korg/linux
0
0
Fork 0
mirror of https://mirrors.bfsu.edu.cn/git/linux.git synced 2024-11-29 07:04:10 +08:00
Code Issues Packages Projects Releases Wiki Activity
dffcac2cb8
linux/mm
History
Shakeel Butt dffcac2cb8 mm/vmscan.c: prevent useless kswapd loops 
In production we have noticed hard lockups on large machines running
large jobs due to kswaps hoarding lru lock within isolate_lru_pages when
sc->reclaim_idx is 0 which is a small zone.  The lru was couple hundred
GiBs and the condition (page_zonenum(page) > sc->reclaim_idx) in
isolate_lru_pages() was basically skipping GiBs of pages while holding
the LRU spinlock with interrupt disabled.

On further inspection, it seems like there are two issues:

(1) If kswapd on the return from balance_pgdat() could not sleep (i.e.
    node is still unbalanced), the classzone_idx is unintentionally set
    to 0 and the whole reclaim cycle of kswapd will try to reclaim only
    the lowest and smallest zone while traversing the whole memory.

(2) Fundamentally isolate_lru_pages() is really bad when the
    allocation has woken kswapd for a smaller zone on a very large machine
    running very large jobs.  It can hoard the LRU spinlock while skipping
    over 100s of GiBs of pages.

This patch only fixes (1).  (2) needs a more fundamental solution.  To
fix (1), in the kswapd context, if pgdat->kswapd_classzone_idx is
invalid use the classzone_idx of the previous kswapd loop otherwise use
the one the waker has requested.

Link: http://lkml.kernel.org/r/20190701201847.251028-1-shakeelb@google.com
Fixes: e716f2eb24 ("mm, vmscan: prevent kswapd sleeping prematurely due to mismatched classzone_idx")
Signed-off-by: Shakeel Butt <shakeelb@google.com>
Reviewed-by: Yang Shi <yang.shi@linux.alibaba.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Roman Gushchin <guro@fb.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-07-05 11:12:07 +09:00
..
kasan kasan: initialize tag to 0xff in __kasan_kmalloc 2019-06-01 15:51:31 -07:00
backing-dev.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
balloon_compaction.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
cleancache.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 482 2019-06-19 17:09:52 +02:00
cma_debug.c mm/cma_debug.c: fix the break condition in cma_maxchunk_get() 2019-05-14 09:47:45 -07:00
cma.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 98 2019-05-24 17:37:54 +02:00
cma.h
compaction.c mm, compaction: make sure we isolate a valid PFN 2019-06-01 15:51:32 -07:00
debug_page_ref.c
debug.c mm: update references to page _refcount 2019-05-14 19:52:47 -07:00
dmapool.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 403 2019-06-05 17:37:13 +02:00
early_ioremap.c
fadvise.c vfs: implement readahead(2) using POSIX_FADV_WILLNEED 2018-08-30 20:01:32 +02:00
failslab.c mm: no need to check return value of debugfs_create functions 2019-03-05 21:07:17 -08:00
filemap.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
frame_vector.c
frontswap.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 482 2019-06-19 17:09:52 +02:00
gup_benchmark.c mm/gup: replace get_user_pages_longterm() with FOLL_LONGTERM 2019-05-14 09:47:45 -07:00
gup.c mm/gup: continue VM_FAULT_RETRY processing even for pre-faults 2019-06-01 15:51:31 -07:00
highmem.c mm: convert totalram_pages and totalhigh_pages variables to atomic 2018-12-28 12:11:47 -08:00
hmm.c mm/devm_memremap_pages: fix final page put race 2019-06-13 17:34:56 -10:00
huge_memory.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 499 2019-06-19 17:09:53 +02:00
hugetlb_cgroup.c mm: rename page_counter's count/limit into usage/max 2018-06-07 17:34:35 -07:00
hugetlb.c mm: hugetlb: soft-offline: dissolve_free_huge_page() return zero on !PageHuge 2019-06-29 16:43:45 +08:00
hwpoison-inject.c treewide: Add SPDX license identifier for more missed files 2019-05-21 10:50:45 +02:00
init-mm.c mm: Allocate the mm_cpumask (mm->cpu_bitmap[]) dynamically based on nr_cpu_ids 2018-07-17 09:35:30 +02:00
internal.h treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
interval_tree.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 248 2019-06-19 17:09:08 +02:00
Kconfig treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
Kconfig.debug treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
khugepaged.c coredump: fix race condition between collapse_huge_page() and core dumping 2019-06-13 17:34:56 -10:00
kmemleak-test.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 333 2019-06-05 17:37:06 +02:00
kmemleak.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 333 2019-06-05 17:37:06 +02:00
ksm.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 482 2019-06-19 17:09:52 +02:00
list_lru.c mm/list_lru.c: fix memory leak in __memcg_init_list_lru_node 2019-06-13 17:34:56 -10:00
maccess.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
madvise.c mm/mmu_notifier: use correct mmu_notifier events for each invalidation 2019-05-14 09:47:49 -07:00
Makefile mm: shuffle initial free memory to improve memory-side-cache utilization 2019-05-14 19:52:48 -07:00
memblock.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
memcontrol.c mm: memcontrol: don't batch updates of local VM stats and events 2019-06-13 17:34:56 -10:00
memfd.c mm: page cache: store only head pages in i_pages 2019-05-14 09:47:45 -07:00
memory_hotplug.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
memory-failure.c mm: hugetlb: soft-offline: dissolve_free_huge_page() return zero on !PageHuge 2019-06-29 16:43:45 +08:00
memory.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
mempolicy.c mm/mempolicy.c: fix an incorrect rebind node in mpol_rebind_nodemask 2019-06-29 16:43:44 +08:00
mempool.c docs/core-api/mm: fix return value descriptions in mm/ 2019-03-05 21:07:20 -08:00
memtest.c
migrate.c mm/mmu_notifier: use correct mmu_notifier events for each invalidation 2019-05-14 09:47:49 -07:00
mincore.c mm/mincore.c: make mincore() more conservative 2019-05-14 19:52:48 -07:00
mlock.c mm/mlock.c: change count_mm_mlocked_page_nr return type 2019-06-13 17:34:56 -10:00
mm_init.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
mmap.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
mmu_context.c
mmu_gather.c mm: mmu_gather: remove __tlb_reset_range() for force flush 2019-06-13 17:34:56 -10:00
mmu_notifier.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 499 2019-06-19 17:09:53 +02:00
mmzone.c
mprotect.c mm/mprotect.c: fix compilation warning because of unused 'mm' variable 2019-05-14 09:47:51 -07:00
mremap.c mm/mmu_notifier: contextual information for event triggering invalidation 2019-05-14 09:47:49 -07:00
msync.c
nommu.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
oom_kill.c mm/oom_kill.c: fix uninitialized oc->constraint 2019-06-29 16:43:45 +08:00
page_alloc.c mm/page_alloc.c: fix regression with deferred struct page init 2019-07-05 11:12:07 +09:00
page_counter.c memcg: introduce memory.min 2018-06-07 17:34:36 -07:00
page_ext.c memblock: drop memblock_alloc_*_nopanic() variants 2019-03-12 10:04:02 -07:00
page_idle.c mm/page_idle.c: fix oops because end_pfn is larger than max_pfn 2019-06-29 16:43:45 +08:00
page_io.c mm, swap: fix THP swap out 2019-06-29 16:43:45 +08:00
page_isolation.c mm/page_isolation.c: remove redundant pfn_valid_within() in __first_valid_page() 2019-05-14 09:47:46 -07:00
page_owner.c mm/page_owner: Simplify stack trace handling 2019-04-29 12:37:50 +02:00
page_poison.c page_poison: play nicely with KASAN 2019-03-05 21:07:13 -08:00
page_vma_mapped.c mm/rmap: map_pte() was not handling private ZONE_DEVICE page properly 2018-10-31 08:54:11 -07:00
page-writeback.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
pagewalk.c mm: kernel-doc: add missing parameter descriptions 2018-04-05 21:36:27 -07:00
percpu-internal.h percpu: convert chunk hints to be based on pcpu_block_md 2019-03-13 12:25:31 -07:00
percpu-km.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 428 2019-06-05 17:37:16 +02:00
percpu-stats.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 428 2019-06-05 17:37:16 +02:00
percpu-vm.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 428 2019-06-05 17:37:16 +02:00
percpu.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 428 2019-06-05 17:37:16 +02:00
pgtable-generic.c x86/mm: Page size aware flush_tlb_mm_range() 2018-10-09 16:51:11 +02:00
process_vm_access.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 2019-05-30 11:26:32 -07:00
quicklist.c
readahead.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
rmap.c mm/rmap.c: use the pra.mapcount to do the check 2019-05-14 09:47:49 -07:00
rodata_test.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 441 2019-06-05 17:37:17 +02:00
shmem.c mm: page cache: store only head pages in i_pages 2019-05-14 09:47:45 -07:00
shuffle.c mm: maintain randomization of page free lists 2019-05-14 19:52:48 -07:00
shuffle.h mm: maintain randomization of page free lists 2019-05-14 19:52:48 -07:00
slab_common.c mm: add support for kmem caches in DMA32 zone 2019-03-29 10:01:37 -07:00
slab.c slab: remove /proc/slab_allocators 2019-05-16 15:51:55 -07:00
slab.h mm: add support for kmem caches in DMA32 zone 2019-03-29 10:01:37 -07:00
slob.c slob: use slab_list instead of lru 2019-05-14 09:47:44 -07:00
slub.c mm/slub.c: update the comment about slab frozen 2019-05-14 09:47:45 -07:00
sparse-vmemmap.c mm: remove include/linux/bootmem.h 2018-10-31 08:54:16 -07:00
sparse.c mm/sparse.c: clean up obsolete code comment 2019-05-14 09:47:48 -07:00
swap_cgroup.c
swap_slots.c mm, swap, get_swap_pages: use entry_size instead of cluster in parameter 2018-08-22 10:52:44 -07:00
swap_state.c mm: page cache: store only head pages in i_pages 2019-05-14 09:47:45 -07:00
swap.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
swapfile.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
truncate.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
usercopy.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 500 2019-06-19 17:09:55 +02:00
userfaultfd.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 499 2019-06-19 17:09:53 +02:00
util.c prctl_set_mm: downgrade mmap_sem to read lock 2019-06-01 15:51:31 -07:00
vmacache.c mm: get rid of vmacache_flush_all() entirely 2018-09-13 15:18:04 -10:00
vmalloc.c mm/vmalloc.c: avoid bogus -Wmaybe-uninitialized warning 2019-06-29 16:43:45 +08:00
vmpressure.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 500 2019-06-19 17:09:55 +02:00
vmscan.c mm/vmscan.c: prevent useless kswapd loops 2019-07-05 11:12:07 +09:00
vmstat.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
workingset.c mm: memcontrol: make cgroup stats and events query API explicitly local 2019-05-14 19:52:53 -07:00
z3fold.c z3fold: fix sheduling while atomic 2019-06-01 15:51:31 -07:00
zbud.c treewide: Add SPDX license identifier for more missed files 2019-05-21 10:50:45 +02:00
zpool.c treewide: Add SPDX license identifier for more missed files 2019-05-21 10:50:45 +02:00
zsmalloc.c mm/zsmalloc.c: fix fall-through annotation 2018-10-26 16:26:35 -07:00
zswap.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 157 2019-05-30 11:26:37 -07:00