Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:7 as permitted sender) client-ip=2620:137:e000::3:7;
Date:   Mon, 6 Nov 2023 07:31:18 -0800
In-Reply-To: <CALMp9eS+kNYYK_1Ufy5vc5PK25q-ny20woxbHz1onStkcfWNVw@mail.gmail.com>
Mime-Version: 1.0
References: <20231104000239.367005-1-seanjc@google.com> <20231104000239.367005-3-seanjc@google.com>
 <CALMp9eS+kNYYK_1Ufy5vc5PK25q-ny20woxbHz1onStkcfWNVw@mail.gmail.com>
Message-ID: <ZUkGxqX8mJPPtxHD@google.com>
Subject: Re: [PATCH v6 02/20] KVM: x86/pmu: Don't enumerate support for fixed
 counters KVM can't virtualize
From:   Sean Christopherson <seanjc@google.com>
To:     Jim Mattson <jmattson@google.com>
Cc:     Paolo Bonzini <pbonzini@redhat.com>, kvm@vger.kernel.org,
        linux-kernel@vger.kernel.org,
        Kan Liang <kan.liang@linux.intel.com>,
        Dapeng Mi <dapeng1.mi@linux.intel.com>,
        Jinrong Liang <cloudliang@tencent.com>,
        Like Xu <likexu@tencent.com>,
        Aaron Lewis <aaronlewis@google.com>
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Sat, Nov 04, 2023, Jim Mattson wrote:
> On Fri, Nov 3, 2023 at 5:02=E2=80=AFPM Sean Christopherson <seanjc@google=
.com> wrote:
> >
> > Hide fixed counters for which perf is incapable of creating the associa=
ted
> > architectural event.  Except for the so called pseudo-architectural eve=
nt
> > for counting TSC reference cycle, KVM virtualizes fixed counters by
> > creating a perf event for the associated general purpose architectural
> > event.  If the associated event isn't supported in hardware, KVM can't
> > actually virtualize the fixed counter because perf will likely not prog=
ram
> > up the correct event.
>=20
> Won't it? My understanding was that perf preferred to use a fixed
> counter when there was a choice of fixed or general purpose counter.
> Unless the fixed counter is already assigned to a perf_event, KVM's
> request should be satisfied by assigning the fixed counter.
>=20
> > Note, this issue is almost certainly limited to running KVM on a funky
> > virtual CPU model, no known real hardware has an asymmetric PMU where a
> > fixed counter is supported but the associated architectural event is no=
t.
>=20
> This seems like a fix looking for a problem. Has the "problem"
> actually been encountered?

Heh, yes, I "encountered" the problem in a curated VM I created.  But I com=
pletely
agree that this is unnecessary, especially since odds are very, very good t=
hat
requesting the architectural general purpose encoding will still work.  E.g=
. in
my goofy setup, the underlying hardware does support the architectural even=
t and
so even if perf doesn't use the fixed counter for whatever reason, the GP c=
ounter
will still count the right event.