From: Con Kolivas <kernel@kolivas.org>
To: Nick Piggin <npiggin@suse.de>
Subject: Re: [rfc][patch] sched: remove smpnice
Date: Wed, 8 Feb 2006 01:57:06 +1100
User-Agent: KMail/1.9.1
Cc: Andrew Morton <akpm@osdl.org>, Ingo Molnar <mingo@elte.hu>,
       Steven Rostedt <rostedt@goodmis.org>,
       Peter Williams <pwil3058@bigpond.net.au>,
       Suresh B <suresh.b.siddha@intel.com>,
       Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
       Linus Torvalds <torvalds@osdl.org>
References: <20060207142828.GA20930@wotan.suse.de>
In-Reply-To: <20060207142828.GA20930@wotan.suse.de>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Message-Id: <200602080157.07823.kernel@kolivas.org>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2053
Lines: 45

On Wednesday 08 February 2006 01:28, Nick Piggin wrote:
> I'd like to get some comments on removing smpnice for 2.6.16. I don't
> think the code is quite ready, which is why I asked for Peter's additions
> to also be merged before I acked it (although it turned out that it still
> isn't quite ready with his additions either).
>
> Basically I have had similar observations to Suresh in that it does not
> play nicely with the rest of the balancing infrastructure (and raised
> similar concerns in my review).
>
> The samples (group of 4) I got for "maximum recorded imbalance" on a 2x2
>
> SMP+HT Xeon are as follows:
>            | Following boot | hackbench 20        | hackbench 40
>
> -----------+----------------+---------------------+---------------------
> 2.6.16-rc2 | 30,37,100,112  | 5600,5530,6020,6090 | 6390,7090,8760,8470
> +nosmpnice |  3, 2,  4,  2  |   28, 150, 294, 132 |  348, 348, 294, 347
>
> Hackbench raw performance is down around 15% with smpnice (but that in
> itself isn't a huge deal because it is just a benchmark). However, the
> samples show that the imbalance passed into move_tasks is increased by
> about a factor of 10-30. I think this would also go some way to
> explaining latency blips turning up in the balancing code (though I
> haven't actually measured that).
>
> We'll probably have to revert this in the SUSE kernel.
>
> The other option for 2.6.16 would be to fast track Peter's stuff, which
> I could put some time into... but that seems a bit risky at this stage
> of the game.
>
> I'd like to hear any other suggestions though. Patch included to aid
> discussion at this stage, rather than to encourage any rash decisions.

I see the demonstrable imbalance but I was wondering if there is there a real 
world benchmark that is currently affected?

Cheers,
Con
-
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/