linux-kernel - Re: [PATCH v5 7/7] perf test: Add a test for default perf stat command

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <ZtX5Mw8Iuc5qgBWz@google.com>
Date: Mon, 2 Sep 2024 10:43:15 -0700
From: Namhyung Kim <namhyung@...nel.org>
To: James Clark <james.clark@...aro.org>
Cc: irogers@...gle.com, linux-perf-users@...r.kernel.org,
	kan.liang@...ux.intel.com, ak@...ux.intel.com,
	John Garry <john.g.garry@...cle.com>, Will Deacon <will@...nel.org>,
	Mike Leach <mike.leach@...aro.org>, Leo Yan <leo.yan@...ux.dev>,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...hat.com>,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Mark Rutland <mark.rutland@....com>,
	Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
	Jiri Olsa <jolsa@...nel.org>,
	Adrian Hunter <adrian.hunter@...el.com>,
	Weilin Wang <weilin.wang@...el.com>,
	Dominique Martinet <asmadeus@...ewreck.org>,
	Athira Rajeev <atrajeev@...ux.vnet.ibm.com>,
	Colin Ian King <colin.i.king@...il.com>,
	Yang Jihong <yangjihong@...edance.com>,
	Ze Gao <zegao2021@...il.com>, Sun Haiyong <sunhaiyong@...ngson.cn>,
	Jing Zhang <renyu.zj@...ux.alibaba.com>,
	Yicong Yang <yangyicong@...ilicon.com>,
	linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v5 7/7] perf test: Add a test for default perf stat
 command

On Fri, Aug 30, 2024 at 09:45:11AM +0100, James Clark wrote:
> 
> 
> On 30/08/2024 9:29 am, James Clark wrote:
> > 
> > 
> > On 30/08/2024 7:32 am, Namhyung Kim wrote:
> > > Hello,
> > > 
> > > On Wed, Aug 28, 2024 at 03:07:21PM +0100, James Clark wrote:
> > > > Test that one cycles event is opened for each core PMU when "perf stat"
> > > > is run without arguments.
> > > > 
> > > > The event line can either be output as "pmu/cycles/" or just "cycles" if
> > > > there is only one PMU. Include 2 spaces for padding in the one PMU case
> > > > to avoid matching when the word cycles is included in metric
> > > > descriptions.
> > > > 
> > > > Signed-off-by: James Clark <james.clark@...aro.org>
> > > > ---
> > > >   tools/perf/tests/shell/stat.sh | 21 +++++++++++++++++++++
> > > >   1 file changed, 21 insertions(+)
> > > > 
> > > > diff --git a/tools/perf/tests/shell/stat.sh
> > > > b/tools/perf/tests/shell/stat.sh
> > > > index 525d0c44fdc6..24ace1de71cc 100755
> > > > --- a/tools/perf/tests/shell/stat.sh
> > > > +++ b/tools/perf/tests/shell/stat.sh
> > > > @@ -148,6 +148,26 @@ test_cputype() {
> > > >     echo "cputype test [Success]"
> > > >   }
> > > > +test_hybrid() {
> > > > +  # Test the default stat command on hybrid devices opens one
> > > > cycles event for
> > > > +  # each CPU type.
> > > > +  echo "hybrid test"
> > > > +
> > > > +  # Count the number of core PMUs
> > > > +  pmus=$(ls /sys/bus/event_source/devices/*/cpus 2>/dev/null | wc -l)
> > > 
> > > Is it working on non-hybrid systems?  I don't think they have cpus file
> > > in the core PMU.
> > > 
> > > Thanks,
> > > Namhyung
> > > 
> > 
> > Good point I only tested on Arm non-hybrid. I can change it to assume 1
> > PMU for no cpus files?
> 
> Or maybe assume 1 if a /sys/bus/event_source/devices/cpu folder exists? Not
> sure which is best but either will work.

Some arch might not have a cpu PMU, I think we can assume 1 and update
only if it finds cpus files.

Thanks,
Namhyung