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
} 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 },
static 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 SEV-ES support */
-int sev_es = IS_ENABLED(CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT);
-module_param(sev_es, int, 0444);
-
bool __read_mostly dump_invalid_vmcb;
module_param(dump_invalid_vmcb, bool, 0644);
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)
* 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 {
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
* 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)skip_emulated_instruction(vcpu);
+ rip = kvm_rip_read(vcpu);
svm->int3_rip = rip + svm->vmcb->save.cs.base;
svm->int3_injected = rip - old_rip;
}
return 0;
}
+ if (pgtable_l5_enabled()) {
+ pr_info("KVM doesn't yet support 5-level paging on AMD SVM\n");
+ return 0;
+ }
+
return 1;
}
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);
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;
__free_page(sd->save_area);
free_cpu_data:
kfree(sd);
- return -ENOMEM;
+ return ret;
}
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;
}
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)
*/
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);
+ kvm_mmu_set_mmio_spte_mask(mask, mask, PT_WRITABLE_MASK | PT_USER_MASK);
}
static void svm_hardware_teardown(void)
{
int cpu;
- if (svm_sev_enabled())
- sev_hardware_teardown();
+ sev_hardware_teardown();
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;
}
if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD) ||
boot_cpu_has(X86_FEATURE_AMD_SSBD))
kvm_cpu_cap_set(X86_FEATURE_VIRT_SSBD);
+
+ /* CPUID 0x8000001F (SME/SEV features) */
+ sev_set_cpu_caps();
}
static __init int svm_hardware_setup(void)
struct page *iopm_pages;
void *iopm_va;
int r;
+ unsigned int order = get_order(IOPM_SIZE);
- iopm_pages = alloc_pages(GFP_KERNEL, IOPM_ALLOC_ORDER);
+ 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 << IOPM_ALLOC_ORDER));
+ memset(iopm_va, 0xff, PAGE_SIZE * (1 << order));
iopm_base = page_to_pfn(iopm_pages) << PAGE_SHIFT;
init_msrpm_offsets();
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;
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
kvm_configure_mmu(npt_enabled, get_max_npt_level(), PG_LEVEL_1G);
pr_info("kvm: Nested Paging %sabled\n", npt_enabled ? "en" : "dis");
+ /* Note, SEV setup consumes npt_enabled. */
+ sev_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;
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;
+ svm->vmcb01.ptr->control.tsc_offset;
+ svm->vmcb01.ptr->control.tsc_offset = offset;
}
trace_kvm_write_tsc_offset(vcpu->vcpu_id,
return svm->vmcb->control.tsc_offset;
}
-static void svm_check_invpcid(struct vcpu_svm *svm)
+/* 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
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 void init_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;
- svm->vcpu.arch.hflags = 0;
+ vcpu->arch.hflags = 0;
svm_set_intercept(svm, INTERCEPT_CR0_READ);
svm_set_intercept(svm, INTERCEPT_CR3_READ);
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);
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);
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);
+ svm_set_cr4(vcpu, 0);
+ svm_set_efer(vcpu, 0);
save->dr6 = 0xffff0ff0;
- kvm_set_rflags(&svm->vcpu, X86_EFLAGS_FIXED);
+ kvm_set_rflags(vcpu, X86_EFLAGS_FIXED);
save->rip = 0x0000fff0;
- svm->vcpu.arch.regs[VCPU_REGS_RIP] = save->rip;
+ 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);
+ svm_set_cr0(vcpu, X86_CR0_NW | X86_CR0_CD | X86_CR0_ET);
+ kvm_mmu_reset_context(vcpu);
save->cr4 = X86_CR4_PAE;
/* rdx = ?? */
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;
+ vcpu->arch.hflags = 0;
- 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;
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);
if (vgif) {
svm_clr_intercept(svm, INTERCEPT_STGI);
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);
}
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;
+ vcpu->arch.apic_base = APIC_DEFAULT_PHYS_BASE |
+ MSR_IA32_APICBASE_ENABLE;
+ if (kvm_vcpu_is_reset_bsp(vcpu))
+ 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);
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)
{
struct vcpu_svm *svm;
- struct page *vmcb_page;
+ struct page *vmcb01_page;
struct page *vmsa_page = NULL;
int err;
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.
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);
if (vmsa_page)
svm->vmsa = page_address(vmsa_page);
- svm->asid_generation = 0;
svm->guest_state_loaded = false;
- init_vmcb(svm);
+
+ svm_switch_vmcb(svm, &svm->vmcb01);
+ init_vmcb(vcpu);
svm_init_osvw(vcpu);
vcpu->arch.microcode_version = 0x01000065;
- if (sev_es_guest(svm->vcpu.kvm))
+ if (sev_es_guest(vcpu->kvm))
/* Perform SEV-ES specific VMCB creation updates */
sev_es_create_vcpu(svm);
if (vmsa_page)
__free_page(vmsa_page);
error_free_vmcb_page:
- __free_page(vmcb_page);
+ __free_page(vmcb01_page);
out:
return err;
}
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)
{
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)) {
+ if (sev_es_guest(vcpu->kvm)) {
sev_es_prepare_guest_switch(svm, vcpu->cpu);
} else {
vmsave(__sme_page_pa(sd->save_area));
}
}
- /* 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 (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)
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();
/* Drop int_ctl fields related to VINTR injection. */
svm->vmcb->control.int_ctl &= mask;
if (is_guest_mode(&svm->vcpu)) {
- svm->nested.hsave->control.int_ctl &= mask;
+ svm->vmcb01.ptr->control.int_ctl &= mask;
WARN_ON((svm->vmcb->control.int_ctl & V_TPR_MASK) !=
(svm->nested.ctl.int_ctl & V_TPR_MASK));
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;
vmcb_mark_dirty(svm->vmcb, VMCB_DT);
}
-static void update_cr0_intercept(struct vcpu_svm *svm)
-{
- ulong gcr0;
- u64 *hcr0;
-
- /*
- * SEV-ES guests must always keep the CR intercepts cleared. CR
- * tracking is done using the CR write traps.
- */
- 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);
- }
-}
-
void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
{
struct vcpu_svm *svm = to_svm(vcpu);
+ u64 hcr0 = cr0;
#ifdef CONFIG_X86_64
if (vcpu->arch.efer & EFER_LME && !vcpu->arch.guest_state_protected) {
vcpu->arch.cr0 = cr0;
if (!npt_enabled)
- cr0 |= X86_CR0_PG | X86_CR0_WP;
+ hcr0 |= X86_CR0_PG | X86_CR0_WP;
/*
* re-enable caching here because the QEMU bios
* 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)
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++;
}
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)
{
+ struct vcpu_svm *svm = to_svm(vcpu);
+
u64 fault_address = __sme_clr(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;
}
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;
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;
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;
}
return true;
}
-static void svm_handle_mce(struct vcpu_svm *svm)
+static void svm_handle_mce(struct kvm_vcpu *vcpu)
{
if (is_erratum_383()) {
/*
*/
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;
}
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;
/*
* so reinitialize it.
*/
clear_page(svm->vmcb);
- init_vmcb(svm);
+ init_vmcb(vcpu);
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;
svm->next_rip = svm->vmcb->control.exit_info_2;
- return kvm_fast_pio(&svm->vcpu, size, port, in);
-}
-
-static int nmi_interception(struct vcpu_svm *svm)
-{
- return 1;
+ return kvm_fast_pio(vcpu, size, port, in);
}
-static int intr_interception(struct vcpu_svm *svm)
+static int nmi_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)
-{
- 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)
+static int vmload_vmsave_interception(struct kvm_vcpu *vcpu, bool vmload)
{
- 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(vcpu);
- ret = kvm_skip_emulated_instruction(&svm->vcpu);
+ if (vmload) {
+ nested_svm_vmloadsave(vmcb12, svm->vmcb);
+ svm->sysenter_eip_hi = 0;
+ svm->sysenter_esp_hi = 0;
+ } else
+ nested_svm_vmloadsave(svm->vmcb, vmcb12);
- nested_svm_vmloadsave(nested_vmcb, svm->vmcb);
- kvm_vcpu_unmap(&svm->vcpu, &map, true);
+ 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 {
[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,
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);
}
/*
* 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;
}
}
-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));
-
- /* Let's treat INVLPGA the same as INVLPG (can be optimized!) */
- kvm_mmu_invlpg(vcpu, kvm_rax_read(&svm->vcpu));
+ gva_t gva = kvm_rax_read(vcpu);
+ u32 asid = kvm_rcx_read(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);
-
- int err = kvm_set_xcr(&svm->vcpu, index, new_bv);
- return kvm_complete_insn_gp(&svm->vcpu, err);
-}
+ trace_kvm_skinit(to_svm(vcpu)->vmcb->save.rip, kvm_rax_read(vcpu));
-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 &
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 &
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;
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 (!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) ||
+ if (!is_guest_mode(vcpu) ||
(!(vmcb_is_intercept(&svm->nested.ctl, INTERCEPT_SELECTIVE_CR0))))
return false;
#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)
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;
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);
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;
*/
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)
switch (msr_info->index) {
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;
/*
!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 &&
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->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,
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);
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;
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:
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;
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)) {
return 0;
}
-static int wrmsr_interception(struct vcpu_svm *svm)
-{
- return kvm_emulate_wrmsr(&svm->vcpu);
-}
-
-static int msr_interception(struct vcpu_svm *svm)
+static int msr_interception(struct kvm_vcpu *vcpu)
{
- 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.
* In this case AVIC was temporarily disabled for
* requesting the IRQ window and we have to re-enable it.
*/
- svm_toggle_avic_for_irq_window(&svm->vcpu, true);
+ svm_toggle_avic_for_irq_window(vcpu, true);
- ++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;
/*
* 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);
kvm_vcpu_on_spin(vcpu, in_kernel);
- return 1;
-}
-
-static int nop_interception(struct vcpu_svm *svm)
-{
- return kvm_skip_emulated_instruction(&(svm->vcpu));
-}
-
-static int monitor_interception(struct vcpu_svm *svm)
-{
- printk_once(KERN_WARNING "kvm: MONITOR instruction emulated as NOP!\n");
- return nop_interception(svm);
+ return kvm_skip_emulated_instruction(vcpu);
}
-static int mwait_interception(struct vcpu_svm *svm)
+static int invpcid_interception(struct kvm_vcpu *vcpu)
{
- 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;
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,
[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] = kvm_emulate_as_nop,
+ [SVM_EXIT_INIT] = kvm_emulate_as_nop,
[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,
[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,
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");
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);
+ save01->tr.selector, save01->tr.attrib,
+ save01->tr.limit, save01->tr.base);
pr_err("cpl: %d efer: %016llx\n",
save->cpl, save->efer);
pr_err("%-15s %016llx %-13s %016llx\n",
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",
return -EINVAL;
}
-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))
+ if (svm_handle_invalid_exit(vcpu, exit_code))
return 0;
#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,
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)
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);
}
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;
}
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;
}
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)
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);
}
}
if (!gif_set(svm))
return true;
- if (sev_es_guest(svm->vcpu.kvm)) {
+ if (sev_es_guest(vcpu->kvm)) {
/*
* SEV-ES guests to not expose RFLAGS. Use the VMCB interrupt mask
* bit to determine the state of the IF flag.
} else 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;
{
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 (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)
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;
* 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:
/*
*/
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;
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 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);
+ kvm_guest_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();
+ kvm_guest_exit_irqoff();
}
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);
* 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);
* 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
* 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;
}
if (unlikely(svm->vmcb->control.exit_code == SVM_EXIT_NMI))
- kvm_before_interrupt(&svm->vcpu);
+ kvm_before_interrupt(vcpu);
kvm_load_host_xsave_state(vcpu);
stgi();
/* 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);
svm->nested.nested_run_pending = 0;
}
/* 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) {
*/
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;
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;
cr3 = vcpu->arch.cr3;
+ } else if (vcpu->arch.mmu->shadow_root_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;
/* 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);
- /* Check again if INVPCID interception if required */
- svm_check_invpcid(svm);
+ svm_recalc_instruction_intercepts(vcpu, svm);
/* For sev guests, the memory encryption bit is not reserved in CR3. */
if (sev_guest(vcpu->kvm)) {
vcpu->arch.reserved_gpa_bits &= ~(1UL << (best->ebx & 0x3f));
}
- if (!kvm_vcpu_apicv_active(vcpu))
- return;
+ 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_request_apicv_update(vcpu->kvm, false,
+ APICV_INHIBIT_REASON_X2APIC);
- /*
- * 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);
+ }
- /*
- * 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 (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);
+ } 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 bool svm_has_wbinvd_exit(void)
if (!(saved_efer & EFER_SVME))
return 1;
- if (kvm_vcpu_map(&svm->vcpu,
+ if (kvm_vcpu_map(vcpu,
gpa_to_gfn(vmcb12_gpa), &map) == -EINVAL)
return 1;
if (svm_allocate_nested(svm))
return 1;
- ret = enter_svm_guest_mode(svm, vmcb12_gpa, map.hva);
- kvm_vcpu_unmap(&svm->vcpu, &map, true);
+ ret = enter_svm_guest_mode(vcpu, vmcb12_gpa, map.hva);
+ kvm_vcpu_unmap(vcpu, &map, true);
}
}
.mem_enc_reg_region = svm_register_enc_region,
.mem_enc_unreg_region = svm_unregister_enc_region,
+ .vm_copy_enc_context_from = svm_vm_copy_asid_from,
+
.can_emulate_instruction = svm_can_emulate_instruction,
.apic_init_signal_blocked = svm_apic_init_signal_blocked,
projects / linux-2.6-microblaze.git / blobdiff