btrfs: fix readdir deadlock with pagefault
Readdir does dir_emit while under the btree lock. dir_emit can trigger
the page fault which means we can deadlock. Fix this by allocating a
buffer on opening a directory and copying the readdir into this buffer
and doing dir_emit from outside of the tree lock.
Thread A
readdir <holding tree lock>
dir_emit
<page fault>
down_read(mmap_sem)
Thread B
mmap write
down_write(mmap_sem)
page_mkwrite
wait_ordered_extents
Process C
finish_ordered_extent
insert_reserved_file_extent
try to lock leaf <hang>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ copy the deadlock scenario to changelog ]
Signed-off-by: David Sterba <dsterba@suse.com>
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index b9d5d95..a3db2b1 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -1264,6 +1264,11 @@ struct btrfs_root {
atomic64_t qgroup_meta_rsv;
};
+struct btrfs_file_private {
+ struct btrfs_trans_handle *trans;
+ void *filldir_buf;
+};
+
static inline u32 btrfs_inode_sectorsize(const struct inode *inode)
{
return btrfs_sb(inode->i_sb)->sectorsize;