Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754104AbZIVFDO (ORCPT ); Tue, 22 Sep 2009 01:03:14 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753798AbZIVFDL (ORCPT ); Tue, 22 Sep 2009 01:03:11 -0400 Received: from e28smtp01.in.ibm.com ([59.145.155.1]:56745 "EHLO e28smtp01.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753785AbZIVFDK (ORCPT ); Tue, 22 Sep 2009 01:03:10 -0400 Message-ID: <4AB85A8F.6010106@in.ibm.com> Date: Tue, 22 Sep 2009 10:33:11 +0530 From: Sachin Sant User-Agent: Thunderbird 2.0.0.22 (X11/20090609) MIME-Version: 1.0 To: Mel Gorman CC: Christoph Lameter , Nick Piggin , Pekka Enberg , heiko.carstens@de.ibm.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org, Tejun Heo , Benjamin Herrenschmidt Subject: Re: [RFC PATCH 0/3] Fix SLQB on memoryless configurations V2 References: <1253549426-917-1-git-send-email-mel@csn.ul.ie> <20090921174656.GS12726@csn.ul.ie> <20090921180739.GT12726@csn.ul.ie> In-Reply-To: <20090921180739.GT12726@csn.ul.ie> Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2403 Lines: 73 Mel Gorman wrote: > On Mon, Sep 21, 2009 at 01:54:12PM -0400, Christoph Lameter wrote: > >> Lets just keep SLQB back until the basic issues with memoryless nodes are >> resolved. >> > > It's not even super-clear that the memoryless nodes issues are entirely > related to SLQB. Sachin for example says that there was a stall issue > with memoryless nodes that could be triggered without SLQB. Sachin, is > that still accurate? > I think there are two different problems that we are dealing with. First one is the SLQB not working on a ppc64 box which seems to be specific to only one machine and i haven't seen that on other power boxes.The patches that you have posted seems to allow the box to boot, but eventually it hits the stall issue(related to percpu dynamic allocator not working on ppc64), which is the second problem we are dealing with. The stall issue seems to be much more critical as it is affecting almost all of the power boxes that i have tested with (4 in all). This issue is seen with Linus tree as well and was first seen with 2.6.31-git5 (0cb583fd..) The stall issue was reported here: http://lists.ozlabs.org/pipermail/linuxppc-dev/2009-September/075791.html Thanks -Sachin > If so, it's possible that SLQB somehow exasperates the problem in some > unknown fashion. > > >> There does not seem to be an easy way to deal with this. Some >> thought needs to go into how memoryless node handling relates to per cpu >> lists and locking. List handling issues need to be addressed before SLQB. >> can work reliably. The same issues can surface on x86 platforms with weird >> NUMA memory setups. >> >> > > Can you spot if there is something fundamentally wrong with patch 2? I.e. what > is wrong with treating the closest node as local instead of only the > closest node? > > >> Or just allow SLQB for !NUMA configurations and merge it now. >> >> > > Forcing SLQB !NUMA will not rattle out any existing list issues > unfortunately :(. > > -- --------------------------------- Sachin Sant IBM Linux Technology Center India Systems and Technology Labs Bangalore, India --------------------------------- -- 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/