linux-kernel - Re: [PATCH v4 0/5] KVM: x86: Intel LBR related perf cleanups

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] [day] [month] [year] [list]

Message-ID: <YzsaOMculnYH1d5l@google.com>
Date:   Mon, 3 Oct 2022 17:22:00 +0000
From:   Sean Christopherson <seanjc@...gle.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Ingo Molnar <mingo@...hat.com>,
        Arnaldo Carvalho de Melo <acme@...nel.org>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Mark Rutland <mark.rutland@....com>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Jiri Olsa <jolsa@...nel.org>,
        Namhyung Kim <namhyung@...nel.org>,
        linux-perf-users@...r.kernel.org, linux-kernel@...r.kernel.org,
        kvm@...r.kernel.org, Like Xu <like.xu.linux@...il.com>
Subject: Re: [PATCH v4 0/5]  KVM: x86: Intel LBR related perf cleanups

On Sat, Sep 17, 2022, Sean Christopherson wrote:
> On Thu, Sep 08, 2022, Peter Zijlstra wrote:
> > On Thu, Sep 01, 2022 at 05:32:53PM +0000, Sean Christopherson wrote:
> > 
> > > Sean Christopherson (5):
> > >   perf/x86/core: Remove unnecessary stubs provided for KVM-only helpers
> > >   perf/x86/core: Drop the unnecessary return value from
> > >     x86_perf_get_lbr()
> > >   KVM: VMX: Move vmx_get_perf_capabilities() definition to vmx.c
> > >   KVM: VMX: Fold vmx_supported_debugctl() into vcpu_supported_debugctl()
> > >   KVM: VMX: Advertise PMU LBRs if and only if perf supports LBRs
> > 
> > These look good to me; how do you want this routed, if through the KVM
> > tree:
> > 
> > Acked-by: Peter Zijlstra (Intel) <peterz@...radead.org>
> 
> Thanks!  If you don't anticipate conflicts in the perf headers, I'll take 'em
> through KVM, patch 01 introduced a new warning that I need to resolve (hopefully
> it doesn't throw a wrench into things).

Rats, patch 01 is flat out wrong.  The stubs for Intel and AMD are necessary
because KVM_{AMD,INTEL} don't strictly require CPU_SUP_{AMD/INTEL}.  KVM_AMD
doesn't have any CPU_SUP_* requirement (which probably should be fixed), and
KVM_INTEL effectively require INTEL || CENTAUR || ZHAOXIN.

x86_perf_get_lbr() can still be cleaned up to fix KVM's benign bug of not checking
the result by zeroing the structure when LBRs are unsupported.

v5 incoming...