btrfs: don't propagate EXTENT_FLAG_LOGGING to split extent maps
PATCH 1/1 btrfs: don't propagate EXTENT_FLAG_LOGGING to split extent maps
When btrfs_drop_extent_map_range() splits an extent map, the new split maps inherit the original map's flags through a local 'flags' variable. Commit f86f7a75e2fb ("btrfs: use the flags of an extent map to identify the compression type") changed the EXTENT_FLAG_LOGGING clearing to operate on em->flags instead of that local 'flags' copy, so a split of an extent map that is currently being logged wrongly inherits EXTENT_FLAG_LOGGING.
The flag is then never cleared on the split, and when it is freed while still on the inode's modified_extents list (for example by the extent map shrinker) it trips the WARN_ON(!list_empty(&em->list)) in btrfs_free_extent_map() and leads to a use-after-free.
Clear EXTENT_FLAG_LOGGING from the local 'flags' copy used for the splits and only clear EXTENT_FLAG_PINNED from em->flags, restoring the behaviour prior to f86f7a75e2fb.
--- fs/btrfs/extent_map.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
fs/btrfs/extent_map.c +2 -2
@@ -866,13 +866,13 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, goto next; } - flags = em->flags;/* * In case we split the extent map, we want to preserve the * EXTENT_FLAG_LOGGING flag on our extent map, but we don't want * it on the new extent maps. */ - em->flags &= ~(EXTENT_FLAG_PINNED |EXTENT_FLAG_LOGGING ); + flags = em->flags & ~EXTENT_FLAG_LOGGING ; + em->flags &= ~EXTENT_FLAG_PINNED ; modified = !list_empty (&em->list);/*
Signature · 1 line
2.53.0-Meta
Quoted 35 lines · Leo Martins
When
btrfs_drop_extent_map_range()splits an extent map, the new split maps inherit the original map's flags through a local 'flags' variable. Commitf86f7a75e2fb("btrfs: use the flags of an extent map to identify the compression type") changed theEXTENT_FLAG_LOGGINGclearing to operate onem->flagsinstead of that local 'flags' copy, so a split of an extent map that is currently being logged wrongly inheritsEXTENT_FLAG_LOGGING.The flag is then never cleared on the split, and when it is freed while still on the inode's
modified_extentslist (for example by the extent map shrinker) it trips the WARN_ON(!list_empty(&em->list)) inbtrfs_free_extent_map()and leads to a use-after-free.Clear
EXTENT_FLAG_LOGGINGfrom the local 'flags' copy used for the splits and only clearEXTENT_FLAG_PINNEDfromem->flags, restoring the behaviour prior tof86f7a75e2fb.Fixes f86f7a75e2fb ("btrfs: use the flags of an extent map to identify the compression type")Cc Jeff Layton <[email protected]>Cc Boris Burkov <[email protected]>Signed-off-by Leo Martins <[email protected]>--- fs/btrfs/extent_map.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)fs/btrfs/extent_map.c +0
-1@@ -866,13 +866,13 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, goto next; } - flags = em->flags;/*
fs/btrfs/extent_map.c void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, · line 878
Quoted 10 lines
fs/btrfs/extent_map.c +2
-1* In case we split the extent map, we want to preserve the * EXTENT_FLAG_LOGGING flag on our extent map, but we don't want * it on the new extent maps. */ - em->flags &= ~(EXTENT_FLAG_PINNED |EXTENT_FLAG_LOGGING ); + flags = em->flags & ~EXTENT_FLAG_LOGGING ; + em->flags &= ~EXTENT_FLAG_PINNED ; modified = !list_empty (&em->list);/*
Quoted 14 lines · Leo Martins
When
btrfs_drop_extent_map_range()splits an extent map, the new split maps inherit the original map's flags through a local 'flags' variable. Commitf86f7a75e2fb("btrfs: use the flags of an extent map to identify the compression type") changed theEXTENT_FLAG_LOGGINGclearing to operate onem->flagsinstead of that local 'flags' copy, so a split of an extent map that is currently being logged wrongly inheritsEXTENT_FLAG_LOGGING.The flag is then never cleared on the split, and when it is freed while still on the inode's
modified_extentslist (for example by the extent map shrinker) it trips the WARN_ON(!list_empty(&em->list)) inbtrfs_free_extent_map()and leads to a use-after-free.
Quoted 8 lines
Clear
EXTENT_FLAG_LOGGINGfrom the local 'flags' copy used for the splits and only clearEXTENT_FLAG_PINNEDfromem->flags, restoring the behaviour prior tof86f7a75e2fb.Fixes f86f7a75e2fb ("btrfs: use the flags of an extent map to identify the compression type")Cc Jeff Layton <[email protected]>Cc Boris Burkov <[email protected]>Signed-off-by Leo Martins <[email protected]>
Adding it to for-next, thanks.
Quoted 28 lines
--- fs/btrfs/extent_map.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)fs/btrfs/extent_map.c +2
-3@@ -866,13 +866,13 @@ void btrfs_drop_extent_map_range(struct btrfs_inode *inode, u64 start, u64 end, goto next; } - flags = em->flags;/* * In case we split the extent map, we want to preserve the * EXTENT_FLAG_LOGGING flag on our extent map, but we don't want * it on the new extent maps. */ - em->flags &= ~(EXTENT_FLAG_PINNED |EXTENT_FLAG_LOGGING ); + flags = em->flags & ~EXTENT_FLAG_LOGGING ; + em->flags &= ~EXTENT_FLAG_PINNED ; modified = !list_empty (&em->list);/* -- 2.53.0-Meta