[PATCH v2] KVM: x86/pmu: Don't retry a counter whose config was rejected

[PATCH v2] KVM: x86/pmu: Don't retry a counter whose config was rejected

[Luka Absandze]

kvm_pmu_handle_event() re-arms the reprogram bit for every failed
reprogram, on the assumption that the failure is transient and a later
refresh will succeed.  That is true for contention, e.g. the -EBUSY from
x86_reserve_hardware(), but not for a configuration the host PMU driver
rejects outright.  A rejected config can never succeed on retry, so the
counter is reprogrammed on every PMU refresh for as long as the guest
leaves it enabled, and every attempt fails the same way.

Re-arm only for the errnos that indicate a transient condition, i.e.
-EBUSY and -ENOMEM.

Suggested-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/all/20260916123315.89042-1-absandze@amazon.de/
Signed-off-by: Luka Absandze <absandze@amazon.de>
---
v2:
 - Use an allow-list of transient errnos (-EBUSY, -ENOMEM) rather than
   excluding -EINVAL, and hoist 'r' out of the loop.  Diff taken from
   Sean's suggestion verbatim. [Sean]
 - Reword the comment to say "fails on a transient condition". [Sean]
 - v1: https://lore.kernel.org/all/20260918160006.61816-1-absandze@amazon.de/

diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
index a7d60c8785cd..03a470c49a74 100644
--- a/arch/x86/kvm/pmu.c
+++ b/arch/x86/kvm/pmu.c
@@ -662,7 +662,7 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
 	DECLARE_BITMAP(bitmap, X86_PMC_IDX_MAX);
 	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
 	struct kvm_pmc *pmc;
-	int bit;
+	int bit, r;
 
 	bitmap_copy(bitmap, pmu->reprogram_pmi, X86_PMC_IDX_MAX);
 
@@ -676,12 +676,14 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
 
 	kvm_for_each_pmc(pmu, pmc, bit, bitmap) {
 		/*
-		 * If reprogramming fails, e.g. due to contention, re-set the
-		 * reprogram bit, i.e. opportunistically try again on the next
-		 * PMU refresh.  Don't make a new request as doing so can stall
-		 * the guest if reprogramming repeatedly fails.
+		 * If reprogramming fails on a transient condition, e.g. due to
+		 * contention, re-set the reprogram bit, i.e. opportunistically
+		 * try again on the next PMU refresh.  Don't make a new request
+		 * as doing so can stall the guest if reprogramming repeatedly
+		 * fails.
 		 */
-		if (reprogram_counter(pmc))
+		r = reprogram_counter(pmc);
+		if (r == -EBUSY || r == -ENOMEM)
 			set_bit(pmc->idx, pmu->reprogram_pmi);
 	}
 
-- 
2.47.3

Re: [PATCH v2] KVM: x86/pmu: Don't retry a counter whose config was rejected

[Sandipan Das]

On 20-09-2026 21:24, Luka Absandze wrote:
> kvm_pmu_handle_event() re-arms the reprogram bit for every failed
> reprogram, on the assumption that the failure is transient and a later
> refresh will succeed.  That is true for contention, e.g. the -EBUSY from
> x86_reserve_hardware(), but not for a configuration the host PMU driver
> rejects outright.  A rejected config can never succeed on retry, so the
> counter is reprogrammed on every PMU refresh for as long as the guest
> leaves it enabled, and every attempt fails the same way.
> 
> Re-arm only for the errnos that indicate a transient condition, i.e.
> -EBUSY and -ENOMEM.
> 
> Suggested-by: Sean Christopherson <seanjc@google.com>
> Link: https://lore.kernel.org/all/20260916123315.89042-1-absandze@amazon.de/
> Signed-off-by: Luka Absandze <absandze@amazon.de>
> ---
> v2:
>  - Use an allow-list of transient errnos (-EBUSY, -ENOMEM) rather than
>    excluding -EINVAL, and hoist 'r' out of the loop.  Diff taken from
>    Sean's suggestion verbatim. [Sean]
>  - Reword the comment to say "fails on a transient condition". [Sean]
>  - v1: https://lore.kernel.org/all/20260918160006.61816-1-absandze@amazon.de/
> 
> diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
> index a7d60c8785cd..03a470c49a74 100644
> --- a/arch/x86/kvm/pmu.c
> +++ b/arch/x86/kvm/pmu.c
> @@ -662,7 +662,7 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
>  	DECLARE_BITMAP(bitmap, X86_PMC_IDX_MAX);
>  	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
>  	struct kvm_pmc *pmc;
> -	int bit;
> +	int bit, r;
>  
>  	bitmap_copy(bitmap, pmu->reprogram_pmi, X86_PMC_IDX_MAX);
>  
> @@ -676,12 +676,14 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
>  
>  	kvm_for_each_pmc(pmu, pmc, bit, bitmap) {
>  		/*
> -		 * If reprogramming fails, e.g. due to contention, re-set the
> -		 * reprogram bit, i.e. opportunistically try again on the next
> -		 * PMU refresh.  Don't make a new request as doing so can stall
> -		 * the guest if reprogramming repeatedly fails.
> +		 * If reprogramming fails on a transient condition, e.g. due to
> +		 * contention, re-set the reprogram bit, i.e. opportunistically
> +		 * try again on the next PMU refresh.  Don't make a new request
> +		 * as doing so can stall the guest if reprogramming repeatedly
> +		 * fails.
>  		 */
> -		if (reprogram_counter(pmc))
> +		r = reprogram_counter(pmc);
> +		if (r == -EBUSY || r == -ENOMEM)
>  			set_bit(pmc->idx, pmu->reprogram_pmi);
>  	}
>  

Reviewed-by: Sandipan Das <sandipan.das@amd.com>

Re: [PATCH v2] KVM: x86/pmu: Don't retry a counter whose config was rejected

[Sandipan Das]

On 20-09-2026 21:24, Luka Absandze wrote:
> kvm_pmu_handle_event() re-arms the reprogram bit for every failed
> reprogram, on the assumption that the failure is transient and a later
> refresh will succeed.  That is true for contention, e.g. the -EBUSY from
> x86_reserve_hardware(), but not for a configuration the host PMU driver
> rejects outright.  A rejected config can never succeed on retry, so the
> counter is reprogrammed on every PMU refresh for as long as the guest
> leaves it enabled, and every attempt fails the same way.
> 
> Re-arm only for the errnos that indicate a transient condition, i.e.
> -EBUSY and -ENOMEM.
> 
> Suggested-by: Sean Christopherson <seanjc@google.com>
> Link: https://lore.kernel.org/all/20260916123315.89042-1-absandze@amazon.de/
> Signed-off-by: Luka Absandze <absandze@amazon.de>
> ---
> v2:
>  - Use an allow-list of transient errnos (-EBUSY, -ENOMEM) rather than
>    excluding -EINVAL, and hoist 'r' out of the loop.  Diff taken from
>    Sean's suggestion verbatim. [Sean]
>  - Reword the comment to say "fails on a transient condition". [Sean]
>  - v1: https://lore.kernel.org/all/20260918160006.61816-1-absandze@amazon.de/
> 
> diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
> index a7d60c8785cd..03a470c49a74 100644
> --- a/arch/x86/kvm/pmu.c
> +++ b/arch/x86/kvm/pmu.c
> @@ -662,7 +662,7 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
>  	DECLARE_BITMAP(bitmap, X86_PMC_IDX_MAX);
>  	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
>  	struct kvm_pmc *pmc;
> -	int bit;
> +	int bit, r;
>  
>  	bitmap_copy(bitmap, pmu->reprogram_pmi, X86_PMC_IDX_MAX);
>  
> @@ -676,12 +676,14 @@ void kvm_pmu_handle_event(struct kvm_vcpu *vcpu)
>  
>  	kvm_for_each_pmc(pmu, pmc, bit, bitmap) {
>  		/*
> -		 * If reprogramming fails, e.g. due to contention, re-set the
> -		 * reprogram bit, i.e. opportunistically try again on the next
> -		 * PMU refresh.  Don't make a new request as doing so can stall
> -		 * the guest if reprogramming repeatedly fails.
> +		 * If reprogramming fails on a transient condition, e.g. due to
> +		 * contention, re-set the reprogram bit, i.e. opportunistically
> +		 * try again on the next PMU refresh.  Don't make a new request
> +		 * as doing so can stall the guest if reprogramming repeatedly
> +		 * fails.
>  		 */
> -		if (reprogram_counter(pmc))
> +		r = reprogram_counter(pmc);
> +		if (r == -EBUSY || r == -ENOMEM)
>  			set_bit(pmc->idx, pmu->reprogram_pmi);
>  	}
>  

Reviewed-by: Sandipan Das <sandipan.das@amd.com>