[kvm-unit-tests PATCH] x86/pmu: Explicitly zero PERF_GLOBAL_CTRL at start of PMU test

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Explicitly zero PERF_GLOBAL_CTRL at the start of the PMU test as the
architectural RESET value of PERF_GLOBAL_CTRL is to set all enable bits
for general purpose counters (for backwards compatibility with software
that was written for v1 PMUs).  Leaving PERF_GLOBAL_CTRL set can result in
false failures due to counters unexpectedly being left active.

Signed-off-by: Sean Christopherson <seanjc@xxxxxxxxxx>
---
 x86/pmu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/x86/pmu.c b/x86/pmu.c
index 8cf26b12..9bd0c186 100644
--- a/x86/pmu.c
+++ b/x86/pmu.c
@@ -905,9 +905,6 @@ static void set_ref_cycle_expectations(void)
 	if (!pmu.nr_gp_counters || !pmu_gp_counter_is_available(2))
 		return;
 
-	if (this_cpu_has_perf_global_ctrl())
-		wrmsr(pmu.msr_global_ctl, 0);
-
 	t0 = fenced_rdtsc();
 	start_event(&cnt);
 	t1 = fenced_rdtsc();
@@ -956,6 +953,9 @@ int main(int ac, char **av)
 	handle_irq(PMI_VECTOR, cnt_overflow);
 	buf = malloc(N*64);
 
+	if (this_cpu_has_perf_global_ctrl())
+		wrmsr(pmu.msr_global_ctl, 0);
+
 	check_invalid_rdpmc_gp();
 
 	if (pmu.is_intel) {

base-commit: 72d110d8286baf1b355301cc8c8bdb42be2663fb
-- 
2.49.0.1204.g71687c7c1d-goog





[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux