[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <ZID7Ro8vTnhAl2jo@google.com>
Date: Wed, 7 Jun 2023 14:48:54 -0700
From: Sean Christopherson <seanjc@...gle.com>
To: Like Xu <like.xu.linux@...il.com>
Cc: Paolo Bonzini <pbonzini@...hat.com>, kvm@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH kvm-unit-tests] x86/pmu: Check counter is working properly
before AMD pmu init
On Tue, Feb 21, 2023, Like Xu wrote:
> From: Like Xu <likexu@...cent.com>
>
> Avoid misleading pmu initialisation when vPMU is globally disabled
> by performing a read/write consistency test on one of AMD's counters.
> Without this check, pmu test will fail rather than be skipped.
This is unnecessary, the testcases can instead check KVM's enable_pmu module param.
Of course, runtime.bash has a bug and doesn't play nice with multiple requirements,
but that's easy enough to fix. Patches incoming.
@@ -206,7 +206,7 @@ extra_params = -cpu max,vendor=GenuineIntel
[pmu]
file = pmu.flat
extra_params = -cpu max
-check = /proc/sys/kernel/nmi_watchdog=0
+check = /sys/module/kvm/parameters/enable_pmu=Y /proc/sys/kernel/nmi_watchdog=0
accel = kvm
groups = pmu
Powered by blists - more mailing lists