Direct reclaim: cpuset aware writeout

During direct reclaim we traverse down a zonelist and are carefully
checking each zone if its a member of the active cpuset. But then we call
pdflush without enforcing the same restriction. In a larger system this
has the effect of pages being dirtied in the cpuset and then either

A. No writeout occurs because global dirty limits have not been reached

or

B. Writeout starts randomly for any dirty inode in the system. Pdflush
may just write out data for nodes in another cpuset and miss doing
proper dirty handling for the current cpuset.

Fix that by requiring pdflush to only flush inodes that have pages
on a node in the cpuset.

Dave: This is an inode based writeout that will avoid the single
page writeout you are concerned about.

Signed-off-by: Christoph Lameter <clameter@sgi.com>

Index: linux-2.6.20-rc4/mm/vmscan.c
===================================================================
--- linux-2.6.20-rc4.orig/mm/vmscan.c	2007-01-08 18:11:16.867231662 -0600
+++ linux-2.6.20-rc4/mm/vmscan.c	2007-01-08 18:11:17.772586575 -0600
@@ -1065,7 +1065,8 @@ unsigned long try_to_free_pages(struct z
 		 */
 		if (total_scanned > sc.swap_cluster_max +
 					sc.swap_cluster_max / 2) {
-			wakeup_pdflush(laptop_mode ? 0 : total_scanned, NULL);
+			wakeup_pdflush(laptop_mode ? 0 : total_scanned,
+				&cpuset_current_mems_allowed);
 			sc.may_writepage = 1;
 		}