diff mbox series

[23/67] KVM: VMX: Suppress PI notifications whenever the vCPU is put

Message ID 20250404193923.1413163-24-seanjc@google.com (mailing list archive)
State New
Headers show
Series KVM: iommu: Overhaul device posted IRQs support | expand

Commit Message

Sean Christopherson April 4, 2025, 7:38 p.m. UTC
Suppress posted interrupt notifications (set PID.SN=1) whenever the vCPU
is put, i.e. unloaded, not just when the vCPU is preempted, as KVM doesn't
do anything in response to a notification IRQ that arrives in the host,
nor does KVM rely on the Outstanding Notification (PID.ON) flag when the
vCPU is unloaded.  And, the cost of scanning the PIR to manually set PID.ON
when loading the vCPU is quite small, especially relative to the cost of
loading (and unloading) a vCPU.

On the flip side, leaving SN clear means a notification for the vCPU will
result in a spurious IRQ for the pCPU, even if vCPU task is scheduled out,
running in userspace, etc.  Even worse, if the pCPU is running a different
vCPU, the spurious IRQ could trigger posted interrupt processing for the
wrong vCPU, which is technically a violation of the architecture, as
setting bits in PIR aren't supposed to be propagated to the vIRR until a
notification IRQ is received.

The saving grace of the current behavior is that hardware sends
notification interrupts if and only if PID.ON=0, i.e. only the first
posted interrupt for a vCPU will trigger a spurious IRQ (for each window
where the vCPU is unloaded).

Ideally, KVM would suppress notifications before enabling IRQs in the
VM-Exit, but KVM relies on PID.ON as an indicator that there is a posted
interrupt pending in PIR, e.g. in vmx_sync_pir_to_irr(), and sadly there
is no way to ask hardware to set PID.ON, but not generate an interrupt.
That could be solved by using pi_has_pending_interrupt() instead of
checking only PID.ON, but it's not at all clear that would be a performance
win, as KVM would end up scanning the entire PIR whenever an interrupt
isn't pending.

And long term, the spurious IRQ window, i.e. where a vCPU is loaded with
IRQs enabled, can effectively be made smaller for hot paths by moving
performance critical VM-Exit handlers into the fastpath, i.e. by never
enabling IRQs for hot path VM-Exits.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 arch/x86/kvm/vmx/posted_intr.c | 30 +++++++++++++++++-------------
 1 file changed, 17 insertions(+), 13 deletions(-)
diff mbox series

Patch

diff --git a/arch/x86/kvm/vmx/posted_intr.c b/arch/x86/kvm/vmx/posted_intr.c
index 1b6b655a2b8a..00818ca30ee0 100644
--- a/arch/x86/kvm/vmx/posted_intr.c
+++ b/arch/x86/kvm/vmx/posted_intr.c
@@ -70,13 +70,10 @@  void vmx_vcpu_pi_load(struct kvm_vcpu *vcpu, int cpu)
 	/*
 	 * If the vCPU wasn't on the wakeup list and wasn't migrated, then the
 	 * full update can be skipped as neither the vector nor the destination
-	 * needs to be changed.
+	 * needs to be changed.  Clear SN even if there is no assigned device,
+	 * again for simplicity.
 	 */
 	if (pi_desc->nv != POSTED_INTR_WAKEUP_VECTOR && vcpu->cpu == cpu) {
-		/*
-		 * Clear SN if it was set due to being preempted.  Again, do
-		 * this even if there is no assigned device for simplicity.
-		 */
 		if (pi_test_and_clear_sn(pi_desc))
 			goto after_clear_sn;
 		return;
@@ -200,15 +197,22 @@  void vmx_vcpu_pi_put(struct kvm_vcpu *vcpu)
 	if (!vmx_needs_pi_wakeup(vcpu))
 		return;
 
-	if (kvm_vcpu_is_blocking(vcpu) && !vmx_interrupt_blocked(vcpu))
+	/*
+	 * If the vCPU is blocking with IRQs enabled and ISN'T being preempted,
+	 * enable the wakeup handler so that notification IRQ wakes the vCPU as
+	 * expected.  There is no need to enable the wakeup handler if the vCPU
+	 * is preempted between setting its wait state and manually scheduling
+	 * out, as the task is still runnable, i.e. doesn't need a wake event
+	 * from KVM to be scheduled in.
+	 *
+	 * If the wakeup handler isn't being enabled, Suppress Notifications as
+	 * the cost of propagating PIR.IRR to PID.ON is negligible compared to
+	 * the cost of a spurious IRQ, and vCPU put/load is a slow path.
+	 */
+	if (!vcpu->preempted && kvm_vcpu_is_blocking(vcpu) &&
+	    !vmx_interrupt_blocked(vcpu))
 		pi_enable_wakeup_handler(vcpu);
-
-	/*
-	 * Set SN when the vCPU is preempted.  Note, the vCPU can both be seen
-	 * as blocking and preempted, e.g. if it's preempted between setting
-	 * its wait state and manually scheduling out.
-	 */
-	if (vcpu->preempted)
+	else
 		pi_set_sn(pi_desc);
 }