From: Sean Christopherson sean.j.christopherson@intel.com
commit 5c911beff20aa8639e7a1f28988736c13e03ed54 upstream.
Skip the Indirect Branch Prediction Barrier that is triggered on a VMCS switch when running with spectre_v2_user=on/auto if the switch is between two VMCSes in the same guest, i.e. between vmcs01 and vmcs02. The IBPB is intended to prevent one guest from attacking another, which is unnecessary in the nested case as it's the same guest from KVM's perspective.
This all but eliminates the overhead observed for nested VMX transitions when running with CONFIG_RETPOLINE=y and spectre_v2_user=on/auto, which can be significant, e.g. roughly 3x on current systems.
Reported-by: Alexander Graf graf@amazon.com Cc: KarimAllah Raslan karahmed@amazon.de Cc: stable@vger.kernel.org Fixes: 15d45071523d ("KVM/x86: Add IBPB support") Signed-off-by: Sean Christopherson sean.j.christopherson@intel.com Message-Id: 20200501163117.4655-1-sean.j.christopherson@intel.com [Invert direction of bool argument. - Paolo] Signed-off-by: Paolo Bonzini pbonzini@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
--- arch/x86/kvm/vmx/nested.c | 2 +- arch/x86/kvm/vmx/vmx.c | 18 ++++++++++++++---- arch/x86/kvm/vmx/vmx.h | 3 ++- 3 files changed, 17 insertions(+), 6 deletions(-)
--- a/arch/x86/kvm/vmx/nested.c +++ b/arch/x86/kvm/vmx/nested.c @@ -303,7 +303,7 @@ static void vmx_switch_vmcs(struct kvm_v cpu = get_cpu(); prev = vmx->loaded_vmcs; vmx->loaded_vmcs = vmcs; - vmx_vcpu_load_vmcs(vcpu, cpu); + vmx_vcpu_load_vmcs(vcpu, cpu, prev); vmx_sync_vmcs_host_state(vmx, prev); put_cpu();
--- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -1306,10 +1306,12 @@ after_clear_sn: pi_set_on(pi_desc); }
-void vmx_vcpu_load_vmcs(struct kvm_vcpu *vcpu, int cpu) +void vmx_vcpu_load_vmcs(struct kvm_vcpu *vcpu, int cpu, + struct loaded_vmcs *buddy) { struct vcpu_vmx *vmx = to_vmx(vcpu); bool already_loaded = vmx->loaded_vmcs->cpu == cpu; + struct vmcs *prev;
if (!already_loaded) { loaded_vmcs_clear(vmx->loaded_vmcs); @@ -1328,10 +1330,18 @@ void vmx_vcpu_load_vmcs(struct kvm_vcpu local_irq_enable(); }
- if (per_cpu(current_vmcs, cpu) != vmx->loaded_vmcs->vmcs) { + prev = per_cpu(current_vmcs, cpu); + if (prev != vmx->loaded_vmcs->vmcs) { per_cpu(current_vmcs, cpu) = vmx->loaded_vmcs->vmcs; vmcs_load(vmx->loaded_vmcs->vmcs); - indirect_branch_prediction_barrier(); + + /* + * No indirect branch prediction barrier needed when switching + * the active VMCS within a guest, e.g. on nested VM-Enter. + * The L1 VMM can protect itself with retpolines, IBPB or IBRS. + */ + if (!buddy || WARN_ON_ONCE(buddy->vmcs != prev)) + indirect_branch_prediction_barrier(); }
if (!already_loaded) { @@ -1368,7 +1378,7 @@ void vmx_vcpu_load(struct kvm_vcpu *vcpu { struct vcpu_vmx *vmx = to_vmx(vcpu);
- vmx_vcpu_load_vmcs(vcpu, cpu); + vmx_vcpu_load_vmcs(vcpu, cpu, NULL);
vmx_vcpu_pi_load(vcpu, cpu);
--- a/arch/x86/kvm/vmx/vmx.h +++ b/arch/x86/kvm/vmx/vmx.h @@ -317,7 +317,8 @@ struct kvm_vmx { };
bool nested_vmx_allowed(struct kvm_vcpu *vcpu); -void vmx_vcpu_load_vmcs(struct kvm_vcpu *vcpu, int cpu); +void vmx_vcpu_load_vmcs(struct kvm_vcpu *vcpu, int cpu, + struct loaded_vmcs *buddy); void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu); int allocate_vpid(void); void free_vpid(int vpid);