From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754719Ab0IFU1O (ORCPT ); Mon, 6 Sep 2010 16:27:14 -0400 Received: from icebox.esperi.org.uk ([81.187.191.129]:35376 "EHLO mail.esperi.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753051Ab0IFU1I (ORCPT ); Mon, 6 Sep 2010 16:27:08 -0400 To: Damien Wyart Cc: linux-kernel@vger.kernel.org Subject: Re: 2.6.35.*: horrible (exponential? >linear) slowdown to unusability (ACPI idle?) References: <87zkvx2ese.fsf@spindle.srvr.nix> <20100906053255.GA23340@brouette> From: Nix Emacs: or perhaps you'd prefer Russian Roulette, after all? Date: Mon, 06 Sep 2010 21:27:04 +0100 In-Reply-To: <20100906053255.GA23340@brouette> (Damien Wyart's message of "Mon, 6 Sep 2010 07:32:55 +0200") Message-ID: <87aanu39ev.fsf@spindle.srvr.nix> User-Agent: Gnus/5.1008 (Gnus v5.10.8) XEmacs/21.5-b29 (linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-DCC-URT-Metrics: spindle 1060; Body=2 Fuz1=2 Fuz2=2 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 6 Sep 2010, Damien Wyart told this: > I think this is related to what has been discussed in this thread: > http://lkml.org/lkml/2010/7/14/11 Quite possibly :/ though I'm on x86-64, not x86, and certainly don't have kmemleak turne don. > I've not seen the problem with 2.6.35.4 yet, and only once with > 2.6.35.3. Zeno said it disappeared for him when he started to use > 2.6.35-stable. Argh. > So if you can reproduce it quite easily, I guess bisection (even > painful) will be the best way to get (hopefuly) an idea of where the > problem might come from... 'Quite easily' is not quite true. It does not show up in qemu -kernel. It does not show up if you boot straight to an emergency boot shell in the initramfs (I was trying to avoid activating the RAID arrays). So I guess it's a full-blown boot each time, which is made pointlessly difficult by the fact that the rpc.mountd on my NFS server is acting up, giving me -ESTALE half the time for an initial mount (??!!!) and thus totally buggering up my client system whenever I reboot. So I may have to debug *that* first. What fun. But I'll get around to it.