Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751750AbZKJHYU (ORCPT ); Tue, 10 Nov 2009 02:24:20 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751457AbZKJHYT (ORCPT ); Tue, 10 Nov 2009 02:24:19 -0500 Received: from fgwmail6.fujitsu.co.jp ([192.51.44.36]:53148 "EHLO fgwmail6.fujitsu.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751034AbZKJHYT (ORCPT ); Tue, 10 Nov 2009 02:24:19 -0500 X-SecurityPolicyCheck-FJ: OK by FujitsuOutboundMailChecker v1.3.1 From: KOSAKI Motohiro To: KAMEZAWA Hiroyuki Subject: Re: [BUGFIX][PATCH] oom-kill: fix NUMA consraint check with nodemask v2 Cc: kosaki.motohiro@jp.fujitsu.com, "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , "akpm@linux-foundation.org" , cl@linux-foundation.org, rientjes@google.com In-Reply-To: <20091106090202.dc2472b3.kamezawa.hiroyu@jp.fujitsu.com> References: <20091104170944.cef988c7.kamezawa.hiroyu@jp.fujitsu.com> <20091106090202.dc2472b3.kamezawa.hiroyu@jp.fujitsu.com> Message-Id: <20091110162121.361B.A69D9226@jp.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Mailer: Becky! ver. 2.50.07 [ja] Date: Tue, 10 Nov 2009 16:24:22 +0900 (JST) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2123 Lines: 70 Hi > =================================================================== > --- mmotm-2.6.32-Nov2.orig/mm/oom_kill.c > +++ mmotm-2.6.32-Nov2/mm/oom_kill.c > @@ -196,27 +196,40 @@ unsigned long badness(struct task_struct > /* > * Determine the type of allocation constraint. > */ > +#ifdef CONFIG_NUMA > static inline enum oom_constraint constrained_alloc(struct zonelist *zonelist, > - gfp_t gfp_mask) > + gfp_t gfp_mask, nodemask_t *nodemask) > { > -#ifdef CONFIG_NUMA > struct zone *zone; > struct zoneref *z; > enum zone_type high_zoneidx = gfp_zone(gfp_mask); > - nodemask_t nodes = node_states[N_HIGH_MEMORY]; > + int ret = CONSTRAINT_NONE; > > - for_each_zone_zonelist(zone, z, zonelist, high_zoneidx) > - if (cpuset_zone_allowed_softwall(zone, gfp_mask)) > - node_clear(zone_to_nid(zone), nodes); > - else > + /* > + * The nodemask here is a nodemask passed to alloc_pages(). Now, > + * cpuset doesn't use this nodemask for its hardwall/softwall/hierarchy > + * feature. Then, only mempolicy use this nodemask. > + */ > + if (nodemask && nodes_equal(*nodemask, node_states[N_HIGH_MEMORY])) > + ret = CONSTRAINT_MEMORY_POLICY; !nodes_equal() ? > + > + /* Check this allocation failure is caused by cpuset's wall function */ > + for_each_zone_zonelist_nodemask(zone, z, zonelist, > + high_zoneidx, nodemask) > + if (!cpuset_zone_allowed_softwall(zone, gfp_mask)) > return CONSTRAINT_CPUSET; If cpuset and MPOL_BIND are both used, Probably CONSTRAINT_MEMORY_POLICY is better choice. > > - if (!nodes_empty(nodes)) > - return CONSTRAINT_MEMORY_POLICY; > -#endif > + /* __GFP_THISNODE never calls OOM.*/ > > + return ret; > +} > +#else > +static inline enum oom_constraint constrained_alloc(struct zonelist *zonelist, > + gfp_t gfp_mask, nodemask_t *nodemask) > +{ > return CONSTRAINT_NONE; > } > +#endif -- 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/