Skip to contentSkip to conversation

KVM: arm64: Correctly handle RES0 bits PMEVTYPER<n>_EL0.evtCount

PATCH Applied Oliver Upton 3 participants 5 messages Jul 13, 2023 – Jul 14, 2023

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 KVM: arm64: Correctly handle RES0 bits PMEVTYPER<n>_EL0.evtCount

The PMU event ID varies from 10 to 16 bits, depending on the PMU version. If the PMU only supports 10 bits of event ID, bits [15:10] of the evtCount field behave as RES0.

While the actual PMU emulation code gets this right (i.e. RES0 bits are masked out when programming the perf event), the sysreg emulation writes an unmasked value to the in-memory cpu context. The net effect is that guest reads and writes of PMEVTYPER<n>_EL0 will see non-RES0 behavior in the reserved bits of the field.

As it so happens, kvm_pmu_set_counter_event_type() already writes a masked value to the in-memory context that gets overwritten by access_pmu_evtyper(). Fix the issue by removing the unnecessary (and incorrect) register write in access_pmu_evtyper().

---
 arch/arm64/kvm/sys_regs.c | 1 -
 1 file changed, 1 deletion(-)
arch/arm64/kvm/sys_regs.c +0 -1
index bd3431823ec5..fa7597882ed8 100644
@@ -986,7 +986,6 @@ static bool access_pmu_evtyper(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
 
 	if (p->is_write) {
 		kvm_pmu_set_counter_event_type(vcpu, p->regval, idx);
-		__vcpu_sys_reg(vcpu, reg) = p->regval & ARMV8_PMU_EVTYPE_MASK;
 		kvm_vcpu_pmu_restore_guest(vcpu);
 	} else {
 		p->regval = __vcpu_sys_reg(vcpu, reg) & ARMV8_PMU_EVTYPE_MASK;

base-commit: dcf89d111199562fa5f31a1bb76f17bc4831f6da

Signature · 1 line
2.41.0.255.g8b1d071c50-goog
maintainer Reviewed-by on v1 1/1
#1 lore raw

AOn Thu, 13 Jul 2023 23:16:49 +0100,

Quoted 11 lines · Oliver Upton

The PMU event ID varies from 10 to 16 bits, depending on the PMU version. If the PMU only supports 10 bits of event ID, bits [15:10] of the evtCount field behave as RES0.

While the actual PMU emulation code gets this right (i.e. RES0 bits are masked out when programming the perf event), the sysreg emulation writes an unmasked value to the in-memory cpu context. The net effect is that guest reads and writes of PMEVTYPER<n>_EL0 will see non-RES0 behavior in the reserved bits of the field.

Quoted 6 lines

As it so happens, kvm_pmu_set_counter_event_type() already writes a masked value to the in-memory context that gets overwritten by access_pmu_evtyper(). Fix the issue by removing the unnecessary (and incorrect) register write in access_pmu_evtyper().

	M.
Signature · 1 line
Without deviation from the norm, progress is not possible.
Reviewed-by on v1 1/1
#2 lore raw
In reply to Oliver Upton
Quoted 11 lines · Oliver Upton

The PMU event ID varies from 10 to 16 bits, depending on the PMU version. If the PMU only supports 10 bits of event ID, bits [15:10] of the evtCount field behave as RES0.

While the actual PMU emulation code gets this right (i.e. RES0 bits are masked out when programming the perf event), the sysreg emulation writes an unmasked value to the in-memory cpu context. The net effect is that guest reads and writes of PMEVTYPER<n>_EL0 will see non-RES0 behavior in the reserved bits of the field.

Quoted 6 lines

As it so happens, kvm_pmu_set_counter_event_type() already writes a masked value to the in-memory context that gets overwritten by access_pmu_evtyper(). Fix the issue by removing the unnecessary (and incorrect) register write in access_pmu_evtyper().

FYI. Looking at their reset values, fortunately, the pseudo "unknown" reset values used by reset_pmevtyper() has the bits [16:11] cleared.
I would think it's a bit fragile, but not a real problem.

Thank you,
Reiji

#3 lore raw
Quoted 12 lines · Reiji Watanabe
Quoted 11 lines · Oliver Upton

The PMU event ID varies from 10 to 16 bits, depending on the PMU version. If the PMU only supports 10 bits of event ID, bits [15:10] of the evtCount field behave as RES0.

While the actual PMU emulation code gets this right (i.e. RES0 bits are masked out when programming the perf event), the sysreg emulation writes an unmasked value to the in-memory cpu context. The net effect is that guest reads and writes of PMEVTYPER<n>_EL0 will see non-RES0 behavior in the reserved bits of the field.

Quoted 8 lines
Quoted 6 lines

As it so happens, kvm_pmu_set_counter_event_type() already writes a masked value to the in-memory context that gets overwritten by access_pmu_evtyper(). Fix the issue by removing the unnecessary (and incorrect) register write in access_pmu_evtyper().

Thanks!

Quoted 4 lines

FYI. Looking at their reset values, fortunately, the pseudo "unknown" reset values used by reset_pmevtyper() has the bits [16:11] cleared.
I would think it's a bit fragile, but not a real problem.

Fragile indeed. I'd like to clean up the handling a bit more in the future, because it makes no sense why we mask the value read from __vcpu_sys_reg() for emulated reads.

Signature · 2 lines
Thanks,
Oliver
author Applied
#4 lore raw
In reply to Oliver Upton
Quoted 11 lines · Oliver Upton

The PMU event ID varies from 10 to 16 bits, depending on the PMU version. If the PMU only supports 10 bits of event ID, bits [15:10] of the evtCount field behave as RES0.

While the actual PMU emulation code gets this right (i.e. RES0 bits are masked out when programming the perf event), the sysreg emulation writes an unmasked value to the in-memory cpu context. The net effect is that guest reads and writes of PMEVTYPER<n>_EL0 will see non-RES0 behavior in the reserved bits of the field.

[...]

Applied to kvmarm/fixes, thanks!

[1/1] KVM: arm64: Correctly handle RES0 bits PMEVTYPER<n>_EL0.evtCount

      https://git.kernel.org/kvmarm/kvmarm/c/6d4f9236cd67

--
Best,
Oliver