ext4: Fix possible recursive locking warning in EXT4_IOC_MOVE_EXT
From: Akira Fujita <[email protected]>
If kernel configuration related to lock debugging is enabled,
we get possible recursive locking warning.
Therefore, use down_write_nested() to guarantee
the lock order explicitly.
This problem has been reported by Brian Rogers before.
http://marc.info/?l=linux-ext4&m=125115356928011&w=1
Reported-by: Brian Rogers <[email protected]>
Signed-off-by: Akira Fujita <[email protected]>
---
fs/ext4/move_extent.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c
index a7410b3..2ca6aa3 100644
--- a/fs/ext4/move_extent.c
+++ b/fs/ext4/move_extent.c
@@ -189,7 +189,7 @@ double_down_write_data_sem(struct inode *orig_inode, struct inode *donor_inode)
}
down_write(&EXT4_I(first)->i_data_sem);
- down_write(&EXT4_I(second)->i_data_sem);
+ down_write_nested(&EXT4_I(second)->i_data_sem, SINGLE_DEPTH_NESTING);
}
/**
On Fri, Oct 30, 2009 at 04:42:01PM +0900, Akira Fujita wrote:
> ext4: Fix possible recursive locking warning in EXT4_IOC_MOVE_EXT
>
> From: Akira Fujita <[email protected]>
Thanks, I've added this patch to the ext4 patch queue, but with the
following updated commit description:
ext4: Fix possible recursive locking warning in EXT4_IOC_MOVE_EXT
From: Akira Fujita <[email protected]>
If CONFIG_PROVE_LOCKING is enabled, the double_down_write_data_sem()
will trigger a false-positive warning of a recursive lock. Since we
take i_data_sem for the two inodes ordered by their inode numbers,
this isn't a problem. Use of down_write_nested() will notify the lock
dependency checker machinery that there is no problem here.
This problem was reported by Brian Rogers:
http://marc.info/?l=linux-ext4&m=125115356928011&w=1
Reported-by: Brian Rogers <[email protected]>
Signed-off-by: Akira Fujita <[email protected]>
Signed-off-by: "Theodore Ts'o" <[email protected]>