linux/include
Mel Gorman 8fb74b9fb2 mm: compaction: partially revert capture of suitable high-order page
Eric Wong reported on 3.7 and 3.8-rc2 that ppoll() got stuck when
waiting for POLLIN on a local TCP socket.  It was easier to trigger if
there was disk IO and dirty pages at the same time and he bisected it to
commit 1fb3f8ca0e ("mm: compaction: capture a suitable high-order page
immediately when it is made available").

The intention of that patch was to improve high-order allocations under
memory pressure after changes made to reclaim in 3.6 drastically hurt
THP allocations but the approach was flawed.  For Eric, the problem was
that page->pfmemalloc was not being cleared for captured pages leading
to a poor interaction with swap-over-NFS support causing the packets to
be dropped.  However, I identified a few more problems with the patch
including the fact that it can increase contention on zone->lock in some
cases which could result in async direct compaction being aborted early.

In retrospect the capture patch took the wrong approach.  What it should
have done is mark the pageblock being migrated as MIGRATE_ISOLATE if it
was allocating for THP and avoided races that way.  While the patch was
showing to improve allocation success rates at the time, the benefit is
marginal given the relative complexity and it should be revisited from
scratch in the context of the other reclaim-related changes that have
taken place since the patch was first written and tested.  This patch
partially reverts commit 1fb3f8ca0e ("mm: compaction: capture a
suitable high-order page immediately when it is made available").

Reported-and-tested-by: Eric Wong <normalperson@yhbt.net>
Tested-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-01-11 14:54:56 -08:00
..
acpi PCI changes for the v3.8 merge window: 2012-12-13 12:14:47 -08:00
asm-generic mm: limit mmu_gather batching to fix soft lockups on !CONFIG_PREEMPT 2013-01-04 16:11:46 -08:00
clocksource
crypto crypto: cast5/cast6 - move lookup tables to shared module 2012-12-06 17:16:26 +08:00
drm drm/exynos: change file license to GPL 2013-01-04 15:54:32 +09:00
keys
linux mm: compaction: partially revert capture of suitable high-order page 2013-01-11 14:54:56 -08:00
math-emu
media [media] V4L: Add driver for S3C24XX/S3C64XX SoC series camera interface 2012-11-28 09:43:34 -02:00
memory
misc
net Merge branch 'master' of git://1984.lsi.us.es/nf 2012-12-28 14:28:17 -08:00
pcmcia
ras
rdma UAPI: Remove empty Kbuild files 2013-01-02 17:36:10 -08:00
rxrpc
scsi SCSI misc on 20121212 2012-12-13 19:20:31 -08:00
sound UAPI: Remove empty Kbuild files 2013-01-02 17:36:10 -08:00
target target/iscsi_target: Add NodeACL tags for initiator group support 2012-12-13 14:18:09 -08:00
trace Various bug fixes for ext4. Perhaps the most serious bug fixed is one 2013-01-02 09:57:34 -08:00
uapi linux/audit.h: move ptrace.h include to kernel header 2013-01-11 14:54:56 -08:00
video Merge branch 'omap-for-v3.8/fixes-for-merge-window' into omap-for-v3.8/fixes-for-merge-window-v2 2012-12-16 11:28:10 -08:00
xen Bugfixes: 2012-12-18 12:26:54 -08:00
Kbuild UAPI: Remove empty Kbuild files 2013-01-02 17:36:10 -08:00