[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d7310ec0-4c4c-0d1c-5725-5377b539344a@redhat.com>
Date: Mon, 29 Mar 2021 11:39:51 +0200
From: Paolo Bonzini <pbonzini@...hat.com>
To: Vitaly Kuznetsov <vkuznets@...hat.com>, kvm@...r.kernel.org
Cc: Sean Christopherson <seanjc@...gle.com>,
Wanpeng Li <wanpengli@...cent.com>,
Jim Mattson <jmattson@...gle.com>,
Wei Huang <wei.huang2@....com>, Joerg Roedel <joro@...tes.org>,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] KVM: x86/vPMU: Forbid writing to MSR_F15H_PERF MSRs when
guest doesn't have X86_FEATURE_PERFCTR_CORE
On 29/03/21 10:52, Vitaly Kuznetsov wrote:
> Paolo Bonzini <pbonzini@...hat.com> writes:
>
>> On 23/03/21 09:45, Vitaly Kuznetsov wrote:
>>> MSR_F15H_PERF_CTL0-5, MSR_F15H_PERF_CTR0-5 MSRs are only available when
>>> X86_FEATURE_PERFCTR_CORE CPUID bit was exposed to the guest. KVM, however,
>>> allows these MSRs unconditionally because kvm_pmu_is_valid_msr() ->
>>> amd_msr_idx_to_pmc() check always passes and because kvm_pmu_set_msr() ->
>>> amd_pmu_set_msr() doesn't fail.
>>>
>>> In case of a counter (CTRn), no big harm is done as we only increase
>>> internal PMC's value but in case of an eventsel (CTLn), we go deep into
>>> perf internals with a non-existing counter.
>>>
>>> Note, kvm_get_msr_common() just returns '0' when these MSRs don't exist
>>> and this also seems to contradict architectural behavior which is #GP
>>> (I did check one old Opteron host) but changing this status quo is a bit
>>> scarier.
>>
>> Hmm, since these do have a cpuid bit it may not be that scary.
>
> Well, if you're not scared I can send a patch)
Go ahead.
Paolo
Powered by blists - more mailing lists