diff options
Diffstat (limited to 'arch/x86/hyperv')
-rw-r--r-- | arch/x86/hyperv/Makefile | 1 | ||||
-rw-r--r-- | arch/x86/hyperv/hv_apic.c | 12 | ||||
-rw-r--r-- | arch/x86/hyperv/hv_init.c | 18 | ||||
-rw-r--r-- | arch/x86/hyperv/hv_vtl.c | 227 | ||||
-rw-r--r-- | arch/x86/hyperv/ivm.c | 148 | ||||
-rw-r--r-- | arch/x86/hyperv/mmu.c | 11 |
6 files changed, 336 insertions, 81 deletions
diff --git a/arch/x86/hyperv/Makefile b/arch/x86/hyperv/Makefile index 5d2de10809ae..3a1548054b48 100644 --- a/arch/x86/hyperv/Makefile +++ b/arch/x86/hyperv/Makefile @@ -1,6 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only obj-y := hv_init.o mmu.o nested.o irqdomain.o ivm.o obj-$(CONFIG_X86_64) += hv_apic.o hv_proc.o +obj-$(CONFIG_HYPERV_VTL_MODE) += hv_vtl.o ifdef CONFIG_X86_64 obj-$(CONFIG_PARAVIRT_SPINLOCKS) += hv_spinlock.o diff --git a/arch/x86/hyperv/hv_apic.c b/arch/x86/hyperv/hv_apic.c index fb8b2c088681..1fbda2f94184 100644 --- a/arch/x86/hyperv/hv_apic.c +++ b/arch/x86/hyperv/hv_apic.c @@ -96,6 +96,11 @@ static void hv_apic_eoi_write(u32 reg, u32 val) wrmsr(HV_X64_MSR_EOI, val, 0); } +static bool cpu_is_self(int cpu) +{ + return cpu == smp_processor_id(); +} + /* * IPI implementation on Hyper-V. */ @@ -128,10 +133,9 @@ static bool __send_ipi_mask_ex(const struct cpumask *mask, int vector, */ if (!cpumask_equal(mask, cpu_present_mask) || exclude_self) { ipi_arg->vp_set.format = HV_GENERIC_SET_SPARSE_4K; - if (exclude_self) - nr_bank = cpumask_to_vpset_noself(&(ipi_arg->vp_set), mask); - else - nr_bank = cpumask_to_vpset(&(ipi_arg->vp_set), mask); + + nr_bank = cpumask_to_vpset_skip(&(ipi_arg->vp_set), mask, + exclude_self ? cpu_is_self : NULL); /* * 'nr_bank <= 0' means some CPUs in cpumask can't be diff --git a/arch/x86/hyperv/hv_init.c b/arch/x86/hyperv/hv_init.c index 41ef036ebb7b..a5f9474f08e1 100644 --- a/arch/x86/hyperv/hv_init.c +++ b/arch/x86/hyperv/hv_init.c @@ -29,7 +29,6 @@ #include <linux/syscore_ops.h> #include <clocksource/hyperv_timer.h> #include <linux/highmem.h> -#include <linux/swiotlb.h> int hyperv_init_cpuhp; u64 hv_current_partition_id = ~0ull; @@ -64,7 +63,10 @@ static int hyperv_init_ghcb(void) * memory boundary and map it here. */ rdmsrl(MSR_AMD64_SEV_ES_GHCB, ghcb_gpa); - ghcb_va = memremap(ghcb_gpa, HV_HYP_PAGE_SIZE, MEMREMAP_WB); + + /* Mask out vTOM bit. ioremap_cache() maps decrypted */ + ghcb_gpa &= ~ms_hyperv.shared_gpa_boundary; + ghcb_va = (void *)ioremap_cache(ghcb_gpa, HV_HYP_PAGE_SIZE); if (!ghcb_va) return -ENOMEM; @@ -218,7 +220,7 @@ static int hv_cpu_die(unsigned int cpu) if (hv_ghcb_pg) { ghcb_va = (void **)this_cpu_ptr(hv_ghcb_pg); if (*ghcb_va) - memunmap(*ghcb_va); + iounmap(*ghcb_va); *ghcb_va = NULL; } @@ -504,16 +506,6 @@ void __init hyperv_init(void) /* Query the VMs extended capability once, so that it can be cached. */ hv_query_ext_cap(0); -#ifdef CONFIG_SWIOTLB - /* - * Swiotlb bounce buffer needs to be mapped in extra address - * space. Map function doesn't work in the early place and so - * call swiotlb_update_mem_attributes() here. - */ - if (hv_is_isolation_supported()) - swiotlb_update_mem_attributes(); -#endif - return; clean_guest_os_id: diff --git a/arch/x86/hyperv/hv_vtl.c b/arch/x86/hyperv/hv_vtl.c new file mode 100644 index 000000000000..1ba5d3b99b16 --- /dev/null +++ b/arch/x86/hyperv/hv_vtl.c @@ -0,0 +1,227 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2023, Microsoft Corporation. + * + * Author: + * Saurabh Sengar <ssengar@microsoft.com> + */ + +#include <asm/apic.h> +#include <asm/boot.h> +#include <asm/desc.h> +#include <asm/i8259.h> +#include <asm/mshyperv.h> +#include <asm/realmode.h> + +extern struct boot_params boot_params; +static struct real_mode_header hv_vtl_real_mode_header; + +void __init hv_vtl_init_platform(void) +{ + pr_info("Linux runs in Hyper-V Virtual Trust Level\n"); + + x86_init.irqs.pre_vector_init = x86_init_noop; + x86_init.timers.timer_init = x86_init_noop; + + x86_platform.get_wallclock = get_rtc_noop; + x86_platform.set_wallclock = set_rtc_noop; + x86_platform.get_nmi_reason = hv_get_nmi_reason; + + x86_platform.legacy.i8042 = X86_LEGACY_I8042_PLATFORM_ABSENT; + x86_platform.legacy.rtc = 0; + x86_platform.legacy.warm_reset = 0; + x86_platform.legacy.reserve_bios_regions = 0; + x86_platform.legacy.devices.pnpbios = 0; +} + +static inline u64 hv_vtl_system_desc_base(struct ldttss_desc *desc) +{ + return ((u64)desc->base3 << 32) | ((u64)desc->base2 << 24) | + (desc->base1 << 16) | desc->base0; +} + +static inline u32 hv_vtl_system_desc_limit(struct ldttss_desc *desc) +{ + return ((u32)desc->limit1 << 16) | (u32)desc->limit0; +} + +typedef void (*secondary_startup_64_fn)(void*, void*); +static void hv_vtl_ap_entry(void) +{ + ((secondary_startup_64_fn)secondary_startup_64)(&boot_params, &boot_params); +} + +static int hv_vtl_bringup_vcpu(u32 target_vp_index, u64 eip_ignored) +{ + u64 status; + int ret = 0; + struct hv_enable_vp_vtl *input; + unsigned long irq_flags; + + struct desc_ptr gdt_ptr; + struct desc_ptr idt_ptr; + + struct ldttss_desc *tss; + struct ldttss_desc *ldt; + struct desc_struct *gdt; + + u64 rsp = current->thread.sp; + u64 rip = (u64)&hv_vtl_ap_entry; + + native_store_gdt(&gdt_ptr); + store_idt(&idt_ptr); + + gdt = (struct desc_struct *)((void *)(gdt_ptr.address)); + tss = (struct ldttss_desc *)(gdt + GDT_ENTRY_TSS); + ldt = (struct ldttss_desc *)(gdt + GDT_ENTRY_LDT); + + local_irq_save(irq_flags); + + input = *this_cpu_ptr(hyperv_pcpu_input_arg); + memset(input, 0, sizeof(*input)); + + input->partition_id = HV_PARTITION_ID_SELF; + input->vp_index = target_vp_index; + input->target_vtl.target_vtl = HV_VTL_MGMT; + + /* + * The x86_64 Linux kernel follows the 16-bit -> 32-bit -> 64-bit + * mode transition sequence after waking up an AP with SIPI whose + * vector points to the 16-bit AP startup trampoline code. Here in + * VTL2, we can't perform that sequence as the AP has to start in + * the 64-bit mode. + * + * To make this happen, we tell the hypervisor to load a valid 64-bit + * context (most of which is just magic numbers from the CPU manual) + * so that AP jumps right to the 64-bit entry of the kernel, and the + * control registers are loaded with values that let the AP fetch the + * code and data and carry on with work it gets assigned. + */ + + input->vp_context.rip = rip; + input->vp_context.rsp = rsp; + input->vp_context.rflags = 0x0000000000000002; + input->vp_context.efer = __rdmsr(MSR_EFER); + input->vp_context.cr0 = native_read_cr0(); + input->vp_context.cr3 = __native_read_cr3(); + input->vp_context.cr4 = native_read_cr4(); + input->vp_context.msr_cr_pat = __rdmsr(MSR_IA32_CR_PAT); + input->vp_context.idtr.limit = idt_ptr.size; + input->vp_context.idtr.base = idt_ptr.address; + input->vp_context.gdtr.limit = gdt_ptr.size; + input->vp_context.gdtr.base = gdt_ptr.address; + + /* Non-system desc (64bit), long, code, present */ + input->vp_context.cs.selector = __KERNEL_CS; + input->vp_context.cs.base = 0; + input->vp_context.cs.limit = 0xffffffff; + input->vp_context.cs.attributes = 0xa09b; + /* Non-system desc (64bit), data, present, granularity, default */ + input->vp_context.ss.selector = __KERNEL_DS; + input->vp_context.ss.base = 0; + input->vp_context.ss.limit = 0xffffffff; + input->vp_context.ss.attributes = 0xc093; + + /* System desc (128bit), present, LDT */ + input->vp_context.ldtr.selector = GDT_ENTRY_LDT * 8; + input->vp_context.ldtr.base = hv_vtl_system_desc_base(ldt); + input->vp_context.ldtr.limit = hv_vtl_system_desc_limit(ldt); + input->vp_context.ldtr.attributes = 0x82; + + /* System desc (128bit), present, TSS, 0x8b - busy, 0x89 -- default */ + input->vp_context.tr.selector = GDT_ENTRY_TSS * 8; + input->vp_context.tr.base = hv_vtl_system_desc_base(tss); + input->vp_context.tr.limit = hv_vtl_system_desc_limit(tss); + input->vp_context.tr.attributes = 0x8b; + + status = hv_do_hypercall(HVCALL_ENABLE_VP_VTL, input, NULL); + + if (!hv_result_success(status) && + hv_result(status) != HV_STATUS_VTL_ALREADY_ENABLED) { + pr_err("HVCALL_ENABLE_VP_VTL failed for VP : %d ! [Err: %#llx\n]", + target_vp_index, status); + ret = -EINVAL; + goto free_lock; + } + + status = hv_do_hypercall(HVCALL_START_VP, input, NULL); + + if (!hv_result_success(status)) { + pr_err("HVCALL_START_VP failed for VP : %d ! [Err: %#llx]\n", + target_vp_index, status); + ret = -EINVAL; + } + +free_lock: + local_irq_restore(irq_flags); + + return ret; +} + +static int hv_vtl_apicid_to_vp_id(u32 apic_id) +{ + u64 control; + u64 status; + unsigned long irq_flags; + struct hv_get_vp_from_apic_id_in *input; + u32 *output, ret; + + local_irq_save(irq_flags); + + input = *this_cpu_ptr(hyperv_pcpu_input_arg); + memset(input, 0, sizeof(*input)); + input->partition_id = HV_PARTITION_ID_SELF; + input->apic_ids[0] = apic_id; + + output = (u32 *)input; + + control = HV_HYPERCALL_REP_COMP_1 | HVCALL_GET_VP_ID_FROM_APIC_ID; + status = hv_do_hypercall(control, input, output); + ret = output[0]; + + local_irq_restore(irq_flags); + + if (!hv_result_success(status)) { + pr_err("failed to get vp id from apic id %d, status %#llx\n", + apic_id, status); + return -EINVAL; + } + + return ret; +} + +static int hv_vtl_wakeup_secondary_cpu(int apicid, unsigned long start_eip) +{ + int vp_id; + + pr_debug("Bringing up CPU with APIC ID %d in VTL2...\n", apicid); + vp_id = hv_vtl_apicid_to_vp_id(apicid); + + if (vp_id < 0) { + pr_err("Couldn't find CPU with APIC ID %d\n", apicid); + return -EINVAL; + } + if (vp_id > ms_hyperv.max_vp_index) { + pr_err("Invalid CPU id %d for APIC ID %d\n", vp_id, apicid); + return -EINVAL; + } + + return hv_vtl_bringup_vcpu(vp_id, start_eip); +} + +static int __init hv_vtl_early_init(void) +{ + /* + * `boot_cpu_has` returns the runtime feature support, + * and here is the earliest it can be used. + */ + if (cpu_feature_enabled(X86_FEATURE_XSAVE)) + panic("XSAVE has to be disabled as it is not supported by this module.\n" + "Please add 'noxsave' to the kernel command line.\n"); + + real_mode_header = &hv_vtl_real_mode_header; + apic->wakeup_secondary_cpu_64 = hv_vtl_wakeup_secondary_cpu; + + return 0; +} +early_initcall(hv_vtl_early_init); diff --git a/arch/x86/hyperv/ivm.c b/arch/x86/hyperv/ivm.c index 4f79dc76042d..cc92388b7a99 100644 --- a/arch/x86/hyperv/ivm.c +++ b/arch/x86/hyperv/ivm.c @@ -13,6 +13,8 @@ #include <asm/svm.h> #include <asm/sev.h> #include <asm/io.h> +#include <asm/coco.h> +#include <asm/mem_encrypt.h> #include <asm/mshyperv.h> #include <asm/hypervisor.h> @@ -233,41 +235,6 @@ void hv_ghcb_msr_read(u64 msr, u64 *value) local_irq_restore(flags); } EXPORT_SYMBOL_GPL(hv_ghcb_msr_read); -#endif - -enum hv_isolation_type hv_get_isolation_type(void) -{ - if (!(ms_hyperv.priv_high & HV_ISOLATION)) - return HV_ISOLATION_TYPE_NONE; - return FIELD_GET(HV_ISOLATION_TYPE, ms_hyperv.isolation_config_b); -} -EXPORT_SYMBOL_GPL(hv_get_isolation_type); - -/* - * hv_is_isolation_supported - Check system runs in the Hyper-V - * isolation VM. - */ -bool hv_is_isolation_supported(void) -{ - if (!cpu_feature_enabled(X86_FEATURE_HYPERVISOR)) - return false; - - if (!hypervisor_is_type(X86_HYPER_MS_HYPERV)) - return false; - - return hv_get_isolation_type() != HV_ISOLATION_TYPE_NONE; -} - -DEFINE_STATIC_KEY_FALSE(isolation_type_snp); - -/* - * hv_isolation_type_snp - Check system runs in the AMD SEV-SNP based - * isolation VM. - */ -bool hv_isolation_type_snp(void) -{ - return static_branch_unlikely(&isolation_type_snp); -} /* * hv_mark_gpa_visibility - Set pages visible to host via hvcall. @@ -320,27 +287,25 @@ static int hv_mark_gpa_visibility(u16 count, const u64 pfn[], } /* - * hv_set_mem_host_visibility - Set specified memory visible to host. + * hv_vtom_set_host_visibility - Set specified memory visible to host. * * In Isolation VM, all guest memory is encrypted from host and guest * needs to set memory visible to host via hvcall before sharing memory * with host. This function works as wrap of hv_mark_gpa_visibility() * with memory base and size. */ -int hv_set_mem_host_visibility(unsigned long kbuffer, int pagecount, bool visible) +static bool hv_vtom_set_host_visibility(unsigned long kbuffer, int pagecount, bool enc) { - enum hv_mem_host_visibility visibility = visible ? - VMBUS_PAGE_VISIBLE_READ_WRITE : VMBUS_PAGE_NOT_VISIBLE; + enum hv_mem_host_visibility visibility = enc ? + VMBUS_PAGE_NOT_VISIBLE : VMBUS_PAGE_VISIBLE_READ_WRITE; u64 *pfn_array; int ret = 0; + bool result = true; int i, pfn; - if (!hv_is_isolation_supported() || !hv_hypercall_pg) - return 0; - pfn_array = kmalloc(HV_HYP_PAGE_SIZE, GFP_KERNEL); if (!pfn_array) - return -ENOMEM; + return false; for (i = 0, pfn = 0; i < pagecount; i++) { pfn_array[pfn] = virt_to_hvpfn((void *)kbuffer + i * HV_HYP_PAGE_SIZE); @@ -349,41 +314,98 @@ int hv_set_mem_host_visibility(unsigned long kbuffer, int pagecount, bool visibl if (pfn == HV_MAX_MODIFY_GPA_REP_COUNT || i == pagecount - 1) { ret = hv_mark_gpa_visibility(pfn, pfn_array, visibility); - if (ret) + if (ret) { + result = false; goto err_free_pfn_array; + } pfn = 0; } } err_free_pfn_array: kfree(pfn_array); - return ret; + return result; } -/* - * hv_map_memory - map memory to extra space in the AMD SEV-SNP Isolation VM. - */ -void *hv_map_memory(void *addr, unsigned long size) +static bool hv_vtom_tlb_flush_required(bool private) { - unsigned long *pfns = kcalloc(size / PAGE_SIZE, - sizeof(unsigned long), GFP_KERNEL); - void *vaddr; - int i; + return true; +} + +static bool hv_vtom_cache_flush_required(void) +{ + return false; +} - if (!pfns) - return NULL; +static bool hv_is_private_mmio(u64 addr) +{ + /* + * Hyper-V always provides a single IO-APIC in a guest VM. + * When a paravisor is used, it is emulated by the paravisor + * in the guest context and must be mapped private. + */ + if (addr >= HV_IOAPIC_BASE_ADDRESS && + addr < (HV_IOAPIC_BASE_ADDRESS + PAGE_SIZE)) + return true; + + /* Same with a vTPM */ + if (addr >= VTPM_BASE_ADDRESS && + addr < (VTPM_BASE_ADDRESS + PAGE_SIZE)) + return true; + + return false; +} + +void __init hv_vtom_init(void) +{ + /* + * By design, a VM using vTOM doesn't see the SEV setting, + * so SEV initialization is bypassed and sev_status isn't set. + * Set it here to indicate a vTOM VM. + */ + sev_status = MSR_AMD64_SNP_VTOM; + cc_set_vendor(CC_VENDOR_AMD); + cc_set_mask(ms_hyperv.shared_gpa_boundary); + physical_mask &= ms_hyperv.shared_gpa_boundary - 1; + + x86_platform.hyper.is_private_mmio = hv_is_private_mmio; + x86_platform.guest.enc_cache_flush_required = hv_vtom_cache_flush_required; + x86_platform.guest.enc_tlb_flush_required = hv_vtom_tlb_flush_required; + x86_platform.guest.enc_status_change_finish = hv_vtom_set_host_visibility; +} + +#endif /* CONFIG_AMD_MEM_ENCRYPT */ + +enum hv_isolation_type hv_get_isolation_type(void) +{ + if (!(ms_hyperv.priv_high & HV_ISOLATION)) + return HV_ISOLATION_TYPE_NONE; + return FIELD_GET(HV_ISOLATION_TYPE, ms_hyperv.isolation_config_b); +} +EXPORT_SYMBOL_GPL(hv_get_isolation_type); - for (i = 0; i < size / PAGE_SIZE; i++) - pfns[i] = vmalloc_to_pfn(addr + i * PAGE_SIZE) + - (ms_hyperv.shared_gpa_boundary >> PAGE_SHIFT); +/* + * hv_is_isolation_supported - Check system runs in the Hyper-V + * isolation VM. + */ +bool hv_is_isolation_supported(void) +{ + if (!cpu_feature_enabled(X86_FEATURE_HYPERVISOR)) + return false; - vaddr = vmap_pfn(pfns, size / PAGE_SIZE, PAGE_KERNEL_IO); - kfree(pfns); + if (!hypervisor_is_type(X86_HYPER_MS_HYPERV)) + return false; - return vaddr; + return hv_get_isolation_type() != HV_ISOLATION_TYPE_NONE; } -void hv_unmap_memory(void *addr) +DEFINE_STATIC_KEY_FALSE(isolation_type_snp); + +/* + * hv_isolation_type_snp - Check system runs in the AMD SEV-SNP based + * isolation VM. + */ +bool hv_isolation_type_snp(void) { - vunmap(addr); + return static_branch_unlikely(&isolation_type_snp); } diff --git a/arch/x86/hyperv/mmu.c b/arch/x86/hyperv/mmu.c index 0ad2378fe6ad..8460bd35e10c 100644 --- a/arch/x86/hyperv/mmu.c +++ b/arch/x86/hyperv/mmu.c @@ -52,6 +52,11 @@ static inline int fill_gva_list(u64 gva_list[], int offset, return gva_n - offset; } +static bool cpu_is_lazy(int cpu) +{ + return per_cpu(cpu_tlbstate_shared.is_lazy, cpu); +} + static void hyperv_flush_tlb_multi(const struct cpumask *cpus, const struct flush_tlb_info *info) { @@ -60,6 +65,7 @@ static void hyperv_flush_tlb_multi(const struct cpumask *cpus, struct hv_tlb_flush *flush; u64 status; unsigned long flags; + bool do_lazy = !info->freed_tables; trace_hyperv_mmu_flush_tlb_multi(cpus, info); @@ -112,6 +118,8 @@ static void hyperv_flush_tlb_multi(const struct cpumask *cpus, goto do_ex_hypercall; for_each_cpu(cpu, cpus) { + if (do_lazy && cpu_is_lazy(cpu)) + continue; vcpu = hv_cpu_number_to_vp_number(cpu); if (vcpu == VP_INVAL) { local_irq_restore(flags); @@ -198,7 +206,8 @@ static u64 hyperv_flush_tlb_others_ex(const struct cpumask *cpus, flush->hv_vp_set.valid_bank_mask = 0; flush->hv_vp_set.format = HV_GENERIC_SET_SPARSE_4K; - nr_bank = cpumask_to_vpset(&(flush->hv_vp_set), cpus); + nr_bank = cpumask_to_vpset_skip(&flush->hv_vp_set, cpus, + info->freed_tables ? NULL : cpu_is_lazy); if (nr_bank < 0) return HV_STATUS_INVALID_PARAMETER; |