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;
MIME-Version: 1.0
References: <20201114023643.211313-1-namhyung@kernel.org> <20201115130524.GL894261@tassilo.jf.intel.com>
In-Reply-To: <20201115130524.GL894261@tassilo.jf.intel.com>
From:   Namhyung Kim <namhyung@kernel.org>
Date:   Tue, 17 Nov 2020 01:40:45 +0900
Message-ID: <CAM9d7cibrsM-Nv-QnFzxKsZ43+sUAm=vE-Caqy3usKjzQ2phug@mail.gmail.com>
Subject: Re: [PATCH] perf stat: Take cgroups into account for shadow stats
To:     Andi Kleen <ak@linux.intel.com>
Cc:     Arnaldo Carvalho de Melo <acme@kernel.org>,
        Jiri Olsa <jolsa@redhat.com>, Ingo Molnar <mingo@kernel.org>,
        Peter Zijlstra <peterz@infradead.org>,
        Mark Rutland <mark.rutland@arm.com>,
        Alexander Shishkin <alexander.shishkin@linux.intel.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Stephane Eranian <eranian@google.com>,
        Ian Rogers <irogers@google.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hi Andi,

On Sun, Nov 15, 2020 at 10:05 PM Andi Kleen <ak@linux.intel.com> wrote:
>
> > @@ -57,6 +59,9 @@ static int saved_value_cmp(struct rb_node *rb_node, const void *entry)
> >       if (a->ctx != b->ctx)
> >               return a->ctx - b->ctx;
> >
> > +     if (a->cgrp != b->cgrp)
> > +             return (char *)a->cgrp < (char *)b->cgrp ? -1 : +1;
>
> This means the sort order will depend on heap randomization,
> which will make it harder to debug.
>
> Better use something stable like the inode number of the cgroup.

I don't think they are used for sorting.  It's just to compare to find a
matching event value.

For heap randomization, we already used the same technique
for evsel and runtime_stat pointers.  I can make it use cgroup id
but not sure it is really worth it.

>
> Do we have the same problem with other filters?

I'm not aware of it.

>
> The rest of the patch looks good to me.

Thanks for the review!

Namhyung