KVM: x86/pmu: Update AMD PMC sample period to fix guest NMI-watchdog
authorLike Xu <likexu@tencent.com>
Sat, 9 Apr 2022 01:52:26 +0000 (09:52 +0800)
committerPaolo Bonzini <pbonzini@redhat.com>
Thu, 21 Apr 2022 17:16:14 +0000 (13:16 -0400)
commit75189d1de1b377e580ebd2d2c55914631eac9c64
tree411e9a78ba15411a89c4529c6c7046aafa1bb4d1
parent0361bdfddca20c8855ea3bdbbbc9c999912b10ff
KVM: x86/pmu: Update AMD PMC sample period to fix guest NMI-watchdog

NMI-watchdog is one of the favorite features of kernel developers,
but it does not work in AMD guest even with vPMU enabled and worse,
the system misrepresents this capability via /proc.

This is a PMC emulation error. KVM does not pass the latest valid
value to perf_event in time when guest NMI-watchdog is running, thus
the perf_event corresponding to the watchdog counter will enter the
old state at some point after the first guest NMI injection, forcing
the hardware register PMC0 to be constantly written to 0x800000000001.

Meanwhile, the running counter should accurately reflect its new value
based on the latest coordinated pmc->counter (from vPMC's point of view)
rather than the value written directly by the guest.

Fixes: 168d918f2643 ("KVM: x86: Adjust counter sample period after a wrmsr")
Reported-by: Dongli Cao <caodongli@kingsoft.com>
Signed-off-by: Like Xu <likexu@tencent.com>
Reviewed-by: Yanan Wang <wangyanan55@huawei.com>
Tested-by: Yanan Wang <wangyanan55@huawei.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Message-Id: <20220409015226.38619-1-likexu@tencent.com>
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
arch/x86/kvm/pmu.h
arch/x86/kvm/svm/pmu.c
arch/x86/kvm/vmx/pmu_intel.c