linux-kernel - Re: [PATCH 1/3] KVM: x86: Refresh PMU after writes to MSR_IA32_PERF

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 3 Aug 2022 18:47:46 +0000
From:   Sean Christopherson <seanjc@...gle.com>
To:     Like Xu <like.xu.linux@...il.com>
Cc:     kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
        Paolo Bonzini <pbonzini@...hat.com>
Subject: Re: [PATCH 1/3] KVM: x86: Refresh PMU after writes to
 MSR_IA32_PERF_CAPABILITIES

On Wed, Aug 03, 2022, Sean Christopherson wrote:
> On Wed, Aug 03, 2022, Like Xu wrote:
> > Now, all the dots have been connected. As punishment, I'd like to cook this
> > patch set more with trackable tests so that you have more time for other
> > things that are not housekeeping.
> 
> Let me post v2, I've already done all the work and testing.  If there's more to
> be done, we can figure out next steps from there.

I partially take that back, I forgot about the "disallow writes to feature MSRs
after KVM_RUN".  I haven't addressed that and don't have bandwidth to work on it
for the foreseeable future.  If you can address the issue, that'd be awesome.

I'm still going to post v2, all of the PMU fixes/cleanups are valid regardless of
the KVM_RUN issue, i.e. they can go in sooner to fix slightly less theoretical
problems (I doubt there's a userspace that actually changes PERF_CAPABILITIES
after KVM_RUN).  I'll note in the changelog that KVM should disallow changing
feature MSRs after KVM_RUN.