Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Fri, 15 Jan 2021 09:57:41 -0800
From:   Sean Christopherson <seanjc@google.com>
To:     "Xu, Like" <like.xu@intel.com>
Cc:     Andi Kleen <andi@firstfloor.org>,
        Kan Liang <kan.liang@linux.intel.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Paolo Bonzini <pbonzini@redhat.com>, eranian@google.com,
        kvm@vger.kernel.org, Ingo Molnar <mingo@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Jim Mattson <jmattson@google.com>,
        Joerg Roedel <joro@8bytes.org>, wei.w.wang@intel.com,
        luwei.kang@intel.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 00/17] KVM: x86/pmu: Add support to enable Guest PEBS
 via DS
Message-ID: <YAHXlWmeR9p6JZm2@google.com>
References: <20210104131542.495413-1-like.xu@linux.intel.com>
 <YACXQwBPI8OFV1T+@google.com>
 <f8a8e4e2-e0b1-8e68-81d4-044fb62045d5@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <f8a8e4e2-e0b1-8e68-81d4-044fb62045d5@intel.com>
Precedence: bulk

On Fri, Jan 15, 2021, Xu, Like wrote:
> Hi Sean,
> 
> Thanks for your comments !
> 
> On 2021/1/15 3:10, Sean Christopherson wrote:
> > On Mon, Jan 04, 2021, Like Xu wrote:
> > > 2) Slow path (part 3, patch 0012-0017)
> > > 
> > > This is when the host assigned physical PMC has a different index
> > > from the virtual PMC (e.g. using physical PMC1 to emulate virtual PMC0)
> > > In this case, KVM needs to rewrite the PEBS records to change the
> > > applicable counter indexes to the virtual PMC indexes, which would
> > > otherwise contain the physical counter index written by PEBS facility,
> > > and switch the counter reset values to the offset corresponding to
> > > the physical counter indexes in the DS data structure.
> > > 
> > > Large PEBS needs to be disabled by KVM rewriting the
> > > pebs_interrupt_threshold filed in DS to only one record in
> > > the slow path.  This is because a guest may implicitly drain PEBS buffer,
> > > e.g., context switch. KVM doesn't get a chance to update the PEBS buffer.
> > Are the PEBS record write, PEBS index update, and subsequent PMI atomic with
> > respect to instruction execution?  If not, doesn't this approach still leave a
> > window where the guest could see the wrong counter?
> 
> First, KVM would limit/rewrite guest DS pebs_interrupt_threshold to one
> record before vm-entry,
> (see patch [PATCH v3 14/17] KVM: vmx/pmu: Limit pebs_interrupt_threshold in
> the guest DS area)
> which means once a PEBS record is written into the guest pebs buffer,
> a PEBS PMI will be generated immediately and thus vm-exit.

I'm asking about ucode/hardare.  Is the "guest pebs buffer write -> PEBS PMI"
guaranteed to be atomic?

In practice, under what scenarios will guest counters get cross-mapped?  And,
how does this support affect guest accuracy?  I.e. how bad do things get for the
guest if we simply disable guest counters if they can't have a 1:1 association
with their physical counter?