From: "Wangnan (F)" <wangnan0@huawei.com>
To: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: He Kuang <hekuang@huawei.com>, Jiri Olsa <jolsa@kernel.org>,
"Linux Kernel Mailing List" <linux-kernel@vger.kernel.org>
Subject: Re: 'perf test backward' failing with older kernels, should Skip
Date: Fri, 1 Jul 2016 09:53:56 +0800 [thread overview]
Message-ID: <5775CD34.6030107@huawei.com> (raw)
In-Reply-To: <20160630210052.GE5324@kernel.org>
On 2016/7/1 5:00, Arnaldo Carvalho de Melo wrote:
> Hi Wang,
>
> I booted a machine with the f24 distro kernel and the backward
> ring buffer test started FAILing, when it should've noticed that the
> kernel doesn't support it and Skip instead:
>
> [root@jouet ~]# perf test -v backward
> 45: Test backward reading from ring buffer :
> --- start ---
> <SNIP>
> mmap size 1052672B
> Unexpected counter: sample_count=0, comm_count=0
> ---- end ----
> Test backward reading from ring buffer: FAILED!
> [root@jouet ~]#
>
I didn't test it on old kernel, but I tested it by forcing
sys_perf_event_open
return -EINVAL when attr->backward is true by following patch:
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 1d8f2bb..3e91c63 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -1418,6 +1418,9 @@ retry_open:
pr_debug2("sys_perf_event_open: pid %d cpu %d
group_fd %d flags %#lx\n",
pid, cpus->map[cpu], group_fd, flags);
+ if (evsel->attr.write_backward)
+ FD(evsel, cpu, thread) = -EINVAL;
+ else
FD(evsel, cpu, thread) =
sys_perf_event_open(&evsel->attr,
pid,
cpus->map[cpu],
'perf test' skips this test.
Let me switch to an old kernel and try again...
Oops... It does FAILED...
Seems the code I use to emulate old kernel is wrong. I should set errno
to EINVAL and
make FD(...) = -1.
I agree with your patch. Thank you for pointing it out.
> Ok, so I think the patch below should do the trick, is that right? With it
> I get:
>
> [root@jouet ~]# perf test backward
> 45: Test backward reading from ring buffer : Skip
> [root@jouet ~]#
>
> diff --git a/tools/perf/tests/backward-ring-buffer.c b/tools/perf/tests/backward-ring-buffer.c
> index e70313fac5a5..0731b0bf9552 100644
> --- a/tools/perf/tests/backward-ring-buffer.c
> +++ b/tools/perf/tests/backward-ring-buffer.c
> @@ -118,8 +118,10 @@ int test__backward_ring_buffer(int subtest __maybe_unused)
> perf_evlist__config(evlist, &opts, NULL);
>
> /* Set backward bit, ring buffer should be writing from end */
> - evlist__for_each_entry(evlist, evsel)
> + evlist__for_each_entry(evlist, evsel) {
> + evsel->overwrite = true;
> evsel->attr.write_backward = 1;
> + }
>
> err = perf_evlist__open(evlist);
> if (err < 0) {
prev parent reply other threads:[~2016-07-01 1:55 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-30 21:00 'perf test backward' failing with older kernels, should Skip Arnaldo Carvalho de Melo
2016-07-01 1:53 ` Wangnan (F) [this message]
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=5775CD34.6030107@huawei.com \
--to=wangnan0@huawei.com \
--cc=acme@kernel.org \
--cc=hekuang@huawei.com \
--cc=jolsa@kernel.org \
--cc=linux-kernel@vger.kernel.org \
/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