From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Sat, 23 Feb 2008 00:05:57 -0800 From: Andrew Morton Subject: Re: [PATCH 08/28] mm: system wide ALLOC_NO_WATERMARK Message-Id: <20080223000557.82125b3c.akpm@linux-foundation.org> In-Reply-To: <20080220150306.297640000@chello.nl> References: <20080220144610.548202000@chello.nl> <20080220150306.297640000@chello.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: Peter Zijlstra Cc: Linus Torvalds , linux-kernel@vger.kernel.org, linux-mm@kvack.org, netdev@vger.kernel.org, trond.myklebust@fys.uio.no List-ID: On Wed, 20 Feb 2008 15:46:18 +0100 Peter Zijlstra wrote: > Change ALLOC_NO_WATERMARK page allocation such that the reserves are system > wide - which they are per setup_per_zone_pages_min(), when we scrape the > barrel, do it properly. > The changelog is fairly incomprehensible. > mm/page_alloc.c | 6 ++++++ > 1 file changed, 6 insertions(+) > > Index: linux-2.6/mm/page_alloc.c > =================================================================== > --- linux-2.6.orig/mm/page_alloc.c > +++ linux-2.6/mm/page_alloc.c > @@ -1552,6 +1552,12 @@ restart: > rebalance: > if (alloc_flags & ALLOC_NO_WATERMARKS) { > nofail_alloc: > + /* > + * break out of mempolicy boundaries > + */ > + zonelist = NODE_DATA(numa_node_id())->node_zonelists + > + gfp_zone(gfp_mask); > + > /* go through the zonelist yet again, ignoring mins */ > page = get_page_from_freelist(gfp_mask, order, zonelist, > ALLOC_NO_WATERMARKS); As is the patch. People who care about mempolicies will want a better explanation, please, so they can check that we're not busting their stuff. -- 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