linux/mm
Johannes Weiner 6457474624 vmscan: detect mapped file pages used only once
The VM currently assumes that an inactive, mapped and referenced file page
is in use and promotes it to the active list.

However, every mapped file page starts out like this and thus a problem
arises when workloads create a stream of such pages that are used only for
a short time.  By flooding the active list with those pages, the VM
quickly gets into trouble finding eligible reclaim canditates.  The result
is long allocation latencies and eviction of the wrong pages.

This patch reuses the PG_referenced page flag (used for unmapped file
pages) to implement a usage detection that scales with the speed of LRU
list cycling (i.e.  memory pressure).

If the scanner encounters those pages, the flag is set and the page cycled
again on the inactive list.  Only if it returns with another page table
reference it is activated.  Otherwise it is reclaimed as 'not recently
used cache'.

This effectively changes the minimum lifetime of a used-once mapped file
page from a full memory cycle to an inactive list cycle, which allows it
to occur in linear streams without affecting the stable working set of the
system.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: OSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-03-06 11:26:27 -08:00
..
backing-dev.c flusher: Fix PF_FROZEN race 2009-12-03 13:49:43 +01:00
bootmem.c x86: Make 64 bit use early_res instead of bootmem before slab 2010-02-12 09:41:59 -08:00
bounce.c block: remove some includings of blktrace_api.h 2009-06-16 11:19:36 +02:00
debug-pagealloc.c generic debug pagealloc 2009-04-01 08:59:13 -07:00
dmapool.c dmapools: protect page_list walk in show_pools() 2009-06-30 18:56:00 -07:00
fadvise.c readahead: introduce FMODE_RANDOM for POSIX_FADV_RANDOM 2010-03-06 11:26:25 -08:00
failslab.c failslab: add ability to filter slab caches 2010-02-26 19:19:39 +02:00
filemap_xip.c mm: clean up mm_counter 2010-03-06 11:26:23 -08:00
filemap.c mm: use rlimit helpers 2010-03-06 11:26:24 -08:00
fremap.c mm: clean up mm_counter 2010-03-06 11:26:23 -08:00
highmem.c highmem: Fix debug_kmap_atomic() to also handle KM_IRQ_PTE, KM_NMI, and KM_NMI_PTE 2009-11-10 04:15:47 +01:00
hugetlb.c Merge branch 'for-linus' of master.kernel.org:/home/rmk/linux-2.6-arm 2010-03-01 09:15:15 -08:00
hwpoison-inject.c HWPOISON: Don't do early filtering if filter is disabled 2009-12-16 12:20:01 +01:00
init-mm.c mm: consolidate init_mm definition 2009-06-16 19:47:28 -07:00
internal.h HWPOISON: add an interface to switch off/on all the page filters 2009-12-16 12:19:59 +01:00
Kconfig Merge branch 'x86-bootmem-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip 2010-03-03 08:15:05 -08:00
Kconfig.debug trivial: improve help text for mm debug config options 2009-09-21 15:14:57 +02:00
kmemcheck.c kmemcheck: add hooks for the page allocator 2009-06-15 15:48:33 +02:00
kmemleak-test.c percpu: clean up percpu variable definitions 2009-06-24 15:13:48 +09:00
kmemleak.c Merge branch 'kmemleak' of git://linux-arm.org/linux-2.6 2009-12-17 16:00:19 -08:00
ksm.c mm: change anon_vma linking to fix multi-process server scalability issue 2010-03-06 11:26:26 -08:00
maccess.c maccess,probe_kernel: Allow arch specific override probe_kernel_(read|write) 2010-01-07 11:58:36 -06:00
madvise.c HWPOISON: Add a madvise() injector for soft page offlining 2009-12-16 12:20:00 +01:00
Makefile make generic_acl slightly more generic 2009-12-16 12:16:49 -05:00
memcontrol.c mm/memcontrol.c: fix "integer as NULL pointer" sparse warning 2010-03-06 11:26:26 -08:00
memory_hotplug.c memory-hotplug: create /sys/firmware/memmap entry for new memory 2010-03-06 11:26:25 -08:00
memory-failure.c mm: change anon_vma linking to fix multi-process server scalability issue 2010-03-06 11:26:26 -08:00
memory.c rmap: move exclusively owned pages to own anon_vma in do_wp_page() 2010-03-06 11:26:26 -08:00
mempolicy.c mm/mempolicy.c: fix indentation of the comments of do_migrate_pages 2010-03-06 11:26:25 -08:00
mempool.c mm: remove broken 'kzalloc' mempool 2009-09-22 07:17:35 -07:00
migrate.c mm/migrate.c: kill anon local variable from migrate_page_copy 2010-03-06 11:26:25 -08:00
mincore.c mm: hugetlb: fix hugepage memory leak in mincore() 2009-12-15 08:53:24 -08:00
mlock.c mm: use rlimit helpers 2010-03-06 11:26:24 -08:00
mm_init.c
mmap.c mm: remove VM_LOCK_RMAP code 2010-03-06 11:26:26 -08:00
mmu_context.c mm: export use_mm/unuse_mm to modules 2010-01-15 01:43:28 -08:00
mmu_notifier.c ksm: add mmu_notifier set_pte_at_notify() 2009-09-22 07:17:31 -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: Do the big rename: Performance Counters -> Performance Events 2009-09-21 14:28:04 +02:00
mremap.c mm: change anon_vma linking to fix multi-process server scalability issue 2010-03-06 11:26:26 -08:00
msync.c
nommu.c mm: change anon_vma linking to fix multi-process server scalability issue 2010-03-06 11:26:26 -08:00
oom_kill.c mm: clean up mm_counter 2010-03-06 11:26:23 -08:00
page_alloc.c mm: suppress pfn range output for zones without pages 2010-03-06 11:26:26 -08:00
page_cgroup.c memory hotplug: alloc page from other node in memory online 2009-09-22 07:17:26 -07:00
page_io.c swap: rework map_swap_page() again 2009-12-15 08:53:16 -08:00
page_isolation.c
page-writeback.c writeback: remove unused nonblocking and congestion checks 2009-12-03 13:54:25 +01:00
pagewalk.c mm hugetlb: add hugepage support to pagemap 2009-12-15 08:53:24 -08:00
percpu.c early_res: Add free_early_partial() 2010-02-26 08:25:35 +01:00
prio_tree.c
quicklist.c cpumask: use new-style cpumask ops in mm/quicklist. 2009-09-24 09:34:52 +09:30
readahead.c readahead: introduce FMODE_RANDOM for POSIX_FADV_RANDOM 2010-03-06 11:26:25 -08:00
rmap.c vmscan: detect mapped file pages used only once 2010-03-06 11:26:27 -08:00
shmem.c Fix breakage in shmem.c 2009-12-16 19:48:48 -05:00
slab.c Merge branches 'slab/cleanups', 'slab/failslab', 'slab/fixes' and 'slub/percpu' into slab-for-linus 2010-03-04 12:07:50 +02:00
slob.c slab: remove duplicate kmem_cache_init_late() declarations 2009-08-06 11:36:25 +03:00
slub.c SLUB: Fix per-cpu merge conflict 2010-03-04 12:09:43 +02:00
sparse-vmemmap.c sparsemem: Put mem map for one node together. 2010-02-12 09:42:38 -08:00
sparse.c sparsemem: Fix compilation on PowerPC 2010-03-01 17:59:24 -08:00
swap_state.c mm: add_to_swap_cache() does not return -EEXIST 2009-09-22 07:17:35 -07:00
swap.c mm: remove free_hot_page() 2010-03-06 11:26:25 -08:00
swapfile.c mm/swapfile.c: fix swapon size off-by-one 2010-03-06 11:26:26 -08:00
thrash.c mm: pass mm to grab_swap_token 2009-06-23 12:50:05 -07:00
truncate.c vfs: Fix vmtruncate() regression 2010-01-13 16:09:33 -08:00
util.c nommu: don't need get_unmapped_area() for NOMMU 2010-01-16 12:15:40 -08:00
vmalloc.c mm: purge fragmented percpu vmap blocks 2010-02-02 12:50:47 -08:00
vmscan.c vmscan: detect mapped file pages used only once 2010-03-06 11:26:27 -08:00
vmstat.c mm: restore zone->all_unreclaimable to independence word 2010-03-06 11:26:25 -08:00