From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1755385AbdJITC1 (ORCPT <rfc822;w@1wt.eu>);
        Mon, 9 Oct 2017 15:02:27 -0400
Received: from usa-sjc-mx-foss1.foss.arm.com ([217.140.101.70]:34364 "EHLO
        foss.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1754210AbdJITCZ (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
        Mon, 9 Oct 2017 15:02:25 -0400
Date: Mon, 9 Oct 2017 20:02:28 +0100
From: Will Deacon <will.deacon@arm.com>
To: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>, linux-kernel@vger.kernel.org,
        jolsa@kernel.org, stable@vger.kernel.org,
        Adrian Hunter <adrian.hunter@intel.com>,
        Borislav Petkov <bp@alien8.de>, David Ahern <dsahern@gmail.com>,
        Namhyung Kim <namhyung@kernel.org>,
        Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: Re: [PATCH] perf tools: unbreak perf record for arm/arm64
Message-ID: <20171009190227.GG30828@arm.com>
References: <1507315102-5942-1-git-send-email-mark.rutland@arm.com>
 <20171009102159.GA6882@arm.com>
 <20171009190000.GL2121@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20171009190000.GL2121@redhat.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Mon, Oct 09, 2017 at 04:00:00PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Mon, Oct 09, 2017 at 11:22:00AM +0100, Will Deacon escreveu:
> > On Fri, Oct 06, 2017 at 07:38:22PM +0100, Mark Rutland wrote:
> > > Currently, perf record is broken on arm/arm64 systems when the PMU is
> > > specified explicitly as part of the event, e.g.
> > > 
> > > $ ./perf record -e armv8_cortex_a53/cpu_cycles/u true
> > > 
> > > In such cases, perf record fails to open events unless
> > > perf_event_paranoid is set to -1, even if the PMU in question supports
> > > mode exclusion. Further, even when perf_event_paranoid is toggled, no
> > > samples are recorded.
> > > 
> > > This is an unintended side effect of commit:
> > > 
> > >   e3ba76deef23064f ("perf tools: Force uncore events to system wide monitoring)
> > > 
> > > ... which assumes that if a PMU has an associated cpu_map, it is an
> > > uncore PMU, and forces events for such PMUs to be system-wide.
> > > 
> > > This is not true for arm/arm64 systems, which can have heterogeneous
> > > CPUs. To account for this, multiple CPU PMUs are exposed, each with a
> > > "cpus" field under sysfs, which the perf tool parses into a cpu_map. ARM
> > > PMUs do not have a "cpumask" file, and only have a "cpus" file. For the
> > > gory details as to why, see commit:
> > > 
> > >  7e3fcffe95544010 ("perf pmu: Support alternative sysfs cpumask")
> > > 
> > > Given all of this, we can instead identify uncore PMUs by explicitly
> > > checking for a "cpumask" file, and restore arm/arm64 PMU support back to
> > > a working state. This patch does so, adding a new perf_pmu::is_uncore
> > > field, and splitting the existing cpumask parsing so that it can be
> > > reused.
> > > 
> > > Signed-off-by: Mark Rutland <mark.rutland@arm.com>
> > > Fixes: e3ba76deef23064f ("perf tools: Force uncore events to system wide monitoring)
> > 
> > It sucks that we haven't noticed this being broken for so long, but I can
> > confirm that this fixes the issue:
> > 
> > Acked-by: Will Deacon <will.deacon@arm.com>
> > Tested-by Will Deacon <will.deacon@arm.com>
> > 
> > Any chance we can get this into 4.14? You'll probably need to do some stable
> > backports too, since this is a bit spread out.
> 
> Sure, I've added this to my perf/urgent branch, that, together with a
> few other fixes is undergoing testing now.

Thanks, Arnaldo.

Will