[PATCH 2/4] KVM: x86/pmu: Simplify intel_hw_event_available()

[Sean Christopherson]

Walk only the "real", i.e. non-pseudo, architectural events when checking
if a hardware event is available, i.e. isn't disabled by guest CPUID.
Skipping pseudo-arch events in the loop body is unnecessarily convoluted,
especially now that KVM has enums that delineate between real and pseudo
events.

Signed-off-by: Sean Christopherson <seanjc@xxxxxxxxxx>
---
 arch/x86/kvm/vmx/pmu_intel.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/arch/x86/kvm/vmx/pmu_intel.c b/arch/x86/kvm/vmx/pmu_intel.c
index 0050d71c9c01..f281e634af3c 100644
--- a/arch/x86/kvm/vmx/pmu_intel.c
+++ b/arch/x86/kvm/vmx/pmu_intel.c
@@ -122,17 +122,16 @@ static bool intel_hw_event_available(struct kvm_pmc *pmc)
 
 	BUILD_BUG_ON(ARRAY_SIZE(intel_arch_events) != NR_INTEL_ARCH_EVENTS);
 
-	for (i = 0; i < NR_INTEL_ARCH_EVENTS; i++) {
+	/*
+	 * Disallow events reported as unavailable in guest CPUID.  Note, this
+	 * doesn't apply to pseudo-architectural events.
+	 */
+	for (i = 0; i < NR_REAL_INTEL_ARCH_EVENTS; i++) {
 		if (intel_arch_events[i].eventsel != event_select ||
 		    intel_arch_events[i].unit_mask != unit_mask)
 			continue;
 
-		/* disable event that reported as not present by cpuid */
-		if ((i < PSEUDO_ARCH_REFERENCE_CYCLES) &&
-		    !(pmu->available_event_types & (1 << i)))
-			return false;
-
-		break;
+		return pmu->available_event_types & BIT(i);
 	}
 
 	return true;
-- 
2.41.0.162.gfafddb0af9-goog