4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jeff Mahoney jeffm@suse.com
commit 08db141b5313ac2f64b844fb5725b8d81744b417 upstream.
The main loop in __discard_prealloc is protected by the reiserfs write lock which is dropped across schedules like the BKL it replaced. The problem is that it checks the value, calls a routine that schedules, and then adjusts the state. As a result, two threads that are calling reiserfs_prealloc_discard at the same time can race when one calls reiserfs_free_prealloc_block, the lock is dropped, and the other calls reiserfs_free_prealloc_block with the same block number. In the right circumstances, it can cause the prealloc count to go negative.
Signed-off-by: Jeff Mahoney jeffm@suse.com Signed-off-by: Jan Kara jack@suse.cz Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
--- fs/reiserfs/bitmap.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-)
--- a/fs/reiserfs/bitmap.c +++ b/fs/reiserfs/bitmap.c @@ -513,9 +513,17 @@ static void __discard_prealloc(struct re "inode has negative prealloc blocks count."); #endif while (ei->i_prealloc_count > 0) { - reiserfs_free_prealloc_block(th, inode, ei->i_prealloc_block); - ei->i_prealloc_block++; + b_blocknr_t block_to_free; + + /* + * reiserfs_free_prealloc_block can drop the write lock, + * which could allow another caller to free the same block. + * We can protect against it by modifying the prealloc + * state before calling it. + */ + block_to_free = ei->i_prealloc_block++; ei->i_prealloc_count--; + reiserfs_free_prealloc_block(th, inode, block_to_free); dirty = 1; } if (dirty)