Message-Id: <200602141831.k1EIVog26783@unix-os.sc.intel.com>
From: "Chen, Kenneth W" <kenneth.w.chen@intel.com>
To: "'Ingo Molnar'" <mingo@elte.hu>, "Andrew Morton" <akpm@osdl.org>
Cc: <nickpiggin@yahoo.com.au>, <linux-kernel@vger.kernel.org>
Subject: RE: [patch 0/2] fix perf. bug in wake-up load balancing for aim7 and db workload
Date: Tue, 14 Feb 2006 10:31:49 -0800
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
Thread-Index: AcYxZTiFr/lqF8LPSFaf+2LOMmjbwAALsM6w
In-Reply-To: <20060214124743.GA5586@elte.hu>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1146
Lines: 26

Ingo Molnar wrote on Tuesday, February 14, 2006 4:48 AM
> * Andrew Morton <akpm@osdl.org> wrote:
> > Ingo, what's your plan here?
> 
> I really dont like the sysctl hack. Firstly, which precise kernel 
> version was tested - do we know that it wasnt e.g. the smpnice 
> regression interfering? Secondly, i dont like the sysctl concept itself: 
> i really think we should try to find a way for _applications_ to be 
> woken up according to their workload.


Ingo, you have tried to come up with an algorithm, Nick tried equally,
and I have tried.  None of the "smart algorithms" works right now.  The
workload requirements are on two opposite end of spectrum.  If you do
one thing, it will hurt the other and vice versa.  I don't see a solution
other than a sysctl.  If you have implementation idea, I would be more
than happy to code it up and test. But I think we are really running dry
here.

- Ken

-
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/