aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation/filesystems/Locking
diff options
context:
space:
mode:
Diffstat (limited to 'Documentation/filesystems/Locking')
-rw-r--r--Documentation/filesystems/Locking6
1 files changed, 3 insertions, 3 deletions
diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index 59c14159cc47..d866551be037 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -54,7 +54,7 @@ ata *);
54 54
55locking rules: 55locking rules:
56 all may block, none have BKL 56 all may block, none have BKL
57 i_sem(inode) 57 i_mutex(inode)
58lookup: yes 58lookup: yes
59create: yes 59create: yes
60link: yes (both) 60link: yes (both)
@@ -74,7 +74,7 @@ setxattr: yes
74getxattr: no 74getxattr: no
75listxattr: no 75listxattr: no
76removexattr: yes 76removexattr: yes
77 Additionally, ->rmdir(), ->unlink() and ->rename() have ->i_sem on 77 Additionally, ->rmdir(), ->unlink() and ->rename() have ->i_mutex on
78victim. 78victim.
79 cross-directory ->rename() has (per-superblock) ->s_vfs_rename_sem. 79 cross-directory ->rename() has (per-superblock) ->s_vfs_rename_sem.
80 ->truncate() is never called directly - it's a callback, not a 80 ->truncate() is never called directly - it's a callback, not a
@@ -461,7 +461,7 @@ doesn't take the BKL.
461->read on directories probably must go away - we should just enforce -EISDIR 461->read on directories probably must go away - we should just enforce -EISDIR
462in sys_read() and friends. 462in sys_read() and friends.
463 463
464->fsync() has i_sem on inode. 464->fsync() has i_mutex on inode.
465 465
466--------------------------- dquot_operations ------------------------------- 466--------------------------- dquot_operations -------------------------------
467prototypes: 467prototypes: