linux/mm
Mel Gorman 90afa5de6f vmscan: properly account for the number of page cache pages zone_reclaim() can reclaim
A bug was brought to my attention against a distro kernel but it affects
mainline and I believe problems like this have been reported in various
guises on the mailing lists although I don't have specific examples at the
moment.

The reported problem was that malloc() stalled for a long time (minutes in
some cases) if a large tmpfs mount was occupying a large percentage of
memory overall.  The pages did not get cleaned or reclaimed by
zone_reclaim() because the zone_reclaim_mode was unsuitable, but the lists
are uselessly scanned frequencly making the CPU spin at near 100%.

This patchset intends to address that bug and bring the behaviour of
zone_reclaim() more in line with expectations which were noticed during
investigation.  It is based on top of mmotm and takes advantage of
Kosaki's work with respect to zone_reclaim().

Patch 1 fixes the heuristics that zone_reclaim() uses to determine if the
	scan should go ahead. The broken heuristic is what was causing the
	malloc() stall as it uselessly scanned the LRU constantly. Currently,
	zone_reclaim is assuming zone_reclaim_mode is 1 and historically it
	could not deal with tmpfs pages at all. This fixes up the heuristic so
	that an unnecessary scan is more likely to be correctly avoided.

Patch 2 notes that zone_reclaim() returning a failure automatically means
	the zone is marked full. This is not always true. It could have
	failed because the GFP mask or zone_reclaim_mode were unsuitable.

Patch 3 introduces a counter zreclaim_failed that will increment each
	time the zone_reclaim scan-avoidance heuristics fail. If that
	counter is rapidly increasing, then zone_reclaim_mode should be
	set to 0 as a temporarily resolution and a bug reported because
	the scan-avoidance heuristic is still broken.

This patch:

On NUMA machines, the administrator can configure zone_reclaim_mode that
is a more targetted form of direct reclaim.  On machines with large NUMA
distances for example, a zone_reclaim_mode defaults to 1 meaning that
clean unmapped pages will be reclaimed if the zone watermarks are not
being met.

There is a heuristic that determines if the scan is worthwhile but the
problem is that the heuristic is not being properly applied and is
basically assuming zone_reclaim_mode is 1 if it is enabled.  The lack of
proper detection can manfiest as high CPU usage as the LRU list is scanned
uselessly.

Historically, once enabled it was depending on NR_FILE_PAGES which may
include swapcache pages that the reclaim_mode cannot deal with.  Patch
vmscan-change-the-number-of-the-unmapped-files-in-zone-reclaim.patch by
Kosaki Motohiro noted that zone_page_state(zone, NR_FILE_PAGES) included
pages that were not file-backed such as swapcache and made a calculation
based on the inactive, active and mapped files.  This is far superior when
zone_reclaim==1 but if RECLAIM_SWAP is set, then NR_FILE_PAGES is a
reasonable starting figure.

This patch alters how zone_reclaim() works out how many pages it might be
able to reclaim given the current reclaim_mode.  If RECLAIM_SWAP is set in
the reclaim_mode it will either consider NR_FILE_PAGES as potential
candidates or else use NR_{IN}ACTIVE}_PAGES-NR_FILE_MAPPED to discount
swapcache and other non-file-backed pages.  If RECLAIM_WRITE is not set,
then NR_FILE_DIRTY number of pages are not candidates.  If RECLAIM_SWAP is
not set, then NR_FILE_MAPPED are not.

[kosaki.motohiro@jp.fujitsu.com: Estimate unmapped pages minus tmpfs pages]
[fengguang.wu@intel.com: Fix underflow problem in Kosaki's estimate]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: Rik van Riel <riel@redhat.com>
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-06-16 19:47:45 -07:00
..
allocpercpu.c percpu: __percpu_depopulate_mask can take a const mask 2009-04-06 13:44:15 -07:00
backing-dev.c block: change the request allocation/congestion logic to be sync/async based 2009-04-06 08:04:53 -07:00
bootmem.c bootmem: fix slab fallback on numa 2009-06-11 19:15:54 +03:00
bounce.c Merge branch 'for-2.6.31' of git://git.kernel.dk/linux-2.6-block 2009-06-11 11:10:35 -07:00
debug-pagealloc.c generic debug pagealloc 2009-04-01 08:59:13 -07:00
dmapool.c dmapool: enable debugging for CONFIG_SLUB_DEBUG_ON too 2008-04-28 08:58:20 -07:00
fadvise.c readahead: move max_sane_readahead() calls into force_page_cache_readahead() 2009-06-16 19:47:28 -07:00
failslab.c kmemtrace, mm: fix slab.h dependency problem in mm/failslab.c 2009-04-03 12:23:01 +02:00
filemap_xip.c mm: do_xip_mapping_read: fix length calculation 2009-04-02 19:04:49 -07:00
filemap.c page allocator: do not check NUMA node ID when the caller knows the node is valid 2009-06-16 19:47:32 -07:00
fremap.c Do not account for the address space used by hugetlbfs using VM_ACCOUNT 2009-02-10 10:48:42 -08:00
highmem.c mm: introduce debug_kmap_atomic 2009-04-01 08:59:14 -07:00
hugetlb.c mm: introduce PageHuge() for testing huge/gigantic pages 2009-06-16 19:47:36 -07:00
init-mm.c mm: consolidate init_mm definition 2009-06-16 19:47:28 -07:00
internal.h mm: remove CONFIG_UNEVICTABLE_LRU config option 2009-06-16 19:47:42 -07:00
Kconfig mm: remove CONFIG_UNEVICTABLE_LRU config option 2009-06-16 19:47:42 -07:00
Kconfig.debug generic debug pagealloc: build fix 2009-04-02 19:04:48 -07:00
kmemleak-test.c kmemleak: Simple testing module for kmemleak 2009-06-11 17:04:19 +01:00
kmemleak.c kmemleak: Add the base support 2009-06-11 17:03:28 +01:00
maccess.c [S390] maccess: add weak attribute to probe_kernel_write 2009-06-12 10:27:37 +02:00
madvise.c mm: madvise(): correct return code 2009-06-16 19:47:40 -07:00
Makefile mm: consolidate init_mm definition 2009-06-16 19:47:28 -07:00
memcontrol.c vmscan: evict use-once pages first 2009-06-16 19:47:38 -07:00
memory_hotplug.c page-allocator: reset wmark_min and inactive ratio of zone when hotplug happens 2009-06-16 19:47:42 -07:00
memory.c mm: introduce follow_pfn() 2009-06-16 19:47:40 -07:00
mempolicy.c page allocator: do not check NUMA node ID when the caller knows the node is valid 2009-06-16 19:47:32 -07:00
mempool.c
migrate.c migration: only migrate_prep() once per move_pages() 2009-06-16 19:47:41 -07:00
mincore.c [CVE-2009-0029] System call wrappers part 14 2009-01-14 14:15:24 +01:00
mlock.c mm: remove CONFIG_UNEVICTABLE_LRU config option 2009-06-16 19:47:42 -07:00
mm_init.c mm: mminit_loglevel cannot be __meminitdata anymore 2008-08-20 15:40:30 -07:00
mmap.c Merge branch 'perfcounters-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2009-06-11 14:01:07 -07:00
mmu_notifier.c mmu-notifiers: core 2008-07-28 16:30:21 -07:00
mmzone.c [ARM] Double check memmap is actually valid with a memmap has unexpected holes V2 2009-05-18 11:22:24 +01:00
mprotect.c perf_counter: Add mmap event hooks to mprotect() 2009-06-08 23:10:43 +02:00
mremap.c [CVE-2009-0029] System call wrappers part 13 2009-01-14 14:15:23 +01:00
msync.c [CVE-2009-0029] System call wrappers part 13 2009-01-14 14:15:23 +01:00
nommu.c nommu: Provide mmap_min_addr definition. 2009-06-10 09:24:09 +10:00
oom_kill.c oom: only oom kill exiting tasks with attached memory 2009-06-16 19:47:45 -07:00
page_alloc.c page-allocator: clear N_HIGH_MEMORY map before we set it again 2009-06-16 19:47:43 -07:00
page_cgroup.c memcg: fix page_cgroup fatal error in FLATMEM 2009-06-12 11:00:54 +03:00
page_io.c mm: remove file argument from swap_readpage() 2009-06-16 19:47:44 -07:00
page_isolation.c memory hotplug: fix page_zone() calculation in test_pages_isolated() 2008-11-06 15:41:19 -08:00
page-writeback.c mm/page-writeback.c: dirty limit type should be unsigned long 2009-06-16 19:47:31 -07:00
pagewalk.c pagemap: pass mm into pagewalkers 2008-06-12 18:05:41 -07:00
pdflush.c Revert "mm: add /proc controls for pdflush threads" 2009-05-15 11:32:24 +02:00
percpu.c percpu: remove rbtree and use page->index instead 2009-04-08 18:31:31 +02:00
prio_tree.c
quicklist.c cpumask: replace node_to_cpumask with cpumask_of_node. 2009-03-13 14:49:46 +10:30
readahead.c readahead: introduce context readahead algorithm 2009-06-16 19:47:30 -07:00
rmap.c vmscan: report vm_flags in page_referenced() 2009-06-16 19:47:44 -07:00
shmem_acl.c [PATCH] sanitize ->permission() prototype 2008-07-26 20:53:14 -04:00
shmem.c mm cleanup: shmem_file_setup: 'char *' -> 'const char *' for name argument 2009-06-16 19:47:44 -07:00
slab.c page allocator: slab: use nr_online_nodes to check for a NUMA platform 2009-06-16 19:47:35 -07:00
slob.c page allocator: do not check NUMA node ID when the caller knows the node is valid 2009-06-16 19:47:32 -07:00
slub.c page allocator: use a pre-calculated value instead of num_online_nodes() in fast paths 2009-06-16 19:47:35 -07:00
sparse-vmemmap.c vmemmap: warn about page_structs with remote distance 2008-11-06 15:41:19 -08:00
sparse.c mm: mminit_validate_memmodel_limits(): remove redundant test 2009-04-01 08:59:11 -07:00
swap_state.c mm: remove file argument from swap_readpage() 2009-06-16 19:47:44 -07:00
swap.c mm: fix Committed_AS underflow on large NR_CPUS environment 2009-05-02 15:36:10 -07:00
swapfile.c mm: reuse unused swap entry if necessary 2009-06-16 19:47:42 -07:00
thrash.c
truncate.c mm: remove __invalidate_mapping_pages variant 2009-06-16 19:47:43 -07:00
util.c mm: clean up get_user_pages_fast() documentation 2009-06-16 19:47:30 -07:00
vmalloc.c Merge branch 'for-linus' of git://linux-arm.org/linux-2.6 2009-06-11 14:15:57 -07:00
vmscan.c vmscan: properly account for the number of page cache pages zone_reclaim() can reclaim 2009-06-16 19:47:45 -07:00
vmstat.c mm: remove CONFIG_UNEVICTABLE_LRU config option 2009-06-16 19:47:42 -07:00