From: Arnd Bergmann <arnd@arndb.de>
To: matthieu castet <castet.matthieu@free.fr>
Subject: Re: msleep_interruptible vs msleep
Date: Mon, 4 Sep 2006 21:38:26 +0200
User-Agent: KMail/1.9.1
Cc: Linux Kernel list <linux-kernel@vger.kernel.org>
References: <44FC7EAE.6020300@free.fr>
In-Reply-To: <44FC7EAE.6020300@free.fr>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Message-Id: <200609042138.26603.arnd@arndb.de>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 910
Lines: 31

Am Monday 04 September 2006 21:29 schrieb matthieu castet:
>
> But why if I have a kernel thread that do [1] :
>
> while (true) {
> Do some stuff
> msleep(1000)
> }
>
> the load average is high (near 100%).
>
> and if I use msleep_interruptible the load average is normal.

These are the traditional semantics of incorruptible vs. noninterruptible
sleep. A process that sleep noninterruptible contributes to the load
average but does not consume actual CPU cycles.

I guess you can take that as a hint that the code you're describing
above is a bad thing to do.

> Does the same applies to wait_event_timeout vs
> wait_event_interruptible_timeout ?

yes.

	Arnd <><
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/