| Message ID | 20240608001003.3296640-1-seanjc@google.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | KVM: VMX: Remove unnecessary INVEPT[GLOBAL] from hardware enable path | expand |
On Sat, Jun 8, 2024 at 2:10 AM Sean Christopherson <seanjc@google.com> wrote: > > Remove the completely pointess global INVEPT, i.e. EPT TLB flush, from > KVM's VMX enablement path. KVM always does a targeted TLB flush when > using a "new" EPT root, in quotes because "new" simply means a root that > isn't currently being used by the vCPU. > > KVM also _deliberately_ runs with stale TLB entries for defunct roots, > i.e. doesn't do a TLB flush when vCPUs stop using roots, precisely because > KVM does the flush on first use. As called out by the comment in > kvm_mmu_load(), the reason KVM flushes on first use is because KVM can't > guarantee the correctness of past hypervisors. > > Jumping back to the global INVEPT, when the painfully terse commit > 1439442c7b25 ("KVM: VMX: Enable EPT feature for KVM") was added, the > effective TLB flush being performed was: > > static void vmx_flush_tlb(struct kvm_vcpu *vcpu) > { > vpid_sync_vcpu_all(to_vmx(vcpu)); > } > > I.e. KVM was not flushing EPT TLB entries when allocating a "new" root, > which very strongly suggests that the global INVEPT during hardware > enabling was a misguided hack that addressed the most obvious symptom, > but failed to fix the underlying bug. Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Paolo > Signed-off-by: Sean Christopherson <seanjc@google.com> > --- > arch/x86/kvm/vmx/vmx.c | 3 --- > 1 file changed, 3 deletions(-) > > diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c > index 0e3aaf520db2..21dbe20f50ba 100644 > --- a/arch/x86/kvm/vmx/vmx.c > +++ b/arch/x86/kvm/vmx/vmx.c > @@ -2832,9 +2832,6 @@ int vmx_hardware_enable(void) > return r; > } > > - if (enable_ept) > - ept_sync_global(); > - > return 0; > } > > > base-commit: af0903ab52ee6d6f0f63af67fa73d5eb00f79b9a > -- > 2.45.2.505.gda0bf45e8d-goog >
On Fri, 07 Jun 2024 17:10:03 -0700, Sean Christopherson wrote: > Remove the completely pointess global INVEPT, i.e. EPT TLB flush, from > KVM's VMX enablement path. KVM always does a targeted TLB flush when > using a "new" EPT root, in quotes because "new" simply means a root that > isn't currently being used by the vCPU. > > KVM also _deliberately_ runs with stale TLB entries for defunct roots, > i.e. doesn't do a TLB flush when vCPUs stop using roots, precisely because > KVM does the flush on first use. As called out by the comment in > kvm_mmu_load(), the reason KVM flushes on first use is because KVM can't > guarantee the correctness of past hypervisors. > > [...] Applied to kvm-x86 vmx, thanks! [1/1] KVM: VMX: Remove unnecessary INVEPT[GLOBAL] from hardware enable path https://github.com/kvm-x86/linux/commit/23b2c5088d01 -- https://github.com/kvm-x86/linux/tree/next
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index 0e3aaf520db2..21dbe20f50ba 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -2832,9 +2832,6 @@ int vmx_hardware_enable(void) return r; } - if (enable_ept) - ept_sync_global(); - return 0; }
Remove the completely pointess global INVEPT, i.e. EPT TLB flush, from KVM's VMX enablement path. KVM always does a targeted TLB flush when using a "new" EPT root, in quotes because "new" simply means a root that isn't currently being used by the vCPU. KVM also _deliberately_ runs with stale TLB entries for defunct roots, i.e. doesn't do a TLB flush when vCPUs stop using roots, precisely because KVM does the flush on first use. As called out by the comment in kvm_mmu_load(), the reason KVM flushes on first use is because KVM can't guarantee the correctness of past hypervisors. Jumping back to the global INVEPT, when the painfully terse commit 1439442c7b25 ("KVM: VMX: Enable EPT feature for KVM") was added, the effective TLB flush being performed was: static void vmx_flush_tlb(struct kvm_vcpu *vcpu) { vpid_sync_vcpu_all(to_vmx(vcpu)); } I.e. KVM was not flushing EPT TLB entries when allocating a "new" root, which very strongly suggests that the global INVEPT during hardware enabling was a misguided hack that addressed the most obvious symptom, but failed to fix the underlying bug. Signed-off-by: Sean Christopherson <seanjc@google.com> --- arch/x86/kvm/vmx/vmx.c | 3 --- 1 file changed, 3 deletions(-) base-commit: af0903ab52ee6d6f0f63af67fa73d5eb00f79b9a