linux/mm
Hugh Dickins b1dea800ac tmpfs: fix race between umount and writepage
Konstanin Khlebnikov reports that a dangerous race between umount and
shmem_writepage can be reproduced by this script:

  for i in {1..300} ; do
	mkdir $i
	while true ; do
		mount -t tmpfs none $i
		dd if=/dev/zero of=$i/test bs=1M count=$(($RANDOM % 100))
		umount $i
	done &
  done

on a 6xCPU node with 8Gb RAM: kernel very unstable after this accident. =)

Kernel log:

  VFS: Busy inodes after unmount of tmpfs.
                 Self-destruct in 5 seconds.  Have a nice day...

  WARNING: at lib/list_debug.c:53 __list_del_entry+0x8d/0x98()
  list_del corruption. prev->next should be ffff880222fdaac8, but was (null)
  Pid: 11222, comm: mount.tmpfs Not tainted 2.6.39-rc2+ #4
  Call Trace:
   warn_slowpath_common+0x80/0x98
   warn_slowpath_fmt+0x41/0x43
   __list_del_entry+0x8d/0x98
   evict+0x50/0x113
   iput+0x138/0x141
  ...
  BUG: unable to handle kernel paging request at ffffffffffffffff
  IP: shmem_free_blocks+0x18/0x4c
  Pid: 10422, comm: dd Tainted: G        W   2.6.39-rc2+ #4
  Call Trace:
   shmem_recalc_inode+0x61/0x66
   shmem_writepage+0xba/0x1dc
   pageout+0x13c/0x24c
   shrink_page_list+0x28e/0x4be
   shrink_inactive_list+0x21f/0x382
  ...

shmem_writepage() calls igrab() on the inode for the page which came from
page reclaim, to add it later into shmem_swaplist for swapoff operation.

This igrab() can race with super-block deactivating process:

  shrink_inactive_list()          deactivate_super()
  pageout()                       tmpfs_fs_type->kill_sb()
  shmem_writepage()               kill_litter_super()
                                  generic_shutdown_super()
                                   evict_inodes()
   igrab()
                                    atomic_read(&inode->i_count)
                                     skip-inode
   iput()
                                   if (!list_empty(&sb->s_inodes))
                                          printk("VFS: Busy inodes after...

This igrap-iput pair was added in commit 1b1b32f2c6 "tmpfs: fix
shmem_swaplist races" based on incorrect assumptions: igrab() protects the
inode from concurrent eviction by deletion, but it does nothing to protect
it from concurrent unmounting, which goes ahead despite the raised
i_count.

So this use of igrab() was wrong all along, but the race made much worse
in 2.6.37 when commit 63997e98a3 "split invalidate_inodes()" replaced
two attempts at invalidate_inodes() by a single evict_inodes().

Konstantin posted a plausible patch, raising sb->s_active too: I'm unsure
whether it was correct or not; but burnt once by igrab(), I am sure that
we don't want to rely more deeply upon externals here.

Fix it by adding the inode to shmem_swaplist earlier, while the page lock
on page in page cache still secures the inode against eviction, without
artifically raising i_count.  It was originally added later because
shmem_unuse_inode() is liable to remove an inode from the list while it's
unswapped; but we can guard against that by taking spinlock before
dropping mutex.

Reported-by: Konstantin Khlebnikov <khlebnikov@openvz.org>
Signed-off-by: Hugh Dickins <hughd@google.com>
Tested-by: Konstantin Khlebnikov <khlebnikov@openvz.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-05-11 18:50:45 -07:00
..
backing-dev.c Fix common misspellings 2011-03-31 11:26:23 -03:00
bootmem.c crash_dump: export is_kdump_kernel to modules, consolidate elfcorehdr_addr, setup_elfcorehdr and saved_max_pfn 2011-03-23 19:47:19 -07:00
bounce.c
compaction.c mm: compaction: minimise the time IRQs are disabled while isolating pages for migration 2011-03-22 17:44:05 -07:00
debug-pagealloc.c
dmapool.c mm/dmapool.c: use TASK_UNINTERRUPTIBLE in dma_pool_alloc() 2011-01-13 17:32:48 -08:00
fadvise.c
failslab.c
filemap_xip.c
filemap.c Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6 2011-03-24 19:01:30 -07:00
fremap.c
highmem.c
huge_memory.c mm: thp: fix /dev/zero MAP_PRIVATE and vm_flags cleanups 2011-04-28 11:28:20 -07:00
hugetlb.c Fix common misspellings 2011-03-31 11:26:23 -03:00
hwpoison-inject.c Fix common misspellings 2011-03-31 11:26:23 -03:00
init-mm.c
internal.h Fix common misspellings 2011-03-31 11:26:23 -03:00
Kconfig mm: compaction: don't depend on HUGETLB_PAGE 2011-01-26 10:50:02 +10:00
Kconfig.debug mm: debug-pagealloc: fix kconfig dependency warning 2011-03-22 17:44:02 -07:00
kmemcheck.c
kmemleak-test.c kmemleak: remove memset by using kzalloc 2011-01-27 18:31:51 +00:00
kmemleak.c Fix common misspellings 2011-03-31 11:26:23 -03:00
ksm.c Fix common misspellings 2011-03-31 11:26:23 -03:00
maccess.c
madvise.c thp: khugepaged: make khugepaged aware about madvise 2011-01-13 17:32:47 -08:00
Makefile bootmem: Separate out CONFIG_NO_BOOTMEM code into nobootmem.c 2011-02-24 14:43:05 +01:00
memblock.c mm/memblock: properly handle overlaps and fix error path 2011-03-22 17:44:09 -07:00
memcontrol.c Fix common misspellings 2011-03-31 11:26:23 -03:00
memory_hotplug.c mm: optimize pfn calculation in online_page() 2011-04-14 16:06:54 -07:00
memory-failure.c Fix common misspellings 2011-03-31 11:26:23 -03:00
memory.c Don't lock guardpage if the stack is growing up 2011-05-09 16:22:07 -07:00
mempolicy.c mempolicy: remove redundant check in __mpol_equal() 2011-03-22 17:44:04 -07:00
mempool.c
migrate.c Fix common misspellings 2011-03-31 11:26:23 -03:00
mincore.c thp: mincore transparent hugepage support 2011-01-13 17:32:44 -08:00
mlock.c VM: skip the stack guard page lookup in get_user_pages only for mlock 2011-05-04 21:30:28 -07:00
mm_init.c
mmap.c vm: fix vm_pgoff wrap in upward expansion 2011-05-09 17:52:17 -07:00
mmu_context.c
mmu_notifier.c thp: mmu_notifier_test_young 2011-01-13 17:32:46 -08:00
mmzone.c mm: page allocator: adjust the per-cpu counter threshold when memory is low 2011-01-13 17:32:31 -08:00
mprotect.c thp: mprotect: transparent huge page support 2011-01-13 17:32:44 -08:00
mremap.c mm: avoid wrapping vm_pgoff in mremap() 2011-04-07 07:35:51 -07:00
msync.c
nobootmem.c Fix common misspellings 2011-03-31 11:26:23 -03:00
nommu.c NOMMU: implement access_remote_vm 2011-03-29 14:05:12 +01:00
oom_kill.c oom: use pte pages in OOM score 2011-04-28 11:28:21 -07:00
page_alloc.c mm: add alloc_pages_exact_nid() 2011-05-11 18:50:45 -07:00
page_cgroup.c memcg: allocate memory cgroup structures in local nodes 2011-05-11 18:50:45 -07:00
page_io.c block: kill off REQ_UNPLUG 2011-03-10 08:52:27 +01:00
page_isolation.c
page-writeback.c Merge branch 'for-2.6.39/core' of git://git.kernel.dk/linux-2.6-block 2011-03-24 10:16:26 -07:00
pagewalk.c pagewalk: only split huge pages when necessary 2011-03-22 17:44:04 -07:00
percpu-km.c
percpu-vm.c mm: remove gfp mask from pcpu_get_vm_areas 2011-01-13 17:32:34 -08:00
percpu.c Fix common misspellings 2011-03-31 11:26:23 -03:00
pgtable-generic.c mm/pgtable-generic.c: fix CONFIG_SWAP=n build 2011-01-26 10:49:58 +10:00
prio_tree.c
quicklist.c
readahead.c read-ahead: use plugging 2011-03-10 08:52:26 +01:00
rmap.c fs: move i_wb_list out from under inode_lock 2011-03-24 21:17:51 -04:00
shmem.c tmpfs: fix race between umount and writepage 2011-05-11 18:50:45 -07:00
slab.c Fix common misspellings 2011-03-31 11:26:23 -03:00
slob.c mm: Remove support for kmem_cache_name() 2011-01-23 21:00:05 +02:00
slub.c slub: Fix the lockless code on 32-bit platforms with no 64-bit cmpxchg 2011-05-04 14:20:20 -07:00
sparse-vmemmap.c
sparse.c Fix common misspellings 2011-03-31 11:26:23 -03:00
swap_state.c block: remove per-queue plugging 2011-03-10 08:52:07 +01:00
swap.c mm: check PageUnevictable in lru_deactivate_fn() 2011-05-11 18:50:44 -07:00
swapfile.c Merge branch 'for-2.6.39/core' of git://git.kernel.dk/linux-2.6-block 2011-03-24 10:16:26 -07:00
thrash.c
truncate.c mm: deactivate invalidated pages 2011-03-22 17:44:03 -07:00
util.c Fix common misspellings 2011-03-31 11:26:23 -03:00
vmalloc.c vmalloc: remove confusing comment on vwrite() 2011-03-22 17:44:09 -07:00
vmscan.c vmscan: all_unreclaimable() use zone->all_unreclaimable as a name 2011-04-14 16:06:56 -07:00
vmstat.c mm: add VM counters for transparent hugepages 2011-04-14 16:06:55 -07:00