Kernel KVM virtualization development
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Jeremiah McReynolds <jmcrey@google.com>
Cc: kvm@vger.kernel.org
Subject: Re: [kvm-unit-tests PATCH] x86/msr: Disable PMU for MSR tests
Date: Thu, 17 Sep 2026 12:36:23 -0700	[thread overview]
Message-ID: <aqxBNxlhV8WbFxo7@google.com> (raw)
In-Reply-To: <20260917172651.3322899-1-jmcrey@google.com>

On Thu, Sep 17, 2026, Jeremiah McReynolds wrote:
> QEMUv11 introduced a check when PMU is enabled to ensure that the
> virtualized CPU is the same as the underlying host CPU[1]. The MSR test
> always runs under GenuineIntel with the max CPU model, so PMU is
> implicitly enabled. However, when run on AMD, this new check in QEMU
> causes the test to fail with the log: "host doesn't support requested
> feature: vPMU".
> 
> Explicitly disable PMU on the MSR test to avoid this failure.
> 
> [1]: https://gitlab.com/qemu-project/qemu/-/commit/4c7f05232cd1db5db114d36be2b210d24f1b9ad3
> 
> Signed-off-by: Jeremiah McReynolds <jmcrey@google.com>
> ---
>  x86/unittests.cfg | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/x86/unittests.cfg b/x86/unittests.cfg
> index ae731a93..06fe92f9 100644
> --- a/x86/unittests.cfg
> +++ b/x86/unittests.cfg
> @@ -174,9 +174,10 @@ arch = x86_64
>  # Use GenuineIntel to ensure SYSENTER MSRs are fully preserved, and to test
>  # SVM emulation of Intel CPU behavior.  Use the max CPU model so that 64-bit
>  # support follows the host kernel.  Running a 32-bit guest on a 64-bit host
> -# will fail due to shortcomings in KVM.
> +# will fail due to shortcomings in KVM.  The max CPU model turns on PMU, but
> +# Intel PMU registers cannot be virtualized on AMD, so turn it off.
>  file = msr.flat
> -qemu_params = -cpu max,vendor=GenuineIntel
> +qemu_params = -cpu max,vendor=GenuineIntel,pmu=off

I double checked that "pmu" has existed for a while; it goes back all the way to
QEMU v1.6.0, so we should be good on that front.

>  
>  [msr64]
>  # Same as above, but with a minimal 64-bit CPU model to validate cases where an
> -- 
> 2.55.0.1082.g2b9226bbc0-goog
> 

  reply	other threads:[~2026-09-17 19:36 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-17 17:26 [kvm-unit-tests PATCH] x86/msr: Disable PMU for MSR tests Jeremiah McReynolds
2026-09-17 19:36 ` Sean Christopherson [this message]
2026-09-25 22:44 ` Sean Christopherson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=aqxBNxlhV8WbFxo7@google.com \
    --to=seanjc@google.com \
    --cc=jmcrey@google.com \
    --cc=kvm@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox