1	Locking scheme used for directory operations is based on two
2kinds of locks - per-inode (->i_rwsem) and per-filesystem
3(->s_vfs_rename_mutex).
4
5	When taking the i_rwsem on multiple non-directory objects, we
6always acquire the locks in order by increasing address.  We'll call
7that "inode pointer" order in the following.
8
9	For our purposes all operations fall in 5 classes:
10
111) read access.  Locking rules: caller locks directory we are accessing.
12The lock is taken shared.
13
142) object creation.  Locking rules: same as above, but the lock is taken
15exclusive.
16
173) object removal.  Locking rules: caller locks parent, finds victim,
18locks victim and calls the method.  Locks are exclusive.
19
204) rename() that is _not_ cross-directory.  Locking rules: caller locks
21the parent and finds source and target.  In case of exchange (with
22RENAME_EXCHANGE in flags argument) lock both.  In any case,
23if the target already exists, lock it.  If the source is a non-directory,
24lock it.  If we need to lock both, lock them in inode pointer order.
25Then call the method.  All locks are exclusive.
26NB: we might get away with locking the the source (and target in exchange
27case) shared.
28
295) link creation.  Locking rules:
30	* lock parent
31	* check that source is not a directory
32	* lock source
33	* call the method.
34All locks are exclusive.
35
366) cross-directory rename.  The trickiest in the whole bunch.  Locking
37rules:
38	* lock the filesystem
39	* lock parents in "ancestors first" order.
40	* find source and target.
41	* if old parent is equal to or is a descendent of target
42		fail with -ENOTEMPTY
43	* if new parent is equal to or is a descendent of source
44		fail with -ELOOP
45	* If it's an exchange, lock both the source and the target.
46	* If the target exists, lock it.  If the source is a non-directory,
47	  lock it.  If we need to lock both, do so in inode pointer order.
48	* call the method.
49All ->i_rwsem are taken exclusive.  Again, we might get away with locking
50the the source (and target in exchange case) shared.
51
52The rules above obviously guarantee that all directories that are going to be
53read, modified or removed by method will be locked by caller.
54
55
56If no directory is its own ancestor, the scheme above is deadlock-free.
57Proof:
58
59	First of all, at any moment we have a partial ordering of the
60objects - A < B iff A is an ancestor of B.
61
62	That ordering can change.  However, the following is true:
63
64(1) if object removal or non-cross-directory rename holds lock on A and
65    attempts to acquire lock on B, A will remain the parent of B until we
66    acquire the lock on B.  (Proof: only cross-directory rename can change
67    the parent of object and it would have to lock the parent).
68
69(2) if cross-directory rename holds the lock on filesystem, order will not
70    change until rename acquires all locks.  (Proof: other cross-directory
71    renames will be blocked on filesystem lock and we don't start changing
72    the order until we had acquired all locks).
73
74(3) locks on non-directory objects are acquired only after locks on
75    directory objects, and are acquired in inode pointer order.
76    (Proof: all operations but renames take lock on at most one
77    non-directory object, except renames, which take locks on source and
78    target in inode pointer order in the case they are not directories.)
79
80	Now consider the minimal deadlock.  Each process is blocked on
81attempt to acquire some lock and already holds at least one lock.  Let's
82consider the set of contended locks.  First of all, filesystem lock is
83not contended, since any process blocked on it is not holding any locks.
84Thus all processes are blocked on ->i_rwsem.
85
86	By (3), any process holding a non-directory lock can only be
87waiting on another non-directory lock with a larger address.  Therefore
88the process holding the "largest" such lock can always make progress, and
89non-directory objects are not included in the set of contended locks.
90
91	Thus link creation can't be a part of deadlock - it can't be
92blocked on source and it means that it doesn't hold any locks.
93
94	Any contended object is either held by cross-directory rename or
95has a child that is also contended.  Indeed, suppose that it is held by
96operation other than cross-directory rename.  Then the lock this operation
97is blocked on belongs to child of that object due to (1).
98
99	It means that one of the operations is cross-directory rename.
100Otherwise the set of contended objects would be infinite - each of them
101would have a contended child and we had assumed that no object is its
102own descendent.  Moreover, there is exactly one cross-directory rename
103(see above).
104
105	Consider the object blocking the cross-directory rename.  One
106of its descendents is locked by cross-directory rename (otherwise we
107would again have an infinite set of contended objects).  But that
108means that cross-directory rename is taking locks out of order.  Due
109to (2) the order hadn't changed since we had acquired filesystem lock.
110But locking rules for cross-directory rename guarantee that we do not
111try to acquire lock on descendent before the lock on ancestor.
112Contradiction.  I.e.  deadlock is impossible.  Q.E.D.
113
114
115	These operations are guaranteed to avoid loop creation.  Indeed,
116the only operation that could introduce loops is cross-directory rename.
117Since the only new (parent, child) pair added by rename() is (new parent,
118source), such loop would have to contain these objects and the rest of it
119would have to exist before rename().  I.e. at the moment of loop creation
120rename() responsible for that would be holding filesystem lock and new parent
121would have to be equal to or a descendent of source.  But that means that
122new parent had been equal to or a descendent of source since the moment when
123we had acquired filesystem lock and rename() would fail with -ELOOP in that
124case.
125
126	While this locking scheme works for arbitrary DAGs, it relies on
127ability to check that directory is a descendent of another object.  Current
128implementation assumes that directory graph is a tree.  This assumption is
129also preserved by all operations (cross-directory rename on a tree that would
130not introduce a cycle will leave it a tree and link() fails for directories).
131
132	Notice that "directory" in the above == "anything that might have
133children", so if we are going to introduce hybrid objects we will need
134either to make sure that link(2) doesn't work for them or to make changes
135in is_subdir() that would make it work even in presence of such beasts.
136