Btrfs: Add more synchronization before creating a snapshot
File data checksums are only done during writepage, so we have to make sure
all pages are written when the snapshot is taken. This also adds some
locking so that new writes don't race in and add new dirty pages.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index cd75c90..03e1c4a 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -333,6 +333,7 @@
struct inode *inode;
struct kobject root_kobj;
struct completion kobj_unregister;
+ struct rw_semaphore snap_sem;
u64 objectid;
u64 last_trans;
u32 blocksize;