From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760243AbXKAKMi (ORCPT ); Thu, 1 Nov 2007 06:12:38 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755926AbXKAKMa (ORCPT ); Thu, 1 Nov 2007 06:12:30 -0400 Received: from viefep18-int.chello.at ([213.46.255.22]:45905 "EHLO viefep12-int.chello.at" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1756141AbXKAKM3 (ORCPT ); Thu, 1 Nov 2007 06:12:29 -0400 Subject: Re: [PATCH 3/6] sched: high-res preemption tick From: Peter Zijlstra To: Andi Kleen Cc: linux-kernel@vger.kernel.org, Ingo Molnar , Mike Galbraith , Dmitry Adamushko , Thomas Gleixner In-Reply-To: References: <20071031211030.310581000@chello.nl> <20071031211249.142859000@chello.nl> Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-ISBkyK7YYNfiFtZQoO26" Date: Thu, 01 Nov 2007 11:12:26 +0100 Message-Id: <1193911946.27652.252.camel@twins> Mime-Version: 1.0 X-Mailer: Evolution 2.10.1 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org --=-ISBkyK7YYNfiFtZQoO26 Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Wed, 2007-10-31 at 22:53 +0100, Andi Kleen wrote: > Peter Zijlstra writes: >=20 > > Use HR-timers (when available) to deliver an accurate preemption tick. > > > > The regular scheduler tick that runs at 1/HZ can be too coarse when nic= e > > level are used. The fairness system will still keep the cpu utilisation= 'fair' > > by then delaying the task that got an excessive amount of CPU time but = try to > > minimize this by delivering preemption points spot-on. >=20 > This might be costly when hrtimers happen to use an more expensive > to reprogram time source. Even an APIC timer access is fairly slow.=20 > And you'll potentially add the to lots of context switces. >=20 > Not sure that is a good idea for performance in general. Right, now I remember.=20 The idea was to run the rest of the kernel at HZ=3D50 or so, nothing but scheduling needs it anymore, and with this patch the scheduler doesn't need it anymore either. Should be good for power. This new hrtick thing only does a lot of ticks when there are a lot of runnable tasks, it starts at 2 with a tick per latency/2. --=-ISBkyK7YYNfiFtZQoO26 Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQBHKaaKXA2jU0ANEf4RAiRDAJ99SJJyXrohnF6U8CPcZ6HgB2e07wCfROFT t8RkHGv4HWF+p8FotQLxj30= =NDa5 -----END PGP SIGNATURE----- --=-ISBkyK7YYNfiFtZQoO26--