From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261457AbULPSPd (ORCPT ); Thu, 16 Dec 2004 13:15:33 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261763AbULPSPd (ORCPT ); Thu, 16 Dec 2004 13:15:33 -0500 Received: from relay02.pair.com ([209.68.5.16]:4616 "HELO relay02.pair.com") by vger.kernel.org with SMTP id S261457AbULPSP1 (ORCPT ); Thu, 16 Dec 2004 13:15:27 -0500 X-pair-Authenticated: 66.134.112.218 Subject: Re: [ACPI] [ACPI][2.6.10-rc3][SUSPEND] S3 mode - Cannot resume from PCI devices From: Daniel Gryniewicz To: Pavel Machek Cc: Shawn Starr , linux-kernel@vger.kernel.org, Len Brown , acpi-devel@lists.sourceforge.net In-Reply-To: <20041214181819.GB5267@openzaurus.ucw.cz> References: <200412100315.21725.shawn.starr@rogers.com> <20041214110648.GA2291@elf.ucw.cz> <1103039717.10857.53.camel@athena.fprintf.net> <20041214181819.GB5267@openzaurus.ucw.cz> Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Thu, 16 Dec 2004 13:15:24 -0500 Message-Id: <1103220924.23829.8.camel@athena.fprintf.net> Mime-Version: 1.0 X-Mailer: Evolution 2.1.1 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2004-12-14 at 19:18 +0100, Pavel Machek wrote: > Hi! > > > > > This isn't a nice regression. > > > > > > So what was the last kernel where it worked? > > > > For me, 2.6.9-rc4. I've tried every -rc and -mm since, and it cannot > > resume. (I get no video on resume, even with 2.6.9-rc4, until X > > Okay, perhaps you are lucky, diff between 2.6.9-rc4 and 2.6.9 > should be fairly small. Could you find which changeset is responsible? > bkcvs should be really usefull here. > > Pavel So, I started out by building 2.6.9 to test it failed. It failed. Then, I started binary searching between 2.6.9-rc4 and 2.6.9, and worked my way all the way back up to 2.6.9 with it working. I'm now running the current CVS as of yesterday, with no problems, and with suspend and resume working. I have no idea what changed between the original build of 2.6.9 that failed, and the final build that worked, but it would appear to have been something that was my fault. Now if I could only get 10-rc4-ck1 to not hang on boot... Daniel