[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 02 Feb 2009 09:19:23 +0100
From: Peter Zijlstra <a.p.zijlstra@...llo.nl>
To: Thomas Pilarski <thomas.pi@...or.de>
Cc: Mike Galbraith <efault@....de>,
Andrew Morton <akpm@...ux-foundation.org>,
Gregory Haskins <ghaskins@...ell.com>,
bugme-daemon@...zilla.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [Bugme-new] [Bug 12562] New: High overhead while switching or
synchronizing threads on different cores
On Mon, 2009-02-02 at 08:43 +0100, Thomas Pilarski wrote:
> Am Freitag, den 30.01.2009, 08:57 +0100 schrieb Mike Galbraith:
> > One of those "don't _ever_ do that" things?
>
> I did not known random() uses a system call. It's rather unrealistic to
> have five million system calls in a second. By adding a small loop with
> some calculations near the random, the problem disappears too.
> It is a unlucky chosen data generator.
I suppose you'll have to go bug the glibc people about their random()
implementation.
If you really need random() to perform for your application (monte-carlo
stuff?) You might be better off writing a PRNG with TLS state or
something.
--
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