]> git.proxmox.com Git - mirror_ubuntu-eoan-kernel.git/commit
mm: page_alloc: exclude unreclaimable allocations from zone fairness policy
authorJohannes Weiner <hannes@cmpxchg.org>
Thu, 19 Dec 2013 01:08:47 +0000 (17:08 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 19 Dec 2013 03:04:51 +0000 (19:04 -0800)
commit73f038b863dfe98acabc7c36c17342b84ad52e94
tree4cd2132b22b48308bf3676f80fdce1f7691de06b
parentb0943d61b8fa420180f92f64ef67662b4f6cc493
mm: page_alloc: exclude unreclaimable allocations from zone fairness policy

Dave Hansen noted a regression in a microbenchmark that loops around
open() and close() on an 8-node NUMA machine and bisected it down to
commit 81c0a2bb515f ("mm: page_alloc: fair zone allocator policy").
That change forces the slab allocations of the file descriptor to spread
out to all 8 nodes, causing remote references in the page allocator and
slab.

The round-robin policy is only there to provide fairness among memory
allocations that are reclaimed involuntarily based on pressure in each
zone.  It does not make sense to apply it to unreclaimable kernel
allocations that are freed manually, in this case instantly after the
allocation, and incur the remote reference costs twice for no reason.

Only round-robin allocations that are usually freed through page reclaim
or slab shrinking.

Bisected by Dave Hansen.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/page_alloc.c