linux-toradex.git/mm/migrate.c, branch v3.0.37

mm: fix race between mremap and removing migration entry

2011-10-25T05:10:17+00:00

commit 486cf46f3f9be5f2a966016c1a8fe01e32cde09e upstream.

I don't usually pay much attention to the stale "? " addresses in
stack backtraces, but this lucky report from Pawel Sikora hints that
mremap's move_ptes() has inadequate locking against page migration.

 3.0 BUG_ON(!PageLocked(p)) in migration_entry_to_page():
 kernel BUG at include/linux/swapops.h:105!
 RIP: 0010:[]  []
                       migration_entry_wait+0x156/0x160
  [] handle_pte_fault+0xae1/0xaf0
  [] ? __pte_alloc+0x42/0x120
  [] ? do_huge_pmd_anonymous_page+0xab/0x310
  [] handle_mm_fault+0x181/0x310
  [] ? vma_adjust+0x537/0x570
  [] do_page_fault+0x11d/0x4e0
  [] ? do_mremap+0x2d5/0x570
  [] page_fault+0x1f/0x30

mremap's down_write of mmap_sem, together with i_mmap_mutex or lock,
and pagetable locks, were good enough before page migration (with its
requirement that every migration entry be found) came in, and enough
while migration always held mmap_sem; but not enough nowadays, when
there's memory hotremove and compaction.

The danger is that move_ptes() lets a migration entry dodge around
behind remove_migration_pte()'s back, so it's in the old location when
looking at the new, then in the new location when looking at the old.

Either mremap's move_ptes() must additionally take anon_vma lock(), or
migration's remove_migration_pte() must stop peeking for is_swap_entry()
before it takes pagetable lock.

Consensus chooses the latter: we prefer to add overhead to migration
than to mremapping, which gets used by JVMs and by exec stack setup.

Reported-and-tested-by: Paweł Sikora 
Signed-off-by: Hugh Dickins 
Acked-by: Andrea Arcangeli 
Acked-by: Mel Gorman 
Signed-off-by: Linus Torvalds 
Signed-off-by: Greg Kroah-Hartman

migrate: don't account swapcache as shmem

2011-06-16T22:01:24+00:00

swapcache will reach the below code path in migrate_page_move_mapping,
and swapcache is accounted as NR_FILE_PAGES but it's not accounted as
NR_SHMEM.

Hugh pointed out we must use PageSwapCache instead of comparing
mapping to &swapper_space, to avoid build failure with CONFIG_SWAP=n.

Signed-off-by: Andrea Arcangeli 
Acked-by: Hugh Dickins 
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds

mm: use refcounts for page_lock_anon_vma()

2011-05-25T15:39:19+00:00

Convert page_lock_anon_vma() over to use refcounts.  This is done to
prepare for the conversion of anon_vma from spinlock to mutex.

Sadly this inceases the cost of page_lock_anon_vma() from one to two
atomics, a follow up patch addresses this, lets keep that simple for now.

Signed-off-by: Peter Zijlstra 
Reviewed-by: KAMEZAWA Hiroyuki 
Reviewed-by: KOSAKI Motohiro 
Acked-by: Hugh Dickins 
Cc: Benjamin Herrenschmidt 
Cc: David Miller 
Cc: Martin Schwidefsky 
Cc: Russell King 
Cc: Paul Mundt 
Cc: Jeff Dike 
Cc: Richard Weinberger 
Cc: Tony Luck 
Cc: Mel Gorman 
Cc: Nick Piggin 
Cc: Namhyung Kim 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

Fix common misspellings

2011-03-31T14:26:23+00:00

Fixes generated by 'codespell' and manually reviewed.

Signed-off-by: Lucas De Marchi

memcg: fix ugly initialization of return value is in caller

2011-03-24T02:46:22+00:00

Remove initialization of vaiable in caller of memory cgroup function.
Actually, it's return value of memcg function but it's initialized in
caller.

Some memory cgroup uses following style to bring the result of start
function to the end function for avoiding races.

   mem_cgroup_start_A(&(*ptr))
   /* Something very complicated can happen here. */
   mem_cgroup_end_A(*ptr)

In some calls, *ptr should be initialized to NULL be caller.  But it's
ugly.  This patch fixes that *ptr is initialized by _start function.

Signed-off-by: KAMEZAWA Hiroyuki 
Acked-by: Johannes Weiner 
Acked-by: Daisuke Nishimura 
Cc: Balbir Singh 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: compaction: Use async migration for __GFP_NO_KSWAPD and enforce no writeback

2011-03-23T00:44:05+00:00

__GFP_NO_KSWAPD allocations are usually very expensive and not mandatory
to succeed as they have graceful fallback.  Waiting for I/O in those,
tends to be overkill in terms of latencies, so we can reduce their latency
by disabling sync migrate.

Unfortunately, even with async migration it's still possible for the
process to be blocked waiting for a request slot (e.g.  get_request_wait
in the block layer) when ->writepage is called.  To prevent
__GFP_NO_KSWAPD blocking, this patch prevents ->writepage being called on
dirty page cache for asynchronous migration.

Addresses https://bugzilla.kernel.org/show_bug.cgi?id=31142

[mel@csn.ul.ie: Avoid writebacks for NFS, retry locked pages, use bool]
Signed-off-by: Andrea Arcangeli 
Signed-off-by: Mel Gorman 
Cc: Arthur Marsh 
Cc: Clemens Ladisch 
Cc: Johannes Weiner 
Cc: KAMEZAWA Hiroyuki 
Cc: Minchan Kim 
Reported-by: Alex Villacis Lasso 
Tested-by: Alex Villacis Lasso 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: rename drop_anon_vma() to put_anon_vma()

2011-03-23T00:44:03+00:00

The normal code pattern used in the kernel is: get/put.

Signed-off-by: Peter Zijlstra 
Reviewed-by: KAMEZAWA Hiroyuki 
Acked-by: Hugh Dickins 
Reviewed-by: Rik van Riel 
Acked-by: Mel Gorman 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: add replace_page_cache_page() function

2011-03-23T00:44:02+00:00

This function basically does:

     remove_from_page_cache(old);
     page_cache_release(old);
     add_to_page_cache_locked(new);

Except it does this atomically, so there's no possibility for the "add" to
fail because of a race.

If memory cgroups are enabled, then the memory cgroup charge is also moved
from the old page to the new.

This function is currently used by fuse to move pages into the page cache
on read, instead of copying the page contents.

[minchan.kim@gmail.com: add freepage() hook to replace_page_cache_page()]
Signed-off-by: Miklos Szeredi 
Acked-by: Rik van Riel 
Acked-by: KAMEZAWA Hiroyuki 
Cc: Mel Gorman 
Signed-off-by: Minchan Kim 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm: grab rcu read lock in move_pages()

2011-02-25T23:07:36+00:00

The move_pages() usage of find_task_by_vpid() requires rcu_read_lock() to
prevent free_pid() from reclaiming the pid.

Without this patch, RCU warnings are printed in v2.6.38-rc4 move_pages()
with:

  CONFIG_LOCKUP_DETECTOR=y
  CONFIG_PREEMPT=y
  CONFIG_LOCKDEP=y
  CONFIG_PROVE_LOCKING=y
  CONFIG_PROVE_RCU=y

Previously, migrate_pages() went through a similar transformation
replacing usage of tasklist_lock with rcu read lock:

  commit 55cfaa3cbdd29c4919ecb5fb8965c310f357e48c
  Author: Zeng Zhaoming 
  Date:   Thu Dec 2 14:31:13 2010 -0800

      mm/mempolicy.c: add rcu read lock to protect pid structure

  commit 1e50df39f6e2c3a4a3394df62baa8a213df16c54
  Author: KOSAKI Motohiro 
  Date:   Thu Jan 13 15:46:14 2011 -0800

      mempolicy: remove tasklist_lock from migrate_pages

Signed-off-by: Greg Thelen 
Cc: Mel Gorman 
Cc: Minchan Kim 
Cc: Rik van Riel 
Cc: KAMEZAWA Hiroyuki 
Cc: "Paul E. McKenney" 
Cc: Tetsuo Handa 
Cc: Sergey Senozhatsky 
Cc: Oleg Nesterov 
Cc: Zeng Zhaoming 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm/migration: fix page corruption during hugepage migration

2011-02-03T00:03:18+00:00

If migrate_huge_page by memory-failure fails , it calls put_page in itself
to decrease page reference and caller of migrate_huge_page also calls
putback_lru_pages.  It can do double free of page so it can make page
corruption on page holder.

In addtion, clean of pages on caller is consistent behavior with
migrate_pages by cf608ac19c ("mm: compaction: fix COMPACTPAGEFAILED
counting").

Signed-off-by: Minchan Kim 
Cc: Andrea Arcangeli 
Cc: Christoph Lameter 
Cc: Mel Gorman 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds