Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262229AbVDXCLR (ORCPT ); Sat, 23 Apr 2005 22:11:17 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262231AbVDXCLR (ORCPT ); Sat, 23 Apr 2005 22:11:17 -0400 Received: from gate.crashing.org ([63.228.1.57]:60336 "EHLO gate.crashing.org") by vger.kernel.org with ESMTP id S262229AbVDXCLP (ORCPT ); Sat, 23 Apr 2005 22:11:15 -0400 Subject: Re: 2.6.12-rc2-mm3 From: Benjamin Herrenschmidt To: Juergen Kreileder Cc: Andrew Morton , Linux Kernel list , Oleg Nesterov In-Reply-To: <87fyxhj5p1.fsf@blackdown.de> References: <20050411012532.58593bc1.akpm@osdl.org> <87wtr8rdvu.fsf@blackdown.de> <87u0m7aogx.fsf@blackdown.de> <1113607416.5462.212.camel@gaston> <877jj1aj99.fsf@blackdown.de> <20050423170152.6b308c74.akpm@osdl.org> <87fyxhj5p1.fsf@blackdown.de> Content-Type: text/plain Date: Sun, 24 Apr 2005 12:15:28 +1000 Message-Id: <1114308928.5443.13.camel@gaston> Mime-Version: 1.0 X-Mailer: Evolution 2.0.4 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1347 Lines: 33 On Sun, 2005-04-24 at 03:59 +0200, Juergen Kreileder wrote: > I'm might be the only one using evdev on ppc64. > > And I don't know how popular LVM2 is on disks with Macintosh labels. > I had to set it up manually when I installed the machine, Debian's > installer couldn't handle it at that time. > > Workload is normal, the lockups happen with just X and Azaereus. > (The machine also runs mysqld, apache, and a few other daemons. But I > don't have to put load on these to make the machine lock up.) If you make sure you have CONFIG_XMON enabled and CONFIG_BOOTX_TEXT, and make sure X has "UseFBDev" option, do you drop into xmon before the lockup ? Also, do you have another machine at hand ? if yes, then we can try to revive my old firewire based debug tools we used to track things down in linus tree. I'll have a look at the timer patch next week, they might have some subtle race caused by a lack of memory barrier. I've had to debug some of those in early timer code, and those are really nasty, they usually only trigger under some subtle conditions, like ... heavy networking. Ben. - 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/