From: Zhao Liu <zhao1.liu@intel.com>
To: Dongli Zhang <dongli.zhang@oracle.com>
Cc: qemu-devel@nongnu.org, kvm@vger.kernel.org, pbonzini@redhat.com,
mtosatti@redhat.com, sandipan.das@amd.com, babu.moger@amd.com,
likexu@tencent.com, like.xu.linux@gmail.com, groug@kaod.org,
khorenko@virtuozzo.com, alexander.ivanov@virtuozzo.com,
den@virtuozzo.com, davydov-max@yandex-team.ru,
xiaoyao.li@intel.com, dapeng1.mi@linux.intel.com,
joe.jin@oracle.com, ewanhai-oc@zhaoxin.com, ewanhai@zhaoxin.com
Subject: Re: [PATCH v7 0/9] target/i386/kvm/pmu: PMU Enhancement, Bugfix and Cleanup
Date: Tue, 16 Dec 2025 16:25:52 +0800 [thread overview]
Message-ID: <aUEXkDDOba+oZ4v+@intel.com> (raw)
In-Reply-To: <20251111061532.36702-1-dongli.zhang@oracle.com>
On Mon, Nov 10, 2025 at 10:14:49PM -0800, Dongli Zhang wrote:
> Date: Mon, 10 Nov 2025 22:14:49 -0800
> From: Dongli Zhang <dongli.zhang@oracle.com>
> Subject: [PATCH v7 0/9] target/i386/kvm/pmu: PMU Enhancement, Bugfix and
> Cleanup
> X-Mailer: git-send-email 2.43.5
>
> This patchset addresses four bugs related to AMD PMU virtualization.
>
> 1. The PerfMonV2 is still available if PERCORE if disabled via
> "-cpu host,-perfctr-core".
>
> 2. The VM 'cpuid' command still returns PERFCORE although "-pmu" is
> configured.
>
> 3. The third issue is that using "-cpu host,-pmu" does not disable AMD PMU
> virtualization. When using "-cpu EPYC" or "-cpu host,-pmu", AMD PMU
> virtualization remains enabled. On the VM's Linux side, you might still
> see:
>
> [ 0.510611] Performance Events: Fam17h+ core perfctr, AMD PMU driver.
>
> instead of:
>
> [ 0.596381] Performance Events: PMU not available due to virtualization, using software events only.
> [ 0.600972] NMI watchdog: Perf NMI watchdog permanently disabled
>
> To address this, KVM_CAP_PMU_CAPABILITY is used to set KVM_PMU_CAP_DISABLE
> when "-pmu" is configured.
>
> 4. The fourth issue is that unreclaimed performance events (after a QEMU
> system_reset) in KVM may cause random, unwanted, or unknown NMIs to be
> injected into the VM.
>
> The AMD PMU registers are not reset during QEMU system_reset.
>
> (1) If the VM is reset (e.g., via QEMU system_reset or VM kdump/kexec) while
> running "perf top", the PMU registers are not disabled properly.
>
> (2) Despite x86_cpu_reset() resetting many registers to zero, kvm_put_msrs()
> does not handle AMD PMU registers, causing some PMU events to remain
> enabled in KVM.
>
> (3) The KVM kvm_pmc_speculative_in_use() function consistently returns true,
> preventing the reclamation of these events. Consequently, the
> kvm_pmc->perf_event remains active.
>
> (4) After a reboot, the VM kernel may report the following error:
>
> [ 0.092011] Performance Events: Fam17h+ core perfctr, Broken BIOS detected, complain to your hardware vendor.
> [ 0.092023] [Firmware Bug]: the BIOS has corrupted hw-PMU resources (MSR c0010200 is 530076)
>
> (5) In the worst case, the active kvm_pmc->perf_event may inject unknown
> NMIs randomly into the VM kernel:
>
> [...] Uhhuh. NMI received for unknown reason 30 on CPU 0.
>
> To resolve these issues, we propose resetting AMD PMU registers during the
> VM reset process
Hi Dongli,
Except for Patch 1 & 2 which need compatibility options (if you think
it's okay, I could help take these 2 and fix them when v11.0's compat
array is ready).
The other patches still LGTM. Maybe it's better to have a v8 excluding
patch 1 & 2?
Regards,
Zhao
next prev parent reply other threads:[~2025-12-16 8:01 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-11 6:14 [PATCH v7 0/9] target/i386/kvm/pmu: PMU Enhancement, Bugfix and Cleanup Dongli Zhang
2025-11-11 6:14 ` [PATCH v7 1/9] target/i386: disable PerfMonV2 when PERFCORE unavailable Dongli Zhang
2025-11-11 6:14 ` [PATCH v7 2/9] target/i386: disable PERFCORE when "-pmu" is configured Dongli Zhang
2025-11-19 11:06 ` Zhao Liu
2025-11-20 0:22 ` Dongli Zhang
2025-11-25 3:47 ` Zhao Liu
2025-11-11 6:14 ` [PATCH v7 3/9] target/i386/kvm: set KVM_PMU_CAP_DISABLE if " Dongli Zhang
2025-11-11 6:14 ` [PATCH v7 4/9] target/i386/kvm: extract unrelated code out of kvm_x86_build_cpuid() Dongli Zhang
2025-11-11 6:14 ` [PATCH v7 5/9] target/i386/kvm: rename architectural PMU variables Dongli Zhang
2025-11-11 6:14 ` [PATCH v7 6/9] target/i386/kvm: query kvm.enable_pmu parameter Dongli Zhang
2025-11-11 6:14 ` [PATCH v7 7/9] target/i386/kvm: reset AMD PMU registers during VM reset Dongli Zhang
2025-11-11 6:14 ` [PATCH v7 8/9] target/i386/kvm: support perfmon-v2 for reset Dongli Zhang
2025-11-11 6:14 ` [PATCH v7 9/9] target/i386/kvm: don't stop Intel PMU counters Dongli Zhang
2025-11-19 2:07 ` [PATCH v7 0/9] target/i386/kvm/pmu: PMU Enhancement, Bugfix and Cleanup Dongli Zhang
2025-12-16 8:25 ` Zhao Liu [this message]
2025-12-18 6:36 ` Dongli Zhang
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=aUEXkDDOba+oZ4v+@intel.com \
--to=zhao1.liu@intel.com \
--cc=alexander.ivanov@virtuozzo.com \
--cc=babu.moger@amd.com \
--cc=dapeng1.mi@linux.intel.com \
--cc=davydov-max@yandex-team.ru \
--cc=den@virtuozzo.com \
--cc=dongli.zhang@oracle.com \
--cc=ewanhai-oc@zhaoxin.com \
--cc=ewanhai@zhaoxin.com \
--cc=groug@kaod.org \
--cc=joe.jin@oracle.com \
--cc=khorenko@virtuozzo.com \
--cc=kvm@vger.kernel.org \
--cc=like.xu.linux@gmail.com \
--cc=likexu@tencent.com \
--cc=mtosatti@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=sandipan.das@amd.com \
--cc=xiaoyao.li@intel.com \
/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