[PATCH v2 33/59] KVM: x86: Skip IOMMU IRTE updates if there's no old or new vCPU being targeted

Don't "reconfigure" an IRTE into host controlled mode when it's already in
the state, i.e. if KVM's GSI routing changes but the IRQ wasn't and still
isn't being posted to a vCPU.

Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 arch/x86/kvm/irq.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c
index 3a0f28e98429..67fc8901d15f 100644
--- a/arch/x86/kvm/irq.c
+++ b/arch/x86/kvm/irq.c
@@ -542,6 +542,9 @@ static int kvm_pi_update_irte(struct kvm_kernel_irqfd *irqfd,
 			vcpu = NULL;
 	}
 
+	if (!irqfd->irq_bypass_vcpu && !vcpu)
+		return 0;
+
 	r = kvm_x86_call(pi_update_irte)(irqfd, irqfd->kvm, host_irq, irqfd->gsi,
 					 vcpu, irq.vector);
 	if (r) {
-- 
2.49.0.1151.ga128411c76-goog

• [PATCH v2 01/59] KVM: x86: Pass new routing entries and irqfd when updating IRTEs
• [PATCH v2 02/59] KVM: SVM: Track per-vCPU IRTEs using kvm_kernel_irqfd structure
• [PATCH v2 03/59] KVM: SVM: Delete IRTE link from previous vCPU before setting new IRTE
• [PATCH v2 04/59] iommu/amd: KVM: SVM: Delete now-unused cached/previous GA tag fields
• [PATCH v2 05/59] KVM: SVM: Delete IRTE link from previous vCPU irrespective of new routing
• [PATCH v2 06/59] KVM: SVM: Drop pointless masking of default APIC base when setting V_APIC_BAR
• [PATCH v2 07/59] KVM: SVM: Drop pointless masking of kernel page pa's with AVIC HPA masks
• [PATCH v2 08/59] KVM: SVM: Add helper to deduplicate code for getting AVIC backing page
• [PATCH v2 09/59] KVM: SVM: Drop vcpu_svm's pointless avic_backing_page field
• [PATCH v2 10/59] KVM: SVM: Inhibit AVIC if ID is too big instead of rejecting vCPU creation
• [PATCH v2 11/59] KVM: SVM: Drop redundant check in AVIC code on ID during vCPU creation
• [PATCH v2 12/59] KVM: SVM: Track AVIC tables as natively sized pointers, not "struct pages"
• [PATCH v2 13/59] KVM: SVM: Drop superfluous "cache" of AVIC Physical ID entry pointer
• [PATCH v2 14/59] KVM: VMX: Move enable_ipiv knob to common x86
• [PATCH v2 15/59] KVM: SVM: Add enable_ipiv param, never set IsRunning if disabled
• [PATCH v2 16/59] KVM: SVM: Disable (x2)AVIC IPI virtualization if CPU has erratum #1235
• [PATCH v2 17/59] KVM: VMX: Suppress PI notifications whenever the vCPU is put
• [PATCH v2 18/59] KVM: SVM: Add a comment to explain why avic_vcpu_blocking() ignores IRQ blocking
• [PATCH v2 19/59] iommu/amd: KVM: SVM: Use pi_desc_addr to derive ga_root_ptr
• [PATCH v2 20/59] iommu/amd: KVM: SVM: Pass NULL @vcpu_info to indicate "not guest mode"
• [PATCH v2 21/59] KVM: SVM: Stop walking list of routing table entries when updating IRTE
• [PATCH v2 22/59] KVM: VMX: Stop walking list of routing table entries when updating IRTE
• [PATCH v2 23/59] KVM: SVM: Extract SVM specific code out of get_pi_vcpu_info()
• [PATCH v2 24/59] KVM: x86: Move IRQ routing/delivery APIs from x86.c => irq.c
• [PATCH v2 25/59] KVM: x86: Nullify irqfd->producer after updating IRTEs
• [PATCH v2 26/59] KVM: x86: Dedup AVIC vs. PI code for identifying target vCPU
• [PATCH v2 27/59] KVM: x86: Move posted interrupt tracepoint to common code
• [PATCH v2 28/59] KVM: SVM: Clean up return handling in avic_pi_update_irte()
• [PATCH v2 29/59] iommu: KVM: Split "struct vcpu_data" into separate AMD vs. Intel structs
• [PATCH v2 30/59] KVM: Don't WARN if updating IRQ bypass route fails
• [PATCH v2 31/59] KVM: Fold kvm_arch_irqfd_route_changed() into kvm_arch_update_irqfd_routing()
• [PATCH v2 32/59] KVM: x86: Track irq_bypass_vcpu in common x86 code
• [PATCH v2 33/59] KVM: x86: Skip IOMMU IRTE updates if there's no old or new vCPU being targeted
• [PATCH v2 34/59] KVM: x86: Don't update IRTE entries when old and new routes were !MSI
• [PATCH v2 35/59] KVM: SVM: Revert IRTE to legacy mode if IOMMU doesn't provide IR metadata
• [PATCH v2 36/59] KVM: SVM: Take and hold ir_list_lock across IRTE updates in IOMMU
• [PATCH v2 37/59] iommu/amd: Document which IRTE fields amd_iommu_update_ga() can modify
• [PATCH v2 38/59] iommu/amd: KVM: SVM: Infer IsRun from validity of pCPU destination
• [PATCH v2 39/59] iommu/amd: Factor out helper for manipulating IRTE GA/CPU info
• [PATCH v2 40/59] iommu/amd: KVM: SVM: Set pCPU info in IRTE when setting vCPU affinity
• [PATCH v2 41/59] iommu/amd: KVM: SVM: Add IRTE metadata to affined vCPU's list if AVIC is inhibited
• [PATCH v2 42/59] KVM: SVM: Don't check for assigned device(s) when updating affinity
• [PATCH v2 43/59] KVM: SVM: Don't check for assigned device(s) when activating AVIC
• [PATCH v2 44/59] KVM: SVM: WARN if (de)activating guest mode in IOMMU fails
• [PATCH v2 45/59] KVM: SVM: Process all IRTEs on affinity change even if one update fails
• [PATCH v2 46/59] KVM: SVM: WARN if updating IRTE GA fields in IOMMU fails
• [PATCH v2 47/59] KVM: x86: Drop superfluous "has assigned device" check in kvm_pi_update_irte()
• [PATCH v2 48/59] KVM: x86: WARN if IRQ bypass isn't supported in kvm_pi_update_irte()
• [PATCH v2 49/59] KVM: x86: WARN if IRQ bypass routing is updated without in-kernel local APIC
• [PATCH v2 50/59] KVM: SVM: WARN if ir_list is non-empty at vCPU free
• [PATCH v2 51/59] KVM: x86: Decouple device assignment from IRQ bypass
• [PATCH v2 52/59] KVM: VMX: WARN if VT-d Posted IRQs aren't possible when starting IRQ bypass
• [PATCH v2 53/59] KVM: SVM: Use vcpu_idx, not vcpu_id, for GA log tag/metadata
• [PATCH v2 54/59] iommu/amd: WARN if KVM calls GA IRTE helpers without virtual APIC support
• [PATCH v2 55/59] KVM: SVM: Fold avic_set_pi_irte_mode() into its sole caller
• [PATCH v2 56/59] KVM: SVM: Don't check vCPU's blocking status when toggling AVIC on/off
• [PATCH v2 57/59] KVM: SVM: Consolidate IRTE update when toggling AVIC on/off
• [PATCH v2 58/59] iommu/amd: KVM: SVM: Allow KVM to control need for GA log interrupts
• [PATCH v2 59/59] KVM: SVM: Generate GA log IRQs only if the associated vCPUs is blocking