From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757175Ab0BKUyK (ORCPT ); Thu, 11 Feb 2010 15:54:10 -0500 Received: from one.firstfloor.org ([213.235.205.2]:36916 "EHLO one.firstfloor.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757127Ab0BKUyG (ORCPT ); Thu, 11 Feb 2010 15:54:06 -0500 From: Andi Kleen References: <20100211953.850854588@firstfloor.org> In-Reply-To: <20100211953.850854588@firstfloor.org> To: penberg@cs.helsinki.fi, linux-kernel@vger.kernel.org, linux-mm@kvack.org, haicheng.li@intel.com, rientjes@google.com Subject: [PATCH] [1/4] SLAB: Handle node-not-up case in fallback_alloc() v2 Message-Id: <20100211205401.002CFB1978@basil.firstfloor.org> Date: Thu, 11 Feb 2010 21:54:00 +0100 (CET) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org When fallback_alloc() runs the node of the CPU might not be initialized yet. Handle this case by allocating in another node. v2: Try to allocate from all nodes (David Rientjes) Signed-off-by: Andi Kleen --- mm/slab.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) Index: linux-2.6.32-memhotadd/mm/slab.c =================================================================== --- linux-2.6.32-memhotadd.orig/mm/slab.c +++ linux-2.6.32-memhotadd/mm/slab.c @@ -3188,7 +3188,24 @@ retry: if (local_flags & __GFP_WAIT) local_irq_enable(); kmem_flagcheck(cache, flags); - obj = kmem_getpages(cache, local_flags, numa_node_id()); + + /* + * Node not set up yet? Try one that the cache has been set up + * for. + */ + nid = numa_node_id(); + if (cache->nodelists[nid] == NULL) { + for_each_zone_zonelist(zone, z, zonelist, high_zoneidx) { + nid = zone_to_nid(zone); + if (cache->nodelists[nid]) { + obj = kmem_getpages(cache, local_flags, nid); + if (obj) + break; + } + } + } else + obj = kmem_getpages(cache, local_flags, nid); + if (local_flags & __GFP_WAIT) local_irq_disable(); if (obj) { From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail190.messagelabs.com (mail190.messagelabs.com [216.82.249.51]) by kanga.kvack.org (Postfix) with ESMTP id 0C0FF6B0078 for ; Thu, 11 Feb 2010 15:54:04 -0500 (EST) From: Andi Kleen References: <20100211953.850854588@firstfloor.org> In-Reply-To: <20100211953.850854588@firstfloor.org> Subject: [PATCH] [1/4] SLAB: Handle node-not-up case in fallback_alloc() v2 Message-Id: <20100211205401.002CFB1978@basil.firstfloor.org> Date: Thu, 11 Feb 2010 21:54:00 +0100 (CET) Sender: owner-linux-mm@kvack.org To: penberg@cs.helsinki.fi, linux-kernel@vger.kernel.org, linux-mm@kvack.org, haicheng.li@intel.com, rientjes@google.com List-ID: When fallback_alloc() runs the node of the CPU might not be initialized yet. Handle this case by allocating in another node. v2: Try to allocate from all nodes (David Rientjes) Signed-off-by: Andi Kleen --- mm/slab.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) Index: linux-2.6.32-memhotadd/mm/slab.c =================================================================== --- linux-2.6.32-memhotadd.orig/mm/slab.c +++ linux-2.6.32-memhotadd/mm/slab.c @@ -3188,7 +3188,24 @@ retry: if (local_flags & __GFP_WAIT) local_irq_enable(); kmem_flagcheck(cache, flags); - obj = kmem_getpages(cache, local_flags, numa_node_id()); + + /* + * Node not set up yet? Try one that the cache has been set up + * for. + */ + nid = numa_node_id(); + if (cache->nodelists[nid] == NULL) { + for_each_zone_zonelist(zone, z, zonelist, high_zoneidx) { + nid = zone_to_nid(zone); + if (cache->nodelists[nid]) { + obj = kmem_getpages(cache, local_flags, nid); + if (obj) + break; + } + } + } else + obj = kmem_getpages(cache, local_flags, nid); + if (local_flags & __GFP_WAIT) local_irq_disable(); if (obj) { -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org