cpusets: update mems allowed in page allocator
commite33c3b5e172e2e45456f42fba47227d48745543f
authorDavid Rientjes <rientjes@google.com>
Wed, 12 Nov 2008 21:25:37 +0000 (12 13:25 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 13 Nov 2008 01:17:16 +0000 (12 17:17 -0800)
tree0dc15f10c63b36b12de1dee6b3b70108286a5058
parentf0f7e0dc7393268947dc3ed285defc3d375487b9
cpusets: update mems allowed in page allocator

If all allowable memory is unreclaimable, it is possible to loop forever
in the page allocator for ~__GFP_NORETRY allocations.

During this time, it is also possible for a task's cpuset to expand its
set of allowable nodes so that it now includes free memory.  The cached
copy of this set, current->mems_allowed, is stale, however, since there
has not been a subsequent call to cpuset_update_task_memory_state().

The cached copy of the set of allowable nodes is now updated in the page
allocator's slow path so the additional memory is available to
get_page_from_freelist().

[akpm@linux-foundation.org: add comment]
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Paul Menage <menage@google.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/page_alloc.c