From: Jiri Olsa <jolsa@redhat.com>
To: Andi Kleen <andi@firstfloor.org>
Cc: acme@infradead.org, mingo@kernel.org,
linux-kernel@vger.kernel.org, peterz@infradead.org,
eranian@google.com, namhyung@kernel.org,
Andi Kleen <ak@linux.intel.com>,
fweisbec@gmail.com
Subject: Re: [PATCH 7/8] perf, tools, record: Always allow to overide default period
Date: Tue, 11 Mar 2014 15:26:30 +0100 [thread overview]
Message-ID: <20140311142630.GH22678@krava.redhat.com> (raw)
In-Reply-To: <1394048978-15909-8-git-send-email-andi@firstfloor.org>
On Wed, Mar 05, 2014 at 11:49:37AM -0800, Andi Kleen wrote:
> From: Andi Kleen <ak@linux.intel.com>
>
> Fix the logic to allow overriding event default periods with -c or -F
> on the command line. I'm not sure what the previous if was supposed
> to do, it didn't work and seemed bogus. I just remove the extra
> check and unconditionally allow overiding when the user set
> an option.
>
> This fixed specifying -c / -F with json event list events,
> which have a default period. It should do the same
> for trace point events.
>
> Cc: fweisbec@gmail.com
> Signed-off-by: Andi Kleen <ak@linux.intel.com>
> ---
> tools/perf/util/evsel.c | 18 +++++++-----------
> 1 file changed, 7 insertions(+), 11 deletions(-)
>
> diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
> index adc94dd..ca4b976 100644
> --- a/tools/perf/util/evsel.c
> +++ b/tools/perf/util/evsel.c
> @@ -561,19 +561,15 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)
> }
>
> /*
> - * We default some events to a 1 default interval. But keep
> - * it a weak assumption overridable by the user.
> + * Let the user override any default periods.
> */
> - if (!attr->sample_period || (opts->user_freq != UINT_MAX &&
> - opts->user_interval != ULLONG_MAX)) {
> - if (opts->freq) {
> - perf_evsel__set_sample_bit(evsel, PERIOD);
> - attr->freq = 1;
> - attr->sample_freq = opts->freq;
> - } else {
> - attr->sample_period = opts->default_interval;
> - }
> + if (opts->user_freq != UINT_MAX) {
> + perf_evsel__set_sample_bit(evsel, PERIOD);
> + attr->freq = 1;
> + attr->sample_freq = opts->freq;
> }
> + if (opts->user_interval != ULLONG_MAX)
> + attr->sample_period = opts->default_interval;
and this also makes the default frequency not be set ;-)
$ ./perf record ls
...
$ ./perf report -D | grep SAMPLE
$
and 'perf test attr' fails for this reason:
$ ./perf test attr
13: struct perf_event_attr setup :FAILED './tests/attr/test-record-branch-filter-k' - match failure
FAILED!
jirka
next prev parent reply other threads:[~2014-03-11 14:27 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-03-05 19:49 perf: Add support for full Intel event lists Andi Kleen
2014-03-05 19:49 ` [PATCH 1/8] perf, tools: Add jsmn `jasmine' JSON parser Andi Kleen
2014-03-11 10:23 ` Jiri Olsa
2014-03-11 13:37 ` Jiri Olsa
2014-03-11 13:38 ` Jiri Olsa
2014-03-13 20:06 ` Andi Kleen
2014-03-11 13:38 ` Jiri Olsa
2014-03-13 20:07 ` Andi Kleen
2014-03-11 13:38 ` Jiri Olsa
2014-03-05 19:49 ` [PATCH 2/8] perf, tools: Add support for reading JSON event files Andi Kleen
2014-03-11 13:37 ` Jiri Olsa
2014-03-05 19:49 ` [PATCH 3/8] perf, tools: Automatically look for event file name for cpu Andi Kleen
2014-03-11 13:37 ` Jiri Olsa
2014-03-05 19:49 ` [PATCH 4/8] perf, tools: Add perf download to download event files Andi Kleen
2014-03-05 19:49 ` [PATCH 5/8] perf, tools: Allow events with dot Andi Kleen
2014-03-05 19:49 ` [PATCH 6/8] perf, tools: Query terminal width and use in perf list Andi Kleen
2014-03-11 13:41 ` Jiri Olsa
2014-03-05 19:49 ` [PATCH 7/8] perf, tools, record: Always allow to overide default period Andi Kleen
2014-03-11 14:26 ` Jiri Olsa [this message]
2014-03-05 19:49 ` [PATCH 8/8] perf, tools, test: Add test case for alias and JSON parsing Andi Kleen
2014-03-11 14:31 ` Jiri Olsa
2014-03-14 21:12 ` Andi Kleen
2014-03-06 9:37 ` perf: Add support for full Intel event lists Jiri Olsa
2014-03-10 19:39 ` Ben Hutchings
2014-03-10 22:14 ` Andi Kleen
2014-03-10 23:13 ` Ben Hutchings
2014-03-10 23:31 ` Andi Kleen
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=20140311142630.GH22678@krava.redhat.com \
--to=jolsa@redhat.com \
--cc=acme@infradead.org \
--cc=ak@linux.intel.com \
--cc=andi@firstfloor.org \
--cc=eranian@google.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.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