From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934206AbXFEXzy (ORCPT ); Tue, 5 Jun 2007 19:55:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S934142AbXFEXzj (ORCPT ); Tue, 5 Jun 2007 19:55:39 -0400 Received: from netops-testserver-3-out.sgi.com ([192.48.171.28]:47100 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S934119AbXFEXzi (ORCPT ); Tue, 5 Jun 2007 19:55:38 -0400 Date: Tue, 5 Jun 2007 16:55:38 -0700 From: Paul Jackson To: David Rientjes Cc: clameter@sgi.com, 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 Message-Id: <20070605165538.0a3ce6fe.pj@sgi.com> In-Reply-To: References: <20070605160149.348b5c4a.pj@sgi.com> Organization: SGI X-Mailer: Sylpheed version 2.2.4 (GTK+ 2.8.3; i686-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > If that fails, we can't allocate elsewhere because then we have taken > exclusive memory from other applications and is contrary to the definition > of mem_exclusive. Well, I can't speak to the 'real' meaning of TIF_MEMDIE with authority, but I can speak to the meaning of cpuset flags. The mem_exclusive flag doesn't mean this. It means that you cannot overlap the memory of a sibling cpuset. You will, necessarily, still overlap the memory of your ancestor cpusets. Whether or not you make any use of the mem_exclusive flag, you still get the same (limited) guarantees of memory usage -- namely that your memory won't be used by tasks in non-overlapping cpusets, with some exceptions, such as: 1) memory handed out to interrupt code, 2) memory handed out for GFP_ATOMIC requests, and 3) tasks marked PF_EXITING -- will soon free up memory Tasks in cpusets ancestor to your tasks cpuset can always, easily, use memory on the same nodes your task is on. -- I won't rest till it's the best ... Programmer, Linux Scalability Paul Jackson 1.925.600.0401