From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nigel Cunningham Subject: Re: [linux-pm] [RFC] Asynchronous suspend/resume - test results Date: Sat, 26 Dec 2009 07:57:17 +1100 Message-ID: <4B35272D.7040207@crca.org.au> References: <200912210140.19713.rjw@sisk.pl> <200912232146.29339.rjw@sisk.pl> <4B32D2D6.3030707@crca.org.au> <200912242310.42615.rjw@sisk.pl> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from mail.crca.org.au ([67.207.131.56]:43204 "EHLO crca.org.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756630AbZLYU5F (ORCPT ); Fri, 25 Dec 2009 15:57:05 -0500 In-Reply-To: <200912242310.42615.rjw@sisk.pl> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: "Rafael J. Wysocki" Cc: Jiri Slaby , Dmitry Torokhov , LKML , ACPI Devel Maling List , pm list , Linus Torvalds Rafael J. Wysocki wrote: > On Thursday 24 December 2009, Nigel Cunningham wrote: >> Hi. > > Hi, > >> I built the async branch of your tree and tested it, also running >> 2.6.33-rc1 + TuxOnIce for comparison. Dmesg for both are attached. Is >> there anything I can/should be doing for you on top of this? > > No, thanks a lot. > >> I'll try Dmitry's patch on top of this a little later - other things to do first. > > No need for that, the patchset contains an equivalent of the Dmitry's patch. > >> I noticed that you were doing standard deviations in your stats - how >> many runs were you basing them on? > > I usually run 10 iterations of suspend-resume for each configuration. > The raw data are at http://www.sisk.pl/kernel/data/async-suspend-updated.pdf > if you're interested. > >> Not sure that I can be bothered to do too many - too much else to do! > > Sure, thanks a lot anyway. Your data confirn that there's a measurable gain > from suspending and resuming devices asynchronously. It did? I thought it showed no difference at all! I'll see if I can find the time to do the other computers, then. We're going away for a couple of weeks on Monday, though, so I'm not sure that I'll get the time beforehand. Nigel