@@ -3561,6 +3561,52 @@ Returns: 0 on success,
-ENOENT on deassign if the conn_id isn't registered
-EEXIST on assign if the conn_id is already registered
+4.114 KVM_GET_NESTED_STATE
+
+Capability: KVM_CAP_NESTED_STATE
+Architectures: x86
+Type: vcpu ioctl
+Parameters: struct kvm_nested_state (in/out)
+Returns: 0 on success, -1 on error
+Errors:
+ E2BIG: the data size exceeds the value of 'size' specified by
+ the user (the size required will be written into size).
+
+struct kvm_nested_state {
+ __u16 flags;
+ __u16 format;
+ __u32 size;
+ union {
+ struct kvm_vmx_nested_state vmx;
+ struct kvm_svm_nested_state svm;
+ __u8 pad[120];
+ };
+ __u8 data[0];
+};
+
+This ioctl copies the vcpu's kvm_nested_state struct from the kernel to userspace.
+
+4.115 KVM_SET_NESTED_STATE
+
+Capability: KVM_CAP_NESTED_STATE
+Architectures: x86
+Type: vcpu ioctl
+Parameters: struct kvm_nested_state (in)
+Returns: 0 on success, -1 on error
+
+struct kvm_nested_state {
+ __u16 flags;
+ __u16 format;
+ __u32 size;
+ union {
+ struct kvm_vmx_nested_state vmx;
+ struct kvm_svm_nested_state svm;
+ __u8 pad[120];
+ };
+ __u8 data[0];
+};
+
+This copies the vcpu's kvm_nested_state struct from userspace to the kernel.
5. The kvm_run structure
------------------------
@@ -75,6 +75,7 @@
#define KVM_REQ_HV_EXIT KVM_ARCH_REQ(21)
#define KVM_REQ_HV_STIMER KVM_ARCH_REQ(22)
#define KVM_REQ_LOAD_EOI_EXITMAP KVM_ARCH_REQ(23)
+#define KVM_REQ_GET_VMCS12_PAGES KVM_ARCH_REQ(24)
#define CR0_RESERVED_BITS \
(~(unsigned long)(X86_CR0_PE | X86_CR0_MP | X86_CR0_EM | X86_CR0_TS \
@@ -1087,6 +1088,12 @@ struct kvm_x86_ops {
void (*setup_mce)(struct kvm_vcpu *vcpu);
+ int (*get_nested_state)(struct kvm_vcpu *vcpu,
+ struct kvm_nested_state __user *user_kvm_nested_state);
+ int (*set_nested_state)(struct kvm_vcpu *vcpu,
+ struct kvm_nested_state __user *user_kvm_nested_state);
+ void (*get_vmcs12_pages)(struct kvm_vcpu *vcpu);
+
int (*smi_allowed)(struct kvm_vcpu *vcpu);
int (*pre_enter_smm)(struct kvm_vcpu *vcpu, char *smstate);
int (*pre_leave_smm)(struct kvm_vcpu *vcpu, u64 smbase);
@@ -378,4 +378,49 @@ struct kvm_sync_regs {
#define KVM_X86_QUIRK_LINT0_REENABLED (1 << 0)
#define KVM_X86_QUIRK_CD_NW_CLEARED (1 << 1)
+#define KVM_STATE_NESTED_GUEST_MODE 0x00000001
+#define KVM_STATE_NESTED_RUN_PENDING 0x00000002
+#define KVM_STATE_NESTED_GIF 0x00000004
+
+#define KVM_STATE_NESTED_SMM_GUEST_MODE 0x00000001
+#define KVM_STATE_NESTED_SMM_VMXON 0x00000002
+
+struct kvm_vmx_nested_state {
+ __u64 vmxon_pa;
+ __u64 vmcs_pa;
+
+ struct {
+ __u16 flags;
+ } smm;
+};
+
+struct kvm_svm_nested_state {
+ __u64 hsave_pa;
+ __u64 vmcb_pa;
+};
+
+/* for KVM_CAP_STATE */
+struct kvm_nested_state {
+ /* KVM_STATE_* flags */
+ __u16 flags;
+
+ /* 0 for VMX, 1 for SVM. */
+ __u16 format;
+
+ /* 128 for SVM, 128 + VMCS size for VMX. */
+ __u32 size;
+
+ union {
+ /* VMXON, VMCS */
+ struct kvm_vmx_nested_state vmx;
+ /* HSAVE_PA, VMCB */
+ struct kvm_svm_nested_state svm;
+
+ /* Pad the union to 120 bytes. */
+ __u8 pad[120];
+ };
+
+ __u8 data[0];
+};
+
#endif /* _ASM_X86_KVM_H */
@@ -10635,9 +10635,9 @@ static void vmx_inject_page_fault_nested(struct kvm_vcpu *vcpu,
static inline bool nested_vmx_prepare_msr_bitmap(struct kvm_vcpu *vcpu,
struct vmcs12 *vmcs12);
-static void nested_get_vmcs12_pages(struct kvm_vcpu *vcpu,
- struct vmcs12 *vmcs12)
+static void nested_get_vmcs12_pages(struct kvm_vcpu *vcpu)
{
+ struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
struct vcpu_vmx *vmx = to_vmx(vcpu);
struct page *page;
u64 hpa;
@@ -11772,7 +11772,6 @@ static int enter_vmx_non_root_mode(struct kvm_vcpu *vcpu)
if (prepare_vmcs02(vcpu, vmcs12, &exit_qual))
goto fail;
- nested_get_vmcs12_pages(vcpu, vmcs12);
r = EXIT_REASON_MSR_LOAD_FAIL;
msr_entry_idx = nested_vmx_load_msr(vcpu,
@@ -11878,6 +11877,8 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch)
return ret;
}
+ nested_get_vmcs12_pages(vcpu);
+
/*
* If we're entering a halted L2 vcpu and the L2 vcpu won't be woken
* by event injection, halt vcpu.
@@ -12976,6 +12977,197 @@ static int enable_smi_window(struct kvm_vcpu *vcpu)
return 0;
}
+static int get_vmcs_cache(struct kvm_vcpu *vcpu,
+ struct kvm_nested_state __user *user_kvm_nested_state)
+{
+ struct vcpu_vmx *vmx = to_vmx(vcpu);
+ struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
+
+ /*
+ * When running L2, the authoritative vmcs12 state is in the
+ * vmcs02. When running L1, the authoritative vmcs12 state is
+ * in the shadow vmcs linked to vmcs01, unless
+ * sync_shadow_vmcs is set, in which case, the authoritative
+ * vmcs12 state is in the vmcs12 already.
+ */
+ if (is_guest_mode(vcpu))
+ sync_vmcs12(vcpu, vmcs12);
+ else if (enable_shadow_vmcs && !vmx->nested.sync_shadow_vmcs)
+ copy_shadow_to_vmcs12(vmx);
+
+ if (copy_to_user(user_kvm_nested_state->data, vmcs12, sizeof(*vmcs12)))
+ return -EFAULT;
+
+ return 0;
+}
+
+static int vmx_get_nested_state(struct kvm_vcpu *vcpu,
+ struct kvm_nested_state __user *user_kvm_nested_state)
+{
+ u32 user_data_size;
+ struct vcpu_vmx *vmx = to_vmx(vcpu);
+ struct kvm_nested_state kvm_state = {
+ .flags = 0,
+ .format = 0,
+ .size = sizeof(kvm_state),
+ .vmx.vmxon_pa = -1ull,
+ .vmx.vmcs_pa = -1ull,
+ };
+
+ if (copy_from_user(&user_data_size, &user_kvm_nested_state->size,
+ sizeof(user_data_size)))
+ return -EFAULT;
+
+ if (nested_vmx_allowed(vcpu) &&
+ (vmx->nested.vmxon || vmx->nested.smm.vmxon)) {
+ kvm_state.vmx.vmxon_pa = vmx->nested.vmxon_ptr;
+ kvm_state.vmx.vmcs_pa = vmx->nested.current_vmptr;
+
+ if (vmx->nested.current_vmptr != -1ull)
+ kvm_state.size += VMCS12_SIZE;
+
+ if (vmx->nested.smm.vmxon)
+ kvm_state.vmx.smm.flags |= KVM_STATE_NESTED_SMM_VMXON;
+
+ if (vmx->nested.smm.guest_mode)
+ kvm_state.vmx.smm.flags |= KVM_STATE_NESTED_SMM_GUEST_MODE;
+
+ if (is_guest_mode(vcpu)) {
+ kvm_state.flags |= KVM_STATE_NESTED_GUEST_MODE;
+
+ if (vmx->nested.nested_run_pending)
+ kvm_state.flags |= KVM_STATE_NESTED_RUN_PENDING;
+ }
+ }
+
+ if (user_data_size < kvm_state.size) {
+ if (copy_to_user(&user_kvm_nested_state->size, &kvm_state.size,
+ sizeof(kvm_state.size)))
+ return -EFAULT;
+ return -E2BIG;
+ }
+
+ if (copy_to_user(user_kvm_nested_state, &kvm_state, sizeof(kvm_state)))
+ return -EFAULT;
+
+ if (vmx->nested.current_vmptr == -1ull)
+ return 0;
+
+ return get_vmcs_cache(vcpu, user_kvm_nested_state);
+}
+
+static int set_vmcs_cache(struct kvm_vcpu *vcpu,
+ struct kvm_nested_state __user *user_kvm_nested_state,
+ struct kvm_nested_state *kvm_state)
+
+{
+ struct vcpu_vmx *vmx = to_vmx(vcpu);
+ struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
+ u32 exit_qual;
+ int ret;
+
+ if ((kvm_state->size < (sizeof(*vmcs12) + sizeof(*kvm_state))) ||
+ kvm_state->vmx.vmcs_pa == kvm_state->vmx.vmxon_pa ||
+ !page_address_valid(vcpu, kvm_state->vmx.vmcs_pa))
+ return -EINVAL;
+
+ if (copy_from_user(vmcs12, user_kvm_nested_state->data, sizeof(*vmcs12)))
+ return -EFAULT;
+
+ if (vmcs12->revision_id != VMCS12_REVISION)
+ return -EINVAL;
+
+ set_current_vmptr(vmx, kvm_state->vmx.vmcs_pa);
+
+ if ((kvm_state->vmx.smm.flags & KVM_STATE_NESTED_SMM_GUEST_MODE) &&
+ (kvm_state->flags & KVM_STATE_NESTED_GUEST_MODE))
+ return -EINVAL;
+
+ if (vmx->nested.vmxon &&
+ (kvm_state->vmx.smm.flags & KVM_STATE_NESTED_SMM_VMXON))
+ return -EINVAL;
+
+ if (kvm_state->vmx.smm.flags & KVM_STATE_NESTED_SMM_GUEST_MODE)
+ vmx->nested.smm.guest_mode = true;
+
+ if (kvm_state->vmx.smm.flags & KVM_STATE_NESTED_SMM_VMXON) {
+ vmx->nested.smm.vmxon = true;
+ vmx->nested.vmxon = false;
+ }
+
+ if (!(kvm_state->flags & KVM_STATE_NESTED_GUEST_MODE))
+ return 0;
+
+ if (kvm_state->flags & KVM_STATE_NESTED_RUN_PENDING)
+ vmx->nested.nested_run_pending = 1;
+
+ if (check_vmentry_prereqs(vcpu, vmcs12) ||
+ check_vmentry_postreqs(vcpu, vmcs12, &exit_qual))
+ return -EINVAL;
+
+ ret = enter_vmx_non_root_mode(vcpu);
+ if (ret)
+ return ret;
+
+ /*
+ * The MMU is not initialized to point at the right entities yet and
+ * "get pages" would need to read data from the guest (i.e. we will
+ * need to perform gpa to hpa translation). So, This request will
+ * result in a call to nested_get_vmcs12_pages before the next
+ * VM-entry.
+ */
+ kvm_make_request(KVM_REQ_GET_VMCS12_PAGES, vcpu);
+
+ vmx->nested.nested_run_pending = 1;
+
+ return 0;
+}
+
+static int vmx_set_nested_state(struct kvm_vcpu *vcpu,
+ struct kvm_nested_state __user *user_kvm_nested_state)
+{
+ struct vcpu_vmx *vmx = to_vmx(vcpu);
+ struct kvm_nested_state kvm_state;
+ int ret;
+
+ if (copy_from_user(&kvm_state, user_kvm_nested_state, sizeof(kvm_state)))
+ return -EFAULT;
+
+ if (kvm_state.size < sizeof(kvm_state))
+ return -EINVAL;
+
+ if (kvm_state.format != 0)
+ return -EINVAL;
+
+ if (kvm_state.flags &
+ ~(KVM_STATE_NESTED_RUN_PENDING | KVM_STATE_NESTED_GUEST_MODE))
+ return -EINVAL;
+
+ if (!nested_vmx_allowed(vcpu))
+ return kvm_state.vmx.vmxon_pa == -1ull ? 0 : -EINVAL;
+
+ vmx_leave_nested(vcpu);
+
+ vmx->nested.nested_run_pending =
+ !!(kvm_state.flags & KVM_STATE_NESTED_RUN_PENDING);
+
+ if (kvm_state.vmx.vmxon_pa == -1ull)
+ return 0;
+
+ if (!page_address_valid(vcpu, kvm_state.vmx.vmxon_pa))
+ return -EINVAL;
+
+ vmx->nested.vmxon_ptr = kvm_state.vmx.vmxon_pa;
+ ret = enter_vmx_operation(vcpu);
+ if (ret)
+ return ret;
+
+ if (kvm_state.vmx.vmcs_pa == -1ull)
+ return 0;
+
+ return set_vmcs_cache(vcpu, user_kvm_nested_state, &kvm_state);
+}
+
static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
.cpu_has_kvm_support = cpu_has_kvm_support,
.disabled_by_bios = vmx_disabled_by_bios,
@@ -13110,6 +13302,10 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
.setup_mce = vmx_setup_mce,
+ .get_nested_state = vmx_get_nested_state,
+ .set_nested_state = vmx_set_nested_state,
+ .get_vmcs12_pages = nested_get_vmcs12_pages,
+
.smi_allowed = vmx_smi_allowed,
.pre_enter_smm = vmx_pre_enter_smm,
.pre_leave_smm = vmx_pre_leave_smm,
@@ -2943,6 +2943,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
case KVM_CAP_X2APIC_API:
r = KVM_X2APIC_API_VALID_FLAGS;
break;
+ case KVM_CAP_STATE:
+ r = !!kvm_x86_ops->get_nested_state;
+ break;
default:
break;
}
@@ -3961,6 +3964,22 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
r = kvm_vcpu_ioctl_enable_cap(vcpu, &cap);
break;
}
+ case KVM_GET_NESTED_STATE: {
+ struct kvm_nested_state __user *user_kvm_nested_state = argp;
+
+ r = -EINVAL;
+ if (kvm_x86_ops->get_nested_state)
+ r = kvm_x86_ops->get_nested_state(vcpu, user_kvm_nested_state);
+ break;
+ }
+ case KVM_SET_NESTED_STATE: {
+ struct kvm_nested_state __user *user_kvm_nested_state = argp;
+
+ r = -EINVAL;
+ if (kvm_x86_ops->set_nested_state)
+ r = kvm_x86_ops->set_nested_state(vcpu, user_kvm_nested_state);
+ break;
+ }
default:
r = -EINVAL;
}
@@ -7309,6 +7328,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
bool req_immediate_exit = false;
if (kvm_request_pending(vcpu)) {
+ if (kvm_check_request(KVM_REQ_GET_VMCS12_PAGES, vcpu))
+ kvm_x86_ops->get_vmcs12_pages(vcpu);
if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu))
kvm_mmu_unload(vcpu);
if (kvm_check_request(KVM_REQ_MIGRATE_TIMER, vcpu))
@@ -963,6 +963,7 @@ struct kvm_ppc_resize_hpt {
#define KVM_CAP_GET_MSR_FEATURES 153
#define KVM_CAP_HYPERV_EVENTFD 154
#define KVM_CAP_HYPERV_TLBFLUSH 155
+#define KVM_CAP_STATE 156
#ifdef KVM_CAP_IRQ_ROUTING
@@ -1405,6 +1406,9 @@ struct kvm_enc_region {
/* Available with KVM_CAP_HYPERV_EVENTFD */
#define KVM_HYPERV_EVENTFD _IOW(KVMIO, 0xbd, struct kvm_hyperv_eventfd)
+/* Available with KVM_CAP_STATE */
+#define KVM_GET_NESTED_STATE _IOWR(KVMIO, 0xbe, struct kvm_nested_state)
+#define KVM_SET_NESTED_STATE _IOW(KVMIO, 0xbf, struct kvm_nested_state)
/* Secure Encrypted Virtualization command */
enum sev_cmd_id {