From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933893AbXFFBc7 (ORCPT ); Tue, 5 Jun 2007 21:32:59 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758510AbXFFBcw (ORCPT ); Tue, 5 Jun 2007 21:32:52 -0400 Received: from netops-testserver-4-out.sgi.com ([192.48.171.29]:51896 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1761085AbXFFBcv (ORCPT ); Tue, 5 Jun 2007 21:32:51 -0400 Date: Tue, 5 Jun 2007 18:32:50 -0700 (PDT) From: Christoph Lameter X-X-Sender: clameter@schroedinger.engr.sgi.com To: David Rientjes cc: Paul Jackson , akpm@linux-foundation.org, ak@suse.de, clameter@cthulhu.engr.sgi.com, linux-kernel@vger.kernel.org Subject: Re: [patch] cpusets: do not allow TIF_MEMDIE tasks to allocate globally In-Reply-To: Message-ID: References: <20070605160149.348b5c4a.pj@sgi.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 5 Jun 2007, David Rientjes wrote: > Obviously GFP_KERNEL allocations can allocate regardless of our memory > exclusivity, but the point is that a job in one exclusive cpuset should > not have the ability to effect the performance (in terms of reclaim and > swap), memory usage, or survival of jobs in other exclusive cpusets > because it was out of memory. Right but the process is terminating thus only requiring limited resources to get finished. The process does not have the ability to affect the other cpusets. I.e. it cannot directly allocate outside of the cpuset. The system has that capability and the system is handling the termination of the process and should terminate the process in a clean way if possible. > > Processes stuck in D state is another issue with reliability. > But it's a reality that we need to respect. It happens and when it does > it has the potential to hamper other cpusets that we setup to be exclusive > themselves. The fact that process hang because of some software deficiency is an exceptional failure scenario. The system in general is not fully operational anymore anyways. Typically one or the other lock is held which makes certain kernel functionality inaccessible.