[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4F3BEB9F.2090502@gmail.com>
Date: Wed, 15 Feb 2012 10:30:07 -0700
From: David Ahern <dsahern@...il.com>
To: Luigi Semenzato <semenzato@...omium.org>
CC: Peter Zijlstra <a.p.zijlstra@...llo.nl>,
Alexander Viro <viro@...iv.linux.org.uk>,
Paul Mackerras <paulus@...ba.org>, Ingo Molnar <mingo@...e.hu>,
Arnaldo Carvalho de Melo <acme@...stprotocols.net>,
Andrew Morton <akpm@...ux-foundation.org>,
Vasiliy Kulikov <segoon@...nwall.com>,
Stephen Wilson <wilsons@...rt.ca>,
Oleg Nesterov <oleg@...hat.com>, Tejun Heo <tj@...nel.org>,
Paul Gortmaker <paul.gortmaker@...driver.com>,
Andi Kleen <ak@...ux.intel.com>,
Lucas De Marchi <lucas.demarchi@...fusion.mobi>,
Greg Kroah-Hartman <gregkh@...e.de>,
"Eric W. Biederman" <ebiederm@...ssion.com>,
"Rafael J. Wysocki" <rjw@...k.pl>,
Frederic Weisbecker <fweisbec@...il.com>,
Namhyung Kim <namhyung@...il.com>,
Robert Richter <robert.richter@....com>,
linux-kernel@...r.kernel.org, sonnyrao@...omium.org,
olofj@...omium.org, eranian@...gle.com
Subject: Re: [PATCH] Perf: bug fix: distinguish between rename and exec
On 2/15/12 10:22 AM, Luigi Semenzato wrote:
>
> Any process that makes a prctl(PR_SET_NAME) call loses its mappings,
> no matter when makes the call. The perf records for that process look
> like this:
>
> COMM (for the initial exec)
> MMAP (the executable)
> MMAP (1st dll)
> MMAP (2nd dll)
> ...
> COMM (for the prctl)
>
> The second COMM flushes the old mappings, and all samples from then on
> cannot be classified. This is easily reproducible with a small
> program, which I would be happy to send.
Ok, I see that now.
Thanks,
David
>
> I found this while trying to use perf with Chrome on Chrome OS.
> Chrome forks and execs all the time, and calls prctl() in the thread
> library.
--
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