BackPup/Linux
1
0 Fork 0
mirror of https://github.com/torvalds/linux.git synced 2026-10-07 05:31:08 -04:00
Code Issues Packages Projects Releases Wiki Activity
1,233,815 Commits 1 Branch 949 Tags
v6.7-rc2
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Johannes Weiner 8b39d20ece sched: psi: fix unprivileged polling against cgroups 
519fabc7aa ("psi: remove 500ms min window size limitation for
triggers") breaks unprivileged psi polling on cgroups.

Historically, we had a privilege check for polling in the open() of a
pressure file in /proc, but were erroneously missing it for the open()
of cgroup pressure files.

When unprivileged polling was introduced in d82caa2735 ("sched/psi:
Allow unprivileged polling of N*2s period"), it needed to filter
privileges depending on the exact polling parameters, and as such
moved the CAP_SYS_RESOURCE check from the proc open() callback to
psi_trigger_create(). Both the proc files as well as cgroup files go
through this during write(). This implicitly added the missing check
for privileges required for HT polling for cgroups.

When 519fabc7aa ("psi: remove 500ms min window size limitation for
triggers") followed right after to remove further restrictions on the
RT polling window, it incorrectly assumed the cgroup privilege check
was still missing and added it to the cgroup open(), mirroring what we
used to do for proc files in the past.

As a result, unprivileged poll requests that would be supported now
get rejected when opening the cgroup pressure file for writing.

Remove the cgroup open() check. psi_trigger_create() handles it.

Fixes: 519fabc7aa ("psi: remove 500ms min window size limitation for triggers")
Reported-by: Luca Boccassi <[email protected]>
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Acked-by: Luca Boccassi <[email protected]>
Acked-by: Suren Baghdasaryan <[email protected]>
Cc: [email protected] # 6.5+
Link: https://lore.kernel.org/r/[email protected]
 2023-11-14 22:27:00 +01:00
Johannes Weiner 969d63e1af mm: zswap: fix pool refcount bug around shrink_worker() 
When a zswap store fails due to the limit, it acquires a pool reference
and queues the shrinker.  When the shrinker runs, it drops the reference. 
However, there can be multiple store attempts before the shrinker wakes up
and runs once.  This results in reference leaks and eventual saturation
warnings for the pool refcount.

Fix this by dropping the reference again when the shrinker is already
queued.  This ensures one reference per shrinker run.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 45190f01dd ("mm/zswap.c: add allocation hysteresis if pool limit is hit")
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Chris Mason <[email protected]>
Acked-by: Nhat Pham <[email protected]>
Cc: Vitaly Wool <[email protected]>
Cc: Domenico Cerasuolo <[email protected]>
Cc: <[email protected]>	[5.6+]
Signed-off-by: Andrew Morton <[email protected]>
 2023-10-18 12:12:40 -07:00
Johannes Weiner 9ea9cb00a8 mm: memcontrol: fix GFP_NOFS recursion in memory.high enforcement 
Breno and Josef report a deadlock scenario from cgroup reclaim
re-entering the filesystem:

[  361.546690] ======================================================
[  361.559210] WARNING: possible circular locking dependency detected
[  361.571703] 6.5.0-0_fbk700_debug_rc0_kbuilder_13159_gbf787a128001 #1 Tainted: G S          E
[  361.589704] ------------------------------------------------------
[  361.602277] find/9315 is trying to acquire lock:
[  361.611625] ffff88837ba140c0 (&delayed_node->mutex){+.+.}-{4:4}, at: __btrfs_release_delayed_node+0x68/0x4f0
[  361.631437]
[  361.631437] but task is already holding lock:
[  361.643243] ffff8881765b8678 (btrfs-tree-01){++++}-{4:4}, at: btrfs_tree_read_lock+0x1e/0x40

[  362.904457]  mutex_lock_nested+0x1c/0x30
[  362.912414]  __btrfs_release_delayed_node+0x68/0x4f0
[  362.922460]  btrfs_evict_inode+0x301/0x770
[  362.982726]  evict+0x17c/0x380
[  362.988944]  prune_icache_sb+0x100/0x1d0
[  363.005559]  super_cache_scan+0x1f8/0x260
[  363.013695]  do_shrink_slab+0x2a2/0x540
[  363.021489]  shrink_slab_memcg+0x237/0x3d0
[  363.050606]  shrink_slab+0xa7/0x240
[  363.083382]  shrink_node_memcgs+0x262/0x3b0
[  363.091870]  shrink_node+0x1a4/0x720
[  363.099150]  shrink_zones+0x1f6/0x5d0
[  363.148798]  do_try_to_free_pages+0x19b/0x5e0
[  363.157633]  try_to_free_mem_cgroup_pages+0x266/0x370
[  363.190575]  reclaim_high+0x16f/0x1f0
[  363.208409]  mem_cgroup_handle_over_high+0x10b/0x270
[  363.246678]  try_charge_memcg+0xaf2/0xc70
[  363.304151]  charge_memcg+0xf0/0x350
[  363.320070]  __mem_cgroup_charge+0x28/0x40
[  363.328371]  __filemap_add_folio+0x870/0xd50
[  363.371303]  filemap_add_folio+0xdd/0x310
[  363.399696]  __filemap_get_folio+0x2fc/0x7d0
[  363.419086]  pagecache_get_page+0xe/0x30
[  363.427048]  alloc_extent_buffer+0x1cd/0x6a0
[  363.435704]  read_tree_block+0x43/0xc0
[  363.443316]  read_block_for_search+0x361/0x510
[  363.466690]  btrfs_search_slot+0xc8c/0x1520

This is caused by the mem_cgroup_handle_over_high() not respecting the
gfp_mask of the allocation context.  We used to only call this function on
resume to userspace, where no locks were held.  But c9afe31ec4 ("memcg:
synchronously enforce memory.high for large overcharges") added a call
from the allocation context without considering the gfp.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: c9afe31ec4 ("memcg: synchronously enforce memory.high for large overcharges")
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Breno Leitao <[email protected]>
Reported-by: Josef Bacik <[email protected]>
Acked-by: Shakeel Butt <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: <[email protected]>	[5.17+]
Signed-off-by: Andrew Morton <[email protected]>
 2023-09-19 13:21:34 -07:00
Johannes Weiner 7b086755fb mm: page_alloc: fix CMA and HIGHATOMIC landing on the wrong buddy list 
Commit 4b23a68f95 ("mm/page_alloc: protect PCP lists with a spinlock")
bypasses the pcplist on lock contention and returns the page directly to
the buddy list of the page's migratetype.

For pages that don't have their own pcplist, such as CMA and HIGHATOMIC,
the migratetype is temporarily updated such that the page can hitch a ride
on the MOVABLE pcplist.  Their true type is later reassessed when flushing
in free_pcppages_bulk().  However, when lock contention is detected after
the type was already overridden, the bypass will then put the page on the
wrong buddy list.

Once on the MOVABLE buddy list, the page becomes eligible for fallbacks
and even stealing.  In the case of HIGHATOMIC, otherwise ineligible
allocations can dip into the highatomic reserves.  In the case of CMA, the
page can be lost from the CMA region permanently.

Use a separate pcpmigratetype variable for the pcplist override.  Use the
original migratetype when going directly to the buddy.  This fixes the bug
and should make the intentions more obvious in the code.

Originally sent here to address the HIGHATOMIC case:
https://lore.kernel.org/lkml/[email protected]/

Changelog updated in response to the CMA-specific bug report.

[[email protected]: updated changelog]
Link: https://lkml.kernel.org/r/[email protected]
Fixes: 4b23a68f95 ("mm/page_alloc: protect PCP lists with a spinlock")
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Joe Liu <[email protected]>
Reviewed-by: Vlastimil Babka <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-09-19 13:21:32 -07:00
Johannes Weiner 6f0df8e16e memcontrol: ensure memcg acquired by id is properly set up 
In the eviction recency check, we attempt to retrieve the memcg to which
the folio belonged when it was evicted, by the memcg id stored in the
shadow entry.  However, there is a chance that the retrieved memcg is not
the original memcg that has been killed, but a new one which happens to
have the same id.

This is a somewhat unfortunate, but acceptable and rare inaccuracy in the
heuristics.  However, if we retrieve this new memcg between its allocation
and when it is properly attached to the memcg hierarchy, we could run into
the following NULL pointer exception during the memcg hierarchy traversal
done in mem_cgroup_get_nr_swap_pages():

[ 155757.793456] BUG: kernel NULL pointer dereference, address: 00000000000000c0
[ 155757.807568] #PF: supervisor read access in kernel mode
[ 155757.818024] #PF: error_code(0x0000) - not-present page
[ 155757.828482] PGD 401f77067 P4D 401f77067 PUD 401f76067 PMD 0
[ 155757.839985] Oops: 0000 [#1] SMP
[ 155757.887870] RIP: 0010:mem_cgroup_get_nr_swap_pages+0x3d/0xb0
[ 155757.899377] Code: 29 19 4a 02 48 39 f9 74 63 48 8b 97 c0 00 00 00 48 8b b7 58 02 00 00 48 2b b7 c0 01 00 00 48 39 f0 48 0f 4d c6 48 39 d1 74 42 <48> 8b b2 c0 00 00 00 48 8b ba 58 02 00 00 48 2b ba c0 01 00 00 48
[ 155757.937125] RSP: 0018:ffffc9002ecdfbc8 EFLAGS: 00010286
[ 155757.947755] RAX: 00000000003a3b1c RBX: 000007ffffffffff RCX: ffff888280183000
[ 155757.962202] RDX: 0000000000000000 RSI: 0007ffffffffffff RDI: ffff888bbc2d1000
[ 155757.976648] RBP: 0000000000000001 R08: 000000000000000b R09: ffff888ad9cedba0
[ 155757.991094] R10: ffffea0039c07900 R11: 0000000000000010 R12: ffff888b23a7b000
[ 155758.005540] R13: 0000000000000000 R14: ffff888bbc2d1000 R15: 000007ffffc71354
[ 155758.019991] FS:  00007f6234c68640(0000) GS:ffff88903f9c0000(0000) knlGS:0000000000000000
[ 155758.036356] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 155758.048023] CR2: 00000000000000c0 CR3: 0000000a83eb8004 CR4: 00000000007706e0
[ 155758.062473] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 155758.076924] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 155758.091376] PKRU: 55555554
[ 155758.096957] Call Trace:
[ 155758.102016]  <TASK>
[ 155758.106502]  ? __die+0x78/0xc0
[ 155758.112793]  ? page_fault_oops+0x286/0x380
[ 155758.121175]  ? exc_page_fault+0x5d/0x110
[ 155758.129209]  ? asm_exc_page_fault+0x22/0x30
[ 155758.137763]  ? mem_cgroup_get_nr_swap_pages+0x3d/0xb0
[ 155758.148060]  workingset_test_recent+0xda/0x1b0
[ 155758.157133]  workingset_refault+0xca/0x1e0
[ 155758.165508]  filemap_add_folio+0x4d/0x70
[ 155758.173538]  page_cache_ra_unbounded+0xed/0x190
[ 155758.182919]  page_cache_sync_ra+0xd6/0x1e0
[ 155758.191738]  filemap_read+0x68d/0xdf0
[ 155758.199495]  ? mlx5e_napi_poll+0x123/0x940
[ 155758.207981]  ? __napi_schedule+0x55/0x90
[ 155758.216095]  __x64_sys_pread64+0x1d6/0x2c0
[ 155758.224601]  do_syscall_64+0x3d/0x80
[ 155758.232058]  entry_SYSCALL_64_after_hwframe+0x46/0xb0
[ 155758.242473] RIP: 0033:0x7f62c29153b5
[ 155758.249938] Code: e8 48 89 75 f0 89 7d f8 48 89 4d e0 e8 b4 e6 f7 ff 41 89 c0 4c 8b 55 e0 48 8b 55 e8 48 8b 75 f0 8b 7d f8 b8 11 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 33 44 89 c7 48 89 45 f8 e8 e7 e6 f7 ff 48 8b
[ 155758.288005] RSP: 002b:00007f6234c5ffd0 EFLAGS: 00000293 ORIG_RAX: 0000000000000011
[ 155758.303474] RAX: ffffffffffffffda RBX: 00007f628c4e70c0 RCX: 00007f62c29153b5
[ 155758.318075] RDX: 000000000003c041 RSI: 00007f61d2986000 RDI: 0000000000000076
[ 155758.332678] RBP: 00007f6234c5fff0 R08: 0000000000000000 R09: 0000000064d5230c
[ 155758.347452] R10: 000000000027d450 R11: 0000000000000293 R12: 000000000003c041
[ 155758.362044] R13: 00007f61d2986000 R14: 00007f629e11b060 R15: 000000000027d450
[ 155758.376661]  </TASK>

This patch fixes the issue by moving the memcg's id publication from the
alloc stage to online stage, ensuring that any memcg acquired via id must
be connected to the memcg tree.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: f78dfc7b77 ("workingset: fix confusion around eviction vs refault container")
Signed-off-by: Johannes Weiner <[email protected]>
Co-developed-by: Nhat Pham <[email protected]>
Signed-off-by: Nhat Pham <[email protected]>
Acked-by: Shakeel Butt <[email protected]>
Cc: Yosry Ahmed <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-09-05 10:13:45 -07:00
Johannes Weiner f945116e4e mm: page_alloc: remove stale CMA guard code 
In the past, movable allocations could be disallowed from CMA through
PF_MEMALLOC_PIN.  As CMA pages are funneled through the MOVABLE pcplist,
this required filtering that cornercase during allocations, such that
pinnable allocations wouldn't accidentally get a CMA page.

However, since 8e3560d963 ("mm: honor PF_MEMALLOC_PIN for all movable
pages"), PF_MEMALLOC_PIN automatically excludes __GFP_MOVABLE.  Once
again, MOVABLE implies CMA is allowed.

Remove the stale filtering code.  Also remove a stale comment that was
introduced as part of the filtering code, because the filtering let
order-0 pages fall through to the buddy allocator.  See 1d91df85f3
("mm/page_alloc: handle a missing case for memalloc_nocma_{save/restore}
APIs") for context.  The comment's been obsolete since the introduction of
the explicit ALLOC_HIGHATOMIC flag in eb2e2b425c ("mm/page_alloc:
explicitly record high-order atomic allocations in alloc_flags").

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Miaohe Lin <[email protected]>
Cc: Pasha Tatashin <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-09-02 15:17:34 -07:00
Johannes Weiner 98804a944a mm: zswap: kill zswap_get_swap_cache_page() 
The __read_swap_cache_async() interface isn't more difficult to understand
than what the helper abstracts.  Save the indirection and a level of
indentation for the primary work of the writeback func.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Yosry Ahmed <[email protected]>
Cc: Vitaly Wool <[email protected]>
Cc: Barry Song <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Domenico Cerasuolo <[email protected]>
Cc: Nhat Pham <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-08-21 13:37:28 -07:00
Johannes Weiner 7310895779 mm: zswap: tighten up entry invalidation 
Removing a zswap entry from the tree is tied to an explicit operation
that's supposed to drop the base reference: swap invalidation, exclusive
load, duplicate store.  Don't silently remove the entry on final put, but
instead warn if an entry is in tree without reference.

While in that diff context, convert a BUG_ON to a WARN_ON_ONCE.  No need
to crash on a refcount underflow.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Yosry Ahmed <[email protected]>
Cc: Barry Song <[email protected]>
Cc: Domenico Cerasuolo <[email protected]>
Cc: Nhat Pham <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Vitaly Wool <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-08-21 13:37:28 -07:00
Johannes Weiner 56c67049c0 mm: zswap: use zswap_invalidate_entry() for duplicates 
Patch series "mm: zswap: three cleanups".

Three small cleanups to zswap, the first one suggested by Yosry during the
frontswap removal.


This patch (of 3):

Minor cleanup.  Instead of open-coding the tree deletion and the put, use
the zswap_invalidate_entry() convenience helper.

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Suggested-by: Yosry Ahmed <[email protected]>
Reviewed-by: Yosry Ahmed <[email protected]>
Cc: Domenico Cerasuolo <[email protected]>
Cc: Nhat Pham <[email protected]>
Cc: Barry Song <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Vitaly Wool <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-08-21 13:37:28 -07:00
Johannes Weiner 42c06a0e8e mm: kill frontswap 
The only user of frontswap is zswap, and has been for a long time.  Have
swap call into zswap directly and remove the indirection.

[[email protected]: remove obsolete comment, per Yosry]
  Link: https://lkml.kernel.org/r/[email protected]
[[email protected]: don't warn if none swapcache folio is passed to zswap_load]
  Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Yin Fengwei <[email protected]>
Acked-by: Konrad Rzeszutek Wilk <[email protected]>
Acked-by: Nhat Pham <[email protected]>
Acked-by: Yosry Ahmed <[email protected]>
Acked-by: Christoph Hellwig <[email protected]>
Cc: Domenico Cerasuolo <[email protected]>
Cc: Matthew Wilcox (Oracle) <[email protected]>
Cc: Vitaly Wool <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-08-21 13:37:26 -07:00
Johannes Weiner fac2650276 selftests: cgroup: fix test_kmem_basic false positives 
This test fails routinely in our prod testing environment, and I can
reproduce it locally as well.

The test allocates dcache inside a cgroup, then drops the memory limit
and checks that usage drops correspondingly. The reason it fails is
because dentries are freed with an RCU delay - a debugging sleep shows
that usage drops as expected shortly after.

Insert a 1s sleep after dropping the limit. This should be good
enough, assuming that machines running those tests are otherwise not
very busy.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Paul E. McKenney <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-08-04 13:03:42 -07:00
Johannes Weiner 493614da0d mm: compaction: fix endless looping over same migrate block 
During stress testing, the following situation was observed:

     70 root      39  19       0      0      0 R 100.0   0.0 959:29.92 khugepaged
 310936 root      20   0   84416  25620    512 R  99.7   1.5 642:37.22 hugealloc

Tracing shows isolate_migratepages_block() endlessly looping over the
first block in the DMA zone:

       hugealloc-310936  [001] ..... 237297.415718: mm_compaction_finished: node=0 zone=DMA      order=9 ret=no_suitable_page
       hugealloc-310936  [001] ..... 237297.415718: mm_compaction_isolate_migratepages: range=(0x1 ~ 0x400) nr_scanned=513 nr_taken=0
       hugealloc-310936  [001] ..... 237297.415718: mm_compaction_finished: node=0 zone=DMA      order=9 ret=no_suitable_page
       hugealloc-310936  [001] ..... 237297.415718: mm_compaction_isolate_migratepages: range=(0x1 ~ 0x400) nr_scanned=513 nr_taken=0
       hugealloc-310936  [001] ..... 237297.415718: mm_compaction_finished: node=0 zone=DMA      order=9 ret=no_suitable_page
       hugealloc-310936  [001] ..... 237297.415718: mm_compaction_isolate_migratepages: range=(0x1 ~ 0x400) nr_scanned=513 nr_taken=0
       hugealloc-310936  [001] ..... 237297.415718: mm_compaction_finished: node=0 zone=DMA      order=9 ret=no_suitable_page
       hugealloc-310936  [001] ..... 237297.415718: mm_compaction_isolate_migratepages: range=(0x1 ~ 0x400) nr_scanned=513 nr_taken=0

The problem is that the functions tries to test and set the skip bit once
on the block, to avoid skipping on its own skip-set, using
pageblock_aligned() on the pfn as a test.  But because this is the DMA
zone which starts at pfn 1, this is never true for the first block, and
the skip bit isn't set or tested at all.  As a result,
fast_find_migrateblock() returns the same pageblock over and over.

If the pfn isn't pageblock-aligned, also check if it's the start of the
zone to ensure test-and-set-exactly-once on unaligned ranges.

Thanks to Vlastimil Babka for the help in debugging this.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 90ed667c03 ("Revert "Revert "mm/compaction: fix set skip in fast_find_migrateblock""")
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Vlastimil Babka <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Reviewed-by: Baolin Wang <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-08-04 13:03:42 -07:00
Johannes Weiner 113d0a6b39 MAINTAINERS: Add Peter explicitly to the psi section 
Peter is kind enough to route the low-volume psi patches through the
scheduler tree, but he is frequently not CC'd on them.

While he is matched through the SCHEDULER maintainers and reviewers on
kern/sched/*, that list is long, and mostly not applicable to psi
code. Thus, patch submitters often just CC the explicit PSI entries.

Add him to that section, to make sure he gets those patches.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
 2023-08-02 16:19:25 +02:00
Johannes Weiner ce5df7764b mm: page_isolation: write proper kerneldoc 
And remove the incorrect header comments.

[[email protected]: s/lower/first/, s/upper/last/, per Mike]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Cc: Mike Rapoport <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-06-19 16:18:59 -07:00
Johannes Weiner 4fbbb3fde3 mm: compaction: avoid GFP_NOFS ABBA deadlock 
During stress testing with higher-order allocations, a deadlock scenario
was observed in compaction: One GFP_NOFS allocation was sleeping on
mm/compaction.c::too_many_isolated(), while all CPUs in the system were
busy with compactors spinning on buffer locks held by the sleeping
GFP_NOFS allocation.

Reclaim is susceptible to this same deadlock; we fixed it by granting
GFP_NOFS allocations additional LRU isolation headroom, to ensure it makes
forward progress while holding fs locks that other reclaimers might
acquire.  Do the same here.

This code has been like this since compaction was initially merged, and I
only managed to trigger this with out-of-tree patches that dramatically
increase the contexts that do GFP_NOFS compaction.  While the issue is
real, it seems theoretical in nature given existing allocation sites. 
Worth fixing now, but no Fixes tag or stable CC.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-06-09 16:25:37 -07:00
Johannes Weiner 3cf0493752 mm: compaction: have compaction_suitable() return bool 
Since it only returns COMPACT_CONTINUE or COMPACT_SKIPPED now, a bool
return value simplifies the callsites.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Suggested-by: Vlastimil Babka <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Baolin Wang <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-06-09 16:25:37 -07:00
Johannes Weiner 1c9568e806 mm: compaction: drop redundant watermark check in compaction_zonelist_suitable() 
The watermark check in compaction_zonelist_suitable(), called from
should_compact_retry(), is sandwiched between two watermark checks
already: before, there are freelist attempts as part of direct reclaim and
direct compaction; after, there is a last-minute freelist attempt in
__alloc_pages_may_oom().

The check in compaction_zonelist_suitable() isn't necessary. Kill it.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Baolin Wang <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-06-09 16:25:37 -07:00
Johannes Weiner f98a497e1f mm: compaction: remove unnecessary is_via_compact_memory() checks 
Remove from all paths not reachable via /proc/sys/vm/compact_memory.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Baolin Wang <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-06-09 16:25:36 -07:00
Johannes Weiner e8606320e9 mm: compaction: refactor __compaction_suitable() 
__compaction_suitable() is supposed to check for available migration
targets.  However, it also checks whether the operation was requested via
/proc/sys/vm/compact_memory, and whether the original allocation request
can already succeed.  These don't apply to all callsites.

Move the checks out to the callers, so that later patches can deal with
them one by one.  No functional change intended.

[[email protected]: fix comment, per Vlastimil]
  Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-06-09 16:25:36 -07:00
Johannes Weiner 511a69b27f mm: compaction: simplify should_compact_retry() 
The different branches for retry are unnecessarily complicated.  There are
really only three outcomes: progress (retry n times), skipped (retry if
reclaim can help), failed (retry with higher priority).

Rearrange the branches and the retry counter to make it simpler.

[[email protected]: restore behavior when hitting max_retries]
  Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-06-09 16:25:36 -07:00
Johannes Weiner ecd8b2928f mm: compaction: remove compaction result helpers 
Patch series "mm: compaction: cleanups & simplifications".

These compaction cleanups are split out from the huge page allocator
series[1], as requested by reviewer feedback.

[1] https://lore.kernel.org/linux-mm/[email protected]/


This patch (of 5):

The compaction result helpers encode quirks that are specific to the
allocator's retry logic.  E.g.  COMPACT_SUCCESS and COMPACT_COMPLETE
actually represent failures that should be retried upon, and so on.  I
frequently found myself pulling up the helper implementation in order to
understand and work on the retry logic.  They're not quite clean
abstractions; rather they split the retry logic into two locations.

Remove the helpers and inline the checks.  Then comment on the result
interpretations directly where the decision making happens.

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-06-09 16:25:36 -07:00
Johannes Weiner f78dfc7b77 workingset: fix confusion around eviction vs refault container 
Refault decisions are made based on the lruvec where the page was evicted,
as that determined its LRU order while it was alive.  Stats and workingset
aging must then occur on the lruvec of the new page, as that's the node
and cgroup that experience the refault and that's the lruvec whose
nonresident info ages out by a new resident page.  Those lruvecs could be
different when a page is shared between cgroups, or the refaulting page is
allocated on a different node.

There are currently two mix-ups:

1. When swap is available, the resident anon set must be considered
   when comparing the refault distance. The comparison is made against
   the right anon set, but the check for swap is not. When pages get
   evicted from a cgroup with swap, and refault in one without, this
   can incorrectly consider a hot refault as cold - and vice
   versa. Fix that by using the eviction cgroup for the swap check.

2. The stats and workingset age are updated against the wrong lruvec
   altogether: the right cgroup but the wrong NUMA node. When a page
   refaults on a different NUMA node, this will have confusing stats
   and distort the workingset age on a different lruvec - again
   possibly resulting in hot/cold misclassifications down the line.

Fix the swap check and the refault pgdat to address both concerns.

This was found during code review.  It hasn't caused notable issues in
production, suggesting that those refault-migrations are relatively rare
in practice.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Co-developed-by: Nhat Pham <[email protected]>
Signed-off-by: Nhat Pham <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-01-18 17:12:53 -08:00
Johannes Weiner da34a8484d mm: memcontrol: deprecate charge moving 
Charge moving mode in cgroup1 allows memory to follow tasks as they
migrate between cgroups.  This is, and always has been, a questionable
thing to do - for several reasons.

First, it's expensive.  Pages need to be identified, locked and isolated
from various MM operations, and reassigned, one by one.

Second, it's unreliable.  Once pages are charged to a cgroup, there isn't
always a clear owner task anymore.  Cache isn't moved at all, for example.
Mapped memory is moved - but if trylocking or isolating a page fails,
it's arbitrarily left behind.  Frequent moving between domains may leave a
task's memory scattered all over the place.

Third, it isn't really needed.  Launcher tasks can kick off workload tasks
directly in their target cgroup.  Using dedicated per-workload groups
allows fine-grained policy adjustments - no need to move tasks and their
physical pages between control domains.  The feature was never
forward-ported to cgroup2, and it hasn't been missed.

Despite it being a niche usecase, the maintenance overhead of supporting
it is enormous.  Because pages are moved while they are live and subject
to various MM operations, the synchronization rules are complicated. 
There are lock_page_memcg() in MM and FS code, which non-cgroup people
don't understand.  In some cases we've been able to shift code and cgroup
API calls around such that we can rely on native locking as much as
possible.  But that's fragile, and sometimes we need to hold MM locks for
longer than we otherwise would (pte lock e.g.).

Mark the feature deprecated. Hopefully we can remove it soon.

And backport into -stable kernels so that people who develop against
earlier kernels are warned about this deprecation as early as possible.

[[email protected]: fix memory.rst underlining]
Link: https://lkml.kernel.org/r/Y5COd+qXwk/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Shakeel Butt <[email protected]>
Acked-by: Hugh Dickins <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-01-18 17:12:42 -08:00
Johannes Weiner c7c3dec1c9 mm: rmap: remove lock_page_memcg() 
The previous patch made sure charge moving only touches pages for which
page_mapped() is stable.  lock_page_memcg() is no longer needed.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Hugh Dickins <[email protected]>
Acked-by: Shakeel Butt <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Muchun Song <[email protected]>
Cc: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-01-18 17:12:42 -08:00
Johannes Weiner 4e0cf05f60 mm: memcontrol: skip moving non-present pages that are mapped elsewhere 
Patch series "mm: push down lock_page_memcg()", v2.


This patch (of 3):

During charge moving, the pte lock and the page lock cover nearly all
cases of stabilizing page_mapped().  The only exception is when we're
looking at a non-present pte and find a page in the page cache or in the
swapcache: if the page is mapped elsewhere, it can become unmapped outside
of our control.  For this reason, rmap needs lock_page_memcg().

We don't like cgroup-specific locks in generic MM code - especially in
performance-critical MM code - and for a legacy feature that's unlikely to
have many users left - if any.

So remove the exception.  Arguably that's better semantics anyway: the
page is shared, and another process seems to be the more active user.

Once we stop moving such pages, rmap doesn't need lock_page_memcg()
anymore.  The next patch will remove it.

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Suggested-by: Hugh Dickins <[email protected]>
Acked-by: Hugh Dickins <[email protected]>
Acked-by: Shakeel Butt <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Muchun Song <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2023-01-18 17:12:42 -08:00
Johannes Weiner 6a05aa3010 zpool: clean out dead code 
There is a lot of provision for flexibility that isn't actually needed or
used.  Zswap (the only zpool user) always passes zpool_ops with an .evict
method set.  The backends who reclaim only do so for zswap, so they can
also directly call zpool_ops without indirection or checks.

Finally, there is no need to check the retries parameters and bail with
-EINVAL in the reclaim function, when that's called just a few lines below
with a hard-coded 8.  There is no need to duplicate the evictable and
sleep_mapped attrs from the driver in zpool_ops.

Link: https://lkml.kernel.org/r/[email protected]
Reviewed-by: Sergey Senozhatsky <[email protected]>
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Nhat Pham <[email protected]>
Cc: Dan Streetman <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Nitin Gupta <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Vitaly Wool <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-12-11 18:12:10 -08:00
Johannes Weiner 6b3379e8dc zswap: fix writeback lock ordering for zsmalloc 
Patch series "Implement writeback for zsmalloc", v7.

Unlike other zswap allocators such as zbud or z3fold, zsmalloc currently
lacks the writeback mechanism.  This means that when the zswap pool is
full, it will simply reject further allocations, and the pages will be
written directly to swap.

This series of patches implements writeback for zsmalloc. When the zswap
pool becomes full, zsmalloc will attempt to evict all the compressed
objects in the least-recently used zspages.


This patch (of 6):

zswap's customary lock order is tree->lock before pool->lock, because the
tree->lock protects the entries' refcount, and the free callbacks in the
backends acquire their respective pool locks to dispatch the backing
object.  zsmalloc's map callback takes the pool lock, so zswap must not
grab the tree->lock while a handle is mapped.  This currently only happens
during writeback, which isn't implemented for zsmalloc.  In preparation
for it, move the tree->lock section out of the mapped entry section

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Nhat Pham <[email protected]>
Reviewed-by: Sergey Senozhatsky <[email protected]>
Cc: Dan Streetman <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Nitin Gupta <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Vitaly Wool <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-12-11 18:12:09 -08:00
Johannes Weiner 57e9cc50f4 mm: vmscan: split khugepaged stats from direct reclaim stats 
Direct reclaim stats are useful for identifying a potential source for
application latency, as well as spotting issues with kswapd.  However,
khugepaged currently distorts the picture: as a kernel thread it doesn't
impose allocation latencies on userspace, and it explicitly opts out of
kswapd reclaim.  Its activity showing up in the direct reclaim stats is
misleading.  Counting it as kswapd reclaim could also cause confusion when
trying to understand actual kswapd behavior.

Break out khugepaged from the direct reclaim counters into new
pgsteal_khugepaged, pgdemote_khugepaged, pgscan_khugepaged counters.

Test with a huge executable (CONFIG_READ_ONLY_THP_FOR_FS):

pgsteal_kswapd 1342185
pgsteal_direct 0
pgsteal_khugepaged 3623
pgscan_kswapd 1345025
pgscan_direct 0
pgscan_khugepaged 3623

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Eric Bergen <[email protected]>
Cc: Matthew Wilcox (Oracle) <[email protected]>
Cc: Yang Shi <[email protected]>
Cc: Yosry Ahmed <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-11-30 15:58:41 -08:00
Johannes Weiner f53af4285d mm: vmscan: fix extreme overreclaim and swap floods 
During proactive reclaim, we sometimes observe severe overreclaim, with
several thousand times more pages reclaimed than requested.

This trace was obtained from shrink_lruvec() during such an instance:

    prio:0 anon_cost:1141521 file_cost:7767
    nr_reclaimed:4387406 nr_to_reclaim:1047 (or_factor:4190)
    nr=[7161123 345 578 1111]

While he reclaimer requested 4M, vmscan reclaimed close to 16G, most of it
by swapping.  These requests take over a minute, during which the write()
to memory.reclaim is unkillably stuck inside the kernel.

Digging into the source, this is caused by the proportional reclaim
bailout logic.  This code tries to resolve a fundamental conflict: to
reclaim roughly what was requested, while also aging all LRUs fairly and
in accordance to their size, swappiness, refault rates etc.  The way it
attempts fairness is that once the reclaim goal has been reached, it stops
scanning the LRUs with the smaller remaining scan targets, and adjusts the
remainder of the bigger LRUs according to how much of the smaller LRUs was
scanned.  It then finishes scanning that remainder regardless of the
reclaim goal.

This works fine if priority levels are low and the LRU lists are
comparable in size.  However, in this instance, the cgroup that is
targeted by proactive reclaim has almost no files left - they've already
been squeezed out by proactive reclaim earlier - and the remaining anon
pages are hot.  Anon rotations cause the priority level to drop to 0,
which results in reclaim targeting all of anon (a lot) and all of file
(almost nothing).  By the time reclaim decides to bail, it has scanned
most or all of the file target, and therefor must also scan most or all of
the enormous anon target.  This target is thousands of times larger than
the reclaim goal, thus causing the overreclaim.

The bailout code hasn't changed in years, why is this failing now?  The
most likely explanations are two other recent changes in anon reclaim:

1. Before the series starting with commit 5df741963d ("mm: fix LRU
   balancing effect of new transparent huge pages"), the VM was
   overall relatively reluctant to swap at all, even if swap was
   configured. This means the LRU balancing code didn't come into play
   as often as it does now, and mostly in high pressure situations
   where pronounced swap activity wouldn't be as surprising.

2. For historic reasons, shrink_lruvec() loops on the scan targets of
   all LRU lists except the active anon one, meaning it would bail if
   the only remaining pages to scan were active anon - even if there
   were a lot of them.

   Before the series starting with commit ccc5dc6734 ("mm/vmscan:
   make active/inactive ratio as 1:1 for anon lru"), most anon pages
   would live on the active LRU; the inactive one would contain only a
   handful of preselected reclaim candidates. After the series, anon
   gets aged similarly to file, and the inactive list is the default
   for new anon pages as well, making it often the much bigger list.

   As a result, the VM is now more likely to actually finish large
   anon targets than before.

Change the code such that only one SWAP_CLUSTER_MAX-sized nudge toward the
larger LRU lists is made before bailing out on a met reclaim goal.

This fixes the extreme overreclaim problem.

Fairness is more subtle and harder to evaluate.  No obvious misbehavior
was observed on the test workload, in any case.  Conceptually, fairness
should primarily be a cumulative effect from regular, lower priority
scans.  Once the VM is in trouble and needs to escalate scan targets to
make forward progress, fairness needs to take a backseat.  This is also
acknowledged by the myriad exceptions in get_scan_count().  This patch
makes fairness decrease gradually, as it keeps fairness work static over
increasing priority levels with growing scan targets.  This should make
more sense - although we may have to re-visit the exact values.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Rik van Riel <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-11-22 18:50:41 -08:00
Johannes Weiner 0538a82c39 mm: vmscan: make rotations a secondary factor in balancing anon vs file 
We noticed a 2% webserver throughput regression after upgrading from 5.6. 
This could be tracked down to a shift in the anon/file reclaim balance
(confirmed with swappiness) that resulted in worse reclaim efficiency and
thus more kswapd activity for the same outcome.

The change that exposed the problem is aae466b005 ("mm/swap: implement
workingset detection for anonymous LRU").  By qualifying swapins based on
their refault distance, it lowered the cost of anon reclaim in this
workload, in turn causing (much) more anon scanning than before.  Scanning
the anon list is more expensive due to the higher ratio of mmapped pages
that may rotate during reclaim, and so the result was an increase in %sys
time.

Right now, rotations aren't considered a cost when balancing scan pressure
between LRUs.  We can end up with very few file refaults putting all the
scan pressure on hot anon pages that are rotated en masse, don't get
reclaimed, and never push back on the file LRU again.  We still only
reclaim file cache in that case, but we burn a lot CPU rotating anon
pages.  It's "fair" from an LRU age POV, but doesn't reflect the real cost
it imposes on the system.

Consider rotations as a secondary factor in balancing the LRUs.  This
doesn't attempt to make a precise comparison between IO cost and CPU cost,
it just says: if reloads are about comparable between the lists, or
rotations are overwhelmingly different, adjust for CPU work.

This fixed the regression on our webservers.  It has since been deployed
to the entire Meta fleet and hasn't caused any problems.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Cc: Rik van Riel <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-11-08 17:37:11 -08:00
Johannes Weiner 82e60d00b7 fs: fix leaked psi pressure state 
When psi annotations were added to to btrfs compression reads, the psi
state tracking over add_ra_bio_pages and btrfs_submit_compressed_read was
faulty.  A pressure state, once entered, is never left.  This results in
incorrectly elevated pressure, which triggers OOM kills.

pflags record the *previous* memstall state when we enter a new one.  The
code tried to initialize pflags to 1, and then optimize the leave call
when we either didn't enter a memstall, or were already inside a nested
stall.  However, there can be multiple PageWorkingset pages in the bio, at
which point it's that path itself that enters repeatedly and overwrites
pflags.  This causes us to miss the exit.

Enter the stall only once if needed, then unwind correctly.

erofs has the same problem, fix that up too.  And move the memstall exit
past submit_bio() to restore submit accounting originally added by
b8e24a9300 ("block: annotate refault stalls from IO submission").

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 4088a47e78 ("btrfs: add manual PSI accounting for compressed reads")
Fixes: 99486c511f ("erofs: add manual PSI accounting for the compressed address space")
Fixes: 118f3663fb ("block: remove PSI accounting from the bio layer")
Link: https://lore.kernel.org/r/[email protected]/
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Thorsten Leemhuis <[email protected]>
Tested-by: Thorsten Leemhuis <[email protected]>
Cc: Chao Yu <[email protected]>
Cc: Chris Mason <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: David Sterba <[email protected]>
Cc: Gao Xiang <[email protected]>
Cc: Jens Axboe <[email protected]>
Cc: Josef Bacik <[email protected]>
Cc: Suren Baghdasaryan <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-11-08 15:57:25 -08:00
Johannes Weiner e55b9f9686 mm: memcontrol: drop dead CONFIG_MEMCG_SWAP config symbol 
Since 2d1c498072 ("mm: memcontrol: make swap tracking an integral part
of memory control"), CONFIG_MEMCG_SWAP hasn't been a user-visible config
option anymore, it just means CONFIG_MEMCG && CONFIG_SWAP.

Update the sites accordingly and drop the symbol.

[ While touching the docs, remove two references to CONFIG_MEMCG_KMEM,
  which hasn't been a user-visible symbol for over half a decade. ]

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Shakeel Butt <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-10-03 14:03:36 -07:00
Johannes Weiner b94c4e949c mm: memcontrol: use do_memsw_account() in a few more places 
It's slightly more descriptive and consistent with other places that
distinguish cgroup1's combined memory+swap accounting scheme from
cgroup2's dedicated swap accounting.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Shakeel Butt <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-10-03 14:03:36 -07:00
Johannes Weiner b25806dcd3 mm: memcontrol: deprecate swapaccounting=0 mode 
The swapaccounting= commandline option already does very little today.  To
close a trivial containment failure case, the swap ownership tracking part
of the swap controller has recently become mandatory (see commit
2d1c498072 ("mm: memcontrol: make swap tracking an integral part of
memory control") for details), which makes up the majority of the work
during swapout, swapin, and the swap slot map.

The only thing left under this flag is the page_counter operations and the
visibility of the swap control files in the first place, which are rather
meager savings.  There also aren't many scenarios, if any, where
controlling the memory of a cgroup while allowing it unlimited access to a
global swap space is a workable resource isolation strategy.

On the other hand, there have been several bugs and confusion around the
many possible swap controller states (cgroup1 vs cgroup2 behavior, memory
accounting without swap accounting, memcg runtime disabled).

This puts the maintenance overhead of retaining the toggle above its
practical benefits.  Deprecate it.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Suggested-by: Shakeel Butt <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-10-03 14:03:36 -07:00
Johannes Weiner c91bdc9358 mm: memcontrol: don't allocate cgroup swap arrays when memcg is disabled 
Patch series "memcg swap fix & cleanups".


This patch (of 4):

Since commit 2d1c498072 ("mm: memcontrol: make swap tracking an integral
part of memory control"), the cgroup swap arrays are used to track memory
ownership at the time of swap readahead and swapoff, even if swap space
*accounting* has been turned off by the user via swapaccount=0 (which sets
cgroup_memory_noswap).

However, the patch was overzealous: by simply dropping the
cgroup_memory_noswap conditionals in the swapon, swapoff and uncharge
path, it caused the cgroup arrays being allocated even when the memory
controller as a whole is disabled.  This is a waste of that memory.

Restore mem_cgroup_disabled() checks, implied previously by
cgroup_memory_noswap, in the swapon, swapoff, and swap_entry_free
callbacks.

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Fixes: 2d1c498072 ("mm: memcontrol: make swap tracking an integral part of memory control")
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Hugh Dickins <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Hugh Dickins <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-10-03 14:03:36 -07:00
Johannes Weiner 71dbdde791 sched/psi: Remove NR_ONCPU task accounting 
We put all fields updated by the scheduler in the first cacheline of
struct psi_group_cpu for performance.

Since we want add another PSI_IRQ_FULL to track IRQ/SOFTIRQ pressure,
we need to reclaim space first. This patch remove NR_ONCPU task accounting
in struct psi_group_cpu, use one bit in state_mask to track instead.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Chengming Zhou <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Chengming Zhou <[email protected]>
Tested-by: Chengming Zhou <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
 2022-09-09 11:08:32 +02:00
Johannes Weiner 3f1509c57b Revert "mm/vmscan: never demote for memcg reclaim" 
This reverts commit 3a235693d3.

Its premise was that cgroup reclaim cares about freeing memory inside the
cgroup, and demotion just moves them around within the cgroup limit. 
Hence, pages from toptier nodes should be reclaimed directly.

However, with NUMA balancing now doing tier promotions, demotion is part
of the page aging process.  Global reclaim demotes the coldest toptier
pages to secondary memory, where their life continues and from which they
have a chance to get promoted back.  Essentially, tiered memory systems
have an LRU order that spans multiple nodes.

When cgroup reclaims pages coming off the toptier directly, there can be
colder pages on lower tier nodes that were demoted by global reclaim. 
This is an aging inversion, not unlike if cgroups were to reclaim directly
from the active lists while there are inactive pages.

Proactive reclaim is another factor.  The goal of that it is to offload
colder pages from expensive RAM to cheaper storage.  When lower tier
memory is available as an intermediate layer, we want offloading to take
advantage of it instead of bypassing to storage.

Revert the patch so that cgroups respect the LRU order spanning the memory
hierarchy.

Of note is a specific undercommit scenario, where all cgroup limits in the
system add up to <= available toptier memory.  In that case, shuffling
pages out to lower tiers first to reclaim them from there is inefficient. 
This is something could be optimized/short-circuited later on (although
care must be taken not to accidentally recreate the aging inversion). 
Let's ensure correctness first.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Dave Hansen <[email protected]>
Reviewed-by: Yang Shi <[email protected]>
Acked-by: Roman Gushchin <[email protected]>
Reviewed-by: "Huang, Ying" <[email protected]>
Reviewed-by: Muchun Song <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Shakeel Butt <[email protected]>
Acked-by: Tim Chen <[email protected]>
Cc: Zi Yan <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-05-25 10:47:48 -07:00
Johannes Weiner f4840ccfca zswap: memcg accounting 
Applications can currently escape their cgroup memory containment when
zswap is enabled.  This patch adds per-cgroup tracking and limiting of
zswap backend memory to rectify this.

The existing cgroup2 memory.stat file is extended to show zswap statistics
analogous to what's in meminfo and vmstat.  Furthermore, two new control
files, memory.zswap.current and memory.zswap.max, are added to allow
tuning zswap usage on a per-workload basis.  This is important since not
all workloads benefit from zswap equally; some even suffer compared to
disk swap when memory contents don't compress well.  The optimal size of
the zswap pool, and the threshold for writeback, also depends on the size
of the workload's warm set.

The implementation doesn't use a traditional page_counter transaction. 
zswap is unconventional as a memory consumer in that we only know the
amount of memory to charge once expensive compression has occurred.  If
zwap is disabled or the limit is already exceeded we obviously don't want
to compress page upon page only to reject them all.  Instead, the limit is
checked against current usage, then we compress and charge.  This allows
some limit overrun, but not enough to matter in practice.

[[email protected]: fix for CONFIG_SLOB builds]
  Link: https://lkml.kernel.org/r/[email protected]
[[email protected]: opt out of cgroups v1]
  Link: https://lkml.kernel.org/r/Yn6it9mBYFA+/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Dan Streetman <[email protected]>
Cc: Minchan Kim <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-05-19 14:08:53 -07:00
Johannes Weiner f6498b776d mm: zswap: add basic meminfo and vmstat coverage 
Currently it requires poking at debugfs to figure out the size and
population of the zswap cache on a host.  There are no counters for reads
and writes against the cache.  As a result, it's difficult to understand
zswap behavior on production systems.

Print zswap memory consumption and how many pages are zswapped out in
/proc/meminfo.  Count zswapouts and zswapins in /proc/vmstat.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Cc: Dan Streetman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Shakeel Butt <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-05-19 14:08:53 -07:00
Johannes Weiner b3fbd58fcb mm: Kconfig: simplify zswap configuration 
- CONFIG_ZRAM: Zram is a user-facing feature, whereas zsmalloc is
  not. Don't make the user chase down a technical dependency like
  that, just select it in automatically when zram is requested. The
  CONFIG_CRYPTO dependency is redundant due to more specific deps.

- CONFIG_ZPOOL: This is not a user-facing feature. Hide the symbol and
  have it selected in as needed.

- CONFIG_ZSWAP: Select CRYPTO instead of depend. Common pattern.

- Make the ZSWAP suboptions and their descriptions (compression,
  allocation backend) a bit more straight-forward for the user.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Cc: Dan Streetman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Shakeel Butt <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-05-19 14:08:53 -07:00
Johannes Weiner 519bcb7979 mm: Kconfig: group swap, slab, hotplug and thp options into submenus 
There are several clusters of related config options spread throughout the
mostly flat MM submenu.  Group them together and put specialization
options into further subdirectories to make the MM submenu a bit more
organized and easier to navigate.

[[email protected]: fix kbuild warnings]
  Link: https://lkml.kernel.org/r/[email protected]
[[email protected]: fix more kbuild warnings]
  Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Cc: Dan Streetman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Shakeel Butt <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-05-19 14:08:53 -07:00
Johannes Weiner 7b42f1041c mm: Kconfig: move swap and slab config options to the MM section 
These are currently under General Setup. MM seems like a better fit.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Cc: Dan Streetman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Shakeel Butt <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-05-19 14:08:53 -07:00
Johannes Weiner 39799b6409 Documentation: filesystems: proc: update meminfo section 
Patch series "zswap: accounting & cgroup control", v2.

Zswap can consume nearly a quarter of RAM in the default configuration,
yet it's neither listed in /proc/meminfo, nor is it accounted and
manageable on a per-cgroup basis.

This makes reasoning about the memory situation on a host in general
rather difficult.  On shared/cgrouped hosts, the consequences are worse. 
First, workloads can escape memory containment and cause resource priority
inversions: a lo-pri group can fill the global zswap pool and force a
hi-pri group out to disk.  Second, not all workloads benefit from zswap
equally.  Some even suffer when memory contents compress poorly, and are
better off going to disk swap directly.  On a host with mixed workloads,
it's currently not possible to enable zswap for one workload but not for
the other.

This series implements the missing global accounting as well as cgroup
tracking & control for zswap backing memory:

- Patch 1 refreshes the very out-of-date meminfo documentation in
  Documentation/filesystems/proc.rst.

- Patches 2-4 clean up related and adjacent options in Kconfig. Not
  actual dependencies, just things I noticed during development.

- Patch 5 adds meminfo and vmstat coverage for zswap consumption and
  activity.

- Patch 6 implements per-cgroup tracking & control of zswap memory.


This patch (of 6):

Add new entries.  Minor corrections and cleanups.

[[email protected]: fix htmldocs warnings]
  Link: https://lkml.kernel.org/r/[email protected]
[[email protected]: change `Unevictable' wording, per David]
  Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: David Hildenbrand <[email protected]>
Cc: Dan Streetman <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Seth Jennings <[email protected]>
Cc: Shakeel Butt <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
 2022-05-19 14:08:52 -07:00
Johannes Weiner 9457056ac4 mm: madvise: MADV_DONTNEED_LOCKED 
MADV_DONTNEED historically rejects mlocked ranges, but with MLOCK_ONFAULT
and MCL_ONFAULT allowing to mlock without populating, there are valid use
cases for depopulating locked ranges as well.

Users mlock memory to protect secrets.  There are allocators for secure
buffers that want in-use memory generally mlocked, but cleared and
invalidated memory to give up the physical pages.  This could be done with
explicit munlock -> mlock calls on free -> alloc of course, but that adds
two unnecessary syscalls, heavy mmap_sem write locks, vma splits and
re-merges - only to get rid of the backing pages.

Users also mlockall(MCL_ONFAULT) to suppress sustained paging, but are
okay with on-demand initial population.  It seems valid to selectively
free some memory during the lifetime of such a process, without having to
mess with its overall policy.

Why add a separate flag? Isn't this a pretty niche usecase?

- MADV_DONTNEED has been bailing on locked vmas forever. It's at least
  conceivable that someone, somewhere is relying on mlock to protect
  data from perhaps broader invalidation calls. Changing this behavior
  now could lead to quiet data corruption.

- It also clarifies expectations around MADV_FREE and maybe
  MADV_REMOVE. It avoids the situation where one quietly behaves
  different than the others. MADV_FREE_LOCKED can be added later.

- The combination of mlock() and madvise() in the first place is
  probably niche. But where it happens, I'd say that dropping pages
  from a locked region once they don't contain secrets or won't page
  anymore is much saner than relying on mlock to protect memory from
  speculative or errant invalidation calls. It's just that we can't
  change the default behavior because of the two previous points.

Given that, an explicit new flag seems to make the most sense.

[[email protected]: fix mips build]

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Reviewed-by: Mike Kravetz <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: Nadav Amit <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: Dr. David Alan Gilbert <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2022-03-24 19:06:51 -07:00
Johannes Weiner d8c47cc7bf mm: page_io: fix psi memory pressure error on cold swapins 
Once upon a time, all swapins counted toward memory pressure[1].  Then
Joonsoo introduced workingset detection for anonymous pages and we gained
the ability to distinguish hot from cold swapins[2][3].  But we failed to
update swap_readpage() accordingly, and now we account partial memory
pressure in the swapin path of cold memory.

Not for all situations - which adds more inconsistency: paths using the
conventional submit_bio() and lock_page() route will not see much pressure
- unless storage itself is heavily congested and the bio submissions
stall.  ZRAM and ZSWAP do most of the work directly from swap_readpage()
and will see all swapins reflected as pressure.

IOW, a workload doing cold swapins could see little to no pressure
reported with on-disk swap, but potentially high pressure with a zram or
zswap backend.  That confuses any psi-based health monitoring, load
shedding, proactive reclaim, or userspace OOM killing schemes that might
be in place for the workload.

Restore consistency by making all swapin stall accounting conditional on
the page actually being part of the workingset.

[1] commit 937790699b ("mm/page_io.c: annotate refault stalls from swap_readpage")
[2] commit aae466b005 ("mm/swap: implement workingset detection for anonymous LRU")
[3] commit cad8320b4b ("mm/swap: don't SetPageWorkingset unconditionally during swapin")

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: CGEL <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Yu Zhao <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2022-03-22 15:57:09 -07:00
Johannes Weiner af9a3b69e8 mm/memcg: opencode the inner part of obj_cgroup_uncharge_pages() in drain_obj_stock() 
Provide the inner part of refill_stock() as __refill_stock() without
disabling interrupts.  This eases the integration of local_lock_t where
recursive locking must be avoided.

Open code obj_cgroup_uncharge_pages() in drain_obj_stock() and use
__refill_stock().  The caller of drain_obj_stock() already disables
interrupts.

[[email protected]: patch body around Johannes' diff]

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Reviewed-by: Roman Gushchin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: kernel test robot <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Michal Koutný <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Vladimir Davydov <[email protected]>
Cc: Waiman Long <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2022-03-22 15:57:02 -07:00
Johannes Weiner 16c8fd64c3 MAINTAINERS: add Suren as psi co-maintainer 
Suren wrote the poll() interface, which is a significant part of the
psi code and represents a large user of psi itself (Android). It's a
good idea to have him look at psi patches as well, and it's good to
have two people following things in case one of us is traveling.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
 2022-01-27 12:57:18 +01:00
Johannes Weiner 51b8c1fe25 vfs: keep inodes with page cache off the inode shrinker LRU 
Historically (pre-2.5), the inode shrinker used to reclaim only empty
inodes and skip over those that still contained page cache.  This caused
problems on highmem hosts: struct inode could put fill lowmem zones
before the cache was getting reclaimed in the highmem zones.

To address this, the inode shrinker started to strip page cache to
facilitate reclaiming lowmem.  However, this comes with its own set of
problems: the shrinkers may drop actively used page cache just because
the inodes are not currently open or dirty - think working with a large
git tree.  It further doesn't respect cgroup memory protection settings
and can cause priority inversions between containers.

Nowadays, the page cache also holds non-resident info for evicted cache
pages in order to detect refaults.  We've come to rely heavily on this
data inside reclaim for protecting the cache workingset and driving swap
behavior.  We also use it to quantify and report workload health through
psi.  The latter in turn is used for fleet health monitoring, as well as
driving automated memory sizing of workloads and containers, proactive
reclaim and memory offloading schemes.

The consequences of dropping page cache prematurely is that we're seeing
subtle and not-so-subtle failures in all of the above-mentioned
scenarios, with the workload generally entering unexpected thrashing
states while losing the ability to reliably detect it.

To fix this on non-highmem systems at least, going back to rotating
inodes on the LRU isn't feasible.  We've tried (commit a76cf1a474
("mm: don't reclaim inodes with many attached pages")) and failed
(commit 69056ee6a8 ("Revert "mm: don't reclaim inodes with many
attached pages"")).

The issue is mostly that shrinker pools attract pressure based on their
size, and when objects get skipped the shrinkers remember this as
deferred reclaim work.  This accumulates excessive pressure on the
remaining inodes, and we can quickly eat into heavily used ones, or
dirty ones that require IO to reclaim, when there potentially is plenty
of cold, clean cache around still.

Instead, this patch keeps populated inodes off the inode LRU in the
first place - just like an open file or dirty state would.  An otherwise
clean and unused inode then gets queued when the last cache entry
disappears.  This solves the problem without reintroducing the reclaim
issues, and generally is a bit more scalable than having to wade through
potentially hundreds of thousands of busy inodes.

Locking is a bit tricky because the locks protecting the inode state
(i_lock) and the inode LRU (lru_list.lock) don't nest inside the
irq-safe page cache lock (i_pages.xa_lock).  Page cache deletions are
serialized through i_lock, taken before the i_pages lock, to make sure
depopulated inodes are queued reliably.  Additions may race with
deletions, but we'll check again in the shrinker.  If additions race
with the shrinker itself, we're protected by the i_lock: if find_inode()
or iput() win, the shrinker will bail on the elevated i_count or
I_REFERENCED; if the shrinker wins and goes ahead with the inode, it
will set I_FREEING and inhibit further igets(), which will cause the
other side to create a new instance of the inode instead.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Dave Chinner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-11-09 10:02:48 -08:00
Johannes Weiner 7ae12c809f fs: inode: count invalidated shadow pages in pginodesteal 
pginodesteal is supposed to capture the impact that inode reclaim has on
the page cache state.  Currently, it doesn't consider shadow pages that
get dropped this way, even though this can have a significant impact on
paging behavior, memory pressure calculations etc.

To improve visibility into these effects, make sure shadow pages get
counted when they get dropped through inode reclaim.

This changes the return value semantics of invalidate_mapping_pages()
semantics slightly, but the only two users are the inode shrinker itsel
and a usb driver that logs it for debugging purposes.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-09-03 09:58:10 -07:00
Johannes Weiner 16e2df2a05 fs: drop_caches: fix skipping over shadow cache inodes 
When drop_caches truncates the page cache in an inode it also includes any
shadow entries for evicted pages.  However, there is a preliminary check
on whether the inode has pages: if it has *only* shadow entries, it will
skip running truncation on the inode and leave it behind.

Fix the check to mapping_empty(), such that it runs truncation on any
inode that has cache entries at all.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Roman Gushchin <[email protected]>
Acked-by: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-09-03 09:58:10 -07:00
Johannes Weiner 3047250972 mm: remove irqsave/restore locking from contexts with irqs enabled 
The page cache deletion paths all have interrupts enabled, so no need to
use irqsafe/irqrestore locking variants.

They used to have irqs disabled by the memcg lock added in commit
c4843a7593 ("memcg: add per cgroup dirty page accounting"), but that has
since been replaced by memcg taking the page lock instead, commit
0a31bc97c8 ("mm: memcontrol: rewrite uncharge AP").

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-09-03 09:58:10 -07:00
Johannes Weiner 57f29762cd mm: vmscan: fix missing psi annotation for node_reclaim() 
In a debugging session the other day, Rik noticed that node_reclaim()
was missing memstall annotations.  This means we'll miss pressure and
lost productivity resulting from reclaim on an overloaded local NUMA
node when vm.zone_reclaim_mode is enabled.

There haven't been any reports, but that's likely because
vm.zone_reclaim_mode hasn't been a commonly used feature recently, and
the intersection between such setups and psi users is probably nil.

But secondary memory such as CXL-connected DIMMS, persistent memory etc,
and the page demotion patches that handle them
(https://lore.kernel.org/lkml/[email protected]/)
could soon make this a more common codepath again.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Rik van Riel <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-08-20 11:31:42 -07:00
Johannes Weiner f56ce412a5 mm: memcontrol: fix occasional OOMs due to proportional memory.low reclaim 
We've noticed occasional OOM killing when memory.low settings are in
effect for cgroups.  This is unexpected and undesirable as memory.low is
supposed to express non-OOMing memory priorities between cgroups.

The reason for this is proportional memory.low reclaim.  When cgroups
are below their memory.low threshold, reclaim passes them over in the
first round, and then retries if it couldn't find pages anywhere else.
But when cgroups are slightly above their memory.low setting, page scan
force is scaled down and diminished in proportion to the overage, to the
point where it can cause reclaim to fail as well - only in that case we
currently don't retry, and instead trigger OOM.

To fix this, hook proportional reclaim into the same retry logic we have
in place for when cgroups are skipped entirely.  This way if reclaim
fails and some cgroups were scanned with diminished pressure, we'll try
another full-force cycle before giving up and OOMing.

[[email protected]: coding-style fixes]

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 9783aa9917 ("mm, memcg: proportional memory.{low,min} reclaim")
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Leon Yang <[email protected]>
Reviewed-by: Rik van Riel <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Roman Gushchin <[email protected]>
Acked-by: Chris Down <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: <[email protected]>		[5.4+]
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-08-20 11:31:42 -07:00
Johannes Weiner 30def93565 mm: memcontrol: fix blocking rstat function called from atomic cgroup1 thresholding code 
Dan Carpenter reports:

    The patch 2d146aa3aa: "mm: memcontrol: switch to rstat" from Apr
    29, 2021, leads to the following static checker warning:

	    kernel/cgroup/rstat.c:200 cgroup_rstat_flush()
	    warn: sleeping in atomic context

    mm/memcontrol.c
      3572  static unsigned long mem_cgroup_usage(struct mem_cgroup *memcg, bool swap)
      3573  {
      3574          unsigned long val;
      3575
      3576          if (mem_cgroup_is_root(memcg)) {
      3577                  cgroup_rstat_flush(memcg->css.cgroup);
			    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

    This is from static analysis and potentially a false positive.  The
    problem is that mem_cgroup_usage() is called from __mem_cgroup_threshold()
    which holds an rcu_read_lock().  And the cgroup_rstat_flush() function
    can sleep.

      3578                  val = memcg_page_state(memcg, NR_FILE_PAGES) +
      3579                          memcg_page_state(memcg, NR_ANON_MAPPED);
      3580                  if (swap)
      3581                          val += memcg_page_state(memcg, MEMCG_SWAP);
      3582          } else {
      3583                  if (!swap)
      3584                          val = page_counter_read(&memcg->memory);
      3585                  else
      3586                          val = page_counter_read(&memcg->memsw);
      3587          }
      3588          return val;
      3589  }

__mem_cgroup_threshold() indeed holds the rcu lock.  In addition, the
thresholding code is invoked during stat changes, and those contexts
have irqs disabled as well.  If the lock breaking occurs inside the
flush function, it will result in a sleep from an atomic context.

Use the irqsafe flushing variant in mem_cgroup_usage() to fix this.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 2d146aa3aa ("mm: memcontrol: switch to rstat")
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Dan Carpenter <[email protected]>
Acked-by: Chris Down <[email protected]>
Reviewed-by: Rik van Riel <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-07-30 10:14:39 -07:00
Johannes Weiner d583d360a6 psi: Fix psi state corruption when schedule() races with cgroup move 
4117cebf1a ("psi: Optimize task switch inside shared cgroups")
introduced a race condition that corrupts internal psi state. This
manifests as kernel warnings, sometimes followed by bogusly high IO
pressure:

  psi: task underflow! cpu=1 t=2 tasks=[0 0 0 0] clear=c set=0
  (schedule() decreasing RUNNING and ONCPU, both of which are 0)

  psi: incosistent task state! task=2412744:systemd cpu=17 psi_flags=e clear=3 set=0
  (cgroup_move_task() clearing MEMSTALL and IOWAIT, but task is MEMSTALL | RUNNING | ONCPU)

What the offending commit does is batch the two psi callbacks in
schedule() to reduce the number of cgroup tree updates. When prev is
deactivated and removed from the runqueue, nothing is done in psi at
first; when the task switch completes, TSK_RUNNING and TSK_IOWAIT are
updated along with TSK_ONCPU.

However, the deactivation and the task switch inside schedule() aren't
atomic: pick_next_task() may drop the rq lock for load balancing. When
this happens, cgroup_move_task() can run after the task has been
physically dequeued, but the psi updates are still pending. Since it
looks at the task's scheduler state, it doesn't move everything to the
new cgroup that the task switch that follows is about to clear from
it. cgroup_move_task() will leak the TSK_RUNNING count in the old
cgroup, and psi_sched_switch() will underflow it in the new cgroup.

A similar thing can happen for iowait. TSK_IOWAIT is usually set when
a p->in_iowait task is dequeued, but again this update is deferred to
the switch. cgroup_move_task() can see an unqueued p->in_iowait task
and move a non-existent TSK_IOWAIT. This results in the inconsistent
task state warning, as well as a counter underflow that will result in
permanent IO ghost pressure being reported.

Fix this bug by making cgroup_move_task() use task->psi_flags instead
of looking at the potentially mismatching scheduler state.

[ We used the scheduler state historically in order to not rely on
  task->psi_flags for anything but debugging. But that ship has sailed
  anyway, and this is simpler and more robust.

  We previously already batched TSK_ONCPU clearing with the
  TSK_RUNNING update inside the deactivation call from schedule(). But
  that ordering was safe and didn't result in TSK_ONCPU corruption:
  unlike most places in the scheduler, cgroup_move_task() only checked
  task_current() and handled TSK_ONCPU if the task was still queued. ]

Fixes: 4117cebf1a ("psi: Optimize task switch inside shared cgroups")
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
 2021-05-06 15:33:26 +02:00
Johannes Weiner 9317d0fffe mm: page_counter: mitigate consequences of a page_counter underflow 
When the unsigned page_counter underflows, even just by a few pages, a
cgroup will not be able to run anything afterwards and trigger the OOM
killer in a loop.

Underflows shouldn't happen, but when they do in practice, we may just be
off by a small amount that doesn't interfere with the normal operation -
consequences don't need to be that dire.

Reset the page_counter to 0 upon underflow.  We'll issue a warning that
the accounting will be off and then try to keep limping along.

[ We used to do this with the original res_counter, where it was a
  more straight-forward correction inside the spinlock section. I
  didn't carry it forward into the lockless page counters for
  simplicity, but it turns out this is quite useful in practice. ]

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Chris Down <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:38 -07:00
Johannes Weiner 4bbcc5a41c kselftests: cgroup: update kmem test for new vmstat implementation 
With memcg having switched to rstat, memory.stat output is precise.
Update the cgroup selftest to reflect the expectations and error
tolerances of the new implementation.

Also add newly tracked types of memory to the memory.stat side of the
equation, since they're included in memory.current and could throw false
positives.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Reviewed-by: Michal Koutný <[email protected]>
Acked-by: Roman Gushchin <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:38 -07:00
Johannes Weiner 2cd21c8980 mm: memcontrol: consolidate lruvec stat flushing 
There are two functions to flush the per-cpu data of an lruvec into the
rest of the cgroup tree: when the cgroup is being freed, and when a CPU
disappears during hotplug.  The difference is whether all CPUs or just
one is being collected, but the rest of the flushing code is the same.
Merge them into one function and share the common code.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Roman Gushchin <[email protected]>
Cc: Michal Koutný <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:38 -07:00
Johannes Weiner 2d146aa3aa mm: memcontrol: switch to rstat 
Replace the memory controller's custom hierarchical stats code with the
generic rstat infrastructure provided by the cgroup core.

The current implementation does batched upward propagation from the
write side (i.e.  as stats change).  The per-cpu batches introduce an
error, which is multiplied by the number of subgroups in a tree.  In
systems with many CPUs and sizable cgroup trees, the error can be large
enough to confuse users (e.g.  32 batch pages * 32 CPUs * 32 subgroups
results in an error of up to 128M per stat item).  This can entirely
swallow allocation bursts inside a workload that the user is expecting
to see reflected in the statistics.

In the past, we've done read-side aggregation, where a memory.stat read
would have to walk the entire subtree and add up per-cpu counts.  This
became problematic with lazily-freed cgroups: we could have large
subtrees where most cgroups were entirely idle.  Hence the switch to
change-driven upward propagation.  Unfortunately, it needed to trade
accuracy for speed due to the write side being so hot.

Rstat combines the best of both worlds: from the write side, it cheaply
maintains a queue of cgroups that have pending changes, so that the read
side can do selective tree aggregation.  This way the reported stats
will always be precise and recent as can be, while the aggregation can
skip over potentially large numbers of idle cgroups.

The way rstat works is that it implements a tree for tracking cgroups
with pending local changes, as well as a flush function that walks the
tree upwards.  The controller then drives this by 1) telling rstat when
a local cgroup stat changes (e.g.  mod_memcg_state) and 2) when a flush
is required to get uptodate hierarchy stats for a given subtree (e.g.
when memory.stat is read).  The controller also provides a flush
callback that is called during the rstat flush walk for each cgroup and
aggregates its local per-cpu counters and propagates them upwards.

This adds a second vmstats to struct mem_cgroup (MEMCG_NR_STAT +
NR_VM_EVENT_ITEMS) to track pending subtree deltas during upward
aggregation.  It removes 3 words from the per-cpu data.  It eliminates
memcg_exact_page_state(), since memcg_page_state() is now exact.

[[email protected]: merge fix]
[[email protected]: fix a sleep in atomic section problem]
  Link: https://lkml.kernel.org/r/[email protected]

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Roman Gushchin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Reviewed-by: Michal Koutný <[email protected]>
Acked-by: Balbir Singh <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:38 -07:00
Johannes Weiner dc26532aed cgroup: rstat: punt root-level optimization to individual controllers 
Current users of the rstat code can source root-level statistics from
the native counters of their respective subsystem, allowing them to
forego aggregation at the root level.  This optimization is currently
implemented inside the generic rstat code, which doesn't track the root
cgroup and doesn't invoke the subsystem flush callbacks on it.

However, the memory controller cannot do this optimization, because
cgroup1 breaks out memory specifically for the local level, including at
the root level.  In preparation for the memory controller switching to
rstat, move the optimization from rstat core to the controllers.

Afterwards, rstat will always track the root cgroup for changes and
invoke the subsystem callbacks on it; and it's up to the subsystem to
special-case and skip aggregation of the root cgroup if it can source
this information through other, cheaper means.

This is the case for the io controller and the cgroup base stats.  In
their respective flush callbacks, check whether the parent is the root
cgroup, and if so, skip the unnecessary upward propagation.

The extra cost of tracking the root cgroup is negligible: on stat
changes, we actually remove a branch that checks for the root.  The
queueing for a flush touches only per-cpu data, and only the first stat
change since a flush requires a (per-cpu) lock.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Michal Koutný <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:37 -07:00
Johannes Weiner a7df69b81a cgroup: rstat: support cgroup1 
Rstat currently only supports the default hierarchy in cgroup2.  In
order to replace memcg's private stats infrastructure - used in both
cgroup1 and cgroup2 - with rstat, the latter needs to support cgroup1.

The initialization and destruction callbacks for regular cgroups are
already in place.  Remove the cgroup_on_dfl() guards to handle cgroup1.

The initialization of the root cgroup is currently hardcoded to only
handle cgrp_dfl_root.cgrp.  Move those callbacks to cgroup_setup_root()
and cgroup_destroy_root() to handle the default root as well as the
various cgroup1 roots we may set up during mounting.

The linking of css to cgroups happens in code shared between cgroup1 and
cgroup2 as well.  Simply remove the cgroup_on_dfl() guard.

Linkage of the root css to the root cgroup is a bit trickier: per
default, the root css of a subsystem controller belongs to the default
hierarchy (i.e.  the cgroup2 root).  When a controller is mounted in its
cgroup1 version, the root css is stolen and moved to the cgroup1 root;
on unmount, the css moves back to the default hierarchy.  Annotate
rebind_subsystems() to move the root css linkage along between roots.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Roman Gushchin <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Reviewed-by: Michal Koutný <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:37 -07:00
Johannes Weiner a18e6e6e15 mm: memcontrol: privatize memcg_page_state query functions 
There are no users outside of the memory controller itself. The rest
of the kernel cares either about node or lruvec stats.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Reviewed-by: Roman Gushchin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Reviewed-by: Michal Koutný <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:37 -07:00
Johannes Weiner a3747b53b1 mm: memcontrol: kill mem_cgroup_nodeinfo() 
No need to encapsulate a simple struct member access.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Reviewed-by: Roman Gushchin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Reviewed-by: Michal Koutný <[email protected]>
Cc: Tejun Heo <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:37 -07:00
Johannes Weiner a3d4c05a44 mm: memcontrol: fix cpuhotplug statistics flushing 
Patch series "mm: memcontrol: switch to rstat", v3.

This series converts memcg stats tracking to the streamlined rstat
infrastructure provided by the cgroup core code.  rstat is already used by
the CPU controller and the IO controller.  This change is motivated by
recent accuracy problems in memcg's custom stats code, as well as the
benefits of sharing common infra with other controllers.

The current memcg implementation does batched tree aggregation on the
write side: local stat changes are cached in per-cpu counters, which are
then propagated upward in batches when a threshold (32 pages) is exceeded.
This is cheap, but the error introduced by the lazy upward propagation
adds up: 32 pages times CPUs times cgroups in the subtree.  We've had
complaints from service owners that the stats do not reliably track and
react to allocation behavior as expected, sometimes swallowing the results
of entire test applications.

The original memcg stat implementation used to do tree aggregation
exclusively on the read side: local stats would only ever be tracked in
per-cpu counters, and a memory.stat read would iterate the entire subtree
and sum those counters up.  This didn't keep up with the times:

 - Cgroup trees are much bigger now. We switched to lazily-freed
   cgroups, where deleted groups would hang around until their remaining
   page cache has been reclaimed. This can result in large subtrees that
   are expensive to walk, while most of the groups are idle and their
   statistics don't change much anymore.

 - Automated monitoring increased. With the proliferation of userspace
   oom killing, proactive reclaim, and higher-resolution logging of
   workload trends in general, top-level stat files are polled at least
   once a second in many deployments.

 - The lifetime of cgroups got shorter. Where most cgroup setups in the
   past would have a few large policy-oriented cgroups for everything
   running on the system, newer cgroup deployments tend to create one
   group per application - which gets deleted again as the processes
   exit. An aggregation scheme that doesn't retain child data inside the
   parents loses event history of the subtree.

Rstat addresses all three of those concerns through intelligent,
persistent read-side aggregation.  As statistics change at the local
level, rstat tracks - on a per-cpu basis - only those parts of a subtree
that have changes pending and require aggregation.  The actual
aggregation occurs on the colder read side - which can now skip over
(potentially large) numbers of recently idle cgroups.

===

The test_kmem cgroup selftest is currently failing due to excessive
cumulative vmstat drift from 100 subgroups:

    ok 1 test_kmem_basic
    memory.current = 8810496
    slab + anon + file + kernel_stack = 17074568
    slab = 6101384
    anon = 946176
    file = 0
    kernel_stack = 10027008
    not ok 2 test_kmem_memcg_deletion
    ok 3 test_kmem_proc_kpagecgroup
    ok 4 test_kmem_kernel_stacks
    ok 5 test_kmem_dead_cgroups
    ok 6 test_percpu_basic

As you can see, memory.stat items far exceed memory.current.  The kernel
stack alone is bigger than all of charged memory.  That's because the
memory of the test has been uncharged from memory.current, but the
negative vmstat deltas are still sitting in the percpu caches.

The test at this time isn't even counting percpu, pagetables etc.  yet,
which would further contribute to the error.  The last patch in the series
updates the test to include them - as well as reduces the vmstat
tolerances in general to only expect page_counter batching.

With all patches applied, the (now more stringent) test succeeds:

    ok 1 test_kmem_basic
    ok 2 test_kmem_memcg_deletion
    ok 3 test_kmem_proc_kpagecgroup
    ok 4 test_kmem_kernel_stacks
    ok 5 test_kmem_dead_cgroups
    ok 6 test_percpu_basic

===

A kernel build test confirms that overhead is comparable.  Two kernels are
built simultaneously in a nested tree with several idle siblings:

root - kernelbuild - one - two - three - four - build-a (defconfig, make -j16)
                                             `- build-b (defconfig, make -j16)
                                             `- idle-1
                                             `- ...
                                             `- idle-9

During the builds, kernelbuild/memory.stat is read once a second.

A perf diff shows that the changes in cycle distribution is
minimal. Top 10 kernel symbols:

     0.09%     +0.08%  [kernel.kallsyms]                       [k] __mod_memcg_lruvec_state
     0.00%     +0.06%  [kernel.kallsyms]                       [k] cgroup_rstat_updated
     0.08%     -0.05%  [kernel.kallsyms]                       [k] __mod_memcg_state.part.0
     0.16%     -0.04%  [kernel.kallsyms]                       [k] release_pages
     0.00%     +0.03%  [kernel.kallsyms]                       [k] __count_memcg_events
     0.01%     +0.03%  [kernel.kallsyms]                       [k] mem_cgroup_charge_statistics.constprop.0
     0.10%     -0.02%  [kernel.kallsyms]                       [k] get_mem_cgroup_from_mm
     0.05%     -0.02%  [kernel.kallsyms]                       [k] mem_cgroup_update_lru_size
     0.57%     +0.01%  [kernel.kallsyms]                       [k] asm_exc_page_fault

===

The on-demand aggregated stats are now fully accurate:

$ grep -e nr_inactive_file /proc/vmstat | awk '{print($1,$2*4096)}'; \
  grep -e inactive_file /sys/fs/cgroup/memory.stat

vanilla:                              patched:
nr_inactive_file 1574105088           nr_inactive_file 1027801088
   inactive_file 1577410560              inactive_file 1027801088

===

This patch (of 8):

The memcg hotunplug callback erroneously flushes counts on the local CPU,
not the counts of the CPU going away; those counts will be lost.

Flush the CPU that is actually going away.

Also simplify the code a bit by using mod_memcg_state() and
count_memcg_events() instead of open-coding the upward flush - this is
comparable to how vmstat.c handles hotunplug flushing.

Link: https://lkml.kernel.org/r/[email protected]
Link: https://lkml.kernel.org/r/[email protected]
Fixes: a983b5ebee ("mm: memcontrol: fix excessive complexity in memory.stat reporting")
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Reviewed-by: Roman Gushchin <[email protected]>
Reviewed-by: Michal Koutný <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:37 -07:00
Johannes Weiner 1c824a680b mm: page-writeback: simplify memcg handling in test_clear_page_writeback() 
Page writeback doesn't hold a page reference, which allows truncate to
free a page the second PageWriteback is cleared.  This used to require
special attention in test_clear_page_writeback(), where we had to be
careful not to rely on the unstable page->memcg binding and look up all
the necessary information before clearing the writeback flag.

Since commit 073861ed77 ("mm: fix VM_BUG_ON(PageTail) and
BUG_ON(PageWriteback)") test_clear_page_writeback() is called with an
explicit reference on the page, and this dance is no longer needed.

Use unlock_page_memcg() and dec_lruvec_page_state() directly.

This removes the last user of the lock_page_memcg() return value, change
it to void.  Touch up the comments in there as well.  This also removes
the last extern user of __unlock_page_memcg(), make it static.  Further,
it removes the last user of dec_lruvec_state(), delete it, along with a
few other unused helpers.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Hugh Dickins <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-04-30 11:20:37 -07:00
Johannes Weiner fddc8bab53 psi: Pressure states are unlikely 
Move the unlikely branches out of line. This eliminates undesirable
jumps during wakeup and sleeps for workloads that aren't under any
sort of resource pressure.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Chengming Zhou <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
 2021-03-06 12:40:23 +01:00
Johannes Weiner 629484ae73 mm: vmstat: add some comments on internal storage of byte items 
Byte-accounted items are used for slab object accounting at the cgroup
level, because the objects in a slab page can belong to different cgroups.
At the global level these items always change in multiples of whole slab
pages.  The vmstat code exploits this and stores these items as pages
internally, which allows for more compact per-cpu data.

This optimization isn't self-evident from the asserts and the division in
the stat update functions.  Provide the reader with some context.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-02-26 09:41:00 -08:00
Johannes Weiner 2bbd00aef0 mm: vmstat: fix NOHZ wakeups for node stat changes 
On NOHZ, the periodic vmstat flushers on each CPU can go to sleep and
won't wake up until stat changes are detected in the per-cpu deltas of the
zone vmstat counters.

In commit 75ef718405 ("mm, vmstat: add infrastructure for per-node
vmstats") per-node counters were introduced, and subsequently most stats
were moved from the zone to the node level.  However, the node counters
weren't added to the NOHZ wakeup detection.

In theory this can cause per-cpu errors to remain in the user-reported
stats indefinitely.  In practice this only affects a handful of sub
counters (file_mapped, dirty and writeback e.g.) because other page state
changes at the node level likely involve a change at the zone level as
well (alloc and free, lru ops).  Also, nobody has complained.

Fix it up for completeness: wake up vmstat refreshing on node changes.
Also remove the BUILD_BUG_ONs that assert counter size; we haven't relied
on it since we added sizeof() to the range calculation in commit
13c9aaf7fa ("mm/vmstat.c: fix NUMA statistics updates").

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-02-26 09:41:00 -08:00
Johannes Weiner 6eeb104e11 fs: buffer: use raw page_memcg() on locked page 
alloc_page_buffers() currently uses get_mem_cgroup_from_page() for
charging the buffers to the page owner, which does an rcu-protected
page->memcg lookup and acquires a reference.  But buffer allocation has
the page lock held throughout, which pins the page to the memcg and
thereby the memcg - neither rcu nor holding an extra reference during the
allocation are necessary.  Use a raw page_memcg() instead.

This was the last user of get_mem_cgroup_from_page(), delete it.

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Muchun Song <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Roman Gushchin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-02-24 13:38:30 -08:00
Johannes Weiner e82553c10b Revert "mm: memcontrol: avoid workload stalls when lowering memory.high" 
This reverts commit 536d3bf261, as it can
cause writers to memory.high to get stuck in the kernel forever,
performing page reclaim and consuming excessive amounts of CPU cycles.

Before the patch, a write to memory.high would first put the new limit
in place for the workload, and then reclaim the requested delta.  After
the patch, the kernel tries to reclaim the delta before putting the new
limit into place, in order to not overwhelm the workload with a sudden,
large excess over the limit.  However, if reclaim is actively racing
with new allocations from the uncurbed workload, it can keep the write()
working inside the kernel indefinitely.

This is causing problems in Facebook production.  A privileged
system-level daemon that adjusts memory.high for various workloads
running on a host can get unexpectedly stuck in the kernel and
essentially turn into a sort of involuntary kswapd for one of the
workloads.  We've observed that daemon busy-spin in a write() for
minutes at a time, neglecting its other duties on the system, and
expending privileged system resources on behalf of a workload.

To remedy this, we have first considered changing the reclaim logic to
break out after a couple of loops - whether the workload has converged
to the new limit or not - and bound the write() call this way.  However,
the root cause that inspired the sequence change in the first place has
been fixed through other means, and so a revert back to the proven
limit-setting sequence, also used by memory.max, is preferable.

The sequence was changed to avoid extreme latencies in the workload when
the limit was lowered: the sudden, large excess created by the limit
lowering would erroneously trigger the penalty sleeping code that is
meant to throttle excessive growth from below.  Allocating threads could
end up sleeping long after the write() had already reclaimed the delta
for which they were being punished.

However, erroneous throttling also caused problems in other scenarios at
around the same time.  This resulted in commit b3ff92916a ("mm, memcg:
reclaim more aggressively before high allocator throttling"), included
in the same release as the offending commit.  When allocating threads
now encounter large excess caused by a racing write() to memory.high,
instead of entering punitive sleeps, they will simply be tasked with
helping reclaim down the excess, and will be held no longer than it
takes to accomplish that.  This is in line with regular limit
enforcement - i.e.  if the workload allocates up against or over an
otherwise unchanged limit from below.

With the patch breaking userspace, and the root cause addressed by other
means already, revert it again.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 536d3bf261 ("mm: memcontrol: avoid workload stalls when lowering memory.high")
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Tejun Heo <[email protected]>
Acked-by: Chris Down <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Michal Koutný <[email protected]>
Cc: <[email protected]>	[5.8+]
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2021-02-09 17:26:44 -08:00
Johannes Weiner 597c892038 mm: don't wake kswapd prematurely when watermark boosting is disabled 
On 2-node NUMA hosts we see bursts of kswapd reclaim and subsequent
pressure spikes and stalls from cache refaults while there is plenty of
free memory in the system.

Usually, kswapd is woken up when all eligible nodes in an allocation are
full.  But the code related to watermark boosting can wake kswapd on one
full node while the other one is mostly empty.  This may be justified to
fight fragmentation, but is currently unconditionally done whether
watermark boosting is occurring or not.

In our case, many of our workloads' throughput scales with available
memory, and pure utilization is a more tangible concern than trends
around longer-term fragmentation.  As a result we generally disable
watermark boosting.

Wake kswapd only woken when watermark boosting is requested.

Link: https://lkml.kernel.org/r/[email protected]
Fixes: 1c30844d2d ("mm: reclaim small amounts of memory when an external fragmentation event occurs")
Signed-off-by: Johannes Weiner <[email protected]>
Acked-by: Mel Gorman <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2020-12-15 12:13:45 -08:00
Johannes Weiner b8eddff888 mm: memcontrol: add file_thp, shmem_thp to memory.stat 
As huge page usage in the page cache and for shmem files proliferates in
our production environment, the performance monitoring team has asked for
per-cgroup stats on those pages.

We already track and export anon_thp per cgroup.  We already track file
THP and shmem THP per node, so making them per-cgroup is only a matter of
switching from node to lruvec counters.  All callsites are in places where
the pages are charged and locked, so page->memcg is stable.

[[email protected]: add documentation]
  Link: https://lkml.kernel.org/r/[email protected]

Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Reviewed-by: Rik van Riel <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: David Rientjes <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Song Liu <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2020-12-15 12:13:39 -08:00
Johannes Weiner 9f45717924 mm: memcontrol: fix warning when allocating the root cgroup 
Commit 3e38e0aaca ("mm: memcg: charge memcg percpu memory to the
parent cgroup") adds memory tracking to the memcg kernel structures
themselves to make cgroups liable for the memory they are consuming
through the allocation of child groups (which can be significant).

This code is a bit awkward as it's spread out through several functions:
The outermost function does memalloc_use_memcg(parent) to set up
current->active_memcg, which designates which cgroup to charge, and the
inner functions pass GFP_ACCOUNT to request charging for specific
allocations.  To make sure this dependency is satisfied at all times -
to make sure we don't randomly charge whoever is calling the functions -
the inner functions warn on !current->active_memcg.

However, this triggers a false warning when the root memcg itself is
allocated.  No parent exists in this case, and so current->active_memcg
is rightfully NULL.  It's a false positive, not indicative of a bug.

Delete the warnings for now, we can revisit this later.

Fixes: 3e38e0aaca ("mm: memcg: charge memcg percpu memory to the parent cgroup")
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Stephen Rothwell <[email protected]>
Acked-by: Roman Gushchin <[email protected]>
Cc: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2020-08-13 12:15:21 -07:00
Johannes Weiner e22c6ed90a mm: memcontrol: don't count limit-setting reclaim as memory pressure 
When an outside process lowers one of the memory limits of a cgroup (or
uses the force_empty knob in cgroup1), direct reclaim is performed in the
context of the write(), in order to directly enforce the new limit and
have it being met by the time the write() returns.

Currently, this reclaim activity is accounted as memory pressure in the
cgroup that the writer(!) belongs to.  This is unexpected.  It
specifically causes problems for senpai
(https://github.com/facebookincubator/senpai), which is an agent that
routinely adjusts the memory limits and performs associated reclaim work
in tens or even hundreds of cgroups running on the host.  The cgroup that
senpai is running in itself will report elevated levels of memory
pressure, even though it itself is under no memory shortage or any sort of
distress.

Move the psi annotation from the central cgroup reclaim function to
callsites in the allocation context, and thereby no longer count any
limit-setting reclaim as memory pressure.  If the newly set limit causes
the workload inside the cgroup into direct reclaim, that of course will
continue to count as memory pressure.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Reviewed-by: Roman Gushchin <[email protected]>
Acked-by: Chris Down <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-08-07 11:33:26 -07:00
Johannes Weiner 19ce33acbb mm: memcontrol: restore proper dirty throttling when memory.high changes 
Commit 8c8c383c04 ("mm: memcontrol: try harder to set a new
memory.high") inadvertently removed a callback to recalculate the
writeback cache size in light of a newly configured memory.high limit.

Without letting the writeback cache know about a potentially heavily
reduced limit, it may permit too many dirty pages, which can cause
unnecessary reclaim latencies or even avoidable OOM situations.

This was spotted while reading the code, it hasn't knowingly caused any
problems in practice so far.

Fixes: 8c8c383c04 ("mm: memcontrol: try harder to set a new memory.high")
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Chris Down <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-08-07 11:33:26 -07:00
Johannes Weiner 1a3e1f4096 mm: memcontrol: decouple reference counting from page accounting 
The reference counting of a memcg is currently coupled directly to how
many 4k pages are charged to it.  This doesn't work well with Roman's new
slab controller, which maintains pools of objects and doesn't want to keep
an extra balance sheet for the pages backing those objects.

This unusual refcounting design (reference counts usually track pointers
to an object) is only for historical reasons: memcg used to not take any
css references and simply stalled offlining until all charges had been
reparented and the page counters had dropped to zero.  When we got rid of
the reparenting requirement, the simple mechanical translation was to take
a reference for every charge.

More historical context can be found in commit e8ea14cc6e ("mm:
memcontrol: take a css reference for each charged page"), commit
64f2199389 ("mm: memcontrol: remove obsolete kmemcg pinning tricks") and
commit b2052564e6 ("mm: memcontrol: continue cache reclaim from offlined
groups").

The new slab controller exposes the limitations in this scheme, so let's
switch it to a more idiomatic reference counting model based on actual
kernel pointers to the memcg:

- The per-cpu stock holds a reference to the memcg its caching

- User pages hold a reference for their page->mem_cgroup. Transparent
  huge pages will no longer acquire tail references in advance, we'll
  get them if needed during the split.

- Kernel pages hold a reference for their page->mem_cgroup

- Pages allocated in the root cgroup will acquire and release css
  references for simplicity. css_get() and css_put() optimize that.

- The current memcg_charge_slab() already hacked around the per-charge
  references; this change gets rid of that as well.

- tcp accounting will handle reference in mem_cgroup_sk_{alloc,free}

Roman:
1) Rebased on top of the current mm tree: added css_get() in
   mem_cgroup_charge(), dropped mem_cgroup_try_charge() part
2) I've reformatted commit references in the commit log to make
   checkpatch.pl happy.

[[email protected]: remove css_put_many() from __mem_cgroup_clear_mc()]
  Link: http://lkml.kernel.org/r/[email protected]

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Roman Gushchin <[email protected]>
Signed-off-by: Hugh Dickins <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Roman Gushchin <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Vlastimil Babka <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-08-07 11:33:24 -07:00
Johannes Weiner 31d8fcac00 mm: workingset: age nonresident information alongside anonymous pages 
Patch series "fix for "mm: balance LRU lists based on relative
thrashing" patchset"

This patchset fixes some problems of the patchset, "mm: balance LRU
lists based on relative thrashing", which is now merged on the mainline.

Patch "mm: workingset: let cache workingset challenge anon fix" is the
result of discussion with Johannes.  See following link.

  http://lkml.kernel.org/r/[email protected]

And, the other two are minor things which are found when I try to rebase
my patchset.

This patch (of 3):

After ("mm: workingset: let cache workingset challenge anon fix"), we
compare refault distances to active_file + anon.  But age of the
non-resident information is only driven by the file LRU.  As a result,
we may overestimate the recency of any incoming refaults and activate
them too eagerly, causing unnecessary LRU churn in certain situations.

Make anon aging drive nonresident age as well to address that.

Link: http://lkml.kernel.org/r/[email protected]
Link: http://lkml.kernel.org/r/[email protected]
Fixes: 34e58cac6d ("mm: workingset: let cache workingset challenge anon")
Reported-by: Joonsoo Kim <[email protected]>
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Joonsoo Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Michal Hocko <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-26 00:27:37 -07:00
Johannes Weiner cd324edce5 mm: memcontrol: handle div0 crash race condition in memory.low 
Tejun reports seeing rare div0 crashes in memory.low stress testing:

  RIP: 0010:mem_cgroup_calculate_protection+0xed/0x150
  Code: 0f 46 d1 4c 39 d8 72 57 f6 05 16 d6 42 01 40 74 1f 4c 39 d8 76 1a 4c 39 d1 76 15 4c 29 d1 4c 29 d8 4d 29 d9 31 d2 48 0f af c1 <49> f7 f1 49 01 c2 4c 89 96 38 01 00 00 5d c3 48 0f af c7 31 d2 49
  RSP: 0018:ffffa14e01d6fcd0 EFLAGS: 00010246
  RAX: 000000000243e384 RBX: 0000000000000000 RCX: 0000000000008f4b
  RDX: 0000000000000000 RSI: ffff8b89bee84000 RDI: 0000000000000000
  RBP: ffffa14e01d6fcd0 R08: ffff8b89ca7d40f8 R09: 0000000000000000
  R10: 0000000000000000 R11: 00000000006422f7 R12: 0000000000000000
  R13: ffff8b89d9617000 R14: ffff8b89bee84000 R15: ffffa14e01d6fdb8
  FS:  0000000000000000(0000) GS:ffff8b8a1f1c0000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 00007f93b1fc175b CR3: 000000016100a000 CR4: 0000000000340ea0
  Call Trace:
    shrink_node+0x1e5/0x6c0
    balance_pgdat+0x32d/0x5f0
    kswapd+0x1d7/0x3d0
    kthread+0x11c/0x160
    ret_from_fork+0x1f/0x30

This happens when parent_usage == siblings_protected.

We check that usage is bigger than protected, which should imply
parent_usage being bigger than siblings_protected.  However, we don't
read (or even update) these values atomically, and they can be out of
sync as the memory state changes under us.  A bit of fluctuation around
the target protection isn't a big deal, but we need to handle the div0
case.

Check the parent state explicitly to make sure we have a reasonable
positive value for the divisor.

Link: http://lkml.kernel.org/r/[email protected]
Fixes: 8a931f8013 ("mm: memcontrol: recursive memory.low protection")
Signed-off-by: Johannes Weiner <[email protected]>
Reported-by: Tejun Heo <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Chris Down <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-26 00:27:37 -07:00
Johannes Weiner d483a5dd00 mm: vmscan: limit the range of LRU type balancing 
When LRU cost only shows up on one list, we abruptly stop scanning that
list altogether.  That's an extreme reaction: by the time the other list
starts thrashing and the pendulum swings back, we may have no recent age
information on the first list anymore, and we could have significant
latencies until the scanner has caught up.

Soften this change in the feedback system by ensuring that no list
receives less than a third of overall pressure, and only distribute the
other 66% according to LRU cost.  This ensures that we maintain a minimum
rate of aging on the entire workingset while it's being pressured, while
still allowing a generous rate of convergence when the relative sizes of
the lists need to adjust.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:49 -07:00
Johannes Weiner 96f8bf4fb1 mm: vmscan: reclaim writepage is IO cost 
The VM tries to balance reclaim pressure between anon and file so as to
reduce the amount of IO incurred due to the memory shortage.  It already
counts refaults and swapins, but in addition it should also count
writepage calls during reclaim.

For swap, this is obvious: it's IO that wouldn't have occurred if the
anonymous memory hadn't been under memory pressure.  From a relative
balancing point of view this makes sense as well: even if anon is cold and
reclaimable, a cache that isn't thrashing may have equally cold pages that
don't require IO to reclaim.

For file writeback, it's trickier: some of the reclaim writepage IO would
have likely occurred anyway due to dirty expiration.  But not all of it -
premature writeback reduces batching and generates additional writes.
Since the flushers are already woken up by the time the VM starts writing
cache pages one by one, let's assume that we'e likely causing writes that
wouldn't have happened without memory pressure.  In addition, the per-page
cost of IO would have probably been much cheaper if written in larger
batches from the flusher thread rather than the single-page-writes from
kswapd.

For our purposes - getting the trend right to accelerate convergence on a
stable state that doesn't require paging at all - this is sufficiently
accurate.  If we later wanted to optimize for sustained thrashing, we can
still refine the measurements.

Count all writepage calls from kswapd as IO cost toward the LRU that the
page belongs to.

Why do this dynamically?  Don't we know in advance that anon pages require
IO to reclaim, and so could build in a static bias?

First, scanning is not the same as reclaiming.  If all the anon pages are
referenced, we may not swap for a while just because we're scanning the
anon list.  During this time, however, it's important that we age
anonymous memory and the page cache at the same rate so that their
hot-cold gradients are comparable.  Everything else being equal, we still
want to reclaim the coldest memory overall.

Second, we keep copies in swap unless the page changes.  If there is
swap-backed data that's mostly read (tmpfs file) and has been swapped out
before, we can reclaim it without incurring additional IO.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:49 -07:00
Johannes Weiner 7cf111bc39 mm: vmscan: determine anon/file pressure balance at the reclaim root 
We split the LRU lists into anon and file, and we rebalance the scan
pressure between them when one of them begins thrashing: if the file cache
experiences workingset refaults, we increase the pressure on anonymous
pages; if the workload is stalled on swapins, we increase the pressure on
the file cache instead.

With cgroups and their nested LRU lists, we currently don't do this
correctly.  While recursive cgroup reclaim establishes a relative LRU
order among the pages of all involved cgroups, LRU pressure balancing is
done on an individual cgroup LRU level.  As a result, when one cgroup is
thrashing on the filesystem cache while a sibling may have cold anonymous
pages, pressure doesn't get equalized between them.

This patch moves LRU balancing decision to the root of reclaim - the same
level where the LRU order is established.

It does this by tracking LRU cost recursively, so that every level of the
cgroup tree knows the aggregate LRU cost of all memory within its domain.
When the page scanner calculates the scan balance for any given individual
cgroup's LRU list, it uses the values from the ancestor cgroup that
initiated the reclaim cycle.

If one sibling is then thrashing on the cache, it will tip the pressure
balance inside its ancestors, and the next hierarchical reclaim iteration
will go more after the anon pages in the tree.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:49 -07:00
Johannes Weiner 314b57fb04 mm: balance LRU lists based on relative thrashing 
Since the LRUs were split into anon and file lists, the VM has been
balancing between page cache and anonymous pages based on per-list ratios
of scanned vs.  rotated pages.  In most cases that tips page reclaim
towards the list that is easier to reclaim and has the fewest actively
used pages, but there are a few problems with it:

1. Refaults and LRU rotations are weighted the same way, even though
   one costs IO and the other costs a bit of CPU.

2. The less we scan an LRU list based on already observed rotations,
   the more we increase the sampling interval for new references, and
   rotations become even more likely on that list. This can enter a
   death spiral in which we stop looking at one list completely until
   the other one is all but annihilated by page reclaim.

Since commit a528910e12 ("mm: thrash detection-based file cache sizing")
we have refault detection for the page cache.  Along with swapin events,
they are good indicators of when the file or anon list, respectively, is
too small for its workingset and needs to grow.

For example, if the page cache is thrashing, the cache pages need more
time in memory, while there may be colder pages on the anonymous list.
Likewise, if swapped pages are faulting back in, it indicates that we
reclaim anonymous pages too aggressively and should back off.

Replace LRU rotations with refaults and swapins as the basis for relative
reclaim cost of the two LRUs.  This will have the VM target list balances
that incur the least amount of IO on aggregate.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:49 -07:00
Johannes Weiner 264e90cc07 mm: only count actual rotations as LRU reclaim cost 
When shrinking the active file list we rotate referenced pages only when
they're in an executable mapping.  The others get deactivated.  When it
comes to balancing scan pressure, though, we count all referenced pages as
rotated, even the deactivated ones.  Yet they do not carry the same cost
to the system: the deactivated page *might* refault later on, but the
deactivation is tangible progress toward freeing pages; rotations on the
other hand cost time and effort without getting any closer to freeing
memory.

Don't treat both events as equal.  The following patch will hook up LRU
balancing to cache and anon refaults, which are a much more concrete cost
signal for reclaiming one list over the other.  Thus, remove the maybe-IO
cost bias from page references, and only note the CPU cost for actual
rotations that prevent the pages from getting reclaimed.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:49 -07:00
Johannes Weiner fbbb602e40 mm: deactivations shouldn't bias the LRU balance 
Operations like MADV_FREE, FADV_DONTNEED etc.  currently move any affected
active pages to the inactive list to accelerate their reclaim (good) but
also steer page reclaim toward that LRU type, or away from the other
(bad).

The reason why this is undesirable is that such operations are not part of
the regular page aging cycle, and rather a fluke that doesn't say much
about the remaining pages on that list; they might all be in heavy use,
and once the chunk of easy victims has been purged, the VM continues to
apply elevated pressure on those remaining hot pages.  The other LRU,
meanwhile, might have easily reclaimable pages, and there was never a need
to steer away from it in the first place.

As the previous patch outlined, we should focus on recording actually
observed cost to steer the balance rather than speculating about the
potential value of one LRU list over the other.  In that spirit, leave
explicitely deactivated pages to the LRU algorithm to pick up, and let
rotations decide which list is the easiest to reclaim.

[[email protected]: fix set-but-not-used warning]
  Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Qian Cai <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:49 -07:00
Johannes Weiner 1431d4d11a mm: base LRU balancing on an explicit cost model 
Currently, scan pressure between the anon and file LRU lists is balanced
based on a mixture of reclaim efficiency and a somewhat vague notion of
"value" of having certain pages in memory over others.  That concept of
value is problematic, because it has caused us to count any event that
remotely makes one LRU list more or less preferrable for reclaim, even
when these events are not directly comparable and impose very different
costs on the system.  One example is referenced file pages that we still
deactivate and referenced anonymous pages that we actually rotate back to
the head of the list.

There is also conceptual overlap with the LRU algorithm itself.  By
rotating recently used pages instead of reclaiming them, the algorithm
already biases the applied scan pressure based on page value.  Thus, when
rebalancing scan pressure due to rotations, we should think of reclaim
cost, and leave assessing the page value to the LRU algorithm.

Lastly, considering both value-increasing as well as value-decreasing
events can sometimes cause the same type of event to be counted twice,
i.e.  how rotating a page increases the LRU value, while reclaiming it
succesfully decreases the value.  In itself this will balance out fine,
but it quietly skews the impact of events that are only recorded once.

The abstract metric of "value", the murky relationship with the LRU
algorithm, and accounting both negative and positive events make the
current pressure balancing model hard to reason about and modify.

This patch switches to a balancing model of accounting the concrete,
actually observed cost of reclaiming one LRU over another.  For now, that
cost includes pages that are scanned but rotated back to the list head.
Subsequent patches will add consideration for IO caused by refaulting of
recently evicted pages.

Replace struct zone_reclaim_stat with two cost counters in the lruvec, and
make everything that affects cost go through a new lru_note_cost()
function.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner a4fe1631f3 mm: vmscan: drop unnecessary div0 avoidance rounding in get_scan_count() 
When we calculate the relative scan pressure between the anon and file LRU
lists, we have to assume that reclaim_stat can contain zeroes.  To avoid
div0 crashes, we add 1 to all denominators like so:

        anon_prio = swappiness;
        file_prio = 200 - anon_prio;

	[...]

        /*
         * The amount of pressure on anon vs file pages is inversely
         * proportional to the fraction of recently scanned pages on
         * each list that were recently referenced and in active use.
         */
        ap = anon_prio * (reclaim_stat->recent_scanned[0] + 1);
        ap /= reclaim_stat->recent_rotated[0] + 1;

        fp = file_prio * (reclaim_stat->recent_scanned[1] + 1);
        fp /= reclaim_stat->recent_rotated[1] + 1;
        spin_unlock_irq(&pgdat->lru_lock);

        fraction[0] = ap;
        fraction[1] = fp;
        denominator = ap + fp + 1;

While reclaim_stat can contain 0, it's not actually possible for ap + fp
to be 0.  One of anon_prio or file_prio could be zero, but they must still
add up to 200.  And the reclaim_stat fraction, due to the +1 in there, is
always at least 1.  So if one of the two numerators is 0, the other one
can't be.  ap + fp is always at least 1.  Drop the + 1.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner 9682468747 mm: remove use-once cache bias from LRU balancing 
When the splitlru patches divided page cache and swap-backed pages into
separate LRU lists, the pressure balance between the lists was biased to
account for the fact that streaming IO can cause memory pressure with a
flood of pages that are used only once.  New page cache additions would
tip the balance toward the file LRU, and repeat access would neutralize
that bias again.  This ensured that page reclaim would always go for
used-once cache first.

Since e986850598 ("mm,vmscan: only evict file pages when we have
plenty"), page reclaim generally skips over swap-backed memory entirely as
long as there is used-once cache present, and will apply the LRU balancing
when only repeatedly accessed cache pages are left - at which point the
previous use-once bias will have been neutralized.  This makes the
use-once cache balancing bias unnecessary.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner 34e58cac6d mm: workingset: let cache workingset challenge anon 
We activate cache refaults with reuse distances in pages smaller than the
size of the total cache.  This allows new pages with competitive access
frequencies to establish themselves, as well as challenge and potentially
displace pages on the active list that have gone cold.

However, that assumes that active cache can only replace other active
cache in a competition for the hottest memory.  This is not a great
default assumption.  The page cache might be thrashing while there are
enough completely cold and unused anonymous pages sitting around that we'd
only have to write to swap once to stop all IO from the cache.

Activate cache refaults when their reuse distance in pages is smaller than
the total userspace workingset, including anonymous pages.

Reclaim can still decide how to balance pressure among the two LRUs
depending on the IO situation.  Rotational drives will prefer avoiding
random IO from swap and go harder after cache.  But fundamentally, hot
cache should be able to compete with anon pages for a place in RAM.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner 6058eaec81 mm: fold and remove lru_cache_add_anon() and lru_cache_add_file() 
They're the same function, and for the purpose of all callers they are
equivalent to lru_cache_add().

[[email protected]: fix it for local_lock changes]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Rik van Riel <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner c843966c55 mm: allow swappiness that prefers reclaiming anon over the file workingset 
With the advent of fast random IO devices (SSDs, PMEM) and in-memory swap
devices such as zswap, it's possible for swap to be much faster than
filesystems, and for swapping to be preferable over thrashing filesystem
caches.

Allow setting swappiness - which defines the rough relative IO cost of
cache misses between page cache and swap-backed pages - to reflect such
situations by making the swap-preferred range configurable.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner 497a6c1b09 mm: keep separate anon and file statistics on page reclaim activity 
Having statistics on pages scanned and pages reclaimed for both anon and
file pages makes it easier to evaluate changes to LRU balancing.

While at it, clean up the stat-keeping mess for isolation, putback,
reclaim stats etc.  a bit: first the physical LRU operation (isolation and
putback), followed by vmstats, reclaim_stats, and then vm events.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Minchan Kim <[email protected]>
Cc: Rik van Riel <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner 5df741963d mm: fix LRU balancing effect of new transparent huge pages 
The reclaim code that balances between swapping and cache reclaim tries to
predict likely reuse based on in-memory reference patterns alone.  This
works in many cases, but when it fails it cannot detect when the cache is
thrashing pathologically, or when we're in the middle of a swap storm.

The high seek cost of rotational drives under which the algorithm evolved
also meant that mistakes could quickly result in lockups from too
aggressive swapping (which is predominantly random IO).  As a result, the
balancing code has been tuned over time to a point where it mostly goes
for page cache and defers swapping until the VM is under significant
memory pressure.

The resulting strategy doesn't make optimal caching decisions - where
optimal is the least amount of IO required to execute the workload.

The proliferation of fast random IO devices such as SSDs, in-memory
compression such as zswap, and persistent memory technologies on the
horizon, has made this undesirable behavior very noticable: Even in the
presence of large amounts of cold anonymous memory and a capable swap
device, the VM refuses to even seriously scan these pages, and can leave
the page cache thrashing needlessly.

This series sets out to address this.  Since commit ("a528910e12ec mm:
thrash detection-based file cache sizing") we have exact tracking of
refault IO - the ultimate cost of reclaiming the wrong pages.  This allows
us to use an IO cost based balancing model that is more aggressive about
scanning anonymous memory when the cache is thrashing, while being able to
avoid unnecessary swap storms.

These patches base the LRU balance on the rate of refaults on each list,
times the relative IO cost between swap device and filesystem
(swappiness), in order to optimize reclaim for least IO cost incurred.

	History

I floated these changes in 2016.  At the time they were incomplete and
full of workarounds due to a lack of infrastructure in the reclaim code:
We didn't have PageWorkingset, we didn't have hierarchical cgroup
statistics, and problems with the cgroup swap controller.  As swapping
wasn't too high a priority then, the patches stalled out.  With all
dependencies in place now, here we are again with much cleaner,
feature-complete patches.

I kept the acks for patches that stayed materially the same :-)

Below is a series of test results that demonstrate certain problematic
behavior of the current code, as well as showcase the new code's more
predictable and appropriate balancing decisions.

	Test #1: No convergence

This test shows an edge case where the VM currently doesn't converge at
all on a new file workingset with a stale anon/tmpfs set.

The test sets up a cold anon set the size of 3/4 RAM, then tries to
establish a new file set half the size of RAM (flat access pattern).

The vanilla kernel refuses to even scan anon pages and never converges.
The file set is perpetually served from the filesystem.

The first test kernel is with the series up to the workingset patch
applied.  This allows thrashing page cache to challenge the anonymous
workingset.  The VM then scans the lists based on the current
scanned/rotated balancing algorithm.  It converges on a stable state where
all cold anon pages are pushed out and the fileset is served entirely from
cache:

			    noconverge/5.7-rc5-mm	noconverge/5.7-rc5-mm-workingset
Scanned			417719308.00 (    +0.00%)		64091155.00 (   -84.66%)
Reclaimed		417711094.00 (    +0.00%)		61640308.00 (   -85.24%)
Reclaim efficiency %	      100.00 (    +0.00%)		      96.18 (    -3.78%)
Scanned file		417719308.00 (    +0.00%)		59211118.00 (   -85.83%)
Scanned anon			0.00 (    +0.00%)	         4880037.00 (          )
Swapouts			0.00 (    +0.00%)	         2439957.00 (          )
Swapins				0.00 (    +0.00%)		     257.00 (          )
Refaults		415246605.00 (    +0.00%)		59183722.00 (   -85.75%)
Restore refaults		0.00 (    +0.00%)	        54988252.00 (          )

The second test kernel is with the full patch series applied, which
replaces the scanned/rotated ratios with refault/swapin rate-based
balancing.  It evicts the cold anon pages more aggressively in the
presence of a thrashing cache and the absence of swapins, and so converges
with about 60% of the IO and reclaim activity:

			noconverge/5.7-rc5-mm-workingset	noconverge/5.7-rc5-mm-lrubalance
Scanned				64091155.00 (    +0.00%)		37579741.00 (   -41.37%)
Reclaimed			61640308.00 (    +0.00%)		35129293.00 (   -43.01%)
Reclaim efficiency %		      96.18 (    +0.00%)		      93.48 (    -2.78%)
Scanned file			59211118.00 (    +0.00%)		32708385.00 (   -44.76%)
Scanned anon			 4880037.00 (    +0.00%)		 4871356.00 (    -0.18%)
Swapouts			 2439957.00 (    +0.00%)		 2435565.00 (    -0.18%)
Swapins				     257.00 (    +0.00%)		     262.00 (    +1.94%)
Refaults			59183722.00 (    +0.00%)		32675667.00 (   -44.79%)
Restore refaults		54988252.00 (    +0.00%)		28480430.00 (   -48.21%)

We're triggering this case in host sideloading scenarios: When a host's
primary workload is not saturating the machine (primary load is usually
driven by user activity), we can optimistically sideload a batch job; if
user activity picks up and the primary workload needs the whole host
during this time, we freeze the sideload and rely on it getting pushed to
swap.  Frequently that swapping doesn't happen and the completely inactive
sideload simply stays resident while the expanding primary worklad is
struggling to gain ground.

	Test #2: Kernel build

This test is a a kernel build that is slightly memory-restricted (make -j4
inside a 400M cgroup).

Despite the very aggressive swapping of cold anon pages in test #1, this
test shows that the new kernel carefully balances swap against cache
refaults when both the file and the cache set are pressured.

It shows the patched kernel to be slightly better at finding the coldest
memory from the combined anon and file set to evict under pressure.  The
result is lower aggregate reclaim and paging activity:

z				    5.7-rc5-mm	5.7-rc5-mm-lrubalance
Real time		   210.60 (    +0.00%)	   210.97 (    +0.18%)
User time		   745.42 (    +0.00%)	   746.48 (    +0.14%)
System time		    69.78 (    +0.00%)	    69.79 (    +0.02%)
Scanned file		354682.00 (    +0.00%)	293661.00 (   -17.20%)
Scanned anon		465381.00 (    +0.00%)	378144.00 (   -18.75%)
Swapouts		185920.00 (    +0.00%)	147801.00 (   -20.50%)
Swapins			 34583.00 (    +0.00%)	 32491.00 (    -6.05%)
Refaults		212664.00 (    +0.00%)	172409.00 (   -18.93%)
Restore refaults	 48861.00 (    +0.00%)	 80091.00 (   +63.91%)
Total paging IO		433167.00 (    +0.00%)	352701.00 (   -18.58%)

	Test #3: Overload

This next test is not about performance, but rather about the
predictability of the algorithm.  The current balancing behavior doesn't
always lead to comprehensible results, which makes performance analysis
and parameter tuning (swappiness e.g.) very difficult.

The test shows the balancing behavior under equivalent anon and file
input.  Anon and file sets are created of equal size (3/4 RAM), have the
same access patterns (a hot-cold gradient), and synchronized access rates.
Swappiness is raised from the default of 60 to 100 to indicate equal IO
cost between swap and cache.

With the vanilla balancing code, anon scans make up around 9% of the total
pages scanned, or a ~1:10 ratio.  This is a surprisingly skewed ratio, and
it's an outcome that is hard to explain given the input parameters to the
VM.

The new balancing model targets a 1:2 balance: All else being equal,
reclaiming a file page costs one page IO - the refault; reclaiming an anon
page costs two IOs - the swapout and the swapin.  In the test we observe a
~1:3 balance.

The scanned and paging IO numbers indicate that the anon LRU algorithm we
have in place right now does a slightly worse job at picking the coldest
pages compared to the file algorithm.  There is ongoing work to improve
this, like Joonsoo's anon workingset patches; however, it's difficult to
compare the two aging strategies when the balancing between them is
behaving unintuitively.

The slightly less efficient anon reclaim results in a deviation from the
optimal 1:2 scan ratio we would like to see here - however, 1:3 is much
closer to what we'd want to see in this test than the vanilla kernel's
aging of 10+ cache pages for every anonymous one:

			overload-100/5.7-rc5-mm-workingset	overload-100/5.7-rc5-mm-lrubalance-realfile
Scanned				 533633725.00 (    +0.00%)			  595687785.00 (   +11.63%)
Reclaimed			 494325440.00 (    +0.00%)			  518154380.00 (    +4.82%)
Reclaim efficiency %			92.63 (    +0.00%)				 86.98 (    -6.03%)
Scanned file			 484532894.00 (    +0.00%)			  456937722.00 (    -5.70%)
Scanned anon			  49100831.00 (    +0.00%)			  138750063.00 (  +182.58%)
Swapouts			   8096423.00 (    +0.00%)			   48982142.00 (  +504.98%)
Swapins				  10027384.00 (    +0.00%)			   62325044.00 (  +521.55%)
Refaults			 479819973.00 (    +0.00%)			  451309483.00 (    -5.94%)
Restore refaults		 426422087.00 (    +0.00%)			  399914067.00 (    -6.22%)
Total paging IO			 497943780.00 (    +0.00%)			  562616669.00 (   +12.99%)

	Test #4: Parallel IO

It's important to note that these patches only affect the situation where
the kernel has to reclaim workingset memory, which is usually a
transitionary period.  The vast majority of page reclaim occuring in a
system is from trimming the ever-expanding page cache.

These patches don't affect cache trimming behavior.  We never swap as long
as we only have use-once cache moving through the file LRU, we only
consider swapping when the cache is actively thrashing.

The following test demonstrates this.  It has an anon workingset that
takes up half of RAM and then writes a file that is twice the size of RAM
out to disk.

As the cache is funneled through the inactive file list, no anon pages are
scanned (aside from apparently some background noise of 10 pages):

					  5.7-rc5-mm		          5.7-rc5-mm-lrubalance
Scanned			    10714722.00 (    +0.00%)		       10723445.00 (    +0.08%)
Reclaimed		    10703596.00 (    +0.00%)		       10712166.00 (    +0.08%)
Reclaim efficiency %		  99.90 (    +0.00%)			     99.89 (    -0.00%)
Scanned file		    10714722.00 (    +0.00%)		       10723435.00 (    +0.08%)
Scanned anon			   0.00 (    +0.00%)			     10.00 (          )
Swapouts			   0.00 (    +0.00%)			      7.00 (          )
Swapins				   0.00 (    +0.00%)			      0.00 (    +0.00%)
Refaults			  92.00 (    +0.00%)			     41.00 (   -54.84%)
Restore refaults		   0.00 (    +0.00%)			      0.00 (    +0.00%)
Total paging IO			  92.00 (    +0.00%)			     48.00 (   -47.31%)

This patch (of 14):

Currently, THP are counted as single pages until they are split right
before being swapped out.  However, at that point the VM is already in the
middle of reclaim, and adjusting the LRU balance then is useless.

Always account THP by the number of basepages, and remove the fixup from
the splitting path.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Shakeel Butt <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Rik van Riel <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner a0b5b4147f mm: memcontrol: update page->mem_cgroup stability rules 
The previous patches have simplified the access rules around
page->mem_cgroup somewhat:

1. We never change page->mem_cgroup while the page is isolated by
   somebody else.  This was by far the biggest exception to our rules and
   it didn't stop at lock_page() or lock_page_memcg().

2. We charge pages before they get put into page tables now, so the
   somewhat fishy rule about "can be in page table as long as it's still
   locked" is now gone and boiled down to having an exclusive reference to
   the page.

Document the new rules.  Any of the following will stabilize the
page->mem_cgroup association:

- the page lock
- LRU isolation
- lock_page_memcg()
- exclusive access to the page

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Alex Shi <[email protected]>
Reviewed-by: Joonsoo Kim <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Balbir Singh <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner d9eb1ea2bf mm: memcontrol: delete unused lrucare handling 
Swapin faults were the last event to charge pages after they had already
been put on the LRU list.  Now that we charge directly on swapin, the
lrucare portion of the charge code is unused.

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Joonsoo Kim <[email protected]>
Cc: Alex Shi <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Shakeel Butt <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner 4c6355b25e mm: memcontrol: charge swapin pages on instantiation 
Right now, users that are otherwise memory controlled can easily escape
their containment and allocate significant amounts of memory that they're
not being charged for.  That's because swap readahead pages are not being
charged until somebody actually faults them into their page table.  This
can be exploited with MADV_WILLNEED, which triggers arbitrary readahead
allocations without charging the pages.

There are additional problems with the delayed charging of swap pages:

1. To implement refault/workingset detection for anonymous pages, we
   need to have a target LRU available at swapin time, but the LRU is not
   determinable until the page has been charged.

2. To implement per-cgroup LRU locking, we need page->mem_cgroup to be
   stable when the page is isolated from the LRU; otherwise, the locks
   change under us.  But swapcache gets charged after it's already on the
   LRU, and even if we cannot isolate it ourselves (since charging is not
   exactly optional).

The previous patch ensured we always maintain cgroup ownership records for
swap pages.  This patch moves the swapcache charging point from the fault
handler to swapin time to fix all of the above problems.

v2: simplify swapin error checking (Joonsoo)

[[email protected]: fix livelock in __read_swap_cache_async()]
  Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Hugh Dickins <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Alex Shi <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Joonsoo Kim <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Rafael Aquini <[email protected]>
Cc: Alex Shi <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner 2d1c498072 mm: memcontrol: make swap tracking an integral part of memory control 
Without swap page tracking, users that are otherwise memory controlled can
easily escape their containment and allocate significant amounts of memory
that they're not being charged for.  That's because swap does readahead,
but without the cgroup records of who owned the page at swapout, readahead
pages don't get charged until somebody actually faults them into their
page table and we can identify an owner task.  This can be maliciously
exploited with MADV_WILLNEED, which triggers arbitrary readahead
allocations without charging the pages.

Make swap swap page tracking an integral part of memcg and remove the
Kconfig options.  In the first place, it was only made configurable to
allow users to save some memory.  But the overhead of tracking cgroup
ownership per swap page is minimal - 2 byte per page, or 512k per 1G of
swap, or 0.04%.  Saving that at the expense of broken containment
semantics is not something we should present as a coequal option.

The swapaccount=0 boot option will continue to exist, and it will
eliminate the page_counter overhead and hide the swap control files, but
it won't disable swap slot ownership tracking.

This patch makes sure we always have the cgroup records at swapin time;
the next patch will fix the actual bug by charging readahead swap pages at
swapin time rather than at fault time.

v2: fix double swap charge bug in cgroup1/cgroup2 code gating

[[email protected]: fix crash with cgroup_disable=memory]
  Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Joonsoo Kim <[email protected]>
Cc: Alex Shi <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Naresh Kamboju <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Debugged-by: Hugh Dickins <[email protected]>
Debugged-by: Michal Hocko <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner eccb52e788 mm: memcontrol: prepare swap controller setup for integration 
A few cleanups to streamline the swap controller setup:

- Replace the do_swap_account flag with cgroup_memory_noswap. This
  brings it in line with other functionality that is usually available
  unless explicitly opted out of - nosocket, nokmem.

- Remove the really_do_swap_account flag that stores the boot option
  and is later used to switch the do_swap_account. It's not clear why
  this indirection is/was necessary. Use do_swap_account directly.

- Minor coding style polishing

Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Joonsoo Kim <[email protected]>
Cc: Alex Shi <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Balbir Singh <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner f0e45fb4da mm: memcontrol: drop unused try/commit/cancel charge API 
There are no more users. RIP in peace.

[[email protected]: fix an unused-function warning]
  Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Arnd Bergmann <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Joonsoo Kim <[email protected]>
Cc: Alex Shi <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Balbir Singh <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner 9d82c69438 mm: memcontrol: convert anon and file-thp to new mem_cgroup_charge() API 
With the page->mapping requirement gone from memcg, we can charge anon and
file-thp pages in one single step, right after they're allocated.

This removes two out of three API calls - especially the tricky commit
step that needed to happen at just the right time between when the page is
"set up" and when it's "published" - somewhat vague and fluid concepts
that varied by page type.  All we need is a freshly allocated page and a
memcg context to charge.

v2: prevent double charges on pre-allocated hugepages in khugepaged

[[email protected]: Fix crash - *hpage could be ERR_PTR instead of NULL]
  Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Joonsoo Kim <[email protected]>
Cc: Alex Shi <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Balbir Singh <[email protected]>
Cc: Qian Cai <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:48 -07:00
Johannes Weiner 468c398233 mm: memcontrol: switch to native NR_ANON_THPS counter 
With rmap memcg locking already in place for NR_ANON_MAPPED, it's just a
small step to remove the MEMCG_RSS_HUGE wart and switch memcg to the
native NR_ANON_THPS accounting sites.

[[email protected]: fixes]
  Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Johannes Weiner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Tested-by: Naresh Kamboju <[email protected]>
Reviewed-by: Joonsoo Kim <[email protected]>
Acked-by: Randy Dunlap <[email protected]>	[build-tested]
Cc: Alex Shi <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: "Kirill A. Shutemov" <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Roman Gushchin <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: Balbir Singh <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
 2020-06-03 20:09:47 -07:00