change system file

hc

2024-10-09 244b2c5ca8b14627e4a17755e5922221e121c771

 kernel/Documentation/filesystems/directory-locking.rst
..
..
@@ -22,12 +22,11 @@
22
22
 3) object removal.  Locking rules: caller locks parent, finds victim,
23
23
 locks victim and calls the method.  Locks are exclusive.
24
24
 
25
-4) rename() that is _not_ cross-directory.  Locking rules: caller locks
26
-the parent and finds source and target.  In case of exchange (with
27
-RENAME_EXCHANGE in flags argument) lock both.  In any case,
28
-if the target already exists, lock it.  If the source is a non-directory,
29
-lock it.  If we need to lock both, lock them in inode pointer order.
30
-Then call the method.  All locks are exclusive.
25
+4) rename() that is _not_ cross-directory.  Locking rules: caller locks the
26
+parent and finds source and target.  We lock both (provided they exist).  If we
27
+need to lock two inodes of different type (dir vs non-dir), we lock directory
28
+first.  If we need to lock two inodes of the same type, lock them in inode
29
+pointer order.  Then call the method.  All locks are exclusive.
31
30
 NB: we might get away with locking the source (and target in exchange
32
31
 case) shared.
33
32
 
..
..
@@ -44,15 +43,17 @@
44
43
 rules:
45
44
 
46
45
 	* lock the filesystem
47
-	* lock parents in "ancestors first" order.
46
+	* lock parents in "ancestors first" order. If one is not ancestor of
47
+	  the other, lock them in inode pointer order.
48
48
 	* find source and target.
49
49
 	* if old parent is equal to or is a descendent of target
50
50
 	  fail with -ENOTEMPTY
51
51
 	* if new parent is equal to or is a descendent of source
52
52
 	  fail with -ELOOP
53
-	* If it's an exchange, lock both the source and the target.
54
-	* If the target exists, lock it.  If the source is a non-directory,
55
-	  lock it.  If we need to lock both, do so in inode pointer order.
53
+	* Lock both the source and the target provided they exist. If we
54
+	  need to lock two inodes of different type (dir vs non-dir), we lock
55
+	  the directory first. If we need to lock two inodes of the same type,
56
+	  lock them in inode pointer order.
56
57
 	* call the method.
57
58
 
58
59
 All ->i_rwsem are taken exclusive.  Again, we might get away with locking
..
..
@@ -66,8 +67,9 @@
66
67
 
67
68
 Proof:
68
69
 
69
-	First of all, at any moment we have a partial ordering of the
70
-	objects - A < B iff A is an ancestor of B.
70
+	First of all, at any moment we have a linear ordering of the
71
+	objects - A < B iff (A is an ancestor of B) or (B is not an ancestor
72
+        of A and ptr(A) < ptr(B)).
71
73
 
72
74
 	That ordering can change.  However, the following is true:
73
75