Skip to content

Commit

Permalink
btrfs: mask out gfp flags in releasepage
Browse files Browse the repository at this point in the history
btree_releasepage is a callback and can be passed unknown gfp flags and then
they may end up in kmem_cache_alloc called from alloc_extent_state, slab
allocator will BUG_ON when there is HIGHMEM or DMA32 flag set.

This may happen when btrfs is mounted from a loop device, which masks out
__GFP_IO flag. The check in try_release_extent_state

3399                 if ((mask & GFP_NOFS) == GFP_NOFS)
3400                         mask = GFP_NOFS;

will not work and passes unfiltered flags further resulting in crash at
mm/slab.c:2963

 [<000000000024ae4c>] cache_alloc_refill+0x3b4/0x5c8
 [<000000000024c810>] kmem_cache_alloc+0x204/0x294
 [<00000000001fd3c2>] mempool_alloc+0x52/0x170
 [<000003c000ced0b0>] alloc_extent_state+0x40/0xd4 [btrfs]
 [<000003c000cee5ae>] __clear_extent_bit+0x38a/0x4cc [btrfs]
 [<000003c000cee78c>] try_release_extent_state+0x9c/0xd4 [btrfs]
 [<000003c000cc4c66>] btree_releasepage+0x7e/0xd0 [btrfs]
 [<0000000000210d84>] shrink_page_list+0x6a0/0x724
 [<0000000000211394>] shrink_inactive_list+0x230/0x578
 [<0000000000211bb8>] shrink_list+0x6c/0x120
 [<0000000000211e4e>] shrink_zone+0x1e2/0x228
 [<0000000000211f24>] shrink_zones+0x90/0x254
 [<0000000000213410>] do_try_to_free_pages+0xac/0x420
 [<0000000000213ae0>] try_to_free_pages+0x13c/0x1b0
 [<0000000000204e6c>] __alloc_pages_nodemask+0x5b4/0x9a8
 [<00000000001fb04a>] grab_cache_page_write_begin+0x7e/0xe8

Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
  • Loading branch information
David Sterba authored and Chris Mason committed Jan 26, 2012
1 parent 9e622d6 commit 0c4e538
Showing 1 changed file with 7 additions and 0 deletions.
7 changes: 7 additions & 0 deletions fs/btrfs/disk-io.c
Original file line number Diff line number Diff line change
Expand Up @@ -961,6 +961,13 @@ static int btree_releasepage(struct page *page, gfp_t gfp_flags)
tree = &BTRFS_I(page->mapping->host)->io_tree;
map = &BTRFS_I(page->mapping->host)->extent_tree;

/*
* We need to mask out eg. __GFP_HIGHMEM and __GFP_DMA32 as we're doing
* slab allocation from alloc_extent_state down the callchain where
* it'd hit a BUG_ON as those flags are not allowed.
*/
gfp_flags &= ~GFP_SLAB_BUG_MASK;

ret = try_release_extent_state(map, tree, page, gfp_flags);
if (!ret)
return 0;
Expand Down

0 comments on commit 0c4e538

Please sign in to comment.