1
0
Fork 0
alistair23-linux/mm
Joonsoo Kim 7cf91a98e6 mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous
There is a performance drop report due to hugepage allocation and in
there half of cpu time are spent on pageblock_pfn_to_page() in
compaction [1].

In that workload, compaction is triggered to make hugepage but most of
pageblocks are un-available for compaction due to pageblock type and
skip bit so compaction usually fails.  Most costly operations in this
case is to find valid pageblock while scanning whole zone range.  To
check if pageblock is valid to compact, valid pfn within pageblock is
required and we can obtain it by calling pageblock_pfn_to_page().  This
function checks whether pageblock is in a single zone and return valid
pfn if possible.  Problem is that we need to check it every time before
scanning pageblock even if we re-visit it and this turns out to be very
expensive in this workload.

Although we have no way to skip this pageblock check in the system where
hole exists at arbitrary position, we can use cached value for zone
continuity and just do pfn_to_page() in the system where hole doesn't
exist.  This optimization considerably speeds up in above workload.

Before vs After
  Max: 1096 MB/s vs 1325 MB/s
  Min: 635 MB/s 1015 MB/s
  Avg: 899 MB/s 1194 MB/s

Avg is improved by roughly 30% [2].

[1]: http://www.spinics.net/lists/linux-mm/msg97378.html
[2]: https://lkml.org/lkml/2015/12/9/23

[akpm@linux-foundation.org: don't forget to restore zone->contiguous on error path, per Vlastimil]
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Reported-by: Aaron Lu <aaron.lu@intel.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Aaron Lu <aaron.lu@intel.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-15 16:55:16 -07:00
..
kasan kasan: add functions to clear stack poison 2016-03-09 15:43:42 -08:00
Kconfig mm/Kconfig: correct description of DEFERRED_STRUCT_PAGE_INIT 2016-02-05 18:10:40 -08:00
Kconfig.debug mm/page_poisoning.c: allow for zero poisoning 2016-03-15 16:55:16 -07:00
Makefile mm/page_poison.c: enable PAGE_POISONING as a separate option 2016-03-15 16:55:16 -07:00
backing-dev.c mm/backing-dev.c: fix error path in wb_init() 2016-02-11 18:35:48 -08:00
balloon_compaction.c virtio_balloon: fix race between migration and ballooning 2016-01-12 20:47:06 +02:00
bootmem.c x86/mm: Introduce max_possible_pfn 2015-12-06 12:46:31 +01:00
cleancache.c cleancache: constify cleancache_ops structure 2016-01-27 09:09:57 -05:00
cma.c mm/cma.c: suppress warning 2015-11-05 19:34:48 -08:00
cma.h mm: cma: mark cma_bitmap_maxno() inline in header 2015-08-14 15:56:32 -07:00
cma_debug.c mm/cma_debug: correct size input to bitmap function 2015-07-17 16:39:54 -07:00
compaction.c mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous 2016-03-15 16:55:16 -07:00
debug.c mm, debug: move bad flags printing to bad_page() 2016-03-15 16:55:16 -07:00
dmapool.c mm, page_alloc: distinguish between being unable to sleep, unwilling to sleep and avoiding waking kswapd 2015-11-06 17:50:42 -08:00
early_ioremap.c mm/early_ioremap: use offset_in_page macro 2015-11-05 19:34:48 -08:00
fadvise.c writeback: implement and use inode_congested() 2015-06-02 08:33:35 -06:00
failslab.c mm: fault-inject take over bootstrap kmem_cache check 2016-03-15 16:55:16 -07:00
filemap.c mm: remove unnecessary uses of lock_page_memcg() 2016-03-15 16:55:16 -07:00
frame_vector.c mm: fix docbook comment for get_vaddr_frames() 2015-11-05 19:34:48 -08:00
frontswap.c frontswap: allow multiple backends 2015-06-24 17:49:45 -07:00
gup.c mm: retire GUP WARN_ON_ONCE that outlived its usefulness 2016-02-03 08:57:14 -08:00
highmem.c mm/highmem: make kmap cache coloring aware 2014-08-06 18:01:22 -07:00
huge_memory.c thp: cleanup split_huge_page() 2016-03-15 16:55:16 -07:00
hugetlb.c mm/hugetlb: use EOPNOTSUPP in hugetlb sysctl handlers 2016-03-09 15:43:42 -08:00
hugetlb_cgroup.c mm: make compound_head() robust 2015-11-06 17:50:42 -08:00
hwpoison-inject.c hwpoison: use page_cgroup_ino for filtering by memcg 2015-09-10 13:29:01 -07:00
init-mm.c
internal.h mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous 2016-03-15 16:55:16 -07:00
interval_tree.c mm: replace vma->sharead.linear with vma->shared 2015-02-10 14:30:31 -08:00
kmemcheck.c mm: kmemcheck skip object if slab allocation failed 2016-03-15 16:55:16 -07:00
kmemleak-test.c mm/kmemleak-test.c: use pr_fmt for logging 2014-06-06 16:08:18 -07:00
kmemleak.c Revert "gfp: add __GFP_NOACCOUNT" 2016-01-14 16:00:49 -08:00
ksm.c mm/ksm.c: mark stable page dirty 2016-01-15 17:56:32 -08:00
list_lru.c mm: memcontrol: move kmem accounting code to CONFIG_MEMCG 2016-01-20 17:09:18 -08:00
maccess.c mm/maccess.c: actually return -EFAULT from strncpy_from_unsafe 2015-11-05 19:34:48 -08:00
madvise.c mm/madvise: update comment on sys_madvise() 2016-03-15 16:55:16 -07:00
memblock.c mm/memblock.c: remove unnecessary memblock_type variable 2016-03-15 16:55:16 -07:00
memcontrol.c mm: simplify lock_page_memcg() 2016-03-15 16:55:16 -07:00
memory-failure.c mm/memory-failure.c: remove useless "undef"s 2016-03-15 16:55:16 -07:00
memory.c mm: use linear_page_index() in do_fault() 2016-03-15 16:55:16 -07:00
memory_hotplug.c mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous 2016-03-15 16:55:16 -07:00
mempolicy.c mm/mempolicy.c: skip VM_HUGETLB and VM_MIXEDMAP VMA for lazy mbind 2016-03-15 16:55:16 -07:00
mempool.c mm/mempool: avoid KASAN marking mempool poison checks as use-after-free 2016-03-11 16:17:47 -08:00
memtest.c memtest: remove unused header files 2015-09-08 15:35:28 -07:00
migrate.c mm: migrate: do not touch page->mem_cgroup of live pages 2016-03-15 16:55:16 -07:00
mincore.c thp: change pmd_trans_huge_lock() interface to return ptl 2016-01-21 17:20:51 -08:00
mlock.c mm: fix mlock accouting 2016-01-21 17:20:51 -08:00
mm_init.c mm: meminit: remove mminit_verify_page_links 2015-06-30 19:44:56 -07:00
mmap.c mm: fix regression in remap_file_pages() emulation 2016-02-18 16:23:24 -08:00
mmu_context.c sched/mm: call finish_arch_post_lock_switch in idle_task_exit and use_mm 2014-02-21 08:50:17 +01:00
mmu_notifier.c mmu-notifier: add clear_young callback 2015-09-10 13:29:01 -07:00
mmzone.c mm/mmzone.c: memmap_valid_within() can be boolean 2016-01-14 16:00:49 -08:00
mprotect.c mm, dax: check for pmd_none() after split_huge_pmd() 2016-02-11 18:35:48 -08:00
mremap.c mm, dax: check for pmd_none() after split_huge_pmd() 2016-02-11 18:35:48 -08:00
msync.c mm/msync: use offset_in_page macro 2015-11-05 19:34:48 -08:00
nobootmem.c x86/mm: Introduce max_possible_pfn 2015-12-06 12:46:31 +01:00
nommu.c kmemcg: account certain kmem allocations to memcg 2016-01-14 16:00:49 -08:00
oom_kill.c mm, oom: print symbolic gfp_flags in oom warning 2016-03-15 16:55:16 -07:00
page-writeback.c mm: remove unnecessary uses of lock_page_memcg() 2016-03-15 16:55:16 -07:00
page_alloc.c mm/compaction: speed up pageblock_pfn_to_page() when zone is contiguous 2016-03-15 16:55:16 -07:00
page_counter.c mm: page_counter: let page_counter_try_charge() return bool 2015-11-05 19:34:48 -08:00
page_ext.c mm/page_poisoning.c: allow for zero poisoning 2016-03-15 16:55:16 -07:00
page_idle.c mm: add page_check_address_transhuge() helper 2016-01-15 17:56:32 -08:00
page_io.c fs: use helper bio_add_page() instead of open coding on bi_io_vec 2015-08-13 12:32:00 -06:00
page_isolation.c mm/page_isolation: do some cleanup in "undo_isolate_page_range" 2016-01-15 17:56:32 -08:00
page_owner.c mm, page_owner: dump page owner info from dump_page() 2016-03-15 16:55:16 -07:00
page_poison.c mm/page_poisoning.c: allow for zero poisoning 2016-03-15 16:55:16 -07:00
pagewalk.c thp: rename split_huge_page_pmd() to split_huge_pmd() 2016-01-15 17:56:32 -08:00
percpu-km.c percpu: implmeent pcpu_nr_empty_pop_pages and chunk->nr_populated 2014-09-02 14:46:05 -04:00
percpu-vm.c percpu: move region iterations out of pcpu_[de]populate_chunk() 2014-09-02 14:46:02 -04:00
percpu.c tree wide: use kvfree() than conditional kfree()/vfree() 2016-01-22 17:02:18 -08:00
pgtable-generic.c mm,thp: fix spellos in describing __HAVE_ARCH_FLUSH_PMD_TLB_RANGE 2016-02-11 18:35:48 -08:00
process_vm_access.c ptrace: use fsuid, fsgid, effective creds for fs access checks 2016-01-20 17:09:18 -08:00
quicklist.c
readahead.c mm: move lru_to_page to mm_inline.h 2016-01-14 16:00:49 -08:00
rmap.c mm: simplify lock_page_memcg() 2016-03-15 16:55:16 -07:00
shmem.c mm: migrate: do not touch page->mem_cgroup of live pages 2016-03-15 16:55:16 -07:00
slab.c mm, sl[au]b: print gfp_flags as strings in slab_out_of_memory() 2016-03-15 16:55:16 -07:00
slab.h slub: convert SLAB_DEBUG_FREE to SLAB_CONSISTENCY_CHECKS 2016-03-15 16:55:16 -07:00
slab_common.c mm: new API kfree_bulk() for SLAB+SLUB allocators 2016-03-15 16:55:16 -07:00
slob.c mm: slab: free kmem_cache_node after destroy sysfs file 2016-02-18 16:23:24 -08:00
slub.c mm, sl[au]b: print gfp_flags as strings in slab_out_of_memory() 2016-03-15 16:55:16 -07:00
sparse-vmemmap.c x86, mm: introduce vmem_altmap to augment vmemmap_populate() 2016-01-15 17:56:32 -08:00
sparse.c x86, mm: introduce vmem_altmap to augment vmemmap_populate() 2016-01-15 17:56:32 -08:00
swap.c mm, x86: get_user_pages() for dax mappings 2016-01-15 17:56:32 -08:00
swap_cgroup.c mm: page_cgroup: rename file to mm/swap_cgroup.c 2014-12-10 17:41:09 -08:00
swap_state.c mm: memcontrol: charge swap to cgroup2 2016-01-20 17:09:18 -08:00
swapfile.c wrappers for ->i_mutex access 2016-01-22 18:04:28 -05:00
truncate.c mm: remove unnecessary uses of lock_page_memcg() 2016-03-15 16:55:16 -07:00
userfaultfd.c memcg: adjust to support new THP refcounting 2016-01-15 17:56:32 -08:00
util.c proc: revert /proc/<pid>/maps [stack:TID] annotation 2016-02-03 08:28:43 -08:00
vmacache.c mm/vmacache: inline vmacache_valid_mm() 2015-11-05 19:34:48 -08:00
vmalloc.c mm/vmalloc.c: use macro IS_ALIGNED to judge the aligment 2016-01-15 17:56:32 -08:00
vmpressure.c mm/vmpressure.c: fix subtree pressure detection 2016-02-03 08:28:43 -08:00
vmscan.c mm: remove unnecessary uses of lock_page_memcg() 2016-03-15 16:55:16 -07:00
vmstat.c mm, page_owner: convert page_owner_inited to static key 2016-03-15 16:55:16 -07:00
workingset.c mm: simplify lock_page_memcg() 2016-03-15 16:55:16 -07:00
zbud.c mm/zbud.c: use list_last_entry() instead of list_tail_entry() 2016-01-15 11:40:52 -08:00
zpool.c mm: zsmalloc: constify struct zs_pool name 2015-11-06 17:50:42 -08:00
zsmalloc.c zsmalloc: fix migrate_zspage-zs_free race condition 2016-01-20 17:09:18 -08:00
zswap.c mm/zswap: change incorrect strncmp use to strcmp 2015-12-18 14:25:40 -08:00