memory-barriers.txt 110.8 KB
Newer Older
1 2 3 4 5
			 ============================
			 LINUX KERNEL MEMORY BARRIERS
			 ============================

By: David Howells <dhowells@redhat.com>
6
    Paul E. McKenney <paulmck@linux.vnet.ibm.com>
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

Contents:

 (*) Abstract memory access model.

     - Device operations.
     - Guarantees.

 (*) What are memory barriers?

     - Varieties of memory barrier.
     - What may not be assumed about memory barriers?
     - Data dependency barriers.
     - Control dependencies.
     - SMP barrier pairing.
     - Examples of memory barrier sequences.
23
     - Read memory barriers vs load speculation.
24
     - Transitivity
25 26 27 28

 (*) Explicit kernel barriers.

     - Compiler barrier.
29
     - CPU memory barriers.
30 31 32 33 34 35
     - MMIO write barrier.

 (*) Implicit kernel memory barriers.

     - Locking functions.
     - Interrupt disabling functions.
36
     - Sleep and wake-up functions.
37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
     - Miscellaneous functions.

 (*) Inter-CPU locking barrier effects.

     - Locks vs memory accesses.
     - Locks vs I/O accesses.

 (*) Where are memory barriers needed?

     - Interprocessor interaction.
     - Atomic operations.
     - Accessing devices.
     - Interrupts.

 (*) Kernel I/O barrier effects.

 (*) Assumed minimum execution ordering model.

 (*) The effects of the cpu cache.

     - Cache coherency.
     - Cache coherency vs DMA.
     - Cache coherency vs MMIO.

 (*) The things CPUs get up to.

     - And then there's the Alpha.

65 66 67 68
 (*) Example uses.

     - Circular buffers.

69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117
 (*) References.


============================
ABSTRACT MEMORY ACCESS MODEL
============================

Consider the following abstract model of the system:

		            :                :
		            :                :
		            :                :
		+-------+   :   +--------+   :   +-------+
		|       |   :   |        |   :   |       |
		|       |   :   |        |   :   |       |
		| CPU 1 |<----->| Memory |<----->| CPU 2 |
		|       |   :   |        |   :   |       |
		|       |   :   |        |   :   |       |
		+-------+   :   +--------+   :   +-------+
		    ^       :       ^        :       ^
		    |       :       |        :       |
		    |       :       |        :       |
		    |       :       v        :       |
		    |       :   +--------+   :       |
		    |       :   |        |   :       |
		    |       :   |        |   :       |
		    +---------->| Device |<----------+
		            :   |        |   :
		            :   |        |   :
		            :   +--------+   :
		            :                :

Each CPU executes a program that generates memory access operations.  In the
abstract CPU, memory operation ordering is very relaxed, and a CPU may actually
perform the memory operations in any order it likes, provided program causality
appears to be maintained.  Similarly, the compiler may also arrange the
instructions it emits in any order it likes, provided it doesn't affect the
apparent operation of the program.

So in the above diagram, the effects of the memory operations performed by a
CPU are perceived by the rest of the system as the operations cross the
interface between the CPU and rest of the system (the dotted lines).


For example, consider the following sequence of events:

	CPU 1		CPU 2
	===============	===============
	{ A == 1; B == 2 }
118 119
	A = 3;		x = B;
	B = 4;		y = A;
120 121 122 123

The set of accesses as seen by the memory system in the middle can be arranged
in 24 different combinations:

124 125 126 127 128 129 130
	STORE A=3,	STORE B=4,	y=LOAD A->3,	x=LOAD B->4
	STORE A=3,	STORE B=4,	x=LOAD B->4,	y=LOAD A->3
	STORE A=3,	y=LOAD A->3,	STORE B=4,	x=LOAD B->4
	STORE A=3,	y=LOAD A->3,	x=LOAD B->2,	STORE B=4
	STORE A=3,	x=LOAD B->2,	STORE B=4,	y=LOAD A->3
	STORE A=3,	x=LOAD B->2,	y=LOAD A->3,	STORE B=4
	STORE B=4,	STORE A=3,	y=LOAD A->3,	x=LOAD B->4
131 132 133 134 135
	STORE B=4, ...
	...

and can thus result in four different combinations of values:

136 137 138 139
	x == 2, y == 1
	x == 2, y == 3
	x == 4, y == 1
	x == 4, y == 3
140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196


Furthermore, the stores committed by a CPU to the memory system may not be
perceived by the loads made by another CPU in the same order as the stores were
committed.


As a further example, consider this sequence of events:

	CPU 1		CPU 2
	===============	===============
	{ A == 1, B == 2, C = 3, P == &A, Q == &C }
	B = 4;		Q = P;
	P = &B		D = *Q;

There is an obvious data dependency here, as the value loaded into D depends on
the address retrieved from P by CPU 2.  At the end of the sequence, any of the
following results are possible:

	(Q == &A) and (D == 1)
	(Q == &B) and (D == 2)
	(Q == &B) and (D == 4)

Note that CPU 2 will never try and load C into D because the CPU will load P
into Q before issuing the load of *Q.


DEVICE OPERATIONS
-----------------

Some devices present their control interfaces as collections of memory
locations, but the order in which the control registers are accessed is very
important.  For instance, imagine an ethernet card with a set of internal
registers that are accessed through an address port register (A) and a data
port register (D).  To read internal register 5, the following code might then
be used:

	*A = 5;
	x = *D;

but this might show up as either of the following two sequences:

	STORE *A = 5, x = LOAD *D
	x = LOAD *D, STORE *A = 5

the second of which will almost certainly result in a malfunction, since it set
the address _after_ attempting to read the register.


GUARANTEES
----------

There are some minimal guarantees that may be expected of a CPU:

 (*) On any given CPU, dependent memory accesses will be issued in order, with
     respect to itself.  This means that for:

197
	WRITE_ONCE(Q, P); smp_read_barrier_depends(); D = READ_ONCE(*Q);
198 199 200 201 202

     the CPU will issue the following memory operations:

	Q = LOAD P, D = LOAD *Q

203
     and always in that order.  On most systems, smp_read_barrier_depends()
204 205 206 207
     does nothing, but it is required for DEC Alpha.  The READ_ONCE()
     and WRITE_ONCE() are required to prevent compiler mischief.  Please
     note that you should normally use something like rcu_dereference()
     instead of open-coding smp_read_barrier_depends().
208 209 210 211

 (*) Overlapping loads and stores within a particular CPU will appear to be
     ordered within that CPU.  This means that for:

212
	a = READ_ONCE(*X); WRITE_ONCE(*X, b);
213 214 215 216 217 218 219

     the CPU will only issue the following sequence of memory operations:

	a = LOAD *X, STORE *X = b

     And for:

220
	WRITE_ONCE(*X, c); d = READ_ONCE(*X);
221 222 223 224 225

     the CPU will only issue:

	STORE *X = c, d = LOAD *X

M
Matt LaPlante 已提交
226
     (Loads and stores overlap if they are targeted at overlapping pieces of
227 228 229 230
     memory).

And there are a number of things that _must_ or _must_not_ be assumed:

231 232 233 234 235
 (*) It _must_not_ be assumed that the compiler will do what you want
     with memory references that are not protected by READ_ONCE() and
     WRITE_ONCE().  Without them, the compiler is within its rights to
     do all sorts of "creative" transformations, which are covered in
     the Compiler Barrier section.
236

237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263
 (*) It _must_not_ be assumed that independent loads and stores will be issued
     in the order given.  This means that for:

	X = *A; Y = *B; *D = Z;

     we may get any of the following sequences:

	X = LOAD *A,  Y = LOAD *B,  STORE *D = Z
	X = LOAD *A,  STORE *D = Z, Y = LOAD *B
	Y = LOAD *B,  X = LOAD *A,  STORE *D = Z
	Y = LOAD *B,  STORE *D = Z, X = LOAD *A
	STORE *D = Z, X = LOAD *A,  Y = LOAD *B
	STORE *D = Z, Y = LOAD *B,  X = LOAD *A

 (*) It _must_ be assumed that overlapping memory accesses may be merged or
     discarded.  This means that for:

	X = *A; Y = *(A + 4);

     we may get any one of the following sequences:

	X = LOAD *A; Y = LOAD *(A + 4);
	Y = LOAD *(A + 4); X = LOAD *A;
	{X, Y} = LOAD {*A, *(A + 4) };

     And for:

264
	*A = X; *(A + 4) = Y;
265

266
     we may get any of:
267

268 269 270
	STORE *A = X; STORE *(A + 4) = Y;
	STORE *(A + 4) = Y; STORE *A = X;
	STORE {*A, *(A + 4) } = {X, Y};
271

272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315
And there are anti-guarantees:

 (*) These guarantees do not apply to bitfields, because compilers often
     generate code to modify these using non-atomic read-modify-write
     sequences.  Do not attempt to use bitfields to synchronize parallel
     algorithms.

 (*) Even in cases where bitfields are protected by locks, all fields
     in a given bitfield must be protected by one lock.  If two fields
     in a given bitfield are protected by different locks, the compiler's
     non-atomic read-modify-write sequences can cause an update to one
     field to corrupt the value of an adjacent field.

 (*) These guarantees apply only to properly aligned and sized scalar
     variables.  "Properly sized" currently means variables that are
     the same size as "char", "short", "int" and "long".  "Properly
     aligned" means the natural alignment, thus no constraints for
     "char", two-byte alignment for "short", four-byte alignment for
     "int", and either four-byte or eight-byte alignment for "long",
     on 32-bit and 64-bit systems, respectively.  Note that these
     guarantees were introduced into the C11 standard, so beware when
     using older pre-C11 compilers (for example, gcc 4.6).  The portion
     of the standard containing this guarantee is Section 3.14, which
     defines "memory location" as follows:

     	memory location
		either an object of scalar type, or a maximal sequence
		of adjacent bit-fields all having nonzero width

		NOTE 1: Two threads of execution can update and access
		separate memory locations without interfering with
		each other.

		NOTE 2: A bit-field and an adjacent non-bit-field member
		are in separate memory locations. The same applies
		to two bit-fields, if one is declared inside a nested
		structure declaration and the other is not, or if the two
		are separated by a zero-length bit-field declaration,
		or if they are separated by a non-bit-field member
		declaration. It is not safe to concurrently update two
		bit-fields in the same structure if all members declared
		between them are also bit-fields, no matter what the
		sizes of those intervening bit-fields happen to be.

316 317 318 319 320 321 322 323 324 325 326

=========================
WHAT ARE MEMORY BARRIERS?
=========================

As can be seen above, independent memory operations are effectively performed
in random order, but this can be a problem for CPU-CPU interaction and for I/O.
What is required is some way of intervening to instruct the compiler and the
CPU to restrict the order.

Memory barriers are such interventions.  They impose a perceived partial
327 328 329
ordering over the memory operations on either side of the barrier.

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,
331 332 333 334
deferral and combination of memory operations; speculative loads; speculative
branch prediction and various types of caching.  Memory barriers are used to
override or suppress these tricks, allowing the code to sanely control the
interaction of multiple CPUs and/or devices.
335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351


VARIETIES OF MEMORY BARRIER
---------------------------

Memory barriers come in four basic varieties:

 (1) Write (or store) memory barriers.

     A write memory barrier gives a guarantee that all the STORE operations
     specified before the barrier will appear to happen before all the STORE
     operations specified after the barrier with respect to the other
     components of the system.

     A write barrier is a partial ordering on stores only; it is not required
     to have any effect on loads.

352
     A CPU can be viewed as committing a sequence of store operations to the
353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414
     memory system as time progresses.  All stores before a write barrier will
     occur in the sequence _before_ all the stores after the write barrier.

     [!] Note that write barriers should normally be paired with read or data
     dependency barriers; see the "SMP barrier pairing" subsection.


 (2) Data dependency barriers.

     A data dependency barrier is a weaker form of read barrier.  In the case
     where two loads are performed such that the second depends on the result
     of the first (eg: the first load retrieves the address to which the second
     load will be directed), a data dependency barrier would be required to
     make sure that the target of the second load is updated before the address
     obtained by the first load is accessed.

     A data dependency barrier is a partial ordering on interdependent loads
     only; it is not required to have any effect on stores, independent loads
     or overlapping loads.

     As mentioned in (1), the other CPUs in the system can be viewed as
     committing sequences of stores to the memory system that the CPU being
     considered can then perceive.  A data dependency barrier issued by the CPU
     under consideration guarantees that for any load preceding it, if that
     load touches one of a sequence of stores from another CPU, then by the
     time the barrier completes, the effects of all the stores prior to that
     touched by the load will be perceptible to any loads issued after the data
     dependency barrier.

     See the "Examples of memory barrier sequences" subsection for diagrams
     showing the ordering constraints.

     [!] Note that the first load really has to have a _data_ dependency and
     not a control dependency.  If the address for the second load is dependent
     on the first load, but the dependency is through a conditional rather than
     actually loading the address itself, then it's a _control_ dependency and
     a full read barrier or better is required.  See the "Control dependencies"
     subsection for more information.

     [!] Note that data dependency barriers should normally be paired with
     write barriers; see the "SMP barrier pairing" subsection.


 (3) Read (or load) memory barriers.

     A read barrier is a data dependency barrier plus a guarantee that all the
     LOAD operations specified before the barrier will appear to happen before
     all the LOAD operations specified after the barrier with respect to the
     other components of the system.

     A read barrier is a partial ordering on loads only; it is not required to
     have any effect on stores.

     Read memory barriers imply data dependency barriers, and so can substitute
     for them.

     [!] Note that read barriers should normally be paired with write barriers;
     see the "SMP barrier pairing" subsection.


 (4) General memory barriers.

415 416 417 418 419 420
     A general memory barrier gives a guarantee that all the LOAD and STORE
     operations specified before the barrier will appear to happen before all
     the LOAD and STORE operations specified after the barrier with respect to
     the other components of the system.

     A general memory barrier is a partial ordering over both loads and stores.
421 422 423 424 425 426 427

     General memory barriers imply both read and write memory barriers, and so
     can substitute for either.


And a couple of implicit varieties:

428
 (5) ACQUIRE operations.
429 430

     This acts as a one-way permeable barrier.  It guarantees that all memory
431 432 433 434
     operations after the ACQUIRE operation will appear to happen after the
     ACQUIRE operation with respect to the other components of the system.
     ACQUIRE operations include LOCK operations and smp_load_acquire()
     operations.
435

436 437
     Memory operations that occur before an ACQUIRE operation may appear to
     happen after it completes.
438

439 440
     An ACQUIRE operation should almost always be paired with a RELEASE
     operation.
441 442


443
 (6) RELEASE operations.
444 445

     This also acts as a one-way permeable barrier.  It guarantees that all
446 447 448 449
     memory operations before the RELEASE operation will appear to happen
     before the RELEASE operation with respect to the other components of the
     system. RELEASE operations include UNLOCK operations and
     smp_store_release() operations.
450

451
     Memory operations that occur after a RELEASE operation may appear to
452 453
     happen before it completes.

454 455 456 457 458 459 460 461 462
     The use of ACQUIRE and RELEASE operations generally precludes the need
     for other sorts of memory barrier (but note the exceptions mentioned in
     the subsection "MMIO write barrier").  In addition, a RELEASE+ACQUIRE
     pair is -not- guaranteed to act as a full memory barrier.  However, after
     an ACQUIRE on a given variable, all memory accesses preceding any prior
     RELEASE on that same variable are guaranteed to be visible.  In other
     words, within a given variable's critical section, all accesses of all
     previous critical sections for that variable are guaranteed to have
     completed.
463

464 465
     This means that ACQUIRE acts as a minimal "acquire" operation and
     RELEASE acts as a minimal "release" operation.
466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493


Memory barriers are only required where there's a possibility of interaction
between two CPUs or between a CPU and a device.  If it can be guaranteed that
there won't be any such interaction in any particular piece of code, then
memory barriers are unnecessary in that piece of code.


Note that these are the _minimum_ guarantees.  Different architectures may give
more substantial guarantees, but they may _not_ be relied upon outside of arch
specific code.


WHAT MAY NOT BE ASSUMED ABOUT MEMORY BARRIERS?
----------------------------------------------

There are certain things that the Linux kernel memory barriers do not guarantee:

 (*) There is no guarantee that any of the memory accesses specified before a
     memory barrier will be _complete_ by the completion of a memory barrier
     instruction; the barrier can be considered to draw a line in that CPU's
     access queue that accesses of the appropriate type may not cross.

 (*) There is no guarantee that issuing a memory barrier on one CPU will have
     any direct effect on another CPU or any other hardware in the system.  The
     indirect effect will be the order in which the second CPU sees the effects
     of the first CPU's accesses occur, but see the next point:

494
 (*) There is no guarantee that a CPU will see the correct order of effects
495 496 497 498 499 500 501 502 503 504 505
     from a second CPU's accesses, even _if_ the second CPU uses a memory
     barrier, unless the first CPU _also_ uses a matching memory barrier (see
     the subsection on "SMP Barrier Pairing").

 (*) There is no guarantee that some intervening piece of off-the-CPU
     hardware[*] will not reorder the memory accesses.  CPU cache coherency
     mechanisms should propagate the indirect effects of a memory barrier
     between CPUs, but might not do so in order.

	[*] For information on bus mastering DMA and coherency please read:

506
	    Documentation/PCI/pci.txt
P
Paul Bolle 已提交
507
	    Documentation/DMA-API-HOWTO.txt
508 509 510 511 512 513 514 515 516 517
	    Documentation/DMA-API.txt


DATA DEPENDENCY BARRIERS
------------------------

The usage requirements of data dependency barriers are a little subtle, and
it's not always obvious that they're needed.  To illustrate, consider the
following sequence of events:

518 519
	CPU 1		      CPU 2
	===============	      ===============
520 521 522
	{ A == 1, B == 2, C = 3, P == &A, Q == &C }
	B = 4;
	<write barrier>
523 524
	WRITE_ONCE(P, &B)
			      Q = READ_ONCE(P);
525
			      D = *Q;
526 527 528 529 530 531 532

There's a clear data dependency here, and it would seem that by the end of the
sequence, Q must be either &A or &B, and that:

	(Q == &A) implies (D == 1)
	(Q == &B) implies (D == 4)

533
But!  CPU 2's perception of P may be updated _before_ its perception of B, thus
534 535 536 537 538 539 540 541
leading to the following situation:

	(Q == &B) and (D == 2) ????

Whilst this may seem like a failure of coherency or causality maintenance, it
isn't, and this behaviour can be observed on certain real CPUs (such as the DEC
Alpha).

542 543
To deal with this, a data dependency barrier or better must be inserted
between the address load and the data load:
544

545 546
	CPU 1		      CPU 2
	===============	      ===============
547 548 549
	{ A == 1, B == 2, C = 3, P == &A, Q == &C }
	B = 4;
	<write barrier>
550 551
	WRITE_ONCE(P, &B);
			      Q = READ_ONCE(P);
552 553
			      <data dependency barrier>
			      D = *Q;
554 555 556 557 558 559 560 561 562 563 564

This enforces the occurrence of one of the two implications, and prevents the
third possibility from arising.

[!] Note that this extremely counterintuitive situation arises most easily on
machines with split caches, so that, for example, one cache bank processes
even-numbered cache lines and the other bank processes odd-numbered cache
lines.  The pointer P might be stored in an odd-numbered cache line, and the
variable B might be stored in an even-numbered cache line.  Then, if the
even-numbered bank of the reading CPU's cache is extremely busy while the
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
569 570 571
number is read from memory and then used to calculate the index for an array
access:

572 573
	CPU 1		      CPU 2
	===============	      ===============
574 575 576
	{ M[0] == 1, M[1] == 2, M[3] = 3, P == 0, Q == 3 }
	M[1] = 4;
	<write barrier>
577 578
	WRITE_ONCE(P, 1);
			      Q = READ_ONCE(P);
579 580
			      <data dependency barrier>
			      D = M[Q];
581 582


583 584 585 586 587
The data dependency barrier is very important to the RCU system,
for example.  See rcu_assign_pointer() and rcu_dereference() in
include/linux/rcupdate.h.  This permits the current target of an RCU'd
pointer to be replaced with a new modified target, without the replacement
target appearing to be incompletely initialised.
588 589 590 591 592 593 594

See also the subsection on "Cache Coherency" for a more thorough example.


CONTROL DEPENDENCIES
--------------------

595 596 597
A load-load control dependency requires a full read memory barrier, not
simply a data dependency barrier to make it work correctly.  Consider the
following bit of code:
598

599
	q = READ_ONCE(a);
600 601
	if (q) {
		<data dependency barrier>  /* BUG: No data dependency!!! */
602
		p = READ_ONCE(b);
603
	}
604 605

This will not have the desired effect because there is no actual data
606 607 608 609
dependency, but rather a control dependency that the CPU may short-circuit
by attempting to predict the outcome in advance, so that other CPUs see
the load from b as having happened before the load from a.  In such a
case what's actually required is:
610

611
	q = READ_ONCE(a);
612
	if (q) {
613
		<read barrier>
614
		p = READ_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 = READ_ONCE_CTRL(a);
621
	if (q) {
622
		WRITE_ONCE(b, p);
623 624
	}

625 626 627 628 629
Control dependencies pair normally with other types of barriers.  That
said, please note that READ_ONCE_CTRL() is not optional!  Without the
READ_ONCE_CTRL(), the compiler might combine the load from 'a' with
other loads from 'a', and the store to 'b' with other stores to 'b',
with possible highly counterintuitive effects on ordering.
630 631 632 633 634 635 636

Worse yet, if the compiler is able to prove (say) that the value of
variable 'a' is always non-zero, it would be well within its rights
to optimize the original example by eliminating the "if" statement
as follows:

	q = a;
637 638
	b = p;  /* BUG: Compiler and CPU can both reorder!!! */

639 640 641 642
Finally, the READ_ONCE_CTRL() includes an smp_read_barrier_depends()
that DEC Alpha needs in order to respect control depedencies.

So don't leave out the READ_ONCE_CTRL().
643

644 645
It is tempting to try to enforce ordering on identical stores on both
branches of the "if" statement as follows:
646

647
	q = READ_ONCE_CTRL(a);
648
	if (q) {
649
		barrier();
650
		WRITE_ONCE(b, p);
651 652
		do_something();
	} else {
653
		barrier();
654
		WRITE_ONCE(b, p);
655 656 657
		do_something_else();
	}

658 659
Unfortunately, current compilers will transform this as follows at high
optimization levels:
660

661
	q = READ_ONCE_CTRL(a);
662
	barrier();
663
	WRITE_ONCE(b, p);  /* BUG: No ordering vs. load from a!!! */
664
	if (q) {
665
		/* WRITE_ONCE(b, p); -- moved up, BUG!!! */
666 667
		do_something();
	} else {
668
		/* WRITE_ONCE(b, p); -- moved up, BUG!!! */
669 670 671
		do_something_else();
	}

672 673 674 675 676 677
Now there is no conditional between the load from 'a' and the store to
'b', which means that the CPU is within its rights to reorder them:
The conditional is absolutely required, and must be present in the
assembly code even after all compiler optimizations have been applied.
Therefore, if you need ordering in this example, you need explicit
memory barriers, for example, smp_store_release():
678

679
	q = READ_ONCE(a);
680 681
	if (q) {
		smp_store_release(&b, p);
682 683
		do_something();
	} else {
684
		smp_store_release(&b, p);
685 686 687
		do_something_else();
	}

688 689 690
In contrast, without explicit memory barriers, two-legged-if control
ordering is guaranteed only when the stores differ, for example:

691
	q = READ_ONCE_CTRL(a);
692
	if (q) {
693
		WRITE_ONCE(b, p);
694 695
		do_something();
	} else {
696
		WRITE_ONCE(b, r);
697 698 699
		do_something_else();
	}

700 701
The initial READ_ONCE_CTRL() is still required to prevent the compiler
from proving the value of 'a'.
702 703 704 705 706

In addition, you need to be careful what you do with the local variable 'q',
otherwise the compiler might be able to guess the value and again remove
the needed conditional.  For example:

707
	q = READ_ONCE_CTRL(a);
708
	if (q % MAX) {
709
		WRITE_ONCE(b, p);
710 711
		do_something();
	} else {
712
		WRITE_ONCE(b, r);
713 714 715 716 717 718 719
		do_something_else();
	}

If MAX is defined to be 1, then the compiler knows that (q % MAX) is
equal to zero, in which case the compiler is within its rights to
transform the above code into the following:

720
	q = READ_ONCE_CTRL(a);
721
	WRITE_ONCE(b, p);
722 723
	do_something_else();

724 725 726 727 728 729
Given this transformation, the CPU is not required to respect the ordering
between the load from variable 'a' and the store to variable 'b'.  It is
tempting to add a barrier(), but this does not help.  The conditional
is gone, and the barrier won't bring it back.  Therefore, if you are
relying on this ordering, you should make sure that MAX is greater than
one, perhaps as follows:
730

731
	q = READ_ONCE_CTRL(a);
732 733
	BUILD_BUG_ON(MAX <= 1); /* Order load from a with store to b. */
	if (q % MAX) {
734
		WRITE_ONCE(b, p);
735 736
		do_something();
	} else {
737
		WRITE_ONCE(b, r);
738 739 740
		do_something_else();
	}

741 742 743 744
Please note once again that the stores to 'b' differ.  If they were
identical, as noted earlier, the compiler could pull this store outside
of the 'if' statement.

745 746 747
You must also be careful not to rely too much on boolean short-circuit
evaluation.  Consider this example:

748
	q = READ_ONCE_CTRL(a);
749
	if (q || 1 > 0)
750
		WRITE_ONCE(b, 1);
751

752 753 754
Because the first condition cannot fault and the second condition is
always true, the compiler can transform this example as following,
defeating control dependency:
755

756
	q = READ_ONCE_CTRL(a);
757
	WRITE_ONCE(b, 1);
758 759

This example underscores the need to ensure that the compiler cannot
760
out-guess your code.  More generally, although READ_ONCE() does force
761 762 763
the compiler to actually emit code for a given load, it does not force
the compiler to use the results.

764
Finally, control dependencies do -not- provide transitivity.  This is
765 766
demonstrated by two related examples, with the initial values of
x and y both being zero:
767 768

	CPU 0                     CPU 1
769 770
	=======================   =======================
	r1 = READ_ONCE_CTRL(x);   r2 = READ_ONCE_CTRL(y);
771
	if (r1 > 0)               if (r2 > 0)
772
	  WRITE_ONCE(y, 1);         WRITE_ONCE(x, 1);
773 774 775 776 777

	assert(!(r1 == 1 && r2 == 1));

The above two-CPU example will never trigger the assert().  However,
if control dependencies guaranteed transitivity (which they do not),
778
then adding the following CPU would guarantee a related assertion:
779

780 781
	CPU 2
	=====================
782
	WRITE_ONCE(x, 2);
783 784

	assert(!(r1 == 2 && r2 == 1 && x == 2)); /* FAILS!!! */
785

786 787 788 789
But because control dependencies do -not- provide transitivity, the above
assertion can fail after the combined three-CPU example completes.  If you
need the three-CPU example to provide ordering, you will need smp_mb()
between the loads and stores in the CPU 0 and CPU 1 code fragments,
790 791
that is, just before or just after the "if" statements.  Furthermore,
the original two-CPU example is very fragile and should be avoided.
792

793 794 795
These two examples are the LB and WWC litmus tests from this paper:
http://www.cl.cam.ac.uk/users/pes20/ppc-supplemental/test6.pdf and this
site: https://www.cl.cam.ac.uk/~pes20/ppcmem/index.html.
796 797 798

In summary:

799 800
  (*) Control dependencies must be headed by READ_ONCE_CTRL().
      Or, as a much less preferable alternative, interpose
801
      smp_read_barrier_depends() between a READ_ONCE() and the
802 803
      control-dependent write.

804 805 806 807
  (*) Control dependencies can order prior loads against later stores.
      However, they do -not- guarantee any other sort of ordering:
      Not prior loads against later loads, nor prior stores against
      later anything.  If you need these other forms of ordering,
808
      use smp_rmb(), smp_wmb(), or, in the case of prior stores and
809 810
      later loads, smp_mb().

811 812 813 814
  (*) If both legs of the "if" statement begin with identical stores
      to the same variable, a barrier() statement is required at the
      beginning of each leg of the "if" statement.

815
  (*) Control dependencies require at least one run-time conditional
816
      between the prior load and the subsequent store, and this
817 818 819 820
      conditional must involve the prior load.  If the compiler is able
      to optimize the conditional away, it will have also optimized
      away the ordering.  Careful use of READ_ONCE_CTRL() READ_ONCE(),
      and WRITE_ONCE() can help to preserve the needed conditional.
821 822

  (*) Control dependencies require that the compiler avoid reordering the
823 824 825 826
      dependency into nonexistence.  Careful use of READ_ONCE_CTRL()
      or smp_read_barrier_depends() can help to preserve your control
      dependency.  Please see the Compiler Barrier section for more
      information.
827

828 829
  (*) Control dependencies pair normally with other types of barriers.

830 831
  (*) Control dependencies do -not- provide transitivity.  If you
      need transitivity, use smp_mb().
832 833 834 835 836 837 838 839


SMP BARRIER PAIRING
-------------------

When dealing with CPU-CPU interactions, certain types of memory barrier should
always be paired.  A lack of appropriate pairing is almost certainly an error.

840 841 842 843 844 845 846 847
General barriers pair with each other, though they also pair with most
other types of barriers, albeit without transitivity.  An acquire barrier
pairs with a release barrier, but both may also pair with other barriers,
including of course general barriers.  A write barrier pairs with a data
dependency barrier, a control dependency, an acquire barrier, a release
barrier, a read barrier, or a general barrier.  Similarly a read barrier,
control dependency, or a data dependency barrier pairs with a write
barrier, an acquire barrier, a release barrier, or a general barrier:
848

849 850
	CPU 1		      CPU 2
	===============	      ===============
851
	WRITE_ONCE(a, 1);
852
	<write barrier>
853
	WRITE_ONCE(b, 2);     x = READ_ONCE(b);
854
			      <read barrier>
855
			      y = READ_ONCE(a);
856 857 858

Or:

859 860
	CPU 1		      CPU 2
	===============	      ===============================
861 862
	a = 1;
	<write barrier>
863
	WRITE_ONCE(b, &a);    x = READ_ONCE(b);
864 865
			      <data dependency barrier>
			      y = *x;
866

867 868 869 870
Or even:

	CPU 1		      CPU 2
	===============	      ===============================
871
	r1 = READ_ONCE(y);
872
	<general barrier>
873
	WRITE_ONCE(y, 1);     if (r2 = READ_ONCE(x)) {
874
			         <implicit control dependency>
875
			         WRITE_ONCE(y, 1);
876 877 878 879
			      }

	assert(r1 == 0 || r2 == 0);

880 881 882
Basically, the read barrier always has to be there, even though it can be of
the "weaker" type.

883
[!] Note that the stores before the write barrier would normally be expected to
884
match the loads after the read barrier or the data dependency barrier, and vice
885 886
versa:

887 888
	CPU 1                               CPU 2
	===================                 ===================
889 890
	WRITE_ONCE(a, 1);    }----   --->{  v = READ_ONCE(c);
	WRITE_ONCE(b, 2);    }    \ /    {  w = READ_ONCE(d);
891
	<write barrier>            \        <read barrier>
892 893
	WRITE_ONCE(c, 3);    }    / \    {  x = READ_ONCE(a);
	WRITE_ONCE(d, 4);    }----   --->{  y = READ_ONCE(b);
894

895 896 897 898

EXAMPLES OF MEMORY BARRIER SEQUENCES
------------------------------------

899
Firstly, write barriers act as partial orderings on store operations.
900 901 902 903 904 905 906 907 908 909 910 911 912
Consider the following sequence of events:

	CPU 1
	=======================
	STORE A = 1
	STORE B = 2
	STORE C = 3
	<write barrier>
	STORE D = 4
	STORE E = 5

This sequence of events is committed to the memory coherence system in an order
that the rest of the system might perceive as the unordered set of { STORE A,
A
Adrian Bunk 已提交
913
STORE B, STORE C } all occurring before the unordered set of { STORE D, STORE E
914 915 916 917 918
}:

	+-------+       :      :
	|       |       +------+
	|       |------>| C=3  |     }     /\
919 920
	|       |  :    +------+     }-----  \  -----> Events perceptible to
	|       |  :    | A=1  |     }        \/       the rest of the system
921 922 923 924 925 926
	|       |  :    +------+     }
	| CPU 1 |  :    | B=2  |     }
	|       |       +------+     }
	|       |   wwwwwwwwwwwwwwww }   <--- At this point the write barrier
	|       |       +------+     }        requires all stores prior to the
	|       |  :    | E=5  |     }        barrier to be committed before
927
	|       |  :    +------+     }        further stores may take place
928 929 930 931
	|       |------>| D=4  |     }
	|       |       +------+
	+-------+       :      :
	                   |
932 933
	                   | Sequence in which stores are committed to the
	                   | memory system by CPU 1
934 935 936
	                   V


937
Secondly, data dependency barriers act as partial orderings on data-dependent
938 939 940 941
loads.  Consider the following sequence of events:

	CPU 1			CPU 2
	=======================	=======================
942
		{ B = 7; X = 9; Y = 8; C = &Y }
943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980
	STORE A = 1
	STORE B = 2
	<write barrier>
	STORE C = &B		LOAD X
	STORE D = 4		LOAD C (gets &B)
				LOAD *C (reads B)

Without intervention, CPU 2 may perceive the events on CPU 1 in some
effectively random order, despite the write barrier issued by CPU 1:

	+-------+       :      :                :       :
	|       |       +------+                +-------+  | Sequence of update
	|       |------>| B=2  |-----       --->| Y->8  |  | of perception on
	|       |  :    +------+     \          +-------+  | CPU 2
	| CPU 1 |  :    | A=1  |      \     --->| C->&Y |  V
	|       |       +------+       |        +-------+
	|       |   wwwwwwwwwwwwwwww   |        :       :
	|       |       +------+       |        :       :
	|       |  :    | C=&B |---    |        :       :       +-------+
	|       |  :    +------+   \   |        +-------+       |       |
	|       |------>| D=4  |    ----------->| C->&B |------>|       |
	|       |       +------+       |        +-------+       |       |
	+-------+       :      :       |        :       :       |       |
	                               |        :       :       |       |
	                               |        :       :       | CPU 2 |
	                               |        +-------+       |       |
	    Apparently incorrect --->  |        | B->7  |------>|       |
	    perception of B (!)        |        +-------+       |       |
	                               |        :       :       |       |
	                               |        +-------+       |       |
	    The load of X holds --->    \       | X->9  |------>|       |
	    up the maintenance           \      +-------+       |       |
	    of coherence of B             ----->| B->2  |       +-------+
	                                        +-------+
	                                        :       :


In the above example, CPU 2 perceives that B is 7, despite the load of *C
981
(which would be B) coming after the LOAD of C.
982 983

If, however, a data dependency barrier were to be placed between the load of C
984 985 986 987 988 989 990 991 992 993 994 995 996 997
and the load of *C (ie: B) on CPU 2:

	CPU 1			CPU 2
	=======================	=======================
		{ B = 7; X = 9; Y = 8; C = &Y }
	STORE A = 1
	STORE B = 2
	<write barrier>
	STORE C = &B		LOAD X
	STORE D = 4		LOAD C (gets &B)
				<data dependency barrier>
				LOAD *C (reads B)

then the following will occur:
998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014

	+-------+       :      :                :       :
	|       |       +------+                +-------+
	|       |------>| B=2  |-----       --->| Y->8  |
	|       |  :    +------+     \          +-------+
	| CPU 1 |  :    | A=1  |      \     --->| C->&Y |
	|       |       +------+       |        +-------+
	|       |   wwwwwwwwwwwwwwww   |        :       :
	|       |       +------+       |        :       :
	|       |  :    | C=&B |---    |        :       :       +-------+
	|       |  :    +------+   \   |        +-------+       |       |
	|       |------>| D=4  |    ----------->| C->&B |------>|       |
	|       |       +------+       |        +-------+       |       |
	+-------+       :      :       |        :       :       |       |
	                               |        :       :       |       |
	                               |        :       :       | CPU 2 |
	                               |        +-------+       |       |
1015 1016 1017 1018 1019 1020
	                               |        | X->9  |------>|       |
	                               |        +-------+       |       |
	  Makes sure all effects --->   \   ddddddddddddddddd   |       |
	  prior to the store of C        \      +-------+       |       |
	  are perceptible to              ----->| B->2  |------>|       |
	  subsequent loads                      +-------+       |       |
1021 1022 1023 1024 1025 1026 1027 1028
	                                        :       :       +-------+


And thirdly, a read barrier acts as a partial order on loads.  Consider the
following sequence of events:

	CPU 1			CPU 2
	=======================	=======================
1029
		{ A = 0, B = 9 }
1030 1031
	STORE A=1
	<write barrier>
1032
	STORE B=2
1033
				LOAD B
1034
				LOAD A
1035 1036 1037 1038

Without intervention, CPU 2 may then choose to perceive the events on CPU 1 in
some effectively random order, despite the write barrier issued by CPU 1:

1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
	+-------+       :      :                :       :
	|       |       +------+                +-------+
	|       |------>| A=1  |------      --->| A->0  |
	|       |       +------+      \         +-------+
	| CPU 1 |   wwwwwwwwwwwwwwww   \    --->| B->9  |
	|       |       +------+        |       +-------+
	|       |------>| B=2  |---     |       :       :
	|       |       +------+   \    |       :       :       +-------+
	+-------+       :      :    \   |       +-------+       |       |
	                             ---------->| B->2  |------>|       |
	                                |       +-------+       | CPU 2 |
	                                |       | A->0  |------>|       |
	                                |       +-------+       |       |
	                                |       :       :       +-------+
	                                 \      :       :
	                                  \     +-------+
	                                   ---->| A->1  |
	                                        +-------+
	                                        :       :
1058

1059

1060
If, however, a read barrier were to be placed between the load of B and the
1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182
load of A on CPU 2:

	CPU 1			CPU 2
	=======================	=======================
		{ A = 0, B = 9 }
	STORE A=1
	<write barrier>
	STORE B=2
				LOAD B
				<read barrier>
				LOAD A

then the partial ordering imposed by CPU 1 will be perceived correctly by CPU
2:

	+-------+       :      :                :       :
	|       |       +------+                +-------+
	|       |------>| A=1  |------      --->| A->0  |
	|       |       +------+      \         +-------+
	| CPU 1 |   wwwwwwwwwwwwwwww   \    --->| B->9  |
	|       |       +------+        |       +-------+
	|       |------>| B=2  |---     |       :       :
	|       |       +------+   \    |       :       :       +-------+
	+-------+       :      :    \   |       +-------+       |       |
	                             ---------->| B->2  |------>|       |
	                                |       +-------+       | CPU 2 |
	                                |       :       :       |       |
	                                |       :       :       |       |
	  At this point the read ---->   \  rrrrrrrrrrrrrrrrr   |       |
	  barrier causes all effects      \     +-------+       |       |
	  prior to the storage of B        ---->| A->1  |------>|       |
	  to be perceptible to CPU 2            +-------+       |       |
	                                        :       :       +-------+


To illustrate this more completely, consider what could happen if the code
contained a load of A either side of the read barrier:

	CPU 1			CPU 2
	=======================	=======================
		{ A = 0, B = 9 }
	STORE A=1
	<write barrier>
	STORE B=2
				LOAD B
				LOAD A [first load of A]
				<read barrier>
				LOAD A [second load of A]

Even though the two loads of A both occur after the load of B, they may both
come up with different values:

	+-------+       :      :                :       :
	|       |       +------+                +-------+
	|       |------>| A=1  |------      --->| A->0  |
	|       |       +------+      \         +-------+
	| CPU 1 |   wwwwwwwwwwwwwwww   \    --->| B->9  |
	|       |       +------+        |       +-------+
	|       |------>| B=2  |---     |       :       :
	|       |       +------+   \    |       :       :       +-------+
	+-------+       :      :    \   |       +-------+       |       |
	                             ---------->| B->2  |------>|       |
	                                |       +-------+       | CPU 2 |
	                                |       :       :       |       |
	                                |       :       :       |       |
	                                |       +-------+       |       |
	                                |       | A->0  |------>| 1st   |
	                                |       +-------+       |       |
	  At this point the read ---->   \  rrrrrrrrrrrrrrrrr   |       |
	  barrier causes all effects      \     +-------+       |       |
	  prior to the storage of B        ---->| A->1  |------>| 2nd   |
	  to be perceptible to CPU 2            +-------+       |       |
	                                        :       :       +-------+


But it may be that the update to A from CPU 1 becomes perceptible to CPU 2
before the read barrier completes anyway:

	+-------+       :      :                :       :
	|       |       +------+                +-------+
	|       |------>| A=1  |------      --->| A->0  |
	|       |       +------+      \         +-------+
	| CPU 1 |   wwwwwwwwwwwwwwww   \    --->| B->9  |
	|       |       +------+        |       +-------+
	|       |------>| B=2  |---     |       :       :
	|       |       +------+   \    |       :       :       +-------+
	+-------+       :      :    \   |       +-------+       |       |
	                             ---------->| B->2  |------>|       |
	                                |       +-------+       | CPU 2 |
	                                |       :       :       |       |
	                                 \      :       :       |       |
	                                  \     +-------+       |       |
	                                   ---->| A->1  |------>| 1st   |
	                                        +-------+       |       |
	                                    rrrrrrrrrrrrrrrrr   |       |
	                                        +-------+       |       |
	                                        | A->1  |------>| 2nd   |
	                                        +-------+       |       |
	                                        :       :       +-------+


The guarantee is that the second load will always come up with A == 1 if the
load of B came up with B == 2.  No such guarantee exists for the first load of
A; that may come up with either A == 0 or A == 1.


READ MEMORY BARRIERS VS LOAD SPECULATION
----------------------------------------

Many CPUs speculate with loads: that is they see that they will need to load an
item from memory, and they find a time where they're not using the bus for any
other loads, and so do the load in advance - even though they haven't actually
got to that point in the instruction execution flow yet.  This permits the
actual load instruction to potentially complete immediately because the CPU
already has the value to hand.

It may turn out that the CPU didn't actually need the value - perhaps because a
branch circumvented the load - in which case it can discard the value or just
cache it for later use.

Consider:

1183
	CPU 1			CPU 2
1184
	=======================	=======================
1185 1186 1187 1188
				LOAD B
				DIVIDE		} Divide instructions generally
				DIVIDE		} take a long time to perform
				LOAD A
1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210

Which might appear as this:

	                                        :       :       +-------+
	                                        +-------+       |       |
	                                    --->| B->2  |------>|       |
	                                        +-------+       | CPU 2 |
	                                        :       :DIVIDE |       |
	                                        +-------+       |       |
	The CPU being busy doing a --->     --->| A->0  |~~~~   |       |
	division speculates on the              +-------+   ~   |       |
	LOAD of A                               :       :   ~   |       |
	                                        :       :DIVIDE |       |
	                                        :       :   ~   |       |
	Once the divisions are complete -->     :       :   ~-->|       |
	the CPU can then perform the            :       :       |       |
	LOAD with immediate effect              :       :       +-------+


Placing a read barrier or a data dependency barrier just before the second
load:

1211
	CPU 1			CPU 2
1212
	=======================	=======================
1213 1214 1215
				LOAD B
				DIVIDE
				DIVIDE
1216
				<read barrier>
1217
				LOAD A
1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261

will force any value speculatively obtained to be reconsidered to an extent
dependent on the type of barrier used.  If there was no change made to the
speculated memory location, then the speculated value will just be used:

	                                        :       :       +-------+
	                                        +-------+       |       |
	                                    --->| B->2  |------>|       |
	                                        +-------+       | CPU 2 |
	                                        :       :DIVIDE |       |
	                                        +-------+       |       |
	The CPU being busy doing a --->     --->| A->0  |~~~~   |       |
	division speculates on the              +-------+   ~   |       |
	LOAD of A                               :       :   ~   |       |
	                                        :       :DIVIDE |       |
	                                        :       :   ~   |       |
	                                        :       :   ~   |       |
	                                    rrrrrrrrrrrrrrrr~   |       |
	                                        :       :   ~   |       |
	                                        :       :   ~-->|       |
	                                        :       :       |       |
	                                        :       :       +-------+


but if there was an update or an invalidation from another CPU pending, then
the speculation will be cancelled and the value reloaded:

	                                        :       :       +-------+
	                                        +-------+       |       |
	                                    --->| B->2  |------>|       |
	                                        +-------+       | CPU 2 |
	                                        :       :DIVIDE |       |
	                                        +-------+       |       |
	The CPU being busy doing a --->     --->| A->0  |~~~~   |       |
	division speculates on the              +-------+   ~   |       |
	LOAD of A                               :       :   ~   |       |
	                                        :       :DIVIDE |       |
	                                        :       :   ~   |       |
	                                        :       :   ~   |       |
	                                    rrrrrrrrrrrrrrrrr   |       |
	                                        +-------+       |       |
	The speculation is discarded --->   --->| A->1  |------>|       |
	and an updated value is                 +-------+       |       |
	retrieved                               :       :       +-------+
1262 1263


1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320
TRANSITIVITY
------------

Transitivity is a deeply intuitive notion about ordering that is not
always provided by real computer systems.  The following example
demonstrates transitivity (also called "cumulativity"):

	CPU 1			CPU 2			CPU 3
	=======================	=======================	=======================
		{ X = 0, Y = 0 }
	STORE X=1		LOAD X			STORE Y=1
				<general barrier>	<general barrier>
				LOAD Y			LOAD X

Suppose that CPU 2's load from X returns 1 and its load from Y returns 0.
This indicates that CPU 2's load from X in some sense follows CPU 1's
store to X and that CPU 2's load from Y in some sense preceded CPU 3's
store to Y.  The question is then "Can CPU 3's load from X return 0?"

Because CPU 2's load from X in some sense came after CPU 1's store, it
is natural to expect that CPU 3's load from X must therefore return 1.
This expectation is an example of transitivity: if a load executing on
CPU A follows a load from the same variable executing on CPU B, then
CPU A's load must either return the same value that CPU B's load did,
or must return some later value.

In the Linux kernel, use of general memory barriers guarantees
transitivity.  Therefore, in the above example, if CPU 2's load from X
returns 1 and its load from Y returns 0, then CPU 3's load from X must
also return 1.

However, transitivity is -not- guaranteed for read or write barriers.
For example, suppose that CPU 2's general barrier in the above example
is changed to a read barrier as shown below:

	CPU 1			CPU 2			CPU 3
	=======================	=======================	=======================
		{ X = 0, Y = 0 }
	STORE X=1		LOAD X			STORE Y=1
				<read barrier>		<general barrier>
				LOAD Y			LOAD X

This substitution destroys transitivity: in this example, it is perfectly
legal for CPU 2's load from X to return 1, its load from Y to return 0,
and CPU 3's load from X to return 0.

The key point is that although CPU 2's read barrier orders its pair
of loads, it does not guarantee to order CPU 1's store.  Therefore, if
this example runs on a system where CPUs 1 and 2 share a store buffer
or a level of cache, CPU 2 might have early access to CPU 1's writes.
General barriers are therefore required to ensure that all CPUs agree
on the combined order of CPU 1's and CPU 2's accesses.

To reiterate, if your code requires transitivity, use general barriers
throughout.


1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342
========================
EXPLICIT KERNEL BARRIERS
========================

The Linux kernel has a variety of different barriers that act at different
levels:

  (*) Compiler barrier.

  (*) CPU memory barriers.

  (*) MMIO write barrier.


COMPILER BARRIER
----------------

The Linux kernel has an explicit compiler barrier function that prevents the
compiler from moving the memory accesses either side of it to the other side:

	barrier();

1343 1344 1345 1346
This is a general barrier -- there are no read-read or write-write
variants of barrier().  However, READ_ONCE() and WRITE_ONCE() can be
thought of as weak forms of barrier() that affect only the specific
accesses flagged by the READ_ONCE() or WRITE_ONCE().
1347

1348 1349 1350 1351 1352 1353 1354 1355 1356 1357
The barrier() function has the following effects:

 (*) Prevents the compiler from reordering accesses following the
     barrier() to precede any accesses preceding the barrier().
     One example use for this property is to ease communication between
     interrupt-handler code and the code that was interrupted.

 (*) Within a loop, forces the compiler to load the variables used
     in that loop's conditional on each pass through that loop.

1358 1359 1360 1361
The READ_ONCE() and WRITE_ONCE() functions can prevent any number of
optimizations that, while perfectly safe in single-threaded code, can
be fatal in concurrent code.  Here are some examples of these sorts
of optimizations:
1362

1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373
 (*) The compiler is within its rights to reorder loads and stores
     to the same variable, and in some cases, the CPU is within its
     rights to reorder loads to the same variable.  This means that
     the following code:

	a[0] = x;
	a[1] = x;

     Might result in an older value of x stored in a[1] than in a[0].
     Prevent both the compiler and the CPU from doing this as follows:

1374 1375
	a[0] = READ_ONCE(x);
	a[1] = READ_ONCE(x);
1376

1377 1378
     In short, READ_ONCE() and WRITE_ONCE() provide cache coherence for
     accesses from multiple CPUs to a single variable.
1379

1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394
 (*) The compiler is within its rights to merge successive loads from
     the same variable.  Such merging can cause the compiler to "optimize"
     the following code:

	while (tmp = a)
		do_something_with(tmp);

     into the following code, which, although in some sense legitimate
     for single-threaded code, is almost certainly not what the developer
     intended:

	if (tmp = a)
		for (;;)
			do_something_with(tmp);

1395
     Use READ_ONCE() to prevent the compiler from doing this to you:
1396

1397
	while (tmp = READ_ONCE(a))
1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418
		do_something_with(tmp);

 (*) The compiler is within its rights to reload a variable, for example,
     in cases where high register pressure prevents the compiler from
     keeping all data of interest in registers.  The compiler might
     therefore optimize the variable 'tmp' out of our previous example:

	while (tmp = a)
		do_something_with(tmp);

     This could result in the following code, which is perfectly safe in
     single-threaded code, but can be fatal in concurrent code:

	while (a)
		do_something_with(a);

     For example, the optimized version of this code could result in
     passing a zero to do_something_with() in the case where the variable
     a was modified by some other CPU between the "while" statement and
     the call to do_something_with().

1419
     Again, use READ_ONCE() to prevent the compiler from doing this:
1420

1421
	while (tmp = READ_ONCE(a))
1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440
		do_something_with(tmp);

     Note that if the compiler runs short of registers, it might save
     tmp onto the stack.  The overhead of this saving and later restoring
     is why compilers reload variables.  Doing so is perfectly safe for
     single-threaded code, so you need to tell the compiler about cases
     where it is not safe.

 (*) The compiler is within its rights to omit a load entirely if it knows
     what the value will be.  For example, if the compiler can prove that
     the value of variable 'a' is always zero, it can optimize this code:

	while (tmp = a)
		do_something_with(tmp);

     Into this:

	do { } while (0);

1441 1442 1443 1444 1445 1446
     This transformation is a win for single-threaded code because it
     gets rid of a load and a branch.  The problem is that the compiler
     will carry out its proof assuming that the current CPU is the only
     one updating variable 'a'.  If variable 'a' is shared, then the
     compiler's proof will be erroneous.  Use READ_ONCE() to tell the
     compiler that it doesn't know as much as it thinks it does:
1447

1448
	while (tmp = READ_ONCE(a))
1449 1450 1451
		do_something_with(tmp);

     But please note that the compiler is also closely watching what you
1452
     do with the value after the READ_ONCE().  For example, suppose you
1453 1454
     do the following and MAX is a preprocessor macro with the value 1:

1455
	while ((tmp = READ_ONCE(a)) % MAX)
1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478
		do_something_with(tmp);

     Then the compiler knows that the result of the "%" operator applied
     to MAX will always be zero, again allowing the compiler to optimize
     the code into near-nonexistence.  (It will still load from the
     variable 'a'.)

 (*) Similarly, the compiler is within its rights to omit a store entirely
     if it knows that the variable already has the value being stored.
     Again, the compiler assumes that the current CPU is the only one
     storing into the variable, which can cause the compiler to do the
     wrong thing for shared variables.  For example, suppose you have
     the following:

	a = 0;
	/* Code that does not store to variable a. */
	a = 0;

     The compiler sees that the value of variable 'a' is already zero, so
     it might well omit the second store.  This would come as a fatal
     surprise if some other CPU might have stored to variable 'a' in the
     meantime.

1479
     Use WRITE_ONCE() to prevent the compiler from making this sort of
1480 1481
     wrong guess:

1482
	WRITE_ONCE(a, 0);
1483
	/* Code that does not store to variable a. */
1484
	WRITE_ONCE(a, 0);
1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501

 (*) The compiler is within its rights to reorder memory accesses unless
     you tell it not to.  For example, consider the following interaction
     between process-level code and an interrupt handler:

	void process_level(void)
	{
		msg = get_message();
		flag = true;
	}

	void interrupt_handler(void)
	{
		if (flag)
			process_message(msg);
	}

M
Masanari Iida 已提交
1502
     There is nothing to prevent the compiler from transforming
1503 1504 1505 1506 1507 1508 1509 1510 1511 1512
     process_level() to the following, in fact, this might well be a
     win for single-threaded code:

	void process_level(void)
	{
		flag = true;
		msg = get_message();
	}

     If the interrupt occurs between these two statement, then
1513
     interrupt_handler() might be passed a garbled msg.  Use WRITE_ONCE()
1514 1515 1516 1517
     to prevent this as follows:

	void process_level(void)
	{
1518 1519
		WRITE_ONCE(msg, get_message());
		WRITE_ONCE(flag, true);
1520 1521 1522 1523
	}

	void interrupt_handler(void)
	{
1524 1525
		if (READ_ONCE(flag))
			process_message(READ_ONCE(msg));
1526 1527
	}

1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549
     Note that the READ_ONCE() and WRITE_ONCE() wrappers in
     interrupt_handler() are needed if this interrupt handler can itself
     be interrupted by something that also accesses 'flag' and 'msg',
     for example, a nested interrupt or an NMI.  Otherwise, READ_ONCE()
     and WRITE_ONCE() are not needed in interrupt_handler() other than
     for documentation purposes.  (Note also that nested interrupts
     do not typically occur in modern Linux kernels, in fact, if an
     interrupt handler returns with interrupts enabled, you will get a
     WARN_ONCE() splat.)

     You should assume that the compiler can move READ_ONCE() and
     WRITE_ONCE() past code not containing READ_ONCE(), WRITE_ONCE(),
     barrier(), or similar primitives.

     This effect could also be achieved using barrier(), but READ_ONCE()
     and WRITE_ONCE() are more selective:  With READ_ONCE() and
     WRITE_ONCE(), the compiler need only forget the contents of the
     indicated memory locations, while with barrier() the compiler must
     discard the value of all memory locations that it has currented
     cached in any machine registers.  Of course, the compiler must also
     respect the order in which the READ_ONCE()s and WRITE_ONCE()s occur,
     though the CPU of course need not do so.
1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568

 (*) The compiler is within its rights to invent stores to a variable,
     as in the following example:

	if (a)
		b = a;
	else
		b = 42;

     The compiler might save a branch by optimizing this as follows:

	b = 42;
	if (a)
		b = a;

     In single-threaded code, this is not only safe, but also saves
     a branch.  Unfortunately, in concurrent code, this optimization
     could cause some other CPU to see a spurious value of 42 -- even
     if variable 'a' was never zero -- when loading variable 'b'.
1569
     Use WRITE_ONCE() to prevent this as follows:
1570 1571

	if (a)
1572
		WRITE_ONCE(b, a);
1573
	else
1574
		WRITE_ONCE(b, 42);
1575 1576 1577

     The compiler can also invent loads.  These are usually less
     damaging, but they can result in cache-line bouncing and thus in
1578
     poor performance and scalability.  Use READ_ONCE() to prevent
1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596
     invented loads.

 (*) For aligned memory locations whose size allows them to be accessed
     with a single memory-reference instruction, prevents "load tearing"
     and "store tearing," in which a single large access is replaced by
     multiple smaller accesses.  For example, given an architecture having
     16-bit store instructions with 7-bit immediate fields, the compiler
     might be tempted to use two 16-bit store-immediate instructions to
     implement the following 32-bit store:

	p = 0x00010002;

     Please note that GCC really does use this sort of optimization,
     which is not surprising given that it would likely take more
     than two instructions to build the constant and then store it.
     This optimization can therefore be a win in single-threaded code.
     In fact, a recent bug (since fixed) caused GCC to incorrectly use
     this optimization in a volatile store.  In the absence of such bugs,
1597
     use of WRITE_ONCE() prevents store tearing in the following example:
1598

1599
	WRITE_ONCE(p, 0x00010002);
1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615

     Use of packed structures can also result in load and store tearing,
     as in this example:

	struct __attribute__((__packed__)) foo {
		short a;
		int b;
		short c;
	};
	struct foo foo1, foo2;
	...

	foo2.a = foo1.a;
	foo2.b = foo1.b;
	foo2.c = foo1.c;

1616 1617 1618 1619 1620 1621
     Because there are no READ_ONCE() or WRITE_ONCE() wrappers and no
     volatile markings, the compiler would be well within its rights to
     implement these three assignment statements as a pair of 32-bit
     loads followed by a pair of 32-bit stores.  This would result in
     load tearing on 'foo1.b' and store tearing on 'foo2.b'.  READ_ONCE()
     and WRITE_ONCE() again prevent tearing in this example:
1622 1623

	foo2.a = foo1.a;
1624
	WRITE_ONCE(foo2.b, READ_ONCE(foo1.b));
1625 1626
	foo2.c = foo1.c;

1627 1628 1629 1630 1631 1632
All that aside, it is never necessary to use READ_ONCE() and
WRITE_ONCE() on a variable that has been marked volatile.  For example,
because 'jiffies' is marked volatile, it is never necessary to
say READ_ONCE(jiffies).  The reason for this is that READ_ONCE() and
WRITE_ONCE() are implemented as volatile casts, which has no effect when
its argument is already marked volatile.
1633 1634 1635

Please note that these compiler barriers have no direct effect on the CPU,
which may then reorder things however it wishes.
1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650


CPU MEMORY BARRIERS
-------------------

The Linux kernel has eight basic CPU memory barriers:

	TYPE		MANDATORY		SMP CONDITIONAL
	===============	=======================	===========================
	GENERAL		mb()			smp_mb()
	WRITE		wmb()			smp_wmb()
	READ		rmb()			smp_rmb()
	DATA DEPENDENCY	read_barrier_depends()	smp_read_barrier_depends()


N
Nick Piggin 已提交
1651 1652 1653
All memory barriers except the data dependency barriers imply a compiler
barrier. Data dependencies do not impose any additional compiler ordering.

1654 1655 1656 1657 1658 1659 1660 1661 1662
Aside: In the case of data dependencies, the compiler would be expected
to issue the loads in the correct order (eg. `a[b]` would have to load
the value of b before loading a[b]), however there is no guarantee in
the C specification that the compiler may not speculate the value of b
(eg. is equal to 1) and load a before b (eg. tmp = a[1]; if (b != 1)
tmp = a[b]; ). There is also the problem of a compiler reloading b after
having loaded a[b], thus having a newer copy of b than a[b]. A consensus
has not yet been reached about these problems, however the READ_ONCE()
macro is a good place to start looking.
1663 1664

SMP memory barriers are reduced to compiler barriers on uniprocessor compiled
1665
systems because it is assumed that a CPU will appear to be self-consistent,
1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681
and will order overlapping accesses correctly with respect to itself.

[!] Note that SMP memory barriers _must_ be used to control the ordering of
references to shared memory on SMP systems, though the use of locking instead
is sufficient.

Mandatory barriers should not be used to control SMP effects, since mandatory
barriers unnecessarily impose overhead on UP systems. They may, however, be
used to control MMIO effects on accesses through relaxed memory I/O windows.
These are required even on non-SMP systems as they affect the order in which
memory operations appear to a device by prohibiting both the compiler and the
CPU from reordering them.


There are some more advanced barrier functions:

1682
 (*) smp_store_mb(var, value)
1683

1684
     This assigns the value to the variable and then inserts a full memory
1685
     barrier after it, depending on the function.  It isn't guaranteed to
1686 1687 1688
     insert anything more than a compiler barrier in a UP compilation.


P
Peter Zijlstra 已提交
1689 1690
 (*) smp_mb__before_atomic();
 (*) smp_mb__after_atomic();
1691

P
Peter Zijlstra 已提交
1692 1693 1694 1695 1696 1697
     These are for use with atomic (such as add, subtract, increment and
     decrement) functions that don't return a value, especially when used for
     reference counting.  These functions do not imply memory barriers.

     These are also used for atomic bitop functions that do not return a
     value (such as set_bit and clear_bit).
1698 1699 1700 1701 1702

     As an example, consider a piece of code that marks an object as being dead
     and then decrements the object's reference count:

	obj->dead = 1;
P
Peter Zijlstra 已提交
1703
	smp_mb__before_atomic();
1704 1705 1706 1707 1708 1709 1710 1711 1712
	atomic_dec(&obj->ref_count);

     This makes sure that the death mark on the object is perceived to be set
     *before* the reference counter is decremented.

     See Documentation/atomic_ops.txt for more information.  See the "Atomic
     operations" subsection for information on where to use these.


1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746
 (*) dma_wmb();
 (*) dma_rmb();

     These are for use with consistent memory to guarantee the ordering
     of writes or reads of shared memory accessible to both the CPU and a
     DMA capable device.

     For example, consider a device driver that shares memory with a device
     and uses a descriptor status value to indicate if the descriptor belongs
     to the device or the CPU, and a doorbell to notify it when new
     descriptors are available:

	if (desc->status != DEVICE_OWN) {
		/* do not read data until we own descriptor */
		dma_rmb();

		/* read/modify data */
		read_data = desc->data;
		desc->data = write_data;

		/* flush modifications before status update */
		dma_wmb();

		/* assign ownership */
		desc->status = DEVICE_OWN;

		/* force memory to sync before notifying device via MMIO */
		wmb();

		/* notify device of new descriptors */
		writel(DESC_NOTIFY, doorbell);
	}

     The dma_rmb() allows us guarantee the device has released ownership
1747
     before we read the data from the descriptor, and the dma_wmb() allows
1748 1749 1750 1751 1752 1753 1754
     us to guarantee the data is written to the descriptor before the device
     can see it now has ownership.  The wmb() is needed to guarantee that the
     cache coherent memory writes have completed before attempting a write to
     the cache incoherent MMIO region.

     See Documentation/DMA-API.txt for more information on consistent memory.

1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774
MMIO WRITE BARRIER
------------------

The Linux kernel also has a special barrier for use with memory-mapped I/O
writes:

	mmiowb();

This is a variation on the mandatory write barrier that causes writes to weakly
ordered I/O regions to be partially ordered.  Its effects may go beyond the
CPU->Hardware interface and actually affect the hardware at some level.

See the subsection "Locks vs I/O accesses" for more information.


===============================
IMPLICIT KERNEL MEMORY BARRIERS
===============================

Some of the other functions in the linux kernel imply memory barriers, amongst
1775
which are locking and scheduling functions.
1776 1777 1778 1779 1780 1781

This specification is a _minimum_ guarantee; any particular architecture may
provide more substantial guarantees, but these may not be relied upon outside
of arch specific code.


1782 1783
ACQUIRING FUNCTIONS
-------------------
1784 1785 1786 1787 1788 1789 1790 1791 1792 1793

The Linux kernel has a number of locking constructs:

 (*) spin locks
 (*) R/W spin locks
 (*) mutexes
 (*) semaphores
 (*) R/W semaphores
 (*) RCU

1794
In all cases there are variants on "ACQUIRE" operations and "RELEASE" operations
1795 1796
for each construct.  These operations all imply certain barriers:

1797
 (1) ACQUIRE operation implication:
1798

1799 1800
     Memory operations issued after the ACQUIRE will be completed after the
     ACQUIRE operation has completed.
1801

1802 1803
     Memory operations issued before the ACQUIRE may be completed after
     the ACQUIRE operation has completed.  An smp_mb__before_spinlock(),
1804 1805 1806
     combined with a following ACQUIRE, orders prior stores against
     subsequent loads and stores. Note that this is weaker than smp_mb()!
     The smp_mb__before_spinlock() primitive is free on many architectures.
1807

1808
 (2) RELEASE operation implication:
1809

1810 1811
     Memory operations issued before the RELEASE will be completed before the
     RELEASE operation has completed.
1812

1813 1814
     Memory operations issued after the RELEASE may be completed before the
     RELEASE operation has completed.
1815

1816
 (3) ACQUIRE vs ACQUIRE implication:
1817

1818 1819
     All ACQUIRE operations issued before another ACQUIRE operation will be
     completed before that ACQUIRE operation.
1820

1821
 (4) ACQUIRE vs RELEASE implication:
1822

1823 1824
     All ACQUIRE operations issued before a RELEASE operation will be
     completed before the RELEASE operation.
1825

1826
 (5) Failed conditional ACQUIRE implication:
1827

1828 1829
     Certain locking variants of the ACQUIRE operation may fail, either due to
     being unable to get the lock immediately, or due to receiving an unblocked
1830 1831 1832
     signal whilst asleep waiting for the lock to become available.  Failed
     locks do not imply any sort of barrier.

1833 1834 1835
[!] Note: one of the consequences of lock ACQUIREs and RELEASEs being only
one-way barriers is that the effects of instructions outside of a critical
section may seep into the inside of the critical section.
1836

1837 1838 1839 1840
An ACQUIRE followed by a RELEASE may not be assumed to be full memory barrier
because it is possible for an access preceding the ACQUIRE to happen after the
ACQUIRE, and an access following the RELEASE to happen before the RELEASE, and
the two accesses can themselves then cross:
1841 1842

	*A = a;
1843 1844
	ACQUIRE M
	RELEASE M
1845 1846 1847 1848
	*B = b;

may occur as:

1849
	ACQUIRE M, STORE *B, STORE *A, RELEASE M
1850

1851 1852 1853 1854 1855 1856 1857 1858 1859 1860
When the ACQUIRE and RELEASE are a lock acquisition and release,
respectively, this same reordering can occur if the lock's ACQUIRE and
RELEASE are to the same lock variable, but only from the perspective of
another CPU not holding that lock.  In short, a ACQUIRE followed by an
RELEASE may -not- be assumed to be a full memory barrier.

Similarly, the reverse case of a RELEASE followed by an ACQUIRE does not
imply a full memory barrier.  If it is necessary for a RELEASE-ACQUIRE
pair to produce a full barrier, the ACQUIRE can be followed by an
smp_mb__after_unlock_lock() invocation.  This will produce a full barrier
1861 1862 1863 1864 1865 1866
(including transitivity) if either (a) the RELEASE and the ACQUIRE are
executed by the same CPU or task, or (b) the RELEASE and ACQUIRE act on
the same variable.  The smp_mb__after_unlock_lock() primitive is free
on many architectures.  Without smp_mb__after_unlock_lock(), the CPU's
execution of the critical sections corresponding to the RELEASE and the
ACQUIRE can cross, so that:
1867 1868

	*A = a;
1869 1870
	RELEASE M
	ACQUIRE N
1871 1872 1873 1874
	*B = b;

could occur as:

1875
	ACQUIRE N, STORE *B, STORE *A, RELEASE M
1876

1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906
It might appear that this reordering could introduce a deadlock.
However, this cannot happen because if such a deadlock threatened,
the RELEASE would simply complete, thereby avoiding the deadlock.

	Why does this work?

	One key point is that we are only talking about the CPU doing
	the reordering, not the compiler.  If the compiler (or, for
	that matter, the developer) switched the operations, deadlock
	-could- occur.

	But suppose the CPU reordered the operations.  In this case,
	the unlock precedes the lock in the assembly code.  The CPU
	simply elected to try executing the later lock operation first.
	If there is a deadlock, this lock operation will simply spin (or
	try to sleep, but more on that later).	The CPU will eventually
	execute the unlock operation (which preceded the lock operation
	in the assembly code), which will unravel the potential deadlock,
	allowing the lock operation to succeed.

	But what if the lock is a sleeplock?  In that case, the code will
	try to enter the scheduler, where it will eventually encounter
	a memory barrier, which will force the earlier unlock operation
	to complete, again unraveling the deadlock.  There might be
	a sleep-unlock race, but the locking primitive needs to resolve
	such races properly in any case.

With smp_mb__after_unlock_lock(), the two critical sections cannot overlap.
For example, with the following code, the store to *A will always be
seen by other CPUs before the store to *B:
1907 1908

	*A = a;
1909 1910
	RELEASE M
	ACQUIRE N
1911 1912 1913
	smp_mb__after_unlock_lock();
	*B = b;

1914
The operations will always occur in one of the following orders:
1915

1916 1917 1918
	STORE *A, RELEASE, ACQUIRE, smp_mb__after_unlock_lock(), STORE *B
	STORE *A, ACQUIRE, RELEASE, smp_mb__after_unlock_lock(), STORE *B
	ACQUIRE, STORE *A, RELEASE, smp_mb__after_unlock_lock(), STORE *B
1919

1920
If the RELEASE and ACQUIRE were instead both operating on the same lock
1921 1922 1923 1924 1925
variable, only the first of these alternatives can occur.  In addition,
the more strongly ordered systems may rule out some of the above orders.
But in any case, as noted earlier, the smp_mb__after_unlock_lock()
ensures that the store to *A will always be seen as happening before
the store to *B.
1926

1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938
Locks and semaphores may not provide any guarantee of ordering on UP compiled
systems, and so cannot be counted on in such a situation to actually achieve
anything at all - especially with respect to I/O accesses - unless combined
with interrupt disabling operations.

See also the section on "Inter-CPU locking barrier effects".


As an example, consider the following:

	*A = a;
	*B = b;
1939
	ACQUIRE
1940 1941
	*C = c;
	*D = d;
1942
	RELEASE
1943 1944 1945 1946 1947
	*E = e;
	*F = f;

The following sequence of events is acceptable:

1948
	ACQUIRE, {*F,*A}, *E, {*C,*D}, *B, RELEASE
1949 1950 1951 1952 1953

	[+] Note that {*F,*A} indicates a combined access.

But none of the following are:

1954 1955 1956 1957
	{*F,*A}, *B,	ACQUIRE, *C, *D,	RELEASE, *E
	*A, *B, *C,	ACQUIRE, *D,		RELEASE, *E, *F
	*A, *B,		ACQUIRE, *C,		RELEASE, *D, *E, *F
	*B,		ACQUIRE, *C, *D,	RELEASE, {*F,*A}, *E
1958 1959 1960 1961 1962 1963



INTERRUPT DISABLING FUNCTIONS
-----------------------------

1964 1965
Functions that disable interrupts (ACQUIRE equivalent) and enable interrupts
(RELEASE equivalent) will act as compiler barriers only.  So if memory or I/O
1966 1967 1968 1969
barriers are required in such a situation, they must be provided from some
other means.


1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
SLEEP AND WAKE-UP FUNCTIONS
---------------------------

Sleeping and waking on an event flagged in global data can be viewed as an
interaction between two pieces of data: the task state of the task waiting for
the event and the global data used to indicate the event.  To make sure that
these appear to happen in the right order, the primitives to begin the process
of going to sleep, and the primitives to initiate a wake up imply certain
barriers.

Firstly, the sleeper normally follows something like this sequence of events:

	for (;;) {
		set_current_state(TASK_UNINTERRUPTIBLE);
		if (event_indicated)
			break;
		schedule();
	}

A general memory barrier is interpolated automatically by set_current_state()
after it has altered the task state:

	CPU 1
	===============================
	set_current_state();
1995
	  smp_store_mb();
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035
	    STORE current->state
	    <general barrier>
	LOAD event_indicated

set_current_state() may be wrapped by:

	prepare_to_wait();
	prepare_to_wait_exclusive();

which therefore also imply a general memory barrier after setting the state.
The whole sequence above is available in various canned forms, all of which
interpolate the memory barrier in the right place:

	wait_event();
	wait_event_interruptible();
	wait_event_interruptible_exclusive();
	wait_event_interruptible_timeout();
	wait_event_killable();
	wait_event_timeout();
	wait_on_bit();
	wait_on_bit_lock();


Secondly, code that performs a wake up normally follows something like this:

	event_indicated = 1;
	wake_up(&event_wait_queue);

or:

	event_indicated = 1;
	wake_up_process(event_daemon);

A write memory barrier is implied by wake_up() and co. if and only if they wake
something up.  The barrier occurs before the task state is cleared, and so sits
between the STORE to indicate the event and the STORE to set TASK_RUNNING:

	CPU 1				CPU 2
	===============================	===============================
	set_current_state();		STORE event_indicated
2036
	  smp_store_mb();		wake_up();
2037 2038 2039 2040
	    STORE current->state	  <write barrier>
	    <general barrier>		  STORE current->state
	LOAD event_indicated

2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055
To repeat, this write memory barrier is present if and only if something
is actually awakened.  To see this, consider the following sequence of
events, where X and Y are both initially zero:

	CPU 1				CPU 2
	===============================	===============================
	X = 1;				STORE event_indicated
	smp_mb();			wake_up();
	Y = 1;				wait_event(wq, Y == 1);
	wake_up();			  load from Y sees 1, no memory barrier
					load from X might see 0

In contrast, if a wakeup does occur, CPU 2's load from X would be guaranteed
to see 1.

2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110
The available waker functions include:

	complete();
	wake_up();
	wake_up_all();
	wake_up_bit();
	wake_up_interruptible();
	wake_up_interruptible_all();
	wake_up_interruptible_nr();
	wake_up_interruptible_poll();
	wake_up_interruptible_sync();
	wake_up_interruptible_sync_poll();
	wake_up_locked();
	wake_up_locked_poll();
	wake_up_nr();
	wake_up_poll();
	wake_up_process();


[!] Note that the memory barriers implied by the sleeper and the waker do _not_
order multiple stores before the wake-up with respect to loads of those stored
values after the sleeper has called set_current_state().  For instance, if the
sleeper does:

	set_current_state(TASK_INTERRUPTIBLE);
	if (event_indicated)
		break;
	__set_current_state(TASK_RUNNING);
	do_something(my_data);

and the waker does:

	my_data = value;
	event_indicated = 1;
	wake_up(&event_wait_queue);

there's no guarantee that the change to event_indicated will be perceived by
the sleeper as coming after the change to my_data.  In such a circumstance, the
code on both sides must interpolate its own memory barriers between the
separate data accesses.  Thus the above sleeper ought to do:

	set_current_state(TASK_INTERRUPTIBLE);
	if (event_indicated) {
		smp_rmb();
		do_something(my_data);
	}

and the waker should do:

	my_data = value;
	smp_wmb();
	event_indicated = 1;
	wake_up(&event_wait_queue);


2111 2112 2113 2114 2115 2116 2117 2118
MISCELLANEOUS FUNCTIONS
-----------------------

Other functions that imply barriers:

 (*) schedule() and similar imply full memory barriers.


2119 2120 2121
===================================
INTER-CPU ACQUIRING BARRIER EFFECTS
===================================
2122 2123 2124 2125 2126 2127

On SMP systems locking primitives give a more substantial form of barrier: one
that does affect memory access ordering on other CPUs, within the context of
conflict on any particular lock.


2128 2129
ACQUIRES VS MEMORY ACCESSES
---------------------------
2130

2131
Consider the following: the system has a pair of spinlocks (M) and (Q), and
2132 2133 2134 2135
three CPUs; then should the following sequence of events occur:

	CPU 1				CPU 2
	===============================	===============================
2136
	WRITE_ONCE(*A, a);		WRITE_ONCE(*E, e);
2137
	ACQUIRE M			ACQUIRE Q
2138 2139
	WRITE_ONCE(*B, b);		WRITE_ONCE(*F, f);
	WRITE_ONCE(*C, c);		WRITE_ONCE(*G, g);
2140
	RELEASE M			RELEASE Q
2141
	WRITE_ONCE(*D, d);		WRITE_ONCE(*H, h);
2142

2143
Then there is no guarantee as to what order CPU 3 will see the accesses to *A
2144 2145 2146
through *H occur in, other than the constraints imposed by the separate locks
on the separate CPUs. It might, for example, see:

2147
	*E, ACQUIRE M, ACQUIRE Q, *G, *C, *F, *A, *B, RELEASE Q, *D, *H, RELEASE M
2148 2149 2150

But it won't see any of:

2151 2152 2153 2154
	*B, *C or *D preceding ACQUIRE M
	*A, *B or *C following RELEASE M
	*F, *G or *H preceding ACQUIRE Q
	*E, *F or *G following RELEASE Q
2155 2156 2157 2158 2159 2160


However, if the following occurs:

	CPU 1				CPU 2
	===============================	===============================
2161
	WRITE_ONCE(*A, a);
2162
	ACQUIRE M		     [1]
2163 2164
	WRITE_ONCE(*B, b);
	WRITE_ONCE(*C, c);
2165
	RELEASE M	     [1]
2166
	WRITE_ONCE(*D, d);		WRITE_ONCE(*E, e);
2167
					ACQUIRE M		     [2]
2168
					smp_mb__after_unlock_lock();
2169 2170
					WRITE_ONCE(*F, f);
					WRITE_ONCE(*G, g);
2171
					RELEASE M	     [2]
2172
					WRITE_ONCE(*H, h);
2173

2174
CPU 3 might see:
2175

2176 2177
	*E, ACQUIRE M [1], *C, *B, *A, RELEASE M [1],
		ACQUIRE M [2], *H, *F, *G, RELEASE M [2], *D
2178

2179
But assuming CPU 1 gets the lock first, CPU 3 won't see any of:
2180

2181 2182 2183 2184
	*B, *C, *D, *F, *G or *H preceding ACQUIRE M [1]
	*A, *B or *C following RELEASE M [1]
	*F, *G or *H preceding ACQUIRE M [2]
	*A, *B, *C, *E, *F or *G following RELEASE M [2]
2185

2186 2187 2188 2189 2190
Note that the smp_mb__after_unlock_lock() is critically important
here: Without it CPU 3 might see some of the above orderings.
Without smp_mb__after_unlock_lock(), the accesses are not guaranteed
to be seen in order unless CPU 3 holds lock M.

2191

2192 2193
ACQUIRES VS I/O ACCESSES
------------------------
2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236

Under certain circumstances (especially involving NUMA), I/O accesses within
two spinlocked sections on two different CPUs may be seen as interleaved by the
PCI bridge, because the PCI bridge does not necessarily participate in the
cache-coherence protocol, and is therefore incapable of issuing the required
read memory barriers.

For example:

	CPU 1				CPU 2
	===============================	===============================
	spin_lock(Q)
	writel(0, ADDR)
	writel(1, DATA);
	spin_unlock(Q);
					spin_lock(Q);
					writel(4, ADDR);
					writel(5, DATA);
					spin_unlock(Q);

may be seen by the PCI bridge as follows:

	STORE *ADDR = 0, STORE *ADDR = 4, STORE *DATA = 1, STORE *DATA = 5

which would probably cause the hardware to malfunction.


What is necessary here is to intervene with an mmiowb() before dropping the
spinlock, for example:

	CPU 1				CPU 2
	===============================	===============================
	spin_lock(Q)
	writel(0, ADDR)
	writel(1, DATA);
	mmiowb();
	spin_unlock(Q);
					spin_lock(Q);
					writel(4, ADDR);
					writel(5, DATA);
					mmiowb();
					spin_unlock(Q);

2237 2238
this will ensure that the two stores issued on CPU 1 appear at the PCI bridge
before either of the stores issued on CPU 2.
2239 2240


2241 2242
Furthermore, following a store by a load from the same device obviates the need
for the mmiowb(), because the load forces the store to complete before the load
2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265
is performed:

	CPU 1				CPU 2
	===============================	===============================
	spin_lock(Q)
	writel(0, ADDR)
	a = readl(DATA);
	spin_unlock(Q);
					spin_lock(Q);
					writel(4, ADDR);
					b = readl(DATA);
					spin_unlock(Q);


See Documentation/DocBook/deviceiobook.tmpl for more information.


=================================
WHERE ARE MEMORY BARRIERS NEEDED?
=================================

Under normal operation, memory operation reordering is generally not going to
be a problem as a single-threaded linear piece of code will still appear to
2266
work correctly, even if it's in an SMP kernel.  There are, however, four
2267 2268 2269 2270 2271 2272
circumstances in which reordering definitely _could_ be a problem:

 (*) Interprocessor interaction.

 (*) Atomic operations.

2273
 (*) Accessing devices.
2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308

 (*) Interrupts.


INTERPROCESSOR INTERACTION
--------------------------

When there's a system with more than one processor, more than one CPU in the
system may be working on the same data set at the same time.  This can cause
synchronisation problems, and the usual way of dealing with them is to use
locks.  Locks, however, are quite expensive, and so it may be preferable to
operate without the use of a lock if at all possible.  In such a case
operations that affect both CPUs may have to be carefully ordered to prevent
a malfunction.

Consider, for example, the R/W semaphore slow path.  Here a waiting process is
queued on the semaphore, by virtue of it having a piece of its stack linked to
the semaphore's list of waiting processes:

	struct rw_semaphore {
		...
		spinlock_t lock;
		struct list_head waiters;
	};

	struct rwsem_waiter {
		struct list_head list;
		struct task_struct *task;
	};

To wake up a particular waiter, the up_read() or up_write() functions have to:

 (1) read the next pointer from this waiter's record to know as to where the
     next waiter record is;

2309
 (2) read the pointer to the waiter's task structure;
2310 2311 2312 2313 2314 2315 2316

 (3) clear the task pointer to tell the waiter it has been given the semaphore;

 (4) call wake_up_process() on the task; and

 (5) release the reference held on the waiter's task struct.

2317
In other words, it has to perform this sequence of events:
2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374

	LOAD waiter->list.next;
	LOAD waiter->task;
	STORE waiter->task;
	CALL wakeup
	RELEASE task

and if any of these steps occur out of order, then the whole thing may
malfunction.

Once it has queued itself and dropped the semaphore lock, the waiter does not
get the lock again; it instead just waits for its task pointer to be cleared
before proceeding.  Since the record is on the waiter's stack, this means that
if the task pointer is cleared _before_ the next pointer in the list is read,
another CPU might start processing the waiter and might clobber the waiter's
stack before the up*() function has a chance to read the next pointer.

Consider then what might happen to the above sequence of events:

	CPU 1				CPU 2
	===============================	===============================
					down_xxx()
					Queue waiter
					Sleep
	up_yyy()
	LOAD waiter->task;
	STORE waiter->task;
					Woken up by other event
	<preempt>
					Resume processing
					down_xxx() returns
					call foo()
					foo() clobbers *waiter
	</preempt>
	LOAD waiter->list.next;
	--- OOPS ---

This could be dealt with using the semaphore lock, but then the down_xxx()
function has to needlessly get the spinlock again after being woken up.

The way to deal with this is to insert a general SMP memory barrier:

	LOAD waiter->list.next;
	LOAD waiter->task;
	smp_mb();
	STORE waiter->task;
	CALL wakeup
	RELEASE task

In this case, the barrier makes a guarantee that all memory accesses before the
barrier will appear to happen before all the memory accesses after the barrier
with respect to the other CPUs on the system.  It does _not_ guarantee that all
the memory accesses before the barrier will be complete by the time the barrier
instruction itself is complete.

On a UP system - where this wouldn't be a problem - the smp_mb() is just a
compiler barrier, thus making sure the compiler emits the instructions in the
2375 2376
right order without actually intervening in the CPU.  Since there's only one
CPU, that CPU's dependency ordering logic will take care of everything else.
2377 2378 2379 2380 2381


ATOMIC OPERATIONS
-----------------

2382 2383 2384 2385 2386 2387 2388
Whilst they are technically interprocessor interaction considerations, atomic
operations are noted specially as some of them imply full memory barriers and
some don't, but they're very heavily relied on as a group throughout the
kernel.

Any atomic operation that modifies some state in memory and returns information
about the state (old or new) implies an SMP-conditional general memory barrier
N
Nick Piggin 已提交
2389 2390
(smp_mb()) on each side of the actual operation (with the exception of
explicit lock operations, described later).  These include:
2391 2392 2393

	xchg();
	cmpxchg();
2394 2395 2396 2397 2398 2399 2400 2401 2402 2403
	atomic_xchg();			atomic_long_xchg();
	atomic_cmpxchg();		atomic_long_cmpxchg();
	atomic_inc_return();		atomic_long_inc_return();
	atomic_dec_return();		atomic_long_dec_return();
	atomic_add_return();		atomic_long_add_return();
	atomic_sub_return();		atomic_long_sub_return();
	atomic_inc_and_test();		atomic_long_inc_and_test();
	atomic_dec_and_test();		atomic_long_dec_and_test();
	atomic_sub_and_test();		atomic_long_sub_and_test();
	atomic_add_negative();		atomic_long_add_negative();
2404 2405 2406 2407
	test_and_set_bit();
	test_and_clear_bit();
	test_and_change_bit();

2408 2409 2410
	/* when succeeds (returns 1) */
	atomic_add_unless();		atomic_long_add_unless();

2411
These are used for such things as implementing ACQUIRE-class and RELEASE-class
2412 2413
operations and adjusting reference counters towards object destruction, and as
such the implicit memory barrier effects are necessary.
2414 2415


2416
The following operations are potential problems as they do _not_ imply memory
2417
barriers, but might be used for implementing such things as RELEASE-class
2418
operations:
2419

2420
	atomic_set();
2421 2422 2423
	set_bit();
	clear_bit();
	change_bit();
2424 2425

With these the appropriate explicit memory barrier should be used if necessary
P
Peter Zijlstra 已提交
2426
(smp_mb__before_atomic() for instance).
2427 2428


2429
The following also do _not_ imply memory barriers, and so may require explicit
P
Peter Zijlstra 已提交
2430
memory barriers under some circumstances (smp_mb__before_atomic() for
2431
instance):
2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450

	atomic_add();
	atomic_sub();
	atomic_inc();
	atomic_dec();

If they're used for statistics generation, then they probably don't need memory
barriers, unless there's a coupling between statistical data.

If they're used for reference counting on an object to control its lifetime,
they probably don't need memory barriers because either the reference count
will be adjusted inside a locked section, or the caller will already hold
sufficient references to make the lock, and thus a memory barrier unnecessary.

If they're used for constructing a lock of some description, then they probably
do need memory barriers as a lock primitive generally has to do things in a
specific order.

Basically, each usage case has to be carefully considered as to whether memory
2451 2452
barriers are needed or not.

N
Nick Piggin 已提交
2453 2454 2455 2456 2457 2458
The following operations are special locking primitives:

	test_and_set_bit_lock();
	clear_bit_unlock();
	__clear_bit_unlock();

2459
These implement ACQUIRE-class and RELEASE-class operations. These should be used in
N
Nick Piggin 已提交
2460 2461 2462
preference to other operations when implementing locking primitives, because
their implementations can be optimised on many architectures.

2463 2464 2465 2466
[!] Note that special memory barrier primitives are available for these
situations because on some CPUs the atomic instructions used imply full memory
barriers, and so barrier instructions are superfluous in conjunction with them,
and in such cases the special barrier primitives will be no-ops.
2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563

See Documentation/atomic_ops.txt for more information.


ACCESSING DEVICES
-----------------

Many devices can be memory mapped, and so appear to the CPU as if they're just
a set of memory locations.  To control such a device, the driver usually has to
make the right memory accesses in exactly the right order.

However, having a clever CPU or a clever compiler creates a potential problem
in that the carefully sequenced accesses in the driver code won't reach the
device in the requisite order if the CPU or the compiler thinks it is more
efficient to reorder, combine or merge accesses - something that would cause
the device to malfunction.

Inside of the Linux kernel, I/O should be done through the appropriate accessor
routines - such as inb() or writel() - which know how to make such accesses
appropriately sequential.  Whilst this, for the most part, renders the explicit
use of memory barriers unnecessary, there are a couple of situations where they
might be needed:

 (1) On some systems, I/O stores are not strongly ordered across all CPUs, and
     so for _all_ general drivers locks should be used and mmiowb() must be
     issued prior to unlocking the critical section.

 (2) If the accessor functions are used to refer to an I/O memory window with
     relaxed memory access properties, then _mandatory_ memory barriers are
     required to enforce ordering.

See Documentation/DocBook/deviceiobook.tmpl for more information.


INTERRUPTS
----------

A driver may be interrupted by its own interrupt service routine, and thus the
two parts of the driver may interfere with each other's attempts to control or
access the device.

This may be alleviated - at least in part - by disabling local interrupts (a
form of locking), such that the critical operations are all contained within
the interrupt-disabled section in the driver.  Whilst the driver's interrupt
routine is executing, the driver's core may not run on the same CPU, and its
interrupt is not permitted to happen again until the current interrupt has been
handled, thus the interrupt handler does not need to lock against that.

However, consider a driver that was talking to an ethernet card that sports an
address register and a data register.  If that driver's core talks to the card
under interrupt-disablement and then the driver's interrupt handler is invoked:

	LOCAL IRQ DISABLE
	writew(ADDR, 3);
	writew(DATA, y);
	LOCAL IRQ ENABLE
	<interrupt>
	writew(ADDR, 4);
	q = readw(DATA);
	</interrupt>

The store to the data register might happen after the second store to the
address register if ordering rules are sufficiently relaxed:

	STORE *ADDR = 3, STORE *ADDR = 4, STORE *DATA = y, q = LOAD *DATA


If ordering rules are relaxed, it must be assumed that accesses done inside an
interrupt disabled section may leak outside of it and may interleave with
accesses performed in an interrupt - and vice versa - unless implicit or
explicit barriers are used.

Normally this won't be a problem because the I/O accesses done inside such
sections will include synchronous load operations on strictly ordered I/O
registers that form implicit I/O barriers. If this isn't sufficient then an
mmiowb() may need to be used explicitly.


A similar situation may occur between an interrupt routine and two routines
running on separate CPUs that communicate with each other. If such a case is
likely, then interrupt-disabling locks should be used to guarantee ordering.


==========================
KERNEL I/O BARRIER EFFECTS
==========================

When accessing I/O memory, drivers should use the appropriate accessor
functions:

 (*) inX(), outX():

     These are intended to talk to I/O space rather than memory space, but
     that's primarily a CPU-specific concept. The i386 and x86_64 processors do
     indeed have special I/O space access cycles and instructions, but many
     CPUs don't have such a concept.

2564 2565
     The PCI bus, amongst others, defines an I/O space concept which - on such
     CPUs as i386 and x86_64 - readily maps to the CPU's concept of I/O
2566 2567 2568
     space.  However, it may also be mapped as a virtual I/O space in the CPU's
     memory map, particularly on those CPUs that don't support alternate I/O
     spaces.
2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586

     Accesses to this space may be fully synchronous (as on i386), but
     intermediary bridges (such as the PCI host bridge) may not fully honour
     that.

     They are guaranteed to be fully ordered with respect to each other.

     They are not guaranteed to be fully ordered with respect to other types of
     memory and I/O operation.

 (*) readX(), writeX():

     Whether these are guaranteed to be fully ordered and uncombined with
     respect to each other on the issuing CPU depends on the characteristics
     defined for the memory window through which they're accessing. On later
     i386 architecture machines, for example, this is controlled by way of the
     MTRR registers.

2587
     Ordinarily, these will be guaranteed to be fully ordered and uncombined,
2588 2589 2590 2591 2592 2593 2594 2595
     provided they're not accessing a prefetchable device.

     However, intermediary hardware (such as a PCI bridge) may indulge in
     deferral if it so wishes; to flush a store, a load from the same location
     is preferred[*], but a load from the same device or from configuration
     space should suffice for PCI.

     [*] NOTE! attempting to load from the same location as was written to may
2596 2597
	 cause a malfunction - consider the 16550 Rx/Tx serial registers for
	 example.
2598 2599 2600 2601 2602 2603 2604

     Used with prefetchable I/O memory, an mmiowb() barrier may be required to
     force stores to be ordered.

     Please refer to the PCI specification for more information on interactions
     between PCI transactions.

2605 2606 2607 2608 2609 2610 2611 2612 2613
 (*) readX_relaxed(), writeX_relaxed()

     These are similar to readX() and writeX(), but provide weaker memory
     ordering guarantees. Specifically, they do not guarantee ordering with
     respect to normal memory accesses (e.g. DMA buffers) nor do they guarantee
     ordering with respect to LOCK or UNLOCK operations. If the latter is
     required, an mmiowb() barrier can be used. Note that relaxed accesses to
     the same peripheral are guaranteed to be ordered with respect to each
     other.
2614 2615 2616

 (*) ioreadX(), iowriteX()

2617
     These will perform appropriately for the type of access they're actually
2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632
     doing, be it inX()/outX() or readX()/writeX().


========================================
ASSUMED MINIMUM EXECUTION ORDERING MODEL
========================================

It has to be assumed that the conceptual CPU is weakly-ordered but that it will
maintain the appearance of program causality with respect to itself.  Some CPUs
(such as i386 or x86_64) are more constrained than others (such as powerpc or
frv), and so the most relaxed case (namely DEC Alpha) must be assumed outside
of arch-specific code.

This means that it must be considered that the CPU will execute its instruction
stream in any order it feels like - or even in parallel - provided that if an
2633
instruction in the stream depends on an earlier instruction, then that
2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668
earlier instruction must be sufficiently complete[*] before the later
instruction may proceed; in other words: provided that the appearance of
causality is maintained.

 [*] Some instructions have more than one effect - such as changing the
     condition codes, changing registers or changing memory - and different
     instructions may depend on different effects.

A CPU may also discard any instruction sequence that winds up having no
ultimate effect.  For example, if two adjacent instructions both load an
immediate value into the same register, the first may be discarded.


Similarly, it has to be assumed that compiler might reorder the instruction
stream in any way it sees fit, again provided the appearance of causality is
maintained.


============================
THE EFFECTS OF THE CPU CACHE
============================

The way cached memory operations are perceived across the system is affected to
a certain extent by the caches that lie between CPUs and memory, and by the
memory coherence system that maintains the consistency of state in the system.

As far as the way a CPU interacts with another part of the system through the
caches goes, the memory system has to include the CPU's caches, and memory
barriers for the most part act at the interface between the CPU and its cache
(memory barriers logically act on the dotted line in the following diagram):

	    <--- CPU --->         :       <----------- Memory ----------->
	                          :
	+--------+    +--------+  :   +--------+    +-----------+
	|        |    |        |  :   |        |    |           |    +--------+
2669 2670
	|  CPU   |    | Memory |  :   | CPU    |    |           |    |        |
	|  Core  |--->| Access |----->| Cache  |<-->|           |    |        |
2671
	|        |    | Queue  |  :   |        |    |           |--->| Memory |
2672 2673
	|        |    |        |  :   |        |    |           |    |        |
	+--------+    +--------+  :   +--------+    |           |    |        |
2674 2675 2676 2677 2678 2679
	                          :                 | Cache     |    +--------+
	                          :                 | Coherency |
	                          :                 | Mechanism |    +--------+
	+--------+    +--------+  :   +--------+    |           |    |	      |
	|        |    |        |  :   |        |    |           |    |        |
	|  CPU   |    | Memory |  :   | CPU    |    |           |--->| Device |
2680 2681
	|  Core  |--->| Access |----->| Cache  |<-->|           |    |        |
	|        |    | Queue  |  :   |        |    |           |    |        |
2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722
	|        |    |        |  :   |        |    |           |    +--------+
	+--------+    +--------+  :   +--------+    +-----------+
	                          :
	                          :

Although any particular load or store may not actually appear outside of the
CPU that issued it since it may have been satisfied within the CPU's own cache,
it will still appear as if the full memory access had taken place as far as the
other CPUs are concerned since the cache coherency mechanisms will migrate the
cacheline over to the accessing CPU and propagate the effects upon conflict.

The CPU core may execute instructions in any order it deems fit, provided the
expected program causality appears to be maintained.  Some of the instructions
generate load and store operations which then go into the queue of memory
accesses to be performed.  The core may place these in the queue in any order
it wishes, and continue execution until it is forced to wait for an instruction
to complete.

What memory barriers are concerned with is controlling the order in which
accesses cross from the CPU side of things to the memory side of things, and
the order in which the effects are perceived to happen by the other observers
in the system.

[!] Memory barriers are _not_ needed within a given CPU, as CPUs always see
their own loads and stores as if they had happened in program order.

[!] MMIO or other device accesses may bypass the cache system.  This depends on
the properties of the memory window through which devices are accessed and/or
the use of any special device communication instructions the CPU may have.


CACHE COHERENCY
---------------

Life isn't quite as simple as it may appear above, however: for while the
caches are expected to be coherent, there's no guarantee that that coherency
will be ordered.  This means that whilst changes made on one CPU will
eventually become visible on all CPUs, there's no guarantee that they will
become apparent in the same order on those other CPUs.


2723 2724
Consider dealing with a system that has a pair of CPUs (1 & 2), each of which
has a pair of parallel data caches (CPU 1 has A/B, and CPU 2 has C/D):
2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762

	            :
	            :                          +--------+
	            :      +---------+         |        |
	+--------+  : +--->| Cache A |<------->|        |
	|        |  : |    +---------+         |        |
	|  CPU 1 |<---+                        |        |
	|        |  : |    +---------+         |        |
	+--------+  : +--->| Cache B |<------->|        |
	            :      +---------+         |        |
	            :                          | Memory |
	            :      +---------+         | System |
	+--------+  : +--->| Cache C |<------->|        |
	|        |  : |    +---------+         |        |
	|  CPU 2 |<---+                        |        |
	|        |  : |    +---------+         |        |
	+--------+  : +--->| Cache D |<------->|        |
	            :      +---------+         |        |
	            :                          +--------+
	            :

Imagine the system has the following properties:

 (*) an odd-numbered cache line may be in cache A, cache C or it may still be
     resident in memory;

 (*) an even-numbered cache line may be in cache B, cache D or it may still be
     resident in memory;

 (*) whilst the CPU core is interrogating one cache, the other cache may be
     making use of the bus to access the rest of the system - perhaps to
     displace a dirty cacheline or to do a speculative load;

 (*) each cache has a queue of operations that need to be applied to that cache
     to maintain coherency with the rest of the system;

 (*) the coherency queue is not flushed by normal loads to lines already
     present in the cache, even though the contents of the queue may
2763
     potentially affect those loads.
2764 2765 2766 2767 2768 2769 2770 2771 2772

Imagine, then, that two writes are made on the first CPU, with a write barrier
between them to guarantee that they will appear to reach that CPU's caches in
the requisite order:

	CPU 1		CPU 2		COMMENT
	===============	===============	=======================================
					u == 0, v == 1 and p == &u, q == &u
	v = 2;
2773
	smp_wmb();			Make sure change to v is visible before
2774 2775 2776 2777 2778 2779 2780
					 change to p
	<A:modify v=2>			v is now in cache A exclusively
	p = &v;
	<B:modify p=&v>			p is now in cache B exclusively

The write memory barrier forces the other CPUs in the system to perceive that
the local CPU's caches have apparently been updated in the correct order.  But
2781
now imagine that the second CPU wants to read those values:
2782 2783 2784 2785 2786 2787 2788

	CPU 1		CPU 2		COMMENT
	===============	===============	=======================================
	...
			q = p;
			x = *q;

2789
The above pair of reads may then fail to happen in the expected order, as the
2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800
cacheline holding p may get updated in one of the second CPU's caches whilst
the update to the cacheline holding v is delayed in the other of the second
CPU's caches by some other cache event:

	CPU 1		CPU 2		COMMENT
	===============	===============	=======================================
					u == 0, v == 1 and p == &u, q == &u
	v = 2;
	smp_wmb();
	<A:modify v=2>	<C:busy>
			<C:queue v=2>
2801
	p = &v;		q = p;
2802 2803
			<D:request p>
	<B:modify p=&v>	<D:commit p=&v>
2804
			<D:read p>
2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825
			x = *q;
			<C:read *q>	Reads from v before v updated in cache
			<C:unbusy>
			<C:commit v=2>

Basically, whilst both cachelines will be updated on CPU 2 eventually, there's
no guarantee that, without intervention, the order of update will be the same
as that committed on CPU 1.


To intervene, we need to interpolate a data dependency barrier or a read
barrier between the loads.  This will force the cache to commit its coherency
queue before processing any further requests:

	CPU 1		CPU 2		COMMENT
	===============	===============	=======================================
					u == 0, v == 1 and p == &u, q == &u
	v = 2;
	smp_wmb();
	<A:modify v=2>	<C:busy>
			<C:queue v=2>
2826
	p = &v;		q = p;
2827 2828
			<D:request p>
	<B:modify p=&v>	<D:commit p=&v>
2829
			<D:read p>
2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842
			smp_read_barrier_depends()
			<C:unbusy>
			<C:commit v=2>
			x = *q;
			<C:read *q>	Reads from v after v updated in cache


This sort of problem can be encountered on DEC Alpha processors as they have a
split cache that improves performance by making better use of the data bus.
Whilst most CPUs do imply a data dependency barrier on the read when a memory
access depends on a read, not all do, so it may not be relied on.

Other CPUs may also have split caches, but must coordinate between the various
2843
cachelets for normal memory accesses.  The semantics of the Alpha removes the
2844
need for coordination in the absence of memory barriers.
2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858


CACHE COHERENCY VS DMA
----------------------

Not all systems maintain cache coherency with respect to devices doing DMA.  In
such cases, a device attempting DMA may obtain stale data from RAM because
dirty cache lines may be resident in the caches of various CPUs, and may not
have been written back to RAM yet.  To deal with this, the appropriate part of
the kernel must flush the overlapping bits of cache on each CPU (and maybe
invalidate them as well).

In addition, the data DMA'd to RAM by a device may be overwritten by dirty
cache lines being written back to RAM from a CPU's cache after the device has
2859 2860 2861 2862
installed its own data, or cache lines present in the CPU's cache may simply
obscure the fact that RAM has been updated, until at such time as the cacheline
is discarded from the CPU's cache and reloaded.  To deal with this, the
appropriate part of the kernel must invalidate the overlapping bits of the
2863 2864 2865 2866 2867 2868 2869 2870 2871
cache on each CPU.

See Documentation/cachetlb.txt for more information on cache management.


CACHE COHERENCY VS MMIO
-----------------------

Memory mapped I/O usually takes place through memory locations that are part of
2872
a window in the CPU's memory space that has different properties assigned than
2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887
the usual RAM directed window.

Amongst these properties is usually the fact that such accesses bypass the
caching entirely and go directly to the device buses.  This means MMIO accesses
may, in effect, overtake accesses to cached memory that were emitted earlier.
A memory barrier isn't sufficient in such a case, but rather the cache must be
flushed between the cached memory write and the MMIO access if the two are in
any way dependent.


=========================
THE THINGS CPUS GET UP TO
=========================

A programmer might take it for granted that the CPU will perform memory
2888
operations in exactly the order specified, so that if the CPU is, for example,
2889 2890
given the following piece of code to execute:

2891 2892 2893 2894 2895
	a = READ_ONCE(*A);
	WRITE_ONCE(*B, b);
	c = READ_ONCE(*C);
	d = READ_ONCE(*D);
	WRITE_ONCE(*E, e);
2896

2897
they would then expect that the CPU will complete the memory operation for each
2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913
instruction before moving on to the next one, leading to a definite sequence of
operations as seen by external observers in the system:

	LOAD *A, STORE *B, LOAD *C, LOAD *D, STORE *E.


Reality is, of course, much messier.  With many CPUs and compilers, the above
assumption doesn't hold because:

 (*) loads are more likely to need to be completed immediately to permit
     execution progress, whereas stores can often be deferred without a
     problem;

 (*) loads may be done speculatively, and the result discarded should it prove
     to have been unnecessary;

2914 2915
 (*) loads may be done speculatively, leading to the result having been fetched
     at the wrong time in the expected sequence of events;
2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941

 (*) the order of the memory accesses may be rearranged to promote better use
     of the CPU buses and caches;

 (*) loads and stores may be combined to improve performance when talking to
     memory or I/O hardware that can do batched accesses of adjacent locations,
     thus cutting down on transaction setup costs (memory and PCI devices may
     both be able to do this); and

 (*) the CPU's data cache may affect the ordering, and whilst cache-coherency
     mechanisms may alleviate this - once the store has actually hit the cache
     - there's no guarantee that the coherency management will be propagated in
     order to other CPUs.

So what another CPU, say, might actually observe from the above piece of code
is:

	LOAD *A, ..., LOAD {*C,*D}, STORE *E, STORE *B

	(Where "LOAD {*C,*D}" is a combined load)


However, it is guaranteed that a CPU will be self-consistent: it will see its
_own_ accesses appear to be correctly ordered, without the need for a memory
barrier.  For instance with the following code:

2942 2943 2944 2945 2946 2947
	U = READ_ONCE(*A);
	WRITE_ONCE(*A, V);
	WRITE_ONCE(*A, W);
	X = READ_ONCE(*A);
	WRITE_ONCE(*A, Y);
	Z = READ_ONCE(*A);
2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962

and assuming no intervention by an external influence, it can be assumed that
the final result will appear to be:

	U == the original value of *A
	X == W
	Z == Y
	*A == Y

The code above may cause the CPU to generate the full sequence of memory
accesses:

	U=LOAD *A, STORE *A=V, STORE *A=W, X=LOAD *A, STORE *A=Y, Z=LOAD *A

in that order, but, without intervention, the sequence may have almost any
2963 2964 2965 2966 2967 2968 2969 2970
combination of elements combined or discarded, provided the program's view
of the world remains consistent.  Note that READ_ONCE() and WRITE_ONCE()
are -not- optional in the above example, as there are architectures
where a given CPU might reorder successive loads to the same location.
On such architectures, READ_ONCE() and WRITE_ONCE() do whatever is
necessary to prevent this, for example, on Itanium the volatile casts
used by READ_ONCE() and WRITE_ONCE() cause GCC to emit the special ld.acq
and st.rel instructions (respectively) that prevent such reordering.
2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983

The compiler may also combine, discard or defer elements of the sequence before
the CPU even sees them.

For instance:

	*A = V;
	*A = W;

may be reduced to:

	*A = W;

2984
since, without either a write barrier or an WRITE_ONCE(), it can be
2985
assumed that the effect of the storage of V to *A is lost.  Similarly:
2986 2987 2988 2989

	*A = Y;
	Z = *A;

2990 2991
may, without a memory barrier or an READ_ONCE() and WRITE_ONCE(), be
reduced to:
2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003

	*A = Y;
	Z = Y;

and the LOAD operation never appear outside of the CPU.


AND THEN THERE'S THE ALPHA
--------------------------

The DEC Alpha CPU is one of the most relaxed CPUs there is.  Not only that,
some versions of the Alpha CPU have a split data cache, permitting them to have
3004
two semantically-related cache lines updated at separate times.  This is where
3005 3006 3007 3008
the data dependency barrier really becomes necessary as this synchronises both
caches with the memory coherence system, thus making it seem like pointer
changes vs new data occur in the right order.

3009
The Alpha defines the Linux kernel's memory barrier model.
3010 3011 3012 3013

See the subsection on "Cache Coherency" above.


3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028
============
EXAMPLE USES
============

CIRCULAR BUFFERS
----------------

Memory barriers can be used to implement circular buffering without the need
of a lock to serialise the producer with the consumer.  See:

	Documentation/circular-buffers.txt

for details.


3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083
==========
REFERENCES
==========

Alpha AXP Architecture Reference Manual, Second Edition (Sites & Witek,
Digital Press)
	Chapter 5.2: Physical Address Space Characteristics
	Chapter 5.4: Caches and Write Buffers
	Chapter 5.5: Data Sharing
	Chapter 5.6: Read/Write Ordering

AMD64 Architecture Programmer's Manual Volume 2: System Programming
	Chapter 7.1: Memory-Access Ordering
	Chapter 7.4: Buffering and Combining Memory Writes

IA-32 Intel Architecture Software Developer's Manual, Volume 3:
System Programming Guide
	Chapter 7.1: Locked Atomic Operations
	Chapter 7.2: Memory Ordering
	Chapter 7.4: Serializing Instructions

The SPARC Architecture Manual, Version 9
	Chapter 8: Memory Models
	Appendix D: Formal Specification of the Memory Models
	Appendix J: Programming with the Memory Models

UltraSPARC Programmer Reference Manual
	Chapter 5: Memory Accesses and Cacheability
	Chapter 15: Sparc-V9 Memory Models

UltraSPARC III Cu User's Manual
	Chapter 9: Memory Models

UltraSPARC IIIi Processor User's Manual
	Chapter 8: Memory Models

UltraSPARC Architecture 2005
	Chapter 9: Memory
	Appendix D: Formal Specifications of the Memory Models

UltraSPARC T1 Supplement to the UltraSPARC Architecture 2005
	Chapter 8: Memory Models
	Appendix F: Caches and Cache Coherency

Solaris Internals, Core Kernel Architecture, p63-68:
	Chapter 3.3: Hardware Considerations for Locks and
			Synchronization

Unix Systems for Modern Architectures, Symmetric Multiprocessing and Caching
for Kernel Programmers:
	Chapter 13: Other Memory Models

Intel Itanium Architecture Software Developer's Manual: Volume 1:
	Section 2.6: Speculation
	Section 4.4: Memory Access