Prior to this commit, we only directly check the affected cgroup's
memory.high against its usage.  However, it's possible that we are being
reclaimed as a result of hitting an ancestor memory.high and should be
penalised based on that, instead.

This patch changes memory.high overage throttling to use the largest
overage in its ancestors when considering how many penalty jiffies to
charge.  This makes sure that we penalise poorly behaving cgroups in the
same way regardless of at what level of the hierarchy memory.high was
breached.

Fixes: 0e4b01df ("mm, memcg: throttle allocators when failing reclaim over memory.high")
Reported-by: NJohannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NChris Down <chris@chrisdown.name>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Nathan Chancellor <natechancellor@gmail.com>
Cc: Roman Gushchin <guro@fb.com>
Cc: <stable@vger.kernel.org>	[5.4.x+]
Link: http://lkml.kernel.org/r/8cd132f84bd7e16cdb8fde3378cdbf05ba00d387.1584036142.git.chris@chrisdown.nameSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 0e4b01df had a bunch of fixups to use the right division
method.  However, it seems that after all that it still wasn't right --
div_u64 takes a 32-bit divisor.

The headroom is still large (2^32 pages), so on mundane systems you
won't hit this, but this should definitely be fixed.

Fixes: 0e4b01df ("mm, memcg: throttle allocators when failing reclaim over memory.high")
Reported-by: NJohannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NChris Down <chris@chrisdown.name>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Nathan Chancellor <natechancellor@gmail.com>
Cc: <stable@vger.kernel.org>	[5.4.x+]
Link: http://lkml.kernel.org/r/80780887060514967d414b3cd91f9a316a16ab98.1584036142.git.chris@chrisdown.nameSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

An eventfd monitors multiple memory thresholds of the cgroup, closes them,
the kernel deletes all events related to this eventfd.  Before all events
are deleted, another eventfd monitors the memory threshold of this cgroup,
leading to a crash:

  BUG: kernel NULL pointer dereference, address: 0000000000000004
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 800000033058e067 P4D 800000033058e067 PUD 3355ce067 PMD 0
  Oops: 0002 [#1] SMP PTI
  CPU: 2 PID: 14012 Comm: kworker/2:6 Kdump: loaded Not tainted 5.6.0-rc4 #3
  Hardware name: LENOVO 20AWS01K00/20AWS01K00, BIOS GLET70WW (2.24 ) 05/21/2014
  Workqueue: events memcg_event_remove
  RIP: 0010:__mem_cgroup_usage_unregister_event+0xb3/0x190
  RSP: 0018:ffffb47e01c4fe18 EFLAGS: 00010202
  RAX: 0000000000000001 RBX: ffff8bb223a8a000 RCX: 0000000000000001
  RDX: 0000000000000001 RSI: ffff8bb22fb83540 RDI: 0000000000000001
  RBP: ffffb47e01c4fe48 R08: 0000000000000000 R09: 0000000000000010
  R10: 000000000000000c R11: 071c71c71c71c71c R12: ffff8bb226aba880
  R13: ffff8bb223a8a480 R14: 0000000000000000 R15: 0000000000000000
  FS:  0000000000000000(0000) GS:ffff8bb242680000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000004 CR3: 000000032c29c003 CR4: 00000000001606e0
  Call Trace:
    memcg_event_remove+0x32/0x90
    process_one_work+0x172/0x380
    worker_thread+0x49/0x3f0
    kthread+0xf8/0x130
    ret_from_fork+0x35/0x40
  CR2: 0000000000000004

We can reproduce this problem in the following ways:

1. We create a new cgroup subdirectory and a new eventfd, and then we
   monitor multiple memory thresholds of the cgroup through this eventfd.

2.  closing this eventfd, and __mem_cgroup_usage_unregister_event ()
   will be called multiple times to delete all events related to this
   eventfd.

The first time __mem_cgroup_usage_unregister_event() is called, the
kernel will clear all items related to this eventfd in thresholds->
primary.

Since there is currently only one eventfd, thresholds-> primary becomes
empty, so the kernel will set thresholds-> primary and hresholds-> spare
to NULL.  If at this time, the user creates a new eventfd and monitor
the memory threshold of this cgroup, kernel will re-initialize
thresholds-> primary.

Then when __mem_cgroup_usage_unregister_event () is called for the
second time, because thresholds-> primary is not empty, the system will
access thresholds-> spare, but thresholds-> spare is NULL, which will
trigger a crash.

In general, the longer it takes to delete all events related to this
eventfd, the easier it is to trigger this problem.

The solution is to check whether the thresholds associated with the
eventfd has been cleared when deleting the event.  If so, we do nothing.

[akpm@linux-foundation.org: fix comment, per Kirill]
Fixes: 907860ed ("cgroups: make cftype.unregister_event() void-returning")
Signed-off-by: NChunguang Xu <brookxu@tencent.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Acked-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/077a6f67-aefa-4591-efec-f2f3af2b0b02@gmail.comSigned-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

If a TCP socket is allocated in IRQ context or cloned from unassociated
(i.e. not associated to a memcg) in IRQ context then it will remain
unassociated for its whole life. Almost half of the TCPs created on the
system are created in IRQ context, so, memory used by such sockets will
not be accounted by the memcg.

This issue is more widespread in cgroup v1 where network memory
accounting is opt-in but it can happen in cgroup v2 if the source socket
for the cloning was created in root memcg.

To fix the issue, just do the association of the sockets at the accept()
time in the process context and then force charge the memory buffer
already used and reserved by the socket.
Signed-off-by: NShakeel Butt <shakeelb@google.com>
Reviewed-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We are testing network memory accounting in our setup and noticed
inconsistent network memory usage and often unrelated cgroups network
usage correlates with testing workload. On further inspection, it
seems like mem_cgroup_sk_alloc() and cgroup_sk_alloc() are broken in
irq context specially for cgroup v1.

mem_cgroup_sk_alloc() and cgroup_sk_alloc() can be called in irq context
and kind of assumes that this can only happen from sk_clone_lock()
and the source sock object has already associated cgroup. However in
cgroup v1, where network memory accounting is opt-in, the source sock
can be unassociated with any cgroup and the new cloned sock can get
associated with unrelated interrupted cgroup.

Cgroup v2 can also suffer if the source sock object was created by
process in the root cgroup or if sk_alloc() is called in irq context.
The fix is to just do nothing in interrupt.

WARNING: Please note that about half of the TCP sockets are allocated
from the IRQ context, so, memory used by such sockets will not be
accouted by the memcg.

The stack trace of mem_cgroup_sk_alloc() from IRQ-context:

CPU: 70 PID: 12720 Comm: ssh Tainted:  5.6.0-smp-DEV #1
Hardware name: ...
Call Trace:
 <IRQ>
 dump_stack+0x57/0x75
 mem_cgroup_sk_alloc+0xe9/0xf0
 sk_clone_lock+0x2a7/0x420
 inet_csk_clone_lock+0x1b/0x110
 tcp_create_openreq_child+0x23/0x3b0
 tcp_v6_syn_recv_sock+0x88/0x730
 tcp_check_req+0x429/0x560
 tcp_v6_rcv+0x72d/0xa40
 ip6_protocol_deliver_rcu+0xc9/0x400
 ip6_input+0x44/0xd0
 ? ip6_protocol_deliver_rcu+0x400/0x400
 ip6_rcv_finish+0x71/0x80
 ipv6_rcv+0x5b/0xe0
 ? ip6_sublist_rcv+0x2e0/0x2e0
 process_backlog+0x108/0x1e0
 net_rx_action+0x26b/0x460
 __do_softirq+0x104/0x2a6
 do_softirq_own_stack+0x2a/0x40
 </IRQ>
 do_softirq.part.19+0x40/0x50
 __local_bh_enable_ip+0x51/0x60
 ip6_finish_output2+0x23d/0x520
 ? ip6table_mangle_hook+0x55/0x160
 __ip6_finish_output+0xa1/0x100
 ip6_finish_output+0x30/0xd0
 ip6_output+0x73/0x120
 ? __ip6_finish_output+0x100/0x100
 ip6_xmit+0x2e3/0x600
 ? ipv6_anycast_cleanup+0x50/0x50
 ? inet6_csk_route_socket+0x136/0x1e0
 ? skb_free_head+0x1e/0x30
 inet6_csk_xmit+0x95/0xf0
 __tcp_transmit_skb+0x5b4/0xb20
 __tcp_send_ack.part.60+0xa3/0x110
 tcp_send_ack+0x1d/0x20
 tcp_rcv_state_process+0xe64/0xe80
 ? tcp_v6_connect+0x5d1/0x5f0
 tcp_v6_do_rcv+0x1b1/0x3f0
 ? tcp_v6_do_rcv+0x1b1/0x3f0
 __release_sock+0x7f/0xd0
 release_sock+0x30/0xa0
 __inet_stream_connect+0x1c3/0x3b0
 ? prepare_to_wait+0xb0/0xb0
 inet_stream_connect+0x3b/0x60
 __sys_connect+0x101/0x120
 ? __sys_getsockopt+0x11b/0x140
 __x64_sys_connect+0x1a/0x20
 do_syscall_64+0x51/0x200
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

The stack trace of mem_cgroup_sk_alloc() from IRQ-context:
Fixes: 2d758073 ("mm: memcontrol: consolidate cgroup socket tracking")
Fixes: d979a39d ("cgroup: duplicate cgroup reference when cloning sockets")
Signed-off-by: NShakeel Butt <shakeelb@google.com>
Reviewed-by: NRoman Gushchin <guro@fb.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

for_each_mem_cgroup() increases css reference counter for memory cgroup
and requires to use mem_cgroup_iter_break() if the walk is cancelled.

Link: http://lkml.kernel.org/r/c98414fb-7e1f-da0f-867a-9340ec4bd30b@virtuozzo.com
Fixes: 0a4465d3 ("mm, memcg: assign memcg-aware shrinkers bitmap to memcg")
Signed-off-by: NVasily Averin <vvs@virtuozzo.com>
Acked-by: NKirill Tkhai <ktkhai@virtuozzo.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Reviewed-by: NRoman Gushchin <guro@fb.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Compound pages handling in mem_cgroup_migrate is more convoluted than
necessary.  The state is duplicated in compound variable and the same
could be achieved by PageTransHuge check which is trivial and
hpage_nr_pages is already PageTransHuge aware.

It is much simpler to just use hpage_nr_pages for nr_pages and replace
the local variable by PageTransHuge check directly

Link: http://lkml.kernel.org/r/20191210160450.3395-1-pilgrimtao@gmail.comSigned-off-by: NKaitao Cheng <pilgrimtao@gmail.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

If compound is true, this means it is a PMD mapped THP.  Which implies
the page is not linked to any defer list.  So the first code chunk will
not be executed.

Also with this reason, it would not be proper to add this page to a
defer list.  So the second code chunk is not correct.

Based on this, we should remove the defer list related code.

[yang.shi@linux.alibaba.com: better patch title]
Link: http://lkml.kernel.org/r/20200117233836.3434-1-richardw.yang@linux.intel.com
Fixes: 87eaceb3 ("mm: thp: make deferred split shrinker memcg aware")
Signed-off-by: NWei Yang <richardw.yang@linux.intel.com>
Suggested-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: NYang Shi <yang.shi@linux.alibaba.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: <stable@vger.kernel.org>    [5.4+]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently slab percpu vmstats are flushed twice: during the memcg
offlining and just before freeing the memcg structure.  Each time percpu
counters are summed, added to the atomic counterparts and propagated up
by the cgroup tree.

The second flushing is required due to how recursive vmstats are
implemented: counters are batched in percpu variables on a local level,
and once a percpu value is crossing some predefined threshold, it spills
over to atomic values on the local and each ascendant levels.  It means
that without flushing some numbers cached in percpu variables will be
dropped on floor each time a cgroup is destroyed.  And with uptime the
error on upper levels might become noticeable.

The first flushing aims to make counters on ancestor levels more
precise.  Dying cgroups may resume in the dying state for a long time.
After kmem_cache reparenting which is performed during the offlining
slab counters of the dying cgroup don't have any chances to be updated,
because any slab operations will be performed on the parent level.  It
means that the inaccuracy caused by percpu batching will not decrease up
to the final destruction of the cgroup.  By the original idea flushing
slab counters during the offlining should minimize the visible
inaccuracy of slab counters on the parent level.

The problem is that percpu counters are not zeroed after the first
flushing.  So every cached percpu value is summed twice.  It creates a
small error (up to 32 pages per cpu, but usually less) which accumulates
on parent cgroup level.  After creating and destroying of thousands of
child cgroups, slab counter on parent level can be way off the real
value.

For now, let's just stop flushing slab counters on memcg offlining.  It
can't be done correctly without scheduling a work on each cpu: reading
and zeroing it during css offlining can race with an asynchronous
update, which doesn't expect values to be changed underneath.

With this change, slab counters on parent level will become eventually
consistent.  Once all dying children are gone, values are correct.  And
if not, the error is capped by 32 * NR_CPUS pages per dying cgroup.

It's not perfect, as slab are reparented, so any updates after the
reparenting will happen on the parent level.  It means that if a slab
page was allocated, a counter on child level was bumped, then the page
was reparented and freed, the annihilation of positive and negative
counter values will not happen until the child cgroup is released.  It
makes slab counters different from others, and it might want us to
implement flushing in a correct form again.  But it's also a question of
performance: scheduling a work on each cpu isn't free, and it's an open
question if the benefit of having more accurate counters is worth it.

We might also consider flushing all counters on offlining, not only slab
counters.

So let's fix the main problem now: make the slab counters eventually
consistent, so at least the error won't grow with uptime (or more
precisely the number of created and destroyed cgroups).  And think about
the accuracy of counters separately.

Link: http://lkml.kernel.org/r/20191220042728.1045881-1-guro@fb.com
Fixes: bee07b33 ("mm: memcontrol: flush percpu slab vmstats on kmem offlining")
Signed-off-by: NRoman Gushchin <guro@fb.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Use common names from vmstat array when possible.  This gives not much
difference in code size for now, but should help in keeping interfaces
consistent.

  add/remove: 0/2 grow/shrink: 2/0 up/down: 70/-72 (-2)
  Function                                     old     new   delta
  memory_stat_format                           984    1050     +66
  memcg_stat_show                              957     961      +4
  memcg1_event_names                            32       -     -32
  mem_cgroup_lru_names                          40       -     -40
  Total: Before=14485337, After=14485335, chg -0.00%

Link: http://lkml.kernel.org/r/157113012508.453.80391533767219371.stgit@buzzSigned-off-by: NKonstantin Khlebnikov <khlebnikov@yandex-team.ru>
Acked-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

There is a per-memcg lruvec and a NUMA node lruvec.  Which one is being
used is somewhat confusing right now, and it's easy to make mistakes -
especially when it comes to global reclaim.

How it works: when memory cgroups are enabled, we always use the
root_mem_cgroup's per-node lruvecs.  When memory cgroups are not compiled
in or disabled at runtime, we use pgdat->lruvec.

Document that in a comment.

Due to the way the reclaim code is generalized, all lookups use the
mem_cgroup_lruvec() helper function, and nobody should have to find the
right lruvec manually right now.  But to avoid future mistakes, rename the
pgdat->lruvec member to pgdat->__lruvec and delete the convenience wrapper
that suggests it's a commonly accessed member.

While in this area, swap the mem_cgroup_lruvec() argument order.  The name
suggests a memcg operation, yet it takes a pgdat first and a memcg second.
I have to double take every time I call this.  Fix that.

Link: http://lkml.kernel.org/r/20191022144803.302233-3-hannes@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NMichal Hocko <mhocko@suse.com>
Reviewed-by: NShakeel Butt <shakeelb@google.com>
Cc: Roman Gushchin <guro@fb.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Since commit 1ba6fc9a ("mm: vmscan: do not share cgroup iteration
between reclaimers"), the memcg reclaim does not bail out earlier based
on sc->nr_reclaimed and will traverse all the nodes.  All the
reclaimable pages of the memcg on all the nodes will be scanned relative
to the reclaim priority.  So, there is no need to maintain state
regarding which node to start the memcg reclaim from.

This patch effectively reverts the commit 889976db ("memcg: reclaim
memory from nodes in round-robin order") and commit 453a9bf3
("memcg: fix numa scan information update to be triggered by memory
event").

[shakeelb@google.com: v2]
  Link: http://lkml.kernel.org/r/20191030204232.139424-1-shakeelb@google.com
Link: http://lkml.kernel.org/r/20191029234753.224143-1-shakeelb@google.comSigned-off-by: NShakeel Butt <shakeelb@google.com>
Acked-by: NRoman Gushchin <guro@fb.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Greg Thelen <gthelen@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Setting a memory.high limit below the usage makes almost no effort to
shrink the cgroup to the new target size.

While memory.high is a "soft" limit that isn't supposed to cause OOM
situations, we should still try harder to meet a user request through
persistent reclaim.

For example, after setting a 10M memory.high on an 800M cgroup full of
file cache, the usage shrinks to about 350M:

  + cat /cgroup/workingset/memory.current
  841568256
  + echo 10M
  + cat /cgroup/workingset/memory.current
  355729408

This isn't exactly what the user would expect to happen. Setting the
value a few more times eventually whittles the usage down to what we
are asking for:

  + echo 10M
  + cat /cgroup/workingset/memory.current
  104181760
  + echo 10M
  + cat /cgroup/workingset/memory.current
  31801344
  + echo 10M
  + cat /cgroup/workingset/memory.current
  10440704

To improve this, add reclaim retry loops to the memory.high write()
callback, similar to what we do for memory.max, to make a reasonable
effort that the usage meets the requested size after the call returns.

Afterwards, a single write() to memory.high is enough in all but extreme
cases:

  + cat /cgroup/workingset/memory.current
  841609216
  + echo 10M
  + cat /cgroup/workingset/memory.current
  10182656

790M is not a reasonable reclaim target to ask of a single reclaim
invocation.  And it wouldn't be reasonable to optimize the reclaim code
for it.  So asking for the full size but retrying is not a bad choice
here: we express our intent, and benefit if reclaim becomes better at
handling larger requests, but we also acknowledge that some of the
deltas we can encounter in memory_high_write() are just too ridiculously
big for a single reclaim invocation to manage.

Link: http://lkml.kernel.org/r/20191022201518.341216-2-hannes@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

When the reclaim loop in memory_max_write() is ^C'd or similar, we set err
to -EINTR.  But we don't return err.  Once the limit is set, we always
return success (nbytes).  Delete the dead code.

Link: http://lkml.kernel.org/r/20191022201518.341216-1-hannes@cmpxchg.orgSigned-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The mem_cgroup_reclaim_cookie is only used in memcg softlimit reclaim now,
and the priority of the reclaim is always 0.  We don't need to define the
iter in struct mem_cgroup_per_node as an array any more.  That could make
the code more clear and save some space.

Link: http://lkml.kernel.org/r/1569897728-1686-1-git-send-email-laoar.shao@gmail.comSigned-off-by: NYafang Shao <laoar.shao@gmail.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We've encountered a rcu stall in get_mem_cgroup_from_mm():

  rcu: INFO: rcu_sched self-detected stall on CPU
  rcu: 33-....: (21000 ticks this GP) idle=6c6/1/0x4000000000000002 softirq=35441/35441 fqs=5017
  (t=21031 jiffies g=324821 q=95837) NMI backtrace for cpu 33
  <...>
  RIP: 0010:get_mem_cgroup_from_mm+0x2f/0x90
  <...>
   __memcg_kmem_charge+0x55/0x140
   __alloc_pages_nodemask+0x267/0x320
   pipe_write+0x1ad/0x400
   new_sync_write+0x127/0x1c0
   __kernel_write+0x4f/0xf0
   dump_emit+0x91/0xc0
   writenote+0xa0/0xc0
   elf_core_dump+0x11af/0x1430
   do_coredump+0xc65/0xee0
   get_signal+0x132/0x7c0
   do_signal+0x36/0x640
   exit_to_usermode_loop+0x61/0xd0
   do_syscall_64+0xd4/0x100
   entry_SYSCALL_64_after_hwframe+0x44/0xa9

The problem is caused by an exiting task which is associated with an
offline memcg.  We're iterating over and over in the do {} while
(!css_tryget_online()) loop, but obviously the memcg won't become online
and the exiting task won't be migrated to a live memcg.

Let's fix it by switching from css_tryget_online() to css_tryget().

As css_tryget_online() cannot guarantee that the memcg won't go offline,
the check is usually useless, except some rare cases when for example it
determines if something should be presented to a user.

A similar problem is described by commit 18fa84a2 ("cgroup: Use
css_tryget() instead of css_tryget_online() in task_get_css()").

Johannes:

: The bug aside, it doesn't matter whether the cgroup is online for the
: callers.  It used to matter when offlining needed to evacuate all charges
: from the memcg, and so needed to prevent new ones from showing up, but we
: don't care now.

Link: http://lkml.kernel.org/r/20191106225131.3543616-1-guro@fb.comSigned-off-by: NRoman Gushchin <guro@fb.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NTejun Heo <tj@kernel.org>
Reviewed-by: NShakeel Butt <shakeeb@google.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Michal Koutn <mkoutny@suse.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

While upgrading from 4.16 to 5.2, we noticed these allocation errors in
the log of the new kernel:

  SLUB: Unable to allocate memory on node -1, gfp=0xa20(GFP_ATOMIC)
    cache: tw_sock_TCPv6(960:helper-logs), object size: 232, buffer size: 240, default order: 1, min order: 0
    node 0: slabs: 5, objs: 170, free: 0

        slab_out_of_memory+1
        ___slab_alloc+969
        __slab_alloc+14
        kmem_cache_alloc+346
        inet_twsk_alloc+60
        tcp_time_wait+46
        tcp_fin+206
        tcp_data_queue+2034
        tcp_rcv_state_process+784
        tcp_v6_do_rcv+405
        __release_sock+118
        tcp_close+385
        inet_release+46
        __sock_release+55
        sock_close+17
        __fput+170
        task_work_run+127
        exit_to_usermode_loop+191
        do_syscall_64+212
        entry_SYSCALL_64_after_hwframe+68

accompanied by an increase in machines going completely radio silent
under memory pressure.

One thing that changed since 4.16 is e699e2c6 ("net, mm: account
sock objects to kmemcg"), which made these slab caches subject to cgroup
memory accounting and control.

The problem with that is that cgroups, unlike the page allocator, do not
maintain dedicated atomic reserves.  As a cgroup's usage hovers at its
limit, atomic allocations - such as done during network rx - can fail
consistently for extended periods of time.  The kernel is not able to
operate under these conditions.

We don't want to revert the culprit patch, because it indeed tracks a
potentially substantial amount of memory used by a cgroup.

We also don't want to implement dedicated atomic reserves for cgroups.
There is no point in keeping a fixed margin of unused bytes in the
cgroup's memory budget to accomodate a consumer that is impossible to
predict - we'd be wasting memory and get into configuration headaches,
not unlike what we have going with min_free_kbytes.  We do this for
physical mem because we have to, but cgroups are an accounting game.

Instead, account these privileged allocations to the cgroup, but let
them bypass the configured limit if they have to.  This way, we get the
benefits of accounting the consumed memory and have it exert pressure on
the rest of the cgroup, but like with the page allocator, we shift the
burden of reclaimining on behalf of atomic allocations onto the regular
allocations that can block.

Link: http://lkml.kernel.org/r/20191022233708.365764-1-hannes@cmpxchg.org
Fixes: e699e2c6 ("net, mm: account sock objects to kmemcg")
Signed-off-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NShakeel Butt <shakeelb@google.com>
Cc: Suleiman Souhlal <suleiman@google.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: <stable@vger.kernel.org>	[4.18+]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

page_cgroup_ino() doesn't return a valid memcg pointer for non-compound
slab pages, because it depends on PgHead AND PgSlab flags to be set to
determine the memory cgroup from the kmem_cache.  It's correct for
compound pages, but not for generic small pages.  Those don't have PgHead
set, so it ends up returning zero.

Fix this by replacing the condition to PageSlab() && !PageTail().

Before this patch:
  [root@localhost ~]# ./page-types -c /sys/fs/cgroup/user.slice/user-0.slice/user@0.service/ | grep slab
  0x0000000000000080	        38        0  _______S___________________________________	slab

After this patch:
  [root@localhost ~]# ./page-types -c /sys/fs/cgroup/user.slice/user-0.slice/user@0.service/ | grep slab
  0x0000000000000080	       147        0  _______S___________________________________	slab

Also, hwpoison_filter_task() uses output of page_cgroup_ino() in order
to filter error injection events based on memcg.  So if
page_cgroup_ino() fails to return memcg pointer, we just fail to inject
memory error.  Considering that hwpoison filter is for testing, affected
users are limited and the impact should be marginal.

[n-horiguchi@ah.jp.nec.com: changelog additions]
Link: http://lkml.kernel.org/r/20191031012151.2722280-1-guro@fb.com
Fixes: 4d96ba35 ("mm: memcg/slab: stop setting page->mem_cgroup pointer for slab pages")
Signed-off-by: NRoman Gushchin <guro@fb.com>
Reviewed-by: NShakeel Butt <shakeelb@google.com>
Acked-by: NDavid Rientjes <rientjes@google.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Daniel Jordan <daniel.m.jordan@oracle.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

__mem_cgroup_free() can be called on the failure path in
mem_cgroup_alloc().  However memcg_flush_percpu_vmstats() and
memcg_flush_percpu_vmevents() which are called from __mem_cgroup_free()
access the fields of memcg which can potentially be null if called from
failure path from mem_cgroup_alloc().  Indeed syzbot has reported the
following crash:

	kasan: CONFIG_KASAN_INLINE enabled
	kasan: GPF could be caused by NULL-ptr deref or user memory access
	general protection fault: 0000 [#1] PREEMPT SMP KASAN
	CPU: 0 PID: 30393 Comm: syz-executor.1 Not tainted 5.4.0-rc2+ #0
	Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
	RIP: 0010:memcg_flush_percpu_vmstats+0x4ae/0x930 mm/memcontrol.c:3436
	Code: 05 41 89 c0 41 0f b6 04 24 41 38 c7 7c 08 84 c0 0f 85 5d 03 00 00 44 3b 05 33 d5 12 08 0f 83 e2 00 00 00 4c 89 f0 48 c1 e8 03 <42> 80 3c 28 00 0f 85 91 03 00 00 48 8b 85 10 fe ff ff 48 8b b0 90
	RSP: 0018:ffff888095c27980 EFLAGS: 00010206
	RAX: 0000000000000012 RBX: ffff888095c27b28 RCX: ffffc90008192000
	RDX: 0000000000040000 RSI: ffffffff8340fae7 RDI: 0000000000000007
	RBP: ffff888095c27be0 R08: 0000000000000000 R09: ffffed1013f0da33
	R10: ffffed1013f0da32 R11: ffff88809f86d197 R12: fffffbfff138b760
	R13: dffffc0000000000 R14: 0000000000000090 R15: 0000000000000007
	FS:  00007f5027170700(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000
	CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
	CR2: 0000000000710158 CR3: 00000000a7b18000 CR4: 00000000001406f0
	DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
	DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
	Call Trace:
	__mem_cgroup_free+0x1a/0x190 mm/memcontrol.c:5021
	mem_cgroup_free mm/memcontrol.c:5033 [inline]
	mem_cgroup_css_alloc+0x3a1/0x1ae0 mm/memcontrol.c:5160
	css_create kernel/cgroup/cgroup.c:5156 [inline]
	cgroup_apply_control_enable+0x44d/0xc40 kernel/cgroup/cgroup.c:3119
	cgroup_mkdir+0x899/0x11b0 kernel/cgroup/cgroup.c:5401
	kernfs_iop_mkdir+0x14d/0x1d0 fs/kernfs/dir.c:1124
	vfs_mkdir+0x42e/0x670 fs/namei.c:3807
	do_mkdirat+0x234/0x2a0 fs/namei.c:3830
	__do_sys_mkdir fs/namei.c:3846 [inline]
	__se_sys_mkdir fs/namei.c:3844 [inline]
	__x64_sys_mkdir+0x5c/0x80 fs/namei.c:3844
	do_syscall_64+0xfa/0x760 arch/x86/entry/common.c:290
	entry_SYSCALL_64_after_hwframe+0x49/0xbe

Fixing this by moving the flush to mem_cgroup_free as there is no need
to flush anything if we see failure in mem_cgroup_alloc().

Link: http://lkml.kernel.org/r/20191018165231.249872-1-shakeelb@google.com
Fixes: bb65f89b ("mm: memcontrol: flush percpu vmevents before releasing memcg")
Fixes: c350a99e ("mm: memcontrol: flush percpu vmstats before releasing memcg")
Signed-off-by: NShakeel Butt <shakeelb@google.com>
Reported-by: syzbot+515d5bcfe179cdf049b2@syzkaller.appspotmail.com
Reviewed-by: NRoman Gushchin <guro@fb.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Mapped, dirty and writeback pages are also counted in per-lruvec stats.
These counters needs update when page is moved between cgroups.

Currently is nobody *consuming* the lruvec versions of these counters and
that there is no user-visible effect.

Link: http://lkml.kernel.org/r/157112699975.7360.1062614888388489788.stgit@buzz
Fixes: 00f3ca2c ("mm: memcontrol: per-lruvec stats infrastructure")
Signed-off-by: NKonstantin Khlebnikov <khlebnikov@yandex-team.ru>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Since the following commit:

  b4adfe8e ("locking/lockdep: Remove unused argument in __lock_release")

@nested is no longer used in lock_release(), so remove it from all
lock_release() calls and friends.
Signed-off-by: NQian Cai <cai@lca.pw>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: NWill Deacon <will@kernel.org>
Acked-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: airlied@linux.ie
Cc: akpm@linux-foundation.org
Cc: alexander.levin@microsoft.com
Cc: daniel@iogearbox.net
Cc: davem@davemloft.net
Cc: dri-devel@lists.freedesktop.org
Cc: duyuyang@gmail.com
Cc: gregkh@linuxfoundation.org
Cc: hannes@cmpxchg.org
Cc: intel-gfx@lists.freedesktop.org
Cc: jack@suse.com
Cc: jlbec@evilplan.or
Cc: joonas.lahtinen@linux.intel.com
Cc: joseph.qi@linux.alibaba.com
Cc: jslaby@suse.com
Cc: juri.lelli@redhat.com
Cc: maarten.lankhorst@linux.intel.com
Cc: mark@fasheh.com
Cc: mhocko@kernel.org
Cc: mripard@kernel.org
Cc: ocfs2-devel@oss.oracle.com
Cc: rodrigo.vivi@intel.com
Cc: sean@poorly.run
Cc: st@kernel.org
Cc: tj@kernel.org
Cc: tytso@mit.edu
Cc: vdavydov.dev@gmail.com
Cc: vincent.guittot@linaro.org
Cc: viro@zeniv.linux.org.uk
Link: https://lkml.kernel.org/r/1568909380-32199-1-git-send-email-cai@lca.pwSigned-off-by: NIngo Molnar <mingo@kernel.org>

cgroup v2 introduces two memory protection thresholds: memory.low
(best-effort) and memory.min (hard protection).  While they generally do
what they say on the tin, there is a limitation in their implementation
that makes them difficult to use effectively: that cliff behaviour often
manifests when they become eligible for reclaim.  This patch implements
more intuitive and usable behaviour, where we gradually mount more
reclaim pressure as cgroups further and further exceed their protection
thresholds.

This cliff edge behaviour happens because we only choose whether or not
to reclaim based on whether the memcg is within its protection limits
(see the use of mem_cgroup_protected in shrink_node), but we don't vary
our reclaim behaviour based on this information.  Imagine the following
timeline, with the numbers the lruvec size in this zone:

1. memory.low=1000000, memory.current=999999. 0 pages may be scanned.
2. memory.low=1000000, memory.current=1000000. 0 pages may be scanned.
3. memory.low=1000000, memory.current=1000001. 1000001* pages may be
   scanned. (?!)

* Of course, we won't usually scan all available pages in the zone even
  without this patch because of scan control priority, over-reclaim
  protection, etc.  However, as shown by the tests at the end, these
  techniques don't sufficiently throttle such an extreme change in input,
  so cliff-like behaviour isn't really averted by their existence alone.

Here's an example of how this plays out in practice.  At Facebook, we are
trying to protect various workloads from "system" software, like
configuration management tools, metric collectors, etc (see this[0] case
study).  In order to find a suitable memory.low value, we start by
determining the expected memory range within which the workload will be
comfortable operating.  This isn't an exact science -- memory usage deemed
"comfortable" will vary over time due to user behaviour, differences in
composition of work, etc, etc.  As such we need to ballpark memory.low,
but doing this is currently problematic:

1. If we end up setting it too low for the workload, it won't have
   *any* effect (see discussion above).  The group will receive the full
   weight of reclaim and won't have any priority while competing with the
   less important system software, as if we had no memory.low configured
   at all.

2. Because of this behaviour, we end up erring on the side of setting
   it too high, such that the comfort range is reliably covered.  However,
   protected memory is completely unavailable to the rest of the system,
   so we might cause undue memory and IO pressure there when we *know* we
   have some elasticity in the workload.

3. Even if we get the value totally right, smack in the middle of the
   comfort zone, we get extreme jumps between no pressure and full
   pressure that cause unpredictable pressure spikes in the workload due
   to the current binary reclaim behaviour.

With this patch, we can set it to our ballpark estimation without too much
worry.  Any undesirable behaviour, such as too much or too little reclaim
pressure on the workload or system will be proportional to how far our
estimation is off.  This means we can set memory.low much more
conservatively and thus waste less resources *without* the risk of the
workload falling off a cliff if we overshoot.

As a more abstract technical description, this unintuitive behaviour
results in having to give high-priority workloads a large protection
buffer on top of their expected usage to function reliably, as otherwise
we have abrupt periods of dramatically increased memory pressure which
hamper performance.  Having to set these thresholds so high wastes
resources and generally works against the principle of work conservation.
In addition, having proportional memory reclaim behaviour has other
benefits.  Most notably, before this patch it's basically mandatory to set
memory.low to a higher than desirable value because otherwise as soon as
you exceed memory.low, all protection is lost, and all pages are eligible
to scan again.  By contrast, having a gradual ramp in reclaim pressure
means that you now still get some protection when thresholds are exceeded,
which means that one can now be more comfortable setting memory.low to
lower values without worrying that all protection will be lost.  This is
important because workingset size is really hard to know exactly,
especially with variable workloads, so at least getting *some* protection
if your workingset size grows larger than you expect increases user
confidence in setting memory.low without a huge buffer on top being
needed.

Thanks a lot to Johannes Weiner and Tejun Heo for their advice and
assistance in thinking about how to make this work better.

In testing these changes, I intended to verify that:

1. Changes in page scanning become gradual and proportional instead of
   binary.

   To test this, I experimented stepping further and further down
   memory.low protection on a workload that floats around 19G workingset
   when under memory.low protection, watching page scan rates for the
   workload cgroup:

   +------------+-----------------+--------------------+--------------+
   | memory.low | test (pgscan/s) | control (pgscan/s) | % of control |
   +------------+-----------------+--------------------+--------------+
   |        21G |               0 |                  0 | N/A          |
   |        17G |             867 |               3799 | 23%          |
   |        12G |            1203 |               3543 | 34%          |
   |         8G |            2534 |               3979 | 64%          |
   |         4G |            3980 |               4147 | 96%          |
   |          0 |            3799 |               3980 | 95%          |
   +------------+-----------------+--------------------+--------------+

   As you can see, the test kernel (with a kernel containing this
   patch) ramps up page scanning significantly more gradually than the
   control kernel (without this patch).

2. More gradual ramp up in reclaim aggression doesn't result in
   premature OOMs.

   To test this, I wrote a script that slowly increments the number of
   pages held by stress(1)'s --vm-keep mode until a production system
   entered severe overall memory contention.  This script runs in a highly
   protected slice taking up the majority of available system memory.
   Watching vmstat revealed that page scanning continued essentially
   nominally between test and control, without causing forward reclaim
   progress to become arrested.

[0]: https://facebookmicrosites.github.io/cgroup2/docs/overview.html#case-study-the-fbtax2-project

[akpm@linux-foundation.org: reflow block comments to fit in 80 cols]
[chris@chrisdown.name: handle cgroup_disable=memory when getting memcg protection]
  Link: http://lkml.kernel.org/r/20190201045711.GA18302@chrisdown.name
Link: http://lkml.kernel.org/r/20190124014455.GA6396@chrisdown.nameSigned-off-by: NChris Down <chris@chrisdown.name>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Reviewed-by: NRoman Gushchin <guro@fb.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Thomas has noticed the following NULL ptr dereference when using cgroup
v1 kmem limit:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
PGD 0
P4D 0
Oops: 0000 [#1] PREEMPT SMP PTI
CPU: 3 PID: 16923 Comm: gtk-update-icon Not tainted 4.19.51 #42
Hardware name: Gigabyte Technology Co., Ltd. Z97X-Gaming G1/Z97X-Gaming G1, BIOS F9 07/31/2015
RIP: 0010:create_empty_buffers+0x24/0x100
Code: cd 0f 1f 44 00 00 0f 1f 44 00 00 41 54 49 89 d4 ba 01 00 00 00 55 53 48 89 fb e8 97 fe ff ff 48 89 c5 48 89 c2 eb 03 48 89 ca <48> 8b 4a 08 4c 09 22 48 85 c9 75 f1 48 89 6a 08 48 8b 43 18 48 8d
RSP: 0018:ffff927ac1b37bf8 EFLAGS: 00010286
RAX: 0000000000000000 RBX: fffff2d4429fd740 RCX: 0000000100097149
RDX: 0000000000000000 RSI: 0000000000000082 RDI: ffff9075a99fbe00
RBP: 0000000000000000 R08: fffff2d440949cc8 R09: 00000000000960c0
R10: 0000000000000002 R11: 0000000000000000 R12: 0000000000000000
R13: ffff907601f18360 R14: 0000000000002000 R15: 0000000000001000
FS:  00007fb55b288bc0(0000) GS:ffff90761f8c0000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000008 CR3: 000000007aebc002 CR4: 00000000001606e0
Call Trace:
 create_page_buffers+0x4d/0x60
 __block_write_begin_int+0x8e/0x5a0
 ? ext4_inode_attach_jinode.part.82+0xb0/0xb0
 ? jbd2__journal_start+0xd7/0x1f0
 ext4_da_write_begin+0x112/0x3d0
 generic_perform_write+0xf1/0x1b0
 ? file_update_time+0x70/0x140
 __generic_file_write_iter+0x141/0x1a0
 ext4_file_write_iter+0xef/0x3b0
 __vfs_write+0x17e/0x1e0
 vfs_write+0xa5/0x1a0
 ksys_write+0x57/0xd0
 do_syscall_64+0x55/0x160
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

Tetsuo then noticed that this is because the __memcg_kmem_charge_memcg
fails __GFP_NOFAIL charge when the kmem limit is reached.  This is a wrong
behavior because nofail allocations are not allowed to fail.  Normal
charge path simply forces the charge even if that means to cross the
limit.  Kmem accounting should be doing the same.

Link: http://lkml.kernel.org/r/20190906125608.32129-1-mhocko@kernel.orgSigned-off-by: NMichal Hocko <mhocko@suse.com>
Reported-by: NThomas Lindroth <thomas.lindroth@gmail.com>
Debugged-by: NTetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Thomas Lindroth <thomas.lindroth@gmail.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently THP deferred split shrinker is not memcg aware, this may cause
premature OOM with some configuration.  For example the below test would
run into premature OOM easily:

$ cgcreate -g memory:thp
$ echo 4G > /sys/fs/cgroup/memory/thp/memory/limit_in_bytes
$ cgexec -g memory:thp transhuge-stress 4000

transhuge-stress comes from kernel selftest.

It is easy to hit OOM, but there are still a lot THP on the deferred split
queue, memcg direct reclaim can't touch them since the deferred split
shrinker is not memcg aware.

Convert deferred split shrinker memcg aware by introducing per memcg
deferred split queue.  The THP should be on either per node or per memcg
deferred split queue if it belongs to a memcg.  When the page is
immigrated to the other memcg, it will be immigrated to the target memcg's
deferred split queue too.

Reuse the second tail page's deferred_list for per memcg list since the
same THP can't be on multiple deferred split queues.

[yang.shi@linux.alibaba.com: simplify deferred split queue dereference per Kirill Tkhai]
  Link: http://lkml.kernel.org/r/1566496227-84952-5-git-send-email-yang.shi@linux.alibaba.com
Link: http://lkml.kernel.org/r/1565144277-36240-5-git-send-email-yang.shi@linux.alibaba.comSigned-off-by: NYang Shi <yang.shi@linux.alibaba.com>
Acked-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: NKirill Tkhai <ktkhai@virtuozzo.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Qian Cai <cai@lca.pw>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently shrinker is just allocated and can work when memcg kmem is
enabled.  But, THP deferred split shrinker is not slab shrinker, it
doesn't make too much sense to have such shrinker depend on memcg kmem.
It should be able to reclaim THP even though memcg kmem is disabled.

Introduce a new shrinker flag, SHRINKER_NONSLAB, for non-slab shrinker.
When memcg kmem is disabled, just such shrinkers can be called in
shrinking memcg slab.

[yang.shi@linux.alibaba.com: add comment]
  Link: http://lkml.kernel.org/r/1566496227-84952-4-git-send-email-yang.shi@linux.alibaba.com
Link: http://lkml.kernel.org/r/1565144277-36240-4-git-send-email-yang.shi@linux.alibaba.comSigned-off-by: NYang Shi <yang.shi@linux.alibaba.com>
Acked-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: NKirill Tkhai <ktkhai@virtuozzo.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Qian Cai <cai@lca.pw>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Cgroup v1 memcg controller has exposed a dedicated kmem limit to users
which turned out to be really a bad idea because there are paths which
cannot shrink the kernel memory usage enough to get below the limit (e.g.
because the accounted memory is not reclaimable).  There are cases when
the failure is even not allowed (e.g.  __GFP_NOFAIL).  This means that the
kmem limit is in excess to the hard limit without any way to shrink and
thus completely useless.  OOM killer cannot be invoked to handle the
situation because that would lead to a premature oom killing.

As a result many places might see ENOMEM returning from kmalloc and result
in unexpected errors.  E.g.  a global OOM killer when there is a lot of
free memory because ENOMEM is translated into VM_FAULT_OOM in #PF path and
therefore pagefault_out_of_memory would result in OOM killer.

Please note that the kernel memory is still accounted to the overall limit
along with the user memory so removing the kmem specific limit should
still allow to contain kernel memory consumption.  Unlike the kmem one,
though, it invokes memory reclaim and targeted memcg oom killing if
necessary.

Start the deprecation process by crying to the kernel log.  Let's see
whether there are relevant usecases and simply return to EINVAL in the
second stage if nobody complains in few releases.

[akpm@linux-foundation.org: tweak documentation text]
Link: http://lkml.kernel.org/r/20190911151612.GI4023@dhcp22.suse.czSigned-off-by: NMichal Hocko <mhocko@suse.com>
Reviewed-by: NShakeel Butt <shakeelb@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Thomas Lindroth <thomas.lindroth@gmail.com>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

mem_cgroup_id_get() was introduced in commit 73f576c0 ("mm:memcontrol:
fix cgroup creation failure after many small jobs").

Later, it no longer has any user since the commits,

1f47b61f ("mm: memcontrol: fix swap counter leak on swapout from offline cgroup")
58fa2a55 ("mm: memcontrol: add sanity checks for memcg->id.ref on get/put")

so safe to remove it.

Link: http://lkml.kernel.org/r/1568648453-5482-1-git-send-email-cai@lca.pwSigned-off-by: NQian Cai <cai@lca.pw>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 72f0184c ("mm, memcg: remove hotplug locking from try_charge")
introduced css_tryget()/css_put() calls in drain_all_stock(), which are
supposed to protect the target memory cgroup from being released during
the mem_cgroup_is_descendant() call.

However, it's not completely safe.  In theory, memcg can go away between
reading stock->cached pointer and calling css_tryget().

This can happen if drain_all_stock() races with drain_local_stock()
performed on the remote cpu as a result of a work, scheduled by the
previous invocation of drain_all_stock().

The race is a bit theoretical and there are few chances to trigger it, but
the current code looks a bit confusing, so it makes sense to fix it
anyway.  The code looks like as if css_tryget() and css_put() are used to
protect stocks drainage.  It's not necessary because stocked pages are
holding references to the cached cgroup.  And it obviously won't work for
works, scheduled on other cpus.

So, let's read the stock->cached pointer and evaluate the memory cgroup
inside a rcu read section, and get rid of css_tryget()/css_put() calls.

Link: http://lkml.kernel.org/r/20190802192241.3253165-1-guro@fb.comSigned-off-by: NRoman Gushchin <guro@fb.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Hillf Danton <hdanton@sina.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We're trying to use memory.high to limit workloads, but have found that
containment can frequently fail completely and cause OOM situations
outside of the cgroup.  This happens especially with swap space -- either
when none is configured, or swap is full.  These failures often also don't
have enough warning to allow one to react, whether for a human or for a
daemon monitoring PSI.

Here is output from a simple program showing how long it takes in usec
(column 2) to allocate a megabyte of anonymous memory (column 1) when a
cgroup is already beyond its memory high setting, and no swap is
available:

    [root@ktst ~]# systemd-run -p MemoryHigh=100M -p MemorySwapMax=1 \
    > --wait -t timeout 300 /root/mdf
    [...]
    95  1035
    96  1038
    97  1000
    98  1036
    99  1048
    100 1590
    101 1968
    102 1776
    103 1863
    104 1757
    105 1921
    106 1893
    107 1760
    108 1748
    109 1843
    110 1716
    111 1924
    112 1776
    113 1831
    114 1766
    115 1836
    116 1588
    117 1912
    118 1802
    119 1857
    120 1731
    [...]
    [System OOM in 2-3 seconds]

The delay does go up extremely marginally past the 100MB memory.high
threshold, as now we spend time scanning before returning to usermode, but
it's nowhere near enough to contain growth.  It also doesn't get worse the
more pages you have, since it only considers nr_pages.

The current situation goes against both the expectations of users of
memory.high, and our intentions as cgroup v2 developers.  In
cgroup-v2.txt, we claim that we will throttle and only under "extreme
conditions" will memory.high protection be breached.  Likewise, cgroup v2
users generally also expect that memory.high should throttle workloads as
they exceed their high threshold.  However, as seen above, this isn't
always how it works in practice -- even on banal setups like those with no
swap, or where swap has become exhausted, we can end up with memory.high
being breached and us having no weapons left in our arsenal to combat
runaway growth with, since reclaim is futile.

It's also hard for system monitoring software or users to tell how bad the
situation is, as "high" events for the memcg may in some cases be benign,
and in others be catastrophic.  The current status quo is that we fail
containment in a way that doesn't provide any advance warning that things
are about to go horribly wrong (for example, we are about to invoke the
kernel OOM killer).

This patch introduces explicit throttling when reclaim is failing to keep
memcg size contained at the memory.high setting.  It does so by applying
an exponential delay curve derived from the memcg's overage compared to
memory.high.  In the normal case where the memcg is either below or only
marginally over its memory.high setting, no throttling will be performed.

This composes well with system health monitoring and remediation, as these
allocator delays are factored into PSI's memory pressure calculations.
This both creates a mechanism system administrators or applications
consuming the PSI interface to trivially see that the memcg in question is
struggling and use that to make more reasonable decisions, and permits
them enough time to act.  Either of these can act with significantly more
nuance than that we can provide using the system OOM killer.

This is a similar idea to memory.oom_control in cgroup v1 which would put
the cgroup to sleep if the threshold was violated, but it's also
significantly improved as it results in visible memory pressure, and also
doesn't schedule indefinitely, which previously made tracing and other
introspection difficult (ie.  it's clamped at 2*HZ per allocation through
MEMCG_MAX_HIGH_DELAY_JIFFIES).

Contrast the previous results with a kernel with this patch:

    [root@ktst ~]# systemd-run -p MemoryHigh=100M -p MemorySwapMax=1 \
    > --wait -t timeout 300 /root/mdf
    [...]
    95  1002
    96  1000
    97  1002
    98  1003
    99  1000
    100 1043
    101 84724
    102 330628
    103 610511
    104 1016265
    105 1503969
    106 2391692
    107 2872061
    108 3248003
    109 4791904
    110 5759832
    111 6912509
    112 8127818
    113 9472203
    114 12287622
    115 12480079
    116 14144008
    117 15808029
    118 16384500
    119 16383242
    120 16384979
    [...]

As you can see, in the normal case, memory allocation takes around 1000
usec.  However, as we exceed our memory.high, things start to increase
exponentially, but fairly leniently at first.  Our first megabyte over
memory.high takes us 0.16 seconds, then the next is 0.46 seconds, then the
next is almost an entire second.  This gets worse until we reach our
eventual 2*HZ clamp per batch, resulting in 16 seconds per megabyte.
However, this is still making forward progress, so permits tracing or
further analysis with programs like GDB.

We use an exponential curve for our delay penalty for a few reasons:

1. We run mem_cgroup_handle_over_high to potentially do reclaim after
   we've already performed allocations, which means that temporarily
   going over memory.high by a small amount may be perfectly legitimate,
   even for compliant workloads. We don't want to unduly penalise such
   cases.
2. An exponential curve (as opposed to a static or linear delay) allows
   ramping up memory pressure stats more gradually, which can be useful
   to work out that you have set memory.high too low, without destroying
   application performance entirely.

This patch expands on earlier work by Johannes Weiner. Thanks!

[akpm@linux-foundation.org: fix max() warning]
[akpm@linux-foundation.org: fix __udivdi3 ref on 32-bit]
[akpm@linux-foundation.org: fix it even more]
[chris@chrisdown.name: fix 64-bit divide even more]
Link: http://lkml.kernel.org/r/20190723180700.GA29459@chrisdown.nameSigned-off-by: NChris Down <chris@chrisdown.name>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Roman Gushchin <guro@fb.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Replace 1 << compound_order(page) with compound_nr(page).  Minor
improvements in readability.

Link: http://lkml.kernel.org/r/20190721104612.19120-4-willy@infradead.orgSigned-off-by: NMatthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: NAndrew Morton <akpm@linux-foundation.org>
Reviewed-by: NIra Weiny <ira.weiny@intel.com>
Acked-by: NKirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The mm_walk structure currently mixed data and code.  Split out the
operations vectors into a new mm_walk_ops structure, and while we are
changing the API also declare the mm_walk structure inside the
walk_page_range and walk_page_vma functions.

Based on patch from Linus Torvalds.

Link: https://lore.kernel.org/r/20190828141955.22210-3-hch@lst.deSigned-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NThomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: NSteven Price <steven.price@arm.com>
Reviewed-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Add a new header for the two handful of users of the walk_page_range /
walk_page_vma interface instead of polluting all users of mm.h with it.

Link: https://lore.kernel.org/r/20190828141955.22210-2-hch@lst.deSigned-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NThomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: NSteven Price <steven.price@arm.com>
Reviewed-by: NJason Gunthorpe <jgg@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

Instead of using raw_cpu_read() use per_cpu() to read the actual data of
the corresponding cpu otherwise we will be reading the data of the
current cpu for the number of online CPUs.

Link: http://lkml.kernel.org/r/20190829203110.129263-1-shakeelb@google.com
Fixes: bb65f89b ("mm: memcontrol: flush percpu vmevents before releasing memcg")
Fixes: c350a99e ("mm: memcontrol: flush percpu vmstats before releasing memcg")
Signed-off-by: NShakeel Butt <shakeelb@google.com>
Acked-by: NRoman Gushchin <guro@fb.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

mm, memcg: partially revert "mm/memcontrol.c: keep local VM counters in sync with the hierarchical ones"

Commit 766a4c19 ("mm/memcontrol.c: keep local VM counters in sync
with the hierarchical ones") effectively decreased the precision of
per-memcg vmstats_local and per-memcg-per-node lruvec percpu counters.

That's good for displaying in memory.stat, but brings a serious
regression into the reclaim process.

One issue I've discovered and debugged is the following:
lruvec_lru_size() can return 0 instead of the actual number of pages in
the lru list, preventing the kernel to reclaim last remaining pages.
Result is yet another dying memory cgroups flooding.  The opposite is
also happening: scanning an empty lru list is the waste of cpu time.

Also, inactive_list_is_low() can return incorrect values, preventing the
active lru from being scanned and freed.  It can fail both because the
size of active and inactive lists are inaccurate, and because the number
of workingset refaults isn't precise.  In other words, the result is
pretty random.

I'm not sure, if using the approximate number of slab pages in
count_shadow_number() is acceptable, but issues described above are
enough to partially revert the patch.

Let's keep per-memcg vmstat_local batched (they are only used for
displaying stats to the userspace), but keep lruvec stats precise.  This
change fixes the dead memcg flooding on my setup.

Link: http://lkml.kernel.org/r/20190817004726.2530670-1-guro@fb.com
Fixes: 766a4c19 ("mm/memcontrol.c: keep local VM counters in sync with the hierarchical ones")
Signed-off-by: NRoman Gushchin <guro@fb.com>
Acked-by: NYafang Shao <laoar.shao@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

I've noticed that the "slab" value in memory.stat is sometimes 0, even
if some children memory cgroups have a non-zero "slab" value.  The
following investigation showed that this is the result of the kmem_cache
reparenting in combination with the per-cpu batching of slab vmstats.

At the offlining some vmstat value may leave in the percpu cache, not
being propagated upwards by the cgroup hierarchy.  It means that stats
on ancestor levels are lower than actual.  Later when slab pages are
released, the precise number of pages is substracted on the parent
level, making the value negative.  We don't show negative values, 0 is
printed instead.

To fix this issue, let's flush percpu slab memcg and lruvec stats on
memcg offlining.  This guarantees that numbers on all ancestor levels
are accurate and match the actual number of outstanding slab pages.

Link: http://lkml.kernel.org/r/20190819202338.363363-3-guro@fb.com
Fixes: fb2f2b0a ("mm: memcg/slab: reparent memcg kmem_caches on cgroup removal")
Signed-off-by: NRoman Gushchin <guro@fb.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

cgroup foreign inode handling has quite a bit of heuristics and
internal states which sometimes makes it difficult to understand
what's going on.  Add tracepoints to improve visibility.
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

There's an inherent mismatch between memcg and writeback.  The former
trackes ownership per-page while the latter per-inode.  This was a
deliberate design decision because honoring per-page ownership in the
writeback path is complicated, may lead to higher CPU and IO overheads
and deemed unnecessary given that write-sharing an inode across
different cgroups isn't a common use-case.

Combined with inode majority-writer ownership switching, this works
well enough in most cases but there are some pathological cases.  For
example, let's say there are two cgroups A and B which keep writing to
different but confined parts of the same inode.  B owns the inode and
A's memory is limited far below B's.  A's dirty ratio can rise enough
to trigger balance_dirty_pages() sleeps but B's can be low enough to
avoid triggering background writeback.  A will be slowed down without
a way to make writeback of the dirty pages happen.

This patch implements foreign dirty recording and foreign mechanism so
that when a memcg encounters a condition as above it can trigger
flushes on bdi_writebacks which can clean its pages.  Please see the
comment on top of mem_cgroup_track_foreign_dirty_slowpath() for
details.

A reproducer follows.

write-range.c::

  #include <stdio.h>
  #include <stdlib.h>
  #include <unistd.h>
  #include <fcntl.h>
  #include <sys/types.h>

  static const char *usage = "write-range FILE START SIZE\n";

  int main(int argc, char **argv)
  {
	  int fd;
	  unsigned long start, size, end, pos;
	  char *endp;
	  char buf[4096];

	  if (argc < 4) {
		  fprintf(stderr, usage);
		  return 1;
	  }

	  fd = open(argv[1], O_WRONLY);
	  if (fd < 0) {
		  perror("open");
		  return 1;
	  }

	  start = strtoul(argv[2], &endp, 0);
	  if (*endp != '\0') {
		  fprintf(stderr, usage);
		  return 1;
	  }

	  size = strtoul(argv[3], &endp, 0);
	  if (*endp != '\0') {
		  fprintf(stderr, usage);
		  return 1;
	  }

	  end = start + size;

	  while (1) {
		  for (pos = start; pos < end; ) {
			  long bread, bwritten = 0;

			  if (lseek(fd, pos, SEEK_SET) < 0) {
				  perror("lseek");
				  return 1;
			  }

			  bread = read(0, buf, sizeof(buf) < end - pos ?
					       sizeof(buf) : end - pos);
			  if (bread < 0) {
				  perror("read");
				  return 1;
			  }
			  if (bread == 0)
				  return 0;

			  while (bwritten < bread) {
				  long this;

				  this = write(fd, buf + bwritten,
					       bread - bwritten);
				  if (this < 0) {
					  perror("write");
					  return 1;
				  }

				  bwritten += this;
				  pos += bwritten;
			  }
		  }
	  }
  }

repro.sh::

  #!/bin/bash

  set -e
  set -x

  sysctl -w vm.dirty_expire_centisecs=300000
  sysctl -w vm.dirty_writeback_centisecs=300000
  sysctl -w vm.dirtytime_expire_seconds=300000
  echo 3 > /proc/sys/vm/drop_caches

  TEST=/sys/fs/cgroup/test
  A=$TEST/A
  B=$TEST/B

  mkdir -p $A $B
  echo "+memory +io" > $TEST/cgroup.subtree_control
  echo $((1<<30)) > $A/memory.high
  echo $((32<<30)) > $B/memory.high

  rm -f testfile
  touch testfile
  fallocate -l 4G testfile

  echo "Starting B"

  (echo $BASHPID > $B/cgroup.procs
   pv -q --rate-limit 70M < /dev/urandom | ./write-range testfile $((2<<30)) $((2<<30))) &

  echo "Waiting 10s to ensure B claims the testfile inode"
  sleep 5
  sync
  sleep 5
  sync
  echo "Starting A"

  (echo $BASHPID > $A/cgroup.procs
   pv < /dev/urandom | ./write-range testfile 0 $((2<<30)))

v2: Added comments explaining why the specific intervals are being used.

v3: Use 0 @nr when calling cgroup_writeback_by_id() to use best-effort
    flushing while avoding possible livelocks.

v4: Use get_jiffies_64() and time_before/after64() instead of raw
    jiffies_64 and arthimetic comparisons as suggested by Jan.
Reviewed-by: NJan Kara <jack@suse.cz>
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <axboe@kernel.dk>

Similar to vmstats, percpu caching of local vmevents leads to an
accumulation of errors on non-leaf levels.  This happens because some
leftovers may remain in percpu caches, so that they are never propagated
up by the cgroup tree and just disappear into nonexistence with on
releasing of the memory cgroup.

To fix this issue let's accumulate and propagate percpu vmevents values
before releasing the memory cgroup similar to what we're doing with
vmstats.

Since on cpu hotplug we do flush percpu vmstats anyway, we can iterate
only over online cpus.

Link: http://lkml.kernel.org/r/20190819202338.363363-4-guro@fb.com
Fixes: 42a30035 ("mm: memcontrol: fix recursive statistics correctness & scalabilty")
Signed-off-by: NRoman Gushchin <guro@fb.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Percpu caching of local vmstats with the conditional propagation by the
cgroup tree leads to an accumulation of errors on non-leaf levels.

Let's imagine two nested memory cgroups A and A/B.  Say, a process
belonging to A/B allocates 100 pagecache pages on the CPU 0.  The percpu
cache will spill 3 times, so that 32*3=96 pages will be accounted to A/B
and A atomic vmstat counters, 4 pages will remain in the percpu cache.

Imagine A/B is nearby memory.max, so that every following allocation
triggers a direct reclaim on the local CPU.  Say, each such attempt will
free 16 pages on a new cpu.  That means every percpu cache will have -16
pages, except the first one, which will have 4 - 16 = -12.  A/B and A
atomic counters will not be touched at all.

Now a user removes A/B.  All percpu caches are freed and corresponding
vmstat numbers are forgotten.  A has 96 pages more than expected.

As memory cgroups are created and destroyed, errors do accumulate.  Even
1-2 pages differences can accumulate into large numbers.

To fix this issue let's accumulate and propagate percpu vmstat values
before releasing the memory cgroup.  At this point these numbers are
stable and cannot be changed.

Since on cpu hotplug we do flush percpu vmstats anyway, we can iterate
only over online cpus.

Link: http://lkml.kernel.org/r/20190819202338.363363-2-guro@fb.com
Fixes: 42a30035 ("mm: memcontrol: fix recursive statistics correctness & scalabilty")
Signed-off-by: NRoman Gushchin <guro@fb.com>
Acked-by: NMichal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Memcg counters for shadow nodes are broken because the memcg pointer is
obtained in a wrong way. The following approach is used:
        virt_to_page(xa_node)->mem_cgroup

Since commit 4d96ba35 ("mm: memcg/slab: stop setting
page->mem_cgroup pointer for slab pages") page->mem_cgroup pointer isn't
set for slab pages, so memcg_from_slab_page() should be used instead.

Also I doubt that it ever worked correctly: virt_to_head_page() should
be used instead of virt_to_page().  Otherwise objects residing on tail
pages are not accounted, because only the head page contains a valid
mem_cgroup pointer.  That was a case since the introduction of these
counters by the commit 68d48e6a ("mm: workingset: add vmstat counter
for shadow nodes").

Link: http://lkml.kernel.org/r/20190801233532.138743-1-guro@fb.com
Fixes: 4d96ba35 ("mm: memcg/slab: stop setting page->mem_cgroup pointer for slab pages")
Signed-off-by: NRoman Gushchin <guro@fb.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>