linux/mm
KOSAKI Motohiro 01dbe5c9b1 vmscan: improve reclaim throughput to bail out patch
The vmscan bail out patch move nr_reclaimed variable to struct
scan_control.  Unfortunately, indirect access can easily happen cache
miss.

if heavy memory pressure happend, that's ok.
cache miss already plenty. it is not observable.

but, if memory pressure is lite, performance degression is obserbable.

I compared following three pattern (it was mesured 10 times each)

hackbench 125 process 3000
hackbench 130 process 3000
hackbench 135 process 3000

            2.6.28-rc6                       bail-out

	125	130	135		125	130	135
      ==============================================================
	71.866	75.86	81.274		93.414	73.254	193.382
	74.145	78.295	77.27		74.897	75.021	80.17
	70.305	77.643	75.855		70.134	77.571	79.896
	74.288	73.986	75.955		77.222	78.48	80.619
	72.029	79.947	78.312		75.128	82.172	79.708
	71.499	77.615	77.042		74.177	76.532	77.306
	76.188	74.471	83.562		73.839	72.43	79.833
	73.236	75.606	78.743		76.001	76.557	82.726
	69.427	77.271	76.691		76.236	79.371	103.189
	72.473	76.978	80.643		69.128	78.932	75.736

avg	72.545	76.767	78.534		76.017	77.03	93.256
std	1.89	1.71	2.41		6.29	2.79	34.16
min	69.427	73.986	75.855		69.128	72.43	75.736
max	76.188	79.947	83.562		93.414	82.172	193.382

about 4-5% degression.

Then, this patch introduces a temporary local variable.

result:

            2.6.28-rc6                       this patch

num	125	130	135		125	130	135
      ==============================================================
	71.866	75.86	81.274		67.302	68.269	77.161
	74.145	78.295	77.27   	72.616	72.712	79.06
	70.305	77.643	75.855  	72.475	75.712	77.735
	74.288	73.986	75.955  	69.229	73.062	78.814
	72.029	79.947	78.312  	71.551	74.392	78.564
	71.499	77.615	77.042  	69.227	74.31	78.837
	76.188	74.471	83.562  	70.759	75.256	76.6
	73.236	75.606	78.743  	69.966	76.001	78.464
	69.427	77.271	76.691  	69.068	75.218	80.321
	72.473	76.978	80.643  	72.057	77.151	79.068

avg	72.545	76.767	78.534 		70.425	74.2083	78.462
std 	1.89	1.71	2.41    	1.66	2.34	1.00
min 	69.427	73.986	75.855  	67.302	68.269	76.6
max 	76.188	79.947	83.562  	72.616	77.151	80.321

OK. the degression is disappeared.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: Rik van Riel <riel@redhat.com>
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>
2009-01-06 15:59:06 -08:00
..
allocpercpu.c mm/allocpercpu.c: make 4 functions static 2008-07-26 12:00:12 -07:00
backing-dev.c mm: change dirty limit type specifiers to unsigned long 2009-01-06 15:59:02 -08:00
bootmem.c misc: replace __FUNCTION__ with __func__ 2008-10-16 11:21:30 -07:00
bounce.c bounce: don't rely on a zeroed bio_vec list 2008-12-29 08:29:52 +01:00
dmapool.c dmapool: enable debugging for CONFIG_SLUB_DEBUG_ON too 2008-04-28 08:58:20 -07:00
fadvise.c Remove Andrew Morton's old email accounts 2008-10-16 11:21:32 -07:00
failslab.c SLUB: failslab support 2008-12-29 11:27:46 +02:00
filemap_xip.c mm: xip/ext2 fix block allocation race 2008-08-20 15:40:32 -07:00
filemap.c mm: write_cache_pages integrity fix 2009-01-06 15:58:59 -08:00
fremap.c mmap: handle mlocked pages during map, remap, unmap 2008-10-20 08:52:31 -07:00
highmem.c x86, pat: avoid highmem cache attribute aliasing 2008-08-15 17:22:57 +02:00
hugetlb.c hugetlb: fix sparse warnings 2009-01-06 15:59:06 -08:00
internal.h hugetlb: pull gigantic page initialisation out of the default path 2008-11-06 15:41:18 -08:00
Kconfig Unevictable LRU Infrastructure 2008-10-20 08:50:26 -07:00
maccess.c kgdb: fix optional arch functions and probe_kernel_* 2008-04-17 20:05:39 +02:00
madvise.c madvise: update function comment of madvise_dontneed 2008-07-30 09:41:45 -07:00
Makefile SLUB: failslab support 2008-12-29 11:27:46 +02:00
memcontrol.c mm: make mem_cgroup_resize_limit() static 2009-01-06 15:59:04 -08:00
memory_hotplug.c mm: remove GFP_HIGHUSER_PAGECACHE 2009-01-06 15:59:01 -08:00
memory.c mm: make maddr __iomem 2009-01-06 15:59:04 -08:00
mempolicy.c Merge branch 'master' into next 2008-11-14 11:29:12 +11:00
mempool.c spelling fixes: mm/ 2007-10-20 01:27:18 +02:00
migrate.c mm: add Set,ClearPageSwapCache stubs 2009-01-06 15:59:02 -08:00
mincore.c mm: remove nopage 2008-04-28 08:58:18 -07:00
mlock.c x86, bts: memory accounting 2008-12-20 09:15:47 +01:00
mm_init.c mm: mminit_loglevel cannot be __meminitdata anymore 2008-08-20 15:40:30 -07:00
mmap.c mm: update my address 2009-01-05 17:44:42 -08:00
mmu_notifier.c mmu-notifiers: core 2008-07-28 16:30:21 -07:00
mmzone.c mm: mark the correct zone as full when scanning zonelists 2008-09-13 14:41:52 -07:00
mprotect.c mm: cleanup: remove #ifdef CONFIG_MIGRATION 2009-01-06 15:59:00 -08:00
mremap.c mm: update my address 2009-01-05 17:44:42 -08:00
msync.c add a vfs_fsync helper 2009-01-05 11:54:28 -05:00
nommu.c inode->i_op is never NULL 2009-01-05 11:54:28 -05:00
oom_kill.c oom: print triggering task's cpuset and mems allowed 2009-01-06 15:58:59 -08:00
page_alloc.c mm: make setup_per_zone_inactive_ratio() static 2009-01-06 15:59:05 -08:00
page_cgroup.c mm: make init_section_page_cgroup() static 2009-01-06 15:59:04 -08:00
page_io.c mm: try_to_free_swap replaces remove_exclusive_swap_page 2009-01-06 15:59:03 -08:00
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: add dirty_background_bytes and dirty_bytes sysctls 2009-01-06 15:59:03 -08:00
pagewalk.c pagemap: pass mm into pagewalkers 2008-06-12 18:05:41 -07:00
pdflush.c cpumask: convert mm/ 2009-01-01 10:12:29 +10:30
prio_tree.c spelling fixes: mm/ 2007-10-20 01:27:18 +02:00
quicklist.c mm: size of quicklists shouldn't be proportional to the number of CPUs 2008-09-02 19:21:38 -07:00
readahead.c vmscan: split LRU lists into anon & file sets 2008-10-20 08:50:25 -07:00
rmap.c mm: further cleanup page_add_new_anon_rmap 2009-01-06 15:59:02 -08:00
shmem_acl.c [PATCH] sanitize ->permission() prototype 2008-07-26 20:53:14 -04:00
shmem.c mm: don't mark_page_accessed in shmem_fault 2009-01-06 15:58:58 -08:00
slab.c cpumask: convert mm/ 2009-01-01 10:12:29 +10:30
slob.c slob: do not pass the SLAB flags as GFP in kmem_cache_create() 2008-12-15 16:27:06 -08:00
slub.c cpumask: convert mm/ 2009-01-01 10:12:29 +10:30
sparse-vmemmap.c vmemmap: warn about page_structs with remote distance 2008-11-06 15:41:19 -08:00
sparse.c meminit section warnings 2008-11-30 10:03:35 -08:00
swap_state.c mm: remove gfp_mask from add_to_swap 2009-01-06 15:59:04 -08:00
swap.c mm: try_to_free_swap replaces remove_exclusive_swap_page 2009-01-06 15:59:03 -08:00
swapfile.c swapfile: let others seed random 2009-01-06 15:59:06 -08:00
thrash.c Bug in mm/thrash.c function grab_swap_token() 2007-05-11 08:29:32 -07:00
tiny-shmem.c Export tiny shmem_file_setup for DRM-GEM 2008-10-20 16:17:42 -07:00
truncate.c mmap: handle mlocked pages during map, remap, unmap 2008-10-20 08:52:31 -07:00
util.c mm: Make generic weak get_user_pages_fast and EXPORT_GPL it 2008-08-12 17:52:53 +10:00
vmalloc.c mm: vmalloc make lazy unmapping configurable 2009-01-06 15:59:01 -08:00
vmscan.c vmscan: improve reclaim throughput to bail out patch 2009-01-06 15:59:06 -08:00
vmstat.c cpumask: convert mm/ 2009-01-01 10:12:29 +10:30