linux-stable.git/mm, branch v4.6.6

mm: memcontrol: fix cgroup creation failure after many small jobs

2016-08-10T10:54:43+00:00

commit 73f576c04b9410ed19660f74f97521bee6e1c546 upstream.

The memory controller has quite a bit of state that usually outlives the
cgroup and pins its CSS until said state disappears.  At the same time
it imposes a 16-bit limit on the CSS ID space to economically store IDs
in the wild.  Consequently, when we use cgroups to contain frequent but
small and short-lived jobs that leave behind some page cache, we quickly
run into the 64k limitations of outstanding CSSs.  Creating a new cgroup
fails with -ENOSPC while there are only a few, or even no user-visible
cgroups in existence.

Although pinning CSSs past cgroup removal is common, there are only two
instances that actually need an ID after a cgroup is deleted: cache
shadow entries and swapout records.

Cache shadow entries reference the ID weakly and can deal with the CSS
having disappeared when it's looked up later.  They pose no hurdle.

Swap-out records do need to pin the css to hierarchically attribute
swapins after the cgroup has been deleted; though the only pages that
remain swapped out after offlining are tmpfs/shmem pages.  And those
references are under the user's control, so they are manageable.

This patch introduces a private 16-bit memcg ID and switches swap and
cache shadow entries over to using that.  This ID can then be recycled
after offlining when the CSS remains pinned only by objects that don't
specifically need it.

This script demonstrates the problem by faulting one cache page in a new
cgroup and deleting it again:

  set -e
  mkdir -p pages
  for x in `seq 128000`; do
    [ $((x % 1000)) -eq 0 ] && echo $x
    mkdir /cgroup/foo
    echo $$ >/cgroup/foo/cgroup.procs
    echo trex >pages/$x
    echo $$ >/cgroup/cgroup.procs
    rmdir /cgroup/foo
  done

When run on an unpatched kernel, we eventually run out of possible IDs
even though there are no visible cgroups:

  [root@ham ~]# ./cssidstress.sh
  [...]
  65000
  mkdir: cannot create directory '/cgroup/foo': No space left on device

After this patch, the IDs get released upon cgroup destruction and the
cache and css objects get released once memory reclaim kicks in.

[hannes@cmpxchg.org: init the IDR]
  Link: http://lkml.kernel.org/r/20160621154601.GA22431@cmpxchg.org
Fixes: b2052564e66d ("mm: memcontrol: continue cache reclaim from offlined groups")
Link: http://lkml.kernel.org/r/20160617162516.GD19084@cmpxchg.org
Signed-off-by: Johannes Weiner 
Reported-by: John Garcia 
Reviewed-by: Vladimir Davydov 
Acked-by: Tejun Heo 
Cc: Nikolay Borisov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

mm: thp: refix false positive BUG in page_move_anon_rmap()

2016-08-10T10:54:43+00:00

commit 5a49973d7143ebbabd76e1dcd69ee42e349bb7b9 upstream.

The VM_BUG_ON_PAGE in page_move_anon_rmap() is more trouble than it's
worth: the syzkaller fuzzer hit it again.  It's still wrong for some THP
cases, because linear_page_index() was never intended to apply to
addresses before the start of a vma.

That's easily fixed with a signed long cast inside linear_page_index();
and Dmitry has tested such a patch, to verify the false positive.  But
why extend linear_page_index() just for this case? when the avoidance in
page_move_anon_rmap() has already grown ugly, and there's no reason for
the check at all (nothing else there is using address or index).

Remove address arg from page_move_anon_rmap(), remove VM_BUG_ON_PAGE,
remove CONFIG_DEBUG_VM PageTransHuge adjustment.

And one more thing: should the compound_head(page) be done inside or
outside page_move_anon_rmap()? It's usually pushed down to the lowest
level nowadays (and mm/memory.c shows no other explicit use of it), so I
think it's better done in page_move_anon_rmap() than by caller.

Fixes: 0798d3c022dc ("mm: thp: avoid false positive VM_BUG_ON_PAGE in page_move_anon_rmap()")
Link: http://lkml.kernel.org/r/alpine.LSU.2.11.1607120444540.12528@eggly.anvils
Signed-off-by: Hugh Dickins 
Reported-by: Dmitry Vyukov 
Acked-by: Kirill A. Shutemov 
Cc: Mika Westerberg 
Cc: Andrea Arcangeli 
Cc: Rik van Riel 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

mm, meminit: ensure node is online before checking whether pages are uninitialised

2016-08-10T10:54:43+00:00

commit ef70b6f41cda6270165a6f27b2548ed31cfa3cb2 upstream.

early_page_uninitialised looks up an arbitrary PFN.  While a machine
without node 0 will boot with "mm, page_alloc: Always return a valid
node from early_pfn_to_nid", it works because it assumes that nodes are
always in PFN order.  This is not guaranteed so this patch adds
robustness by always checking if the node being checked is online.

Link: http://lkml.kernel.org/r/1468008031-3848-4-git-send-email-mgorman@techsingularity.net
Signed-off-by: Mel Gorman 
Acked-by: David Rientjes 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

mm, meminit: always return a valid node from early_pfn_to_nid

2016-08-10T10:54:43+00:00

commit e4568d3803852d00effd41dcdd489e726b998879 upstream.

early_pfn_to_nid can return node 0 if a PFN is invalid on machines that
has no node 0.  A machine with only node 1 was observed to crash with
the following message:

   BUG: unable to handle kernel paging request at 000000000002a3c8
   PGD 0
   Modules linked in:
   Hardware name: Supermicro H8DSP-8/H8DSP-8, BIOS 080011  06/30/2006
   task: ffffffff81c0d500 ti: ffffffff81c00000 task.ti: ffffffff81c00000
   RIP: reserve_bootmem_region+0x6a/0xef
   CR2: 000000000002a3c8 CR3: 0000000001c06000 CR4: 00000000000006b0
   Call Trace:
      free_all_bootmem+0x4b/0x12a
      mem_init+0x70/0xa3
      start_kernel+0x25b/0x49b

The problem is that early_page_uninitialised uses the early_pfn_to_nid
helper which returns node 0 for invalid PFNs.  No caller of
early_pfn_to_nid cares except early_page_uninitialised.  This patch has
early_pfn_to_nid always return a valid node.

Link: http://lkml.kernel.org/r/1468008031-3848-3-git-send-email-mgorman@techsingularity.net
Signed-off-by: Mel Gorman 
Acked-by: David Rientjes 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

mm, compaction: prevent VM_BUG_ON when terminating freeing scanner

2016-08-10T10:54:43+00:00

commit a46cbf3bc53b6a93fb84a5ffb288c354fa807954 upstream.

It's possible to isolate some freepages in a pageblock and then fail
split_free_page() due to the low watermark check.  In this case, we hit
VM_BUG_ON() because the freeing scanner terminated early without a
contended lock or enough freepages.

This should never have been a VM_BUG_ON() since it's not a fatal
condition.  It should have been a VM_WARN_ON() at best, or even handled
gracefully.

Regardless, we need to terminate anytime the full pageblock scan was not
done.  The logic belongs in isolate_freepages_block(), so handle its
state gracefully by terminating the pageblock loop and making a note to
restart at the same pageblock next time since it was not possible to
complete the scan this time.

[rientjes@google.com: don't rescan pages in a pageblock]
  Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1607111244150.83138@chino.kir.corp.google.com
Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1606291436300.145590@chino.kir.corp.google.com
Signed-off-by: David Rientjes 
Reported-by: Minchan Kim 
Tested-by: Minchan Kim 
Cc: Joonsoo Kim 
Cc: Hugh Dickins 
Cc: Mel Gorman 
Cc: Vlastimil Babka 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

mm, compaction: abort free scanner if split fails

2016-08-10T10:54:42+00:00

commit a4f04f2c6955aff5e2c08dcb40aca247ff4d7370 upstream.

If the memory compaction free scanner cannot successfully split a free
page (only possible due to per-zone low watermark), terminate the free
scanner rather than continuing to scan memory needlessly.  If the
watermark is insufficient for a free page of order <= cc->order, then
terminate the scanner since all future splits will also likely fail.

This prevents the compaction freeing scanner from scanning all memory on
very large zones (very noticeable for zones > 128GB, for instance) when
all splits will likely fail while holding zone->lock.

compaction_alloc() iterating a 128GB zone has been benchmarked to take
over 400ms on some systems whereas any free page isolated and ready to
be split ends up failing in split_free_page() because of the low
watermark check and thus the iteration continues.

The next time compaction occurs, the freeing scanner will likely start
at the end of the zone again since no success was made previously and we
get the same lengthy iteration until the zone is brought above the low
watermark.  All thp page faults can take >400ms in such a state without
this fix.

Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1606211820350.97086@chino.kir.corp.google.com
Signed-off-by: David Rientjes 
Acked-by: Vlastimil Babka 
Cc: Minchan Kim 
Cc: Joonsoo Kim 
Cc: Mel Gorman 
Cc: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

mm/swap.c: flush lru pvecs on compound page arrival

2016-08-10T10:54:42+00:00

commit 8f182270dfec432e93fae14f9208a6b9af01009f upstream.

Currently we can have compound pages held on per cpu pagevecs, which
leads to a lot of memory unavailable for reclaim when needed.  In the
systems with hundreads of processors it can be GBs of memory.

On of the way of reproducing the problem is to not call munmap
explicitly on all mapped regions (i.e.  after receiving SIGTERM).  After
that some pages (with THP enabled also huge pages) may end up on
lru_add_pvec, example below.

  void main() {
  #pragma omp parallel
  {
	size_t size = 55 * 1000 * 1000; // smaller than  MEM/CPUS
	void *p = mmap(NULL, size, PROT_READ | PROT_WRITE,
		MAP_PRIVATE | MAP_ANONYMOUS , -1, 0);
	if (p != MAP_FAILED)
		memset(p, 0, size);
	//munmap(p, size); // uncomment to make the problem go away
  }
  }

When we run it with THP enabled it will leave significant amount of
memory on lru_add_pvec.  This memory will be not reclaimed if we hit
OOM, so when we run above program in a loop:

	for i in `seq 100`; do ./a.out; done

many processes (95% in my case) will be killed by OOM.

The primary point of the LRU add cache is to save the zone lru_lock
contention with a hope that more pages will belong to the same zone and
so their addition can be batched.  The huge page is already a form of
batched addition (it will add 512 worth of memory in one go) so skipping
the batching seems like a safer option when compared to a potential
excess in the caching which can be quite large and much harder to fix
because lru_add_drain_all is way to expensive and it is not really clear
what would be a good moment to call it.

Similarly we can reproduce the problem on lru_deactivate_pvec by adding:
madvise(p, size, MADV_FREE); after memset.

This patch flushes lru pvecs on compound page arrival making the problem
less severe - after applying it kill rate of above example drops to 0%,
due to reducing maximum amount of memory held on pvec from 28MB (with
THP) to 56kB per CPU.

Suggested-by: Michal Hocko 
Link: http://lkml.kernel.org/r/1466180198-18854-1-git-send-email-lukasz.odzioba@intel.com
Signed-off-by: Lukasz Odzioba 
Acked-by: Michal Hocko 
Cc: Kirill Shutemov 
Cc: Andrea Arcangeli 
Cc: Vladimir Davydov 
Cc: Ming Li 
Cc: Minchan Kim 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

memcg: css_alloc should return an ERR_PTR value on error

2016-08-10T10:54:42+00:00

commit ea3a9645866e12d2b198434f03df3c3e96fb86ce upstream.

mem_cgroup_css_alloc() was returning NULL on failure while cgroup core
expected it to return an ERR_PTR value leading to the following NULL
deref after a css allocation failure.  Fix it by return
ERR_PTR(-ENOMEM) instead.  I'll also update cgroup core so that it
can handle NULL returns.

  mkdir: page allocation failure: order:6, mode:0x240c0c0(GFP_KERNEL|__GFP_COMP|__GFP_ZERO)
  CPU: 0 PID: 8738 Comm: mkdir Not tainted 4.7.0-rc3+ #123
  ...
  Call Trace:
    dump_stack+0x68/0xa1
    warn_alloc_failed+0xd6/0x130
    __alloc_pages_nodemask+0x4c6/0xf20
    alloc_pages_current+0x66/0xe0
    alloc_kmem_pages+0x14/0x80
    kmalloc_order_trace+0x2a/0x1a0
    __kmalloc+0x291/0x310
    memcg_update_all_caches+0x6c/0x130
    mem_cgroup_css_alloc+0x590/0x610
    cgroup_apply_control_enable+0x18b/0x370
    cgroup_mkdir+0x1de/0x2e0
    kernfs_iop_mkdir+0x55/0x80
    vfs_mkdir+0xb9/0x150
    SyS_mkdir+0x66/0xd0
    do_syscall_64+0x53/0x120
    entry_SYSCALL64_slow_path+0x25/0x25
  ...
  BUG: unable to handle kernel NULL pointer dereference at 00000000000000d0
  IP:  init_and_link_css+0x37/0x220
  PGD 34b1e067 PUD 3a109067 PMD 0
  Oops: 0002 [#1] SMP
  Modules linked in:
  CPU: 0 PID: 8738 Comm: mkdir Not tainted 4.7.0-rc3+ #123
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.9.2-20160422_131301-anatol 04/01/2014
  task: ffff88007cbc5200 ti: ffff8800666d4000 task.ti: ffff8800666d4000
  RIP: 0010:[]  [] init_and_link_css+0x37/0x220
  RSP: 0018:ffff8800666d7d90  EFLAGS: 00010246
  RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
  RDX: ffffffff810f2499 RSI: 0000000000000000 RDI: 0000000000000008
  RBP: ffff8800666d7db8 R08: 0000000000000003 R09: 0000000000000000
  R10: 0000000000000001 R11: 0000000000000000 R12: ffff88005a5fb400
  R13: ffffffff81f0f8a0 R14: ffff88005a5fb400 R15: 0000000000000010
  FS:  00007fc944689700(0000) GS:ffff88007fc00000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 00007f3aed0d2b80 CR3: 000000003a1e8000 CR4: 00000000000006f0
  DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
  DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
  Call Trace:
    cgroup_apply_control_enable+0x1ac/0x370
    cgroup_mkdir+0x1de/0x2e0
    kernfs_iop_mkdir+0x55/0x80
    vfs_mkdir+0xb9/0x150
    SyS_mkdir+0x66/0xd0
    do_syscall_64+0x53/0x120
    entry_SYSCALL64_slow_path+0x25/0x25
  Code: 89 f5 48 89 fb 49 89 d4 48 83 ec 08 8b 05 72 3b d8 00 85 c0 0f 85 60 01 00 00 4c 89 e7 e8 72 f7 ff ff 48 8d 7b 08 48 89 d9 31 c0 <48> c7 83 d0 00 00 00 00 00 00 00 48 83 e7 f8 48 29 f9 81 c1 d8
  RIP   init_and_link_css+0x37/0x220
   RSP 
  CR2: 00000000000000d0
  ---[ end trace a2d8836ae1e852d1 ]---

Link: http://lkml.kernel.org/r/20160621165740.GJ3262@mtj.duckdns.org
Signed-off-by: Tejun Heo 
Reported-by: Johannes Weiner 
Reviewed-by: Vladimir Davydov 
Acked-by: Johannes Weiner 
Acked-by: Michal Hocko 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

memcg: mem_cgroup_migrate() may be called with irq disabled

2016-08-10T10:54:42+00:00

commit d93c4130a7d049b234b5d5a15808eaf5406f2789 upstream.

mem_cgroup_migrate() uses local_irq_disable/enable() but can be called
with irq disabled from migrate_page_copy().  This ends up enabling irq
while holding a irq context lock triggering the following lockdep
warning.  Fix it by using irq_save/restore instead.

  =================================
  [ INFO: inconsistent lock state ]
  4.7.0-rc1+ #52 Tainted: G        W
  ---------------------------------
  inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage.
  kcompactd0/151 [HC0[0]:SC0[0]:HE1:SE1] takes:
   (&(&ctx->completion_lock)->rlock){+.?.-.}, at: [<000000000038fd96>] aio_migratepage+0x156/0x1e8
  {IN-SOFTIRQ-W} state was registered at:
     __lock_acquire+0x5b6/0x1930
     lock_acquire+0xee/0x270
     _raw_spin_lock_irqsave+0x66/0xb0
     aio_complete+0x98/0x328
     dio_complete+0xe4/0x1e0
     blk_update_request+0xd4/0x450
     scsi_end_request+0x48/0x1c8
     scsi_io_completion+0x272/0x698
     blk_done_softirq+0xca/0xe8
     __do_softirq+0xc8/0x518
     irq_exit+0xee/0x110
     do_IRQ+0x6a/0x88
     io_int_handler+0x11a/0x25c
     __mutex_unlock_slowpath+0x144/0x1d8
     __mutex_unlock_slowpath+0x140/0x1d8
     kernfs_iop_permission+0x64/0x80
     __inode_permission+0x9e/0xf0
     link_path_walk+0x6e/0x510
     path_lookupat+0xc4/0x1a8
     filename_lookup+0x9c/0x160
     user_path_at_empty+0x5c/0x70
     SyS_readlinkat+0x68/0x140
     system_call+0xd6/0x270
  irq event stamp: 971410
  hardirqs last  enabled at (971409):  migrate_page_move_mapping+0x3ea/0x588
  hardirqs last disabled at (971410):  _raw_spin_lock_irqsave+0x3c/0xb0
  softirqs last  enabled at (970526):  __do_softirq+0x460/0x518
  softirqs last disabled at (970519):  irq_exit+0xee/0x110

  other info that might help us debug this:
   Possible unsafe locking scenario:

	 CPU0
	 ----
    lock(&(&ctx->completion_lock)->rlock);
    
      lock(&(&ctx->completion_lock)->rlock);

    *** DEADLOCK ***

  3 locks held by kcompactd0/151:
   #0:  (&(&mapping->private_lock)->rlock){+.+.-.}, at:  aio_migratepage+0x42/0x1e8
   #1:  (&ctx->ring_lock){+.+.+.}, at:  aio_migratepage+0x5a/0x1e8
   #2:  (&(&ctx->completion_lock)->rlock){+.?.-.}, at:  aio_migratepage+0x156/0x1e8

  stack backtrace:
  CPU: 20 PID: 151 Comm: kcompactd0 Tainted: G        W       4.7.0-rc1+ #52
  Call Trace:
    show_trace+0xea/0xf0
    show_stack+0x72/0xf0
    dump_stack+0x9a/0xd8
    print_usage_bug.part.27+0x2d4/0x2e8
    mark_lock+0x17e/0x758
    mark_held_locks+0xa2/0xd0
    trace_hardirqs_on_caller+0x140/0x1c0
    mem_cgroup_migrate+0x266/0x370
    aio_migratepage+0x16a/0x1e8
    move_to_new_page+0xb0/0x260
    migrate_pages+0x8f4/0x9f0
    compact_zone+0x4dc/0xdc8
    kcompactd_do_work+0x1aa/0x358
    kcompactd+0xba/0x2c8
    kthread+0x10a/0x110
    kernel_thread_starter+0x6/0xc
    kernel_thread_starter+0x0/0xc
  INFO: lockdep is turned off.

Link: http://lkml.kernel.org/r/20160620184158.GO3262@mtj.duckdns.org
Link: http://lkml.kernel.org/g/5767CFE5.7080904@de.ibm.com
Fixes: 74485cf2bc85 ("mm: migrate: consolidate mem_cgroup_migrate() calls")
Signed-off-by: Tejun Heo 
Reported-by: Christian Borntraeger 
Acked-by: Johannes Weiner 
Acked-by: Michal Hocko 
Reviewed-by: Vladimir Davydov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

mm, sl[au]b: add __GFP_ATOMIC to the GFP reclaim mask

2016-08-10T10:54:42+00:00

commit e838a45f9392a5bd2be1cd3ab0b16ae85857461c upstream.

Commit d0164adc89f6 ("mm, page_alloc: distinguish between being unable
to sleep, unwilling to sleep and avoiding waking kswapd") modified
__GFP_WAIT to explicitly identify the difference between atomic callers
and those that were unwilling to sleep.  Later the definition was
removed entirely.

The GFP_RECLAIM_MASK is the set of flags that affect watermark checking
and reclaim behaviour but __GFP_ATOMIC was never added.  Without it,
atomic users of the slab allocator strip the __GFP_ATOMIC flag and
cannot access the page allocator atomic reserves.  This patch addresses
the problem.

The user-visible impact depends on the workload but potentially atomic
allocations unnecessarily fail without this path.

Link: http://lkml.kernel.org/r/20160610093832.GK2527@techsingularity.net
Signed-off-by: Mel Gorman 
Reported-by: Marcin Wojtas 
Acked-by: Vlastimil Babka 
Acked-by: Michal Hocko 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman