From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Lameter Subject: [11/17] GFP_VFALLBACK for zone wait table. Date: Tue, 18 Sep 2007 20:36:16 -0700 Message-ID: <20070919033642.843541917@sgi.com> References: <20070919033605.785839297@sgi.com> Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org To: Christoph Hellwig , Mel Gorman Return-path: Received: from netops-testserver-4-out.sgi.com ([192.48.171.29]:54944 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1752048AbXISDgn (ORCPT ); Tue, 18 Sep 2007 23:36:43 -0400 Cc: David Chinner , Jens Axboe Content-Disposition: inline; filename=vcompound_wait_table_no_vmalloc Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org Currently we have to use vmalloc for the zone wait table possibly generating the need to create lots of TLBs to access the tables. We can now use GFP_VFALLBACK to attempt the use of a physically contiguous page that can then use the large kernel TLBs. Signed-off-by: Christoph Lameter --- mm/page_alloc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) Index: linux-2.6/mm/page_alloc.c =================================================================== --- linux-2.6.orig/mm/page_alloc.c 2007-09-18 14:29:05.000000000 -0700 +++ linux-2.6/mm/page_alloc.c 2007-09-18 14:29:10.000000000 -0700 @@ -2572,7 +2572,9 @@ int zone_wait_table_init(struct zone *zo * To use this new node's memory, further consideration will be * necessary. */ - zone->wait_table = (wait_queue_head_t *)vmalloc(alloc_size); + zone->wait_table = (wait_queue_head_t *) + __get_free_pages(GFP_VFALLBACK, + get_order(alloc_size)); } if (!zone->wait_table) return -ENOMEM; --