diff options
| author | Sean Christopherson <seanjc@google.com> | 2021-12-08 01:52:26 +0000 |
|---|---|---|
| committer | Paolo Bonzini <pbonzini@redhat.com> | 2022-01-19 12:14:45 -0500 |
| commit | 0f65a9d337676b966316db17374fbef910ab8e4a (patch) | |
| tree | ff3b05ff49c38cbb4b12fe43f01abb0f21b46f62 | |
| parent | 782f64558de7bef84b90ea812deb38f0e53a8c7a (diff) | |
| download | linux-stable-0f65a9d337676b966316db17374fbef910ab8e4a.tar.gz linux-stable-0f65a9d337676b966316db17374fbef910ab8e4a.tar.bz2 linux-stable-0f65a9d337676b966316db17374fbef910ab8e4a.zip | |
KVM: VMX: Don't do full kick when triggering posted interrupt "fails"
Replace the full "kick" with just the "wake" in the fallback path when
triggering a virtual interrupt via a posted interrupt fails because the
guest is not IN_GUEST_MODE. If the guest transitions into guest mode
between the check and the kick, then it's guaranteed to see the pending
interrupt as KVM syncs the PIR to IRR (and onto GUEST_RVI) after setting
IN_GUEST_MODE. Kicking the guest in this case is nothing more than an
unnecessary VM-Exit (and host IRQ).
Opportunistically update comments to explain the various ordering rules
and barriers at play.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20211208015236.1616697-17-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| -rw-r--r-- | arch/x86/kvm/vmx/vmx.c | 4 | ||||
| -rw-r--r-- | arch/x86/kvm/x86.c | 9 |
2 files changed, 7 insertions, 6 deletions
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index 9ab98e1bd65a..ac3e943ec2a4 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -3998,7 +3998,7 @@ static int vmx_deliver_nested_posted_interrupt(struct kvm_vcpu *vcpu, /* the PIR and ON have been set by L1. */ if (!kvm_vcpu_trigger_posted_interrupt(vcpu, true)) - kvm_vcpu_kick(vcpu); + kvm_vcpu_wake_up(vcpu); return 0; } return -1; @@ -4036,7 +4036,7 @@ static int vmx_deliver_posted_interrupt(struct kvm_vcpu *vcpu, int vector) * posted interrupt "fails" because vcpu->mode != IN_GUEST_MODE. */ if (!kvm_vcpu_trigger_posted_interrupt(vcpu, false)) - kvm_vcpu_kick(vcpu); + kvm_vcpu_wake_up(vcpu); return 0; } diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 4b9728a53de6..55518b7d3b96 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -9978,10 +9978,11 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) smp_mb__after_srcu_read_unlock(); /* - * This handles the case where a posted interrupt was - * notified with kvm_vcpu_kick. Assigned devices can - * use the POSTED_INTR_VECTOR even if APICv is disabled, - * so do it even if APICv is disabled on this vCPU. + * Process pending posted interrupts to handle the case where the + * notification IRQ arrived in the host, or was never sent (because the + * target vCPU wasn't running). Do this regardless of the vCPU's APICv + * status, KVM doesn't update assigned devices when APICv is inhibited, + * i.e. they can post interrupts even if APICv is temporarily disabled. */ if (kvm_lapic_enabled(vcpu)) static_call_cond(kvm_x86_sync_pir_to_irr)(vcpu); |