[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4EEA45C6.2020106@fb.com>
Date: Thu, 15 Dec 2011 11:08:54 -0800
From: Arun Sharma <asharma@...com>
To: <avagin@...il.com>
CC: <avagin@...nvz.org>, Frederic Weisbecker <fweisbec@...il.com>,
Peter Zijlstra <a.p.zijlstra@...llo.nl>,
<linux-kernel@...r.kernel.org>,
Steven Rostedt <rostedt@...dmis.org>,
Ingo Molnar <mingo@...hat.com>,
Paul Mackerras <paulus@...ba.org>,
Arnaldo Carvalho de Melo <acme@...stprotocols.net>
Subject: Re: [PATCH 3/4] trace: add ability to collect call chain of non-current
task.
On 12/15/11 7:19 AM, Andrew Vagin wrote:
>>> * It doesn't distinguish blocking time and sleeping time.
>>
>> Can't we distinguish based on next->state?
> It will be "Running". We can distinguish based on block_start of
> sleep_start
Sounds good.
>>
>>> * This patch does a bit mess between subsystems...
>>
>> Which subsystems?
> sched_switch belongs to scheduler.
> sched_stat belongs to sched_fair.
>
My reading of the code is that 'struct sched_statistics' is independent
of the scheduling class. But only sched_fair.c is actually populating
the stats. Other classes could populate those fields if they choose to.
> It works for me. You can find output of commands bellow.
>
> # ./perf record -e sched:sched_switch -agP
Thanks. I forgot to use the -P option you added on my cmdline.
I still need to investigate why per-process mode (command line without
-a) doesn't work. Will send an updated patch once that's fixed.
-Arun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists