From: "Liang, Kan" <kan.liang@linux.intel.com>
To: Arnaldo Carvalho de Melo <acme@kernel.org>,
Ian Rogers <irogers@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@redhat.com>,
Namhyung Kim <namhyung@kernel.org>,
Mark Rutland <mark.rutland@arm.com>,
Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Jiri Olsa <jolsa@kernel.org>,
Adrian Hunter <adrian.hunter@intel.com>,
Anne Macedo <retpolanne@posteo.net>,
Changbin Du <changbin.du@huawei.com>,
Andi Kleen <ak@linux.intel.com>,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org
Subject: Re: [PATCH v1 2/2] perf test: Add set of perf record LBR tests
Date: Thu, 8 Aug 2024 12:01:09 -0400 [thread overview]
Message-ID: <5af60713-48a4-4031-b6dc-3680f3a95f8d@linux.intel.com> (raw)
In-Reply-To: <ZrTXftup0H46R8WK@x1>
On 2024-08-08 10:34 a.m., Arnaldo Carvalho de Melo wrote:
> On Wed, Aug 07, 2024 at 10:46:44PM -0700, Ian Rogers wrote:
>> Adds coverage for LBR operations and LBR callgraph.
>>
>> Signed-off-by: Ian Rogers <irogers@google.com>
>> ---
>> tools/perf/tests/shell/record_lbr.sh | 161 +++++++++++++++++++++++++++
>> 1 file changed, 161 insertions(+)
>> create mode 100755 tools/perf/tests/shell/record_lbr.sh
>>
>> diff --git a/tools/perf/tests/shell/record_lbr.sh b/tools/perf/tests/shell/record_lbr.sh
>> new file mode 100755
>> index 000000000000..baf168d0ddbb
>> --- /dev/null
>> +++ b/tools/perf/tests/shell/record_lbr.sh
>> @@ -0,0 +1,161 @@
>> +#!/bin/bash
>> +# perf record LBR tests
>> +# SPDX-License-Identifier: GPL-2.0
>> +
>> +set -e
>> +
>> +if [ ! -f /sys/devices/cpu/caps/branches ]
>> +then
>> + echo "Skip: only x86 CPUs support LBR"
>> + exit 2
>> +fi
>
> root@x1:~# ls -la /sys/devices/cpu*/caps/branches
> -r--r--r--. 1 root root 4096 Jun 4 16:07 /sys/devices/cpu_atom/caps/branches
> -r--r--r--. 1 root root 4096 Aug 8 11:21 /sys/devices/cpu_core/caps/branches
> root@x1:~#
>
> I'm getting:
>
> root@x1:~# perf test -vvvv LBR
> 97: perf record LBR tests:
> --- start ---
> test child forked, pid 2033388
> Skip: only x86 CPUs support LBR
> ---- end(-2) ----
> 97: perf record LBR tests : Skip
> root@x1:~#
>
> So I added the following follow-up patch, ack?
Thanks Arnaldo. We need the test works on hybrid as well.
For both original patch and this patch,
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Thanks,
Kan
>
> - Arnaldo
>
> From f6ea332dd25cc62b9493b2d40040c2fb35253d9e Mon Sep 17 00:00:00 2001
> From: Arnaldo Carvalho de Melo <acme@redhat.com>
> Date: Thu, 8 Aug 2024 11:26:13 -0300
> Subject: [PATCH 1/1] perf test shell lbr: Support hybrid x86 systems too
>
> Running on a:
>
> root@x1:~# grep 'model name' -m1 /proc/cpuinfo
> model name : 13th Gen Intel(R) Core(TM) i7-1365U
> root@x1:~#
>
> It skips all the tests with:
>
> root@x1:~# perf test -vvvv LBR
> 97: perf record LBR tests:
> --- start ---
> test child forked, pid 2033388
> Skip: only x86 CPUs support LBR
> ---- end(-2) ----
> 97: perf record LBR tests : Skip
> root@x1:~#
>
> Because the test checks for the /sys/devices/cpu/caps/branches file,
> that isn't present as we have instead:
>
> root@x1:~# ls -la /sys/devices/cpu*/caps/branches
> -r--r--r--. 1 root root 4096 Aug 8 11:22 /sys/devices/cpu_atom/caps/branches
> -r--r--r--. 1 root root 4096 Aug 8 11:21 /sys/devices/cpu_core/caps/branches
> root@x1:~#
>
> If we check as well for one of those,
> /sys/devices/cpu_core/caps/branches, then we don't skip the tests and
> all are run on these x86 Intel Hybrid systems as well, passing all of
> them:
>
> root@x1:~# perf test -vvvv LBR
> 97: perf record LBR tests:
> --- start ---
> test child forked, pid 2034956
> LBR callgraph
> [ perf record: Woken up 5 times to write data ]
> [ perf record: Captured and wrote 1.812 MB /tmp/__perf_test.perf.data.B2HvQ (8114 samples) ]
> LBR callgraph [Success]
> LBR any branch test
> [ perf record: Woken up 25 times to write data ]
> [ perf record: Captured and wrote 6.382 MB /tmp/__perf_test.perf.data.B2HvQ (8071 samples) ]
> LBR any branch test: 8071 samples
> LBR any branch test [Success]
> LBR any call test
> [ perf record: Woken up 23 times to write data ]
> [ perf record: Captured and wrote 6.208 MB /tmp/__perf_test.perf.data.B2HvQ (8092 samples) ]
> LBR any call test: 8092 samples
> LBR any call test [Success]
> LBR any ret test
> [ perf record: Woken up 24 times to write data ]
> [ perf record: Captured and wrote 6.396 MB /tmp/__perf_test.perf.data.B2HvQ (8093 samples) ]
> LBR any ret test: 8093 samples
> LBR any ret test [Success]
> LBR any indirect call test
> [ perf record: Woken up 25 times to write data ]
> [ perf record: Captured and wrote 6.344 MB /tmp/__perf_test.perf.data.B2HvQ (8067 samples) ]
> LBR any indirect call test: 8067 samples
> LBR any indirect call test [Success]
> LBR any indirect jump test
> [ perf record: Woken up 12 times to write data ]
> [ perf record: Captured and wrote 3.073 MB /tmp/__perf_test.perf.data.B2HvQ (8061 samples) ]
> LBR any indirect jump test: 8061 samples
> LBR any indirect jump test [Success]
> LBR direct calls test
> [ perf record: Woken up 25 times to write data ]
> [ perf record: Captured and wrote 6.380 MB /tmp/__perf_test.perf.data.B2HvQ (8076 samples) ]
> LBR direct calls test: 8076 samples
> LBR direct calls test [Success]
> LBR any indirect user call test
> [ perf record: Woken up 5 times to write data ]
> [ perf record: Captured and wrote 1.597 MB /tmp/__perf_test.perf.data.B2HvQ (8079 samples) ]
> LBR any indirect user call test: 8079 samples
> LBR any indirect user call test [Success]
> LBR system wide any branch test
> [ perf record: Woken up 26 times to write data ]
> [ perf record: Captured and wrote 9.088 MB /tmp/__perf_test.perf.data.B2HvQ (9209 samples) ]
> LBR system wide any branch test: 9209 samples
> LBR system wide any branch test [Success]
> LBR system wide any call test
> [ perf record: Woken up 25 times to write data ]
> [ perf record: Captured and wrote 8.945 MB /tmp/__perf_test.perf.data.B2HvQ (9333 samples) ]
> LBR system wide any call test: 9333 samples
> LBR system wide any call test [Success]
> LBR parallel any branch test
> LBR parallel any call test
> LBR parallel any ret test
> LBR parallel any indirect call test
> LBR parallel any indirect jump test
> LBR parallel direct calls test
> LBR parallel system wide any branch test
> LBR parallel any indirect user call test
> LBR parallel system wide any call test
> [ perf record: Woken up 9 times to write data ]
> [ perf record: Woken up 51 times to write data ]
> [ perf record: Woken up 1 times to write data ]
> [ perf record: Woken up 5 times to write data ]
> [ perf record: Woken up 559 times to write data ]
> [ perf record: Woken up 14 times to write data ]
> [ perf record: Woken up 17 times to write data ]
> [ perf record: Woken up 1 times to write data ]
> [ perf record: Woken up 11 times to write data ]
> [ perf record: Captured and wrote 0.150 MB /tmp/__perf_test.perf.data.lANpR (1909 samples) ]
> [ perf record: Captured and wrote 2.371 MB /tmp/__perf_test.perf.data.Olum8 (3033 samples) ]
> [ perf record: Captured and wrote 1.230 MB /tmp/__perf_test.perf.data.njfJ8 (1742 samples) ]
> [ perf record: Captured and wrote 5.554 MB /tmp/__perf_test.perf.data.4ZTrj (29662 samples) ]
> [ perf record: Captured and wrote 19.906 MB /tmp/__perf_test.perf.data.dlGQt (29576 samples) ]
> [ perf record: Captured and wrote 0.289 MB /tmp/__perf_test.perf.data.CAT7y (4311 samples) ]
> [ perf record: Captured and wrote 3.129 MB /tmp/__perf_test.perf.data.diuKG (3971 samples) ]
> LBR parallel any indirect user call test: 1909 samples
> [ perf record: Captured and wrote 4.858 MB /tmp/__perf_test.perf.data.sVjtN (6130 samples) ]
> LBR parallel any indirect user call test [Success]
> [ perf record: Captured and wrote 3.669 MB /tmp/__perf_test.perf.data.AJtNI (4827 samples) ]
> LBR parallel any indirect jump test: 4311 samples
> LBR parallel any indirect jump test [Success]
> LBR parallel direct calls test: 3033 samples
> LBR parallel direct calls test [Success]
> LBR parallel any indirect call test: 1742 samples
> LBR parallel any indirect call test [Success]
> LBR parallel any call test: 4827 samples
> LBR parallel any call test [Success]
> LBR parallel any branch test: 6130 samples
> LBR parallel any branch test [Success]
> LBR parallel system wide any branch test: 29662 samples
> LBR parallel any ret test: 3971 samples
> LBR parallel any ret test [Success]
> LBR parallel system wide any branch test [Success]
> LBR parallel system wide any call test: 29576 samples
> LBR parallel system wide any call test [Success]
> ---- end(0) ----
> 97: perf record LBR tests : Ok
> root@x1:~#
>
> Cc: Adrian Hunter <adrian.hunter@intel.com>,
> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>,
> Cc: Andi Kleen <ak@linux.intel.com>
> Cc: Anne Macedo <retpolanne@posteo.net>,
> Cc: Changbin Du <changbin.du@huawei.com>,
> Cc: Ian Rogers <irogers@google.com>
> Cc: Ingo Molnar <mingo@redhat.com>,
> Cc: Jiri Olsa <jolsa@kernel.org>,
> Cc: Kan Liang <kan.liang@linux.intel.com>,
> Cc: Mark Rutland <mark.rutland@arm.com>,
> Cc: Namhyung Kim <namhyung@kernel.org>,
> Cc: Peter Zijlstra <peterz@infradead.org>
> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> ---
> tools/perf/tests/shell/record_lbr.sh | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tools/perf/tests/shell/record_lbr.sh b/tools/perf/tests/shell/record_lbr.sh
> index baf168d0ddbbf447..ead87b731898d70b 100755
> --- a/tools/perf/tests/shell/record_lbr.sh
> +++ b/tools/perf/tests/shell/record_lbr.sh
> @@ -4,7 +4,7 @@
>
> set -e
>
> -if [ ! -f /sys/devices/cpu/caps/branches ]
> +if [ ! -f /sys/devices/cpu/caps/branches -a ! -f /sys/devices/cpu_core/caps/branches ]
> then
> echo "Skip: only x86 CPUs support LBR"
> exit 2
next prev parent reply other threads:[~2024-08-08 16:01 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-08 5:46 [PATCH v1 1/2] perf callchain: Fix stitch LBR memory leaks Ian Rogers
2024-08-08 5:46 ` [PATCH v1 2/2] perf test: Add set of perf record LBR tests Ian Rogers
2024-08-08 14:34 ` Arnaldo Carvalho de Melo
2024-08-08 16:00 ` Ian Rogers
2024-08-08 20:28 ` Arnaldo Carvalho de Melo
2024-08-08 16:01 ` Liang, Kan [this message]
2024-08-08 15:59 ` [PATCH v1 1/2] perf callchain: Fix stitch LBR memory leaks Liang, Kan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5af60713-48a4-4031-b6dc-3680f3a95f8d@linux.intel.com \
--to=kan.liang@linux.intel.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=ak@linux.intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=changbin.du@huawei.com \
--cc=irogers@google.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mingo@redhat.com \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=retpolanne@posteo.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox