From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <45B81E5B.1090505@google.com> Date: Wed, 24 Jan 2007 19:04:59 -0800 From: Ethan Solomita MIME-Version: 1.0 Subject: Re: [PATCH 1/5] Add a map to to track dirty pages per node References: <20070123185242.2640.8367.sendpatchset@schroedinger.engr.sgi.com> <20070123185248.2640.87514.sendpatchset@schroedinger.engr.sgi.com> In-Reply-To: <20070123185248.2640.87514.sendpatchset@schroedinger.engr.sgi.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: Christoph Lameter Cc: akpm@osdl.org, Paul Menage , Peter Zijlstra , Nick Piggin , linux-mm@kvack.org, Paul Jackson , Dave Chinner , Andi Kleen List-ID: Do we want this even with WB_SYNC_ALL and WB_SYNC_HOLD? It seems that callers from sync_inodes_sb(), which are the ones that pass in those options, may want to know that everything is written. -- Ethan Christoph Lameter wrote: > Index: linux-2.6.20-rc5/fs/fs-writeback.c > =================================================================== > --- linux-2.6.20-rc5.orig/fs/fs-writeback.c 2007-01-22 13:31:30.440219103 -0600 > +++ linux-2.6.20-rc5/fs/fs-writeback.c 2007-01-23 12:21:44.669179863 -0600 > @@ -22,6 +22,7 @@ > #include > #include > #include > +#include > #include "internal.h" > > /** > @@ -349,6 +350,12 @@ sync_sb_inodes(struct super_block *sb, s > continue; /* blockdev has wrong queue */ > } > > + if (!cpuset_intersects_dirty_nodes(mapping, wbc->nodes)) { > + /* No pages on the nodes under writeback */ > + list_move(&inode->i_list, &sb->s_dirty); > + continue; > + } > + > /* Was this inode dirtied after sync_sb_inodes was called? */ > if (time_after(inode->dirtied_when, start)) > break; > -- 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