: vrefresh is actually not * 1000.Marcin Kościelnicki3-5/+5
2010-02-28tests: Add nouveau to list of supported modules.Marcin Kościelnicki2-1/+2
2010-02-28libkms/nouveau: Add supportMarcin Kościelnicki4-0/+230
2010-02-28libkms/intel: Throw out unused intel_bo fields.Marcin Kościelnicki1-2/+0
Acked-by: Jakob Bornecrantz <jakob@vmware.com>
2010-02-28Add config.h macro HAVE_NOUVEAUMarcin Kościelnicki1-0/+3
2010-02-25intel: Add initial support for Sandybridge, and clean up the #defines.Eric Anholt2-38/+63
2010-02-25nouveau: Update nouveau_class.h.Francisco Jerez1-43/+36
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
2010-02-20nv30: update for 8 texture unitsPatrice Mandin1-9/+9
2010-02-20nv30: update for front/back stencil inversionPatrice Mandin1-48/+48
Signed-off-by: Patrice Mandin <patmandin@gmail.com>
2010-02-18libkms: Use a standard version numberJakob Bornecrantz1-1/+1
2010-02-18libkms: Enable by defaultJakob Bornecrantz1-3/+3
2010-02-18libkms: Add missing include file to libkms sourceJakob Bornecrantz1-0/+1
2010-02-18vmwgfx: Update kernel headerJakob Bornecrantz1-46/+17
2010-02-18radeon: add square-tiling flagMarek Olšák2-0/+2
2010-02-16Increment version to 2.4.18 for release.Carl Worth1-1/+1
The primary motivation of the release is the bug fix in commit 4f0f871730b76730ca58209181d16725b0c40184
2010-02-16nouveau: bump MAX_PUSH to 512Ben Skeggs1-1/+1
2010-02-16nouveau: interface changes for 0.0.16 DRMLuca Barbieri11-404/+376
This commit encompasses the changes necessary to run on top of the 0.0.16 nouveau interface, additional APIs to support the new features of the interface, as well as code from Luca Barbieri to improve the pushbuf interface, which just happens to break nouveau's libdrm ABI so was delayed until now. API changes as a result of 0.0.16 DRM interface: 1. No more bo_pin()/bo_unpin(), these were only there for UMS and we no longer support it. 2. Any random nouveau_bo can be submitted to the GPU as a push buffer. 3. Relocations can be applied on any nouveau_bo This patch changes the pushbuffer ABI to: 1. No longer use/expose nouveau_pushbuffer. Everything is directly in nouveau_channel. This saves the extra "pushbuf" pointer dereference. 2. Use cur/end pointers instead of tracking the remaining size. Pushing data now only needs to alter cur and not both cur and remaining. The goal is to make the *_RING macros faster and make the interface simpler and cleaner in the process. The *_RING APIs are unchanged, but those are inlined and the ABI is changed. Also, anything accessing pushbuf->remaining instead of using AVAIL_RING will need to be fixed.
2010-02-10drm: a some new connector types from the kernelAlex Deucher1-0/+2
Add eDP (embedded displayport) and generic TV Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
2010-02-10intel: Handle resetting of input params after EINTR during SET_TILINGChris Wilson1-7/+4
The SET_TILING is pernicious in that it overwrites the input arguments following an error in order to report the current tiling state of the buffer. This caught us by surprise as we then fed those arguments back into to the ioctl unmodified following an EINTR and so the kernel then reported success for the no-op. We interpreted this success as meaning that the tiling on the buffer had changed so updated our state and started using the buffer incorrectly in the new tiled/untiled manner. This lead to all sorts of random corruption and GPU hangs, even though the batch buffers would look sane (when the GPU had not wandered off into forbidden territory). References: Bug 25475 - [i915] Xorg crash / Execbuf while wedged http://bugs.freedesktop.org/show_bug.cgi?id=25475 Bug 25554 - i830_uxa_prepare_access: gtt bo map failed: Input/output error http://bugs.freedesktop.org/show_bug.cgi?id=25554 (And probably every other weird bug in the last few months.) Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2010-02-09intel: Account for potential pinned buffers hogging fencesChris Wilson1-0/+13
As the kernel reports the total number of fences, we must guess how many fences are likely to be pinned. In the typical system these will be only used by the scanout buffers, of which there may be one per pipe, and any number of manually pinned fenced buffers. So take a conservative guess and reserve two fences for use by the system. Note this reduces the number of fences to 3 for i915 and prior. Reference: http://bugs.freedesktop.org/show_bug.cgi?id=25911 The latest intel driver 2.10.0 causes kernel oops and system hangs Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2010-02-04Tidy up compile warnings by cleaning up types.Matthew W. S. Bell5-30/+26
2010-02-04libdrm/radeon: Fix section size mismatch to reset the section.Pauli Nieminen1-1/+4
If there is section size mismatch reusing the section object makes section start fail. Reseting the object before doing error checking prevents the possible flood of errors.
2010-02-02radeon: enable by default now that kms is out of stagingDave Airlie2-5/+6
2010-02-02intel: check return value for callocDave Airlie1-0/+3
2010-02-01nouveau: Regenerate nouveau_class.h.Francisco Jerez1-491/+541
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
2010-01-31nouveau: add nouveau_resource_destroyMarcin Slusarz2-0/+12
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
2010-01-28libkms: Rework interface to not duplicate fields from kms and make formats ↵Jakob Bornecrantz5-69/+31
explicit List of changes: Fixes the cursor size to 64x64, you still need ti supply width and height Explicitly make the cursor format A8R8G8B8 Explicitly make the scanout format X8R8G8B8
2010-01-27nouveau: Update nouveau_drm.hMarcin Kościelnicki1-0/+1
2010-01-23libkms: Fix return value in vmwgfx_bo_createJakob Bornecrantz1-1/+1
2010-01-21nouveau: fail channel creation if pushbuf init failsBen Skeggs1-1/+5
2010-01-20intel: Add pkg-config dependency on libdrm.so using 'Requires'Chris Wilson1-1/+2
2010-01-16Support gcc's __FUNCTION__ for people using other compilersAlan Coopersmith2-0/+10
Signed-off-by: Alan Coopersmith <alan.coopersmith@sun.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-21 12:42:53 +0000