diff options
| author | Paolo Bonzini <pbonzini@redhat.com> | 2021-04-22 05:11:15 +0300 |
|---|---|---|
| committer | Paolo Bonzini <pbonzini@redhat.com> | 2021-04-26 12:27:15 +0300 |
| commit | d9db0fd6c5c9fa7c9a462a2c54d5e91455a74fca (patch) | |
| tree | 44ca9d2af91b6486a802b4f32d188f191f6b74e3 /arch/x86/kvm/svm | |
| parent | e8126bdaf19400050a57a2c1662a22786b9426f1 (diff) | |
| download | linux-d9db0fd6c5c9fa7c9a462a2c54d5e91455a74fca.tar.xz | |
KVM: SEV: Mask CPUID[0x8000001F].eax according to supported features
Add a reverse-CPUID entry for the memory encryption word, 0x8000001F.EAX,
and use it to override the supported CPUID flags reported to userspace.
Masking the reported CPUID flags avoids over-reporting KVM support, e.g.
without the mask a SEV-SNP capable CPU may incorrectly advertise SNP
support to userspace.
Clear SEV/SEV-ES if their corresponding module parameters are disabled,
and clear the memory encryption leaf completely if SEV is not fully
supported in KVM. Advertise SME_COHERENT in addition to SEV and SEV-ES,
as the guest can use SME_COHERENT to avoid CLFLUSH operations.
Explicitly omit SME and VM_PAGE_FLUSH from the reporting. These features
are used by KVM, but are not exposed to the guest, e.g. guest access to
related MSRs will fault.
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Brijesh Singh <brijesh.singh@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Co-developed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210422021125.3417167-6-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Diffstat (limited to 'arch/x86/kvm/svm')
| -rw-r--r-- | arch/x86/kvm/svm/sev.c | 8 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/svm.c | 3 | ||||
| -rw-r--r-- | arch/x86/kvm/svm/svm.h | 1 |
3 files changed, 12 insertions, 0 deletions
diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index d7088a1d4de6..055aedec2f64 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -1807,6 +1807,14 @@ void sev_vm_destroy(struct kvm *kvm) sev_asid_free(sev); } +void __init sev_set_cpu_caps(void) +{ + if (!sev) + kvm_cpu_cap_clear(X86_FEATURE_SEV); + if (!sev_es) + kvm_cpu_cap_clear(X86_FEATURE_SEV_ES); +} + void __init sev_hardware_setup(void) { unsigned int eax, ebx, ecx, edx, sev_asid_count, sev_es_asid_count; diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index cbddcdb18d97..eed6f9455e1b 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -923,6 +923,9 @@ static __init void svm_set_cpu_caps(void) 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) diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h index afc3166c9027..81894fd649bd 100644 --- a/arch/x86/kvm/svm/svm.h +++ b/arch/x86/kvm/svm/svm.h @@ -575,6 +575,7 @@ int svm_unregister_enc_region(struct kvm *kvm, struct kvm_enc_region *range); int svm_vm_copy_asid_from(struct kvm *kvm, unsigned int source_fd); void pre_sev_run(struct vcpu_svm *svm, int cpu); +void __init sev_set_cpu_caps(void); void __init sev_hardware_setup(void); void sev_hardware_teardown(void); void sev_free_vcpu(struct kvm_vcpu *vcpu); |