Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20181024151116.30935-1-kan.liang@linux.intel.com>
 <CABPqkBQEp9SLQuAonW7-c9HzH4bSgUyJGJ5ZZNXHy-0phHvvZQ@mail.gmail.com>
 <20181025002312.GS3109@worktop.c.hoisthospitality.com> <CABPqkBR1NTcpD3WxFsk4XZ6mT+jteec3kZBgBTxR5gFUL9Zffw@mail.gmail.com>
 <20181025003444.GV3109@worktop.c.hoisthospitality.com>
In-Reply-To: <20181025003444.GV3109@worktop.c.hoisthospitality.com>
From:   Stephane Eranian <eranian@google.com>
Date:   Wed, 24 Oct 2018 17:44:09 -0700
Message-ID: <CABPqkBQYbQV86dWCHMZX0sMspkHXSpYOB6G+c2kSOf3d4R7dNQ@mail.gmail.com>
Subject: Re: [PATCH 1/2] perf: Add munmap callback
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     "Liang, Kan" <kan.liang@linux.intel.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>,
        Arnaldo Carvalho de Melo <acme@kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Borislav Petkov <bp@alien8.de>, Andi Kleen <ak@linux.intel.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Oct 24, 2018 at 5:34 PM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Wed, Oct 24, 2018 at 05:25:59PM -0700, Stephane Eranian wrote:
> > On Wed, Oct 24, 2018 at 5:23 PM Peter Zijlstra <peterz@infradead.org> wrote:
>
> > > That is actually a different problem. And you're right, we never did fix
> > > that.
> > >
> > it is a different problem but the solution is the same: PERF_RECORD_UNMAP!
>
> But he's not actually doing PERF_RECORD_UNMAP, nor is his perf_munmap()
> hook sufficient to actually generate those.
>
You're right. But I saw that and thought, this is going in the right
direction: tracking munmap.
I agree with you that we should use this opportunity to track unmap
for his purpose but also
the issue I raised, which needs a new record type based on the new unmap hook.

> Now I agree that if he's going to do an munmap hook, he should do it
> 'right' and at the very least allow for PERF_RECORD_UNMAP to be done,
> but ideally simply pick up and finish that patch we had back then.
Agreed.