Loading...
Note: File does not exist in v6.13.7.
1Please note that the "What is RCU?" LWN series is an excellent place
2to start learning about RCU:
3
41. What is RCU, Fundamentally? http://lwn.net/Articles/262464/
52. What is RCU? Part 2: Usage http://lwn.net/Articles/263130/
63. RCU part 3: the RCU API http://lwn.net/Articles/264090/
74. The RCU API, 2010 Edition http://lwn.net/Articles/418853/
8 2010 Big API Table http://lwn.net/Articles/419086/
95. The RCU API, 2014 Edition http://lwn.net/Articles/609904/
10 2014 Big API Table http://lwn.net/Articles/609973/
11
12
13What is RCU?
14
15RCU is a synchronization mechanism that was added to the Linux kernel
16during the 2.5 development effort that is optimized for read-mostly
17situations. Although RCU is actually quite simple once you understand it,
18getting there can sometimes be a challenge. Part of the problem is that
19most of the past descriptions of RCU have been written with the mistaken
20assumption that there is "one true way" to describe RCU. Instead,
21the experience has been that different people must take different paths
22to arrive at an understanding of RCU. This document provides several
23different paths, as follows:
24
251. RCU OVERVIEW
262. WHAT IS RCU'S CORE API?
273. WHAT ARE SOME EXAMPLE USES OF CORE RCU API?
284. WHAT IF MY UPDATING THREAD CANNOT BLOCK?
295. WHAT ARE SOME SIMPLE IMPLEMENTATIONS OF RCU?
306. ANALOGY WITH READER-WRITER LOCKING
317. FULL LIST OF RCU APIs
328. ANSWERS TO QUICK QUIZZES
33
34People who prefer starting with a conceptual overview should focus on
35Section 1, though most readers will profit by reading this section at
36some point. People who prefer to start with an API that they can then
37experiment with should focus on Section 2. People who prefer to start
38with example uses should focus on Sections 3 and 4. People who need to
39understand the RCU implementation should focus on Section 5, then dive
40into the kernel source code. People who reason best by analogy should
41focus on Section 6. Section 7 serves as an index to the docbook API
42documentation, and Section 8 is the traditional answer key.
43
44So, start with the section that makes the most sense to you and your
45preferred method of learning. If you need to know everything about
46everything, feel free to read the whole thing -- but if you are really
47that type of person, you have perused the source code and will therefore
48never need this document anyway. ;-)
49
50
511. RCU OVERVIEW
52
53The basic idea behind RCU is to split updates into "removal" and
54"reclamation" phases. The removal phase removes references to data items
55within a data structure (possibly by replacing them with references to
56new versions of these data items), and can run concurrently with readers.
57The reason that it is safe to run the removal phase concurrently with
58readers is the semantics of modern CPUs guarantee that readers will see
59either the old or the new version of the data structure rather than a
60partially updated reference. The reclamation phase does the work of reclaiming
61(e.g., freeing) the data items removed from the data structure during the
62removal phase. Because reclaiming data items can disrupt any readers
63concurrently referencing those data items, the reclamation phase must
64not start until readers no longer hold references to those data items.
65
66Splitting the update into removal and reclamation phases permits the
67updater to perform the removal phase immediately, and to defer the
68reclamation phase until all readers active during the removal phase have
69completed, either by blocking until they finish or by registering a
70callback that is invoked after they finish. Only readers that are active
71during the removal phase need be considered, because any reader starting
72after the removal phase will be unable to gain a reference to the removed
73data items, and therefore cannot be disrupted by the reclamation phase.
74
75So the typical RCU update sequence goes something like the following:
76
77a. Remove pointers to a data structure, so that subsequent
78 readers cannot gain a reference to it.
79
80b. Wait for all previous readers to complete their RCU read-side
81 critical sections.
82
83c. At this point, there cannot be any readers who hold references
84 to the data structure, so it now may safely be reclaimed
85 (e.g., kfree()d).
86
87Step (b) above is the key idea underlying RCU's deferred destruction.
88The ability to wait until all readers are done allows RCU readers to
89use much lighter-weight synchronization, in some cases, absolutely no
90synchronization at all. In contrast, in more conventional lock-based
91schemes, readers must use heavy-weight synchronization in order to
92prevent an updater from deleting the data structure out from under them.
93This is because lock-based updaters typically update data items in place,
94and must therefore exclude readers. In contrast, RCU-based updaters
95typically take advantage of the fact that writes to single aligned
96pointers are atomic on modern CPUs, allowing atomic insertion, removal,
97and replacement of data items in a linked structure without disrupting
98readers. Concurrent RCU readers can then continue accessing the old
99versions, and can dispense with the atomic operations, memory barriers,
100and communications cache misses that are so expensive on present-day
101SMP computer systems, even in absence of lock contention.
102
103In the three-step procedure shown above, the updater is performing both
104the removal and the reclamation step, but it is often helpful for an
105entirely different thread to do the reclamation, as is in fact the case
106in the Linux kernel's directory-entry cache (dcache). Even if the same
107thread performs both the update step (step (a) above) and the reclamation
108step (step (c) above), it is often helpful to think of them separately.
109For example, RCU readers and updaters need not communicate at all,
110but RCU provides implicit low-overhead communication between readers
111and reclaimers, namely, in step (b) above.
112
113So how the heck can a reclaimer tell when a reader is done, given
114that readers are not doing any sort of synchronization operations???
115Read on to learn about how RCU's API makes this easy.
116
117
1182. WHAT IS RCU'S CORE API?
119
120The core RCU API is quite small:
121
122a. rcu_read_lock()
123b. rcu_read_unlock()
124c. synchronize_rcu() / call_rcu()
125d. rcu_assign_pointer()
126e. rcu_dereference()
127
128There are many other members of the RCU API, but the rest can be
129expressed in terms of these five, though most implementations instead
130express synchronize_rcu() in terms of the call_rcu() callback API.
131
132The five core RCU APIs are described below, the other 18 will be enumerated
133later. See the kernel docbook documentation for more info, or look directly
134at the function header comments.
135
136rcu_read_lock()
137
138 void rcu_read_lock(void);
139
140 Used by a reader to inform the reclaimer that the reader is
141 entering an RCU read-side critical section. It is illegal
142 to block while in an RCU read-side critical section, though
143 kernels built with CONFIG_PREEMPT_RCU can preempt RCU
144 read-side critical sections. Any RCU-protected data structure
145 accessed during an RCU read-side critical section is guaranteed to
146 remain unreclaimed for the full duration of that critical section.
147 Reference counts may be used in conjunction with RCU to maintain
148 longer-term references to data structures.
149
150rcu_read_unlock()
151
152 void rcu_read_unlock(void);
153
154 Used by a reader to inform the reclaimer that the reader is
155 exiting an RCU read-side critical section. Note that RCU
156 read-side critical sections may be nested and/or overlapping.
157
158synchronize_rcu()
159
160 void synchronize_rcu(void);
161
162 Marks the end of updater code and the beginning of reclaimer
163 code. It does this by blocking until all pre-existing RCU
164 read-side critical sections on all CPUs have completed.
165 Note that synchronize_rcu() will -not- necessarily wait for
166 any subsequent RCU read-side critical sections to complete.
167 For example, consider the following sequence of events:
168
169 CPU 0 CPU 1 CPU 2
170 ----------------- ------------------------- ---------------
171 1. rcu_read_lock()
172 2. enters synchronize_rcu()
173 3. rcu_read_lock()
174 4. rcu_read_unlock()
175 5. exits synchronize_rcu()
176 6. rcu_read_unlock()
177
178 To reiterate, synchronize_rcu() waits only for ongoing RCU
179 read-side critical sections to complete, not necessarily for
180 any that begin after synchronize_rcu() is invoked.
181
182 Of course, synchronize_rcu() does not necessarily return
183 -immediately- after the last pre-existing RCU read-side critical
184 section completes. For one thing, there might well be scheduling
185 delays. For another thing, many RCU implementations process
186 requests in batches in order to improve efficiencies, which can
187 further delay synchronize_rcu().
188
189 Since synchronize_rcu() is the API that must figure out when
190 readers are done, its implementation is key to RCU. For RCU
191 to be useful in all but the most read-intensive situations,
192 synchronize_rcu()'s overhead must also be quite small.
193
194 The call_rcu() API is a callback form of synchronize_rcu(),
195 and is described in more detail in a later section. Instead of
196 blocking, it registers a function and argument which are invoked
197 after all ongoing RCU read-side critical sections have completed.
198 This callback variant is particularly useful in situations where
199 it is illegal to block or where update-side performance is
200 critically important.
201
202 However, the call_rcu() API should not be used lightly, as use
203 of the synchronize_rcu() API generally results in simpler code.
204 In addition, the synchronize_rcu() API has the nice property
205 of automatically limiting update rate should grace periods
206 be delayed. This property results in system resilience in face
207 of denial-of-service attacks. Code using call_rcu() should limit
208 update rate in order to gain this same sort of resilience. See
209 checklist.txt for some approaches to limiting the update rate.
210
211rcu_assign_pointer()
212
213 typeof(p) rcu_assign_pointer(p, typeof(p) v);
214
215 Yes, rcu_assign_pointer() -is- implemented as a macro, though it
216 would be cool to be able to declare a function in this manner.
217 (Compiler experts will no doubt disagree.)
218
219 The updater uses this function to assign a new value to an
220 RCU-protected pointer, in order to safely communicate the change
221 in value from the updater to the reader. This function returns
222 the new value, and also executes any memory-barrier instructions
223 required for a given CPU architecture.
224
225 Perhaps just as important, it serves to document (1) which
226 pointers are protected by RCU and (2) the point at which a
227 given structure becomes accessible to other CPUs. That said,
228 rcu_assign_pointer() is most frequently used indirectly, via
229 the _rcu list-manipulation primitives such as list_add_rcu().
230
231rcu_dereference()
232
233 typeof(p) rcu_dereference(p);
234
235 Like rcu_assign_pointer(), rcu_dereference() must be implemented
236 as a macro.
237
238 The reader uses rcu_dereference() to fetch an RCU-protected
239 pointer, which returns a value that may then be safely
240 dereferenced. Note that rcu_dereference() does not actually
241 dereference the pointer, instead, it protects the pointer for
242 later dereferencing. It also executes any needed memory-barrier
243 instructions for a given CPU architecture. Currently, only Alpha
244 needs memory barriers within rcu_dereference() -- on other CPUs,
245 it compiles to nothing, not even a compiler directive.
246
247 Common coding practice uses rcu_dereference() to copy an
248 RCU-protected pointer to a local variable, then dereferences
249 this local variable, for example as follows:
250
251 p = rcu_dereference(head.next);
252 return p->data;
253
254 However, in this case, one could just as easily combine these
255 into one statement:
256
257 return rcu_dereference(head.next)->data;
258
259 If you are going to be fetching multiple fields from the
260 RCU-protected structure, using the local variable is of
261 course preferred. Repeated rcu_dereference() calls look
262 ugly, do not guarantee that the same pointer will be returned
263 if an update happened while in the critical section, and incur
264 unnecessary overhead on Alpha CPUs.
265
266 Note that the value returned by rcu_dereference() is valid
267 only within the enclosing RCU read-side critical section.
268 For example, the following is -not- legal:
269
270 rcu_read_lock();
271 p = rcu_dereference(head.next);
272 rcu_read_unlock();
273 x = p->address; /* BUG!!! */
274 rcu_read_lock();
275 y = p->data; /* BUG!!! */
276 rcu_read_unlock();
277
278 Holding a reference from one RCU read-side critical section
279 to another is just as illegal as holding a reference from
280 one lock-based critical section to another! Similarly,
281 using a reference outside of the critical section in which
282 it was acquired is just as illegal as doing so with normal
283 locking.
284
285 As with rcu_assign_pointer(), an important function of
286 rcu_dereference() is to document which pointers are protected by
287 RCU, in particular, flagging a pointer that is subject to changing
288 at any time, including immediately after the rcu_dereference().
289 And, again like rcu_assign_pointer(), rcu_dereference() is
290 typically used indirectly, via the _rcu list-manipulation
291 primitives, such as list_for_each_entry_rcu().
292
293The following diagram shows how each API communicates among the
294reader, updater, and reclaimer.
295
296
297 rcu_assign_pointer()
298 +--------+
299 +---------------------->| reader |---------+
300 | +--------+ |
301 | | |
302 | | | Protect:
303 | | | rcu_read_lock()
304 | | | rcu_read_unlock()
305 | rcu_dereference() | |
306 +---------+ | |
307 | updater |<---------------------+ |
308 +---------+ V
309 | +-----------+
310 +----------------------------------->| reclaimer |
311 +-----------+
312 Defer:
313 synchronize_rcu() & call_rcu()
314
315
316The RCU infrastructure observes the time sequence of rcu_read_lock(),
317rcu_read_unlock(), synchronize_rcu(), and call_rcu() invocations in
318order to determine when (1) synchronize_rcu() invocations may return
319to their callers and (2) call_rcu() callbacks may be invoked. Efficient
320implementations of the RCU infrastructure make heavy use of batching in
321order to amortize their overhead over many uses of the corresponding APIs.
322
323There are no fewer than three RCU mechanisms in the Linux kernel; the
324diagram above shows the first one, which is by far the most commonly used.
325The rcu_dereference() and rcu_assign_pointer() primitives are used for
326all three mechanisms, but different defer and protect primitives are
327used as follows:
328
329 Defer Protect
330
331a. synchronize_rcu() rcu_read_lock() / rcu_read_unlock()
332 call_rcu() rcu_dereference()
333
334b. synchronize_rcu_bh() rcu_read_lock_bh() / rcu_read_unlock_bh()
335 call_rcu_bh() rcu_dereference_bh()
336
337c. synchronize_sched() rcu_read_lock_sched() / rcu_read_unlock_sched()
338 call_rcu_sched() preempt_disable() / preempt_enable()
339 local_irq_save() / local_irq_restore()
340 hardirq enter / hardirq exit
341 NMI enter / NMI exit
342 rcu_dereference_sched()
343
344These three mechanisms are used as follows:
345
346a. RCU applied to normal data structures.
347
348b. RCU applied to networking data structures that may be subjected
349 to remote denial-of-service attacks.
350
351c. RCU applied to scheduler and interrupt/NMI-handler tasks.
352
353Again, most uses will be of (a). The (b) and (c) cases are important
354for specialized uses, but are relatively uncommon.
355
356
3573. WHAT ARE SOME EXAMPLE USES OF CORE RCU API?
358
359This section shows a simple use of the core RCU API to protect a
360global pointer to a dynamically allocated structure. More-typical
361uses of RCU may be found in listRCU.txt, arrayRCU.txt, and NMI-RCU.txt.
362
363 struct foo {
364 int a;
365 char b;
366 long c;
367 };
368 DEFINE_SPINLOCK(foo_mutex);
369
370 struct foo __rcu *gbl_foo;
371
372 /*
373 * Create a new struct foo that is the same as the one currently
374 * pointed to by gbl_foo, except that field "a" is replaced
375 * with "new_a". Points gbl_foo to the new structure, and
376 * frees up the old structure after a grace period.
377 *
378 * Uses rcu_assign_pointer() to ensure that concurrent readers
379 * see the initialized version of the new structure.
380 *
381 * Uses synchronize_rcu() to ensure that any readers that might
382 * have references to the old structure complete before freeing
383 * the old structure.
384 */
385 void foo_update_a(int new_a)
386 {
387 struct foo *new_fp;
388 struct foo *old_fp;
389
390 new_fp = kmalloc(sizeof(*new_fp), GFP_KERNEL);
391 spin_lock(&foo_mutex);
392 old_fp = rcu_dereference_protected(gbl_foo, lockdep_is_held(&foo_mutex));
393 *new_fp = *old_fp;
394 new_fp->a = new_a;
395 rcu_assign_pointer(gbl_foo, new_fp);
396 spin_unlock(&foo_mutex);
397 synchronize_rcu();
398 kfree(old_fp);
399 }
400
401 /*
402 * Return the value of field "a" of the current gbl_foo
403 * structure. Use rcu_read_lock() and rcu_read_unlock()
404 * to ensure that the structure does not get deleted out
405 * from under us, and use rcu_dereference() to ensure that
406 * we see the initialized version of the structure (important
407 * for DEC Alpha and for people reading the code).
408 */
409 int foo_get_a(void)
410 {
411 int retval;
412
413 rcu_read_lock();
414 retval = rcu_dereference(gbl_foo)->a;
415 rcu_read_unlock();
416 return retval;
417 }
418
419So, to sum up:
420
421o Use rcu_read_lock() and rcu_read_unlock() to guard RCU
422 read-side critical sections.
423
424o Within an RCU read-side critical section, use rcu_dereference()
425 to dereference RCU-protected pointers.
426
427o Use some solid scheme (such as locks or semaphores) to
428 keep concurrent updates from interfering with each other.
429
430o Use rcu_assign_pointer() to update an RCU-protected pointer.
431 This primitive protects concurrent readers from the updater,
432 -not- concurrent updates from each other! You therefore still
433 need to use locking (or something similar) to keep concurrent
434 rcu_assign_pointer() primitives from interfering with each other.
435
436o Use synchronize_rcu() -after- removing a data element from an
437 RCU-protected data structure, but -before- reclaiming/freeing
438 the data element, in order to wait for the completion of all
439 RCU read-side critical sections that might be referencing that
440 data item.
441
442See checklist.txt for additional rules to follow when using RCU.
443And again, more-typical uses of RCU may be found in listRCU.txt,
444arrayRCU.txt, and NMI-RCU.txt.
445
446
4474. WHAT IF MY UPDATING THREAD CANNOT BLOCK?
448
449In the example above, foo_update_a() blocks until a grace period elapses.
450This is quite simple, but in some cases one cannot afford to wait so
451long -- there might be other high-priority work to be done.
452
453In such cases, one uses call_rcu() rather than synchronize_rcu().
454The call_rcu() API is as follows:
455
456 void call_rcu(struct rcu_head * head,
457 void (*func)(struct rcu_head *head));
458
459This function invokes func(head) after a grace period has elapsed.
460This invocation might happen from either softirq or process context,
461so the function is not permitted to block. The foo struct needs to
462have an rcu_head structure added, perhaps as follows:
463
464 struct foo {
465 int a;
466 char b;
467 long c;
468 struct rcu_head rcu;
469 };
470
471The foo_update_a() function might then be written as follows:
472
473 /*
474 * Create a new struct foo that is the same as the one currently
475 * pointed to by gbl_foo, except that field "a" is replaced
476 * with "new_a". Points gbl_foo to the new structure, and
477 * frees up the old structure after a grace period.
478 *
479 * Uses rcu_assign_pointer() to ensure that concurrent readers
480 * see the initialized version of the new structure.
481 *
482 * Uses call_rcu() to ensure that any readers that might have
483 * references to the old structure complete before freeing the
484 * old structure.
485 */
486 void foo_update_a(int new_a)
487 {
488 struct foo *new_fp;
489 struct foo *old_fp;
490
491 new_fp = kmalloc(sizeof(*new_fp), GFP_KERNEL);
492 spin_lock(&foo_mutex);
493 old_fp = rcu_dereference_protected(gbl_foo, lockdep_is_held(&foo_mutex));
494 *new_fp = *old_fp;
495 new_fp->a = new_a;
496 rcu_assign_pointer(gbl_foo, new_fp);
497 spin_unlock(&foo_mutex);
498 call_rcu(&old_fp->rcu, foo_reclaim);
499 }
500
501The foo_reclaim() function might appear as follows:
502
503 void foo_reclaim(struct rcu_head *rp)
504 {
505 struct foo *fp = container_of(rp, struct foo, rcu);
506
507 foo_cleanup(fp->a);
508
509 kfree(fp);
510 }
511
512The container_of() primitive is a macro that, given a pointer into a
513struct, the type of the struct, and the pointed-to field within the
514struct, returns a pointer to the beginning of the struct.
515
516The use of call_rcu() permits the caller of foo_update_a() to
517immediately regain control, without needing to worry further about the
518old version of the newly updated element. It also clearly shows the
519RCU distinction between updater, namely foo_update_a(), and reclaimer,
520namely foo_reclaim().
521
522The summary of advice is the same as for the previous section, except
523that we are now using call_rcu() rather than synchronize_rcu():
524
525o Use call_rcu() -after- removing a data element from an
526 RCU-protected data structure in order to register a callback
527 function that will be invoked after the completion of all RCU
528 read-side critical sections that might be referencing that
529 data item.
530
531If the callback for call_rcu() is not doing anything more than calling
532kfree() on the structure, you can use kfree_rcu() instead of call_rcu()
533to avoid having to write your own callback:
534
535 kfree_rcu(old_fp, rcu);
536
537Again, see checklist.txt for additional rules governing the use of RCU.
538
539
5405. WHAT ARE SOME SIMPLE IMPLEMENTATIONS OF RCU?
541
542One of the nice things about RCU is that it has extremely simple "toy"
543implementations that are a good first step towards understanding the
544production-quality implementations in the Linux kernel. This section
545presents two such "toy" implementations of RCU, one that is implemented
546in terms of familiar locking primitives, and another that more closely
547resembles "classic" RCU. Both are way too simple for real-world use,
548lacking both functionality and performance. However, they are useful
549in getting a feel for how RCU works. See kernel/rcupdate.c for a
550production-quality implementation, and see:
551
552 http://www.rdrop.com/users/paulmck/RCU
553
554for papers describing the Linux kernel RCU implementation. The OLS'01
555and OLS'02 papers are a good introduction, and the dissertation provides
556more details on the current implementation as of early 2004.
557
558
5595A. "TOY" IMPLEMENTATION #1: LOCKING
560
561This section presents a "toy" RCU implementation that is based on
562familiar locking primitives. Its overhead makes it a non-starter for
563real-life use, as does its lack of scalability. It is also unsuitable
564for realtime use, since it allows scheduling latency to "bleed" from
565one read-side critical section to another. It also assumes recursive
566reader-writer locks: If you try this with non-recursive locks, and
567you allow nested rcu_read_lock() calls, you can deadlock.
568
569However, it is probably the easiest implementation to relate to, so is
570a good starting point.
571
572It is extremely simple:
573
574 static DEFINE_RWLOCK(rcu_gp_mutex);
575
576 void rcu_read_lock(void)
577 {
578 read_lock(&rcu_gp_mutex);
579 }
580
581 void rcu_read_unlock(void)
582 {
583 read_unlock(&rcu_gp_mutex);
584 }
585
586 void synchronize_rcu(void)
587 {
588 write_lock(&rcu_gp_mutex);
589 write_unlock(&rcu_gp_mutex);
590 }
591
592[You can ignore rcu_assign_pointer() and rcu_dereference() without missing
593much. But here are simplified versions anyway. And whatever you do,
594don't forget about them when submitting patches making use of RCU!]
595
596 #define rcu_assign_pointer(p, v) \
597 ({ \
598 smp_store_release(&(p), (v)); \
599 })
600
601 #define rcu_dereference(p) \
602 ({ \
603 typeof(p) _________p1 = READ_ONCE(p); \
604 (_________p1); \
605 })
606
607
608The rcu_read_lock() and rcu_read_unlock() primitive read-acquire
609and release a global reader-writer lock. The synchronize_rcu()
610primitive write-acquires this same lock, then immediately releases
611it. This means that once synchronize_rcu() exits, all RCU read-side
612critical sections that were in progress before synchronize_rcu() was
613called are guaranteed to have completed -- there is no way that
614synchronize_rcu() would have been able to write-acquire the lock
615otherwise.
616
617It is possible to nest rcu_read_lock(), since reader-writer locks may
618be recursively acquired. Note also that rcu_read_lock() is immune
619from deadlock (an important property of RCU). The reason for this is
620that the only thing that can block rcu_read_lock() is a synchronize_rcu().
621But synchronize_rcu() does not acquire any locks while holding rcu_gp_mutex,
622so there can be no deadlock cycle.
623
624Quick Quiz #1: Why is this argument naive? How could a deadlock
625 occur when using this algorithm in a real-world Linux
626 kernel? How could this deadlock be avoided?
627
628
6295B. "TOY" EXAMPLE #2: CLASSIC RCU
630
631This section presents a "toy" RCU implementation that is based on
632"classic RCU". It is also short on performance (but only for updates) and
633on features such as hotplug CPU and the ability to run in CONFIG_PREEMPT
634kernels. The definitions of rcu_dereference() and rcu_assign_pointer()
635are the same as those shown in the preceding section, so they are omitted.
636
637 void rcu_read_lock(void) { }
638
639 void rcu_read_unlock(void) { }
640
641 void synchronize_rcu(void)
642 {
643 int cpu;
644
645 for_each_possible_cpu(cpu)
646 run_on(cpu);
647 }
648
649Note that rcu_read_lock() and rcu_read_unlock() do absolutely nothing.
650This is the great strength of classic RCU in a non-preemptive kernel:
651read-side overhead is precisely zero, at least on non-Alpha CPUs.
652And there is absolutely no way that rcu_read_lock() can possibly
653participate in a deadlock cycle!
654
655The implementation of synchronize_rcu() simply schedules itself on each
656CPU in turn. The run_on() primitive can be implemented straightforwardly
657in terms of the sched_setaffinity() primitive. Of course, a somewhat less
658"toy" implementation would restore the affinity upon completion rather
659than just leaving all tasks running on the last CPU, but when I said
660"toy", I meant -toy-!
661
662So how the heck is this supposed to work???
663
664Remember that it is illegal to block while in an RCU read-side critical
665section. Therefore, if a given CPU executes a context switch, we know
666that it must have completed all preceding RCU read-side critical sections.
667Once -all- CPUs have executed a context switch, then -all- preceding
668RCU read-side critical sections will have completed.
669
670So, suppose that we remove a data item from its structure and then invoke
671synchronize_rcu(). Once synchronize_rcu() returns, we are guaranteed
672that there are no RCU read-side critical sections holding a reference
673to that data item, so we can safely reclaim it.
674
675Quick Quiz #2: Give an example where Classic RCU's read-side
676 overhead is -negative-.
677
678Quick Quiz #3: If it is illegal to block in an RCU read-side
679 critical section, what the heck do you do in
680 PREEMPT_RT, where normal spinlocks can block???
681
682
6836. ANALOGY WITH READER-WRITER LOCKING
684
685Although RCU can be used in many different ways, a very common use of
686RCU is analogous to reader-writer locking. The following unified
687diff shows how closely related RCU and reader-writer locking can be.
688
689 @@ -5,5 +5,5 @@ struct el {
690 int data;
691 /* Other data fields */
692 };
693 -rwlock_t listmutex;
694 +spinlock_t listmutex;
695 struct el head;
696
697 @@ -13,15 +14,15 @@
698 struct list_head *lp;
699 struct el *p;
700
701 - read_lock(&listmutex);
702 - list_for_each_entry(p, head, lp) {
703 + rcu_read_lock();
704 + list_for_each_entry_rcu(p, head, lp) {
705 if (p->key == key) {
706 *result = p->data;
707 - read_unlock(&listmutex);
708 + rcu_read_unlock();
709 return 1;
710 }
711 }
712 - read_unlock(&listmutex);
713 + rcu_read_unlock();
714 return 0;
715 }
716
717 @@ -29,15 +30,16 @@
718 {
719 struct el *p;
720
721 - write_lock(&listmutex);
722 + spin_lock(&listmutex);
723 list_for_each_entry(p, head, lp) {
724 if (p->key == key) {
725 - list_del(&p->list);
726 - write_unlock(&listmutex);
727 + list_del_rcu(&p->list);
728 + spin_unlock(&listmutex);
729 + synchronize_rcu();
730 kfree(p);
731 return 1;
732 }
733 }
734 - write_unlock(&listmutex);
735 + spin_unlock(&listmutex);
736 return 0;
737 }
738
739Or, for those who prefer a side-by-side listing:
740
741 1 struct el { 1 struct el {
742 2 struct list_head list; 2 struct list_head list;
743 3 long key; 3 long key;
744 4 spinlock_t mutex; 4 spinlock_t mutex;
745 5 int data; 5 int data;
746 6 /* Other data fields */ 6 /* Other data fields */
747 7 }; 7 };
748 8 rwlock_t listmutex; 8 spinlock_t listmutex;
749 9 struct el head; 9 struct el head;
750
751 1 int search(long key, int *result) 1 int search(long key, int *result)
752 2 { 2 {
753 3 struct list_head *lp; 3 struct list_head *lp;
754 4 struct el *p; 4 struct el *p;
755 5 5
756 6 read_lock(&listmutex); 6 rcu_read_lock();
757 7 list_for_each_entry(p, head, lp) { 7 list_for_each_entry_rcu(p, head, lp) {
758 8 if (p->key == key) { 8 if (p->key == key) {
759 9 *result = p->data; 9 *result = p->data;
76010 read_unlock(&listmutex); 10 rcu_read_unlock();
76111 return 1; 11 return 1;
76212 } 12 }
76313 } 13 }
76414 read_unlock(&listmutex); 14 rcu_read_unlock();
76515 return 0; 15 return 0;
76616 } 16 }
767
768 1 int delete(long key) 1 int delete(long key)
769 2 { 2 {
770 3 struct el *p; 3 struct el *p;
771 4 4
772 5 write_lock(&listmutex); 5 spin_lock(&listmutex);
773 6 list_for_each_entry(p, head, lp) { 6 list_for_each_entry(p, head, lp) {
774 7 if (p->key == key) { 7 if (p->key == key) {
775 8 list_del(&p->list); 8 list_del_rcu(&p->list);
776 9 write_unlock(&listmutex); 9 spin_unlock(&listmutex);
777 10 synchronize_rcu();
77810 kfree(p); 11 kfree(p);
77911 return 1; 12 return 1;
78012 } 13 }
78113 } 14 }
78214 write_unlock(&listmutex); 15 spin_unlock(&listmutex);
78315 return 0; 16 return 0;
78416 } 17 }
785
786Either way, the differences are quite small. Read-side locking moves
787to rcu_read_lock() and rcu_read_unlock, update-side locking moves from
788a reader-writer lock to a simple spinlock, and a synchronize_rcu()
789precedes the kfree().
790
791However, there is one potential catch: the read-side and update-side
792critical sections can now run concurrently. In many cases, this will
793not be a problem, but it is necessary to check carefully regardless.
794For example, if multiple independent list updates must be seen as
795a single atomic update, converting to RCU will require special care.
796
797Also, the presence of synchronize_rcu() means that the RCU version of
798delete() can now block. If this is a problem, there is a callback-based
799mechanism that never blocks, namely call_rcu() or kfree_rcu(), that can
800be used in place of synchronize_rcu().
801
802
8037. FULL LIST OF RCU APIs
804
805The RCU APIs are documented in docbook-format header comments in the
806Linux-kernel source code, but it helps to have a full list of the
807APIs, since there does not appear to be a way to categorize them
808in docbook. Here is the list, by category.
809
810RCU list traversal:
811
812 list_entry_rcu
813 list_first_entry_rcu
814 list_next_rcu
815 list_for_each_entry_rcu
816 list_for_each_entry_continue_rcu
817 hlist_first_rcu
818 hlist_next_rcu
819 hlist_pprev_rcu
820 hlist_for_each_entry_rcu
821 hlist_for_each_entry_rcu_bh
822 hlist_for_each_entry_continue_rcu
823 hlist_for_each_entry_continue_rcu_bh
824 hlist_nulls_first_rcu
825 hlist_nulls_for_each_entry_rcu
826 hlist_bl_first_rcu
827 hlist_bl_for_each_entry_rcu
828
829RCU pointer/list update:
830
831 rcu_assign_pointer
832 list_add_rcu
833 list_add_tail_rcu
834 list_del_rcu
835 list_replace_rcu
836 hlist_add_behind_rcu
837 hlist_add_before_rcu
838 hlist_add_head_rcu
839 hlist_del_rcu
840 hlist_del_init_rcu
841 hlist_replace_rcu
842 list_splice_init_rcu()
843 hlist_nulls_del_init_rcu
844 hlist_nulls_del_rcu
845 hlist_nulls_add_head_rcu
846 hlist_bl_add_head_rcu
847 hlist_bl_del_init_rcu
848 hlist_bl_del_rcu
849 hlist_bl_set_first_rcu
850
851RCU: Critical sections Grace period Barrier
852
853 rcu_read_lock synchronize_net rcu_barrier
854 rcu_read_unlock synchronize_rcu
855 rcu_dereference synchronize_rcu_expedited
856 rcu_read_lock_held call_rcu
857 rcu_dereference_check kfree_rcu
858 rcu_dereference_protected
859
860bh: Critical sections Grace period Barrier
861
862 rcu_read_lock_bh call_rcu_bh rcu_barrier_bh
863 rcu_read_unlock_bh synchronize_rcu_bh
864 rcu_dereference_bh synchronize_rcu_bh_expedited
865 rcu_dereference_bh_check
866 rcu_dereference_bh_protected
867 rcu_read_lock_bh_held
868
869sched: Critical sections Grace period Barrier
870
871 rcu_read_lock_sched synchronize_sched rcu_barrier_sched
872 rcu_read_unlock_sched call_rcu_sched
873 [preempt_disable] synchronize_sched_expedited
874 [and friends]
875 rcu_read_lock_sched_notrace
876 rcu_read_unlock_sched_notrace
877 rcu_dereference_sched
878 rcu_dereference_sched_check
879 rcu_dereference_sched_protected
880 rcu_read_lock_sched_held
881
882
883SRCU: Critical sections Grace period Barrier
884
885 srcu_read_lock synchronize_srcu srcu_barrier
886 srcu_read_unlock call_srcu
887 srcu_dereference synchronize_srcu_expedited
888 srcu_dereference_check
889 srcu_read_lock_held
890
891SRCU: Initialization/cleanup
892 DEFINE_SRCU
893 DEFINE_STATIC_SRCU
894 init_srcu_struct
895 cleanup_srcu_struct
896
897All: lockdep-checked RCU-protected pointer access
898
899 rcu_access_pointer
900 rcu_dereference_raw
901 RCU_LOCKDEP_WARN
902 rcu_sleep_check
903 RCU_NONIDLE
904
905See the comment headers in the source code (or the docbook generated
906from them) for more information.
907
908However, given that there are no fewer than four families of RCU APIs
909in the Linux kernel, how do you choose which one to use? The following
910list can be helpful:
911
912a. Will readers need to block? If so, you need SRCU.
913
914b. What about the -rt patchset? If readers would need to block
915 in an non-rt kernel, you need SRCU. If readers would block
916 in a -rt kernel, but not in a non-rt kernel, SRCU is not
917 necessary. (The -rt patchset turns spinlocks into sleeplocks,
918 hence this distinction.)
919
920c. Do you need to treat NMI handlers, hardirq handlers,
921 and code segments with preemption disabled (whether
922 via preempt_disable(), local_irq_save(), local_bh_disable(),
923 or some other mechanism) as if they were explicit RCU readers?
924 If so, RCU-sched is the only choice that will work for you.
925
926d. Do you need RCU grace periods to complete even in the face
927 of softirq monopolization of one or more of the CPUs? For
928 example, is your code subject to network-based denial-of-service
929 attacks? If so, you need RCU-bh.
930
931e. Is your workload too update-intensive for normal use of
932 RCU, but inappropriate for other synchronization mechanisms?
933 If so, consider SLAB_TYPESAFE_BY_RCU (which was originally
934 named SLAB_DESTROY_BY_RCU). But please be careful!
935
936f. Do you need read-side critical sections that are respected
937 even though they are in the middle of the idle loop, during
938 user-mode execution, or on an offlined CPU? If so, SRCU is the
939 only choice that will work for you.
940
941g. Otherwise, use RCU.
942
943Of course, this all assumes that you have determined that RCU is in fact
944the right tool for your job.
945
946
9478. ANSWERS TO QUICK QUIZZES
948
949Quick Quiz #1: Why is this argument naive? How could a deadlock
950 occur when using this algorithm in a real-world Linux
951 kernel? [Referring to the lock-based "toy" RCU
952 algorithm.]
953
954Answer: Consider the following sequence of events:
955
956 1. CPU 0 acquires some unrelated lock, call it
957 "problematic_lock", disabling irq via
958 spin_lock_irqsave().
959
960 2. CPU 1 enters synchronize_rcu(), write-acquiring
961 rcu_gp_mutex.
962
963 3. CPU 0 enters rcu_read_lock(), but must wait
964 because CPU 1 holds rcu_gp_mutex.
965
966 4. CPU 1 is interrupted, and the irq handler
967 attempts to acquire problematic_lock.
968
969 The system is now deadlocked.
970
971 One way to avoid this deadlock is to use an approach like
972 that of CONFIG_PREEMPT_RT, where all normal spinlocks
973 become blocking locks, and all irq handlers execute in
974 the context of special tasks. In this case, in step 4
975 above, the irq handler would block, allowing CPU 1 to
976 release rcu_gp_mutex, avoiding the deadlock.
977
978 Even in the absence of deadlock, this RCU implementation
979 allows latency to "bleed" from readers to other
980 readers through synchronize_rcu(). To see this,
981 consider task A in an RCU read-side critical section
982 (thus read-holding rcu_gp_mutex), task B blocked
983 attempting to write-acquire rcu_gp_mutex, and
984 task C blocked in rcu_read_lock() attempting to
985 read_acquire rcu_gp_mutex. Task A's RCU read-side
986 latency is holding up task C, albeit indirectly via
987 task B.
988
989 Realtime RCU implementations therefore use a counter-based
990 approach where tasks in RCU read-side critical sections
991 cannot be blocked by tasks executing synchronize_rcu().
992
993Quick Quiz #2: Give an example where Classic RCU's read-side
994 overhead is -negative-.
995
996Answer: Imagine a single-CPU system with a non-CONFIG_PREEMPT
997 kernel where a routing table is used by process-context
998 code, but can be updated by irq-context code (for example,
999 by an "ICMP REDIRECT" packet). The usual way of handling
1000 this would be to have the process-context code disable
1001 interrupts while searching the routing table. Use of
1002 RCU allows such interrupt-disabling to be dispensed with.
1003 Thus, without RCU, you pay the cost of disabling interrupts,
1004 and with RCU you don't.
1005
1006 One can argue that the overhead of RCU in this
1007 case is negative with respect to the single-CPU
1008 interrupt-disabling approach. Others might argue that
1009 the overhead of RCU is merely zero, and that replacing
1010 the positive overhead of the interrupt-disabling scheme
1011 with the zero-overhead RCU scheme does not constitute
1012 negative overhead.
1013
1014 In real life, of course, things are more complex. But
1015 even the theoretical possibility of negative overhead for
1016 a synchronization primitive is a bit unexpected. ;-)
1017
1018Quick Quiz #3: If it is illegal to block in an RCU read-side
1019 critical section, what the heck do you do in
1020 PREEMPT_RT, where normal spinlocks can block???
1021
1022Answer: Just as PREEMPT_RT permits preemption of spinlock
1023 critical sections, it permits preemption of RCU
1024 read-side critical sections. It also permits
1025 spinlocks blocking while in RCU read-side critical
1026 sections.
1027
1028 Why the apparent inconsistency? Because it is it
1029 possible to use priority boosting to keep the RCU
1030 grace periods short if need be (for example, if running
1031 short of memory). In contrast, if blocking waiting
1032 for (say) network reception, there is no way to know
1033 what should be boosted. Especially given that the
1034 process we need to boost might well be a human being
1035 who just went out for a pizza or something. And although
1036 a computer-operated cattle prod might arouse serious
1037 interest, it might also provoke serious objections.
1038 Besides, how does the computer know what pizza parlor
1039 the human being went to???
1040
1041
1042ACKNOWLEDGEMENTS
1043
1044My thanks to the people who helped make this human-readable, including
1045Jon Walpole, Josh Triplett, Serge Hallyn, Suzanne Wood, and Alan Stern.
1046
1047
1048For more information, see http://www.rdrop.com/users/paulmck/RCU.