]> git.proxmox.com Git - mirror_ubuntu-zesty-kernel.git/blob - Documentation/memory-barriers.txt
Merge remote-tracking branches 'regulator/fix/doc', 'regulator/fix/max77686' and...
[mirror_ubuntu-zesty-kernel.git] / Documentation / memory-barriers.txt
1 ============================
2 LINUX KERNEL MEMORY BARRIERS
3 ============================
4
5 By: David Howells <dhowells@redhat.com>
6 Paul E. McKenney <paulmck@linux.vnet.ibm.com>
7
8 Contents:
9
10 (*) Abstract memory access model.
11
12 - Device operations.
13 - Guarantees.
14
15 (*) What are memory barriers?
16
17 - Varieties of memory barrier.
18 - What may not be assumed about memory barriers?
19 - Data dependency barriers.
20 - Control dependencies.
21 - SMP barrier pairing.
22 - Examples of memory barrier sequences.
23 - Read memory barriers vs load speculation.
24 - Transitivity
25
26 (*) Explicit kernel barriers.
27
28 - Compiler barrier.
29 - CPU memory barriers.
30 - MMIO write barrier.
31
32 (*) Implicit kernel memory barriers.
33
34 - Locking functions.
35 - Interrupt disabling functions.
36 - Sleep and wake-up functions.
37 - Miscellaneous functions.
38
39 (*) Inter-CPU locking barrier effects.
40
41 - Locks vs memory accesses.
42 - Locks vs I/O accesses.
43
44 (*) Where are memory barriers needed?
45
46 - Interprocessor interaction.
47 - Atomic operations.
48 - Accessing devices.
49 - Interrupts.
50
51 (*) Kernel I/O barrier effects.
52
53 (*) Assumed minimum execution ordering model.
54
55 (*) The effects of the cpu cache.
56
57 - Cache coherency.
58 - Cache coherency vs DMA.
59 - Cache coherency vs MMIO.
60
61 (*) The things CPUs get up to.
62
63 - And then there's the Alpha.
64
65 (*) Example uses.
66
67 - Circular buffers.
68
69 (*) References.
70
71
72 ============================
73 ABSTRACT MEMORY ACCESS MODEL
74 ============================
75
76 Consider the following abstract model of the system:
77
78 : :
79 : :
80 : :
81 +-------+ : +--------+ : +-------+
82 | | : | | : | |
83 | | : | | : | |
84 | CPU 1 |<----->| Memory |<----->| CPU 2 |
85 | | : | | : | |
86 | | : | | : | |
87 +-------+ : +--------+ : +-------+
88 ^ : ^ : ^
89 | : | : |
90 | : | : |
91 | : v : |
92 | : +--------+ : |
93 | : | | : |
94 | : | | : |
95 +---------->| Device |<----------+
96 : | | :
97 : | | :
98 : +--------+ :
99 : :
100
101 Each CPU executes a program that generates memory access operations. In the
102 abstract CPU, memory operation ordering is very relaxed, and a CPU may actually
103 perform the memory operations in any order it likes, provided program causality
104 appears to be maintained. Similarly, the compiler may also arrange the
105 instructions it emits in any order it likes, provided it doesn't affect the
106 apparent operation of the program.
107
108 So in the above diagram, the effects of the memory operations performed by a
109 CPU are perceived by the rest of the system as the operations cross the
110 interface between the CPU and rest of the system (the dotted lines).
111
112
113 For example, consider the following sequence of events:
114
115 CPU 1 CPU 2
116 =============== ===============
117 { A == 1; B == 2 }
118 A = 3; x = B;
119 B = 4; y = A;
120
121 The set of accesses as seen by the memory system in the middle can be arranged
122 in 24 different combinations:
123
124 STORE A=3, STORE B=4, y=LOAD A->3, x=LOAD B->4
125 STORE A=3, STORE B=4, x=LOAD B->4, y=LOAD A->3
126 STORE A=3, y=LOAD A->3, STORE B=4, x=LOAD B->4
127 STORE A=3, y=LOAD A->3, x=LOAD B->2, STORE B=4
128 STORE A=3, x=LOAD B->2, STORE B=4, y=LOAD A->3
129 STORE A=3, x=LOAD B->2, y=LOAD A->3, STORE B=4
130 STORE B=4, STORE A=3, y=LOAD A->3, x=LOAD B->4
131 STORE B=4, ...
132 ...
133
134 and can thus result in four different combinations of values:
135
136 x == 2, y == 1
137 x == 2, y == 3
138 x == 4, y == 1
139 x == 4, y == 3
140
141
142 Furthermore, the stores committed by a CPU to the memory system may not be
143 perceived by the loads made by another CPU in the same order as the stores were
144 committed.
145
146
147 As a further example, consider this sequence of events:
148
149 CPU 1 CPU 2
150 =============== ===============
151 { A == 1, B == 2, C = 3, P == &A, Q == &C }
152 B = 4; Q = P;
153 P = &B D = *Q;
154
155 There is an obvious data dependency here, as the value loaded into D depends on
156 the address retrieved from P by CPU 2. At the end of the sequence, any of the
157 following results are possible:
158
159 (Q == &A) and (D == 1)
160 (Q == &B) and (D == 2)
161 (Q == &B) and (D == 4)
162
163 Note that CPU 2 will never try and load C into D because the CPU will load P
164 into Q before issuing the load of *Q.
165
166
167 DEVICE OPERATIONS
168 -----------------
169
170 Some devices present their control interfaces as collections of memory
171 locations, but the order in which the control registers are accessed is very
172 important. For instance, imagine an ethernet card with a set of internal
173 registers that are accessed through an address port register (A) and a data
174 port register (D). To read internal register 5, the following code might then
175 be used:
176
177 *A = 5;
178 x = *D;
179
180 but this might show up as either of the following two sequences:
181
182 STORE *A = 5, x = LOAD *D
183 x = LOAD *D, STORE *A = 5
184
185 the second of which will almost certainly result in a malfunction, since it set
186 the address _after_ attempting to read the register.
187
188
189 GUARANTEES
190 ----------
191
192 There are some minimal guarantees that may be expected of a CPU:
193
194 (*) On any given CPU, dependent memory accesses will be issued in order, with
195 respect to itself. This means that for:
196
197 ACCESS_ONCE(Q) = P; smp_read_barrier_depends(); D = ACCESS_ONCE(*Q);
198
199 the CPU will issue the following memory operations:
200
201 Q = LOAD P, D = LOAD *Q
202
203 and always in that order. On most systems, smp_read_barrier_depends()
204 does nothing, but it is required for DEC Alpha. The ACCESS_ONCE()
205 is required to prevent compiler mischief. Please note that you
206 should normally use something like rcu_dereference() instead of
207 open-coding smp_read_barrier_depends().
208
209 (*) Overlapping loads and stores within a particular CPU will appear to be
210 ordered within that CPU. This means that for:
211
212 a = ACCESS_ONCE(*X); ACCESS_ONCE(*X) = b;
213
214 the CPU will only issue the following sequence of memory operations:
215
216 a = LOAD *X, STORE *X = b
217
218 And for:
219
220 ACCESS_ONCE(*X) = c; d = ACCESS_ONCE(*X);
221
222 the CPU will only issue:
223
224 STORE *X = c, d = LOAD *X
225
226 (Loads and stores overlap if they are targeted at overlapping pieces of
227 memory).
228
229 And there are a number of things that _must_ or _must_not_ be assumed:
230
231 (*) It _must_not_ be assumed that the compiler will do what you want with
232 memory references that are not protected by ACCESS_ONCE(). Without
233 ACCESS_ONCE(), the compiler is within its rights to do all sorts
234 of "creative" transformations, which are covered in the Compiler
235 Barrier section.
236
237 (*) It _must_not_ be assumed that independent loads and stores will be issued
238 in the order given. This means that for:
239
240 X = *A; Y = *B; *D = Z;
241
242 we may get any of the following sequences:
243
244 X = LOAD *A, Y = LOAD *B, STORE *D = Z
245 X = LOAD *A, STORE *D = Z, Y = LOAD *B
246 Y = LOAD *B, X = LOAD *A, STORE *D = Z
247 Y = LOAD *B, STORE *D = Z, X = LOAD *A
248 STORE *D = Z, X = LOAD *A, Y = LOAD *B
249 STORE *D = Z, Y = LOAD *B, X = LOAD *A
250
251 (*) It _must_ be assumed that overlapping memory accesses may be merged or
252 discarded. This means that for:
253
254 X = *A; Y = *(A + 4);
255
256 we may get any one of the following sequences:
257
258 X = LOAD *A; Y = LOAD *(A + 4);
259 Y = LOAD *(A + 4); X = LOAD *A;
260 {X, Y} = LOAD {*A, *(A + 4) };
261
262 And for:
263
264 *A = X; *(A + 4) = Y;
265
266 we may get any of:
267
268 STORE *A = X; STORE *(A + 4) = Y;
269 STORE *(A + 4) = Y; STORE *A = X;
270 STORE {*A, *(A + 4) } = {X, Y};
271
272 And there are anti-guarantees:
273
274 (*) These guarantees do not apply to bitfields, because compilers often
275 generate code to modify these using non-atomic read-modify-write
276 sequences. Do not attempt to use bitfields to synchronize parallel
277 algorithms.
278
279 (*) Even in cases where bitfields are protected by locks, all fields
280 in a given bitfield must be protected by one lock. If two fields
281 in a given bitfield are protected by different locks, the compiler's
282 non-atomic read-modify-write sequences can cause an update to one
283 field to corrupt the value of an adjacent field.
284
285 (*) These guarantees apply only to properly aligned and sized scalar
286 variables. "Properly sized" currently means variables that are
287 the same size as "char", "short", "int" and "long". "Properly
288 aligned" means the natural alignment, thus no constraints for
289 "char", two-byte alignment for "short", four-byte alignment for
290 "int", and either four-byte or eight-byte alignment for "long",
291 on 32-bit and 64-bit systems, respectively. Note that these
292 guarantees were introduced into the C11 standard, so beware when
293 using older pre-C11 compilers (for example, gcc 4.6). The portion
294 of the standard containing this guarantee is Section 3.14, which
295 defines "memory location" as follows:
296
297 memory location
298 either an object of scalar type, or a maximal sequence
299 of adjacent bit-fields all having nonzero width
300
301 NOTE 1: Two threads of execution can update and access
302 separate memory locations without interfering with
303 each other.
304
305 NOTE 2: A bit-field and an adjacent non-bit-field member
306 are in separate memory locations. The same applies
307 to two bit-fields, if one is declared inside a nested
308 structure declaration and the other is not, or if the two
309 are separated by a zero-length bit-field declaration,
310 or if they are separated by a non-bit-field member
311 declaration. It is not safe to concurrently update two
312 bit-fields in the same structure if all members declared
313 between them are also bit-fields, no matter what the
314 sizes of those intervening bit-fields happen to be.
315
316
317 =========================
318 WHAT ARE MEMORY BARRIERS?
319 =========================
320
321 As can be seen above, independent memory operations are effectively performed
322 in random order, but this can be a problem for CPU-CPU interaction and for I/O.
323 What is required is some way of intervening to instruct the compiler and the
324 CPU to restrict the order.
325
326 Memory barriers are such interventions. They impose a perceived partial
327 ordering over the memory operations on either side of the barrier.
328
329 Such enforcement is important because the CPUs and other devices in a system
330 can use a variety of tricks to improve performance, including reordering,
331 deferral and combination of memory operations; speculative loads; speculative
332 branch prediction and various types of caching. Memory barriers are used to
333 override or suppress these tricks, allowing the code to sanely control the
334 interaction of multiple CPUs and/or devices.
335
336
337 VARIETIES OF MEMORY BARRIER
338 ---------------------------
339
340 Memory barriers come in four basic varieties:
341
342 (1) Write (or store) memory barriers.
343
344 A write memory barrier gives a guarantee that all the STORE operations
345 specified before the barrier will appear to happen before all the STORE
346 operations specified after the barrier with respect to the other
347 components of the system.
348
349 A write barrier is a partial ordering on stores only; it is not required
350 to have any effect on loads.
351
352 A CPU can be viewed as committing a sequence of store operations to the
353 memory system as time progresses. All stores before a write barrier will
354 occur in the sequence _before_ all the stores after the write barrier.
355
356 [!] Note that write barriers should normally be paired with read or data
357 dependency barriers; see the "SMP barrier pairing" subsection.
358
359
360 (2) Data dependency barriers.
361
362 A data dependency barrier is a weaker form of read barrier. In the case
363 where two loads are performed such that the second depends on the result
364 of the first (eg: the first load retrieves the address to which the second
365 load will be directed), a data dependency barrier would be required to
366 make sure that the target of the second load is updated before the address
367 obtained by the first load is accessed.
368
369 A data dependency barrier is a partial ordering on interdependent loads
370 only; it is not required to have any effect on stores, independent loads
371 or overlapping loads.
372
373 As mentioned in (1), the other CPUs in the system can be viewed as
374 committing sequences of stores to the memory system that the CPU being
375 considered can then perceive. A data dependency barrier issued by the CPU
376 under consideration guarantees that for any load preceding it, if that
377 load touches one of a sequence of stores from another CPU, then by the
378 time the barrier completes, the effects of all the stores prior to that
379 touched by the load will be perceptible to any loads issued after the data
380 dependency barrier.
381
382 See the "Examples of memory barrier sequences" subsection for diagrams
383 showing the ordering constraints.
384
385 [!] Note that the first load really has to have a _data_ dependency and
386 not a control dependency. If the address for the second load is dependent
387 on the first load, but the dependency is through a conditional rather than
388 actually loading the address itself, then it's a _control_ dependency and
389 a full read barrier or better is required. See the "Control dependencies"
390 subsection for more information.
391
392 [!] Note that data dependency barriers should normally be paired with
393 write barriers; see the "SMP barrier pairing" subsection.
394
395
396 (3) Read (or load) memory barriers.
397
398 A read barrier is a data dependency barrier plus a guarantee that all the
399 LOAD operations specified before the barrier will appear to happen before
400 all the LOAD operations specified after the barrier with respect to the
401 other components of the system.
402
403 A read barrier is a partial ordering on loads only; it is not required to
404 have any effect on stores.
405
406 Read memory barriers imply data dependency barriers, and so can substitute
407 for them.
408
409 [!] Note that read barriers should normally be paired with write barriers;
410 see the "SMP barrier pairing" subsection.
411
412
413 (4) General memory barriers.
414
415 A general memory barrier gives a guarantee that all the LOAD and STORE
416 operations specified before the barrier will appear to happen before all
417 the LOAD and STORE operations specified after the barrier with respect to
418 the other components of the system.
419
420 A general memory barrier is a partial ordering over both loads and stores.
421
422 General memory barriers imply both read and write memory barriers, and so
423 can substitute for either.
424
425
426 And a couple of implicit varieties:
427
428 (5) ACQUIRE operations.
429
430 This acts as a one-way permeable barrier. It guarantees that all memory
431 operations after the ACQUIRE operation will appear to happen after the
432 ACQUIRE operation with respect to the other components of the system.
433 ACQUIRE operations include LOCK operations and smp_load_acquire()
434 operations.
435
436 Memory operations that occur before an ACQUIRE operation may appear to
437 happen after it completes.
438
439 An ACQUIRE operation should almost always be paired with a RELEASE
440 operation.
441
442
443 (6) RELEASE operations.
444
445 This also acts as a one-way permeable barrier. It guarantees that all
446 memory operations before the RELEASE operation will appear to happen
447 before the RELEASE operation with respect to the other components of the
448 system. RELEASE operations include UNLOCK operations and
449 smp_store_release() operations.
450
451 Memory operations that occur after a RELEASE operation may appear to
452 happen before it completes.
453
454 The use of ACQUIRE and RELEASE operations generally precludes the need
455 for other sorts of memory barrier (but note the exceptions mentioned in
456 the subsection "MMIO write barrier"). In addition, a RELEASE+ACQUIRE
457 pair is -not- guaranteed to act as a full memory barrier. However, after
458 an ACQUIRE on a given variable, all memory accesses preceding any prior
459 RELEASE on that same variable are guaranteed to be visible. In other
460 words, within a given variable's critical section, all accesses of all
461 previous critical sections for that variable are guaranteed to have
462 completed.
463
464 This means that ACQUIRE acts as a minimal "acquire" operation and
465 RELEASE acts as a minimal "release" operation.
466
467
468 Memory barriers are only required where there's a possibility of interaction
469 between two CPUs or between a CPU and a device. If it can be guaranteed that
470 there won't be any such interaction in any particular piece of code, then
471 memory barriers are unnecessary in that piece of code.
472
473
474 Note that these are the _minimum_ guarantees. Different architectures may give
475 more substantial guarantees, but they may _not_ be relied upon outside of arch
476 specific code.
477
478
479 WHAT MAY NOT BE ASSUMED ABOUT MEMORY BARRIERS?
480 ----------------------------------------------
481
482 There are certain things that the Linux kernel memory barriers do not guarantee:
483
484 (*) There is no guarantee that any of the memory accesses specified before a
485 memory barrier will be _complete_ by the completion of a memory barrier
486 instruction; the barrier can be considered to draw a line in that CPU's
487 access queue that accesses of the appropriate type may not cross.
488
489 (*) There is no guarantee that issuing a memory barrier on one CPU will have
490 any direct effect on another CPU or any other hardware in the system. The
491 indirect effect will be the order in which the second CPU sees the effects
492 of the first CPU's accesses occur, but see the next point:
493
494 (*) There is no guarantee that a CPU will see the correct order of effects
495 from a second CPU's accesses, even _if_ the second CPU uses a memory
496 barrier, unless the first CPU _also_ uses a matching memory barrier (see
497 the subsection on "SMP Barrier Pairing").
498
499 (*) There is no guarantee that some intervening piece of off-the-CPU
500 hardware[*] will not reorder the memory accesses. CPU cache coherency
501 mechanisms should propagate the indirect effects of a memory barrier
502 between CPUs, but might not do so in order.
503
504 [*] For information on bus mastering DMA and coherency please read:
505
506 Documentation/PCI/pci.txt
507 Documentation/DMA-API-HOWTO.txt
508 Documentation/DMA-API.txt
509
510
511 DATA DEPENDENCY BARRIERS
512 ------------------------
513
514 The usage requirements of data dependency barriers are a little subtle, and
515 it's not always obvious that they're needed. To illustrate, consider the
516 following sequence of events:
517
518 CPU 1 CPU 2
519 =============== ===============
520 { A == 1, B == 2, C = 3, P == &A, Q == &C }
521 B = 4;
522 <write barrier>
523 ACCESS_ONCE(P) = &B
524 Q = ACCESS_ONCE(P);
525 D = *Q;
526
527 There's a clear data dependency here, and it would seem that by the end of the
528 sequence, Q must be either &A or &B, and that:
529
530 (Q == &A) implies (D == 1)
531 (Q == &B) implies (D == 4)
532
533 But! CPU 2's perception of P may be updated _before_ its perception of B, thus
534 leading to the following situation:
535
536 (Q == &B) and (D == 2) ????
537
538 Whilst this may seem like a failure of coherency or causality maintenance, it
539 isn't, and this behaviour can be observed on certain real CPUs (such as the DEC
540 Alpha).
541
542 To deal with this, a data dependency barrier or better must be inserted
543 between the address load and the data load:
544
545 CPU 1 CPU 2
546 =============== ===============
547 { A == 1, B == 2, C = 3, P == &A, Q == &C }
548 B = 4;
549 <write barrier>
550 ACCESS_ONCE(P) = &B
551 Q = ACCESS_ONCE(P);
552 <data dependency barrier>
553 D = *Q;
554
555 This enforces the occurrence of one of the two implications, and prevents the
556 third possibility from arising.
557
558 [!] Note that this extremely counterintuitive situation arises most easily on
559 machines with split caches, so that, for example, one cache bank processes
560 even-numbered cache lines and the other bank processes odd-numbered cache
561 lines. The pointer P might be stored in an odd-numbered cache line, and the
562 variable B might be stored in an even-numbered cache line. Then, if the
563 even-numbered bank of the reading CPU's cache is extremely busy while the
564 odd-numbered bank is idle, one can see the new value of the pointer P (&B),
565 but the old value of the variable B (2).
566
567
568 Another example of where data dependency barriers might be required is where a
569 number is read from memory and then used to calculate the index for an array
570 access:
571
572 CPU 1 CPU 2
573 =============== ===============
574 { M[0] == 1, M[1] == 2, M[3] = 3, P == 0, Q == 3 }
575 M[1] = 4;
576 <write barrier>
577 ACCESS_ONCE(P) = 1
578 Q = ACCESS_ONCE(P);
579 <data dependency barrier>
580 D = M[Q];
581
582
583 The data dependency barrier is very important to the RCU system,
584 for example. See rcu_assign_pointer() and rcu_dereference() in
585 include/linux/rcupdate.h. This permits the current target of an RCU'd
586 pointer to be replaced with a new modified target, without the replacement
587 target appearing to be incompletely initialised.
588
589 See also the subsection on "Cache Coherency" for a more thorough example.
590
591
592 CONTROL DEPENDENCIES
593 --------------------
594
595 A load-load control dependency requires a full read memory barrier, not
596 simply a data dependency barrier to make it work correctly. Consider the
597 following bit of code:
598
599 q = ACCESS_ONCE(a);
600 if (q) {
601 <data dependency barrier> /* BUG: No data dependency!!! */
602 p = ACCESS_ONCE(b);
603 }
604
605 This will not have the desired effect because there is no actual data
606 dependency, but rather a control dependency that the CPU may short-circuit
607 by attempting to predict the outcome in advance, so that other CPUs see
608 the load from b as having happened before the load from a. In such a
609 case what's actually required is:
610
611 q = ACCESS_ONCE(a);
612 if (q) {
613 <read barrier>
614 p = ACCESS_ONCE(b);
615 }
616
617 However, stores are not speculated. This means that ordering -is- provided
618 for load-store control dependencies, as in the following example:
619
620 q = ACCESS_ONCE(a);
621 if (q) {
622 ACCESS_ONCE(b) = p;
623 }
624
625 Control dependencies pair normally with other types of barriers.
626 That said, please note that ACCESS_ONCE() is not optional! Without the
627 ACCESS_ONCE(), might combine the load from 'a' with other loads from
628 'a', and the store to 'b' with other stores to 'b', with possible highly
629 counterintuitive effects on ordering.
630
631 Worse yet, if the compiler is able to prove (say) that the value of
632 variable 'a' is always non-zero, it would be well within its rights
633 to optimize the original example by eliminating the "if" statement
634 as follows:
635
636 q = a;
637 b = p; /* BUG: Compiler and CPU can both reorder!!! */
638
639 So don't leave out the ACCESS_ONCE().
640
641 It is tempting to try to enforce ordering on identical stores on both
642 branches of the "if" statement as follows:
643
644 q = ACCESS_ONCE(a);
645 if (q) {
646 barrier();
647 ACCESS_ONCE(b) = p;
648 do_something();
649 } else {
650 barrier();
651 ACCESS_ONCE(b) = p;
652 do_something_else();
653 }
654
655 Unfortunately, current compilers will transform this as follows at high
656 optimization levels:
657
658 q = ACCESS_ONCE(a);
659 barrier();
660 ACCESS_ONCE(b) = p; /* BUG: No ordering vs. load from a!!! */
661 if (q) {
662 /* ACCESS_ONCE(b) = p; -- moved up, BUG!!! */
663 do_something();
664 } else {
665 /* ACCESS_ONCE(b) = p; -- moved up, BUG!!! */
666 do_something_else();
667 }
668
669 Now there is no conditional between the load from 'a' and the store to
670 'b', which means that the CPU is within its rights to reorder them:
671 The conditional is absolutely required, and must be present in the
672 assembly code even after all compiler optimizations have been applied.
673 Therefore, if you need ordering in this example, you need explicit
674 memory barriers, for example, smp_store_release():
675
676 q = ACCESS_ONCE(a);
677 if (q) {
678 smp_store_release(&b, p);
679 do_something();
680 } else {
681 smp_store_release(&b, p);
682 do_something_else();
683 }
684
685 In contrast, without explicit memory barriers, two-legged-if control
686 ordering is guaranteed only when the stores differ, for example:
687
688 q = ACCESS_ONCE(a);
689 if (q) {
690 ACCESS_ONCE(b) = p;
691 do_something();
692 } else {
693 ACCESS_ONCE(b) = r;
694 do_something_else();
695 }
696
697 The initial ACCESS_ONCE() is still required to prevent the compiler from
698 proving the value of 'a'.
699
700 In addition, you need to be careful what you do with the local variable 'q',
701 otherwise the compiler might be able to guess the value and again remove
702 the needed conditional. For example:
703
704 q = ACCESS_ONCE(a);
705 if (q % MAX) {
706 ACCESS_ONCE(b) = p;
707 do_something();
708 } else {
709 ACCESS_ONCE(b) = r;
710 do_something_else();
711 }
712
713 If MAX is defined to be 1, then the compiler knows that (q % MAX) is
714 equal to zero, in which case the compiler is within its rights to
715 transform the above code into the following:
716
717 q = ACCESS_ONCE(a);
718 ACCESS_ONCE(b) = p;
719 do_something_else();
720
721 Given this transformation, the CPU is not required to respect the ordering
722 between the load from variable 'a' and the store to variable 'b'. It is
723 tempting to add a barrier(), but this does not help. The conditional
724 is gone, and the barrier won't bring it back. Therefore, if you are
725 relying on this ordering, you should make sure that MAX is greater than
726 one, perhaps as follows:
727
728 q = ACCESS_ONCE(a);
729 BUILD_BUG_ON(MAX <= 1); /* Order load from a with store to b. */
730 if (q % MAX) {
731 ACCESS_ONCE(b) = p;
732 do_something();
733 } else {
734 ACCESS_ONCE(b) = r;
735 do_something_else();
736 }
737
738 Please note once again that the stores to 'b' differ. If they were
739 identical, as noted earlier, the compiler could pull this store outside
740 of the 'if' statement.
741
742 You must also be careful not to rely too much on boolean short-circuit
743 evaluation. Consider this example:
744
745 q = ACCESS_ONCE(a);
746 if (a || 1 > 0)
747 ACCESS_ONCE(b) = 1;
748
749 Because the second condition is always true, the compiler can transform
750 this example as following, defeating control dependency:
751
752 q = ACCESS_ONCE(a);
753 ACCESS_ONCE(b) = 1;
754
755 This example underscores the need to ensure that the compiler cannot
756 out-guess your code. More generally, although ACCESS_ONCE() does force
757 the compiler to actually emit code for a given load, it does not force
758 the compiler to use the results.
759
760 Finally, control dependencies do -not- provide transitivity. This is
761 demonstrated by two related examples, with the initial values of
762 x and y both being zero:
763
764 CPU 0 CPU 1
765 ===================== =====================
766 r1 = ACCESS_ONCE(x); r2 = ACCESS_ONCE(y);
767 if (r1 > 0) if (r2 > 0)
768 ACCESS_ONCE(y) = 1; ACCESS_ONCE(x) = 1;
769
770 assert(!(r1 == 1 && r2 == 1));
771
772 The above two-CPU example will never trigger the assert(). However,
773 if control dependencies guaranteed transitivity (which they do not),
774 then adding the following CPU would guarantee a related assertion:
775
776 CPU 2
777 =====================
778 ACCESS_ONCE(x) = 2;
779
780 assert(!(r1 == 2 && r2 == 1 && x == 2)); /* FAILS!!! */
781
782 But because control dependencies do -not- provide transitivity, the above
783 assertion can fail after the combined three-CPU example completes. If you
784 need the three-CPU example to provide ordering, you will need smp_mb()
785 between the loads and stores in the CPU 0 and CPU 1 code fragments,
786 that is, just before or just after the "if" statements.
787
788 These two examples are the LB and WWC litmus tests from this paper:
789 http://www.cl.cam.ac.uk/users/pes20/ppc-supplemental/test6.pdf and this
790 site: https://www.cl.cam.ac.uk/~pes20/ppcmem/index.html.
791
792 In summary:
793
794 (*) Control dependencies can order prior loads against later stores.
795 However, they do -not- guarantee any other sort of ordering:
796 Not prior loads against later loads, nor prior stores against
797 later anything. If you need these other forms of ordering,
798 use smp_rmb(), smp_wmb(), or, in the case of prior stores and
799 later loads, smp_mb().
800
801 (*) If both legs of the "if" statement begin with identical stores
802 to the same variable, a barrier() statement is required at the
803 beginning of each leg of the "if" statement.
804
805 (*) Control dependencies require at least one run-time conditional
806 between the prior load and the subsequent store, and this
807 conditional must involve the prior load. If the compiler
808 is able to optimize the conditional away, it will have also
809 optimized away the ordering. Careful use of ACCESS_ONCE() can
810 help to preserve the needed conditional.
811
812 (*) Control dependencies require that the compiler avoid reordering the
813 dependency into nonexistence. Careful use of ACCESS_ONCE() or
814 barrier() can help to preserve your control dependency. Please
815 see the Compiler Barrier section for more information.
816
817 (*) Control dependencies pair normally with other types of barriers.
818
819 (*) Control dependencies do -not- provide transitivity. If you
820 need transitivity, use smp_mb().
821
822
823 SMP BARRIER PAIRING
824 -------------------
825
826 When dealing with CPU-CPU interactions, certain types of memory barrier should
827 always be paired. A lack of appropriate pairing is almost certainly an error.
828
829 General barriers pair with each other, though they also pair with most
830 other types of barriers, albeit without transitivity. An acquire barrier
831 pairs with a release barrier, but both may also pair with other barriers,
832 including of course general barriers. A write barrier pairs with a data
833 dependency barrier, a control dependency, an acquire barrier, a release
834 barrier, a read barrier, or a general barrier. Similarly a read barrier,
835 control dependency, or a data dependency barrier pairs with a write
836 barrier, an acquire barrier, a release barrier, or a general barrier:
837
838 CPU 1 CPU 2
839 =============== ===============
840 ACCESS_ONCE(a) = 1;
841 <write barrier>
842 ACCESS_ONCE(b) = 2; x = ACCESS_ONCE(b);
843 <read barrier>
844 y = ACCESS_ONCE(a);
845
846 Or:
847
848 CPU 1 CPU 2
849 =============== ===============================
850 a = 1;
851 <write barrier>
852 ACCESS_ONCE(b) = &a; x = ACCESS_ONCE(b);
853 <data dependency barrier>
854 y = *x;
855
856 Or even:
857
858 CPU 1 CPU 2
859 =============== ===============================
860 r1 = ACCESS_ONCE(y);
861 <general barrier>
862 ACCESS_ONCE(y) = 1; if (r2 = ACCESS_ONCE(x)) {
863 <implicit control dependency>
864 ACCESS_ONCE(y) = 1;
865 }
866
867 assert(r1 == 0 || r2 == 0);
868
869 Basically, the read barrier always has to be there, even though it can be of
870 the "weaker" type.
871
872 [!] Note that the stores before the write barrier would normally be expected to
873 match the loads after the read barrier or the data dependency barrier, and vice
874 versa:
875
876 CPU 1 CPU 2
877 =================== ===================
878 ACCESS_ONCE(a) = 1; }---- --->{ v = ACCESS_ONCE(c);
879 ACCESS_ONCE(b) = 2; } \ / { w = ACCESS_ONCE(d);
880 <write barrier> \ <read barrier>
881 ACCESS_ONCE(c) = 3; } / \ { x = ACCESS_ONCE(a);
882 ACCESS_ONCE(d) = 4; }---- --->{ y = ACCESS_ONCE(b);
883
884
885 EXAMPLES OF MEMORY BARRIER SEQUENCES
886 ------------------------------------
887
888 Firstly, write barriers act as partial orderings on store operations.
889 Consider the following sequence of events:
890
891 CPU 1
892 =======================
893 STORE A = 1
894 STORE B = 2
895 STORE C = 3
896 <write barrier>
897 STORE D = 4
898 STORE E = 5
899
900 This sequence of events is committed to the memory coherence system in an order
901 that the rest of the system might perceive as the unordered set of { STORE A,
902 STORE B, STORE C } all occurring before the unordered set of { STORE D, STORE E
903 }:
904
905 +-------+ : :
906 | | +------+
907 | |------>| C=3 | } /\
908 | | : +------+ }----- \ -----> Events perceptible to
909 | | : | A=1 | } \/ the rest of the system
910 | | : +------+ }
911 | CPU 1 | : | B=2 | }
912 | | +------+ }
913 | | wwwwwwwwwwwwwwww } <--- At this point the write barrier
914 | | +------+ } requires all stores prior to the
915 | | : | E=5 | } barrier to be committed before
916 | | : +------+ } further stores may take place
917 | |------>| D=4 | }
918 | | +------+
919 +-------+ : :
920 |
921 | Sequence in which stores are committed to the
922 | memory system by CPU 1
923 V
924
925
926 Secondly, data dependency barriers act as partial orderings on data-dependent
927 loads. Consider the following sequence of events:
928
929 CPU 1 CPU 2
930 ======================= =======================
931 { B = 7; X = 9; Y = 8; C = &Y }
932 STORE A = 1
933 STORE B = 2
934 <write barrier>
935 STORE C = &B LOAD X
936 STORE D = 4 LOAD C (gets &B)
937 LOAD *C (reads B)
938
939 Without intervention, CPU 2 may perceive the events on CPU 1 in some
940 effectively random order, despite the write barrier issued by CPU 1:
941
942 +-------+ : : : :
943 | | +------+ +-------+ | Sequence of update
944 | |------>| B=2 |----- --->| Y->8 | | of perception on
945 | | : +------+ \ +-------+ | CPU 2
946 | CPU 1 | : | A=1 | \ --->| C->&Y | V
947 | | +------+ | +-------+
948 | | wwwwwwwwwwwwwwww | : :
949 | | +------+ | : :
950 | | : | C=&B |--- | : : +-------+
951 | | : +------+ \ | +-------+ | |
952 | |------>| D=4 | ----------->| C->&B |------>| |
953 | | +------+ | +-------+ | |
954 +-------+ : : | : : | |
955 | : : | |
956 | : : | CPU 2 |
957 | +-------+ | |
958 Apparently incorrect ---> | | B->7 |------>| |
959 perception of B (!) | +-------+ | |
960 | : : | |
961 | +-------+ | |
962 The load of X holds ---> \ | X->9 |------>| |
963 up the maintenance \ +-------+ | |
964 of coherence of B ----->| B->2 | +-------+
965 +-------+
966 : :
967
968
969 In the above example, CPU 2 perceives that B is 7, despite the load of *C
970 (which would be B) coming after the LOAD of C.
971
972 If, however, a data dependency barrier were to be placed between the load of C
973 and the load of *C (ie: B) on CPU 2:
974
975 CPU 1 CPU 2
976 ======================= =======================
977 { B = 7; X = 9; Y = 8; C = &Y }
978 STORE A = 1
979 STORE B = 2
980 <write barrier>
981 STORE C = &B LOAD X
982 STORE D = 4 LOAD C (gets &B)
983 <data dependency barrier>
984 LOAD *C (reads B)
985
986 then the following will occur:
987
988 +-------+ : : : :
989 | | +------+ +-------+
990 | |------>| B=2 |----- --->| Y->8 |
991 | | : +------+ \ +-------+
992 | CPU 1 | : | A=1 | \ --->| C->&Y |
993 | | +------+ | +-------+
994 | | wwwwwwwwwwwwwwww | : :
995 | | +------+ | : :
996 | | : | C=&B |--- | : : +-------+
997 | | : +------+ \ | +-------+ | |
998 | |------>| D=4 | ----------->| C->&B |------>| |
999 | | +------+ | +-------+ | |
1000 +-------+ : : | : : | |
1001 | : : | |
1002 | : : | CPU 2 |
1003 | +-------+ | |
1004 | | X->9 |------>| |
1005 | +-------+ | |
1006 Makes sure all effects ---> \ ddddddddddddddddd | |
1007 prior to the store of C \ +-------+ | |
1008 are perceptible to ----->| B->2 |------>| |
1009 subsequent loads +-------+ | |
1010 : : +-------+
1011
1012
1013 And thirdly, a read barrier acts as a partial order on loads. Consider the
1014 following sequence of events:
1015
1016 CPU 1 CPU 2
1017 ======================= =======================
1018 { A = 0, B = 9 }
1019 STORE A=1
1020 <write barrier>
1021 STORE B=2
1022 LOAD B
1023 LOAD A
1024
1025 Without intervention, CPU 2 may then choose to perceive the events on CPU 1 in
1026 some effectively random order, despite the write barrier issued by CPU 1:
1027
1028 +-------+ : : : :
1029 | | +------+ +-------+
1030 | |------>| A=1 |------ --->| A->0 |
1031 | | +------+ \ +-------+
1032 | CPU 1 | wwwwwwwwwwwwwwww \ --->| B->9 |
1033 | | +------+ | +-------+
1034 | |------>| B=2 |--- | : :
1035 | | +------+ \ | : : +-------+
1036 +-------+ : : \ | +-------+ | |
1037 ---------->| B->2 |------>| |
1038 | +-------+ | CPU 2 |
1039 | | A->0 |------>| |
1040 | +-------+ | |
1041 | : : +-------+
1042 \ : :
1043 \ +-------+
1044 ---->| A->1 |
1045 +-------+
1046 : :
1047
1048
1049 If, however, a read barrier were to be placed between the load of B and the
1050 load of A on CPU 2:
1051
1052 CPU 1 CPU 2
1053 ======================= =======================
1054 { A = 0, B = 9 }
1055 STORE A=1
1056 <write barrier>
1057 STORE B=2
1058 LOAD B
1059 <read barrier>
1060 LOAD A
1061
1062 then the partial ordering imposed by CPU 1 will be perceived correctly by CPU
1063 2:
1064
1065 +-------+ : : : :
1066 | | +------+ +-------+
1067 | |------>| A=1 |------ --->| A->0 |
1068 | | +------+ \ +-------+
1069 | CPU 1 | wwwwwwwwwwwwwwww \ --->| B->9 |
1070 | | +------+ | +-------+
1071 | |------>| B=2 |--- | : :
1072 | | +------+ \ | : : +-------+
1073 +-------+ : : \ | +-------+ | |
1074 ---------->| B->2 |------>| |
1075 | +-------+ | CPU 2 |
1076 | : : | |
1077 | : : | |
1078 At this point the read ----> \ rrrrrrrrrrrrrrrrr | |
1079 barrier causes all effects \ +-------+ | |
1080 prior to the storage of B ---->| A->1 |------>| |
1081 to be perceptible to CPU 2 +-------+ | |
1082 : : +-------+
1083
1084
1085 To illustrate this more completely, consider what could happen if the code
1086 contained a load of A either side of the read barrier:
1087
1088 CPU 1 CPU 2
1089 ======================= =======================
1090 { A = 0, B = 9 }
1091 STORE A=1
1092 <write barrier>
1093 STORE B=2
1094 LOAD B
1095 LOAD A [first load of A]
1096 <read barrier>
1097 LOAD A [second load of A]
1098
1099 Even though the two loads of A both occur after the load of B, they may both
1100 come up with different values:
1101
1102 +-------+ : : : :
1103 | | +------+ +-------+
1104 | |------>| A=1 |------ --->| A->0 |
1105 | | +------+ \ +-------+
1106 | CPU 1 | wwwwwwwwwwwwwwww \ --->| B->9 |
1107 | | +------+ | +-------+
1108 | |------>| B=2 |--- | : :
1109 | | +------+ \ | : : +-------+
1110 +-------+ : : \ | +-------+ | |
1111 ---------->| B->2 |------>| |
1112 | +-------+ | CPU 2 |
1113 | : : | |
1114 | : : | |
1115 | +-------+ | |
1116 | | A->0 |------>| 1st |
1117 | +-------+ | |
1118 At this point the read ----> \ rrrrrrrrrrrrrrrrr | |
1119 barrier causes all effects \ +-------+ | |
1120 prior to the storage of B ---->| A->1 |------>| 2nd |
1121 to be perceptible to CPU 2 +-------+ | |
1122 : : +-------+
1123
1124
1125 But it may be that the update to A from CPU 1 becomes perceptible to CPU 2
1126 before the read barrier completes anyway:
1127
1128 +-------+ : : : :
1129 | | +------+ +-------+
1130 | |------>| A=1 |------ --->| A->0 |
1131 | | +------+ \ +-------+
1132 | CPU 1 | wwwwwwwwwwwwwwww \ --->| B->9 |
1133 | | +------+ | +-------+
1134 | |------>| B=2 |--- | : :
1135 | | +------+ \ | : : +-------+
1136 +-------+ : : \ | +-------+ | |
1137 ---------->| B->2 |------>| |
1138 | +-------+ | CPU 2 |
1139 | : : | |
1140 \ : : | |
1141 \ +-------+ | |
1142 ---->| A->1 |------>| 1st |
1143 +-------+ | |
1144 rrrrrrrrrrrrrrrrr | |
1145 +-------+ | |
1146 | A->1 |------>| 2nd |
1147 +-------+ | |
1148 : : +-------+
1149
1150
1151 The guarantee is that the second load will always come up with A == 1 if the
1152 load of B came up with B == 2. No such guarantee exists for the first load of
1153 A; that may come up with either A == 0 or A == 1.
1154
1155
1156 READ MEMORY BARRIERS VS LOAD SPECULATION
1157 ----------------------------------------
1158
1159 Many CPUs speculate with loads: that is they see that they will need to load an
1160 item from memory, and they find a time where they're not using the bus for any
1161 other loads, and so do the load in advance - even though they haven't actually
1162 got to that point in the instruction execution flow yet. This permits the
1163 actual load instruction to potentially complete immediately because the CPU
1164 already has the value to hand.
1165
1166 It may turn out that the CPU didn't actually need the value - perhaps because a
1167 branch circumvented the load - in which case it can discard the value or just
1168 cache it for later use.
1169
1170 Consider:
1171
1172 CPU 1 CPU 2
1173 ======================= =======================
1174 LOAD B
1175 DIVIDE } Divide instructions generally
1176 DIVIDE } take a long time to perform
1177 LOAD A
1178
1179 Which might appear as this:
1180
1181 : : +-------+
1182 +-------+ | |
1183 --->| B->2 |------>| |
1184 +-------+ | CPU 2 |
1185 : :DIVIDE | |
1186 +-------+ | |
1187 The CPU being busy doing a ---> --->| A->0 |~~~~ | |
1188 division speculates on the +-------+ ~ | |
1189 LOAD of A : : ~ | |
1190 : :DIVIDE | |
1191 : : ~ | |
1192 Once the divisions are complete --> : : ~-->| |
1193 the CPU can then perform the : : | |
1194 LOAD with immediate effect : : +-------+
1195
1196
1197 Placing a read barrier or a data dependency barrier just before the second
1198 load:
1199
1200 CPU 1 CPU 2
1201 ======================= =======================
1202 LOAD B
1203 DIVIDE
1204 DIVIDE
1205 <read barrier>
1206 LOAD A
1207
1208 will force any value speculatively obtained to be reconsidered to an extent
1209 dependent on the type of barrier used. If there was no change made to the
1210 speculated memory location, then the speculated value will just be used:
1211
1212 : : +-------+
1213 +-------+ | |
1214 --->| B->2 |------>| |
1215 +-------+ | CPU 2 |
1216 : :DIVIDE | |
1217 +-------+ | |
1218 The CPU being busy doing a ---> --->| A->0 |~~~~ | |
1219 division speculates on the +-------+ ~ | |
1220 LOAD of A : : ~ | |
1221 : :DIVIDE | |
1222 : : ~ | |
1223 : : ~ | |
1224 rrrrrrrrrrrrrrrr~ | |
1225 : : ~ | |
1226 : : ~-->| |
1227 : : | |
1228 : : +-------+
1229
1230
1231 but if there was an update or an invalidation from another CPU pending, then
1232 the speculation will be cancelled and the value reloaded:
1233
1234 : : +-------+
1235 +-------+ | |
1236 --->| B->2 |------>| |
1237 +-------+ | CPU 2 |
1238 : :DIVIDE | |
1239 +-------+ | |
1240 The CPU being busy doing a ---> --->| A->0 |~~~~ | |
1241 division speculates on the +-------+ ~ | |
1242 LOAD of A : : ~ | |
1243 : :DIVIDE | |
1244 : : ~ | |
1245 : : ~ | |
1246 rrrrrrrrrrrrrrrrr | |
1247 +-------+ | |
1248 The speculation is discarded ---> --->| A->1 |------>| |
1249 and an updated value is +-------+ | |
1250 retrieved : : +-------+
1251
1252
1253 TRANSITIVITY
1254 ------------
1255
1256 Transitivity is a deeply intuitive notion about ordering that is not
1257 always provided by real computer systems. The following example
1258 demonstrates transitivity (also called "cumulativity"):
1259
1260 CPU 1 CPU 2 CPU 3
1261 ======================= ======================= =======================
1262 { X = 0, Y = 0 }
1263 STORE X=1 LOAD X STORE Y=1
1264 <general barrier> <general barrier>
1265 LOAD Y LOAD X
1266
1267 Suppose that CPU 2's load from X returns 1 and its load from Y returns 0.
1268 This indicates that CPU 2's load from X in some sense follows CPU 1's
1269 store to X and that CPU 2's load from Y in some sense preceded CPU 3's
1270 store to Y. The question is then "Can CPU 3's load from X return 0?"
1271
1272 Because CPU 2's load from X in some sense came after CPU 1's store, it
1273 is natural to expect that CPU 3's load from X must therefore return 1.
1274 This expectation is an example of transitivity: if a load executing on
1275 CPU A follows a load from the same variable executing on CPU B, then
1276 CPU A's load must either return the same value that CPU B's load did,
1277 or must return some later value.
1278
1279 In the Linux kernel, use of general memory barriers guarantees
1280 transitivity. Therefore, in the above example, if CPU 2's load from X
1281 returns 1 and its load from Y returns 0, then CPU 3's load from X must
1282 also return 1.
1283
1284 However, transitivity is -not- guaranteed for read or write barriers.
1285 For example, suppose that CPU 2's general barrier in the above example
1286 is changed to a read barrier as shown below:
1287
1288 CPU 1 CPU 2 CPU 3
1289 ======================= ======================= =======================
1290 { X = 0, Y = 0 }
1291 STORE X=1 LOAD X STORE Y=1
1292 <read barrier> <general barrier>
1293 LOAD Y LOAD X
1294
1295 This substitution destroys transitivity: in this example, it is perfectly
1296 legal for CPU 2's load from X to return 1, its load from Y to return 0,
1297 and CPU 3's load from X to return 0.
1298
1299 The key point is that although CPU 2's read barrier orders its pair
1300 of loads, it does not guarantee to order CPU 1's store. Therefore, if
1301 this example runs on a system where CPUs 1 and 2 share a store buffer
1302 or a level of cache, CPU 2 might have early access to CPU 1's writes.
1303 General barriers are therefore required to ensure that all CPUs agree
1304 on the combined order of CPU 1's and CPU 2's accesses.
1305
1306 To reiterate, if your code requires transitivity, use general barriers
1307 throughout.
1308
1309
1310 ========================
1311 EXPLICIT KERNEL BARRIERS
1312 ========================
1313
1314 The Linux kernel has a variety of different barriers that act at different
1315 levels:
1316
1317 (*) Compiler barrier.
1318
1319 (*) CPU memory barriers.
1320
1321 (*) MMIO write barrier.
1322
1323
1324 COMPILER BARRIER
1325 ----------------
1326
1327 The Linux kernel has an explicit compiler barrier function that prevents the
1328 compiler from moving the memory accesses either side of it to the other side:
1329
1330 barrier();
1331
1332 This is a general barrier -- there are no read-read or write-write variants
1333 of barrier(). However, ACCESS_ONCE() can be thought of as a weak form
1334 for barrier() that affects only the specific accesses flagged by the
1335 ACCESS_ONCE().
1336
1337 The barrier() function has the following effects:
1338
1339 (*) Prevents the compiler from reordering accesses following the
1340 barrier() to precede any accesses preceding the barrier().
1341 One example use for this property is to ease communication between
1342 interrupt-handler code and the code that was interrupted.
1343
1344 (*) Within a loop, forces the compiler to load the variables used
1345 in that loop's conditional on each pass through that loop.
1346
1347 The ACCESS_ONCE() function can prevent any number of optimizations that,
1348 while perfectly safe in single-threaded code, can be fatal in concurrent
1349 code. Here are some examples of these sorts of optimizations:
1350
1351 (*) The compiler is within its rights to reorder loads and stores
1352 to the same variable, and in some cases, the CPU is within its
1353 rights to reorder loads to the same variable. This means that
1354 the following code:
1355
1356 a[0] = x;
1357 a[1] = x;
1358
1359 Might result in an older value of x stored in a[1] than in a[0].
1360 Prevent both the compiler and the CPU from doing this as follows:
1361
1362 a[0] = ACCESS_ONCE(x);
1363 a[1] = ACCESS_ONCE(x);
1364
1365 In short, ACCESS_ONCE() provides cache coherence for accesses from
1366 multiple CPUs to a single variable.
1367
1368 (*) The compiler is within its rights to merge successive loads from
1369 the same variable. Such merging can cause the compiler to "optimize"
1370 the following code:
1371
1372 while (tmp = a)
1373 do_something_with(tmp);
1374
1375 into the following code, which, although in some sense legitimate
1376 for single-threaded code, is almost certainly not what the developer
1377 intended:
1378
1379 if (tmp = a)
1380 for (;;)
1381 do_something_with(tmp);
1382
1383 Use ACCESS_ONCE() to prevent the compiler from doing this to you:
1384
1385 while (tmp = ACCESS_ONCE(a))
1386 do_something_with(tmp);
1387
1388 (*) The compiler is within its rights to reload a variable, for example,
1389 in cases where high register pressure prevents the compiler from
1390 keeping all data of interest in registers. The compiler might
1391 therefore optimize the variable 'tmp' out of our previous example:
1392
1393 while (tmp = a)
1394 do_something_with(tmp);
1395
1396 This could result in the following code, which is perfectly safe in
1397 single-threaded code, but can be fatal in concurrent code:
1398
1399 while (a)
1400 do_something_with(a);
1401
1402 For example, the optimized version of this code could result in
1403 passing a zero to do_something_with() in the case where the variable
1404 a was modified by some other CPU between the "while" statement and
1405 the call to do_something_with().
1406
1407 Again, use ACCESS_ONCE() to prevent the compiler from doing this:
1408
1409 while (tmp = ACCESS_ONCE(a))
1410 do_something_with(tmp);
1411
1412 Note that if the compiler runs short of registers, it might save
1413 tmp onto the stack. The overhead of this saving and later restoring
1414 is why compilers reload variables. Doing so is perfectly safe for
1415 single-threaded code, so you need to tell the compiler about cases
1416 where it is not safe.
1417
1418 (*) The compiler is within its rights to omit a load entirely if it knows
1419 what the value will be. For example, if the compiler can prove that
1420 the value of variable 'a' is always zero, it can optimize this code:
1421
1422 while (tmp = a)
1423 do_something_with(tmp);
1424
1425 Into this:
1426
1427 do { } while (0);
1428
1429 This transformation is a win for single-threaded code because it gets
1430 rid of a load and a branch. The problem is that the compiler will
1431 carry out its proof assuming that the current CPU is the only one
1432 updating variable 'a'. If variable 'a' is shared, then the compiler's
1433 proof will be erroneous. Use ACCESS_ONCE() to tell the compiler
1434 that it doesn't know as much as it thinks it does:
1435
1436 while (tmp = ACCESS_ONCE(a))
1437 do_something_with(tmp);
1438
1439 But please note that the compiler is also closely watching what you
1440 do with the value after the ACCESS_ONCE(). For example, suppose you
1441 do the following and MAX is a preprocessor macro with the value 1:
1442
1443 while ((tmp = ACCESS_ONCE(a)) % MAX)
1444 do_something_with(tmp);
1445
1446 Then the compiler knows that the result of the "%" operator applied
1447 to MAX will always be zero, again allowing the compiler to optimize
1448 the code into near-nonexistence. (It will still load from the
1449 variable 'a'.)
1450
1451 (*) Similarly, the compiler is within its rights to omit a store entirely
1452 if it knows that the variable already has the value being stored.
1453 Again, the compiler assumes that the current CPU is the only one
1454 storing into the variable, which can cause the compiler to do the
1455 wrong thing for shared variables. For example, suppose you have
1456 the following:
1457
1458 a = 0;
1459 /* Code that does not store to variable a. */
1460 a = 0;
1461
1462 The compiler sees that the value of variable 'a' is already zero, so
1463 it might well omit the second store. This would come as a fatal
1464 surprise if some other CPU might have stored to variable 'a' in the
1465 meantime.
1466
1467 Use ACCESS_ONCE() to prevent the compiler from making this sort of
1468 wrong guess:
1469
1470 ACCESS_ONCE(a) = 0;
1471 /* Code that does not store to variable a. */
1472 ACCESS_ONCE(a) = 0;
1473
1474 (*) The compiler is within its rights to reorder memory accesses unless
1475 you tell it not to. For example, consider the following interaction
1476 between process-level code and an interrupt handler:
1477
1478 void process_level(void)
1479 {
1480 msg = get_message();
1481 flag = true;
1482 }
1483
1484 void interrupt_handler(void)
1485 {
1486 if (flag)
1487 process_message(msg);
1488 }
1489
1490 There is nothing to prevent the compiler from transforming
1491 process_level() to the following, in fact, this might well be a
1492 win for single-threaded code:
1493
1494 void process_level(void)
1495 {
1496 flag = true;
1497 msg = get_message();
1498 }
1499
1500 If the interrupt occurs between these two statement, then
1501 interrupt_handler() might be passed a garbled msg. Use ACCESS_ONCE()
1502 to prevent this as follows:
1503
1504 void process_level(void)
1505 {
1506 ACCESS_ONCE(msg) = get_message();
1507 ACCESS_ONCE(flag) = true;
1508 }
1509
1510 void interrupt_handler(void)
1511 {
1512 if (ACCESS_ONCE(flag))
1513 process_message(ACCESS_ONCE(msg));
1514 }
1515
1516 Note that the ACCESS_ONCE() wrappers in interrupt_handler()
1517 are needed if this interrupt handler can itself be interrupted
1518 by something that also accesses 'flag' and 'msg', for example,
1519 a nested interrupt or an NMI. Otherwise, ACCESS_ONCE() is not
1520 needed in interrupt_handler() other than for documentation purposes.
1521 (Note also that nested interrupts do not typically occur in modern
1522 Linux kernels, in fact, if an interrupt handler returns with
1523 interrupts enabled, you will get a WARN_ONCE() splat.)
1524
1525 You should assume that the compiler can move ACCESS_ONCE() past
1526 code not containing ACCESS_ONCE(), barrier(), or similar primitives.
1527
1528 This effect could also be achieved using barrier(), but ACCESS_ONCE()
1529 is more selective: With ACCESS_ONCE(), the compiler need only forget
1530 the contents of the indicated memory locations, while with barrier()
1531 the compiler must discard the value of all memory locations that
1532 it has currented cached in any machine registers. Of course,
1533 the compiler must also respect the order in which the ACCESS_ONCE()s
1534 occur, though the CPU of course need not do so.
1535
1536 (*) The compiler is within its rights to invent stores to a variable,
1537 as in the following example:
1538
1539 if (a)
1540 b = a;
1541 else
1542 b = 42;
1543
1544 The compiler might save a branch by optimizing this as follows:
1545
1546 b = 42;
1547 if (a)
1548 b = a;
1549
1550 In single-threaded code, this is not only safe, but also saves
1551 a branch. Unfortunately, in concurrent code, this optimization
1552 could cause some other CPU to see a spurious value of 42 -- even
1553 if variable 'a' was never zero -- when loading variable 'b'.
1554 Use ACCESS_ONCE() to prevent this as follows:
1555
1556 if (a)
1557 ACCESS_ONCE(b) = a;
1558 else
1559 ACCESS_ONCE(b) = 42;
1560
1561 The compiler can also invent loads. These are usually less
1562 damaging, but they can result in cache-line bouncing and thus in
1563 poor performance and scalability. Use ACCESS_ONCE() to prevent
1564 invented loads.
1565
1566 (*) For aligned memory locations whose size allows them to be accessed
1567 with a single memory-reference instruction, prevents "load tearing"
1568 and "store tearing," in which a single large access is replaced by
1569 multiple smaller accesses. For example, given an architecture having
1570 16-bit store instructions with 7-bit immediate fields, the compiler
1571 might be tempted to use two 16-bit store-immediate instructions to
1572 implement the following 32-bit store:
1573
1574 p = 0x00010002;
1575
1576 Please note that GCC really does use this sort of optimization,
1577 which is not surprising given that it would likely take more
1578 than two instructions to build the constant and then store it.
1579 This optimization can therefore be a win in single-threaded code.
1580 In fact, a recent bug (since fixed) caused GCC to incorrectly use
1581 this optimization in a volatile store. In the absence of such bugs,
1582 use of ACCESS_ONCE() prevents store tearing in the following example:
1583
1584 ACCESS_ONCE(p) = 0x00010002;
1585
1586 Use of packed structures can also result in load and store tearing,
1587 as in this example:
1588
1589 struct __attribute__((__packed__)) foo {
1590 short a;
1591 int b;
1592 short c;
1593 };
1594 struct foo foo1, foo2;
1595 ...
1596
1597 foo2.a = foo1.a;
1598 foo2.b = foo1.b;
1599 foo2.c = foo1.c;
1600
1601 Because there are no ACCESS_ONCE() wrappers and no volatile markings,
1602 the compiler would be well within its rights to implement these three
1603 assignment statements as a pair of 32-bit loads followed by a pair
1604 of 32-bit stores. This would result in load tearing on 'foo1.b'
1605 and store tearing on 'foo2.b'. ACCESS_ONCE() again prevents tearing
1606 in this example:
1607
1608 foo2.a = foo1.a;
1609 ACCESS_ONCE(foo2.b) = ACCESS_ONCE(foo1.b);
1610 foo2.c = foo1.c;
1611
1612 All that aside, it is never necessary to use ACCESS_ONCE() on a variable
1613 that has been marked volatile. For example, because 'jiffies' is marked
1614 volatile, it is never necessary to say ACCESS_ONCE(jiffies). The reason
1615 for this is that ACCESS_ONCE() is implemented as a volatile cast, which
1616 has no effect when its argument is already marked volatile.
1617
1618 Please note that these compiler barriers have no direct effect on the CPU,
1619 which may then reorder things however it wishes.
1620
1621
1622 CPU MEMORY BARRIERS
1623 -------------------
1624
1625 The Linux kernel has eight basic CPU memory barriers:
1626
1627 TYPE MANDATORY SMP CONDITIONAL
1628 =============== ======================= ===========================
1629 GENERAL mb() smp_mb()
1630 WRITE wmb() smp_wmb()
1631 READ rmb() smp_rmb()
1632 DATA DEPENDENCY read_barrier_depends() smp_read_barrier_depends()
1633
1634
1635 All memory barriers except the data dependency barriers imply a compiler
1636 barrier. Data dependencies do not impose any additional compiler ordering.
1637
1638 Aside: In the case of data dependencies, the compiler would be expected to
1639 issue the loads in the correct order (eg. `a[b]` would have to load the value
1640 of b before loading a[b]), however there is no guarantee in the C specification
1641 that the compiler may not speculate the value of b (eg. is equal to 1) and load
1642 a before b (eg. tmp = a[1]; if (b != 1) tmp = a[b]; ). There is also the
1643 problem of a compiler reloading b after having loaded a[b], thus having a newer
1644 copy of b than a[b]. A consensus has not yet been reached about these problems,
1645 however the ACCESS_ONCE macro is a good place to start looking.
1646
1647 SMP memory barriers are reduced to compiler barriers on uniprocessor compiled
1648 systems because it is assumed that a CPU will appear to be self-consistent,
1649 and will order overlapping accesses correctly with respect to itself.
1650
1651 [!] Note that SMP memory barriers _must_ be used to control the ordering of
1652 references to shared memory on SMP systems, though the use of locking instead
1653 is sufficient.
1654
1655 Mandatory barriers should not be used to control SMP effects, since mandatory
1656 barriers unnecessarily impose overhead on UP systems. They may, however, be
1657 used to control MMIO effects on accesses through relaxed memory I/O windows.
1658 These are required even on non-SMP systems as they affect the order in which
1659 memory operations appear to a device by prohibiting both the compiler and the
1660 CPU from reordering them.
1661
1662
1663 There are some more advanced barrier functions:
1664
1665 (*) set_mb(var, value)
1666
1667 This assigns the value to the variable and then inserts a full memory
1668 barrier after it, depending on the function. It isn't guaranteed to
1669 insert anything more than a compiler barrier in a UP compilation.
1670
1671
1672 (*) smp_mb__before_atomic();
1673 (*) smp_mb__after_atomic();
1674
1675 These are for use with atomic (such as add, subtract, increment and
1676 decrement) functions that don't return a value, especially when used for
1677 reference counting. These functions do not imply memory barriers.
1678
1679 These are also used for atomic bitop functions that do not return a
1680 value (such as set_bit and clear_bit).
1681
1682 As an example, consider a piece of code that marks an object as being dead
1683 and then decrements the object's reference count:
1684
1685 obj->dead = 1;
1686 smp_mb__before_atomic();
1687 atomic_dec(&obj->ref_count);
1688
1689 This makes sure that the death mark on the object is perceived to be set
1690 *before* the reference counter is decremented.
1691
1692 See Documentation/atomic_ops.txt for more information. See the "Atomic
1693 operations" subsection for information on where to use these.
1694
1695
1696 (*) dma_wmb();
1697 (*) dma_rmb();
1698
1699 These are for use with consistent memory to guarantee the ordering
1700 of writes or reads of shared memory accessible to both the CPU and a
1701 DMA capable device.
1702
1703 For example, consider a device driver that shares memory with a device
1704 and uses a descriptor status value to indicate if the descriptor belongs
1705 to the device or the CPU, and a doorbell to notify it when new
1706 descriptors are available:
1707
1708 if (desc->status != DEVICE_OWN) {
1709 /* do not read data until we own descriptor */
1710 dma_rmb();
1711
1712 /* read/modify data */
1713 read_data = desc->data;
1714 desc->data = write_data;
1715
1716 /* flush modifications before status update */
1717 dma_wmb();
1718
1719 /* assign ownership */
1720 desc->status = DEVICE_OWN;
1721
1722 /* force memory to sync before notifying device via MMIO */
1723 wmb();
1724
1725 /* notify device of new descriptors */
1726 writel(DESC_NOTIFY, doorbell);
1727 }
1728
1729 The dma_rmb() allows us guarantee the device has released ownership
1730 before we read the data from the descriptor, and the dma_wmb() allows
1731 us to guarantee the data is written to the descriptor before the device
1732 can see it now has ownership. The wmb() is needed to guarantee that the
1733 cache coherent memory writes have completed before attempting a write to
1734 the cache incoherent MMIO region.
1735
1736 See Documentation/DMA-API.txt for more information on consistent memory.
1737
1738 MMIO WRITE BARRIER
1739 ------------------
1740
1741 The Linux kernel also has a special barrier for use with memory-mapped I/O
1742 writes:
1743
1744 mmiowb();
1745
1746 This is a variation on the mandatory write barrier that causes writes to weakly
1747 ordered I/O regions to be partially ordered. Its effects may go beyond the
1748 CPU->Hardware interface and actually affect the hardware at some level.
1749
1750 See the subsection "Locks vs I/O accesses" for more information.
1751
1752
1753 ===============================
1754 IMPLICIT KERNEL MEMORY BARRIERS
1755 ===============================
1756
1757 Some of the other functions in the linux kernel imply memory barriers, amongst
1758 which are locking and scheduling functions.
1759
1760 This specification is a _minimum_ guarantee; any particular architecture may
1761 provide more substantial guarantees, but these may not be relied upon outside
1762 of arch specific code.
1763
1764
1765 ACQUIRING FUNCTIONS
1766 -------------------
1767
1768 The Linux kernel has a number of locking constructs:
1769
1770 (*) spin locks
1771 (*) R/W spin locks
1772 (*) mutexes
1773 (*) semaphores
1774 (*) R/W semaphores
1775 (*) RCU
1776
1777 In all cases there are variants on "ACQUIRE" operations and "RELEASE" operations
1778 for each construct. These operations all imply certain barriers:
1779
1780 (1) ACQUIRE operation implication:
1781
1782 Memory operations issued after the ACQUIRE will be completed after the
1783 ACQUIRE operation has completed.
1784
1785 Memory operations issued before the ACQUIRE may be completed after
1786 the ACQUIRE operation has completed. An smp_mb__before_spinlock(),
1787 combined with a following ACQUIRE, orders prior loads against
1788 subsequent loads and stores and also orders prior stores against
1789 subsequent stores. Note that this is weaker than smp_mb()! The
1790 smp_mb__before_spinlock() primitive is free on many architectures.
1791
1792 (2) RELEASE operation implication:
1793
1794 Memory operations issued before the RELEASE will be completed before the
1795 RELEASE operation has completed.
1796
1797 Memory operations issued after the RELEASE may be completed before the
1798 RELEASE operation has completed.
1799
1800 (3) ACQUIRE vs ACQUIRE implication:
1801
1802 All ACQUIRE operations issued before another ACQUIRE operation will be
1803 completed before that ACQUIRE operation.
1804
1805 (4) ACQUIRE vs RELEASE implication:
1806
1807 All ACQUIRE operations issued before a RELEASE operation will be
1808 completed before the RELEASE operation.
1809
1810 (5) Failed conditional ACQUIRE implication:
1811
1812 Certain locking variants of the ACQUIRE operation may fail, either due to
1813 being unable to get the lock immediately, or due to receiving an unblocked
1814 signal whilst asleep waiting for the lock to become available. Failed
1815 locks do not imply any sort of barrier.
1816
1817 [!] Note: one of the consequences of lock ACQUIREs and RELEASEs being only
1818 one-way barriers is that the effects of instructions outside of a critical
1819 section may seep into the inside of the critical section.
1820
1821 An ACQUIRE followed by a RELEASE may not be assumed to be full memory barrier
1822 because it is possible for an access preceding the ACQUIRE to happen after the
1823 ACQUIRE, and an access following the RELEASE to happen before the RELEASE, and
1824 the two accesses can themselves then cross:
1825
1826 *A = a;
1827 ACQUIRE M
1828 RELEASE M
1829 *B = b;
1830
1831 may occur as:
1832
1833 ACQUIRE M, STORE *B, STORE *A, RELEASE M
1834
1835 When the ACQUIRE and RELEASE are a lock acquisition and release,
1836 respectively, this same reordering can occur if the lock's ACQUIRE and
1837 RELEASE are to the same lock variable, but only from the perspective of
1838 another CPU not holding that lock. In short, a ACQUIRE followed by an
1839 RELEASE may -not- be assumed to be a full memory barrier.
1840
1841 Similarly, the reverse case of a RELEASE followed by an ACQUIRE does not
1842 imply a full memory barrier. If it is necessary for a RELEASE-ACQUIRE
1843 pair to produce a full barrier, the ACQUIRE can be followed by an
1844 smp_mb__after_unlock_lock() invocation. This will produce a full barrier
1845 if either (a) the RELEASE and the ACQUIRE are executed by the same
1846 CPU or task, or (b) the RELEASE and ACQUIRE act on the same variable.
1847 The smp_mb__after_unlock_lock() primitive is free on many architectures.
1848 Without smp_mb__after_unlock_lock(), the CPU's execution of the critical
1849 sections corresponding to the RELEASE and the ACQUIRE can cross, so that:
1850
1851 *A = a;
1852 RELEASE M
1853 ACQUIRE N
1854 *B = b;
1855
1856 could occur as:
1857
1858 ACQUIRE N, STORE *B, STORE *A, RELEASE M
1859
1860 It might appear that this reordering could introduce a deadlock.
1861 However, this cannot happen because if such a deadlock threatened,
1862 the RELEASE would simply complete, thereby avoiding the deadlock.
1863
1864 Why does this work?
1865
1866 One key point is that we are only talking about the CPU doing
1867 the reordering, not the compiler. If the compiler (or, for
1868 that matter, the developer) switched the operations, deadlock
1869 -could- occur.
1870
1871 But suppose the CPU reordered the operations. In this case,
1872 the unlock precedes the lock in the assembly code. The CPU
1873 simply elected to try executing the later lock operation first.
1874 If there is a deadlock, this lock operation will simply spin (or
1875 try to sleep, but more on that later). The CPU will eventually
1876 execute the unlock operation (which preceded the lock operation
1877 in the assembly code), which will unravel the potential deadlock,
1878 allowing the lock operation to succeed.
1879
1880 But what if the lock is a sleeplock? In that case, the code will
1881 try to enter the scheduler, where it will eventually encounter
1882 a memory barrier, which will force the earlier unlock operation
1883 to complete, again unraveling the deadlock. There might be
1884 a sleep-unlock race, but the locking primitive needs to resolve
1885 such races properly in any case.
1886
1887 With smp_mb__after_unlock_lock(), the two critical sections cannot overlap.
1888 For example, with the following code, the store to *A will always be
1889 seen by other CPUs before the store to *B:
1890
1891 *A = a;
1892 RELEASE M
1893 ACQUIRE N
1894 smp_mb__after_unlock_lock();
1895 *B = b;
1896
1897 The operations will always occur in one of the following orders:
1898
1899 STORE *A, RELEASE, ACQUIRE, smp_mb__after_unlock_lock(), STORE *B
1900 STORE *A, ACQUIRE, RELEASE, smp_mb__after_unlock_lock(), STORE *B
1901 ACQUIRE, STORE *A, RELEASE, smp_mb__after_unlock_lock(), STORE *B
1902
1903 If the RELEASE and ACQUIRE were instead both operating on the same lock
1904 variable, only the first of these alternatives can occur. In addition,
1905 the more strongly ordered systems may rule out some of the above orders.
1906 But in any case, as noted earlier, the smp_mb__after_unlock_lock()
1907 ensures that the store to *A will always be seen as happening before
1908 the store to *B.
1909
1910 Locks and semaphores may not provide any guarantee of ordering on UP compiled
1911 systems, and so cannot be counted on in such a situation to actually achieve
1912 anything at all - especially with respect to I/O accesses - unless combined
1913 with interrupt disabling operations.
1914
1915 See also the section on "Inter-CPU locking barrier effects".
1916
1917
1918 As an example, consider the following:
1919
1920 *A = a;
1921 *B = b;
1922 ACQUIRE
1923 *C = c;
1924 *D = d;
1925 RELEASE
1926 *E = e;
1927 *F = f;
1928
1929 The following sequence of events is acceptable:
1930
1931 ACQUIRE, {*F,*A}, *E, {*C,*D}, *B, RELEASE
1932
1933 [+] Note that {*F,*A} indicates a combined access.
1934
1935 But none of the following are:
1936
1937 {*F,*A}, *B, ACQUIRE, *C, *D, RELEASE, *E
1938 *A, *B, *C, ACQUIRE, *D, RELEASE, *E, *F
1939 *A, *B, ACQUIRE, *C, RELEASE, *D, *E, *F
1940 *B, ACQUIRE, *C, *D, RELEASE, {*F,*A}, *E
1941
1942
1943
1944 INTERRUPT DISABLING FUNCTIONS
1945 -----------------------------
1946
1947 Functions that disable interrupts (ACQUIRE equivalent) and enable interrupts
1948 (RELEASE equivalent) will act as compiler barriers only. So if memory or I/O
1949 barriers are required in such a situation, they must be provided from some
1950 other means.
1951
1952
1953 SLEEP AND WAKE-UP FUNCTIONS
1954 ---------------------------
1955
1956 Sleeping and waking on an event flagged in global data can be viewed as an
1957 interaction between two pieces of data: the task state of the task waiting for
1958 the event and the global data used to indicate the event. To make sure that
1959 these appear to happen in the right order, the primitives to begin the process
1960 of going to sleep, and the primitives to initiate a wake up imply certain
1961 barriers.
1962
1963 Firstly, the sleeper normally follows something like this sequence of events:
1964
1965 for (;;) {
1966 set_current_state(TASK_UNINTERRUPTIBLE);
1967 if (event_indicated)
1968 break;
1969 schedule();
1970 }
1971
1972 A general memory barrier is interpolated automatically by set_current_state()
1973 after it has altered the task state:
1974
1975 CPU 1
1976 ===============================
1977 set_current_state();
1978 set_mb();
1979 STORE current->state
1980 <general barrier>
1981 LOAD event_indicated
1982
1983 set_current_state() may be wrapped by:
1984
1985 prepare_to_wait();
1986 prepare_to_wait_exclusive();
1987
1988 which therefore also imply a general memory barrier after setting the state.
1989 The whole sequence above is available in various canned forms, all of which
1990 interpolate the memory barrier in the right place:
1991
1992 wait_event();
1993 wait_event_interruptible();
1994 wait_event_interruptible_exclusive();
1995 wait_event_interruptible_timeout();
1996 wait_event_killable();
1997 wait_event_timeout();
1998 wait_on_bit();
1999 wait_on_bit_lock();
2000
2001
2002 Secondly, code that performs a wake up normally follows something like this:
2003
2004 event_indicated = 1;
2005 wake_up(&event_wait_queue);
2006
2007 or:
2008
2009 event_indicated = 1;
2010 wake_up_process(event_daemon);
2011
2012 A write memory barrier is implied by wake_up() and co. if and only if they wake
2013 something up. The barrier occurs before the task state is cleared, and so sits
2014 between the STORE to indicate the event and the STORE to set TASK_RUNNING:
2015
2016 CPU 1 CPU 2
2017 =============================== ===============================
2018 set_current_state(); STORE event_indicated
2019 set_mb(); wake_up();
2020 STORE current->state <write barrier>
2021 <general barrier> STORE current->state
2022 LOAD event_indicated
2023
2024 To repeat, this write memory barrier is present if and only if something
2025 is actually awakened. To see this, consider the following sequence of
2026 events, where X and Y are both initially zero:
2027
2028 CPU 1 CPU 2
2029 =============================== ===============================
2030 X = 1; STORE event_indicated
2031 smp_mb(); wake_up();
2032 Y = 1; wait_event(wq, Y == 1);
2033 wake_up(); load from Y sees 1, no memory barrier
2034 load from X might see 0
2035
2036 In contrast, if a wakeup does occur, CPU 2's load from X would be guaranteed
2037 to see 1.
2038
2039 The available waker functions include:
2040
2041 complete();
2042 wake_up();
2043 wake_up_all();
2044 wake_up_bit();
2045 wake_up_interruptible();
2046 wake_up_interruptible_all();
2047 wake_up_interruptible_nr();
2048 wake_up_interruptible_poll();
2049 wake_up_interruptible_sync();
2050 wake_up_interruptible_sync_poll();
2051 wake_up_locked();
2052 wake_up_locked_poll();
2053 wake_up_nr();
2054 wake_up_poll();
2055 wake_up_process();
2056
2057
2058 [!] Note that the memory barriers implied by the sleeper and the waker do _not_
2059 order multiple stores before the wake-up with respect to loads of those stored
2060 values after the sleeper has called set_current_state(). For instance, if the
2061 sleeper does:
2062
2063 set_current_state(TASK_INTERRUPTIBLE);
2064 if (event_indicated)
2065 break;
2066 __set_current_state(TASK_RUNNING);
2067 do_something(my_data);
2068
2069 and the waker does:
2070
2071 my_data = value;
2072 event_indicated = 1;
2073 wake_up(&event_wait_queue);
2074
2075 there's no guarantee that the change to event_indicated will be perceived by
2076 the sleeper as coming after the change to my_data. In such a circumstance, the
2077 code on both sides must interpolate its own memory barriers between the
2078 separate data accesses. Thus the above sleeper ought to do:
2079
2080 set_current_state(TASK_INTERRUPTIBLE);
2081 if (event_indicated) {
2082 smp_rmb();
2083 do_something(my_data);
2084 }
2085
2086 and the waker should do:
2087
2088 my_data = value;
2089 smp_wmb();
2090 event_indicated = 1;
2091 wake_up(&event_wait_queue);
2092
2093
2094 MISCELLANEOUS FUNCTIONS
2095 -----------------------
2096
2097 Other functions that imply barriers:
2098
2099 (*) schedule() and similar imply full memory barriers.
2100
2101
2102 ===================================
2103 INTER-CPU ACQUIRING BARRIER EFFECTS
2104 ===================================
2105
2106 On SMP systems locking primitives give a more substantial form of barrier: one
2107 that does affect memory access ordering on other CPUs, within the context of
2108 conflict on any particular lock.
2109
2110
2111 ACQUIRES VS MEMORY ACCESSES
2112 ---------------------------
2113
2114 Consider the following: the system has a pair of spinlocks (M) and (Q), and
2115 three CPUs; then should the following sequence of events occur:
2116
2117 CPU 1 CPU 2
2118 =============================== ===============================
2119 ACCESS_ONCE(*A) = a; ACCESS_ONCE(*E) = e;
2120 ACQUIRE M ACQUIRE Q
2121 ACCESS_ONCE(*B) = b; ACCESS_ONCE(*F) = f;
2122 ACCESS_ONCE(*C) = c; ACCESS_ONCE(*G) = g;
2123 RELEASE M RELEASE Q
2124 ACCESS_ONCE(*D) = d; ACCESS_ONCE(*H) = h;
2125
2126 Then there is no guarantee as to what order CPU 3 will see the accesses to *A
2127 through *H occur in, other than the constraints imposed by the separate locks
2128 on the separate CPUs. It might, for example, see:
2129
2130 *E, ACQUIRE M, ACQUIRE Q, *G, *C, *F, *A, *B, RELEASE Q, *D, *H, RELEASE M
2131
2132 But it won't see any of:
2133
2134 *B, *C or *D preceding ACQUIRE M
2135 *A, *B or *C following RELEASE M
2136 *F, *G or *H preceding ACQUIRE Q
2137 *E, *F or *G following RELEASE Q
2138
2139
2140 However, if the following occurs:
2141
2142 CPU 1 CPU 2
2143 =============================== ===============================
2144 ACCESS_ONCE(*A) = a;
2145 ACQUIRE M [1]
2146 ACCESS_ONCE(*B) = b;
2147 ACCESS_ONCE(*C) = c;
2148 RELEASE M [1]
2149 ACCESS_ONCE(*D) = d; ACCESS_ONCE(*E) = e;
2150 ACQUIRE M [2]
2151 smp_mb__after_unlock_lock();
2152 ACCESS_ONCE(*F) = f;
2153 ACCESS_ONCE(*G) = g;
2154 RELEASE M [2]
2155 ACCESS_ONCE(*H) = h;
2156
2157 CPU 3 might see:
2158
2159 *E, ACQUIRE M [1], *C, *B, *A, RELEASE M [1],
2160 ACQUIRE M [2], *H, *F, *G, RELEASE M [2], *D
2161
2162 But assuming CPU 1 gets the lock first, CPU 3 won't see any of:
2163
2164 *B, *C, *D, *F, *G or *H preceding ACQUIRE M [1]
2165 *A, *B or *C following RELEASE M [1]
2166 *F, *G or *H preceding ACQUIRE M [2]
2167 *A, *B, *C, *E, *F or *G following RELEASE M [2]
2168
2169 Note that the smp_mb__after_unlock_lock() is critically important
2170 here: Without it CPU 3 might see some of the above orderings.
2171 Without smp_mb__after_unlock_lock(), the accesses are not guaranteed
2172 to be seen in order unless CPU 3 holds lock M.
2173
2174
2175 ACQUIRES VS I/O ACCESSES
2176 ------------------------
2177
2178 Under certain circumstances (especially involving NUMA), I/O accesses within
2179 two spinlocked sections on two different CPUs may be seen as interleaved by the
2180 PCI bridge, because the PCI bridge does not necessarily participate in the
2181 cache-coherence protocol, and is therefore incapable of issuing the required
2182 read memory barriers.
2183
2184 For example:
2185
2186 CPU 1 CPU 2
2187 =============================== ===============================
2188 spin_lock(Q)
2189 writel(0, ADDR)
2190 writel(1, DATA);
2191 spin_unlock(Q);
2192 spin_lock(Q);
2193 writel(4, ADDR);
2194 writel(5, DATA);
2195 spin_unlock(Q);
2196
2197 may be seen by the PCI bridge as follows:
2198
2199 STORE *ADDR = 0, STORE *ADDR = 4, STORE *DATA = 1, STORE *DATA = 5
2200
2201 which would probably cause the hardware to malfunction.
2202
2203
2204 What is necessary here is to intervene with an mmiowb() before dropping the
2205 spinlock, for example:
2206
2207 CPU 1 CPU 2
2208 =============================== ===============================
2209 spin_lock(Q)
2210 writel(0, ADDR)
2211 writel(1, DATA);
2212 mmiowb();
2213 spin_unlock(Q);
2214 spin_lock(Q);
2215 writel(4, ADDR);
2216 writel(5, DATA);
2217 mmiowb();
2218 spin_unlock(Q);
2219
2220 this will ensure that the two stores issued on CPU 1 appear at the PCI bridge
2221 before either of the stores issued on CPU 2.
2222
2223
2224 Furthermore, following a store by a load from the same device obviates the need
2225 for the mmiowb(), because the load forces the store to complete before the load
2226 is performed:
2227
2228 CPU 1 CPU 2
2229 =============================== ===============================
2230 spin_lock(Q)
2231 writel(0, ADDR)
2232 a = readl(DATA);
2233 spin_unlock(Q);
2234 spin_lock(Q);
2235 writel(4, ADDR);
2236 b = readl(DATA);
2237 spin_unlock(Q);
2238
2239
2240 See Documentation/DocBook/deviceiobook.tmpl for more information.
2241
2242
2243 =================================
2244 WHERE ARE MEMORY BARRIERS NEEDED?
2245 =================================
2246
2247 Under normal operation, memory operation reordering is generally not going to
2248 be a problem as a single-threaded linear piece of code will still appear to
2249 work correctly, even if it's in an SMP kernel. There are, however, four
2250 circumstances in which reordering definitely _could_ be a problem:
2251
2252 (*) Interprocessor interaction.
2253
2254 (*) Atomic operations.
2255
2256 (*) Accessing devices.
2257
2258 (*) Interrupts.
2259
2260
2261 INTERPROCESSOR INTERACTION
2262 --------------------------
2263
2264 When there's a system with more than one processor, more than one CPU in the
2265 system may be working on the same data set at the same time. This can cause
2266 synchronisation problems, and the usual way of dealing with them is to use
2267 locks. Locks, however, are quite expensive, and so it may be preferable to
2268 operate without the use of a lock if at all possible. In such a case
2269 operations that affect both CPUs may have to be carefully ordered to prevent
2270 a malfunction.
2271
2272 Consider, for example, the R/W semaphore slow path. Here a waiting process is
2273 queued on the semaphore, by virtue of it having a piece of its stack linked to
2274 the semaphore's list of waiting processes:
2275
2276 struct rw_semaphore {
2277 ...
2278 spinlock_t lock;
2279 struct list_head waiters;
2280 };
2281
2282 struct rwsem_waiter {
2283 struct list_head list;
2284 struct task_struct *task;
2285 };
2286
2287 To wake up a particular waiter, the up_read() or up_write() functions have to:
2288
2289 (1) read the next pointer from this waiter's record to know as to where the
2290 next waiter record is;
2291
2292 (2) read the pointer to the waiter's task structure;
2293
2294 (3) clear the task pointer to tell the waiter it has been given the semaphore;
2295
2296 (4) call wake_up_process() on the task; and
2297
2298 (5) release the reference held on the waiter's task struct.
2299
2300 In other words, it has to perform this sequence of events:
2301
2302 LOAD waiter->list.next;
2303 LOAD waiter->task;
2304 STORE waiter->task;
2305 CALL wakeup
2306 RELEASE task
2307
2308 and if any of these steps occur out of order, then the whole thing may
2309 malfunction.
2310
2311 Once it has queued itself and dropped the semaphore lock, the waiter does not
2312 get the lock again; it instead just waits for its task pointer to be cleared
2313 before proceeding. Since the record is on the waiter's stack, this means that
2314 if the task pointer is cleared _before_ the next pointer in the list is read,
2315 another CPU might start processing the waiter and might clobber the waiter's
2316 stack before the up*() function has a chance to read the next pointer.
2317
2318 Consider then what might happen to the above sequence of events:
2319
2320 CPU 1 CPU 2
2321 =============================== ===============================
2322 down_xxx()
2323 Queue waiter
2324 Sleep
2325 up_yyy()
2326 LOAD waiter->task;
2327 STORE waiter->task;
2328 Woken up by other event
2329 <preempt>
2330 Resume processing
2331 down_xxx() returns
2332 call foo()
2333 foo() clobbers *waiter
2334 </preempt>
2335 LOAD waiter->list.next;
2336 --- OOPS ---
2337
2338 This could be dealt with using the semaphore lock, but then the down_xxx()
2339 function has to needlessly get the spinlock again after being woken up.
2340
2341 The way to deal with this is to insert a general SMP memory barrier:
2342
2343 LOAD waiter->list.next;
2344 LOAD waiter->task;
2345 smp_mb();
2346 STORE waiter->task;
2347 CALL wakeup
2348 RELEASE task
2349
2350 In this case, the barrier makes a guarantee that all memory accesses before the
2351 barrier will appear to happen before all the memory accesses after the barrier
2352 with respect to the other CPUs on the system. It does _not_ guarantee that all
2353 the memory accesses before the barrier will be complete by the time the barrier
2354 instruction itself is complete.
2355
2356 On a UP system - where this wouldn't be a problem - the smp_mb() is just a
2357 compiler barrier, thus making sure the compiler emits the instructions in the
2358 right order without actually intervening in the CPU. Since there's only one
2359 CPU, that CPU's dependency ordering logic will take care of everything else.
2360
2361
2362 ATOMIC OPERATIONS
2363 -----------------
2364
2365 Whilst they are technically interprocessor interaction considerations, atomic
2366 operations are noted specially as some of them imply full memory barriers and
2367 some don't, but they're very heavily relied on as a group throughout the
2368 kernel.
2369
2370 Any atomic operation that modifies some state in memory and returns information
2371 about the state (old or new) implies an SMP-conditional general memory barrier
2372 (smp_mb()) on each side of the actual operation (with the exception of
2373 explicit lock operations, described later). These include:
2374
2375 xchg();
2376 cmpxchg();
2377 atomic_xchg(); atomic_long_xchg();
2378 atomic_cmpxchg(); atomic_long_cmpxchg();
2379 atomic_inc_return(); atomic_long_inc_return();
2380 atomic_dec_return(); atomic_long_dec_return();
2381 atomic_add_return(); atomic_long_add_return();
2382 atomic_sub_return(); atomic_long_sub_return();
2383 atomic_inc_and_test(); atomic_long_inc_and_test();
2384 atomic_dec_and_test(); atomic_long_dec_and_test();
2385 atomic_sub_and_test(); atomic_long_sub_and_test();
2386 atomic_add_negative(); atomic_long_add_negative();
2387 test_and_set_bit();
2388 test_and_clear_bit();
2389 test_and_change_bit();
2390
2391 /* when succeeds (returns 1) */
2392 atomic_add_unless(); atomic_long_add_unless();
2393
2394 These are used for such things as implementing ACQUIRE-class and RELEASE-class
2395 operations and adjusting reference counters towards object destruction, and as
2396 such the implicit memory barrier effects are necessary.
2397
2398
2399 The following operations are potential problems as they do _not_ imply memory
2400 barriers, but might be used for implementing such things as RELEASE-class
2401 operations:
2402
2403 atomic_set();
2404 set_bit();
2405 clear_bit();
2406 change_bit();
2407
2408 With these the appropriate explicit memory barrier should be used if necessary
2409 (smp_mb__before_atomic() for instance).
2410
2411
2412 The following also do _not_ imply memory barriers, and so may require explicit
2413 memory barriers under some circumstances (smp_mb__before_atomic() for
2414 instance):
2415
2416 atomic_add();
2417 atomic_sub();
2418 atomic_inc();
2419 atomic_dec();
2420
2421 If they're used for statistics generation, then they probably don't need memory
2422 barriers, unless there's a coupling between statistical data.
2423
2424 If they're used for reference counting on an object to control its lifetime,
2425 they probably don't need memory barriers because either the reference count
2426 will be adjusted inside a locked section, or the caller will already hold
2427 sufficient references to make the lock, and thus a memory barrier unnecessary.
2428
2429 If they're used for constructing a lock of some description, then they probably
2430 do need memory barriers as a lock primitive generally has to do things in a
2431 specific order.
2432
2433 Basically, each usage case has to be carefully considered as to whether memory
2434 barriers are needed or not.
2435
2436 The following operations are special locking primitives:
2437
2438 test_and_set_bit_lock();
2439 clear_bit_unlock();
2440 __clear_bit_unlock();
2441
2442 These implement ACQUIRE-class and RELEASE-class operations. These should be used in
2443 preference to other operations when implementing locking primitives, because
2444 their implementations can be optimised on many architectures.
2445
2446 [!] Note that special memory barrier primitives are available for these
2447 situations because on some CPUs the atomic instructions used imply full memory
2448 barriers, and so barrier instructions are superfluous in conjunction with them,
2449 and in such cases the special barrier primitives will be no-ops.
2450
2451 See Documentation/atomic_ops.txt for more information.
2452
2453
2454 ACCESSING DEVICES
2455 -----------------
2456
2457 Many devices can be memory mapped, and so appear to the CPU as if they're just
2458 a set of memory locations. To control such a device, the driver usually has to
2459 make the right memory accesses in exactly the right order.
2460
2461 However, having a clever CPU or a clever compiler creates a potential problem
2462 in that the carefully sequenced accesses in the driver code won't reach the
2463 device in the requisite order if the CPU or the compiler thinks it is more
2464 efficient to reorder, combine or merge accesses - something that would cause
2465 the device to malfunction.
2466
2467 Inside of the Linux kernel, I/O should be done through the appropriate accessor
2468 routines - such as inb() or writel() - which know how to make such accesses
2469 appropriately sequential. Whilst this, for the most part, renders the explicit
2470 use of memory barriers unnecessary, there are a couple of situations where they
2471 might be needed:
2472
2473 (1) On some systems, I/O stores are not strongly ordered across all CPUs, and
2474 so for _all_ general drivers locks should be used and mmiowb() must be
2475 issued prior to unlocking the critical section.
2476
2477 (2) If the accessor functions are used to refer to an I/O memory window with
2478 relaxed memory access properties, then _mandatory_ memory barriers are
2479 required to enforce ordering.
2480
2481 See Documentation/DocBook/deviceiobook.tmpl for more information.
2482
2483
2484 INTERRUPTS
2485 ----------
2486
2487 A driver may be interrupted by its own interrupt service routine, and thus the
2488 two parts of the driver may interfere with each other's attempts to control or
2489 access the device.
2490
2491 This may be alleviated - at least in part - by disabling local interrupts (a
2492 form of locking), such that the critical operations are all contained within
2493 the interrupt-disabled section in the driver. Whilst the driver's interrupt
2494 routine is executing, the driver's core may not run on the same CPU, and its
2495 interrupt is not permitted to happen again until the current interrupt has been
2496 handled, thus the interrupt handler does not need to lock against that.
2497
2498 However, consider a driver that was talking to an ethernet card that sports an
2499 address register and a data register. If that driver's core talks to the card
2500 under interrupt-disablement and then the driver's interrupt handler is invoked:
2501
2502 LOCAL IRQ DISABLE
2503 writew(ADDR, 3);
2504 writew(DATA, y);
2505 LOCAL IRQ ENABLE
2506 <interrupt>
2507 writew(ADDR, 4);
2508 q = readw(DATA);
2509 </interrupt>
2510
2511 The store to the data register might happen after the second store to the
2512 address register if ordering rules are sufficiently relaxed:
2513
2514 STORE *ADDR = 3, STORE *ADDR = 4, STORE *DATA = y, q = LOAD *DATA
2515
2516
2517 If ordering rules are relaxed, it must be assumed that accesses done inside an
2518 interrupt disabled section may leak outside of it and may interleave with
2519 accesses performed in an interrupt - and vice versa - unless implicit or
2520 explicit barriers are used.
2521
2522 Normally this won't be a problem because the I/O accesses done inside such
2523 sections will include synchronous load operations on strictly ordered I/O
2524 registers that form implicit I/O barriers. If this isn't sufficient then an
2525 mmiowb() may need to be used explicitly.
2526
2527
2528 A similar situation may occur between an interrupt routine and two routines
2529 running on separate CPUs that communicate with each other. If such a case is
2530 likely, then interrupt-disabling locks should be used to guarantee ordering.
2531
2532
2533 ==========================
2534 KERNEL I/O BARRIER EFFECTS
2535 ==========================
2536
2537 When accessing I/O memory, drivers should use the appropriate accessor
2538 functions:
2539
2540 (*) inX(), outX():
2541
2542 These are intended to talk to I/O space rather than memory space, but
2543 that's primarily a CPU-specific concept. The i386 and x86_64 processors do
2544 indeed have special I/O space access cycles and instructions, but many
2545 CPUs don't have such a concept.
2546
2547 The PCI bus, amongst others, defines an I/O space concept which - on such
2548 CPUs as i386 and x86_64 - readily maps to the CPU's concept of I/O
2549 space. However, it may also be mapped as a virtual I/O space in the CPU's
2550 memory map, particularly on those CPUs that don't support alternate I/O
2551 spaces.
2552
2553 Accesses to this space may be fully synchronous (as on i386), but
2554 intermediary bridges (such as the PCI host bridge) may not fully honour
2555 that.
2556
2557 They are guaranteed to be fully ordered with respect to each other.
2558
2559 They are not guaranteed to be fully ordered with respect to other types of
2560 memory and I/O operation.
2561
2562 (*) readX(), writeX():
2563
2564 Whether these are guaranteed to be fully ordered and uncombined with
2565 respect to each other on the issuing CPU depends on the characteristics
2566 defined for the memory window through which they're accessing. On later
2567 i386 architecture machines, for example, this is controlled by way of the
2568 MTRR registers.
2569
2570 Ordinarily, these will be guaranteed to be fully ordered and uncombined,
2571 provided they're not accessing a prefetchable device.
2572
2573 However, intermediary hardware (such as a PCI bridge) may indulge in
2574 deferral if it so wishes; to flush a store, a load from the same location
2575 is preferred[*], but a load from the same device or from configuration
2576 space should suffice for PCI.
2577
2578 [*] NOTE! attempting to load from the same location as was written to may
2579 cause a malfunction - consider the 16550 Rx/Tx serial registers for
2580 example.
2581
2582 Used with prefetchable I/O memory, an mmiowb() barrier may be required to
2583 force stores to be ordered.
2584
2585 Please refer to the PCI specification for more information on interactions
2586 between PCI transactions.
2587
2588 (*) readX_relaxed(), writeX_relaxed()
2589
2590 These are similar to readX() and writeX(), but provide weaker memory
2591 ordering guarantees. Specifically, they do not guarantee ordering with
2592 respect to normal memory accesses (e.g. DMA buffers) nor do they guarantee
2593 ordering with respect to LOCK or UNLOCK operations. If the latter is
2594 required, an mmiowb() barrier can be used. Note that relaxed accesses to
2595 the same peripheral are guaranteed to be ordered with respect to each
2596 other.
2597
2598 (*) ioreadX(), iowriteX()
2599
2600 These will perform appropriately for the type of access they're actually
2601 doing, be it inX()/outX() or readX()/writeX().
2602
2603
2604 ========================================
2605 ASSUMED MINIMUM EXECUTION ORDERING MODEL
2606 ========================================
2607
2608 It has to be assumed that the conceptual CPU is weakly-ordered but that it will
2609 maintain the appearance of program causality with respect to itself. Some CPUs
2610 (such as i386 or x86_64) are more constrained than others (such as powerpc or
2611 frv), and so the most relaxed case (namely DEC Alpha) must be assumed outside
2612 of arch-specific code.
2613
2614 This means that it must be considered that the CPU will execute its instruction
2615 stream in any order it feels like - or even in parallel - provided that if an
2616 instruction in the stream depends on an earlier instruction, then that
2617 earlier instruction must be sufficiently complete[*] before the later
2618 instruction may proceed; in other words: provided that the appearance of
2619 causality is maintained.
2620
2621 [*] Some instructions have more than one effect - such as changing the
2622 condition codes, changing registers or changing memory - and different
2623 instructions may depend on different effects.
2624
2625 A CPU may also discard any instruction sequence that winds up having no
2626 ultimate effect. For example, if two adjacent instructions both load an
2627 immediate value into the same register, the first may be discarded.
2628
2629
2630 Similarly, it has to be assumed that compiler might reorder the instruction
2631 stream in any way it sees fit, again provided the appearance of causality is
2632 maintained.
2633
2634
2635 ============================
2636 THE EFFECTS OF THE CPU CACHE
2637 ============================
2638
2639 The way cached memory operations are perceived across the system is affected to
2640 a certain extent by the caches that lie between CPUs and memory, and by the
2641 memory coherence system that maintains the consistency of state in the system.
2642
2643 As far as the way a CPU interacts with another part of the system through the
2644 caches goes, the memory system has to include the CPU's caches, and memory
2645 barriers for the most part act at the interface between the CPU and its cache
2646 (memory barriers logically act on the dotted line in the following diagram):
2647
2648 <--- CPU ---> : <----------- Memory ----------->
2649 :
2650 +--------+ +--------+ : +--------+ +-----------+
2651 | | | | : | | | | +--------+
2652 | CPU | | Memory | : | CPU | | | | |
2653 | Core |--->| Access |----->| Cache |<-->| | | |
2654 | | | Queue | : | | | |--->| Memory |
2655 | | | | : | | | | | |
2656 +--------+ +--------+ : +--------+ | | | |
2657 : | Cache | +--------+
2658 : | Coherency |
2659 : | Mechanism | +--------+
2660 +--------+ +--------+ : +--------+ | | | |
2661 | | | | : | | | | | |
2662 | CPU | | Memory | : | CPU | | |--->| Device |
2663 | Core |--->| Access |----->| Cache |<-->| | | |
2664 | | | Queue | : | | | | | |
2665 | | | | : | | | | +--------+
2666 +--------+ +--------+ : +--------+ +-----------+
2667 :
2668 :
2669
2670 Although any particular load or store may not actually appear outside of the
2671 CPU that issued it since it may have been satisfied within the CPU's own cache,
2672 it will still appear as if the full memory access had taken place as far as the
2673 other CPUs are concerned since the cache coherency mechanisms will migrate the
2674 cacheline over to the accessing CPU and propagate the effects upon conflict.
2675
2676 The CPU core may execute instructions in any order it deems fit, provided the
2677 expected program causality appears to be maintained. Some of the instructions
2678 generate load and store operations which then go into the queue of memory
2679 accesses to be performed. The core may place these in the queue in any order
2680 it wishes, and continue execution until it is forced to wait for an instruction
2681 to complete.
2682
2683 What memory barriers are concerned with is controlling the order in which
2684 accesses cross from the CPU side of things to the memory side of things, and
2685 the order in which the effects are perceived to happen by the other observers
2686 in the system.
2687
2688 [!] Memory barriers are _not_ needed within a given CPU, as CPUs always see
2689 their own loads and stores as if they had happened in program order.
2690
2691 [!] MMIO or other device accesses may bypass the cache system. This depends on
2692 the properties of the memory window through which devices are accessed and/or
2693 the use of any special device communication instructions the CPU may have.
2694
2695
2696 CACHE COHERENCY
2697 ---------------
2698
2699 Life isn't quite as simple as it may appear above, however: for while the
2700 caches are expected to be coherent, there's no guarantee that that coherency
2701 will be ordered. This means that whilst changes made on one CPU will
2702 eventually become visible on all CPUs, there's no guarantee that they will
2703 become apparent in the same order on those other CPUs.
2704
2705
2706 Consider dealing with a system that has a pair of CPUs (1 & 2), each of which
2707 has a pair of parallel data caches (CPU 1 has A/B, and CPU 2 has C/D):
2708
2709 :
2710 : +--------+
2711 : +---------+ | |
2712 +--------+ : +--->| Cache A |<------->| |
2713 | | : | +---------+ | |
2714 | CPU 1 |<---+ | |
2715 | | : | +---------+ | |
2716 +--------+ : +--->| Cache B |<------->| |
2717 : +---------+ | |
2718 : | Memory |
2719 : +---------+ | System |
2720 +--------+ : +--->| Cache C |<------->| |
2721 | | : | +---------+ | |
2722 | CPU 2 |<---+ | |
2723 | | : | +---------+ | |
2724 +--------+ : +--->| Cache D |<------->| |
2725 : +---------+ | |
2726 : +--------+
2727 :
2728
2729 Imagine the system has the following properties:
2730
2731 (*) an odd-numbered cache line may be in cache A, cache C or it may still be
2732 resident in memory;
2733
2734 (*) an even-numbered cache line may be in cache B, cache D or it may still be
2735 resident in memory;
2736
2737 (*) whilst the CPU core is interrogating one cache, the other cache may be
2738 making use of the bus to access the rest of the system - perhaps to
2739 displace a dirty cacheline or to do a speculative load;
2740
2741 (*) each cache has a queue of operations that need to be applied to that cache
2742 to maintain coherency with the rest of the system;
2743
2744 (*) the coherency queue is not flushed by normal loads to lines already
2745 present in the cache, even though the contents of the queue may
2746 potentially affect those loads.
2747
2748 Imagine, then, that two writes are made on the first CPU, with a write barrier
2749 between them to guarantee that they will appear to reach that CPU's caches in
2750 the requisite order:
2751
2752 CPU 1 CPU 2 COMMENT
2753 =============== =============== =======================================
2754 u == 0, v == 1 and p == &u, q == &u
2755 v = 2;
2756 smp_wmb(); Make sure change to v is visible before
2757 change to p
2758 <A:modify v=2> v is now in cache A exclusively
2759 p = &v;
2760 <B:modify p=&v> p is now in cache B exclusively
2761
2762 The write memory barrier forces the other CPUs in the system to perceive that
2763 the local CPU's caches have apparently been updated in the correct order. But
2764 now imagine that the second CPU wants to read those values:
2765
2766 CPU 1 CPU 2 COMMENT
2767 =============== =============== =======================================
2768 ...
2769 q = p;
2770 x = *q;
2771
2772 The above pair of reads may then fail to happen in the expected order, as the
2773 cacheline holding p may get updated in one of the second CPU's caches whilst
2774 the update to the cacheline holding v is delayed in the other of the second
2775 CPU's caches by some other cache event:
2776
2777 CPU 1 CPU 2 COMMENT
2778 =============== =============== =======================================
2779 u == 0, v == 1 and p == &u, q == &u
2780 v = 2;
2781 smp_wmb();
2782 <A:modify v=2> <C:busy>
2783 <C:queue v=2>
2784 p = &v; q = p;
2785 <D:request p>
2786 <B:modify p=&v> <D:commit p=&v>
2787 <D:read p>
2788 x = *q;
2789 <C:read *q> Reads from v before v updated in cache
2790 <C:unbusy>
2791 <C:commit v=2>
2792
2793 Basically, whilst both cachelines will be updated on CPU 2 eventually, there's
2794 no guarantee that, without intervention, the order of update will be the same
2795 as that committed on CPU 1.
2796
2797
2798 To intervene, we need to interpolate a data dependency barrier or a read
2799 barrier between the loads. This will force the cache to commit its coherency
2800 queue before processing any further requests:
2801
2802 CPU 1 CPU 2 COMMENT
2803 =============== =============== =======================================
2804 u == 0, v == 1 and p == &u, q == &u
2805 v = 2;
2806 smp_wmb();
2807 <A:modify v=2> <C:busy>
2808 <C:queue v=2>
2809 p = &v; q = p;
2810 <D:request p>
2811 <B:modify p=&v> <D:commit p=&v>
2812 <D:read p>
2813 smp_read_barrier_depends()
2814 <C:unbusy>
2815 <C:commit v=2>
2816 x = *q;
2817 <C:read *q> Reads from v after v updated in cache
2818
2819
2820 This sort of problem can be encountered on DEC Alpha processors as they have a
2821 split cache that improves performance by making better use of the data bus.
2822 Whilst most CPUs do imply a data dependency barrier on the read when a memory
2823 access depends on a read, not all do, so it may not be relied on.
2824
2825 Other CPUs may also have split caches, but must coordinate between the various
2826 cachelets for normal memory accesses. The semantics of the Alpha removes the
2827 need for coordination in the absence of memory barriers.
2828
2829
2830 CACHE COHERENCY VS DMA
2831 ----------------------
2832
2833 Not all systems maintain cache coherency with respect to devices doing DMA. In
2834 such cases, a device attempting DMA may obtain stale data from RAM because
2835 dirty cache lines may be resident in the caches of various CPUs, and may not
2836 have been written back to RAM yet. To deal with this, the appropriate part of
2837 the kernel must flush the overlapping bits of cache on each CPU (and maybe
2838 invalidate them as well).
2839
2840 In addition, the data DMA'd to RAM by a device may be overwritten by dirty
2841 cache lines being written back to RAM from a CPU's cache after the device has
2842 installed its own data, or cache lines present in the CPU's cache may simply
2843 obscure the fact that RAM has been updated, until at such time as the cacheline
2844 is discarded from the CPU's cache and reloaded. To deal with this, the
2845 appropriate part of the kernel must invalidate the overlapping bits of the
2846 cache on each CPU.
2847
2848 See Documentation/cachetlb.txt for more information on cache management.
2849
2850
2851 CACHE COHERENCY VS MMIO
2852 -----------------------
2853
2854 Memory mapped I/O usually takes place through memory locations that are part of
2855 a window in the CPU's memory space that has different properties assigned than
2856 the usual RAM directed window.
2857
2858 Amongst these properties is usually the fact that such accesses bypass the
2859 caching entirely and go directly to the device buses. This means MMIO accesses
2860 may, in effect, overtake accesses to cached memory that were emitted earlier.
2861 A memory barrier isn't sufficient in such a case, but rather the cache must be
2862 flushed between the cached memory write and the MMIO access if the two are in
2863 any way dependent.
2864
2865
2866 =========================
2867 THE THINGS CPUS GET UP TO
2868 =========================
2869
2870 A programmer might take it for granted that the CPU will perform memory
2871 operations in exactly the order specified, so that if the CPU is, for example,
2872 given the following piece of code to execute:
2873
2874 a = ACCESS_ONCE(*A);
2875 ACCESS_ONCE(*B) = b;
2876 c = ACCESS_ONCE(*C);
2877 d = ACCESS_ONCE(*D);
2878 ACCESS_ONCE(*E) = e;
2879
2880 they would then expect that the CPU will complete the memory operation for each
2881 instruction before moving on to the next one, leading to a definite sequence of
2882 operations as seen by external observers in the system:
2883
2884 LOAD *A, STORE *B, LOAD *C, LOAD *D, STORE *E.
2885
2886
2887 Reality is, of course, much messier. With many CPUs and compilers, the above
2888 assumption doesn't hold because:
2889
2890 (*) loads are more likely to need to be completed immediately to permit
2891 execution progress, whereas stores can often be deferred without a
2892 problem;
2893
2894 (*) loads may be done speculatively, and the result discarded should it prove
2895 to have been unnecessary;
2896
2897 (*) loads may be done speculatively, leading to the result having been fetched
2898 at the wrong time in the expected sequence of events;
2899
2900 (*) the order of the memory accesses may be rearranged to promote better use
2901 of the CPU buses and caches;
2902
2903 (*) loads and stores may be combined to improve performance when talking to
2904 memory or I/O hardware that can do batched accesses of adjacent locations,
2905 thus cutting down on transaction setup costs (memory and PCI devices may
2906 both be able to do this); and
2907
2908 (*) the CPU's data cache may affect the ordering, and whilst cache-coherency
2909 mechanisms may alleviate this - once the store has actually hit the cache
2910 - there's no guarantee that the coherency management will be propagated in
2911 order to other CPUs.
2912
2913 So what another CPU, say, might actually observe from the above piece of code
2914 is:
2915
2916 LOAD *A, ..., LOAD {*C,*D}, STORE *E, STORE *B
2917
2918 (Where "LOAD {*C,*D}" is a combined load)
2919
2920
2921 However, it is guaranteed that a CPU will be self-consistent: it will see its
2922 _own_ accesses appear to be correctly ordered, without the need for a memory
2923 barrier. For instance with the following code:
2924
2925 U = ACCESS_ONCE(*A);
2926 ACCESS_ONCE(*A) = V;
2927 ACCESS_ONCE(*A) = W;
2928 X = ACCESS_ONCE(*A);
2929 ACCESS_ONCE(*A) = Y;
2930 Z = ACCESS_ONCE(*A);
2931
2932 and assuming no intervention by an external influence, it can be assumed that
2933 the final result will appear to be:
2934
2935 U == the original value of *A
2936 X == W
2937 Z == Y
2938 *A == Y
2939
2940 The code above may cause the CPU to generate the full sequence of memory
2941 accesses:
2942
2943 U=LOAD *A, STORE *A=V, STORE *A=W, X=LOAD *A, STORE *A=Y, Z=LOAD *A
2944
2945 in that order, but, without intervention, the sequence may have almost any
2946 combination of elements combined or discarded, provided the program's view of
2947 the world remains consistent. Note that ACCESS_ONCE() is -not- optional
2948 in the above example, as there are architectures where a given CPU might
2949 reorder successive loads to the same location. On such architectures,
2950 ACCESS_ONCE() does whatever is necessary to prevent this, for example, on
2951 Itanium the volatile casts used by ACCESS_ONCE() cause GCC to emit the
2952 special ld.acq and st.rel instructions that prevent such reordering.
2953
2954 The compiler may also combine, discard or defer elements of the sequence before
2955 the CPU even sees them.
2956
2957 For instance:
2958
2959 *A = V;
2960 *A = W;
2961
2962 may be reduced to:
2963
2964 *A = W;
2965
2966 since, without either a write barrier or an ACCESS_ONCE(), it can be
2967 assumed that the effect of the storage of V to *A is lost. Similarly:
2968
2969 *A = Y;
2970 Z = *A;
2971
2972 may, without a memory barrier or an ACCESS_ONCE(), be reduced to:
2973
2974 *A = Y;
2975 Z = Y;
2976
2977 and the LOAD operation never appear outside of the CPU.
2978
2979
2980 AND THEN THERE'S THE ALPHA
2981 --------------------------
2982
2983 The DEC Alpha CPU is one of the most relaxed CPUs there is. Not only that,
2984 some versions of the Alpha CPU have a split data cache, permitting them to have
2985 two semantically-related cache lines updated at separate times. This is where
2986 the data dependency barrier really becomes necessary as this synchronises both
2987 caches with the memory coherence system, thus making it seem like pointer
2988 changes vs new data occur in the right order.
2989
2990 The Alpha defines the Linux kernel's memory barrier model.
2991
2992 See the subsection on "Cache Coherency" above.
2993
2994
2995 ============
2996 EXAMPLE USES
2997 ============
2998
2999 CIRCULAR BUFFERS
3000 ----------------
3001
3002 Memory barriers can be used to implement circular buffering without the need
3003 of a lock to serialise the producer with the consumer. See:
3004
3005 Documentation/circular-buffers.txt
3006
3007 for details.
3008
3009
3010 ==========
3011 REFERENCES
3012 ==========
3013
3014 Alpha AXP Architecture Reference Manual, Second Edition (Sites & Witek,
3015 Digital Press)
3016 Chapter 5.2: Physical Address Space Characteristics
3017 Chapter 5.4: Caches and Write Buffers
3018 Chapter 5.5: Data Sharing
3019 Chapter 5.6: Read/Write Ordering
3020
3021 AMD64 Architecture Programmer's Manual Volume 2: System Programming
3022 Chapter 7.1: Memory-Access Ordering
3023 Chapter 7.4: Buffering and Combining Memory Writes
3024
3025 IA-32 Intel Architecture Software Developer's Manual, Volume 3:
3026 System Programming Guide
3027 Chapter 7.1: Locked Atomic Operations
3028 Chapter 7.2: Memory Ordering
3029 Chapter 7.4: Serializing Instructions
3030
3031 The SPARC Architecture Manual, Version 9
3032 Chapter 8: Memory Models
3033 Appendix D: Formal Specification of the Memory Models
3034 Appendix J: Programming with the Memory Models
3035
3036 UltraSPARC Programmer Reference Manual
3037 Chapter 5: Memory Accesses and Cacheability
3038 Chapter 15: Sparc-V9 Memory Models
3039
3040 UltraSPARC III Cu User's Manual
3041 Chapter 9: Memory Models
3042
3043 UltraSPARC IIIi Processor User's Manual
3044 Chapter 8: Memory Models
3045
3046 UltraSPARC Architecture 2005
3047 Chapter 9: Memory
3048 Appendix D: Formal Specifications of the Memory Models
3049
3050 UltraSPARC T1 Supplement to the UltraSPARC Architecture 2005
3051 Chapter 8: Memory Models
3052 Appendix F: Caches and Cache Coherency
3053
3054 Solaris Internals, Core Kernel Architecture, p63-68:
3055 Chapter 3.3: Hardware Considerations for Locks and
3056 Synchronization
3057
3058 Unix Systems for Modern Architectures, Symmetric Multiprocessing and Caching
3059 for Kernel Programmers:
3060 Chapter 13: Other Memory Models
3061
3062 Intel Itanium Architecture Software Developer's Manual: Volume 1:
3063 Section 2.6: Speculation
3064 Section 4.4: Memory Access