KVM calls kvm_pmu_set_counter_event_type() when PMCCFILTR is configured. But this function can't deals with PMCCFILTR correctly because the evtCount bit of PMCCFILTR, which is reserved 0, conflits with the SW_INCR event type of other PMXEVTYPER<n> registers. To fix it, when eventsel == 0, KVM shouldn't return immediately, but instead it needs to check further if select_idx is ARMV8_PMU_CYCLE_IDX. Another issue is that KVM shouldn't copy the eventsel bits of PMCCFILTER directly to attr.config. Istead it shoudl convert the request to perf_event of type 0x11 (i.e. the "cpu cycle" event type). Signed-off-by: Wei Huang <wei@xxxxxxxxxx> --- virt/kvm/arm/pmu.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/virt/kvm/arm/pmu.c b/virt/kvm/arm/pmu.c index 6e9c40e..13cc812 100644 --- a/virt/kvm/arm/pmu.c +++ b/virt/kvm/arm/pmu.c @@ -379,7 +379,8 @@ void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data, eventsel = data & ARMV8_PMU_EVTYPE_EVENT; /* Software increment event does't need to be backed by a perf event */ - if (eventsel == ARMV8_PMU_EVTYPE_EVENT_SW_INCR) + if (eventsel == ARMV8_PMU_EVTYPE_EVENT_SW_INCR && + select_idx != ARMV8_PMU_CYCLE_IDX) return; memset(&attr, 0, sizeof(struct perf_event_attr)); @@ -391,7 +392,7 @@ void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data, attr.exclude_kernel = data & ARMV8_PMU_EXCLUDE_EL1 ? 1 : 0; attr.exclude_hv = 1; /* Don't count EL2 events */ attr.exclude_host = 1; /* Don't count host events */ - attr.config = eventsel; + attr.config = (select_idx == ARMV8_PMU_CYCLE_IDX) ? 0x011 : eventsel; counter = kvm_pmu_get_counter_value(vcpu, select_idx); /* The initial sample period (overflow count) of an event. */ -- 2.7.4 -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
