Date: Tue, 5 Jun 2007 21:08:06 +0200
From: Ingo Molnar <mingo@elte.hu>
To: Matt Mackall <mpm@selenic.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>, akpm@linux-foundation.org,
       Linux-kernel@vger.kernel.org
Subject: Re: Interesting interaction between lguest and CFS
Message-ID: <20070605190806.GA18503@elte.hu>
References: <20070604173710.GR11166@waste.org> <1181002734.25878.56.camel@localhost.localdomain> <1181010675.25878.140.camel@localhost.localdomain> <20070605041803.GO11115@waste.org> <20070605104353.GA7309@elte.hu>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20070605104353.GA7309@elte.hu>
User-Agent: Mutt/1.5.14 (2007-02-12)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 840
Lines: 23


* Ingo Molnar <mingo@elte.hu> wrote:

> Matt, could you send me your .config too? Maybe i can reproduce it 
> with your config.

ok, got it (off-list) and you seem to have the default HZ=250. Could you 
change it to HZ=1000 - does the delay get alot smaller?

if yes then this could be roughly the expected worst-case behavior for a 
nice+19 'signalling' task - i'd suggest to try the dont-use-renice patch 
Rusty sent (with the hang it causes fixed), or to change the nice(19) to 
nice(10).

(the rq_clock behavior still looks a bit anomalous, that perhaps 
contributes to the problem.)

	Ingo
-
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/