1 Started Oct 1999 by Kanoj Sarcar <kanoj@sgi.com>
2
3 The intent of this file is to have an uptodate, running commentary
4 from different people about how locking and synchronization is done
5 in the Linux vm code.
6
7 page_table_lock
8 --------------------------------------
9
10 Page stealers pick processes out of the process pool and scan for
11 the best process to steal pages from. To guarantee the existence
12 of the victim mm, a mm_count inc and a mmdrop are done in swap_out().
13 Page stealers hold kernel_lock to protect against a bunch of races.
14 The vma list of the victim mm is also scanned by the stealer,
15 and the page_table_lock is used to preserve list sanity against the
16 process adding/deleting to the list. This also guarantees existence
17 of the vma. Vma existence is not guaranteed once try_to_swap_out()
18 drops the page_table_lock. To guarantee the existence of the underlying
19 file structure, a get_file is done before the swapout() method is
20 invoked. The page passed into swapout() is guaranteed not to be reused
21 for a different purpose because the page reference count due to being
22 present in the user's pte is not released till after swapout() returns.
23
24 Any code that modifies the vmlist, or the vm_start/vm_end/
25 vm_flags:VM_LOCKED/vm_next of any vma *in the list* must prevent
26 kswapd from looking at the chain. This does not include driver mmap()
27 methods, for example, since the vma is still not in the list.
28
29 The rules are:
30 1. To modify the vmlist (add/delete or change fields in an element),
31 you must hold mmap_sem to guard against clones doing mmap/munmap/faults,
32 (ie all vm system calls and faults), and from ptrace, swapin due to
33 swap deletion etc.
34 2. To modify the vmlist (add/delete or change fields in an element),
35 you must also hold page_table_lock, to guard against page stealers
36 scanning the list.
37 3. To scan the vmlist (find_vma()), you must either
38 a. grab mmap_sem, which should be done by all cases except
39 page stealer.
40 or
41 b. grab page_table_lock, only done by page stealer.
42 4. While holding the page_table_lock, you must be able to guarantee
43 that no code path will lead to page stealing. A better guarantee is
44 to claim non sleepability, which ensures that you are not sleeping
45 for a lock, whose holder might in turn be doing page stealing.
46 5. You must be able to guarantee that while holding page_table_lock
47 or page_table_lock of mm A, you will not try to get either lock
48 for mm B.
49
50 The caveats are:
51 1. find_vma() makes use of, and updates, the mmap_cache pointer hint.
52 The update of mmap_cache is racy (page stealer can race with other code
53 that invokes find_vma with mmap_sem held), but that is okay, since it
54 is a hint. This can be fixed, if desired, by having find_vma grab the
55 page_table_lock.
56
57
58 Code that add/delete elements from the vmlist chain are
59 1. callers of insert_vm_struct
60 2. callers of merge_segments
61 3. callers of avl_remove
62
63 Code that changes vm_start/vm_end/vm_flags:VM_LOCKED of vma's on
64 the list:
65 1. expand_stack
66 2. mprotect
67 3. mlock
68 4. mremap
69
70 It is advisable that changes to vm_start/vm_end be protected, although
71 in some cases it is not really needed. Eg, vm_start is modified by
72 expand_stack(), it is hard to come up with a destructive scenario without
73 having the vmlist protection in this case.
74
75 The page_table_lock nests with the inode i_shared_lock and the kmem cache
76 c_spinlock spinlocks. This is okay, since code that holds i_shared_lock
77 never asks for memory, and the kmem code asks for pages after dropping
78 c_spinlock. The page_table_lock also nests with pagecache_lock and
79 pagemap_lru_lock spinlocks, and no code asks for memory with these locks
80 held.
81
82 The page_table_lock is grabbed while holding the kernel_lock spinning monitor.
83
84 The page_table_lock is a spin lock.
85
86 swap_list_lock/swap_device_lock
87 -------------------------------
88 The swap devices are chained in priority order from the "swap_list" header.
89 The "swap_list" is used for the round-robin swaphandle allocation strategy.
90 The #free swaphandles is maintained in "nr_swap_pages". These two together
91 are protected by the swap_list_lock.
92
93 The swap_device_lock, which is per swap device, protects the reference
94 counts on the corresponding swaphandles, maintained in the "swap_map"
95 array, and the "highest_bit" and "lowest_bit" fields.
96
97 Both of these are spinlocks, and are never acquired from intr level. The
98 locking hierarchy is swap_list_lock -> swap_device_lock.
99
100 To prevent races between swap space deletion or async readahead swapins
101 deciding whether a swap handle is being used, ie worthy of being read in
102 from disk, and an unmap -> swap_free making the handle unused, the swap
103 delete and readahead code grabs a temp reference on the swaphandle to
104 prevent warning messages from swap_duplicate <- read_swap_cache_async.
105
106 Swap cache locking
107 ------------------
108 Pages are added into the swap cache with kernel_lock held, to make sure
109 that multiple pages are not being added (and hence lost) by associating
110 all of them with the same swaphandle.
111
112 Pages are guaranteed not to be removed from the scache if the page is
113 "shared": ie, other processes hold reference on the page or the associated
114 swap handle. The only code that does not follow this rule is shrink_mmap,
115 which deletes pages from the swap cache if no process has a reference on
116 the page (multiple processes might have references on the corresponding
117 swap handle though). lookup_swap_cache() races with shrink_mmap, when
118 establishing a reference on a scache page, so, it must check whether the
119 page it located is still in the swapcache, or shrink_mmap deleted it.
120 (This race is due to the fact that shrink_mmap looks at the page ref
121 count with pagecache_lock, but then drops pagecache_lock before deleting
122 the page from the scache).
123
124 do_wp_page and do_swap_page have MP races in them while trying to figure
125 out whether a page is "shared", by looking at the page_count + swap_count.
126 To preserve the sum of the counts, the page lock _must_ be acquired before
127 calling is_page_shared (else processes might switch their swap_count refs
128 to the page count refs, after the page count ref has been snapshotted).
129
130 Swap device deletion code currently breaks all the scache assumptions,
131 since it grabs neither mmap_sem nor page_table_lock.
This page was automatically generated by the
LXR engine.
Visit the LXR main site for more
information.