btrfs: Fix missing close devices
PATCH 1/1 btrfs: Fix missing close devices
From: Zqiang <[email protected]>
When the btrfs fill super error, we should first close devices and then call deactivate_locked_super func to free fs_info.
--- fs/btrfs/super.c | 1 + 1 file changed, 1 insertion(+)
fs/btrfs/super.c +1 -0
@@ -1675,6 +1675,7 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type, error =security_sb_set_mnt_opts (s, new_sec_opts,0 ,NULL );security_free_mnt_opts (&new_sec_opts);if (error) { +btrfs_close_devices (fs_devices);deactivate_locked_super (s);return ERR_PTR (error); }
Signature · 1 line
2.17.1
Quoted 16 lines · On 21/09/2020 13:00, [email protected] wrote:
From: Zqiang <[email protected]>
When the btrfs fill super error, we should first close devices and then call
deactivate_locked_superfunc to freefs_info.Signed-off-by Zqiang <[email protected]>--- fs/btrfs/super.c | 1 + 1 file changed, 1 insertion(+)fs/btrfs/super.c +0
-0
fs/btrfs/super.c static struct dentry *btrfs_mount_root(struct file_system_type *fs_type, · line 1681
Quoted 7 lines
fs/btrfs/super.c +1
-0@@ -1675,6 +1675,7 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type, error =security_sb_set_mnt_opts (s, new_sec_opts,0 ,NULL );security_free_mnt_opts (&new_sec_opts);if (error) { +btrfs_close_devices (fs_devices);deactivate_locked_super (s);return ERR_PTR (error); }
Hmm you didn't change anything, so my report in [1] still exists.
Nikolay Borisov
NAKQuoted 16 lines · On 21.09.20 г. 10:29 ч., [email protected] wrote:
From: Zqiang <[email protected]>
When the btrfs fill super error, we should first close devices and then call
deactivate_locked_superfunc to freefs_info.Signed-off-by Zqiang <[email protected]>--- fs/btrfs/super.c | 1 + 1 file changed, 1 insertion(+)fs/btrfs/super.c +0
-0
fs/btrfs/super.c static struct dentry *btrfs_mount_root(struct file_system_type *fs_type, · line 1681
Quoted 7 lines
fs/btrfs/super.c +1
-0@@ -1675,6 +1675,7 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type, error =security_sb_set_mnt_opts (s, new_sec_opts,0 ,NULL );security_free_mnt_opts (&new_sec_opts);if (error) { +btrfs_close_devices (fs_devices);deactivate_locked_super (s);return ERR_PTR (error); }
NAK,
Devices are properly closed via:
deactivate_locked_super
kill_sb (btrfs_kill_super)
kill_anon_super
generic_shutdown_super
put_super (btrfs_put_super)
close_ctree
It seems you haven't done deep enough analysis of the involved call chains.