mirrors
/
linux-stable
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
1
0
Fork 0
Code Issues Releases Wiki Activity
linux-stable/mm
History
Baolin Wang eebb3dabbb mm: migrate: record the mlocked page status to remove unnecessary lru drain 
When doing compaction, I found the lru_add_drain() is an obvious hotspot
when migrating pages. The distribution of this hotspot is as follows:
   - 18.75% compact_zone
      - 17.39% migrate_pages
         - 13.79% migrate_pages_batch
            - 11.66% migrate_folio_move
               - 7.02% lru_add_drain
                  + 7.02% lru_add_drain_cpu
               + 3.00% move_to_new_folio
                 1.23% rmap_walk
            + 1.92% migrate_folio_unmap
         + 3.20% migrate_pages_sync
      + 0.90% isolate_migratepages

The lru_add_drain() was added by commit c3096e6782 ("mm/migrate:
__unmap_and_move() push good newpage to LRU") to drain the newpage to LRU
immediately, to help to build up the correct newpage->mlock_count in
remove_migration_ptes() for mlocked pages.  However, if there are no
mlocked pages are migrating, then we can avoid this lru drain operation,
especailly for the heavy concurrent scenarios.

So we can record the source pages' mlocked status in
migrate_folio_unmap(), and only drain the lru list when the mlocked status
is set in migrate_folio_move().

In addition, the page was already isolated from lru when migrating, so
checking the mlocked status is stable by folio_test_mlocked() in
migrate_folio_unmap().

After this patch, I can see the hotpot of the lru_add_drain() is gone:
   - 9.41% migrate_pages_batch
      - 6.15% migrate_folio_move
         - 3.64% move_to_new_folio
            + 1.80% migrate_folio_extra
            + 1.70% buffer_migrate_folio
         + 1.41% rmap_walk
         + 0.62% folio_add_lru
      + 3.07% migrate_folio_unmap

Meanwhile, the compaction latency shows some improvements when running
thpscale:
                            base                   patched
Amean     fault-both-1      1131.22 (   0.00%)     1112.55 *   1.65%*
Amean     fault-both-3      2489.75 (   0.00%)     2324.15 *   6.65%*
Amean     fault-both-5      3257.37 (   0.00%)     3183.18 *   2.28%*
Amean     fault-both-7      4257.99 (   0.00%)     4079.04 *   4.20%*
Amean     fault-both-12     6614.02 (   0.00%)     6075.60 *   8.14%*
Amean     fault-both-18    10607.78 (   0.00%)     8978.86 *  15.36%*
Amean     fault-both-24    14911.65 (   0.00%)    11619.55 *  22.08%*
Amean     fault-both-30    14954.67 (   0.00%)    14925.66 *   0.19%*
Amean     fault-both-32    16654.87 (   0.00%)    15580.31 *   6.45%*

Link: https://lkml.kernel.org/r/06e9153a7a4850352ec36602df3a3a844de45698.1697859741.git.baolin.wang@linux.alibaba.com
Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Yin Fengwei <fengwei.yin@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
 		2023-10-25 16:47:14 -07:00
..
damon mm/damon/sysfs: avoid empty scheme tried regions for large apply interval 2023-10-18 14:34:19 -07:00
kasan mm: delete checks for xor_unlock_is_negative_byte() 2023-10-18 14:34:17 -07:00
kfence LoongArch changes for v6.6 2023-09-08 12:16:52 -07:00
kmsan mm: kmsan: panic on failure to allocate early boot metadata 2023-10-25 16:47:10 -07:00
Kconfig mm: restrict the pcp batch scale factor to avoid too long latency 2023-10-25 16:47:10 -07:00
Kconfig.debug mm: page_table_check: Make it dependent on EXCLUSIVE_SYSTEM_RAM 2023-05-29 16:14:28 +01:00
Makefile mm: vmscan: move shrinker-related code into a separate file 2023-10-04 10:32:23 -07:00
backing-dev.c writeback: remove redundant checks for root memcg 2023-08-21 13:37:48 -07:00
balloon_compaction.c
…
bootmem_info.c bootmem: use kmemleak_free_part_phys in put_page_bootmem 2023-10-25 16:47:13 -07:00
cma.c mm/cma: use nth_page() in place of direct struct page manipulation 2023-10-04 10:32:29 -07:00
cma.h
…
cma_debug.c
…
cma_sysfs.c mm: cma: make kobj_type structure constant 2023-03-28 16:20:06 -07:00
compaction.c mm/compaction: factor out code to test if we should run compaction for target order 2023-10-04 10:32:19 -07:00
debug.c mm: update validate_mm() to use vma iterator 2023-06-09 16:25:31 -07:00
debug_page_alloc.c mm: page_alloc: split out DEBUG_PAGEALLOC 2023-06-09 16:25:23 -07:00
debug_page_ref.c
…
debug_vm_pgtable.c mm: fix multiple typos in multiple files 2023-10-25 16:47:14 -07:00
dmapool.c dmapool: create/destroy cleanup 2023-06-09 16:25:17 -07:00
dmapool_test.c dmapool: add alloc/free performance test 2023-04-05 19:42:38 -07:00
early_ioremap.c mm/early_ioremap.c: improve the execution efficiency of early_ioremap_setup() 2023-06-09 16:25:56 -07:00
fadvise.c mm: remove unnecessary pagevec includes 2023-06-23 16:59:31 -07:00
fail_page_alloc.c mm: page_alloc: split out FAIL_PAGE_ALLOC 2023-06-09 16:25:23 -07:00
failslab.c
…
filemap.c mm: drop the assumption that VM_SHARED always implies writable 2023-10-18 14:34:19 -07:00
folio-compat.c filemap: Add fgf_t typedef 2023-07-24 18:04:30 -04:00
gup.c mm/gup: make failure to pin an error if FOLL_NOWAIT not specified 2023-10-18 14:34:15 -07:00
gup_test.c Merge mm-hotfixes-stable into mm-stable to pick up depended-upon changes. 2023-06-23 16:58:19 -07:00
gup_test.h
…
highmem.c mm: ptep_get() conversion 2023-06-19 16:19:25 -07:00
hmm.c mm: enable page walking API to lock vmas during the walk 2023-08-21 13:07:20 -07:00
huge_memory.c mm: huge_memory: use folio_xchg_last_cpupid() in __split_huge_page_tail() 2023-10-25 16:47:12 -07:00
hugetlb.c hugetlb_vmemmap: use folio argument for hugetlb_vmemmap_* functions 2023-10-25 16:47:08 -07:00
hugetlb_cgroup.c mm, hugetlb: remove HUGETLB_CGROUP_MIN_ORDER 2023-10-18 14:34:17 -07:00
hugetlb_vmemmap.c hugetlb_vmemmap: use folio argument for hugetlb_vmemmap_* functions 2023-10-25 16:47:08 -07:00
hugetlb_vmemmap.h mm: hugetlb_vmemmap: fix reference to nonexistent file 2023-10-25 16:47:14 -07:00
hwpoison-inject.c
…
init-mm.c mm: move dummy_vm_ops out of a header 2023-08-21 13:37:46 -07:00
internal.h mm: fix multiple typos in multiple files 2023-10-25 16:47:14 -07:00
interval_tree.c
…
io-mapping.c
…
ioremap.c mm: ioremap: remove unneeded ioremap_allowed and iounmap_allowed 2023-08-18 10:12:36 -07:00
khugepaged.c mm/khugepaged: convert collapse_pte_mapped_thp() to use folios 2023-10-25 16:47:14 -07:00
kmemleak.c mm/kmemleak: move the initialisation of object to __link_object 2023-10-25 16:47:13 -07:00
ksm.c mm/ksm: add pages_skipped metric 2023-10-16 15:44:39 -07:00
list_lru.c
…
maccess.c mm: Fix copy_from_user_nofault(). 2023-04-12 17:36:23 -07:00
madvise.c mm: drop the assumption that VM_SHARED always implies writable 2023-10-18 14:34:19 -07:00
mapping_dirty_helpers.c mm: fix clean_record_shared_mapping_range kernel-doc 2023-08-24 16:20:30 -07:00
memblock.c memblock: introduce MEMBLOCK_RSRV_NOINIT flag 2023-10-04 10:32:30 -07:00
memcontrol.c mm: fix multiple typos in multiple files 2023-10-25 16:47:14 -07:00
memfd.c memfd: drop warning for missing exec-related flags 2023-10-04 10:32:22 -07:00
memory-failure.c mm: convert DAX lock/unlock page to lock/unlock folio 2023-10-04 10:32:20 -07:00
memory-tiers.c dax, kmem: calculate abstract distance with general interface 2023-10-16 15:44:39 -07:00
memory.c mm: use folio_xchg_last_cpupid() in wp_page_reuse() 2023-10-25 16:47:13 -07:00
memory_hotplug.c mm: memory_hotplug: drop memoryless node from fallback lists 2023-10-25 16:47:14 -07:00
mempolicy.c mm: abstract the vma_merge()/split_vma() pattern for mprotect() et al. 2023-10-18 14:34:18 -07:00
mempool.c
…
memremap.c mm/memremap.c: fix outdated comment in devm_memremap_pages 2023-02-09 16:51:46 -08:00
memtest.c mm: memtest: convert to memtest_report_meminfo() 2023-08-21 13:37:47 -07:00
migrate.c mm: migrate: record the mlocked page status to remove unnecessary lru drain 2023-10-25 16:47:14 -07:00
migrate_device.c Add x86 shadow stack support 2023-08-31 12:20:12 -07:00
mincore.c mm: enable page walking API to lock vmas during the walk 2023-08-21 13:07:20 -07:00
mlock.c mm: abstract the vma_merge()/split_vma() pattern for mprotect() et al. 2023-10-18 14:34:18 -07:00
mm_init.c mm: hugetlb: skip initialization of gigantic tail struct pages if freed by HVO 2023-10-04 10:32:30 -07:00
mm_slot.h
…
mmap.c mm: fix multiple typos in multiple files 2023-10-25 16:47:14 -07:00
mmap_lock.c
…
mmu_gather.c mm: fix kernel-doc warning from tlb_flush_rmaps() 2023-08-24 16:20:30 -07:00
mmu_notifier.c mmu_notifiers: rename invalidate_range notifier 2023-08-18 10:12:41 -07:00
mmzone.c mm: remove page_cpupid_xchg_last() 2023-10-25 16:47:13 -07:00
mprotect.c mm: mprotect: use a folio in change_pte_range() 2023-10-25 16:47:12 -07:00
mremap.c mm: abstract VMA merge and extend into vma_merge_extend() helper 2023-10-18 14:34:18 -07:00
msync.c
…
nommu.c mm: make vma_merge() and split_vma() internal 2023-10-18 14:34:18 -07:00
oom_kill.c mm/oom_killer: simplify OOM killer info dump helper 2023-10-25 16:47:10 -07:00
page-writeback.c mm: use folio_xor_flags_has_waiters() in folio_end_writeback() 2023-10-18 14:34:17 -07:00
page_alloc.c mm: page_alloc: check the order of compound page even when the order is zero 2023-10-25 16:47:14 -07:00
page_counter.c
…
page_ext.c mm/page_ext: move functions around for minor cleanups to page_ext 2023-08-18 10:12:31 -07:00
page_idle.c
…
page_io.c mm: memcg: add THP swap out info for anonymous reclaim 2023-10-04 10:32:27 -07:00
page_isolation.c mm/hugetlb: get rid of page_hstate() 2023-08-18 10:12:39 -07:00
page_owner.c mm/page_owner: remove free_ts from page_owner output 2023-10-18 14:34:19 -07:00
page_poison.c mm/page_poison: remove unused page_ext.h from page_poison 2023-08-21 13:37:30 -07:00
page_reporting.c mm, treewide: redefine MAX_ORDER sanely 2023-04-05 19:42:46 -07:00
page_reporting.h
…
page_table_check.c mm: convert page_table_check_pte_set() to page_table_check_ptes_set() 2023-08-24 16:20:18 -07:00
page_vma_mapped.c mm: correct stale comment of function check_pte 2023-08-18 10:12:13 -07:00
pagewalk.c mm/pagewalk: fix bootstopping regression from extra pte_unmap() 2023-09-02 08:39:21 -07:00
percpu-internal.h percpu-internal/pcpu_chunk: re-layout pcpu_chunk structure to reduce false sharing 2023-06-19 16:19:29 -07:00
percpu-km.c
…
percpu-stats.c
…
percpu-vm.c
…
percpu.c percpu: scoped objcg protection 2023-10-25 16:47:11 -07:00
pgalloc-track.h
…
pgtable-generic.c mm/pgtable: notes on pte_offset_map[_lock]() 2023-08-18 10:12:25 -07:00
process_vm_access.c mm/gup: remove unused vmas parameter from pin_user_pages_remote() 2023-06-09 16:25:25 -07:00
ptdump.c mm: ptdump should use ptep_get_lockless() 2023-06-19 16:19:24 -07:00
readahead.c filemap: Allow __filemap_get_folio to allocate large folios 2023-07-24 18:04:30 -04:00
rmap.c mm/rmap: convert page_move_anon_rmap() to folio_move_anon_rmap() 2023-10-18 14:34:14 -07:00
rodata_test.c
…
secretmem.c mm/secretmem: use a folio in secretmem_fault() 2023-08-21 13:38:02 -07:00
shmem.c mm: update memfd seal write check to include F_SEAL_WRITE 2023-10-18 14:34:19 -07:00
shmem_quota.c shmem: Add default quota limit mount options 2023-08-09 09:15:40 +02:00
show_mem.c mm: refactor si_mem_available() 2023-10-04 10:32:19 -07:00
shrinker.c mm: shrinker: convert shrinker_rwsem to mutex 2023-10-04 10:32:26 -07:00
shrinker_debug.c mm: shrinker: convert shrinker_rwsem to mutex 2023-10-04 10:32:26 -07:00
shuffle.c
…
shuffle.h mm, treewide: redefine MAX_ORDER sanely 2023-04-05 19:42:46 -07:00
slab.c Randomized slab caches for kmalloc() 2023-07-18 10:07:47 +02:00
slab.h mm: kmem: scoped objcg protection 2023-10-25 16:47:11 -07:00
slab_common.c slab fixes for 6.6-rc4 2023-09-29 12:10:12 -07:00
slub.c mm/slub: remove freelist_dereference() 2023-07-14 09:57:21 +02:00
sparse-vmemmap.c mm/vmemmap: allow architectures to override how vmemmap optimization works 2023-08-18 10:12:53 -07:00
sparse.c mm/sparse: remove redundant judgments from macro for_each_present_section_nr 2023-08-18 10:12:14 -07:00
swap.c mm: remove references to pagevec 2023-06-23 16:59:30 -07:00
swap.h swap: remove remnants of polling from read_swap_cache_async 2023-08-24 16:20:16 -07:00
swap_cgroup.c
…
swap_slots.c
…
swap_state.c mm/swap: avoid a xa load for swapout path 2023-10-25 16:47:11 -07:00
swapfile.c mm/swap: inline folio_set_swap_entry() and folio_swap_entry() 2023-08-24 16:20:28 -07:00
truncate.c - Some swap cleanups from Ma Wupeng ("fix WARN_ON in add_to_avail_list") 2023-08-29 14:25:26 -07:00
usercopy.c mm: Fix copy_from_user_nofault(). 2023-04-12 17:36:23 -07:00
userfaultfd.c Add x86 shadow stack support 2023-08-31 12:20:12 -07:00
util.c nilfs2: convert nilfs_copy_page() to nilfs_copy_folio() 2023-10-25 16:47:09 -07:00
vmalloc.c mm: hugetlb: add huge page size param to set_huge_pte_at() 2023-09-29 17:20:47 -07:00
vmpressure.c net-memcg: Fix scope of sockmem pressure indicators 2023-08-16 12:21:32 +01:00
vmscan.c mm: multi-gen LRU: reuse some legacy trace events 2023-10-18 14:34:14 -07:00
vmstat.c mm: tune PCP high automatically 2023-10-25 16:47:10 -07:00
workingset.c mm: workingset: dynamically allocate the mm-shadow shrinker 2023-10-04 10:32:24 -07:00
z3fold.c mm/z3fold: remove obsolete comment for struct z3fold_pool 2023-08-21 13:37:51 -07:00
zbud.c mm: zswap: remove shrink from zpool interface 2023-06-19 16:19:27 -07:00
zpool.c mm: zswap: remove shrink from zpool interface 2023-06-19 16:19:27 -07:00
zsmalloc.c zsmalloc: use copy_page for full page copy 2023-10-18 14:34:16 -07:00
zswap.c mm: zswap: fix pool refcount bug around shrink_worker() 2023-10-18 12:12:40 -07:00