diff options
Diffstat (limited to 'arch/x86/kvm/svm/svm.c')
-rw-r--r-- | arch/x86/kvm/svm/svm.c | 2596 |
1 files changed, 1485 insertions, 1111 deletions
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index 58a45bb139f8..87da90360bc7 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -25,6 +25,7 @@ #include <linux/pagemap.h> #include <linux/swap.h> #include <linux/rwsem.h> +#include <linux/cc_platform.h> #include <asm/apic.h> #include <asm/perf_event.h> @@ -36,6 +37,7 @@ #include <asm/spec-ctrl.h> #include <asm/cpu_device_id.h> #include <asm/traps.h> +#include <asm/fpu/api.h> #include <asm/virtext.h> #include "trace.h" @@ -43,7 +45,8 @@ #include "svm.h" #include "svm_ops.h" -#define __ex(x) __kvm_handle_fault_on_reboot(x) +#include "kvm_onhyperv.h" +#include "svm_onhyperv.h" MODULE_AUTHOR("Qumranet"); MODULE_LICENSE("GPL"); @@ -56,26 +59,9 @@ static const struct x86_cpu_id svm_cpu_id[] = { MODULE_DEVICE_TABLE(x86cpu, svm_cpu_id); #endif -#define IOPM_ALLOC_ORDER 2 -#define MSRPM_ALLOC_ORDER 1 - #define SEG_TYPE_LDT 2 #define SEG_TYPE_BUSY_TSS16 3 -#define SVM_FEATURE_LBRV (1 << 1) -#define SVM_FEATURE_SVML (1 << 2) -#define SVM_FEATURE_TSC_RATE (1 << 4) -#define SVM_FEATURE_VMCB_CLEAN (1 << 5) -#define SVM_FEATURE_FLUSH_ASID (1 << 6) -#define SVM_FEATURE_DECODE_ASSIST (1 << 7) -#define SVM_FEATURE_PAUSE_FILTER (1 << 10) - -#define DEBUGCTL_RESERVED_BITS (~(0x3fULL)) - -#define TSC_RATIO_RSVD 0xffffff0000000000ULL -#define TSC_RATIO_MIN 0x0000000000000001ULL -#define TSC_RATIO_MAX 0x000000ffffffffffULL - static bool erratum_383_found __read_mostly; u32 msrpm_offsets[MSRPM_OFFSETS] __read_mostly; @@ -87,7 +73,6 @@ u32 msrpm_offsets[MSRPM_OFFSETS] __read_mostly; static uint64_t osvw_len = 4, osvw_status; static DEFINE_PER_CPU(u64, current_tsc_ratio); -#define TSC_RATIO_DEFAULT 0x0100000000ULL static const struct svm_direct_access_msrs { u32 index; /* Index of the MSR */ @@ -95,6 +80,8 @@ static const struct svm_direct_access_msrs { } direct_access_msrs[MAX_DIRECT_ACCESS_MSRS] = { { .index = MSR_STAR, .always = true }, { .index = MSR_IA32_SYSENTER_CS, .always = true }, + { .index = MSR_IA32_SYSENTER_EIP, .always = false }, + { .index = MSR_IA32_SYSENTER_ESP, .always = false }, #ifdef CONFIG_X86_64 { .index = MSR_GS_BASE, .always = true }, { .index = MSR_FS_BASE, .always = true }, @@ -112,6 +99,7 @@ static const struct svm_direct_access_msrs { { .index = MSR_EFER, .always = false }, { .index = MSR_IA32_CR_PAT, .always = false }, { .index = MSR_AMD64_SEV_ES_GHCB, .always = true }, + { .index = MSR_TSC_AUX, .always = false }, { .index = MSR_INVALID, .always = false }, }; @@ -183,20 +171,34 @@ static int vls = true; module_param(vls, int, 0444); /* enable/disable Virtual GIF */ -static int vgif = true; +int vgif = true; module_param(vgif, int, 0444); -/* enable/disable SEV support */ -int sev = IS_ENABLED(CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT); -module_param(sev, int, 0444); +/* enable/disable LBR virtualization */ +static int lbrv = true; +module_param(lbrv, int, 0444); + +static int tsc_scaling = true; +module_param(tsc_scaling, int, 0444); + +/* + * enable / disable AVIC. Because the defaults differ for APICv + * support between VMX and SVM we cannot use module_param_named. + */ +static bool avic; +module_param(avic, bool, 0444); -/* enable/disable SEV-ES support */ -int sev_es = IS_ENABLED(CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT); -module_param(sev_es, int, 0444); +static bool force_avic; +module_param_unsafe(force_avic, bool, 0444); bool __read_mostly dump_invalid_vmcb; module_param(dump_invalid_vmcb, bool, 0644); + +bool intercept_smi = true; +module_param(intercept_smi, bool, 0444); + + static bool svm_gp_erratum_intercept = true; static u8 rsm_ins_bytes[] = "\x0f\xaa"; @@ -214,6 +216,15 @@ struct kvm_ldttss_desc { DEFINE_PER_CPU(struct svm_cpu_data *, svm_data); +/* + * Only MSR_TSC_AUX is switched via the user return hook. EFER is switched via + * the VMCB, and the SYSCALL/SYSENTER MSRs are handled by VMLOAD/VMSAVE. + * + * RDTSCP and RDPID are not used in the kernel, specifically to allow KVM to + * defer the restoration of TSC_AUX until the CPU returns to userspace. + */ +static int tsc_aux_uret_slot __read_mostly = -1; + static const u32 msrpm_ranges[] = {0, 0xc0000000, 0xc0010000}; #define NUM_MSR_MAPS ARRAY_SIZE(msrpm_ranges) @@ -241,12 +252,12 @@ u32 svm_msrpm_offset(u32 msr) return MSR_INVALID; } -#define MAX_INST_SIZE 15 +static void svm_flush_tlb_current(struct kvm_vcpu *vcpu); -static int get_max_npt_level(void) +static int get_npt_level(void) { #ifdef CONFIG_X86_64 - return PT64_ROOT_4LEVEL; + return pgtable_l5_enabled() ? PT64_ROOT_5LEVEL : PT64_ROOT_4LEVEL; #else return PT32E_ROOT_LEVEL; #endif @@ -268,7 +279,7 @@ int svm_set_efer(struct kvm_vcpu *vcpu, u64 efer) if ((old_efer & EFER_SVME) != (efer & EFER_SVME)) { if (!(efer & EFER_SVME)) { - svm_leave_nested(svm); + svm_leave_nested(vcpu); svm_set_gif(svm, true); /* #GP intercept is still needed for vmware backdoor */ if (!enable_vmware_backdoor) @@ -279,7 +290,7 @@ int svm_set_efer(struct kvm_vcpu *vcpu, u64 efer) * In this case we will return to the nested guest * as soon as we leave SMM. */ - if (!is_smm(&svm->vcpu)) + if (!is_smm(vcpu)) svm_free_nested(svm); } else { @@ -290,7 +301,11 @@ int svm_set_efer(struct kvm_vcpu *vcpu, u64 efer) return ret; } - if (svm_gp_erratum_intercept) + /* + * Never intercept #GP for SEV guests, KVM can't + * decrypt guest memory to workaround the erratum. + */ + if (svm_gp_erratum_intercept && !sev_guest(vcpu->kvm)) set_exception_intercept(svm, GP_VECTOR); } } @@ -327,7 +342,7 @@ static void svm_set_interrupt_shadow(struct kvm_vcpu *vcpu, int mask) } -static int skip_emulated_instruction(struct kvm_vcpu *vcpu) +static int svm_skip_emulated_instruction(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -363,10 +378,10 @@ static void svm_queue_exception(struct kvm_vcpu *vcpu) bool has_error_code = vcpu->arch.exception.has_error_code; u32 error_code = vcpu->arch.exception.error_code; - kvm_deliver_exception_payload(&svm->vcpu); + kvm_deliver_exception_payload(vcpu); if (nr == BP_VECTOR && !nrips) { - unsigned long rip, old_rip = kvm_rip_read(&svm->vcpu); + unsigned long rip, old_rip = kvm_rip_read(vcpu); /* * For guest debugging where we have to reinject #BP if some @@ -375,8 +390,8 @@ static void svm_queue_exception(struct kvm_vcpu *vcpu) * raises a fault that is not intercepted. Still better than * failing in all cases. */ - (void)skip_emulated_instruction(&svm->vcpu); - rip = kvm_rip_read(&svm->vcpu); + (void)svm_skip_emulated_instruction(vcpu); + rip = kvm_rip_read(vcpu); svm->int3_rip = rip + svm->vmcb->save.cs.base; svm->int3_injected = rip - old_rip; } @@ -442,7 +457,7 @@ static int has_svm(void) return 0; } - if (sev_active()) { + if (cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT)) { pr_info("KVM is unsupported when running as an SEV guest\n"); return 0; } @@ -450,11 +465,24 @@ static int has_svm(void) return 1; } +void __svm_write_tsc_multiplier(u64 multiplier) +{ + preempt_disable(); + + if (multiplier == __this_cpu_read(current_tsc_ratio)) + goto out; + + wrmsrl(MSR_AMD64_TSC_RATIO, multiplier); + __this_cpu_write(current_tsc_ratio, multiplier); +out: + preempt_enable(); +} + static void svm_hardware_disable(void) { /* Make sure we clean up behind us */ - if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) - wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT); + if (tsc_scaling) + __svm_write_tsc_multiplier(SVM_TSC_RATIO_DEFAULT); cpu_svm_disable(); @@ -496,8 +524,11 @@ static int svm_hardware_enable(void) wrmsrl(MSR_VM_HSAVE_PA, __sme_page_pa(sd->save_area)); if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) { - wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT); - __this_cpu_write(current_tsc_ratio, TSC_RATIO_DEFAULT); + /* + * Set the default value, even if we don't use TSC scaling + * to avoid having stale value in the msr + */ + __svm_write_tsc_multiplier(SVM_TSC_RATIO_DEFAULT); } @@ -553,23 +584,19 @@ static void svm_cpu_uninit(int cpu) static int svm_cpu_init(int cpu) { struct svm_cpu_data *sd; + int ret = -ENOMEM; sd = kzalloc(sizeof(struct svm_cpu_data), GFP_KERNEL); if (!sd) - return -ENOMEM; + return ret; sd->cpu = cpu; - sd->save_area = alloc_page(GFP_KERNEL); + sd->save_area = alloc_page(GFP_KERNEL | __GFP_ZERO); if (!sd->save_area) goto free_cpu_data; - clear_page(page_address(sd->save_area)); - - if (svm_sev_enabled()) { - sd->sev_vmcbs = kmalloc_array(max_sev_asid + 1, - sizeof(void *), - GFP_KERNEL); - if (!sd->sev_vmcbs) - goto free_save_area; - } + + ret = sev_cpu_init(sd); + if (ret) + goto free_save_area; per_cpu(svm_data, cpu) = sd; @@ -579,7 +606,7 @@ free_save_area: __free_page(sd->save_area); free_cpu_data: kfree(sd); - return -ENOMEM; + return ret; } @@ -642,6 +669,7 @@ static bool msr_write_intercepted(struct kvm_vcpu *vcpu, u32 msr) static void set_msr_interception_bitmap(struct kvm_vcpu *vcpu, u32 *msrpm, u32 msr, int read, int write) { + struct vcpu_svm *svm = to_svm(vcpu); u8 bit_read, bit_write; unsigned long tmp; u32 offset; @@ -670,6 +698,9 @@ static void set_msr_interception_bitmap(struct kvm_vcpu *vcpu, u32 *msrpm, write ? clear_bit(bit_write, &tmp) : set_bit(bit_write, &tmp); msrpm[offset] = tmp; + + svm_hv_vmcb_dirty_nested_enlightenments(vcpu); + svm->nested.force_msr_bitmap_recalc = true; } void set_msr_interception(struct kvm_vcpu *vcpu, u32 *msrpm, u32 msr, @@ -681,14 +712,15 @@ void set_msr_interception(struct kvm_vcpu *vcpu, u32 *msrpm, u32 msr, u32 *svm_vcpu_alloc_msrpm(void) { - struct page *pages = alloc_pages(GFP_KERNEL_ACCOUNT, MSRPM_ALLOC_ORDER); + unsigned int order = get_order(MSRPM_SIZE); + struct page *pages = alloc_pages(GFP_KERNEL_ACCOUNT, order); u32 *msrpm; if (!pages) return NULL; msrpm = page_address(pages); - memset(msrpm, 0xff, PAGE_SIZE * (1 << MSRPM_ALLOC_ORDER)); + memset(msrpm, 0xff, PAGE_SIZE * (1 << order)); return msrpm; } @@ -707,7 +739,7 @@ void svm_vcpu_init_msrpm(struct kvm_vcpu *vcpu, u32 *msrpm) void svm_vcpu_free_msrpm(u32 *msrpm) { - __free_pages(virt_to_page(msrpm), MSRPM_ALLOC_ORDER); + __free_pages(virt_to_page(msrpm), get_order(MSRPM_SIZE)); } static void svm_msr_filter_changed(struct kvm_vcpu *vcpu) @@ -772,6 +804,17 @@ static void init_msrpm_offsets(void) } } +void svm_copy_lbrs(struct vmcb *to_vmcb, struct vmcb *from_vmcb) +{ + to_vmcb->save.dbgctl = from_vmcb->save.dbgctl; + to_vmcb->save.br_from = from_vmcb->save.br_from; + to_vmcb->save.br_to = from_vmcb->save.br_to; + to_vmcb->save.last_excp_from = from_vmcb->save.last_excp_from; + to_vmcb->save.last_excp_to = from_vmcb->save.last_excp_to; + + vmcb_mark_dirty(to_vmcb, VMCB_LBR); +} + static void svm_enable_lbrv(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -781,6 +824,10 @@ static void svm_enable_lbrv(struct kvm_vcpu *vcpu) set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHTOIP, 1, 1); set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTFROMIP, 1, 1); set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTTOIP, 1, 1); + + /* Move the LBR msrs to the vmcb02 so that the guest can see them. */ + if (is_guest_mode(vcpu)) + svm_copy_lbrs(svm->vmcb, svm->vmcb01.ptr); } static void svm_disable_lbrv(struct kvm_vcpu *vcpu) @@ -792,6 +839,67 @@ static void svm_disable_lbrv(struct kvm_vcpu *vcpu) set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTBRANCHTOIP, 0, 0); set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTFROMIP, 0, 0); set_msr_interception(vcpu, svm->msrpm, MSR_IA32_LASTINTTOIP, 0, 0); + + /* + * Move the LBR msrs back to the vmcb01 to avoid copying them + * on nested guest entries. + */ + if (is_guest_mode(vcpu)) + svm_copy_lbrs(svm->vmcb01.ptr, svm->vmcb); +} + +static int svm_get_lbr_msr(struct vcpu_svm *svm, u32 index) +{ + /* + * If the LBR virtualization is disabled, the LBR msrs are always + * kept in the vmcb01 to avoid copying them on nested guest entries. + * + * If nested, and the LBR virtualization is enabled/disabled, the msrs + * are moved between the vmcb01 and vmcb02 as needed. + */ + struct vmcb *vmcb = + (svm->vmcb->control.virt_ext & LBR_CTL_ENABLE_MASK) ? + svm->vmcb : svm->vmcb01.ptr; + + switch (index) { + case MSR_IA32_DEBUGCTLMSR: + return vmcb->save.dbgctl; + case MSR_IA32_LASTBRANCHFROMIP: + return vmcb->save.br_from; + case MSR_IA32_LASTBRANCHTOIP: + return vmcb->save.br_to; + case MSR_IA32_LASTINTFROMIP: + return vmcb->save.last_excp_from; + case MSR_IA32_LASTINTTOIP: + return vmcb->save.last_excp_to; + default: + KVM_BUG(false, svm->vcpu.kvm, + "%s: Unknown MSR 0x%x", __func__, index); + return 0; + } +} + +void svm_update_lbrv(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + bool enable_lbrv = svm_get_lbr_msr(svm, MSR_IA32_DEBUGCTLMSR) & + DEBUGCTLMSR_LBR; + + bool current_enable_lbrv = !!(svm->vmcb->control.virt_ext & + LBR_CTL_ENABLE_MASK); + + if (unlikely(is_guest_mode(vcpu) && svm->lbrv_enabled)) + if (unlikely(svm->nested.ctl.virt_ext & LBR_CTL_ENABLE_MASK)) + enable_lbrv = true; + + if (enable_lbrv == current_enable_lbrv) + return; + + if (enable_lbrv) + svm_enable_lbrv(vcpu); + else + svm_disable_lbrv(vcpu); } void disable_nmi_singlestep(struct vcpu_svm *svm) @@ -813,6 +921,9 @@ static void grow_ple_window(struct kvm_vcpu *vcpu) struct vmcb_control_area *control = &svm->vmcb->control; int old = control->pause_filter_count; + if (kvm_pause_in_guest(vcpu->kvm)) + return; + control->pause_filter_count = __grow_ple_window(old, pause_filter_count, pause_filter_count_grow, @@ -831,6 +942,9 @@ static void shrink_ple_window(struct kvm_vcpu *vcpu) struct vmcb_control_area *control = &svm->vmcb->control; int old = control->pause_filter_count; + if (kvm_pause_in_guest(vcpu->kvm)) + return; + control->pause_filter_count = __shrink_ple_window(old, pause_filter_count, @@ -843,222 +957,20 @@ static void shrink_ple_window(struct kvm_vcpu *vcpu) } } -/* - * The default MMIO mask is a single bit (excluding the present bit), - * which could conflict with the memory encryption bit. Check for - * memory encryption support and override the default MMIO mask if - * memory encryption is enabled. - */ -static __init void svm_adjust_mmio_mask(void) -{ - unsigned int enc_bit, mask_bit; - u64 msr, mask; - - /* If there is no memory encryption support, use existing mask */ - if (cpuid_eax(0x80000000) < 0x8000001f) - return; - - /* If memory encryption is not enabled, use existing mask */ - rdmsrl(MSR_K8_SYSCFG, msr); - if (!(msr & MSR_K8_SYSCFG_MEM_ENCRYPT)) - return; - - enc_bit = cpuid_ebx(0x8000001f) & 0x3f; - mask_bit = boot_cpu_data.x86_phys_bits; - - /* Increment the mask bit if it is the same as the encryption bit */ - if (enc_bit == mask_bit) - mask_bit++; - - /* - * If the mask bit location is below 52, then some bits above the - * physical addressing limit will always be reserved, so use the - * rsvd_bits() function to generate the mask. This mask, along with - * the present bit, will be used to generate a page fault with - * PFER.RSV = 1. - * - * If the mask bit location is 52 (or above), then clear the mask. - */ - mask = (mask_bit < 52) ? rsvd_bits(mask_bit, 51) | PT_PRESENT_MASK : 0; - - kvm_mmu_set_mmio_spte_mask(mask, PT_WRITABLE_MASK | PT_USER_MASK); -} - -static void svm_hardware_teardown(void) +static void svm_hardware_unsetup(void) { int cpu; - if (svm_sev_enabled()) - sev_hardware_teardown(); + sev_hardware_unsetup(); for_each_possible_cpu(cpu) svm_cpu_uninit(cpu); - __free_pages(pfn_to_page(iopm_base >> PAGE_SHIFT), IOPM_ALLOC_ORDER); + __free_pages(pfn_to_page(iopm_base >> PAGE_SHIFT), + get_order(IOPM_SIZE)); iopm_base = 0; } -static __init void svm_set_cpu_caps(void) -{ - kvm_set_cpu_caps(); - - supported_xss = 0; - - /* CPUID 0x80000001 and 0x8000000A (SVM features) */ - if (nested) { - kvm_cpu_cap_set(X86_FEATURE_SVM); - - if (nrips) - kvm_cpu_cap_set(X86_FEATURE_NRIPS); - - if (npt_enabled) - kvm_cpu_cap_set(X86_FEATURE_NPT); - - /* Nested VM can receive #VMEXIT instead of triggering #GP */ - kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK); - } - - /* CPUID 0x80000008 */ - if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD) || - boot_cpu_has(X86_FEATURE_AMD_SSBD)) - kvm_cpu_cap_set(X86_FEATURE_VIRT_SSBD); -} - -static __init int svm_hardware_setup(void) -{ - int cpu; - struct page *iopm_pages; - void *iopm_va; - int r; - - iopm_pages = alloc_pages(GFP_KERNEL, IOPM_ALLOC_ORDER); - - if (!iopm_pages) - return -ENOMEM; - - iopm_va = page_address(iopm_pages); - memset(iopm_va, 0xff, PAGE_SIZE * (1 << IOPM_ALLOC_ORDER)); - iopm_base = page_to_pfn(iopm_pages) << PAGE_SHIFT; - - init_msrpm_offsets(); - - supported_xcr0 &= ~(XFEATURE_MASK_BNDREGS | XFEATURE_MASK_BNDCSR); - - if (boot_cpu_has(X86_FEATURE_NX)) - kvm_enable_efer_bits(EFER_NX); - - if (boot_cpu_has(X86_FEATURE_FXSR_OPT)) - kvm_enable_efer_bits(EFER_FFXSR); - - if (boot_cpu_has(X86_FEATURE_TSCRATEMSR)) { - kvm_has_tsc_control = true; - kvm_max_tsc_scaling_ratio = TSC_RATIO_MAX; - kvm_tsc_scaling_ratio_frac_bits = 32; - } - - /* Check for pause filtering support */ - if (!boot_cpu_has(X86_FEATURE_PAUSEFILTER)) { - pause_filter_count = 0; - pause_filter_thresh = 0; - } else if (!boot_cpu_has(X86_FEATURE_PFTHRESHOLD)) { - pause_filter_thresh = 0; - } - - if (nested) { - printk(KERN_INFO "kvm: Nested Virtualization enabled\n"); - kvm_enable_efer_bits(EFER_SVME | EFER_LMSLE); - } - - if (IS_ENABLED(CONFIG_KVM_AMD_SEV) && sev) { - sev_hardware_setup(); - } else { - sev = false; - sev_es = false; - } - - svm_adjust_mmio_mask(); - - for_each_possible_cpu(cpu) { - r = svm_cpu_init(cpu); - if (r) - goto err; - } - - /* - * KVM's MMU doesn't support using 2-level paging for itself, and thus - * NPT isn't supported if the host is using 2-level paging since host - * CR4 is unchanged on VMRUN. - */ - if (!IS_ENABLED(CONFIG_X86_64) && !IS_ENABLED(CONFIG_X86_PAE)) - npt_enabled = false; - - if (!boot_cpu_has(X86_FEATURE_NPT)) - npt_enabled = false; - - kvm_configure_mmu(npt_enabled, get_max_npt_level(), PG_LEVEL_1G); - pr_info("kvm: Nested Paging %sabled\n", npt_enabled ? "en" : "dis"); - - if (nrips) { - if (!boot_cpu_has(X86_FEATURE_NRIPS)) - nrips = false; - } - - if (avic) { - if (!npt_enabled || - !boot_cpu_has(X86_FEATURE_AVIC) || - !IS_ENABLED(CONFIG_X86_LOCAL_APIC)) { - avic = false; - } else { - pr_info("AVIC enabled\n"); - - amd_iommu_register_ga_log_notifier(&avic_ga_log_notifier); - } - } - - if (vls) { - if (!npt_enabled || - !boot_cpu_has(X86_FEATURE_V_VMSAVE_VMLOAD) || - !IS_ENABLED(CONFIG_X86_64)) { - vls = false; - } else { - pr_info("Virtual VMLOAD VMSAVE supported\n"); - } - } - - if (boot_cpu_has(X86_FEATURE_SVME_ADDR_CHK)) - svm_gp_erratum_intercept = false; - - if (vgif) { - if (!boot_cpu_has(X86_FEATURE_VGIF)) - vgif = false; - else - pr_info("Virtual GIF supported\n"); - } - - svm_set_cpu_caps(); - - /* - * It seems that on AMD processors PTE's accessed bit is - * being set by the CPU hardware before the NPF vmexit. - * This is not expected behaviour and our tests fail because - * of it. - * A workaround here is to disable support for - * GUEST_MAXPHYADDR < HOST_MAXPHYADDR if NPT is enabled. - * In this case userspace can know if there is support using - * KVM_CAP_SMALLER_MAXPHYADDR extension and decide how to handle - * it - * If future AMD CPU models change the behaviour described above, - * this variable can be changed accordingly - */ - allow_smaller_maxphyaddr = !npt_enabled; - - return 0; - -err: - svm_hardware_teardown(); - return r; -} - static void init_seg(struct vmcb_seg *seg) { seg->selector = 0; @@ -1076,29 +988,38 @@ static void init_sys_seg(struct vmcb_seg *seg, uint32_t type) seg->base = 0; } -static u64 svm_write_l1_tsc_offset(struct kvm_vcpu *vcpu, u64 offset) +static u64 svm_get_l2_tsc_offset(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - u64 g_tsc_offset = 0; - if (is_guest_mode(vcpu)) { - /* Write L1's TSC offset. */ - g_tsc_offset = svm->vmcb->control.tsc_offset - - svm->nested.hsave->control.tsc_offset; - svm->nested.hsave->control.tsc_offset = offset; - } + return svm->nested.ctl.tsc_offset; +} - trace_kvm_write_tsc_offset(vcpu->vcpu_id, - svm->vmcb->control.tsc_offset - g_tsc_offset, - offset); +static u64 svm_get_l2_tsc_multiplier(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); - svm->vmcb->control.tsc_offset = offset + g_tsc_offset; + return svm->tsc_ratio_msr; +} + +static void svm_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset) +{ + struct vcpu_svm *svm = to_svm(vcpu); + svm->vmcb01.ptr->control.tsc_offset = vcpu->arch.l1_tsc_offset; + svm->vmcb->control.tsc_offset = offset; vmcb_mark_dirty(svm->vmcb, VMCB_INTERCEPTS); - return svm->vmcb->control.tsc_offset; } -static void svm_check_invpcid(struct vcpu_svm *svm) +static void svm_write_tsc_multiplier(struct kvm_vcpu *vcpu, u64 multiplier) +{ + __svm_write_tsc_multiplier(multiplier); +} + + +/* Evaluate instruction intercepts that depend on guest CPUID features. */ +static void svm_recalc_instruction_intercepts(struct kvm_vcpu *vcpu, + struct vcpu_svm *svm) { /* * Intercept INVPCID if shadow paging is enabled to sync/free shadow @@ -1111,14 +1032,55 @@ static void svm_check_invpcid(struct vcpu_svm *svm) else svm_clr_intercept(svm, INTERCEPT_INVPCID); } + + if (kvm_cpu_cap_has(X86_FEATURE_RDTSCP)) { + if (guest_cpuid_has(vcpu, X86_FEATURE_RDTSCP)) + svm_clr_intercept(svm, INTERCEPT_RDTSCP); + else + svm_set_intercept(svm, INTERCEPT_RDTSCP); + } } -static void init_vmcb(struct vcpu_svm *svm) +static inline void init_vmcb_after_set_cpuid(struct kvm_vcpu *vcpu) { - struct vmcb_control_area *control = &svm->vmcb->control; - struct vmcb_save_area *save = &svm->vmcb->save; + struct vcpu_svm *svm = to_svm(vcpu); - svm->vcpu.arch.hflags = 0; + if (guest_cpuid_is_intel(vcpu)) { + /* + * We must intercept SYSENTER_EIP and SYSENTER_ESP + * accesses because the processor only stores 32 bits. + * For the same reason we cannot use virtual VMLOAD/VMSAVE. + */ + svm_set_intercept(svm, INTERCEPT_VMLOAD); + svm_set_intercept(svm, INTERCEPT_VMSAVE); + svm->vmcb->control.virt_ext &= ~VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK; + + set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_EIP, 0, 0); + set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_ESP, 0, 0); + + svm->v_vmload_vmsave_enabled = false; + } else { + /* + * If hardware supports Virtual VMLOAD VMSAVE then enable it + * in VMCB and clear intercepts to avoid #VMEXIT. + */ + if (vls) { + svm_clr_intercept(svm, INTERCEPT_VMLOAD); + svm_clr_intercept(svm, INTERCEPT_VMSAVE); + svm->vmcb->control.virt_ext |= VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK; + } + /* No need to intercept these MSRs */ + set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_EIP, 1, 1); + set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SYSENTER_ESP, 1, 1); + } +} + +static void init_vmcb(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb *vmcb = svm->vmcb01.ptr; + struct vmcb_control_area *control = &vmcb->control; + struct vmcb_save_area *save = &vmcb->save; svm_set_intercept(svm, INTERCEPT_CR0_READ); svm_set_intercept(svm, INTERCEPT_CR3_READ); @@ -1126,7 +1088,7 @@ static void init_vmcb(struct vcpu_svm *svm) svm_set_intercept(svm, INTERCEPT_CR0_WRITE); svm_set_intercept(svm, INTERCEPT_CR3_WRITE); svm_set_intercept(svm, INTERCEPT_CR4_WRITE); - if (!kvm_vcpu_apicv_active(&svm->vcpu)) + if (!kvm_vcpu_apicv_active(vcpu)) svm_set_intercept(svm, INTERCEPT_CR8_WRITE); set_dr_intercepts(svm); @@ -1140,14 +1102,18 @@ static void init_vmcb(struct vcpu_svm *svm) * Guest access to VMware backdoor ports could legitimately * trigger #GP because of TSS I/O permission bitmap. * We intercept those #GP and allow access to them anyway - * as VMware does. + * as VMware does. Don't intercept #GP for SEV guests as KVM can't + * decrypt guest memory to decode the faulting instruction. */ - if (enable_vmware_backdoor) + if (enable_vmware_backdoor && !sev_guest(vcpu->kvm)) set_exception_intercept(svm, GP_VECTOR); svm_set_intercept(svm, INTERCEPT_INTR); svm_set_intercept(svm, INTERCEPT_NMI); - svm_set_intercept(svm, INTERCEPT_SMI); + + if (intercept_smi) + svm_set_intercept(svm, INTERCEPT_SMI); + svm_set_intercept(svm, INTERCEPT_SELECTIVE_CR0); svm_set_intercept(svm, INTERCEPT_RDPMC); svm_set_intercept(svm, INTERCEPT_CPUID); @@ -1170,12 +1136,12 @@ static void init_vmcb(struct vcpu_svm *svm) svm_set_intercept(svm, INTERCEPT_RDPRU); svm_set_intercept(svm, INTERCEPT_RSM); - if (!kvm_mwait_in_guest(svm->vcpu.kvm)) { + if (!kvm_mwait_in_guest(vcpu->kvm)) { svm_set_intercept(svm, INTERCEPT_MONITOR); svm_set_intercept(svm, INTERCEPT_MWAIT); } - if (!kvm_hlt_in_guest(svm->vcpu.kvm)) + if (!kvm_hlt_in_guest(vcpu->kvm)) svm_set_intercept(svm, INTERCEPT_HLT); control->iopm_base_pa = __sme_set(iopm_base); @@ -1195,29 +1161,14 @@ static void init_vmcb(struct vcpu_svm *svm) SVM_SELECTOR_S_MASK | SVM_SELECTOR_CODE_MASK; save->cs.limit = 0xffff; + save->gdtr.base = 0; save->gdtr.limit = 0xffff; + save->idtr.base = 0; save->idtr.limit = 0xffff; init_sys_seg(&save->ldtr, SEG_TYPE_LDT); init_sys_seg(&save->tr, SEG_TYPE_BUSY_TSS16); - svm_set_cr4(&svm->vcpu, 0); - svm_set_efer(&svm->vcpu, 0); - save->dr6 = 0xffff0ff0; - kvm_set_rflags(&svm->vcpu, X86_EFLAGS_FIXED); - save->rip = 0x0000fff0; - svm->vcpu.arch.regs[VCPU_REGS_RIP] = save->rip; - - /* - * svm_set_cr0() sets PG and WP and clears NW and CD on save->cr0. - * It also updates the guest-visible cr0 value. - */ - svm_set_cr0(&svm->vcpu, X86_CR0_NW | X86_CR0_CD | X86_CR0_ET); - kvm_mmu_reset_context(&svm->vcpu); - - save->cr4 = X86_CR4_PAE; - /* rdx = ?? */ - if (npt_enabled) { /* Setup VMCB for Nested Paging */ control->nested_ctl |= SVM_NESTED_CTL_NP_ENABLE; @@ -1225,17 +1176,16 @@ static void init_vmcb(struct vcpu_svm *svm) clr_exception_intercept(svm, PF_VECTOR); svm_clr_intercept(svm, INTERCEPT_CR3_READ); svm_clr_intercept(svm, INTERCEPT_CR3_WRITE); - save->g_pat = svm->vcpu.arch.pat; + save->g_pat = vcpu->arch.pat; save->cr3 = 0; - save->cr4 = 0; } - svm->asid_generation = 0; + svm->current_vmcb->asid_generation = 0; svm->asid = 0; - svm->nested.vmcb12_gpa = 0; - svm->vcpu.arch.hflags = 0; + svm->nested.vmcb12_gpa = INVALID_GPA; + svm->nested.last_vmcb12_gpa = INVALID_GPA; - if (!kvm_pause_in_guest(svm->vcpu.kvm)) { + if (!kvm_pause_in_guest(vcpu->kvm)) { control->pause_filter_count = pause_filter_count; if (pause_filter_thresh) control->pause_filter_thresh = pause_filter_thresh; @@ -1244,20 +1194,17 @@ static void init_vmcb(struct vcpu_svm *svm) svm_clr_intercept(svm, INTERCEPT_PAUSE); } - svm_check_invpcid(svm); - - if (kvm_vcpu_apicv_active(&svm->vcpu)) - avic_init_vmcb(svm); + svm_recalc_instruction_intercepts(vcpu, svm); /* - * If hardware supports Virtual VMLOAD VMSAVE then enable it - * in VMCB and clear intercepts to avoid #VMEXIT. + * If the host supports V_SPEC_CTRL then disable the interception + * of MSR_IA32_SPEC_CTRL. */ - if (vls) { - svm_clr_intercept(svm, INTERCEPT_VMLOAD); - svm_clr_intercept(svm, INTERCEPT_VMSAVE); - svm->vmcb->control.virt_ext |= VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK; - } + if (boot_cpu_has(X86_FEATURE_V_SPEC_CTRL)) + set_msr_interception(vcpu, svm->msrpm, MSR_IA32_SPEC_CTRL, 1, 1); + + if (kvm_vcpu_apicv_active(vcpu)) + avic_init_vmcb(svm, vmcb); if (vgif) { svm_clr_intercept(svm, INTERCEPT_STGI); @@ -1265,50 +1212,61 @@ static void init_vmcb(struct vcpu_svm *svm) svm->vmcb->control.int_ctl |= V_GIF_ENABLE_MASK; } - if (sev_guest(svm->vcpu.kvm)) { + if (sev_guest(vcpu->kvm)) { svm->vmcb->control.nested_ctl |= SVM_NESTED_CTL_SEV_ENABLE; clr_exception_intercept(svm, UD_VECTOR); - if (sev_es_guest(svm->vcpu.kvm)) { + if (sev_es_guest(vcpu->kvm)) { /* Perform SEV-ES specific VMCB updates */ sev_es_init_vmcb(svm); } } - vmcb_mark_all_dirty(svm->vmcb); + svm_hv_init_vmcb(vmcb); + init_vmcb_after_set_cpuid(vcpu); + + vmcb_mark_all_dirty(vmcb); enable_gif(svm); +} +static void __svm_vcpu_reset(struct kvm_vcpu *vcpu) +{ + struct vcpu_svm *svm = to_svm(vcpu); + + svm_vcpu_init_msrpm(vcpu, svm->msrpm); + + svm_init_osvw(vcpu); + vcpu->arch.microcode_version = 0x01000065; + svm->tsc_ratio_msr = kvm_default_tsc_scaling_ratio; + + if (sev_es_guest(vcpu->kvm)) + sev_es_vcpu_reset(svm); } static void svm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event) { struct vcpu_svm *svm = to_svm(vcpu); - u32 dummy; - u32 eax = 1; svm->spec_ctrl = 0; svm->virt_spec_ctrl = 0; - if (!init_event) { - svm->vcpu.arch.apic_base = APIC_DEFAULT_PHYS_BASE | - MSR_IA32_APICBASE_ENABLE; - if (kvm_vcpu_is_reset_bsp(&svm->vcpu)) - svm->vcpu.arch.apic_base |= MSR_IA32_APICBASE_BSP; - } - init_vmcb(svm); + init_vmcb(vcpu); - kvm_cpuid(vcpu, &eax, &dummy, &dummy, &dummy, false); - kvm_rdx_write(vcpu, eax); + if (!init_event) + __svm_vcpu_reset(vcpu); +} - if (kvm_vcpu_apicv_active(vcpu) && !init_event) - avic_update_vapic_bar(svm, APIC_DEFAULT_PHYS_BASE); +void svm_switch_vmcb(struct vcpu_svm *svm, struct kvm_vmcb_info *target_vmcb) +{ + svm->current_vmcb = target_vmcb; + svm->vmcb = target_vmcb->ptr; } -static int svm_create_vcpu(struct kvm_vcpu *vcpu) +static int svm_vcpu_create(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm; - struct page *vmcb_page; + struct page *vmcb01_page; struct page *vmsa_page = NULL; int err; @@ -1316,11 +1274,11 @@ static int svm_create_vcpu(struct kvm_vcpu *vcpu) svm = to_svm(vcpu); err = -ENOMEM; - vmcb_page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); - if (!vmcb_page) + vmcb01_page = alloc_page(GFP_KERNEL_ACCOUNT | __GFP_ZERO); + if (!vmcb01_page) goto out; - if (sev_es_guest(svm->vcpu.kvm)) { + if (sev_es_guest(vcpu->kvm)) { /* * SEV-ES guests require a separate VMSA page used to contain * the encrypted register state of the guest. @@ -1332,46 +1290,30 @@ static int svm_create_vcpu(struct kvm_vcpu *vcpu) /* * SEV-ES guests maintain an encrypted version of their FPU * state which is restored and saved on VMRUN and VMEXIT. - * Free the fpu structure to prevent KVM from attempting to - * access the FPU state. + * Mark vcpu->arch.guest_fpu->fpstate as scratch so it won't + * do xsave/xrstor on it. */ - kvm_free_guest_fpu(vcpu); + fpstate_set_confidential(&vcpu->arch.guest_fpu); } err = avic_init_vcpu(svm); if (err) goto error_free_vmsa_page; - /* We initialize this flag to true to make sure that the is_running - * bit would be set the first time the vcpu is loaded. - */ - if (irqchip_in_kernel(vcpu->kvm) && kvm_apicv_activated(vcpu->kvm)) - svm->avic_is_running = true; - svm->msrpm = svm_vcpu_alloc_msrpm(); if (!svm->msrpm) { err = -ENOMEM; goto error_free_vmsa_page; } - svm_vcpu_init_msrpm(vcpu, svm->msrpm); - - svm->vmcb = page_address(vmcb_page); - svm->vmcb_pa = __sme_set(page_to_pfn(vmcb_page) << PAGE_SHIFT); + svm->vmcb01.ptr = page_address(vmcb01_page); + svm->vmcb01.pa = __sme_set(page_to_pfn(vmcb01_page) << PAGE_SHIFT); + svm_switch_vmcb(svm, &svm->vmcb01); if (vmsa_page) - svm->vmsa = page_address(vmsa_page); + svm->sev_es.vmsa = page_address(vmsa_page); - svm->asid_generation = 0; svm->guest_state_loaded = false; - init_vmcb(svm); - - svm_init_osvw(vcpu); - vcpu->arch.microcode_version = 0x01000065; - - if (sev_es_guest(svm->vcpu.kvm)) - /* Perform SEV-ES specific VMCB creation updates */ - sev_es_create_vcpu(svm); return 0; @@ -1379,7 +1321,7 @@ error_free_vmsa_page: if (vmsa_page) __free_page(vmsa_page); error_free_vmcb_page: - __free_page(vmcb_page); + __free_page(vmcb01_page); out: return err; } @@ -1392,7 +1334,7 @@ static void svm_clear_current_vmcb(struct vmcb *vmcb) cmpxchg(&per_cpu(svm_data, i)->current_vmcb, vmcb, NULL); } -static void svm_free_vcpu(struct kvm_vcpu *vcpu) +static void svm_vcpu_free(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -1407,68 +1349,45 @@ static void svm_free_vcpu(struct kvm_vcpu *vcpu) sev_free_vcpu(vcpu); - __free_page(pfn_to_page(__sme_clr(svm->vmcb_pa) >> PAGE_SHIFT)); - __free_pages(virt_to_page(svm->msrpm), MSRPM_ALLOC_ORDER); + __free_page(pfn_to_page(__sme_clr(svm->vmcb01.pa) >> PAGE_SHIFT)); + __free_pages(virt_to_page(svm->msrpm), get_order(MSRPM_SIZE)); } -static void svm_prepare_guest_switch(struct kvm_vcpu *vcpu) +static void svm_prepare_switch_to_guest(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu); - unsigned int i; + + if (sev_es_guest(vcpu->kvm)) + sev_es_unmap_ghcb(svm); if (svm->guest_state_loaded) return; /* - * Certain MSRs are restored on VMEXIT (sev-es), or vmload of host save - * area (non-sev-es). Save ones that aren't so we can restore them - * individually later. - */ - for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++) - rdmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]); - - /* * Save additional host state that will be restored on VMEXIT (sev-es) * or subsequent vmload of host save area. */ - if (sev_es_guest(svm->vcpu.kvm)) { - sev_es_prepare_guest_switch(svm, vcpu->cpu); - } else { - vmsave(__sme_page_pa(sd->save_area)); - } + vmsave(__sme_page_pa(sd->save_area)); + if (sev_es_guest(vcpu->kvm)) { + struct sev_es_save_area *hostsa; + hostsa = (struct sev_es_save_area *)(page_address(sd->save_area) + 0x400); - if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) { - u64 tsc_ratio = vcpu->arch.tsc_scaling_ratio; - if (tsc_ratio != __this_cpu_read(current_tsc_ratio)) { - __this_cpu_write(current_tsc_ratio, tsc_ratio); - wrmsrl(MSR_AMD64_TSC_RATIO, tsc_ratio); - } + sev_es_prepare_switch_to_guest(hostsa); } - /* This assumes that the kernel never uses MSR_TSC_AUX */ - if (static_cpu_has(X86_FEATURE_RDTSCP)) - wrmsrl(MSR_TSC_AUX, svm->tsc_aux); + if (tsc_scaling) + __svm_write_tsc_multiplier(vcpu->arch.tsc_scaling_ratio); + + if (likely(tsc_aux_uret_slot >= 0)) + kvm_set_user_return_msr(tsc_aux_uret_slot, svm->tsc_aux, -1ull); svm->guest_state_loaded = true; } static void svm_prepare_host_switch(struct kvm_vcpu *vcpu) { - struct vcpu_svm *svm = to_svm(vcpu); - unsigned int i; - - if (!svm->guest_state_loaded) - return; - - /* - * Certain MSRs are restored on VMEXIT (sev-es), or vmload of host save - * area (non-sev-es). Restore the ones that weren't. - */ - for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++) - wrmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]); - - svm->guest_state_loaded = false; + to_svm(vcpu)->guest_state_loaded = false; } static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu) @@ -1476,21 +1395,19 @@ static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu) struct vcpu_svm *svm = to_svm(vcpu); struct svm_cpu_data *sd = per_cpu(svm_data, cpu); - if (unlikely(cpu != vcpu->cpu)) { - svm->asid_generation = 0; - vmcb_mark_all_dirty(svm->vmcb); - } - if (sd->current_vmcb != svm->vmcb) { sd->current_vmcb = svm->vmcb; indirect_branch_prediction_barrier(); } - avic_vcpu_load(vcpu, cpu); + if (kvm_vcpu_apicv_active(vcpu)) + avic_vcpu_load(vcpu, cpu); } static void svm_vcpu_put(struct kvm_vcpu *vcpu) { - avic_vcpu_put(vcpu); + if (kvm_vcpu_apicv_active(vcpu)) + avic_vcpu_put(vcpu); + svm_prepare_host_switch(vcpu); ++vcpu->stat.host_state_reload; @@ -1524,15 +1441,30 @@ static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) to_svm(vcpu)->vmcb->save.rflags = rflags; } +static bool svm_get_if_flag(struct kvm_vcpu *vcpu) +{ + struct vmcb *vmcb = to_svm(vcpu)->vmcb; + + return sev_es_guest(vcpu->kvm) + ? vmcb->control.int_state & SVM_GUEST_INTERRUPT_MASK + : kvm_get_rflags(vcpu) & X86_EFLAGS_IF; +} + static void svm_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg) { + kvm_register_mark_available(vcpu, reg); + switch (reg) { case VCPU_EXREG_PDPTR: - BUG_ON(!npt_enabled); - load_pdptrs(vcpu, vcpu->arch.walk_mmu, kvm_read_cr3(vcpu)); + /* + * When !npt_enabled, mmu->pdptrs[] is already available since + * it is always updated per SDM when moving to CRs. + */ + if (npt_enabled) + load_pdptrs(vcpu, kvm_read_cr3(vcpu)); break; default: - WARN_ON_ONCE(1); + KVM_BUG_ON(1, vcpu->kvm); } } @@ -1540,8 +1472,11 @@ static void svm_set_vintr(struct vcpu_svm *svm) { struct vmcb_control_area *control; - /* The following fields are ignored when AVIC is enabled */ - WARN_ON(kvm_vcpu_apicv_active(&svm->vcpu)); + /* + * The following fields are ignored when AVIC is enabled + */ + WARN_ON(kvm_vcpu_apicv_activated(&svm->vcpu)); + svm_set_intercept(svm, INTERCEPT_VINTR); /* @@ -1558,17 +1493,20 @@ static void svm_set_vintr(struct vcpu_svm *svm) static void svm_clear_vintr(struct vcpu_svm *svm) { - const u32 mask = V_TPR_MASK | V_GIF_ENABLE_MASK | V_GIF_MASK | V_INTR_MASKING_MASK; svm_clr_intercept(svm, INTERCEPT_VINTR); /* Drop int_ctl fields related to VINTR injection. */ - svm->vmcb->control.int_ctl &= mask; + svm->vmcb->control.int_ctl &= ~V_IRQ_INJECTION_BITS_MASK; if (is_guest_mode(&svm->vcpu)) { - svm->nested.hsave->control.int_ctl &= mask; + svm->vmcb01.ptr->control.int_ctl &= ~V_IRQ_INJECTION_BITS_MASK; WARN_ON((svm->vmcb->control.int_ctl & V_TPR_MASK) != (svm->nested.ctl.int_ctl & V_TPR_MASK)); - svm->vmcb->control.int_ctl |= svm->nested.ctl.int_ctl & ~mask; + + svm->vmcb->control.int_ctl |= svm->nested.ctl.int_ctl & + V_IRQ_INJECTION_BITS_MASK; + + svm->vmcb->control.int_vector = svm->nested.ctl.int_vector; } vmcb_mark_dirty(svm->vmcb, VMCB_INTR); @@ -1577,16 +1515,17 @@ static void svm_clear_vintr(struct vcpu_svm *svm) static struct vmcb_seg *svm_seg(struct kvm_vcpu *vcpu, int seg) { struct vmcb_save_area *save = &to_svm(vcpu)->vmcb->save; + struct vmcb_save_area *save01 = &to_svm(vcpu)->vmcb01.ptr->save; switch (seg) { case VCPU_SREG_CS: return &save->cs; case VCPU_SREG_DS: return &save->ds; case VCPU_SREG_ES: return &save->es; - case VCPU_SREG_FS: return &save->fs; - case VCPU_SREG_GS: return &save->gs; + case VCPU_SREG_FS: return &save01->fs; + case VCPU_SREG_GS: return &save01->gs; case VCPU_SREG_SS: return &save->ss; - case VCPU_SREG_TR: return &save->tr; - case VCPU_SREG_LDTR: return &save->ldtr; + case VCPU_SREG_TR: return &save01->tr; + case VCPU_SREG_LDTR: return &save01->ldtr; } BUG(); return NULL; @@ -1675,6 +1614,15 @@ static int svm_get_cpl(struct kvm_vcpu *vcpu) return save->cpl; } +static void svm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l) +{ + struct kvm_segment cs; + + svm_get_segment(vcpu, &cs, VCPU_SREG_CS); + *db = cs.db; + *l = cs.l; +} + static void svm_get_idt(struct kvm_vcpu *vcpu, struct desc_ptr *dt) { struct vcpu_svm *svm = to_svm(vcpu); @@ -1709,37 +1657,29 @@ static void svm_set_gdt(struct kvm_vcpu *vcpu, struct desc_ptr *dt) vmcb_mark_dirty(svm->vmcb, VMCB_DT); } -static void update_cr0_intercept(struct vcpu_svm *svm) +static void sev_post_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) { - ulong gcr0; - u64 *hcr0; + struct vcpu_svm *svm = to_svm(vcpu); /* - * SEV-ES guests must always keep the CR intercepts cleared. CR - * tracking is done using the CR write traps. + * For guests that don't set guest_state_protected, the cr3 update is + * handled via kvm_mmu_load() while entering the guest. For guests + * that do (SEV-ES/SEV-SNP), the cr3 update needs to be written to + * VMCB save area now, since the save area will become the initial + * contents of the VMSA, and future VMCB save area updates won't be + * seen. */ - if (sev_es_guest(svm->vcpu.kvm)) - return; - - gcr0 = svm->vcpu.arch.cr0; - hcr0 = &svm->vmcb->save.cr0; - *hcr0 = (*hcr0 & ~SVM_CR0_SELECTIVE_MASK) - | (gcr0 & SVM_CR0_SELECTIVE_MASK); - - vmcb_mark_dirty(svm->vmcb, VMCB_CR); - - if (gcr0 == *hcr0) { - svm_clr_intercept(svm, INTERCEPT_CR0_READ); - svm_clr_intercept(svm, INTERCEPT_CR0_WRITE); - } else { - svm_set_intercept(svm, INTERCEPT_CR0_READ); - svm_set_intercept(svm, INTERCEPT_CR0_WRITE); + if (sev_es_guest(vcpu->kvm)) { + svm->vmcb->save.cr3 = cr3; + vmcb_mark_dirty(svm->vmcb, VMCB_CR); } } void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) { struct vcpu_svm *svm = to_svm(vcpu); + u64 hcr0 = cr0; + bool old_paging = is_paging(vcpu); #ifdef CONFIG_X86_64 if (vcpu->arch.efer & EFER_LME && !vcpu->arch.guest_state_protected) { @@ -1756,8 +1696,11 @@ void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) #endif vcpu->arch.cr0 = cr0; - if (!npt_enabled) - cr0 |= X86_CR0_PG | X86_CR0_WP; + if (!npt_enabled) { + hcr0 |= X86_CR0_PG | X86_CR0_WP; + if (old_paging != is_paging(vcpu)) + svm_set_cr4(vcpu, kvm_read_cr4(vcpu)); + } /* * re-enable caching here because the QEMU bios @@ -1765,10 +1708,26 @@ void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) * reboot */ if (kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED)) - cr0 &= ~(X86_CR0_CD | X86_CR0_NW); - svm->vmcb->save.cr0 = cr0; + hcr0 &= ~(X86_CR0_CD | X86_CR0_NW); + + svm->vmcb->save.cr0 = hcr0; vmcb_mark_dirty(svm->vmcb, VMCB_CR); - update_cr0_intercept(svm); + + /* + * SEV-ES guests must always keep the CR intercepts cleared. CR + * tracking is done using the CR write traps. + */ + if (sev_es_guest(vcpu->kvm)) + return; + + if (hcr0 == cr0) { + /* Selective CR0 write remains on. */ + svm_clr_intercept(svm, INTERCEPT_CR0_READ); + svm_clr_intercept(svm, INTERCEPT_CR0_WRITE); + } else { + svm_set_intercept(svm, INTERCEPT_CR0_READ); + svm_set_intercept(svm, INTERCEPT_CR0_WRITE); + } } static bool svm_is_valid_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) @@ -1782,11 +1741,15 @@ void svm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) unsigned long old_cr4 = vcpu->arch.cr4; if (npt_enabled && ((old_cr4 ^ cr4) & X86_CR4_PGE)) - svm_flush_tlb(vcpu); + svm_flush_tlb_current(vcpu); vcpu->arch.cr4 = cr4; - if (!npt_enabled) + if (!npt_enabled) { cr4 |= X86_CR4_PAE; + + if (!is_paging(vcpu)) + cr4 &= ~(X86_CR4_SMEP | X86_CR4_SMAP | X86_CR4_PKE); + } cr4 |= host_cr4_mce; to_svm(vcpu)->vmcb->save.cr4 = cr4; vmcb_mark_dirty(to_svm(vcpu)->vmcb, VMCB_CR); @@ -1847,7 +1810,7 @@ static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *sd) vmcb_mark_dirty(svm->vmcb, VMCB_ASID); } - svm->asid_generation = sd->asid_generation; + svm->current_vmcb->asid_generation = sd->asid_generation; svm->asid = sd->next_asid++; } @@ -1896,39 +1859,43 @@ static void svm_set_dr7(struct kvm_vcpu *vcpu, unsigned long value) vmcb_mark_dirty(svm->vmcb, VMCB_DR); } -static int pf_interception(struct vcpu_svm *svm) +static int pf_interception(struct kvm_vcpu *vcpu) { - u64 fault_address = __sme_clr(svm->vmcb->control.exit_info_2); + struct vcpu_svm *svm = to_svm(vcpu); + + u64 fault_address = svm->vmcb->control.exit_info_2; u64 error_code = svm->vmcb->control.exit_info_1; - return kvm_handle_page_fault(&svm->vcpu, error_code, fault_address, + return kvm_handle_page_fault(vcpu, error_code, fault_address, static_cpu_has(X86_FEATURE_DECODEASSISTS) ? svm->vmcb->control.insn_bytes : NULL, svm->vmcb->control.insn_len); } -static int npf_interception(struct vcpu_svm *svm) +static int npf_interception(struct kvm_vcpu *vcpu) { - u64 fault_address = __sme_clr(svm->vmcb->control.exit_info_2); + struct vcpu_svm *svm = to_svm(vcpu); + + u64 fault_address = svm->vmcb->control.exit_info_2; u64 error_code = svm->vmcb->control.exit_info_1; trace_kvm_page_fault(fault_address, error_code); - return kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code, + return kvm_mmu_page_fault(vcpu, fault_address, error_code, static_cpu_has(X86_FEATURE_DECODEASSISTS) ? svm->vmcb->control.insn_bytes : NULL, svm->vmcb->control.insn_len); } -static int db_interception(struct vcpu_svm *svm) +static int db_interception(struct kvm_vcpu *vcpu) { - struct kvm_run *kvm_run = svm->vcpu.run; - struct kvm_vcpu *vcpu = &svm->vcpu; + struct kvm_run *kvm_run = vcpu->run; + struct vcpu_svm *svm = to_svm(vcpu); - if (!(svm->vcpu.guest_debug & + if (!(vcpu->guest_debug & (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP)) && !svm->nmi_singlestep) { u32 payload = svm->vmcb->save.dr6 ^ DR6_ACTIVE_LOW; - kvm_queue_exception_p(&svm->vcpu, DB_VECTOR, payload); + kvm_queue_exception_p(vcpu, DB_VECTOR, payload); return 1; } @@ -1938,7 +1905,7 @@ static int db_interception(struct vcpu_svm *svm) kvm_make_request(KVM_REQ_EVENT, vcpu); } - if (svm->vcpu.guest_debug & + if (vcpu->guest_debug & (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP)) { kvm_run->exit_reason = KVM_EXIT_DEBUG; kvm_run->debug.arch.dr6 = svm->vmcb->save.dr6; @@ -1952,9 +1919,10 @@ static int db_interception(struct vcpu_svm *svm) return 1; } -static int bp_interception(struct vcpu_svm *svm) +static int bp_interception(struct kvm_vcpu *vcpu) { - struct kvm_run *kvm_run = svm->vcpu.run; + struct vcpu_svm *svm = to_svm(vcpu); + struct kvm_run *kvm_run = vcpu->run; kvm_run->exit_reason = KVM_EXIT_DEBUG; kvm_run->debug.arch.pc = svm->vmcb->save.cs.base + svm->vmcb->save.rip; @@ -1962,14 +1930,14 @@ static int bp_interception(struct vcpu_svm *svm) return 0; } -static int ud_interception(struct vcpu_svm *svm) +static int ud_interception(struct kvm_vcpu *vcpu) { - return handle_ud(&svm->vcpu); + return handle_ud(vcpu); } -static int ac_interception(struct vcpu_svm *svm) +static int ac_interception(struct kvm_vcpu *vcpu) { - kvm_queue_exception_e(&svm->vcpu, AC_VECTOR, 0); + kvm_queue_exception_e(vcpu, AC_VECTOR, 0); return 1; } @@ -2012,7 +1980,7 @@ static bool is_erratum_383(void) return true; } -static void svm_handle_mce(struct vcpu_svm *svm) +static void svm_handle_mce(struct kvm_vcpu *vcpu) { if (is_erratum_383()) { /* @@ -2021,7 +1989,7 @@ static void svm_handle_mce(struct vcpu_svm *svm) */ pr_err("KVM: Guest triggered AMD Erratum 383\n"); - kvm_make_request(KVM_REQ_TRIPLE_FAULT, &svm->vcpu); + kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); return; } @@ -2033,41 +2001,46 @@ static void svm_handle_mce(struct vcpu_svm *svm) kvm_machine_check(); } -static int mc_interception(struct vcpu_svm *svm) +static int mc_interception(struct kvm_vcpu *vcpu) { return 1; } -static int shutdown_interception(struct vcpu_svm *svm) +static int shutdown_interception(struct kvm_vcpu *vcpu) { - struct kvm_run *kvm_run = svm->vcpu.run; + struct kvm_run *kvm_run = vcpu->run; + struct vcpu_svm *svm = to_svm(vcpu); /* * The VM save area has already been encrypted so it * cannot be reinitialized - just terminate. */ - if (sev_es_guest(svm->vcpu.kvm)) + if (sev_es_guest(vcpu->kvm)) return -EINVAL; /* - * VMCB is undefined after a SHUTDOWN intercept - * so reinitialize it. + * VMCB is undefined after a SHUTDOWN intercept. INIT the vCPU to put + * the VMCB in a known good state. Unfortuately, KVM doesn't have + * KVM_MP_STATE_SHUTDOWN and can't add it without potentially breaking + * userspace. At a platform view, INIT is acceptable behavior as + * there exist bare metal platforms that automatically INIT the CPU + * in response to shutdown. */ clear_page(svm->vmcb); - init_vmcb(svm); + kvm_vcpu_reset(vcpu, true); kvm_run->exit_reason = KVM_EXIT_SHUTDOWN; return 0; } -static int io_interception(struct vcpu_svm *svm) +static int io_interception(struct kvm_vcpu *vcpu) { - struct kvm_vcpu *vcpu = &svm->vcpu; + struct vcpu_svm *svm = to_svm(vcpu); u32 io_info = svm->vmcb->control.exit_info_1; /* address size bug? */ int size, in, string; unsigned port; - ++svm->vcpu.stat.io_exits; + ++vcpu->stat.io_exits; string = (io_info & SVM_IOIO_STR_MASK) != 0; in = (io_info & SVM_IOIO_TYPE_MASK) != 0; port = io_info >> 16; @@ -2082,93 +2055,75 @@ static int io_interception(struct vcpu_svm *svm) svm->next_rip = svm->vmcb->control.exit_info_2; - return kvm_fast_pio(&svm->vcpu, size, port, in); + return kvm_fast_pio(vcpu, size, port, in); } -static int nmi_interception(struct vcpu_svm *svm) +static int nmi_interception(struct kvm_vcpu *vcpu) { return 1; } -static int intr_interception(struct vcpu_svm *svm) +static int smi_interception(struct kvm_vcpu *vcpu) { - ++svm->vcpu.stat.irq_exits; return 1; } -static int nop_on_interception(struct vcpu_svm *svm) +static int intr_interception(struct kvm_vcpu *vcpu) { + ++vcpu->stat.irq_exits; return 1; } -static int halt_interception(struct vcpu_svm *svm) +static int vmload_vmsave_interception(struct kvm_vcpu *vcpu, bool vmload) { - return kvm_emulate_halt(&svm->vcpu); -} - -static int vmmcall_interception(struct vcpu_svm *svm) -{ - return kvm_emulate_hypercall(&svm->vcpu); -} - -static int vmload_interception(struct vcpu_svm *svm) -{ - struct vmcb *nested_vmcb; + struct vcpu_svm *svm = to_svm(vcpu); + struct vmcb *vmcb12; struct kvm_host_map map; int ret; - if (nested_svm_check_permissions(svm)) + if (nested_svm_check_permissions(vcpu)) return 1; - ret = kvm_vcpu_map(&svm->vcpu, gpa_to_gfn(svm->vmcb->save.rax), &map); + ret = kvm_vcpu_map(vcpu, gpa_to_gfn(svm->vmcb->save.rax), &map); if (ret) { if (ret == -EINVAL) - kvm_inject_gp(&svm->vcpu, 0); + kvm_inject_gp(vcpu, 0); return 1; } - nested_vmcb = map.hva; + vmcb12 = map.hva; - ret = kvm_skip_emulated_instruction(&svm->vcpu); + ret = kvm_skip_emulated_instruction(vcpu); - nested_svm_vmloadsave(nested_vmcb, svm->vmcb); - kvm_vcpu_unmap(&svm->vcpu, &map, true); + if (vmload) { + svm_copy_vmloadsave_state(svm->vmcb, vmcb12); + svm->sysenter_eip_hi = 0; + svm->sysenter_esp_hi = 0; + } else { + svm_copy_vmloadsave_state(vmcb12, svm->vmcb); + } + + kvm_vcpu_unmap(vcpu, &map, true); return ret; } -static int vmsave_interception(struct vcpu_svm *svm) +static int vmload_interception(struct kvm_vcpu *vcpu) { - struct vmcb *nested_vmcb; - struct kvm_host_map map; - int ret; - - if (nested_svm_check_permissions(svm)) - return 1; - - ret = kvm_vcpu_map(&svm->vcpu, gpa_to_gfn(svm->vmcb->save.rax), &map); - if (ret) { - if (ret == -EINVAL) - kvm_inject_gp(&svm->vcpu, 0); - return 1; - } - - nested_vmcb = map.hva; - - ret = kvm_skip_emulated_instruction(&svm->vcpu); - - nested_svm_vmloadsave(svm->vmcb, nested_vmcb); - kvm_vcpu_unmap(&svm->vcpu, &map, true); + return vmload_vmsave_interception(vcpu, true); +} - return ret; +static int vmsave_interception(struct kvm_vcpu *vcpu) +{ + return vmload_vmsave_interception(vcpu, false); } -static int vmrun_interception(struct vcpu_svm *svm) +static int vmrun_interception(struct kvm_vcpu *vcpu) { - if (nested_svm_check_permissions(svm)) + if (nested_svm_check_permissions(vcpu)) return 1; - return nested_svm_vmrun(svm); + return nested_svm_vmrun(vcpu); } enum { @@ -2207,7 +2162,7 @@ static int emulate_svm_instr(struct kvm_vcpu *vcpu, int opcode) [SVM_INSTR_VMLOAD] = SVM_EXIT_VMLOAD, [SVM_INSTR_VMSAVE] = SVM_EXIT_VMSAVE, }; - int (*const svm_instr_handlers[])(struct vcpu_svm *svm) = { + int (*const svm_instr_handlers[])(struct kvm_vcpu *vcpu) = { [SVM_INSTR_VMRUN] = vmrun_interception, [SVM_INSTR_VMLOAD] = vmload_interception, [SVM_INSTR_VMSAVE] = vmsave_interception, @@ -2216,17 +2171,13 @@ static int emulate_svm_instr(struct kvm_vcpu *vcpu, int opcode) int ret; if (is_guest_mode(vcpu)) { - svm->vmcb->control.exit_code = guest_mode_exit_codes[opcode]; - svm->vmcb->control.exit_info_1 = 0; - svm->vmcb->control.exit_info_2 = 0; - /* Returns '1' or -errno on failure, '0' on success. */ - ret = nested_svm_vmexit(svm); + ret = nested_svm_simple_vmexit(svm, guest_mode_exit_codes[opcode]); if (ret) return ret; return 1; } - return svm_instr_handlers[opcode](svm); + return svm_instr_handlers[opcode](vcpu); } /* @@ -2237,9 +2188,9 @@ static int emulate_svm_instr(struct kvm_vcpu *vcpu, int opcode) * regions (e.g. SMM memory on host). * 2) VMware backdoor */ -static int gp_interception(struct vcpu_svm *svm) +static int gp_interception(struct kvm_vcpu *vcpu) { - struct kvm_vcpu *vcpu = &svm->vcpu; + struct vcpu_svm *svm = to_svm(vcpu); u32 error_code = svm->vmcb->control.exit_info_1; int opcode; @@ -2264,8 +2215,13 @@ static int gp_interception(struct vcpu_svm *svm) if (!is_guest_mode(vcpu)) return kvm_emulate_instruction(vcpu, EMULTYPE_VMWARE_GP | EMULTYPE_NO_DECODE); - } else + } else { + /* All SVM instructions expect page aligned RAX */ + if (svm->vmcb->save.rax & ~PAGE_MASK) + goto reinject; + return emulate_svm_instr(vcpu, opcode); + } reinject: kvm_queue_exception_e(vcpu, GP_VECTOR, error_code); @@ -2281,7 +2237,7 @@ void svm_set_gif(struct vcpu_svm *svm, bool value) * Likewise, clear the VINTR intercept, we will set it * again while processing KVM_REQ_EVENT if needed. */ - if (vgif_enabled(svm)) + if (vgif) svm_clr_intercept(svm, INTERCEPT_STGI); if (svm_is_intercept(svm, INTERCEPT_VINTR)) svm_clear_vintr(svm); @@ -2299,78 +2255,63 @@ void svm_set_gif(struct vcpu_svm *svm, bool value) * in use, we still rely on the VINTR intercept (rather than * STGI) to detect an open interrupt window. */ - if (!vgif_enabled(svm)) + if (!vgif) svm_clear_vintr(svm); } } -static int stgi_interception(struct vcpu_svm *svm) +static int stgi_interception(struct kvm_vcpu *vcpu) { int ret; - if (nested_svm_check_permissions(svm)) + if (nested_svm_check_permissions(vcpu)) return 1; - ret = kvm_skip_emulated_instruction(&svm->vcpu); - svm_set_gif(svm, true); + ret = kvm_skip_emulated_instruction(vcpu); + svm_set_gif(to_svm(vcpu), true); return ret; } -static int clgi_interception(struct vcpu_svm *svm) +static int clgi_interception(struct kvm_vcpu *vcpu) { int ret; - if (nested_svm_check_permissions(svm)) + if (nested_svm_check_permissions(vcpu)) return 1; - ret = kvm_skip_emulated_instruction(&svm->vcpu); - svm_set_gif(svm, false); + ret = kvm_skip_emulated_instruction(vcpu); + svm_set_gif(to_svm(vcpu), false); return ret; } -static int invlpga_interception(struct vcpu_svm *svm) +static int invlpga_interception(struct kvm_vcpu *vcpu) { - struct kvm_vcpu *vcpu = &svm->vcpu; - - trace_kvm_invlpga(svm->vmcb->save.rip, kvm_rcx_read(&svm->vcpu), - kvm_rax_read(&svm->vcpu)); + gva_t gva = kvm_rax_read(vcpu); + u32 asid = kvm_rcx_read(vcpu); - /* Let's treat INVLPGA the same as INVLPG (can be optimized!) */ - kvm_mmu_invlpg(vcpu, kvm_rax_read(&svm->vcpu)); - - return kvm_skip_emulated_instruction(&svm->vcpu); -} + /* FIXME: Handle an address size prefix. */ + if (!is_long_mode(vcpu)) + gva = (u32)gva; -static int skinit_interception(struct vcpu_svm *svm) -{ - trace_kvm_skinit(svm->vmcb->save.rip, kvm_rax_read(&svm->vcpu)); + trace_kvm_invlpga(to_svm(vcpu)->vmcb->save.rip, asid, gva); - kvm_queue_exception(&svm->vcpu, UD_VECTOR); - return 1; -} + /* Let's treat INVLPGA the same as INVLPG (can be optimized!) */ + kvm_mmu_invlpg(vcpu, gva); -static int wbinvd_interception(struct vcpu_svm *svm) -{ - return kvm_emulate_wbinvd(&svm->vcpu); + return kvm_skip_emulated_instruction(vcpu); } -static int xsetbv_interception(struct vcpu_svm *svm) +static int skinit_interception(struct kvm_vcpu *vcpu) { - u64 new_bv = kvm_read_edx_eax(&svm->vcpu); - u32 index = kvm_rcx_read(&svm->vcpu); + trace_kvm_skinit(to_svm(vcpu)->vmcb->save.rip, kvm_rax_read(vcpu)); - int err = kvm_set_xcr(&svm->vcpu, index, new_bv); - return kvm_complete_insn_gp(&svm->vcpu, err); -} - -static int rdpru_interception(struct vcpu_svm *svm) -{ - kvm_queue_exception(&svm->vcpu, UD_VECTOR); + kvm_queue_exception(vcpu, UD_VECTOR); return 1; } -static int task_switch_interception(struct vcpu_svm *svm) +static int task_switch_interception(struct kvm_vcpu *vcpu) { + struct vcpu_svm *svm = to_svm(vcpu); u16 tss_selector; int reason; int int_type = svm->vmcb->control.exit_int_info & @@ -2399,7 +2340,7 @@ static int task_switch_interception(struct vcpu_svm *svm) if (reason == TASK_SWITCH_GATE) { switch (type) { case SVM_EXITINTINFO_TYPE_NMI: - svm->vcpu.arch.nmi_injected = false; + vcpu->arch.nmi_injected = false; break; case SVM_EXITINTINFO_TYPE_EXEPT: if (svm->vmcb->control.exit_info_2 & @@ -2408,10 +2349,10 @@ static int task_switch_interception(struct vcpu_svm *svm) error_code = (u32)svm->vmcb->control.exit_info_2; } - kvm_clear_exception_queue(&svm->vcpu); + kvm_clear_exception_queue(vcpu); break; case SVM_EXITINTINFO_TYPE_INTR: - kvm_clear_interrupt_queue(&svm->vcpu); + kvm_clear_interrupt_queue(vcpu); break; default: break; @@ -2422,78 +2363,59 @@ static int task_switch_interception(struct vcpu_svm *svm) int_type == SVM_EXITINTINFO_TYPE_SOFT || (int_type == SVM_EXITINTINFO_TYPE_EXEPT && (int_vec == OF_VECTOR || int_vec == BP_VECTOR))) { - if (!skip_emulated_instruction(&svm->vcpu)) + if (!svm_skip_emulated_instruction(vcpu)) return 0; } if (int_type != SVM_EXITINTINFO_TYPE_SOFT) int_vec = -1; - return kvm_task_switch(&svm->vcpu, tss_selector, int_vec, reason, + return kvm_task_switch(vcpu, tss_selector, int_vec, reason, has_error_code, error_code); } -static int cpuid_interception(struct vcpu_svm *svm) +static int iret_interception(struct kvm_vcpu *vcpu) { - return kvm_emulate_cpuid(&svm->vcpu); -} + struct vcpu_svm *svm = to_svm(vcpu); -static int iret_interception(struct vcpu_svm *svm) -{ - ++svm->vcpu.stat.nmi_window_exits; - svm->vcpu.arch.hflags |= HF_IRET_MASK; - if (!sev_es_guest(svm->vcpu.kvm)) { + ++vcpu->stat.nmi_window_exits; + vcpu->arch.hflags |= HF_IRET_MASK; + if (!sev_es_guest(vcpu->kvm)) { svm_clr_intercept(svm, INTERCEPT_IRET); - svm->nmi_iret_rip = kvm_rip_read(&svm->vcpu); + svm->nmi_iret_rip = kvm_rip_read(vcpu); } - kvm_make_request(KVM_REQ_EVENT, &svm->vcpu); + kvm_make_request(KVM_REQ_EVENT, vcpu); return 1; } -static int invd_interception(struct vcpu_svm *svm) -{ - /* Treat an INVD instruction as a NOP and just skip it. */ - return kvm_skip_emulated_instruction(&svm->vcpu); -} - -static int invlpg_interception(struct vcpu_svm *svm) +static int invlpg_interception(struct kvm_vcpu *vcpu) { if (!static_cpu_has(X86_FEATURE_DECODEASSISTS)) - return kvm_emulate_instruction(&svm->vcpu, 0); - - kvm_mmu_invlpg(&svm->vcpu, svm->vmcb->control.exit_info_1); - return kvm_skip_emulated_instruction(&svm->vcpu); -} + return kvm_emulate_instruction(vcpu, 0); -static int emulate_on_interception(struct vcpu_svm *svm) -{ - return kvm_emulate_instruction(&svm->vcpu, 0); + kvm_mmu_invlpg(vcpu, to_svm(vcpu)->vmcb->control.exit_info_1); + return kvm_skip_emulated_instruction(vcpu); } -static int rsm_interception(struct vcpu_svm *svm) +static int emulate_on_interception(struct kvm_vcpu *vcpu) { - return kvm_emulate_instruction_from_buffer(&svm->vcpu, rsm_ins_bytes, 2); + return kvm_emulate_instruction(vcpu, 0); } -static int rdpmc_interception(struct vcpu_svm *svm) +static int rsm_interception(struct kvm_vcpu *vcpu) { - int err; - - if (!nrips) - return emulate_on_interception(svm); - - err = kvm_rdpmc(&svm->vcpu); - return kvm_complete_insn_gp(&svm->vcpu, err); + return kvm_emulate_instruction_from_buffer(vcpu, rsm_ins_bytes, 2); } -static bool check_selective_cr0_intercepted(struct vcpu_svm *svm, +static bool check_selective_cr0_intercepted(struct kvm_vcpu *vcpu, unsigned long val) { - unsigned long cr0 = svm->vcpu.arch.cr0; + struct vcpu_svm *svm = to_svm(vcpu); + unsigned long cr0 = vcpu->arch.cr0; bool ret = false; - if (!is_guest_mode(&svm->vcpu) || - (!(vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_SELECTIVE_CR0)))) + if (!is_guest_mode(vcpu) || + (!(vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_SELECTIVE_CR0)))) return false; cr0 &= ~SVM_CR0_SELECTIVE_MASK; @@ -2509,17 +2431,18 @@ static bool check_selective_cr0_intercepted(struct vcpu_svm *svm, #define CR_VALID (1ULL << 63) -static int cr_interception(struct vcpu_svm *svm) +static int cr_interception(struct kvm_vcpu *vcpu) { + struct vcpu_svm *svm = to_svm(vcpu); int reg, cr; unsigned long val; int err; if (!static_cpu_has(X86_FEATURE_DECODEASSISTS)) - return emulate_on_interception(svm); + return emulate_on_interception(vcpu); if (unlikely((svm->vmcb->control.exit_info_1 & CR_VALID) == 0)) - return emulate_on_interception(svm); + return emulate_on_interception(vcpu); reg = svm->vmcb->control.exit_info_1 & SVM_EXITINFO_REG_MASK; if (svm->vmcb->control.exit_code == SVM_EXIT_CR0_SEL_WRITE) @@ -2530,61 +2453,61 @@ static int cr_interception(struct vcpu_svm *svm) err = 0; if (cr >= 16) { /* mov to cr */ cr -= 16; - val = kvm_register_read(&svm->vcpu, reg); + val = kvm_register_read(vcpu, reg); trace_kvm_cr_write(cr, val); switch (cr) { case 0: - if (!check_selective_cr0_intercepted(svm, val)) - err = kvm_set_cr0(&svm->vcpu, val); + if (!check_selective_cr0_intercepted(vcpu, val)) + err = kvm_set_cr0(vcpu, val); else return 1; break; case 3: - err = kvm_set_cr3(&svm->vcpu, val); + err = kvm_set_cr3(vcpu, val); break; case 4: - err = kvm_set_cr4(&svm->vcpu, val); + err = kvm_set_cr4(vcpu, val); break; case 8: - err = kvm_set_cr8(&svm->vcpu, val); + err = kvm_set_cr8(vcpu, val); break; default: WARN(1, "unhandled write to CR%d", cr); - kvm_queue_exception(&svm->vcpu, UD_VECTOR); + kvm_queue_exception(vcpu, UD_VECTOR); return 1; } } else { /* mov from cr */ switch (cr) { case 0: - val = kvm_read_cr0(&svm->vcpu); + val = kvm_read_cr0(vcpu); break; case 2: - val = svm->vcpu.arch.cr2; + val = vcpu->arch.cr2; break; case 3: - val = kvm_read_cr3(&svm->vcpu); + val = kvm_read_cr3(vcpu); break; case 4: - val = kvm_read_cr4(&svm->vcpu); + val = kvm_read_cr4(vcpu); break; case 8: - val = kvm_get_cr8(&svm->vcpu); + val = kvm_get_cr8(vcpu); break; default: WARN(1, "unhandled read from CR%d", cr); - kvm_queue_exception(&svm->vcpu, UD_VECTOR); + kvm_queue_exception(vcpu, UD_VECTOR); return 1; } - kvm_register_write(&svm->vcpu, reg, val); + kvm_register_write(vcpu, reg, val); trace_kvm_cr_read(cr, val); } - return kvm_complete_insn_gp(&svm->vcpu, err); + return kvm_complete_insn_gp(vcpu, err); } -static int cr_trap(struct vcpu_svm *svm) +static int cr_trap(struct kvm_vcpu *vcpu) { - struct kvm_vcpu *vcpu = &svm->vcpu; + struct vcpu_svm *svm = to_svm(vcpu); unsigned long old_value, new_value; unsigned int cr; int ret = 0; @@ -2606,7 +2529,7 @@ static int cr_trap(struct vcpu_svm *svm) kvm_post_set_cr4(vcpu, old_value, new_value); break; case 8: - ret = kvm_set_cr8(&svm->vcpu, new_value); + ret = kvm_set_cr8(vcpu, new_value); break; default: WARN(1, "unhandled CR%d write trap", cr); @@ -2617,57 +2540,57 @@ static int cr_trap(struct vcpu_svm *svm) return kvm_complete_insn_gp(vcpu, ret); } -static int dr_interception(struct vcpu_svm *svm) +static int dr_interception(struct kvm_vcpu *vcpu) { + struct vcpu_svm *svm = to_svm(vcpu); int reg, dr; unsigned long val; int err = 0; - if (svm->vcpu.guest_debug == 0) { + if (vcpu->guest_debug == 0) { /* * No more DR vmexits; force a reload of the debug registers * and reenter on this instruction. The next vmexit will * retrieve the full state of the debug registers. */ clr_dr_intercepts(svm); - svm->vcpu.arch.switch_db_regs |= KVM_DEBUGREG_WONT_EXIT; + vcpu->arch.switch_db_regs |= KVM_DEBUGREG_WONT_EXIT; return 1; } if (!boot_cpu_has(X86_FEATURE_DECODEASSISTS)) - return emulate_on_interception(svm); + return emulate_on_interception(vcpu); reg = svm->vmcb->control.exit_info_1 & SVM_EXITINFO_REG_MASK; dr = svm->vmcb->control.exit_code - SVM_EXIT_READ_DR0; if (dr >= 16) { /* mov to DRn */ dr -= 16; - val = kvm_register_read(&svm->vcpu, reg); - err = kvm_set_dr(&svm->vcpu, dr, val); + val = kvm_register_read(vcpu, reg); + err = kvm_set_dr(vcpu, dr, val); } else { - kvm_get_dr(&svm->vcpu, dr, &val); - kvm_register_write(&svm->vcpu, reg, val); + kvm_get_dr(vcpu, dr, &val); + kvm_register_write(vcpu, reg, val); } - return kvm_complete_insn_gp(&svm->vcpu, err); + return kvm_complete_insn_gp(vcpu, err); } -static int cr8_write_interception(struct vcpu_svm *svm) +static int cr8_write_interception(struct kvm_vcpu *vcpu) { - struct kvm_run *kvm_run = svm->vcpu.run; int r; - u8 cr8_prev = kvm_get_cr8(&svm->vcpu); + u8 cr8_prev = kvm_get_cr8(vcpu); /* instruction emulation calls kvm_set_cr8() */ - r = cr_interception(svm); - if (lapic_in_kernel(&svm->vcpu)) + r = cr_interception(vcpu); + if (lapic_in_kernel(vcpu)) return r; - if (cr8_prev <= kvm_get_cr8(&svm->vcpu)) + if (cr8_prev <= kvm_get_cr8(vcpu)) return r; - kvm_run->exit_reason = KVM_EXIT_SET_TPR; + vcpu->run->exit_reason = KVM_EXIT_SET_TPR; return 0; } -static int efer_trap(struct vcpu_svm *svm) +static int efer_trap(struct kvm_vcpu *vcpu) { struct msr_data msr_info; int ret; @@ -2680,10 +2603,10 @@ static int efer_trap(struct vcpu_svm *svm) */ msr_info.host_initiated = false; msr_info.index = MSR_EFER; - msr_info.data = svm->vmcb->control.exit_info_1 & ~EFER_SVME; - ret = kvm_set_msr_common(&svm->vcpu, &msr_info); + msr_info.data = to_svm(vcpu)->vmcb->control.exit_info_1 & ~EFER_SVME; + ret = kvm_set_msr_common(vcpu, &msr_info); - return kvm_complete_insn_gp(&svm->vcpu, ret); + return kvm_complete_insn_gp(vcpu, ret); } static int svm_get_msr_feature(struct kvm_msr_entry *msr) @@ -2709,56 +2632,50 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) struct vcpu_svm *svm = to_svm(vcpu); switch (msr_info->index) { + case MSR_AMD64_TSC_RATIO: + if (!msr_info->host_initiated && !svm->tsc_scaling_enabled) + return 1; + msr_info->data = svm->tsc_ratio_msr; + break; case MSR_STAR: - msr_info->data = svm->vmcb->save.star; + msr_info->data = svm->vmcb01.ptr->save.star; break; #ifdef CONFIG_X86_64 case MSR_LSTAR: - msr_info->data = svm->vmcb->save.lstar; + msr_info->data = svm->vmcb01.ptr->save.lstar; break; case MSR_CSTAR: - msr_info->data = svm->vmcb->save.cstar; + msr_info->data = svm->vmcb01.ptr->save.cstar; break; case MSR_KERNEL_GS_BASE: - msr_info->data = svm->vmcb->save.kernel_gs_base; + msr_info->data = svm->vmcb01.ptr->save.kernel_gs_base; break; case MSR_SYSCALL_MASK: - msr_info->data = svm->vmcb->save.sfmask; + msr_info->data = svm->vmcb01.ptr->save.sfmask; break; #endif case MSR_IA32_SYSENTER_CS: - msr_info->data = svm->vmcb->save.sysenter_cs; + msr_info->data = svm->vmcb01.ptr->save.sysenter_cs; break; case MSR_IA32_SYSENTER_EIP: - msr_info->data = svm->sysenter_eip; + msr_info->data = (u32)svm->vmcb01.ptr->save.sysenter_eip; + if (guest_cpuid_is_intel(vcpu)) + msr_info->data |= (u64)svm->sysenter_eip_hi << 32; break; case MSR_IA32_SYSENTER_ESP: - msr_info->data = svm->sysenter_esp; + msr_info->data = svm->vmcb01.ptr->save.sysenter_esp; + if (guest_cpuid_is_intel(vcpu)) + msr_info->data |= (u64)svm->sysenter_esp_hi << 32; break; case MSR_TSC_AUX: - if (!boot_cpu_has(X86_FEATURE_RDTSCP)) - return 1; msr_info->data = svm->tsc_aux; break; - /* - * Nobody will change the following 5 values in the VMCB so we can - * safely return them on rdmsr. They will always be 0 until LBRV is - * implemented. - */ case MSR_IA32_DEBUGCTLMSR: - msr_info->data = svm->vmcb->save.dbgctl; - break; case MSR_IA32_LASTBRANCHFROMIP: - msr_info->data = svm->vmcb->save.br_from; - break; case MSR_IA32_LASTBRANCHTOIP: - msr_info->data = svm->vmcb->save.br_to; - break; case MSR_IA32_LASTINTFROMIP: - msr_info->data = svm->vmcb->save.last_excp_from; - break; case MSR_IA32_LASTINTTOIP: - msr_info->data = svm->vmcb->save.last_excp_to; + msr_info->data = svm_get_lbr_msr(svm, msr_info->index); break; case MSR_VM_HSAVE_PA: msr_info->data = svm->nested.hsave_msr; @@ -2771,7 +2688,10 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) !guest_has_spec_ctrl_msr(vcpu)) return 1; - msr_info->data = svm->spec_ctrl; + if (boot_cpu_has(X86_FEATURE_V_SPEC_CTRL)) + msr_info->data = svm->vmcb->save.spec_ctrl; + else + msr_info->data = svm->spec_ctrl; break; case MSR_AMD64_VIRT_SPEC_CTRL: if (!msr_info->host_initiated && @@ -2809,22 +2729,17 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) static int svm_complete_emulated_msr(struct kvm_vcpu *vcpu, int err) { struct vcpu_svm *svm = to_svm(vcpu); - if (!sev_es_guest(svm->vcpu.kvm) || !err) - return kvm_complete_insn_gp(&svm->vcpu, err); + if (!err || !sev_es_guest(vcpu->kvm) || WARN_ON_ONCE(!svm->sev_es.ghcb)) + return kvm_complete_insn_gp(vcpu, err); - ghcb_set_sw_exit_info_1(svm->ghcb, 1); - ghcb_set_sw_exit_info_2(svm->ghcb, + ghcb_set_sw_exit_info_1(svm->sev_es.ghcb, 1); + ghcb_set_sw_exit_info_2(svm->sev_es.ghcb, X86_TRAP_GP | SVM_EVTINJ_TYPE_EXEPT | SVM_EVTINJ_VALID); return 1; } -static int rdmsr_interception(struct vcpu_svm *svm) -{ - return kvm_emulate_rdmsr(&svm->vcpu); -} - static int svm_set_vm_cr(struct kvm_vcpu *vcpu, u64 data) { struct vcpu_svm *svm = to_svm(vcpu); @@ -2853,15 +2768,46 @@ static int svm_set_vm_cr(struct kvm_vcpu *vcpu, u64 data) static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) { struct vcpu_svm *svm = to_svm(vcpu); + int r; u32 ecx = msr->index; u64 data = msr->data; switch (ecx) { + case MSR_AMD64_TSC_RATIO: + + if (!svm->tsc_scaling_enabled) { + + if (!msr->host_initiated) + return 1; + /* + * In case TSC scaling is not enabled, always + * leave this MSR at the default value. + * + * Due to bug in qemu 6.2.0, it would try to set + * this msr to 0 if tsc scaling is not enabled. + * Ignore this value as well. + */ + if (data != 0 && data != svm->tsc_ratio_msr) + return 1; + break; + } + + if (data & SVM_TSC_RATIO_RSVD) + return 1; + + svm->tsc_ratio_msr = data; + + if (svm->tsc_scaling_enabled && is_guest_mode(vcpu)) + nested_svm_update_tsc_ratio_msr(vcpu); + + break; case MSR_IA32_CR_PAT: if (!kvm_mtrr_valid(vcpu, MSR_IA32_CR_PAT, data)) return 1; vcpu->arch.pat = data; - svm->vmcb->save.g_pat = data; + svm->vmcb01.ptr->save.g_pat = data; + if (is_guest_mode(vcpu)) + nested_vmcb02_compute_g_pat(svm); vmcb_mark_dirty(svm->vmcb, VMCB_NPT); break; case MSR_IA32_SPEC_CTRL: @@ -2872,7 +2818,10 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) if (kvm_spec_ctrl_test_value(data)) return 1; - svm->spec_ctrl = data; + if (boot_cpu_has(X86_FEATURE_V_SPEC_CTRL)) + svm->vmcb->save.spec_ctrl = data; + else + svm->spec_ctrl = data; if (!data) break; @@ -2915,47 +2864,56 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) svm->virt_spec_ctrl = data; break; case MSR_STAR: - svm->vmcb->save.star = data; + svm->vmcb01.ptr->save.star = data; break; #ifdef CONFIG_X86_64 case MSR_LSTAR: - svm->vmcb->save.lstar = data; + svm->vmcb01.ptr->save.lstar = data; break; case MSR_CSTAR: - svm->vmcb->save.cstar = data; + svm->vmcb01.ptr->save.cstar = data; break; case MSR_KERNEL_GS_BASE: - svm->vmcb->save.kernel_gs_base = data; + svm->vmcb01.ptr->save.kernel_gs_base = data; break; case MSR_SYSCALL_MASK: - svm->vmcb->save.sfmask = data; + svm->vmcb01.ptr->save.sfmask = data; break; #endif case MSR_IA32_SYSENTER_CS: - svm->vmcb->save.sysenter_cs = data; + svm->vmcb01.ptr->save.sysenter_cs = data; break; case MSR_IA32_SYSENTER_EIP: - svm->sysenter_eip = data; - svm->vmcb->save.sysenter_eip = data; + svm->vmcb01.ptr->save.sysenter_eip = (u32)data; + /* + * We only intercept the MSR_IA32_SYSENTER_{EIP|ESP} msrs + * when we spoof an Intel vendor ID (for cross vendor migration). + * In this case we use this intercept to track the high + * 32 bit part of these msrs to support Intel's + * implementation of SYSENTER/SYSEXIT. + */ + svm->sysenter_eip_hi = guest_cpuid_is_intel(vcpu) ? (data >> 32) : 0; break; case MSR_IA32_SYSENTER_ESP: - svm->sysenter_esp = data; - svm->vmcb->save.sysenter_esp = data; + svm->vmcb01.ptr->save.sysenter_esp = (u32)data; + svm->sysenter_esp_hi = guest_cpuid_is_intel(vcpu) ? (data >> 32) : 0; break; case MSR_TSC_AUX: - if (!boot_cpu_has(X86_FEATURE_RDTSCP)) - return 1; - /* - * This is rare, so we update the MSR here instead of using - * direct_access_msrs. Doing that would require a rdmsr in - * svm_vcpu_put. + * TSC_AUX is usually changed only during boot and never read + * directly. Intercept TSC_AUX instead of exposing it to the + * guest via direct_access_msrs, and switch it via user return. */ + preempt_disable(); + r = kvm_set_user_return_msr(tsc_aux_uret_slot, data, -1ull); + preempt_enable(); + if (r) + return 1; + svm->tsc_aux = data; - wrmsrl(MSR_TSC_AUX, svm->tsc_aux); break; case MSR_IA32_DEBUGCTLMSR: - if (!boot_cpu_has(X86_FEATURE_LBRV)) { + if (!lbrv) { vcpu_unimpl(vcpu, "%s: MSR_IA32_DEBUGCTL 0x%llx, nop\n", __func__, data); break; @@ -2963,15 +2921,25 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) if (data & DEBUGCTL_RESERVED_BITS) return 1; - svm->vmcb->save.dbgctl = data; - vmcb_mark_dirty(svm->vmcb, VMCB_LBR); - if (data & (1ULL<<0)) - svm_enable_lbrv(vcpu); + if (svm->vmcb->control.virt_ext & LBR_CTL_ENABLE_MASK) + svm->vmcb->save.dbgctl = data; else - svm_disable_lbrv(vcpu); + svm->vmcb01.ptr->save.dbgctl = data; + + svm_update_lbrv(vcpu); + break; case MSR_VM_HSAVE_PA: - svm->nested.hsave_msr = data; + /* + * Old kernels did not validate the value written to + * MSR_VM_HSAVE_PA. Allow KVM_SET_MSR to set an invalid + * value to allow live migrating buggy or malicious guests + * originating from those kernels. + */ + if (!msr->host_initiated && !page_address_valid(vcpu, data)) + return 1; + + svm->nested.hsave_msr = data & PAGE_MASK; break; case MSR_VM_CR: return svm_set_vm_cr(vcpu, data); @@ -2996,84 +2964,62 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) svm->msr_decfg = data; break; } - case MSR_IA32_APICBASE: - if (kvm_vcpu_apicv_active(vcpu)) - avic_update_vapic_bar(to_svm(vcpu), data); - fallthrough; default: return kvm_set_msr_common(vcpu, msr); } return 0; } -static int wrmsr_interception(struct vcpu_svm *svm) +static int msr_interception(struct kvm_vcpu *vcpu) { - return kvm_emulate_wrmsr(&svm->vcpu); -} - -static int msr_interception(struct vcpu_svm *svm) -{ - if (svm->vmcb->control.exit_info_1) - return wrmsr_interception(svm); + if (to_svm(vcpu)->vmcb->control.exit_info_1) + return kvm_emulate_wrmsr(vcpu); else - return rdmsr_interception(svm); + return kvm_emulate_rdmsr(vcpu); } -static int interrupt_window_interception(struct vcpu_svm *svm) +static int interrupt_window_interception(struct kvm_vcpu *vcpu) { - kvm_make_request(KVM_REQ_EVENT, &svm->vcpu); - svm_clear_vintr(svm); + kvm_make_request(KVM_REQ_EVENT, vcpu); + svm_clear_vintr(to_svm(vcpu)); /* - * For AVIC, the only reason to end up here is ExtINTs. + * If not running nested, for AVIC, the only reason to end up here is ExtINTs. * In this case AVIC was temporarily disabled for * requesting the IRQ window and we have to re-enable it. + * + * If running nested, still remove the VM wide AVIC inhibit to + * support case in which the interrupt window was requested when the + * vCPU was not running nested. + + * All vCPUs which run still run nested, will remain to have their + * AVIC still inhibited due to per-cpu AVIC inhibition. */ - svm_toggle_avic_for_irq_window(&svm->vcpu, true); + kvm_clear_apicv_inhibit(vcpu->kvm, APICV_INHIBIT_REASON_IRQWIN); - ++svm->vcpu.stat.irq_window_exits; + ++vcpu->stat.irq_window_exits; return 1; } -static int pause_interception(struct vcpu_svm *svm) +static int pause_interception(struct kvm_vcpu *vcpu) { - struct kvm_vcpu *vcpu = &svm->vcpu; bool in_kernel; - /* * CPL is not made available for an SEV-ES guest, therefore * vcpu->arch.preempted_in_kernel can never be true. Just * set in_kernel to false as well. */ - in_kernel = !sev_es_guest(svm->vcpu.kvm) && svm_get_cpl(vcpu) == 0; + in_kernel = !sev_es_guest(vcpu->kvm) && svm_get_cpl(vcpu) == 0; - if (!kvm_pause_in_guest(vcpu->kvm)) - grow_ple_window(vcpu); + grow_ple_window(vcpu); kvm_vcpu_on_spin(vcpu, in_kernel); - return 1; -} - -static int nop_interception(struct vcpu_svm *svm) -{ - return kvm_skip_emulated_instruction(&(svm->vcpu)); + return kvm_skip_emulated_instruction(vcpu); } -static int monitor_interception(struct vcpu_svm *svm) +static int invpcid_interception(struct kvm_vcpu *vcpu) { - printk_once(KERN_WARNING "kvm: MONITOR instruction emulated as NOP!\n"); - return nop_interception(svm); -} - -static int mwait_interception(struct vcpu_svm *svm) -{ - printk_once(KERN_WARNING "kvm: MWAIT instruction emulated as NOP!\n"); - return nop_interception(svm); -} - -static int invpcid_interception(struct vcpu_svm *svm) -{ - struct kvm_vcpu *vcpu = &svm->vcpu; + struct vcpu_svm *svm = to_svm(vcpu); unsigned long type; gva_t gva; @@ -3090,15 +3036,10 @@ static int invpcid_interception(struct vcpu_svm *svm) type = svm->vmcb->control.exit_info_2; gva = svm->vmcb->control.exit_info_1; - if (type > 3) { - kvm_inject_gp(vcpu, 0); - return 1; - } - return kvm_handle_invpcid(vcpu, type, gva); } -static int (*const svm_exit_handlers[])(struct vcpu_svm *svm) = { +static int (*const svm_exit_handlers[])(struct kvm_vcpu *vcpu) = { [SVM_EXIT_READ_CR0] = cr_interception, [SVM_EXIT_READ_CR3] = cr_interception, [SVM_EXIT_READ_CR4] = cr_interception, @@ -3133,15 +3074,14 @@ static int (*const svm_exit_handlers[])(struct vcpu_svm *svm) = { [SVM_EXIT_EXCP_BASE + GP_VECTOR] = gp_interception, [SVM_EXIT_INTR] = intr_interception, [SVM_EXIT_NMI] = nmi_interception, - [SVM_EXIT_SMI] = nop_on_interception, - [SVM_EXIT_INIT] = nop_on_interception, + [SVM_EXIT_SMI] = smi_interception, [SVM_EXIT_VINTR] = interrupt_window_interception, - [SVM_EXIT_RDPMC] = rdpmc_interception, - [SVM_EXIT_CPUID] = cpuid_interception, + [SVM_EXIT_RDPMC] = kvm_emulate_rdpmc, + [SVM_EXIT_CPUID] = kvm_emulate_cpuid, [SVM_EXIT_IRET] = iret_interception, - [SVM_EXIT_INVD] = invd_interception, + [SVM_EXIT_INVD] = kvm_emulate_invd, [SVM_EXIT_PAUSE] = pause_interception, - [SVM_EXIT_HLT] = halt_interception, + [SVM_EXIT_HLT] = kvm_emulate_halt, [SVM_EXIT_INVLPG] = invlpg_interception, [SVM_EXIT_INVLPGA] = invlpga_interception, [SVM_EXIT_IOIO] = io_interception, @@ -3149,17 +3089,18 @@ static int (*const svm_exit_handlers[])(struct vcpu_svm *svm) = { [SVM_EXIT_TASK_SWITCH] = task_switch_interception, [SVM_EXIT_SHUTDOWN] = shutdown_interception, [SVM_EXIT_VMRUN] = vmrun_interception, - [SVM_EXIT_VMMCALL] = vmmcall_interception, + [SVM_EXIT_VMMCALL] = kvm_emulate_hypercall, [SVM_EXIT_VMLOAD] = vmload_interception, [SVM_EXIT_VMSAVE] = vmsave_interception, [SVM_EXIT_STGI] = stgi_interception, [SVM_EXIT_CLGI] = clgi_interception, [SVM_EXIT_SKINIT] = skinit_interception, - [SVM_EXIT_WBINVD] = wbinvd_interception, - [SVM_EXIT_MONITOR] = monitor_interception, - [SVM_EXIT_MWAIT] = mwait_interception, - [SVM_EXIT_XSETBV] = xsetbv_interception, - [SVM_EXIT_RDPRU] = rdpru_interception, + [SVM_EXIT_RDTSCP] = kvm_handle_invalid_op, + [SVM_EXIT_WBINVD] = kvm_emulate_wbinvd, + [SVM_EXIT_MONITOR] = kvm_emulate_monitor, + [SVM_EXIT_MWAIT] = kvm_emulate_mwait, + [SVM_EXIT_XSETBV] = kvm_emulate_xsetbv, + [SVM_EXIT_RDPRU] = kvm_handle_invalid_op, [SVM_EXIT_EFER_WRITE_TRAP] = efer_trap, [SVM_EXIT_CR0_WRITE_TRAP] = cr_trap, [SVM_EXIT_CR4_WRITE_TRAP] = cr_trap, @@ -3177,12 +3118,15 @@ static void dump_vmcb(struct kvm_vcpu *vcpu) struct vcpu_svm *svm = to_svm(vcpu); struct vmcb_control_area *control = &svm->vmcb->control; struct vmcb_save_area *save = &svm->vmcb->save; + struct vmcb_save_area *save01 = &svm->vmcb01.ptr->save; if (!dump_invalid_vmcb) { pr_warn_ratelimited("set kvm_amd.dump_invalid_vmcb=1 to dump internal KVM state.\n"); return; } + pr_err("VMCB %p, last attempted VMRUN on CPU %d\n", + svm->current_vmcb->ptr, vcpu->arch.last_vmentry_cpu); pr_err("VMCB Control Area:\n"); pr_err("%-20s%04x\n", "cr_read:", control->intercepts[INTERCEPT_CR] & 0xffff); pr_err("%-20s%04x\n", "cr_write:", control->intercepts[INTERCEPT_CR] >> 16); @@ -3239,30 +3183,30 @@ static void dump_vmcb(struct kvm_vcpu *vcpu) save->ds.limit, save->ds.base); pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", "fs:", - save->fs.selector, save->fs.attrib, - save->fs.limit, save->fs.base); + save01->fs.selector, save01->fs.attrib, + save01->fs.limit, save01->fs.base); pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", "gs:", - save->gs.selector, save->gs.attrib, - save->gs.limit, save->gs.base); + save01->gs.selector, save01->gs.attrib, + save01->gs.limit, save01->gs.base); pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", "gdtr:", save->gdtr.selector, save->gdtr.attrib, save->gdtr.limit, save->gdtr.base); pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", "ldtr:", - save->ldtr.selector, save->ldtr.attrib, - save->ldtr.limit, save->ldtr.base); + save01->ldtr.selector, save01->ldtr.attrib, + save01->ldtr.limit, save01->ldtr.base); pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", "idtr:", save->idtr.selector, save->idtr.attrib, save->idtr.limit, save->idtr.base); pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", "tr:", - save->tr.selector, save->tr.attrib, - save->tr.limit, save->tr.base); - pr_err("cpl: %d efer: %016llx\n", - save->cpl, save->efer); + save01->tr.selector, save01->tr.attrib, + save01->tr.limit, save01->tr.base); + pr_err("vmpl: %d cpl: %d efer: %016llx\n", + save->vmpl, save->cpl, save->efer); pr_err("%-15s %016llx %-13s %016llx\n", "cr0:", save->cr0, "cr2:", save->cr2); pr_err("%-15s %016llx %-13s %016llx\n", @@ -3274,15 +3218,15 @@ static void dump_vmcb(struct kvm_vcpu *vcpu) pr_err("%-15s %016llx %-13s %016llx\n", "rsp:", save->rsp, "rax:", save->rax); pr_err("%-15s %016llx %-13s %016llx\n", - "star:", save->star, "lstar:", save->lstar); + "star:", save01->star, "lstar:", save01->lstar); pr_err("%-15s %016llx %-13s %016llx\n", - "cstar:", save->cstar, "sfmask:", save->sfmask); + "cstar:", save01->cstar, "sfmask:", save01->sfmask); pr_err("%-15s %016llx %-13s %016llx\n", - "kernel_gs_base:", save->kernel_gs_base, - "sysenter_cs:", save->sysenter_cs); + "kernel_gs_base:", save01->kernel_gs_base, + "sysenter_cs:", save01->sysenter_cs); pr_err("%-15s %016llx %-13s %016llx\n", - "sysenter_esp:", save->sysenter_esp, - "sysenter_eip:", save->sysenter_eip); + "sysenter_esp:", save01->sysenter_esp, + "sysenter_eip:", save01->sysenter_eip); pr_err("%-15s %016llx %-13s %016llx\n", "gpat:", save->g_pat, "dbgctl:", save->dbgctl); pr_err("%-15s %016llx %-13s %016llx\n", @@ -3292,12 +3236,14 @@ static void dump_vmcb(struct kvm_vcpu *vcpu) "excp_to:", save->last_excp_to); } -static int svm_handle_invalid_exit(struct kvm_vcpu *vcpu, u64 exit_code) +static bool svm_check_exit_valid(u64 exit_code) { - if (exit_code < ARRAY_SIZE(svm_exit_handlers) && - svm_exit_handlers[exit_code]) - return 0; + return (exit_code < ARRAY_SIZE(svm_exit_handlers) && + svm_exit_handlers[exit_code]); +} +static int svm_handle_invalid_exit(struct kvm_vcpu *vcpu, u64 exit_code) +{ vcpu_unimpl(vcpu, "svm: unexpected exit reason 0x%llx\n", exit_code); dump_vmcb(vcpu); vcpu->run->exit_reason = KVM_EXIT_INTERNAL_ERROR; @@ -3305,35 +3251,36 @@ static int svm_handle_invalid_exit(struct kvm_vcpu *vcpu, u64 exit_code) vcpu->run->internal.ndata = 2; vcpu->run->internal.data[0] = exit_code; vcpu->run->internal.data[1] = vcpu->arch.last_vmentry_cpu; - - return -EINVAL; + return 0; } -int svm_invoke_exit_handler(struct vcpu_svm *svm, u64 exit_code) +int svm_invoke_exit_handler(struct kvm_vcpu *vcpu, u64 exit_code) { - if (svm_handle_invalid_exit(&svm->vcpu, exit_code)) - return 0; + if (!svm_check_exit_valid(exit_code)) + return svm_handle_invalid_exit(vcpu, exit_code); #ifdef CONFIG_RETPOLINE if (exit_code == SVM_EXIT_MSR) - return msr_interception(svm); + return msr_interception(vcpu); else if (exit_code == SVM_EXIT_VINTR) - return interrupt_window_interception(svm); + return interrupt_window_interception(vcpu); else if (exit_code == SVM_EXIT_INTR) - return intr_interception(svm); + return intr_interception(vcpu); else if (exit_code == SVM_EXIT_HLT) - return halt_interception(svm); + return kvm_emulate_halt(vcpu); else if (exit_code == SVM_EXIT_NPF) - return npf_interception(svm); + return npf_interception(vcpu); #endif - return svm_exit_handlers[exit_code](svm); + return svm_exit_handlers[exit_code](vcpu); } -static void svm_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2, +static void svm_get_exit_info(struct kvm_vcpu *vcpu, u32 *reason, + u64 *info1, u64 *info2, u32 *intr_info, u32 *error_code) { struct vmcb_control_area *control = &to_svm(vcpu)->vmcb->control; + *reason = control->exit_code; *info1 = control->exit_info_1; *info2 = control->exit_info_2; *intr_info = control->exit_int_info; @@ -3344,13 +3291,13 @@ static void svm_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2, *error_code = 0; } -static int handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) +static int svm_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) { struct vcpu_svm *svm = to_svm(vcpu); struct kvm_run *kvm_run = vcpu->run; u32 exit_code = svm->vmcb->control.exit_code; - trace_kvm_exit(exit_code, vcpu, KVM_ISA_SVM); + trace_kvm_exit(vcpu, KVM_ISA_SVM); /* SEV-ES guests must use the CR write traps to track CR registers. */ if (!sev_es_guest(vcpu->kvm)) { @@ -3363,7 +3310,7 @@ static int handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) if (is_guest_mode(vcpu)) { int vmexit; - trace_kvm_nested_vmexit(exit_code, vcpu, KVM_ISA_SVM); + trace_kvm_nested_vmexit(vcpu, KVM_ISA_SVM); vmexit = nested_svm_exit_special(svm); @@ -3395,7 +3342,7 @@ static int handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) if (exit_fastpath != EXIT_FASTPATH_NONE) return 1; - return svm_invoke_exit_handler(svm, exit_code); + return svm_invoke_exit_handler(vcpu, exit_code); } static void reload_tss(struct kvm_vcpu *vcpu) @@ -3406,15 +3353,27 @@ static void reload_tss(struct kvm_vcpu *vcpu) load_TR_desc(); } -static void pre_svm_run(struct vcpu_svm *svm) +static void pre_svm_run(struct kvm_vcpu *vcpu) { - struct svm_cpu_data *sd = per_cpu(svm_data, svm->vcpu.cpu); + struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu); + struct vcpu_svm *svm = to_svm(vcpu); + + /* + * If the previous vmrun of the vmcb occurred on a different physical + * cpu, then mark the vmcb dirty and assign a new asid. Hardware's + * vmcb clean bits are per logical CPU, as are KVM's asid assignments. + */ + if (unlikely(svm->current_vmcb->cpu != vcpu->cpu)) { + svm->current_vmcb->asid_generation = 0; + vmcb_mark_all_dirty(svm->vmcb); + svm->current_vmcb->cpu = vcpu->cpu; + } - if (sev_guest(svm->vcpu.kvm)) - return pre_sev_run(svm, svm->vcpu.cpu); + if (sev_guest(vcpu->kvm)) + return pre_sev_run(svm, vcpu->cpu); /* FIXME: handle wraparound of asid_generation */ - if (svm->asid_generation != sd->asid_generation) + if (svm->current_vmcb->asid_generation != sd->asid_generation) new_asid(svm, sd); } @@ -3424,12 +3383,12 @@ static void svm_inject_nmi(struct kvm_vcpu *vcpu) svm->vmcb->control.event_inj = SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_NMI; vcpu->arch.hflags |= HF_NMI_MASK; - if (!sev_es_guest(svm->vcpu.kvm)) + if (!sev_es_guest(vcpu->kvm)) svm_set_intercept(svm, INTERCEPT_IRET); ++vcpu->stat.nmi_injections; } -static void svm_set_irq(struct kvm_vcpu *vcpu) +static void svm_inject_irq(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -3442,6 +3401,55 @@ static void svm_set_irq(struct kvm_vcpu *vcpu) SVM_EVTINJ_VALID | SVM_EVTINJ_TYPE_INTR; } +void svm_complete_interrupt_delivery(struct kvm_vcpu *vcpu, int delivery_mode, + int trig_mode, int vector) +{ + /* + * vcpu->arch.apicv_active must be read after vcpu->mode. + * Pairs with smp_store_release in vcpu_enter_guest. + */ + bool in_guest_mode = (smp_load_acquire(&vcpu->mode) == IN_GUEST_MODE); + + if (!READ_ONCE(vcpu->arch.apicv_active)) { + /* Process the interrupt via inject_pending_event */ + kvm_make_request(KVM_REQ_EVENT, vcpu); + kvm_vcpu_kick(vcpu); + return; + } + + trace_kvm_apicv_accept_irq(vcpu->vcpu_id, delivery_mode, trig_mode, vector); + if (in_guest_mode) { + /* + * Signal the doorbell to tell hardware to inject the IRQ. If + * the vCPU exits the guest before the doorbell chimes, hardware + * will automatically process AVIC interrupts at the next VMRUN. + */ + avic_ring_doorbell(vcpu); + } else { + /* + * Wake the vCPU if it was blocking. KVM will then detect the + * pending IRQ when checking if the vCPU has a wake event. + */ + kvm_vcpu_wake_up(vcpu); + } +} + +static void svm_deliver_interrupt(struct kvm_lapic *apic, int delivery_mode, + int trig_mode, int vector) +{ + kvm_lapic_set_irr(vector, apic); + + /* + * Pairs with the smp_mb_*() after setting vcpu->guest_mode in + * vcpu_enter_guest() to ensure the write to the vIRR is ordered before + * the read of guest_mode. This guarantees that either VMRUN will see + * and process the new vIRR entry, or that svm_complete_interrupt_delivery + * will signal the doorbell if the CPU has already entered the guest. + */ + smp_mb__after_atomic(); + svm_complete_interrupt_delivery(apic->vcpu, delivery_mode, trig_mode, vector); +} + static void svm_update_cr8_intercept(struct kvm_vcpu *vcpu, int tpr, int irr) { struct vcpu_svm *svm = to_svm(vcpu); @@ -3478,7 +3486,7 @@ bool svm_nmi_blocked(struct kvm_vcpu *vcpu) return false; ret = (vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK) || - (svm->vcpu.arch.hflags & HF_NMI_MASK); + (vcpu->arch.hflags & HF_NMI_MASK); return ret; } @@ -3489,18 +3497,18 @@ static int svm_nmi_allowed(struct kvm_vcpu *vcpu, bool for_injection) if (svm->nested.nested_run_pending) return -EBUSY; + if (svm_nmi_blocked(vcpu)) + return 0; + /* An NMI must not be injected into L2 if it's supposed to VM-Exit. */ if (for_injection && is_guest_mode(vcpu) && nested_exit_on_nmi(svm)) return -EBUSY; - - return !svm_nmi_blocked(vcpu); + return 1; } static bool svm_get_nmi_mask(struct kvm_vcpu *vcpu) { - struct vcpu_svm *svm = to_svm(vcpu); - - return !!(svm->vcpu.arch.hflags & HF_NMI_MASK); + return !!(vcpu->arch.hflags & HF_NMI_MASK); } static void svm_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked) @@ -3508,12 +3516,12 @@ static void svm_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked) struct vcpu_svm *svm = to_svm(vcpu); if (masked) { - svm->vcpu.arch.hflags |= HF_NMI_MASK; - if (!sev_es_guest(svm->vcpu.kvm)) + vcpu->arch.hflags |= HF_NMI_MASK; + if (!sev_es_guest(vcpu->kvm)) svm_set_intercept(svm, INTERCEPT_IRET); } else { - svm->vcpu.arch.hflags &= ~HF_NMI_MASK; - if (!sev_es_guest(svm->vcpu.kvm)) + vcpu->arch.hflags &= ~HF_NMI_MASK; + if (!sev_es_guest(vcpu->kvm)) svm_clr_intercept(svm, INTERCEPT_IRET); } } @@ -3526,17 +3534,10 @@ bool svm_interrupt_blocked(struct kvm_vcpu *vcpu) if (!gif_set(svm)) return true; - if (sev_es_guest(svm->vcpu.kvm)) { - /* - * SEV-ES guests to not expose RFLAGS. Use the VMCB interrupt mask - * bit to determine the state of the IF flag. - */ - if (!(vmcb->control.int_state & SVM_GUEST_INTERRUPT_MASK)) - return true; - } else if (is_guest_mode(vcpu)) { + if (is_guest_mode(vcpu)) { /* As long as interrupts are being delivered... */ if ((svm->nested.ctl.int_ctl & V_INTR_MASKING_MASK) - ? !(svm->nested.hsave->save.rflags & X86_EFLAGS_IF) + ? !(svm->vmcb01.ptr->save.rflags & X86_EFLAGS_IF) : !(kvm_get_rflags(vcpu) & X86_EFLAGS_IF)) return true; @@ -3544,7 +3545,7 @@ bool svm_interrupt_blocked(struct kvm_vcpu *vcpu) if (nested_exit_on_intr(svm)) return false; } else { - if (!(kvm_get_rflags(vcpu) & X86_EFLAGS_IF)) + if (!svm_get_if_flag(vcpu)) return true; } @@ -3554,9 +3555,13 @@ bool svm_interrupt_blocked(struct kvm_vcpu *vcpu) static int svm_interrupt_allowed(struct kvm_vcpu *vcpu, bool for_injection) { struct vcpu_svm *svm = to_svm(vcpu); + if (svm->nested.nested_run_pending) return -EBUSY; + if (svm_interrupt_blocked(vcpu)) + return 0; + /* * An IRQ must not be injected into L2 if it's supposed to VM-Exit, * e.g. if the IRQ arrived asynchronously after checking nested events. @@ -3564,7 +3569,7 @@ static int svm_interrupt_allowed(struct kvm_vcpu *vcpu, bool for_injection) if (for_injection && is_guest_mode(vcpu) && nested_exit_on_intr(svm)) return -EBUSY; - return !svm_interrupt_blocked(vcpu); + return 1; } static void svm_enable_irq_window(struct kvm_vcpu *vcpu) @@ -3579,14 +3584,20 @@ static void svm_enable_irq_window(struct kvm_vcpu *vcpu) * enabled, the STGI interception will not occur. Enable the irq * window under the assumption that the hardware will set the GIF. */ - if (vgif_enabled(svm) || gif_set(svm)) { + if (vgif || gif_set(svm)) { /* * IRQ window is not needed when AVIC is enabled, * unless we have pending ExtINT since it cannot be injected - * via AVIC. In such case, we need to temporarily disable AVIC, + * via AVIC. In such case, KVM needs to temporarily disable AVIC, * and fallback to injecting IRQ via V_IRQ. + * + * If running nested, AVIC is already locally inhibited + * on this vCPU, therefore there is no need to request + * the VM wide AVIC inhibition. */ - svm_toggle_avic_for_irq_window(vcpu, false); + if (!is_guest_mode(vcpu)) + kvm_set_apicv_inhibit(vcpu->kvm, APICV_INHIBIT_REASON_IRQWIN); + svm_set_vintr(svm); } } @@ -3595,12 +3606,11 @@ static void svm_enable_nmi_window(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - if ((svm->vcpu.arch.hflags & (HF_NMI_MASK | HF_IRET_MASK)) - == HF_NMI_MASK) + if ((vcpu->arch.hflags & (HF_NMI_MASK | HF_IRET_MASK)) == HF_NMI_MASK) return; /* IRET will cause a vm exit */ if (!gif_set(svm)) { - if (vgif_enabled(svm)) + if (vgif) svm_set_intercept(svm, INTERCEPT_STGI); return; /* STGI will cause a vm exit */ } @@ -3614,17 +3624,7 @@ static void svm_enable_nmi_window(struct kvm_vcpu *vcpu) svm->vmcb->save.rflags |= (X86_EFLAGS_TF | X86_EFLAGS_RF); } -static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr) -{ - return 0; -} - -static int svm_set_identity_map_addr(struct kvm *kvm, u64 ident_addr) -{ - return 0; -} - -void svm_flush_tlb(struct kvm_vcpu *vcpu) +static void svm_flush_tlb_current(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -3638,7 +3638,7 @@ void svm_flush_tlb(struct kvm_vcpu *vcpu) if (static_cpu_has(X86_FEATURE_FLUSHBYASID)) svm->vmcb->control.tlb_ctl = TLB_CONTROL_FLUSH_ASID; else - svm->asid_generation--; + svm->current_vmcb->asid_generation--; } static void svm_flush_tlb_gva(struct kvm_vcpu *vcpu, gva_t gva) @@ -3675,8 +3675,9 @@ static inline void sync_lapic_to_cr8(struct kvm_vcpu *vcpu) svm->vmcb->control.int_ctl |= cr8 & V_TPR_MASK; } -static void svm_complete_interrupts(struct vcpu_svm *svm) +static void svm_complete_interrupts(struct kvm_vcpu *vcpu) { + struct vcpu_svm *svm = to_svm(vcpu); u8 vector; int type; u32 exitintinfo = svm->vmcb->control.exit_int_info; @@ -3688,28 +3689,28 @@ static void svm_complete_interrupts(struct vcpu_svm *svm) * If we've made progress since setting HF_IRET_MASK, we've * executed an IRET and can allow NMI injection. */ - if ((svm->vcpu.arch.hflags & HF_IRET_MASK) && - (sev_es_guest(svm->vcpu.kvm) || - kvm_rip_read(&svm->vcpu) != svm->nmi_iret_rip)) { - svm->vcpu.arch.hflags &= ~(HF_NMI_MASK | HF_IRET_MASK); - kvm_make_request(KVM_REQ_EVENT, &svm->vcpu); + if ((vcpu->arch.hflags & HF_IRET_MASK) && + (sev_es_guest(vcpu->kvm) || + kvm_rip_read(vcpu) != svm->nmi_iret_rip)) { + vcpu->arch.hflags &= ~(HF_NMI_MASK | HF_IRET_MASK); + kvm_make_request(KVM_REQ_EVENT, vcpu); } - svm->vcpu.arch.nmi_injected = false; - kvm_clear_exception_queue(&svm->vcpu); - kvm_clear_interrupt_queue(&svm->vcpu); + vcpu->arch.nmi_injected = false; + kvm_clear_exception_queue(vcpu); + kvm_clear_interrupt_queue(vcpu); if (!(exitintinfo & SVM_EXITINTINFO_VALID)) return; - kvm_make_request(KVM_REQ_EVENT, &svm->vcpu); + kvm_make_request(KVM_REQ_EVENT, vcpu); vector = exitintinfo & SVM_EXITINTINFO_VEC_MASK; type = exitintinfo & SVM_EXITINTINFO_TYPE_MASK; switch (type) { case SVM_EXITINTINFO_TYPE_NMI: - svm->vcpu.arch.nmi_injected = true; + vcpu->arch.nmi_injected = true; break; case SVM_EXITINTINFO_TYPE_EXEPT: /* @@ -3725,21 +3726,20 @@ static void svm_complete_interrupts(struct vcpu_svm *svm) */ if (kvm_exception_is_soft(vector)) { if (vector == BP_VECTOR && int3_injected && - kvm_is_linear_rip(&svm->vcpu, svm->int3_rip)) - kvm_rip_write(&svm->vcpu, - kvm_rip_read(&svm->vcpu) - - int3_injected); + kvm_is_linear_rip(vcpu, svm->int3_rip)) + kvm_rip_write(vcpu, + kvm_rip_read(vcpu) - int3_injected); break; } if (exitintinfo & SVM_EXITINTINFO_VALID_ERR) { u32 err = svm->vmcb->control.exit_int_info_err; - kvm_requeue_exception_e(&svm->vcpu, vector, err); + kvm_requeue_exception_e(vcpu, vector, err); } else - kvm_requeue_exception(&svm->vcpu, vector); + kvm_requeue_exception(vcpu, vector); break; case SVM_EXITINTINFO_TYPE_INTR: - kvm_queue_interrupt(&svm->vcpu, vector, false); + kvm_queue_interrupt(vcpu, vector, false); break; default: break; @@ -3754,7 +3754,12 @@ static void svm_cancel_injection(struct kvm_vcpu *vcpu) control->exit_int_info = control->event_inj; control->exit_int_info_err = control->event_inj_err; control->event_inj = 0; - svm_complete_interrupts(svm); + svm_complete_interrupts(vcpu); +} + +static int svm_vcpu_pre_run(struct kvm_vcpu *vcpu) +{ + return 1; } static fastpath_t svm_exit_handlers_fastpath(struct kvm_vcpu *vcpu) @@ -3766,57 +3771,32 @@ static fastpath_t svm_exit_handlers_fastpath(struct kvm_vcpu *vcpu) return EXIT_FASTPATH_NONE; } -static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu, - struct vcpu_svm *svm) +static noinstr void svm_vcpu_enter_exit(struct kvm_vcpu *vcpu) { - /* - * VMENTER enables interrupts (host state), but the kernel state is - * interrupts disabled when this is invoked. Also tell RCU about - * it. This is the same logic as for exit_to_user_mode(). - * - * This ensures that e.g. latency analysis on the host observes - * guest mode as interrupt enabled. - * - * guest_enter_irqoff() informs context tracking about the - * transition to guest mode and if enabled adjusts RCU state - * accordingly. - */ - instrumentation_begin(); - trace_hardirqs_on_prepare(); - lockdep_hardirqs_on_prepare(CALLER_ADDR0); - instrumentation_end(); + struct vcpu_svm *svm = to_svm(vcpu); + unsigned long vmcb_pa = svm->current_vmcb->pa; - guest_enter_irqoff(); - lockdep_hardirqs_on(CALLER_ADDR0); + guest_state_enter_irqoff(); - if (sev_es_guest(svm->vcpu.kvm)) { - __svm_sev_es_vcpu_run(svm->vmcb_pa); + if (sev_es_guest(vcpu->kvm)) { + __svm_sev_es_vcpu_run(vmcb_pa); } else { struct svm_cpu_data *sd = per_cpu(svm_data, vcpu->cpu); - __svm_vcpu_run(svm->vmcb_pa, (unsigned long *)&svm->vcpu.arch.regs); + /* + * Use a single vmcb (vmcb01 because it's always valid) for + * context switching guest state via VMLOAD/VMSAVE, that way + * the state doesn't need to be copied between vmcb01 and + * vmcb02 when switching vmcbs for nested virtualization. + */ + vmload(svm->vmcb01.pa); + __svm_vcpu_run(vmcb_pa, (unsigned long *)&vcpu->arch.regs); + vmsave(svm->vmcb01.pa); vmload(__sme_page_pa(sd->save_area)); } - /* - * VMEXIT disables interrupts (host state), but tracing and lockdep - * have them in state 'on' as recorded before entering guest mode. - * Same as enter_from_user_mode(). - * - * guest_exit_irqoff() restores host context and reinstates RCU if - * enabled and required. - * - * This needs to be done before the below as native_read_msr() - * contains a tracepoint and x86_spec_ctrl_restore_host() calls - * into world and some more. - */ - lockdep_hardirqs_off(CALLER_ADDR0); - guest_exit_irqoff(); - - instrumentation_begin(); - trace_hardirqs_off_finish(); - instrumentation_end(); + guest_state_exit_irqoff(); } static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) @@ -3845,7 +3825,7 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) smp_send_reschedule(vcpu->cpu); } - pre_svm_run(svm); + pre_svm_run(vcpu); sync_lapic_to_cr8(vcpu); @@ -3855,11 +3835,13 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) } svm->vmcb->save.cr2 = vcpu->arch.cr2; + svm_hv_update_vp_id(svm->vmcb, vcpu); + /* * Run with all-zero DR6 unless needed, so that we can get the exact cause * of a #DB. */ - if (unlikely(svm->vcpu.arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT)) + if (unlikely(vcpu->arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT)) svm_set_dr6(svm, vcpu->arch.dr6); else svm_set_dr6(svm, DR6_ACTIVE_LOW); @@ -3875,9 +3857,10 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) * is no need to worry about the conditional branch over the wrmsr * being speculatively taken. */ - x86_spec_ctrl_set_guest(svm->spec_ctrl, svm->virt_spec_ctrl); + if (!static_cpu_has(X86_FEATURE_V_SPEC_CTRL)) + x86_spec_ctrl_set_guest(svm->spec_ctrl, svm->virt_spec_ctrl); - svm_vcpu_enter_exit(vcpu, svm); + svm_vcpu_enter_exit(vcpu); /* * We do not use IBRS in the kernel. If this vCPU has used the @@ -3894,23 +3877,26 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) * If the L02 MSR bitmap does not intercept the MSR, then we need to * save it. */ - if (unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL))) + if (!static_cpu_has(X86_FEATURE_V_SPEC_CTRL) && + unlikely(!msr_write_intercepted(vcpu, MSR_IA32_SPEC_CTRL))) svm->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL); - if (!sev_es_guest(svm->vcpu.kvm)) + if (!sev_es_guest(vcpu->kvm)) reload_tss(vcpu); - x86_spec_ctrl_restore_host(svm->spec_ctrl, svm->virt_spec_ctrl); + if (!static_cpu_has(X86_FEATURE_V_SPEC_CTRL)) + x86_spec_ctrl_restore_host(svm->spec_ctrl, svm->virt_spec_ctrl); - if (!sev_es_guest(svm->vcpu.kvm)) { + if (!sev_es_guest(vcpu->kvm)) { vcpu->arch.cr2 = svm->vmcb->save.cr2; vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax; vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip; } + vcpu->arch.regs_dirty = 0; if (unlikely(svm->vmcb->control.exit_code == SVM_EXIT_NMI)) - kvm_before_interrupt(&svm->vcpu); + kvm_before_interrupt(vcpu, KVM_HANDLING_NMI); kvm_load_host_xsave_state(vcpu); stgi(); @@ -3918,13 +3904,19 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) /* Any pending NMI will happen here */ if (unlikely(svm->vmcb->control.exit_code == SVM_EXIT_NMI)) - kvm_after_interrupt(&svm->vcpu); + kvm_after_interrupt(vcpu); sync_cr8_to_lapic(vcpu); svm->next_rip = 0; - if (is_guest_mode(&svm->vcpu)) { - sync_nested_vmcb_control(svm); + if (is_guest_mode(vcpu)) { + nested_sync_control_from_vmcb02(svm); + + /* Track VMRUNs that have made past consistency checking */ + if (svm->nested.nested_run_pending && + svm->vmcb->control.exit_code != SVM_EXIT_ERR) + ++vcpu->stat.nested_run; + svm->nested.nested_run_pending = 0; } @@ -3933,13 +3925,10 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) /* if exit due to PF check for async PF */ if (svm->vmcb->control.exit_code == SVM_EXIT_EXCP_BASE + PF_VECTOR) - svm->vcpu.arch.apf.host_apf_flags = + vcpu->arch.apf.host_apf_flags = kvm_read_and_reset_apf_flags(); - if (npt_enabled) { - vcpu->arch.regs_avail &= ~(1 << VCPU_EXREG_PDPTR); - vcpu->arch.regs_dirty &= ~(1 << VCPU_EXREG_PDPTR); - } + vcpu->arch.regs_avail &= ~SVM_REGS_LAZY_LOAD_SET; /* * We need to handle MC intercepts here before the vcpu has a chance to @@ -3947,9 +3936,9 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) */ if (unlikely(svm->vmcb->control.exit_code == SVM_EXIT_EXCP_BASE + MC_VECTOR)) - svm_handle_mce(svm); + svm_handle_mce(vcpu); - svm_complete_interrupts(svm); + svm_complete_interrupts(vcpu); if (is_guest_mode(vcpu)) return EXIT_FASTPATH_NONE; @@ -3957,21 +3946,25 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) return svm_exit_handlers_fastpath(vcpu); } -static void svm_load_mmu_pgd(struct kvm_vcpu *vcpu, unsigned long root, +static void svm_load_mmu_pgd(struct kvm_vcpu *vcpu, hpa_t root_hpa, int root_level) { struct vcpu_svm *svm = to_svm(vcpu); unsigned long cr3; - cr3 = __sme_set(root); if (npt_enabled) { - svm->vmcb->control.nested_cr3 = cr3; + svm->vmcb->control.nested_cr3 = __sme_set(root_hpa); vmcb_mark_dirty(svm->vmcb, VMCB_NPT); - /* Loading L2's CR3 is handled by enter_svm_guest_mode. */ - if (!test_bit(VCPU_EXREG_CR3, (ulong *)&vcpu->arch.regs_avail)) - return; + hv_track_root_tdp(vcpu, root_hpa); + cr3 = vcpu->arch.cr3; + } else if (vcpu->arch.mmu->root_role.level >= PT64_ROOT_4LEVEL) { + cr3 = __sme_set(root_hpa) | kvm_get_active_pcid(vcpu); + } else { + /* PCID in the guest should be impossible with a 32-bit MMU. */ + WARN_ON_ONCE(kvm_get_active_pcid(vcpu)); + cr3 = root_hpa; } svm->vmcb->save.cr3 = cr3; @@ -4005,11 +3998,6 @@ static int __init svm_check_processor_compat(void) return 0; } -static bool svm_cpu_has_accelerated_tpr(void) -{ - return false; -} - /* * The kvm parameter can be NULL (module initialization, or invocation before * VM creation). Be sure to check the kvm parameter before using it. @@ -4041,6 +4029,7 @@ static void svm_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); struct kvm_cpuid_entry2 *best; + struct kvm *kvm = vcpu->kvm; vcpu->arch.xsaves_enabled = guest_cpuid_has(vcpu, X86_FEATURE_XSAVE) && boot_cpu_has(X86_FEATURE_XSAVE) && @@ -4048,10 +4037,22 @@ static void svm_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu) /* Update nrips enabled cache */ svm->nrips_enabled = kvm_cpu_cap_has(X86_FEATURE_NRIPS) && - guest_cpuid_has(&svm->vcpu, X86_FEATURE_NRIPS); + guest_cpuid_has(vcpu, X86_FEATURE_NRIPS); + + svm->tsc_scaling_enabled = tsc_scaling && guest_cpuid_has(vcpu, X86_FEATURE_TSCRATEMSR); + svm->lbrv_enabled = lbrv && guest_cpuid_has(vcpu, X86_FEATURE_LBRV); + + svm->v_vmload_vmsave_enabled = vls && guest_cpuid_has(vcpu, X86_FEATURE_V_VMSAVE_VMLOAD); + + svm->pause_filter_enabled = kvm_cpu_cap_has(X86_FEATURE_PAUSEFILTER) && + guest_cpuid_has(vcpu, X86_FEATURE_PAUSEFILTER); - /* Check again if INVPCID interception if required */ - svm_check_invpcid(svm); + svm->pause_threshold_enabled = kvm_cpu_cap_has(X86_FEATURE_PFTHRESHOLD) && + guest_cpuid_has(vcpu, X86_FEATURE_PFTHRESHOLD); + + svm->vgif_enabled = vgif && guest_cpuid_has(vcpu, X86_FEATURE_VGIF); + + svm_recalc_instruction_intercepts(vcpu, svm); /* For sev guests, the memory encryption bit is not reserved in CR3. */ if (sev_guest(vcpu->kvm)) { @@ -4060,24 +4061,15 @@ static void svm_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu) vcpu->arch.reserved_gpa_bits &= ~(1UL << (best->ebx & 0x3f)); } - if (!kvm_vcpu_apicv_active(vcpu)) - return; - - /* - * AVIC does not work with an x2APIC mode guest. If the X2APIC feature - * is exposed to the guest, disable AVIC. - */ - if (guest_cpuid_has(vcpu, X86_FEATURE_X2APIC)) - kvm_request_apicv_update(vcpu->kvm, false, - APICV_INHIBIT_REASON_X2APIC); - - /* - * Currently, AVIC does not work with nested virtualization. - * So, we disable AVIC when cpuid for SVM is set in the L1 guest. - */ - if (nested && guest_cpuid_has(vcpu, X86_FEATURE_SVM)) - kvm_request_apicv_update(vcpu->kvm, false, - APICV_INHIBIT_REASON_NESTED); + if (kvm_vcpu_apicv_active(vcpu)) { + /* + * AVIC does not work with an x2APIC mode guest. If the X2APIC feature + * is exposed to the guest, disable AVIC. + */ + if (guest_cpuid_has(vcpu, X86_FEATURE_X2APIC)) + kvm_set_apicv_inhibit(kvm, APICV_INHIBIT_REASON_X2APIC); + } + init_vmcb_after_set_cpuid(vcpu); } static bool svm_has_wbinvd_exit(void) @@ -4182,7 +4174,7 @@ static int svm_check_intercept(struct kvm_vcpu *vcpu, info->intercept == x86_intercept_clts) break; - if (!(vmcb_is_intercept(&svm->nested.ctl, + if (!(vmcb12_is_intercept(&svm->nested.ctl, INTERCEPT_SELECTIVE_CR0))) break; @@ -4271,6 +4263,8 @@ out: static void svm_handle_exit_irqoff(struct kvm_vcpu *vcpu) { + if (to_svm(vcpu)->vmcb->control.exit_code == SVM_EXIT_INTR) + vcpu->arch.at_instruction_boundary = true; } static void svm_sched_in(struct kvm_vcpu *vcpu, int cpu) @@ -4302,65 +4296,123 @@ static int svm_smi_allowed(struct kvm_vcpu *vcpu, bool for_injection) if (svm->nested.nested_run_pending) return -EBUSY; + if (svm_smi_blocked(vcpu)) + return 0; + /* An SMI must not be injected into L2 if it's supposed to VM-Exit. */ if (for_injection && is_guest_mode(vcpu) && nested_exit_on_smi(svm)) return -EBUSY; - return !svm_smi_blocked(vcpu); + return 1; } -static int svm_pre_enter_smm(struct kvm_vcpu *vcpu, char *smstate) +static int svm_enter_smm(struct kvm_vcpu *vcpu, char *smstate) { struct vcpu_svm *svm = to_svm(vcpu); + struct kvm_host_map map_save; int ret; - if (is_guest_mode(vcpu)) { - /* FED8h - SVM Guest */ - put_smstate(u64, smstate, 0x7ed8, 1); - /* FEE0h - SVM Guest VMCB Physical Address */ - put_smstate(u64, smstate, 0x7ee0, svm->nested.vmcb12_gpa); + if (!is_guest_mode(vcpu)) + return 0; - svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; - svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; - svm->vmcb->save.rip = vcpu->arch.regs[VCPU_REGS_RIP]; + /* FED8h - SVM Guest */ + put_smstate(u64, smstate, 0x7ed8, 1); + /* FEE0h - SVM Guest VMCB Physical Address */ + put_smstate(u64, smstate, 0x7ee0, svm->nested.vmcb12_gpa); - ret = nested_svm_vmexit(svm); - if (ret) - return ret; - } + svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; + svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; + svm->vmcb->save.rip = vcpu->arch.regs[VCPU_REGS_RIP]; + + ret = nested_svm_simple_vmexit(svm, SVM_EXIT_SW); + if (ret) + return ret; + + /* + * KVM uses VMCB01 to store L1 host state while L2 runs but + * VMCB01 is going to be used during SMM and thus the state will + * be lost. Temporary save non-VMLOAD/VMSAVE state to the host save + * area pointed to by MSR_VM_HSAVE_PA. APM guarantees that the + * format of the area is identical to guest save area offsetted + * by 0x400 (matches the offset of 'struct vmcb_save_area' + * within 'struct vmcb'). Note: HSAVE area may also be used by + * L1 hypervisor to save additional host context (e.g. KVM does + * that, see svm_prepare_switch_to_guest()) which must be + * preserved. + */ + if (kvm_vcpu_map(vcpu, gpa_to_gfn(svm->nested.hsave_msr), + &map_save) == -EINVAL) + return 1; + + BUILD_BUG_ON(offsetof(struct vmcb, save) != 0x400); + + svm_copy_vmrun_state(map_save.hva + 0x400, + &svm->vmcb01.ptr->save); + + kvm_vcpu_unmap(vcpu, &map_save, true); return 0; } -static int svm_pre_leave_smm(struct kvm_vcpu *vcpu, const char *smstate) +static int svm_leave_smm(struct kvm_vcpu *vcpu, const char *smstate) { struct vcpu_svm *svm = to_svm(vcpu); - struct kvm_host_map map; - int ret = 0; + struct kvm_host_map map, map_save; + u64 saved_efer, vmcb12_gpa; + struct vmcb *vmcb12; + int ret; + + if (!guest_cpuid_has(vcpu, X86_FEATURE_LM)) + return 0; - if (guest_cpuid_has(vcpu, X86_FEATURE_LM)) { - u64 saved_efer = GET_SMSTATE(u64, smstate, 0x7ed0); - u64 guest = GET_SMSTATE(u64, smstate, 0x7ed8); - u64 vmcb12_gpa = GET_SMSTATE(u64, smstate, 0x7ee0); + /* Non-zero if SMI arrived while vCPU was in guest mode. */ + if (!GET_SMSTATE(u64, smstate, 0x7ed8)) + return 0; - if (guest) { - if (!guest_cpuid_has(vcpu, X86_FEATURE_SVM)) - return 1; + if (!guest_cpuid_has(vcpu, X86_FEATURE_SVM)) + return 1; - if (!(saved_efer & EFER_SVME)) - return 1; + saved_efer = GET_SMSTATE(u64, smstate, 0x7ed0); + if (!(saved_efer & EFER_SVME)) + return 1; - if (kvm_vcpu_map(&svm->vcpu, - gpa_to_gfn(vmcb12_gpa), &map) == -EINVAL) - return 1; + vmcb12_gpa = GET_SMSTATE(u64, smstate, 0x7ee0); + if (kvm_vcpu_map(vcpu, gpa_to_gfn(vmcb12_gpa), &map) == -EINVAL) + return 1; - if (svm_allocate_nested(svm)) - return 1; + ret = 1; + if (kvm_vcpu_map(vcpu, gpa_to_gfn(svm->nested.hsave_msr), &map_save) == -EINVAL) + goto unmap_map; - ret = enter_svm_guest_mode(svm, vmcb12_gpa, map.hva); - kvm_vcpu_unmap(&svm->vcpu, &map, true); - } - } + if (svm_allocate_nested(svm)) + goto unmap_save; + + /* + * Restore L1 host state from L1 HSAVE area as VMCB01 was + * used during SMM (see svm_enter_smm()) + */ + + svm_copy_vmrun_state(&svm->vmcb01.ptr->save, map_save.hva + 0x400); + /* + * Enter the nested guest now + */ + + vmcb_mark_all_dirty(svm->vmcb01.ptr); + + vmcb12 = map.hva; + nested_copy_vmcb_control_to_cache(svm, &vmcb12->control); + nested_copy_vmcb_save_to_cache(svm, &vmcb12->save); + ret = enter_svm_guest_mode(vcpu, vmcb12_gpa, vmcb12, false); + + if (ret) + goto unmap_save; + + svm->nested.nested_run_pending = 1; + +unmap_save: + kvm_vcpu_unmap(vcpu, &map_save, true); +unmap_map: + kvm_vcpu_unmap(vcpu, &map, true); return ret; } @@ -4369,7 +4421,7 @@ static void svm_enable_smi_window(struct kvm_vcpu *vcpu) struct vcpu_svm *svm = to_svm(vcpu); if (!gif_set(svm)) { - if (vgif_enabled(svm)) + if (vgif) svm_set_intercept(svm, INTERCEPT_STGI); /* STGI will cause a vm exit */ } else { @@ -4377,79 +4429,140 @@ static void svm_enable_smi_window(struct kvm_vcpu *vcpu) } } -static bool svm_can_emulate_instruction(struct kvm_vcpu *vcpu, void *insn, int insn_len) +static bool svm_can_emulate_instruction(struct kvm_vcpu *vcpu, int emul_type, + void *insn, int insn_len) { bool smep, smap, is_user; unsigned long cr4; + u64 error_code; + + /* Emulation is always possible when KVM has access to all guest state. */ + if (!sev_guest(vcpu->kvm)) + return true; + + /* #UD and #GP should never be intercepted for SEV guests. */ + WARN_ON_ONCE(emul_type & (EMULTYPE_TRAP_UD | + EMULTYPE_TRAP_UD_FORCED | + EMULTYPE_VMWARE_GP)); /* - * When the guest is an SEV-ES guest, emulation is not possible. + * Emulation is impossible for SEV-ES guests as KVM doesn't have access + * to guest register state. */ if (sev_es_guest(vcpu->kvm)) return false; /* + * Emulation is possible if the instruction is already decoded, e.g. + * when completing I/O after returning from userspace. + */ + if (emul_type & EMULTYPE_NO_DECODE) + return true; + + /* + * Emulation is possible for SEV guests if and only if a prefilled + * buffer containing the bytes of the intercepted instruction is + * available. SEV guest memory is encrypted with a guest specific key + * and cannot be decrypted by KVM, i.e. KVM would read cyphertext and + * decode garbage. + * + * Inject #UD if KVM reached this point without an instruction buffer. + * In practice, this path should never be hit by a well-behaved guest, + * e.g. KVM doesn't intercept #UD or #GP for SEV guests, but this path + * is still theoretically reachable, e.g. via unaccelerated fault-like + * AVIC access, and needs to be handled by KVM to avoid putting the + * guest into an infinite loop. Injecting #UD is somewhat arbitrary, + * but its the least awful option given lack of insight into the guest. + */ + if (unlikely(!insn)) { + kvm_queue_exception(vcpu, UD_VECTOR); + return false; + } + + /* + * Emulate for SEV guests if the insn buffer is not empty. The buffer + * will be empty if the DecodeAssist microcode cannot fetch bytes for + * the faulting instruction because the code fetch itself faulted, e.g. + * the guest attempted to fetch from emulated MMIO or a guest page + * table used to translate CS:RIP resides in emulated MMIO. + */ + if (likely(insn_len)) + return true; + + /* * Detect and workaround Errata 1096 Fam_17h_00_0Fh. * * Errata: - * When CPU raise #NPF on guest data access and vCPU CR4.SMAP=1, it is - * possible that CPU microcode implementing DecodeAssist will fail - * to read bytes of instruction which caused #NPF. In this case, - * GuestIntrBytes field of the VMCB on a VMEXIT will incorrectly - * return 0 instead of the correct guest instruction bytes. - * - * This happens because CPU microcode reading instruction bytes - * uses a special opcode which attempts to read data using CPL=0 - * priviledges. The microcode reads CS:RIP and if it hits a SMAP - * fault, it gives up and returns no instruction bytes. + * When CPU raises #NPF on guest data access and vCPU CR4.SMAP=1, it is + * possible that CPU microcode implementing DecodeAssist will fail to + * read guest memory at CS:RIP and vmcb.GuestIntrBytes will incorrectly + * be '0'. This happens because microcode reads CS:RIP using a _data_ + * loap uop with CPL=0 privileges. If the load hits a SMAP #PF, ucode + * gives up and does not fill the instruction bytes buffer. * - * Detection: - * We reach here in case CPU supports DecodeAssist, raised #NPF and - * returned 0 in GuestIntrBytes field of the VMCB. - * First, errata can only be triggered in case vCPU CR4.SMAP=1. - * Second, if vCPU CR4.SMEP=1, errata could only be triggered - * in case vCPU CPL==3 (Because otherwise guest would have triggered - * a SMEP fault instead of #NPF). - * Otherwise, vCPU CR4.SMEP=0, errata could be triggered by any vCPU CPL. - * As most guests enable SMAP if they have also enabled SMEP, use above - * logic in order to attempt minimize false-positive of detecting errata - * while still preserving all cases semantic correctness. + * As above, KVM reaches this point iff the VM is an SEV guest, the CPU + * supports DecodeAssist, a #NPF was raised, KVM's page fault handler + * triggered emulation (e.g. for MMIO), and the CPU returned 0 in the + * GuestIntrBytes field of the VMCB. * - * Workaround: - * To determine what instruction the guest was executing, the hypervisor - * will have to decode the instruction at the instruction pointer. + * This does _not_ mean that the erratum has been encountered, as the + * DecodeAssist will also fail if the load for CS:RIP hits a legitimate + * #PF, e.g. if the guest attempt to execute from emulated MMIO and + * encountered a reserved/not-present #PF. * - * In non SEV guest, hypervisor will be able to read the guest - * memory to decode the instruction pointer when insn_len is zero - * so we return true to indicate that decoding is possible. + * To hit the erratum, the following conditions must be true: + * 1. CR4.SMAP=1 (obviously). + * 2. CR4.SMEP=0 || CPL=3. If SMEP=1 and CPL<3, the erratum cannot + * have been hit as the guest would have encountered a SMEP + * violation #PF, not a #NPF. + * 3. The #NPF is not due to a code fetch, in which case failure to + * retrieve the instruction bytes is legitimate (see abvoe). * - * But in the SEV guest, the guest memory is encrypted with the - * guest specific key and hypervisor will not be able to decode the - * instruction pointer so we will not able to workaround it. Lets - * print the error and request to kill the guest. + * In addition, don't apply the erratum workaround if the #NPF occurred + * while translating guest page tables (see below). */ - if (likely(!insn || insn_len)) - return true; - - /* - * If RIP is invalid, go ahead with emulation which will cause an - * internal error exit. - */ - if (!kvm_vcpu_gfn_to_memslot(vcpu, kvm_rip_read(vcpu) >> PAGE_SHIFT)) - return true; + error_code = to_svm(vcpu)->vmcb->control.exit_info_1; + if (error_code & (PFERR_GUEST_PAGE_MASK | PFERR_FETCH_MASK)) + goto resume_guest; cr4 = kvm_read_cr4(vcpu); smep = cr4 & X86_CR4_SMEP; smap = cr4 & X86_CR4_SMAP; is_user = svm_get_cpl(vcpu) == 3; if (smap && (!smep || is_user)) { - if (!sev_guest(vcpu->kvm)) - return true; - pr_err_ratelimited("KVM: SEV Guest triggered AMD Erratum 1096\n"); - kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); + + /* + * If the fault occurred in userspace, arbitrarily inject #GP + * to avoid killing the guest and to hopefully avoid confusing + * the guest kernel too much, e.g. injecting #PF would not be + * coherent with respect to the guest's page tables. Request + * triple fault if the fault occurred in the kernel as there's + * no fault that KVM can inject without confusing the guest. + * In practice, the triple fault is moot as no sane SEV kernel + * will execute from user memory while also running with SMAP=1. + */ + if (is_user) + kvm_inject_gp(vcpu, 0); + else + kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); } +resume_guest: + /* + * If the erratum was not hit, simply resume the guest and let it fault + * again. While awful, e.g. the vCPU may get stuck in an infinite loop + * if the fault is at CPL=0, it's the lesser of all evils. Exiting to + * userspace will kill the guest, and letting the emulator read garbage + * will yield random behavior and potentially corrupt the guest. + * + * Simply resuming the guest is technically not a violation of the SEV + * architecture. AMD's APM states that all code fetches and page table + * accesses for SEV guest are encrypted, regardless of the C-Bit. The + * APM also states that encrypted accesses to MMIO are "ignored", but + * doesn't explicitly define "ignored", i.e. doing nothing and letting + * the guest spin is technically "ignoring" the access. + */ return false; } @@ -4487,36 +4600,36 @@ static int svm_vm_init(struct kvm *kvm) if (!pause_filter_count || !pause_filter_thresh) kvm->arch.pause_in_guest = true; - if (avic) { + if (enable_apicv) { int ret = avic_vm_init(kvm); if (ret) return ret; } - kvm_apicv_init(kvm, avic); return 0; } static struct kvm_x86_ops svm_x86_ops __initdata = { - .hardware_unsetup = svm_hardware_teardown, + .name = "kvm_amd", + + .hardware_unsetup = svm_hardware_unsetup, .hardware_enable = svm_hardware_enable, .hardware_disable = svm_hardware_disable, - .cpu_has_accelerated_tpr = svm_cpu_has_accelerated_tpr, .has_emulated_msr = svm_has_emulated_msr, - .vcpu_create = svm_create_vcpu, - .vcpu_free = svm_free_vcpu, + .vcpu_create = svm_vcpu_create, + .vcpu_free = svm_vcpu_free, .vcpu_reset = svm_vcpu_reset, .vm_size = sizeof(struct kvm_svm), .vm_init = svm_vm_init, .vm_destroy = svm_vm_destroy, - .prepare_guest_switch = svm_prepare_guest_switch, + .prepare_switch_to_guest = svm_prepare_switch_to_guest, .vcpu_load = svm_vcpu_load, .vcpu_put = svm_vcpu_put, - .vcpu_blocking = svm_vcpu_blocking, - .vcpu_unblocking = svm_vcpu_unblocking, + .vcpu_blocking = avic_vcpu_blocking, + .vcpu_unblocking = avic_vcpu_unblocking, .update_exception_bitmap = svm_update_exception_bitmap, .get_msr_feature = svm_get_msr_feature, @@ -4526,8 +4639,9 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .get_segment = svm_get_segment, .set_segment = svm_set_segment, .get_cpl = svm_get_cpl, - .get_cs_db_l_bits = kvm_get_cs_db_l_bits, + .get_cs_db_l_bits = svm_get_cs_db_l_bits, .set_cr0 = svm_set_cr0, + .post_set_cr3 = sev_post_set_cr3, .is_valid_cr4 = svm_is_valid_cr4, .set_cr4 = svm_set_cr4, .set_efer = svm_set_efer, @@ -4540,21 +4654,23 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .cache_reg = svm_cache_reg, .get_rflags = svm_get_rflags, .set_rflags = svm_set_rflags, + .get_if_flag = svm_get_if_flag, - .tlb_flush_all = svm_flush_tlb, - .tlb_flush_current = svm_flush_tlb, - .tlb_flush_gva = svm_flush_tlb_gva, - .tlb_flush_guest = svm_flush_tlb, + .flush_tlb_all = svm_flush_tlb_current, + .flush_tlb_current = svm_flush_tlb_current, + .flush_tlb_gva = svm_flush_tlb_gva, + .flush_tlb_guest = svm_flush_tlb_current, - .run = svm_vcpu_run, - .handle_exit = handle_exit, - .skip_emulated_instruction = skip_emulated_instruction, + .vcpu_pre_run = svm_vcpu_pre_run, + .vcpu_run = svm_vcpu_run, + .handle_exit = svm_handle_exit, + .skip_emulated_instruction = svm_skip_emulated_instruction, .update_emulated_instruction = NULL, .set_interrupt_shadow = svm_set_interrupt_shadow, .get_interrupt_shadow = svm_get_interrupt_shadow, .patch_hypercall = svm_patch_hypercall, - .set_irq = svm_set_irq, - .set_nmi = svm_inject_nmi, + .inject_irq = svm_inject_irq, + .inject_nmi = svm_inject_nmi, .queue_exception = svm_queue_exception, .cancel_injection = svm_cancel_injection, .interrupt_allowed = svm_interrupt_allowed, @@ -4564,27 +4680,21 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .enable_nmi_window = svm_enable_nmi_window, .enable_irq_window = svm_enable_irq_window, .update_cr8_intercept = svm_update_cr8_intercept, - .set_virtual_apic_mode = svm_set_virtual_apic_mode, - .refresh_apicv_exec_ctrl = svm_refresh_apicv_exec_ctrl, - .check_apicv_inhibit_reasons = svm_check_apicv_inhibit_reasons, - .pre_update_apicv_exec_ctrl = svm_pre_update_apicv_exec_ctrl, - .load_eoi_exitmap = svm_load_eoi_exitmap, - .hwapic_irr_update = svm_hwapic_irr_update, - .hwapic_isr_update = svm_hwapic_isr_update, - .sync_pir_to_irr = kvm_lapic_find_highest_irr, - .apicv_post_state_restore = avic_post_state_restore, - - .set_tss_addr = svm_set_tss_addr, - .set_identity_map_addr = svm_set_identity_map_addr, - .get_mt_mask = svm_get_mt_mask, + .refresh_apicv_exec_ctrl = avic_refresh_apicv_exec_ctrl, + .check_apicv_inhibit_reasons = avic_check_apicv_inhibit_reasons, + .apicv_post_state_restore = avic_apicv_post_state_restore, + .get_mt_mask = svm_get_mt_mask, .get_exit_info = svm_get_exit_info, .vcpu_after_set_cpuid = svm_vcpu_after_set_cpuid, .has_wbinvd_exit = svm_has_wbinvd_exit, - .write_l1_tsc_offset = svm_write_l1_tsc_offset, + .get_l2_tsc_offset = svm_get_l2_tsc_offset, + .get_l2_tsc_multiplier = svm_get_l2_tsc_multiplier, + .write_tsc_offset = svm_write_tsc_offset, + .write_tsc_multiplier = svm_write_tsc_multiplier, .load_mmu_pgd = svm_load_mmu_pgd, @@ -4595,22 +4705,24 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .sched_in = svm_sched_in, - .pmu_ops = &amd_pmu_ops, .nested_ops = &svm_nested_ops, - .deliver_posted_interrupt = svm_deliver_avic_intr, - .dy_apicv_has_pending_interrupt = svm_dy_apicv_has_pending_interrupt, - .update_pi_irte = svm_update_pi_irte, + .deliver_interrupt = svm_deliver_interrupt, + .pi_update_irte = avic_pi_update_irte, .setup_mce = svm_setup_mce, .smi_allowed = svm_smi_allowed, - .pre_enter_smm = svm_pre_enter_smm, - .pre_leave_smm = svm_pre_leave_smm, + .enter_smm = svm_enter_smm, + .leave_smm = svm_leave_smm, .enable_smi_window = svm_enable_smi_window, - .mem_enc_op = svm_mem_enc_op, - .mem_enc_reg_region = svm_register_enc_region, - .mem_enc_unreg_region = svm_unregister_enc_region, + .mem_enc_ioctl = sev_mem_enc_ioctl, + .mem_enc_register_region = sev_mem_enc_register_region, + .mem_enc_unregister_region = sev_mem_enc_unregister_region, + .guest_memory_reclaimed = sev_guest_memory_reclaimed, + + .vm_copy_enc_context_from = sev_vm_copy_enc_context_from, + .vm_move_enc_context_from = sev_vm_move_enc_context_from, .can_emulate_instruction = svm_can_emulate_instruction, @@ -4620,8 +4732,269 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .complete_emulated_msr = svm_complete_emulated_msr, .vcpu_deliver_sipi_vector = svm_vcpu_deliver_sipi_vector, + .vcpu_get_apicv_inhibit_reasons = avic_vcpu_get_apicv_inhibit_reasons, }; +/* + * The default MMIO mask is a single bit (excluding the present bit), + * which could conflict with the memory encryption bit. Check for + * memory encryption support and override the default MMIO mask if + * memory encryption is enabled. + */ +static __init void svm_adjust_mmio_mask(void) +{ + unsigned int enc_bit, mask_bit; + u64 msr, mask; + + /* If there is no memory encryption support, use existing mask */ + if (cpuid_eax(0x80000000) < 0x8000001f) + return; + + /* If memory encryption is not enabled, use existing mask */ + rdmsrl(MSR_AMD64_SYSCFG, msr); + if (!(msr & MSR_AMD64_SYSCFG_MEM_ENCRYPT)) + return; + + enc_bit = cpuid_ebx(0x8000001f) & 0x3f; + mask_bit = boot_cpu_data.x86_phys_bits; + + /* Increment the mask bit if it is the same as the encryption bit */ + if (enc_bit == mask_bit) + mask_bit++; + + /* + * If the mask bit location is below 52, then some bits above the + * physical addressing limit will always be reserved, so use the + * rsvd_bits() function to generate the mask. This mask, along with + * the present bit, will be used to generate a page fault with + * PFER.RSV = 1. + * + * If the mask bit location is 52 (or above), then clear the mask. + */ + mask = (mask_bit < 52) ? rsvd_bits(mask_bit, 51) | PT_PRESENT_MASK : 0; + + kvm_mmu_set_mmio_spte_mask(mask, mask, PT_WRITABLE_MASK | PT_USER_MASK); +} + +static __init void svm_set_cpu_caps(void) +{ + kvm_set_cpu_caps(); + + supported_xss = 0; + + /* CPUID 0x80000001 and 0x8000000A (SVM features) */ + if (nested) { + kvm_cpu_cap_set(X86_FEATURE_SVM); + kvm_cpu_cap_set(X86_FEATURE_VMCBCLEAN); + + if (nrips) + kvm_cpu_cap_set(X86_FEATURE_NRIPS); + + if (npt_enabled) + kvm_cpu_cap_set(X86_FEATURE_NPT); + + if (tsc_scaling) + kvm_cpu_cap_set(X86_FEATURE_TSCRATEMSR); + + if (vls) + kvm_cpu_cap_set(X86_FEATURE_V_VMSAVE_VMLOAD); + if (lbrv) + kvm_cpu_cap_set(X86_FEATURE_LBRV); + + if (boot_cpu_has(X86_FEATURE_PAUSEFILTER)) + kvm_cpu_cap_set(X86_FEATURE_PAUSEFILTER); + + if (boot_cpu_has(X86_FEATURE_PFTHRESHOLD)) + kvm_cpu_cap_set(X86_FEATURE_PFTHRESHOLD); + + if (vgif) + kvm_cpu_cap_set(X86_FEATURE_VGIF); + + /* Nested VM can receive #VMEXIT instead of triggering #GP */ + kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK); + } + + /* CPUID 0x80000008 */ + if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD) || + boot_cpu_has(X86_FEATURE_AMD_SSBD)) + kvm_cpu_cap_set(X86_FEATURE_VIRT_SSBD); + + /* AMD PMU PERFCTR_CORE CPUID */ + if (enable_pmu && boot_cpu_has(X86_FEATURE_PERFCTR_CORE)) + kvm_cpu_cap_set(X86_FEATURE_PERFCTR_CORE); + + /* CPUID 0x8000001F (SME/SEV features) */ + sev_set_cpu_caps(); +} + +static __init int svm_hardware_setup(void) +{ + int cpu; + struct page *iopm_pages; + void *iopm_va; + int r; + unsigned int order = get_order(IOPM_SIZE); + + /* + * NX is required for shadow paging and for NPT if the NX huge pages + * mitigation is enabled. + */ + if (!boot_cpu_has(X86_FEATURE_NX)) { + pr_err_ratelimited("NX (Execute Disable) not supported\n"); + return -EOPNOTSUPP; + } + kvm_enable_efer_bits(EFER_NX); + + iopm_pages = alloc_pages(GFP_KERNEL, order); + + if (!iopm_pages) + return -ENOMEM; + + iopm_va = page_address(iopm_pages); + memset(iopm_va, 0xff, PAGE_SIZE * (1 << order)); + iopm_base = page_to_pfn(iopm_pages) << PAGE_SHIFT; + + init_msrpm_offsets(); + + supported_xcr0 &= ~(XFEATURE_MASK_BNDREGS | XFEATURE_MASK_BNDCSR); + + if (boot_cpu_has(X86_FEATURE_FXSR_OPT)) + kvm_enable_efer_bits(EFER_FFXSR); + + if (tsc_scaling) { + if (!boot_cpu_has(X86_FEATURE_TSCRATEMSR)) { + tsc_scaling = false; + } else { + pr_info("TSC scaling supported\n"); + kvm_has_tsc_control = true; + } + } + kvm_max_tsc_scaling_ratio = SVM_TSC_RATIO_MAX; + kvm_tsc_scaling_ratio_frac_bits = 32; + + tsc_aux_uret_slot = kvm_add_user_return_msr(MSR_TSC_AUX); + + /* Check for pause filtering support */ + if (!boot_cpu_has(X86_FEATURE_PAUSEFILTER)) { + pause_filter_count = 0; + pause_filter_thresh = 0; + } else if (!boot_cpu_has(X86_FEATURE_PFTHRESHOLD)) { + pause_filter_thresh = 0; + } + + if (nested) { + printk(KERN_INFO "kvm: Nested Virtualization enabled\n"); + kvm_enable_efer_bits(EFER_SVME | EFER_LMSLE); + } + + /* + * KVM's MMU doesn't support using 2-level paging for itself, and thus + * NPT isn't supported if the host is using 2-level paging since host + * CR4 is unchanged on VMRUN. + */ + if (!IS_ENABLED(CONFIG_X86_64) && !IS_ENABLED(CONFIG_X86_PAE)) + npt_enabled = false; + + if (!boot_cpu_has(X86_FEATURE_NPT)) + npt_enabled = false; + + /* Force VM NPT level equal to the host's paging level */ + kvm_configure_mmu(npt_enabled, get_npt_level(), + get_npt_level(), PG_LEVEL_1G); + pr_info("kvm: Nested Paging %sabled\n", npt_enabled ? "en" : "dis"); + + /* Setup shadow_me_value and shadow_me_mask */ + kvm_mmu_set_me_spte_mask(sme_me_mask, sme_me_mask); + + /* Note, SEV setup consumes npt_enabled. */ + sev_hardware_setup(); + + svm_hv_hardware_setup(); + + svm_adjust_mmio_mask(); + + for_each_possible_cpu(cpu) { + r = svm_cpu_init(cpu); + if (r) + goto err; + } + + if (nrips) { + if (!boot_cpu_has(X86_FEATURE_NRIPS)) + nrips = false; + } + + enable_apicv = avic = avic && npt_enabled && (boot_cpu_has(X86_FEATURE_AVIC) || force_avic); + + if (enable_apicv) { + if (!boot_cpu_has(X86_FEATURE_AVIC)) { + pr_warn("AVIC is not supported in CPUID but force enabled"); + pr_warn("Your system might crash and burn"); + } else + pr_info("AVIC enabled\n"); + + amd_iommu_register_ga_log_notifier(&avic_ga_log_notifier); + } else { + svm_x86_ops.vcpu_blocking = NULL; + svm_x86_ops.vcpu_unblocking = NULL; + svm_x86_ops.vcpu_get_apicv_inhibit_reasons = NULL; + } + + if (vls) { + if (!npt_enabled || + !boot_cpu_has(X86_FEATURE_V_VMSAVE_VMLOAD) || + !IS_ENABLED(CONFIG_X86_64)) { + vls = false; + } else { + pr_info("Virtual VMLOAD VMSAVE supported\n"); + } + } + + if (boot_cpu_has(X86_FEATURE_SVME_ADDR_CHK)) + svm_gp_erratum_intercept = false; + + if (vgif) { + if (!boot_cpu_has(X86_FEATURE_VGIF)) + vgif = false; + else + pr_info("Virtual GIF supported\n"); + } + + if (lbrv) { + if (!boot_cpu_has(X86_FEATURE_LBRV)) + lbrv = false; + else + pr_info("LBR virtualization supported\n"); + } + + if (!enable_pmu) + pr_info("PMU virtualization is disabled\n"); + + svm_set_cpu_caps(); + + /* + * It seems that on AMD processors PTE's accessed bit is + * being set by the CPU hardware before the NPF vmexit. + * This is not expected behaviour and our tests fail because + * of it. + * A workaround here is to disable support for + * GUEST_MAXPHYADDR < HOST_MAXPHYADDR if NPT is enabled. + * In this case userspace can know if there is support using + * KVM_CAP_SMALLER_MAXPHYADDR extension and decide how to handle + * it + * If future AMD CPU models change the behaviour described above, + * this variable can be changed accordingly + */ + allow_smaller_maxphyaddr = !npt_enabled; + + return 0; + +err: + svm_hardware_unsetup(); + return r; +} + + static struct kvm_x86_init_ops svm_init_ops __initdata = { .cpu_has_kvm_support = has_svm, .disabled_by_bios = is_disabled, @@ -4629,6 +5002,7 @@ static struct kvm_x86_init_ops svm_init_ops __initdata = { .check_processor_compatibility = svm_check_processor_compat, .runtime_ops = &svm_x86_ops, + .pmu_ops = &amd_pmu_ops, }; static int __init svm_init(void) |