1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
2 "http://www.w3.org/TR/html4/loose.dtd">
4 <head><title>A Tour Through RCU's Requirements [LWN.net]</title>
5 <meta HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=utf-8">
7 <h1>A Tour Through RCU's Requirements</h1>
9 <p>Copyright IBM Corporation, 2015</p>
10 <p>Author: Paul E. McKenney</p>
11 <p><i>The initial version of this document appeared in the
12 <a href="https://lwn.net/">LWN</a> articles
13 <a href="https://lwn.net/Articles/652156/">here</a>,
14 <a href="https://lwn.net/Articles/652677/">here</a>, and
15 <a href="https://lwn.net/Articles/653326/">here</a>.</i></p>
20 Read-copy update (RCU) is a synchronization mechanism that is often
21 used as a replacement for reader-writer locking.
22 RCU is unusual in that updaters do not block readers,
23 which means that RCU's read-side primitives can be exceedingly fast
25 In addition, updaters can make useful forward progress concurrently
27 However, all this concurrency between RCU readers and updaters does raise
28 the question of exactly what RCU readers are doing, which in turn
29 raises the question of exactly what RCU's requirements are.
32 This document therefore summarizes RCU's requirements, and can be thought
33 of as an informal, high-level specification for RCU.
34 It is important to understand that RCU's specification is primarily
36 in fact, I learned about many of these requirements the hard way.
37 This situation might cause some consternation, however, not only
38 has this learning process been a lot of fun, but it has also been
39 a great privilege to work with so many people willing to apply
40 technologies in interesting new ways.
43 All that aside, here are the categories of currently known RCU requirements:
47 <li> <a href="#Fundamental Requirements">
48 Fundamental Requirements</a>
49 <li> <a href="#Fundamental Non-Requirements">Fundamental Non-Requirements</a>
50 <li> <a href="#Parallelism Facts of Life">
51 Parallelism Facts of Life</a>
52 <li> <a href="#Quality-of-Implementation Requirements">
53 Quality-of-Implementation Requirements</a>
54 <li> <a href="#Linux Kernel Complications">
55 Linux Kernel Complications</a>
56 <li> <a href="#Software-Engineering Requirements">
57 Software-Engineering Requirements</a>
58 <li> <a href="#Other RCU Flavors">
60 <li> <a href="#Possible Future Changes">
61 Possible Future Changes</a>
65 This is followed by a <a href="#Summary">summary</a>,
66 however, the answers to each quick quiz immediately follows the quiz.
67 Select the big white space with your mouse to see the answer.
69 <h2><a name="Fundamental Requirements">Fundamental Requirements</a></h2>
72 RCU's fundamental requirements are the closest thing RCU has to hard
73 mathematical requirements.
77 <li> <a href="#Grace-Period Guarantee">
78 Grace-Period Guarantee</a>
79 <li> <a href="#Publish-Subscribe Guarantee">
80 Publish-Subscribe Guarantee</a>
81 <li> <a href="#Memory-Barrier Guarantees">
82 Memory-Barrier Guarantees</a>
83 <li> <a href="#RCU Primitives Guaranteed to Execute Unconditionally">
84 RCU Primitives Guaranteed to Execute Unconditionally</a>
85 <li> <a href="#Guaranteed Read-to-Write Upgrade">
86 Guaranteed Read-to-Write Upgrade</a>
89 <h3><a name="Grace-Period Guarantee">Grace-Period Guarantee</a></h3>
92 RCU's grace-period guarantee is unusual in being premeditated:
93 Jack Slingwine and I had this guarantee firmly in mind when we started
94 work on RCU (then called “rclock”) in the early 1990s.
95 That said, the past two decades of experience with RCU have produced
96 a much more detailed understanding of this guarantee.
99 RCU's grace-period guarantee allows updaters to wait for the completion
100 of all pre-existing RCU read-side critical sections.
101 An RCU read-side critical section
102 begins with the marker <tt>rcu_read_lock()</tt> and ends with
103 the marker <tt>rcu_read_unlock()</tt>.
104 These markers may be nested, and RCU treats a nested set as one
105 big RCU read-side critical section.
106 Production-quality implementations of <tt>rcu_read_lock()</tt> and
107 <tt>rcu_read_unlock()</tt> are extremely lightweight, and in
108 fact have exactly zero overhead in Linux kernels built for production
109 use with <tt>CONFIG_PREEMPT=n</tt>.
112 This guarantee allows ordering to be enforced with extremely low
113 overhead to readers, for example:
127 11 void thread1(void)
130 14 synchronize_rcu();
137 Because the <tt>synchronize_rcu()</tt> on line 14 waits for
138 all pre-existing readers, any instance of <tt>thread0()</tt> that
139 loads a value of zero from <tt>x</tt> must complete before
140 <tt>thread1()</tt> stores to <tt>y</tt>, so that instance must
141 also load a value of zero from <tt>y</tt>.
142 Similarly, any instance of <tt>thread0()</tt> that loads a value of
143 one from <tt>y</tt> must have started after the
144 <tt>synchronize_rcu()</tt> started, and must therefore also load
145 a value of one from <tt>x</tt>.
146 Therefore, the outcome:
149 (r1 == 0 && r2 == 1)
155 <tr><th> </th></tr>
156 <tr><th align="left">Quick Quiz:</th></tr>
159 You said that updaters can make useful forward progress concurrently
160 with readers, but pre-existing readers will block
161 <tt>synchronize_rcu()</tt>!!!
162 Just who are you trying to fool???
164 <tr><th align="left">Answer:</th></tr>
165 <tr><td bgcolor="#ffffff"><font color="ffffff">
166 First, if updaters do not wish to be blocked by readers, they can use
167 <tt>call_rcu()</tt> or <tt>kfree_rcu()</tt>, which will
169 Second, even when using <tt>synchronize_rcu()</tt>, the other
170 update-side code does run concurrently with readers, whether
173 <tr><td> </td></tr>
177 This scenario resembles one of the first uses of RCU in
178 <a href="https://en.wikipedia.org/wiki/DYNIX">DYNIX/ptx</a>,
179 which managed a distributed lock manager's transition into
180 a state suitable for handling recovery from node failure,
181 more or less as follows:
185 1 #define STATE_NORMAL 0
186 2 #define STATE_WANT_RECOVERY 1
187 3 #define STATE_RECOVERING 2
188 4 #define STATE_WANT_NORMAL 3
190 6 int state = STATE_NORMAL;
192 8 void do_something_dlm(void)
197 13 state_snap = READ_ONCE(state);
198 14 if (state_snap == STATE_NORMAL)
201 17 do_something_carefully();
202 18 rcu_read_unlock();
205 21 void start_recovery(void)
207 23 WRITE_ONCE(state, STATE_WANT_RECOVERY);
208 24 synchronize_rcu();
209 25 WRITE_ONCE(state, STATE_RECOVERING);
211 27 WRITE_ONCE(state, STATE_WANT_NORMAL);
212 28 synchronize_rcu();
213 29 WRITE_ONCE(state, STATE_NORMAL);
219 The RCU read-side critical section in <tt>do_something_dlm()</tt>
220 works with the <tt>synchronize_rcu()</tt> in <tt>start_recovery()</tt>
221 to guarantee that <tt>do_something()</tt> never runs concurrently
222 with <tt>recovery()</tt>, but with little or no synchronization
223 overhead in <tt>do_something_dlm()</tt>.
226 <tr><th> </th></tr>
227 <tr><th align="left">Quick Quiz:</th></tr>
229 Why is the <tt>synchronize_rcu()</tt> on line 28 needed?
231 <tr><th align="left">Answer:</th></tr>
232 <tr><td bgcolor="#ffffff"><font color="ffffff">
233 Without that extra grace period, memory reordering could result in
234 <tt>do_something_dlm()</tt> executing <tt>do_something()</tt>
235 concurrently with the last bits of <tt>recovery()</tt>.
237 <tr><td> </td></tr>
241 In order to avoid fatal problems such as deadlocks,
242 an RCU read-side critical section must not contain calls to
243 <tt>synchronize_rcu()</tt>.
244 Similarly, an RCU read-side critical section must not
245 contain anything that waits, directly or indirectly, on completion of
246 an invocation of <tt>synchronize_rcu()</tt>.
249 Although RCU's grace-period guarantee is useful in and of itself, with
250 <a href="https://lwn.net/Articles/573497/">quite a few use cases</a>,
251 it would be good to be able to use RCU to coordinate read-side
252 access to linked data structures.
253 For this, the grace-period guarantee is not sufficient, as can
254 be seen in function <tt>add_gp_buggy()</tt> below.
255 We will look at the reader's code later, but in the meantime, just think of
256 the reader as locklessly picking up the <tt>gp</tt> pointer,
257 and, if the value loaded is non-<tt>NULL</tt>, locklessly accessing the
258 <tt>->a</tt> and <tt>->b</tt> fields.
262 1 bool add_gp_buggy(int a, int b)
264 3 p = kmalloc(sizeof(*p), GFP_KERNEL);
267 6 spin_lock(&gp_lock);
268 7 if (rcu_access_pointer(gp)) {
269 8 spin_unlock(&gp_lock);
274 13 gp = p; /* ORDERING BUG */
275 14 spin_unlock(&gp_lock);
282 The problem is that both the compiler and weakly ordered CPUs are within
283 their rights to reorder this code as follows:
287 1 bool add_gp_buggy_optimized(int a, int b)
289 3 p = kmalloc(sizeof(*p), GFP_KERNEL);
292 6 spin_lock(&gp_lock);
293 7 if (rcu_access_pointer(gp)) {
294 8 spin_unlock(&gp_lock);
297 <b>11 gp = p; /* ORDERING BUG */
300 14 spin_unlock(&gp_lock);
307 If an RCU reader fetches <tt>gp</tt> just after
308 <tt>add_gp_buggy_optimized</tt> executes line 11,
309 it will see garbage in the <tt>->a</tt> and <tt>->b</tt>
311 And this is but one of many ways in which compiler and hardware optimizations
313 Therefore, we clearly need some way to prevent the compiler and the CPU from
314 reordering in this manner, which brings us to the publish-subscribe
315 guarantee discussed in the next section.
317 <h3><a name="Publish-Subscribe Guarantee">Publish/Subscribe Guarantee</a></h3>
320 RCU's publish-subscribe guarantee allows data to be inserted
321 into a linked data structure without disrupting RCU readers.
322 The updater uses <tt>rcu_assign_pointer()</tt> to insert the
323 new data, and readers use <tt>rcu_dereference()</tt> to
324 access data, whether new or old.
325 The following shows an example of insertion:
329 1 bool add_gp(int a, int b)
331 3 p = kmalloc(sizeof(*p), GFP_KERNEL);
334 6 spin_lock(&gp_lock);
335 7 if (rcu_access_pointer(gp)) {
336 8 spin_unlock(&gp_lock);
341 13 rcu_assign_pointer(gp, p);
342 14 spin_unlock(&gp_lock);
349 The <tt>rcu_assign_pointer()</tt> on line 13 is conceptually
350 equivalent to a simple assignment statement, but also guarantees
351 that its assignment will
352 happen after the two assignments in lines 11 and 12,
353 similar to the C11 <tt>memory_order_release</tt> store operation.
354 It also prevents any number of “interesting” compiler
355 optimizations, for example, the use of <tt>gp</tt> as a scratch
356 location immediately preceding the assignment.
359 <tr><th> </th></tr>
360 <tr><th align="left">Quick Quiz:</th></tr>
362 But <tt>rcu_assign_pointer()</tt> does nothing to prevent the
363 two assignments to <tt>p->a</tt> and <tt>p->b</tt>
364 from being reordered.
365 Can't that also cause problems?
367 <tr><th align="left">Answer:</th></tr>
368 <tr><td bgcolor="#ffffff"><font color="ffffff">
370 The readers cannot see either of these two fields until
371 the assignment to <tt>gp</tt>, by which time both fields are
373 So reordering the assignments
374 to <tt>p->a</tt> and <tt>p->b</tt> cannot possibly
377 <tr><td> </td></tr>
381 It is tempting to assume that the reader need not do anything special
382 to control its accesses to the RCU-protected data,
383 as shown in <tt>do_something_gp_buggy()</tt> below:
387 1 bool do_something_gp_buggy(void)
390 4 p = gp; /* OPTIMIZATIONS GALORE!!! */
392 6 do_something(p->a, p->b);
396 10 rcu_read_unlock();
403 However, this temptation must be resisted because there are a
404 surprisingly large number of ways that the compiler
406 <a href="https://h71000.www7.hp.com/wizard/wiz_2637.html">DEC Alpha CPUs</a>)
407 can trip this code up.
408 For but one example, if the compiler were short of registers, it
409 might choose to refetch from <tt>gp</tt> rather than keeping
410 a separate copy in <tt>p</tt> as follows:
414 1 bool do_something_gp_buggy_optimized(void)
417 4 if (gp) { /* OPTIMIZATIONS GALORE!!! */
418 <b> 5 do_something(gp->a, gp->b);</b>
429 If this function ran concurrently with a series of updates that
430 replaced the current structure with a new one,
431 the fetches of <tt>gp->a</tt>
432 and <tt>gp->b</tt> might well come from two different structures,
433 which could cause serious confusion.
434 To prevent this (and much else besides), <tt>do_something_gp()</tt> uses
435 <tt>rcu_dereference()</tt> to fetch from <tt>gp</tt>:
439 1 bool do_something_gp(void)
442 4 p = rcu_dereference(gp);
444 6 do_something(p->a, p->b);
448 10 rcu_read_unlock();
455 The <tt>rcu_dereference()</tt> uses volatile casts and (for DEC Alpha)
456 memory barriers in the Linux kernel.
458 <a href="http://www.rdrop.com/users/paulmck/RCU/consume.2015.07.13a.pdf">high-quality implementation of C11 <tt>memory_order_consume</tt> [PDF]</a>
459 ever appear, then <tt>rcu_dereference()</tt> could be implemented
460 as a <tt>memory_order_consume</tt> load.
461 Regardless of the exact implementation, a pointer fetched by
462 <tt>rcu_dereference()</tt> may not be used outside of the
463 outermost RCU read-side critical section containing that
464 <tt>rcu_dereference()</tt>, unless protection of
465 the corresponding data element has been passed from RCU to some
466 other synchronization mechanism, most commonly locking or
467 <a href="https://www.kernel.org/doc/Documentation/RCU/rcuref.txt">reference counting</a>.
470 In short, updaters use <tt>rcu_assign_pointer()</tt> and readers
471 use <tt>rcu_dereference()</tt>, and these two RCU API elements
472 work together to ensure that readers have a consistent view of
473 newly added data elements.
476 Of course, it is also necessary to remove elements from RCU-protected
477 data structures, for example, using the following process:
480 <li> Remove the data element from the enclosing structure.
481 <li> Wait for all pre-existing RCU read-side critical sections
482 to complete (because only pre-existing readers can possibly have
483 a reference to the newly removed data element).
484 <li> At this point, only the updater has a reference to the
485 newly removed data element, so it can safely reclaim
486 the data element, for example, by passing it to <tt>kfree()</tt>.
489 This process is implemented by <tt>remove_gp_synchronous()</tt>:
493 1 bool remove_gp_synchronous(void)
497 5 spin_lock(&gp_lock);
498 6 p = rcu_access_pointer(gp);
500 8 spin_unlock(&gp_lock);
503 11 rcu_assign_pointer(gp, NULL);
504 12 spin_unlock(&gp_lock);
505 13 synchronize_rcu();
513 This function is straightforward, with line 13 waiting for a grace
514 period before line 14 frees the old data element.
515 This waiting ensures that readers will reach line 7 of
516 <tt>do_something_gp()</tt> before the data element referenced by
518 The <tt>rcu_access_pointer()</tt> on line 6 is similar to
519 <tt>rcu_dereference()</tt>, except that:
522 <li> The value returned by <tt>rcu_access_pointer()</tt>
523 cannot be dereferenced.
524 If you want to access the value pointed to as well as
525 the pointer itself, use <tt>rcu_dereference()</tt>
526 instead of <tt>rcu_access_pointer()</tt>.
527 <li> The call to <tt>rcu_access_pointer()</tt> need not be
529 In contrast, <tt>rcu_dereference()</tt> must either be
530 within an RCU read-side critical section or in a code
531 segment where the pointer cannot change, for example, in
532 code protected by the corresponding update-side lock.
536 <tr><th> </th></tr>
537 <tr><th align="left">Quick Quiz:</th></tr>
539 Without the <tt>rcu_dereference()</tt> or the
540 <tt>rcu_access_pointer()</tt>, what destructive optimizations
541 might the compiler make use of?
543 <tr><th align="left">Answer:</th></tr>
544 <tr><td bgcolor="#ffffff"><font color="ffffff">
545 Let's start with what happens to <tt>do_something_gp()</tt>
546 if it fails to use <tt>rcu_dereference()</tt>.
547 It could reuse a value formerly fetched from this same pointer.
548 It could also fetch the pointer from <tt>gp</tt> in a byte-at-a-time
549 manner, resulting in <i>load tearing</i>, in turn resulting a bytewise
550 mash-up of two distinct pointer values.
551 It might even use value-speculation optimizations, where it makes
552 a wrong guess, but by the time it gets around to checking the
553 value, an update has changed the pointer to match the wrong guess.
554 Too bad about any dereferences that returned pre-initialization garbage
558 <p><font color="ffffff">
559 For <tt>remove_gp_synchronous()</tt>, as long as all modifications
560 to <tt>gp</tt> are carried out while holding <tt>gp_lock</tt>,
561 the above optimizations are harmless.
562 However, <tt>sparse</tt> will complain if you
563 define <tt>gp</tt> with <tt>__rcu</tt> and then
564 access it without using
565 either <tt>rcu_access_pointer()</tt> or <tt>rcu_dereference()</tt>.
567 <tr><td> </td></tr>
571 In short, RCU's publish-subscribe guarantee is provided by the combination
572 of <tt>rcu_assign_pointer()</tt> and <tt>rcu_dereference()</tt>.
573 This guarantee allows data elements to be safely added to RCU-protected
574 linked data structures without disrupting RCU readers.
575 This guarantee can be used in combination with the grace-period
576 guarantee to also allow data elements to be removed from RCU-protected
577 linked data structures, again without disrupting RCU readers.
580 This guarantee was only partially premeditated.
581 DYNIX/ptx used an explicit memory barrier for publication, but had nothing
582 resembling <tt>rcu_dereference()</tt> for subscription, nor did it
583 have anything resembling the <tt>smp_read_barrier_depends()</tt>
584 that was later subsumed into <tt>rcu_dereference()</tt> and later
585 still into <tt>READ_ONCE()</tt>.
586 The need for these operations made itself known quite suddenly at a
587 late-1990s meeting with the DEC Alpha architects, back in the days when
588 DEC was still a free-standing company.
589 It took the Alpha architects a good hour to convince me that any sort
590 of barrier would ever be needed, and it then took me a good <i>two</i> hours
591 to convince them that their documentation did not make this point clear.
592 More recent work with the C and C++ standards committees have provided
593 much education on tricks and traps from the compiler.
594 In short, compilers were much less tricky in the early 1990s, but in
595 2015, don't even think about omitting <tt>rcu_dereference()</tt>!
597 <h3><a name="Memory-Barrier Guarantees">Memory-Barrier Guarantees</a></h3>
600 The previous section's simple linked-data-structure scenario clearly
601 demonstrates the need for RCU's stringent memory-ordering guarantees on
602 systems with more than one CPU:
605 <li> Each CPU that has an RCU read-side critical section that
606 begins before <tt>synchronize_rcu()</tt> starts is
607 guaranteed to execute a full memory barrier between the time
608 that the RCU read-side critical section ends and the time that
609 <tt>synchronize_rcu()</tt> returns.
610 Without this guarantee, a pre-existing RCU read-side critical section
611 might hold a reference to the newly removed <tt>struct foo</tt>
612 after the <tt>kfree()</tt> on line 14 of
613 <tt>remove_gp_synchronous()</tt>.
614 <li> Each CPU that has an RCU read-side critical section that ends
615 after <tt>synchronize_rcu()</tt> returns is guaranteed
616 to execute a full memory barrier between the time that
617 <tt>synchronize_rcu()</tt> begins and the time that the RCU
618 read-side critical section begins.
619 Without this guarantee, a later RCU read-side critical section
620 running after the <tt>kfree()</tt> on line 14 of
621 <tt>remove_gp_synchronous()</tt> might
622 later run <tt>do_something_gp()</tt> and find the
623 newly deleted <tt>struct foo</tt>.
624 <li> If the task invoking <tt>synchronize_rcu()</tt> remains
625 on a given CPU, then that CPU is guaranteed to execute a full
626 memory barrier sometime during the execution of
627 <tt>synchronize_rcu()</tt>.
628 This guarantee ensures that the <tt>kfree()</tt> on
629 line 14 of <tt>remove_gp_synchronous()</tt> really does
630 execute after the removal on line 11.
631 <li> If the task invoking <tt>synchronize_rcu()</tt> migrates
632 among a group of CPUs during that invocation, then each of the
633 CPUs in that group is guaranteed to execute a full memory barrier
634 sometime during the execution of <tt>synchronize_rcu()</tt>.
635 This guarantee also ensures that the <tt>kfree()</tt> on
636 line 14 of <tt>remove_gp_synchronous()</tt> really does
637 execute after the removal on
638 line 11, but also in the case where the thread executing the
639 <tt>synchronize_rcu()</tt> migrates in the meantime.
643 <tr><th> </th></tr>
644 <tr><th align="left">Quick Quiz:</th></tr>
646 Given that multiple CPUs can start RCU read-side critical sections
647 at any time without any ordering whatsoever, how can RCU possibly
648 tell whether or not a given RCU read-side critical section starts
649 before a given instance of <tt>synchronize_rcu()</tt>?
651 <tr><th align="left">Answer:</th></tr>
652 <tr><td bgcolor="#ffffff"><font color="ffffff">
653 If RCU cannot tell whether or not a given
654 RCU read-side critical section starts before a
655 given instance of <tt>synchronize_rcu()</tt>,
656 then it must assume that the RCU read-side critical section
658 In other words, a given instance of <tt>synchronize_rcu()</tt>
659 can avoid waiting on a given RCU read-side critical section only
660 if it can prove that <tt>synchronize_rcu()</tt> started first.
663 <p><font color="ffffff">
664 A related question is “When <tt>rcu_read_lock()</tt>
665 doesn't generate any code, why does it matter how it relates
666 to a grace period?”
667 The answer is that it is not the relationship of
668 <tt>rcu_read_lock()</tt> itself that is important, but rather
669 the relationship of the code within the enclosed RCU read-side
670 critical section to the code preceding and following the
672 If we take this viewpoint, then a given RCU read-side critical
673 section begins before a given grace period when some access
674 preceding the grace period observes the effect of some access
675 within the critical section, in which case none of the accesses
676 within the critical section may observe the effects of any
677 access following the grace period.
680 <p><font color="ffffff">
681 As of late 2016, mathematical models of RCU take this
682 viewpoint, for example, see slides 62 and 63
684 <a href="http://www2.rdrop.com/users/paulmck/scalability/paper/LinuxMM.2016.10.04c.LCE.pdf">2016 LinuxCon EU</a>
687 <tr><td> </td></tr>
691 <tr><th> </th></tr>
692 <tr><th align="left">Quick Quiz:</th></tr>
694 The first and second guarantees require unbelievably strict ordering!
695 Are all these memory barriers <i> really</i> required?
697 <tr><th align="left">Answer:</th></tr>
698 <tr><td bgcolor="#ffffff"><font color="ffffff">
699 Yes, they really are required.
700 To see why the first guarantee is required, consider the following
705 <li> <font color="ffffff">
706 CPU 1: <tt>rcu_read_lock()</tt>
708 <li> <font color="ffffff">
709 CPU 1: <tt>q = rcu_dereference(gp);
710 /* Very likely to return p. */</tt>
712 <li> <font color="ffffff">
713 CPU 0: <tt>list_del_rcu(p);</tt>
715 <li> <font color="ffffff">
716 CPU 0: <tt>synchronize_rcu()</tt> starts.
718 <li> <font color="ffffff">
719 CPU 1: <tt>do_something_with(q->a);
720 /* No smp_mb(), so might happen after kfree(). */</tt>
722 <li> <font color="ffffff">
723 CPU 1: <tt>rcu_read_unlock()</tt>
725 <li> <font color="ffffff">
726 CPU 0: <tt>synchronize_rcu()</tt> returns.
728 <li> <font color="ffffff">
729 CPU 0: <tt>kfree(p);</tt>
733 <p><font color="ffffff">
734 Therefore, there absolutely must be a full memory barrier between the
735 end of the RCU read-side critical section and the end of the
739 <p><font color="ffffff">
740 The sequence of events demonstrating the necessity of the second rule
745 <li> <font color="ffffff">CPU 0: <tt>list_del_rcu(p);</tt>
747 <li> <font color="ffffff">CPU 0: <tt>synchronize_rcu()</tt> starts.
749 <li> <font color="ffffff">CPU 1: <tt>rcu_read_lock()</tt>
751 <li> <font color="ffffff">CPU 1: <tt>q = rcu_dereference(gp);
752 /* Might return p if no memory barrier. */</tt>
754 <li> <font color="ffffff">CPU 0: <tt>synchronize_rcu()</tt> returns.
756 <li> <font color="ffffff">CPU 0: <tt>kfree(p);</tt>
758 <li> <font color="ffffff">
759 CPU 1: <tt>do_something_with(q->a); /* Boom!!! */</tt>
761 <li> <font color="ffffff">CPU 1: <tt>rcu_read_unlock()</tt>
765 <p><font color="ffffff">
766 And similarly, without a memory barrier between the beginning of the
767 grace period and the beginning of the RCU read-side critical section,
768 CPU 1 might end up accessing the freelist.
771 <p><font color="ffffff">
772 The “as if” rule of course applies, so that any
773 implementation that acts as if the appropriate memory barriers
774 were in place is a correct implementation.
775 That said, it is much easier to fool yourself into believing
776 that you have adhered to the as-if rule than it is to actually
779 <tr><td> </td></tr>
783 <tr><th> </th></tr>
784 <tr><th align="left">Quick Quiz:</th></tr>
786 You claim that <tt>rcu_read_lock()</tt> and <tt>rcu_read_unlock()</tt>
787 generate absolutely no code in some kernel builds.
788 This means that the compiler might arbitrarily rearrange consecutive
789 RCU read-side critical sections.
790 Given such rearrangement, if a given RCU read-side critical section
791 is done, how can you be sure that all prior RCU read-side critical
793 Won't the compiler rearrangements make that impossible to determine?
795 <tr><th align="left">Answer:</th></tr>
796 <tr><td bgcolor="#ffffff"><font color="ffffff">
797 In cases where <tt>rcu_read_lock()</tt> and <tt>rcu_read_unlock()</tt>
798 generate absolutely no code, RCU infers quiescent states only at
799 special locations, for example, within the scheduler.
800 Because calls to <tt>schedule()</tt> had better prevent calling-code
801 accesses to shared variables from being rearranged across the call to
802 <tt>schedule()</tt>, if RCU detects the end of a given RCU read-side
803 critical section, it will necessarily detect the end of all prior
804 RCU read-side critical sections, no matter how aggressively the
805 compiler scrambles the code.
808 <p><font color="ffffff">
809 Again, this all assumes that the compiler cannot scramble code across
810 calls to the scheduler, out of interrupt handlers, into the idle loop,
811 into user-mode code, and so on.
812 But if your kernel build allows that sort of scrambling, you have broken
813 far more than just RCU!
815 <tr><td> </td></tr>
819 Note that these memory-barrier requirements do not replace the fundamental
820 RCU requirement that a grace period wait for all pre-existing readers.
821 On the contrary, the memory barriers called out in this section must operate in
822 such a way as to <i>enforce</i> this fundamental requirement.
823 Of course, different implementations enforce this requirement in different
824 ways, but enforce it they must.
826 <h3><a name="RCU Primitives Guaranteed to Execute Unconditionally">RCU Primitives Guaranteed to Execute Unconditionally</a></h3>
829 The common-case RCU primitives are unconditional.
830 They are invoked, they do their job, and they return, with no possibility
831 of error, and no need to retry.
832 This is a key RCU design philosophy.
835 However, this philosophy is pragmatic rather than pigheaded.
836 If someone comes up with a good justification for a particular conditional
837 RCU primitive, it might well be implemented and added.
838 After all, this guarantee was reverse-engineered, not premeditated.
839 The unconditional nature of the RCU primitives was initially an
840 accident of implementation, and later experience with synchronization
841 primitives with conditional primitives caused me to elevate this
842 accident to a guarantee.
843 Therefore, the justification for adding a conditional primitive to
844 RCU would need to be based on detailed and compelling use cases.
846 <h3><a name="Guaranteed Read-to-Write Upgrade">Guaranteed Read-to-Write Upgrade</a></h3>
849 As far as RCU is concerned, it is always possible to carry out an
850 update within an RCU read-side critical section.
851 For example, that RCU read-side critical section might search for
852 a given data element, and then might acquire the update-side
853 spinlock in order to update that element, all while remaining
854 in that RCU read-side critical section.
855 Of course, it is necessary to exit the RCU read-side critical section
856 before invoking <tt>synchronize_rcu()</tt>, however, this
857 inconvenience can be avoided through use of the
858 <tt>call_rcu()</tt> and <tt>kfree_rcu()</tt> API members
859 described later in this document.
862 <tr><th> </th></tr>
863 <tr><th align="left">Quick Quiz:</th></tr>
865 But how does the upgrade-to-write operation exclude other readers?
867 <tr><th align="left">Answer:</th></tr>
868 <tr><td bgcolor="#ffffff"><font color="ffffff">
869 It doesn't, just like normal RCU updates, which also do not exclude
872 <tr><td> </td></tr>
876 This guarantee allows lookup code to be shared between read-side
877 and update-side code, and was premeditated, appearing in the earliest
878 DYNIX/ptx RCU documentation.
880 <h2><a name="Fundamental Non-Requirements">Fundamental Non-Requirements</a></h2>
883 RCU provides extremely lightweight readers, and its read-side guarantees,
884 though quite useful, are correspondingly lightweight.
885 It is therefore all too easy to assume that RCU is guaranteeing more
887 Of course, the list of things that RCU does not guarantee is infinitely
888 long, however, the following sections list a few non-guarantees that
889 have caused confusion.
890 Except where otherwise noted, these non-guarantees were premeditated.
893 <li> <a href="#Readers Impose Minimal Ordering">
894 Readers Impose Minimal Ordering</a>
895 <li> <a href="#Readers Do Not Exclude Updaters">
896 Readers Do Not Exclude Updaters</a>
897 <li> <a href="#Updaters Only Wait For Old Readers">
898 Updaters Only Wait For Old Readers</a>
899 <li> <a href="#Grace Periods Don't Partition Read-Side Critical Sections">
900 Grace Periods Don't Partition Read-Side Critical Sections</a>
901 <li> <a href="#Read-Side Critical Sections Don't Partition Grace Periods">
902 Read-Side Critical Sections Don't Partition Grace Periods</a>
903 <li> <a href="#Disabling Preemption Does Not Block Grace Periods">
904 Disabling Preemption Does Not Block Grace Periods</a>
907 <h3><a name="Readers Impose Minimal Ordering">Readers Impose Minimal Ordering</a></h3>
910 Reader-side markers such as <tt>rcu_read_lock()</tt> and
911 <tt>rcu_read_unlock()</tt> provide absolutely no ordering guarantees
912 except through their interaction with the grace-period APIs such as
913 <tt>synchronize_rcu()</tt>.
914 To see this, consider the following pair of threads:
928 11 void thread1(void)
931 14 r1 = READ_ONCE(y);
932 15 rcu_read_unlock();
934 17 r2 = READ_ONCE(x);
935 18 rcu_read_unlock();
941 After <tt>thread0()</tt> and <tt>thread1()</tt> execute
942 concurrently, it is quite possible to have
946 (r1 == 1 && r2 == 0)
950 (that is, <tt>y</tt> appears to have been assigned before <tt>x</tt>),
951 which would not be possible if <tt>rcu_read_lock()</tt> and
952 <tt>rcu_read_unlock()</tt> had much in the way of ordering
954 But they do not, so the CPU is within its rights
955 to do significant reordering.
956 This is by design: Any significant ordering constraints would slow down
957 these fast-path APIs.
960 <tr><th> </th></tr>
961 <tr><th align="left">Quick Quiz:</th></tr>
963 Can't the compiler also reorder this code?
965 <tr><th align="left">Answer:</th></tr>
966 <tr><td bgcolor="#ffffff"><font color="ffffff">
967 No, the volatile casts in <tt>READ_ONCE()</tt> and
968 <tt>WRITE_ONCE()</tt> prevent the compiler from reordering in
969 this particular case.
971 <tr><td> </td></tr>
974 <h3><a name="Readers Do Not Exclude Updaters">Readers Do Not Exclude Updaters</a></h3>
977 Neither <tt>rcu_read_lock()</tt> nor <tt>rcu_read_unlock()</tt>
979 All they do is to prevent grace periods from ending.
980 The following example illustrates this:
989 6 do_something_with_nonzero_x();
991 8 WARN_ON(!r2); /* BUG!!! */
993 10 rcu_read_unlock();
996 13 void thread1(void)
998 15 spin_lock(&my_lock);
1000 17 WRITE_ONCE(y, 1);
1001 18 spin_unlock(&my_lock);
1007 If the <tt>thread0()</tt> function's <tt>rcu_read_lock()</tt>
1008 excluded the <tt>thread1()</tt> function's update,
1009 the <tt>WARN_ON()</tt> could never fire.
1010 But the fact is that <tt>rcu_read_lock()</tt> does not exclude
1011 much of anything aside from subsequent grace periods, of which
1012 <tt>thread1()</tt> has none, so the
1013 <tt>WARN_ON()</tt> can and does fire.
1015 <h3><a name="Updaters Only Wait For Old Readers">Updaters Only Wait For Old Readers</a></h3>
1018 It might be tempting to assume that after <tt>synchronize_rcu()</tt>
1019 completes, there are no readers executing.
1020 This temptation must be avoided because
1021 new readers can start immediately after <tt>synchronize_rcu()</tt>
1022 starts, and <tt>synchronize_rcu()</tt> is under no
1023 obligation to wait for these new readers.
1026 <tr><th> </th></tr>
1027 <tr><th align="left">Quick Quiz:</th></tr>
1029 Suppose that synchronize_rcu() did wait until <i>all</i>
1030 readers had completed instead of waiting only on
1031 pre-existing readers.
1032 For how long would the updater be able to rely on there
1035 <tr><th align="left">Answer:</th></tr>
1036 <tr><td bgcolor="#ffffff"><font color="ffffff">
1038 Even if <tt>synchronize_rcu()</tt> were to wait until
1039 all readers had completed, a new reader might start immediately after
1040 <tt>synchronize_rcu()</tt> completed.
1041 Therefore, the code following
1042 <tt>synchronize_rcu()</tt> can <i>never</i> rely on there being
1045 <tr><td> </td></tr>
1048 <h3><a name="Grace Periods Don't Partition Read-Side Critical Sections">
1049 Grace Periods Don't Partition Read-Side Critical Sections</a></h3>
1052 It is tempting to assume that if any part of one RCU read-side critical
1053 section precedes a given grace period, and if any part of another RCU
1054 read-side critical section follows that same grace period, then all of
1055 the first RCU read-side critical section must precede all of the second.
1056 However, this just isn't the case: A single grace period does not
1057 partition the set of RCU read-side critical sections.
1058 An example of this situation can be illustrated as follows, where
1059 <tt>x</tt>, <tt>y</tt>, and <tt>z</tt> are initially all zero:
1063 1 void thread0(void)
1068 6 rcu_read_unlock();
1071 9 void thread1(void)
1073 11 r1 = READ_ONCE(a);
1074 12 synchronize_rcu();
1075 13 WRITE_ONCE(c, 1);
1078 16 void thread2(void)
1081 19 r2 = READ_ONCE(b);
1082 20 r3 = READ_ONCE(c);
1083 21 rcu_read_unlock();
1089 It turns out that the outcome:
1093 (r1 == 1 && r2 == 0 && r3 == 1)
1097 is entirely possible.
1098 The following figure show how this can happen, with each circled
1099 <tt>QS</tt> indicating the point at which RCU recorded a
1100 <i>quiescent state</i> for each thread, that is, a state in which
1101 RCU knows that the thread cannot be in the midst of an RCU read-side
1102 critical section that started before the current grace period:
1104 <p><img src="GPpartitionReaders1.svg" alt="GPpartitionReaders1.svg" width="60%"></p>
1107 If it is necessary to partition RCU read-side critical sections in this
1108 manner, it is necessary to use two grace periods, where the first
1109 grace period is known to end before the second grace period starts:
1113 1 void thread0(void)
1118 6 rcu_read_unlock();
1121 9 void thread1(void)
1123 11 r1 = READ_ONCE(a);
1124 12 synchronize_rcu();
1125 13 WRITE_ONCE(c, 1);
1128 16 void thread2(void)
1130 18 r2 = READ_ONCE(c);
1131 19 synchronize_rcu();
1132 20 WRITE_ONCE(d, 1);
1135 23 void thread3(void)
1138 26 r3 = READ_ONCE(b);
1139 27 r4 = READ_ONCE(d);
1140 28 rcu_read_unlock();
1146 Here, if <tt>(r1 == 1)</tt>, then
1147 <tt>thread0()</tt>'s write to <tt>b</tt> must happen
1148 before the end of <tt>thread1()</tt>'s grace period.
1149 If in addition <tt>(r4 == 1)</tt>, then
1150 <tt>thread3()</tt>'s read from <tt>b</tt> must happen
1151 after the beginning of <tt>thread2()</tt>'s grace period.
1152 If it is also the case that <tt>(r2 == 1)</tt>, then the
1153 end of <tt>thread1()</tt>'s grace period must precede the
1154 beginning of <tt>thread2()</tt>'s grace period.
1155 This mean that the two RCU read-side critical sections cannot overlap,
1156 guaranteeing that <tt>(r3 == 1)</tt>.
1157 As a result, the outcome:
1161 (r1 == 1 && r2 == 1 && r3 == 0 && r4 == 1)
1168 This non-requirement was also non-premeditated, but became apparent
1169 when studying RCU's interaction with memory ordering.
1171 <h3><a name="Read-Side Critical Sections Don't Partition Grace Periods">
1172 Read-Side Critical Sections Don't Partition Grace Periods</a></h3>
1175 It is also tempting to assume that if an RCU read-side critical section
1176 happens between a pair of grace periods, then those grace periods cannot
1178 However, this temptation leads nowhere good, as can be illustrated by
1179 the following, with all variables initially zero:
1183 1 void thread0(void)
1188 6 rcu_read_unlock();
1191 9 void thread1(void)
1193 11 r1 = READ_ONCE(a);
1194 12 synchronize_rcu();
1195 13 WRITE_ONCE(c, 1);
1198 16 void thread2(void)
1201 19 WRITE_ONCE(d, 1);
1202 20 r2 = READ_ONCE(c);
1203 21 rcu_read_unlock();
1206 24 void thread3(void)
1208 26 r3 = READ_ONCE(d);
1209 27 synchronize_rcu();
1210 28 WRITE_ONCE(e, 1);
1213 31 void thread4(void)
1216 34 r4 = READ_ONCE(b);
1217 35 r5 = READ_ONCE(e);
1218 36 rcu_read_unlock();
1224 In this case, the outcome:
1228 (r1 == 1 && r2 == 1 && r3 == 1 && r4 == 0 && r5 == 1)
1232 is entirely possible, as illustrated below:
1234 <p><img src="ReadersPartitionGP1.svg" alt="ReadersPartitionGP1.svg" width="100%"></p>
1237 Again, an RCU read-side critical section can overlap almost all of a
1238 given grace period, just so long as it does not overlap the entire
1240 As a result, an RCU read-side critical section cannot partition a pair
1241 of RCU grace periods.
1244 <tr><th> </th></tr>
1245 <tr><th align="left">Quick Quiz:</th></tr>
1247 How long a sequence of grace periods, each separated by an RCU
1248 read-side critical section, would be required to partition the RCU
1249 read-side critical sections at the beginning and end of the chain?
1251 <tr><th align="left">Answer:</th></tr>
1252 <tr><td bgcolor="#ffffff"><font color="ffffff">
1253 In theory, an infinite number.
1254 In practice, an unknown number that is sensitive to both implementation
1255 details and timing considerations.
1256 Therefore, even in practice, RCU users must abide by the
1257 theoretical rather than the practical answer.
1259 <tr><td> </td></tr>
1262 <h3><a name="Disabling Preemption Does Not Block Grace Periods">
1263 Disabling Preemption Does Not Block Grace Periods</a></h3>
1266 There was a time when disabling preemption on any given CPU would block
1267 subsequent grace periods.
1268 However, this was an accident of implementation and is not a requirement.
1269 And in the current Linux-kernel implementation, disabling preemption
1270 on a given CPU in fact does not block grace periods, as Oleg Nesterov
1271 <a href="https://lkml.kernel.org/g/20150614193825.GA19582@redhat.com">demonstrated</a>.
1274 If you need a preempt-disable region to block grace periods, you need to add
1275 <tt>rcu_read_lock()</tt> and <tt>rcu_read_unlock()</tt>, for example
1280 1 preempt_disable();
1283 4 rcu_read_unlock();
1286 7 /* Spinlocks implicitly disable preemption. */
1287 8 spin_lock(&mylock);
1290 11 rcu_read_unlock();
1291 12 spin_unlock(&mylock);
1296 In theory, you could enter the RCU read-side critical section first,
1297 but it is more efficient to keep the entire RCU read-side critical
1298 section contained in the preempt-disable region as shown above.
1299 Of course, RCU read-side critical sections that extend outside of
1300 preempt-disable regions will work correctly, but such critical sections
1301 can be preempted, which forces <tt>rcu_read_unlock()</tt> to do
1303 And no, this is <i>not</i> an invitation to enclose all of your RCU
1304 read-side critical sections within preempt-disable regions, because
1305 doing so would degrade real-time response.
1308 This non-requirement appeared with preemptible RCU.
1309 If you need a grace period that waits on non-preemptible code regions, use
1310 <a href="#Sched Flavor">RCU-sched</a>.
1312 <h2><a name="Parallelism Facts of Life">Parallelism Facts of Life</a></h2>
1315 These parallelism facts of life are by no means specific to RCU, but
1316 the RCU implementation must abide by them.
1317 They therefore bear repeating:
1320 <li> Any CPU or task may be delayed at any time,
1321 and any attempts to avoid these delays by disabling
1322 preemption, interrupts, or whatever are completely futile.
1323 This is most obvious in preemptible user-level
1324 environments and in virtualized environments (where
1325 a given guest OS's VCPUs can be preempted at any time by
1326 the underlying hypervisor), but can also happen in bare-metal
1327 environments due to ECC errors, NMIs, and other hardware
1329 Although a delay of more than about 20 seconds can result
1330 in splats, the RCU implementation is obligated to use
1331 algorithms that can tolerate extremely long delays, but where
1332 “extremely long” is not long enough to allow
1333 wrap-around when incrementing a 64-bit counter.
1334 <li> Both the compiler and the CPU can reorder memory accesses.
1335 Where it matters, RCU must use compiler directives and
1336 memory-barrier instructions to preserve ordering.
1337 <li> Conflicting writes to memory locations in any given cache line
1338 will result in expensive cache misses.
1339 Greater numbers of concurrent writes and more-frequent
1340 concurrent writes will result in more dramatic slowdowns.
1341 RCU is therefore obligated to use algorithms that have
1342 sufficient locality to avoid significant performance and
1343 scalability problems.
1344 <li> As a rough rule of thumb, only one CPU's worth of processing
1345 may be carried out under the protection of any given exclusive
1347 RCU must therefore use scalable locking designs.
1348 <li> Counters are finite, especially on 32-bit systems.
1349 RCU's use of counters must therefore tolerate counter wrap,
1350 or be designed such that counter wrap would take way more
1351 time than a single system is likely to run.
1352 An uptime of ten years is quite possible, a runtime
1353 of a century much less so.
1354 As an example of the latter, RCU's dyntick-idle nesting counter
1355 allows 54 bits for interrupt nesting level (this counter
1356 is 64 bits even on a 32-bit system).
1357 Overflowing this counter requires 2<sup>54</sup>
1358 half-interrupts on a given CPU without that CPU ever going idle.
1359 If a half-interrupt happened every microsecond, it would take
1360 570 years of runtime to overflow this counter, which is currently
1361 believed to be an acceptably long time.
1362 <li> Linux systems can have thousands of CPUs running a single
1363 Linux kernel in a single shared-memory environment.
1364 RCU must therefore pay close attention to high-end scalability.
1368 This last parallelism fact of life means that RCU must pay special
1369 attention to the preceding facts of life.
1370 The idea that Linux might scale to systems with thousands of CPUs would
1371 have been met with some skepticism in the 1990s, but these requirements
1372 would have otherwise have been unsurprising, even in the early 1990s.
1374 <h2><a name="Quality-of-Implementation Requirements">Quality-of-Implementation Requirements</a></h2>
1377 These sections list quality-of-implementation requirements.
1378 Although an RCU implementation that ignores these requirements could
1379 still be used, it would likely be subject to limitations that would
1380 make it inappropriate for industrial-strength production use.
1381 Classes of quality-of-implementation requirements are as follows:
1384 <li> <a href="#Specialization">Specialization</a>
1385 <li> <a href="#Performance and Scalability">Performance and Scalability</a>
1386 <li> <a href="#Composability">Composability</a>
1387 <li> <a href="#Corner Cases">Corner Cases</a>
1391 These classes is covered in the following sections.
1393 <h3><a name="Specialization">Specialization</a></h3>
1396 RCU is and always has been intended primarily for read-mostly situations,
1397 which means that RCU's read-side primitives are optimized, often at the
1398 expense of its update-side primitives.
1399 Experience thus far is captured by the following list of situations:
1402 <li> Read-mostly data, where stale and inconsistent data is not
1403 a problem: RCU works great!
1404 <li> Read-mostly data, where data must be consistent:
1406 <li> Read-write data, where data must be consistent:
1407 RCU <i>might</i> work OK.
1409 <li> Write-mostly data, where data must be consistent:
1410 RCU is very unlikely to be the right tool for the job,
1411 with the following exceptions, where RCU can provide:
1413 <li> Existence guarantees for update-friendly mechanisms.
1414 <li> Wait-free read-side primitives for real-time use.
1419 This focus on read-mostly situations means that RCU must interoperate
1420 with other synchronization primitives.
1421 For example, the <tt>add_gp()</tt> and <tt>remove_gp_synchronous()</tt>
1422 examples discussed earlier use RCU to protect readers and locking to
1423 coordinate updaters.
1424 However, the need extends much farther, requiring that a variety of
1425 synchronization primitives be legal within RCU read-side critical sections,
1426 including spinlocks, sequence locks, atomic operations, reference
1427 counters, and memory barriers.
1430 <tr><th> </th></tr>
1431 <tr><th align="left">Quick Quiz:</th></tr>
1433 What about sleeping locks?
1435 <tr><th align="left">Answer:</th></tr>
1436 <tr><td bgcolor="#ffffff"><font color="ffffff">
1437 These are forbidden within Linux-kernel RCU read-side critical
1438 sections because it is not legal to place a quiescent state
1439 (in this case, voluntary context switch) within an RCU read-side
1441 However, sleeping locks may be used within userspace RCU read-side
1442 critical sections, and also within Linux-kernel sleepable RCU
1443 <a href="#Sleepable RCU"><font color="ffffff">(SRCU)</font></a>
1444 read-side critical sections.
1445 In addition, the -rt patchset turns spinlocks into a
1446 sleeping locks so that the corresponding critical sections
1447 can be preempted, which also means that these sleeplockified
1448 spinlocks (but not other sleeping locks!) may be acquire within
1449 -rt-Linux-kernel RCU read-side critical sections.
1452 <p><font color="ffffff">
1453 Note that it <i>is</i> legal for a normal RCU read-side
1454 critical section to conditionally acquire a sleeping locks
1455 (as in <tt>mutex_trylock()</tt>), but only as long as it does
1456 not loop indefinitely attempting to conditionally acquire that
1458 The key point is that things like <tt>mutex_trylock()</tt>
1459 either return with the mutex held, or return an error indication if
1460 the mutex was not immediately available.
1461 Either way, <tt>mutex_trylock()</tt> returns immediately without
1464 <tr><td> </td></tr>
1468 It often comes as a surprise that many algorithms do not require a
1469 consistent view of data, but many can function in that mode,
1470 with network routing being the poster child.
1471 Internet routing algorithms take significant time to propagate
1472 updates, so that by the time an update arrives at a given system,
1473 that system has been sending network traffic the wrong way for
1474 a considerable length of time.
1475 Having a few threads continue to send traffic the wrong way for a
1476 few more milliseconds is clearly not a problem: In the worst case,
1477 TCP retransmissions will eventually get the data where it needs to go.
1478 In general, when tracking the state of the universe outside of the
1479 computer, some level of inconsistency must be tolerated due to
1480 speed-of-light delays if nothing else.
1483 Furthermore, uncertainty about external state is inherent in many cases.
1484 For example, a pair of veterinarians might use heartbeat to determine
1485 whether or not a given cat was alive.
1486 But how long should they wait after the last heartbeat to decide that
1487 the cat is in fact dead?
1488 Waiting less than 400 milliseconds makes no sense because this would
1489 mean that a relaxed cat would be considered to cycle between death
1490 and life more than 100 times per minute.
1491 Moreover, just as with human beings, a cat's heart might stop for
1492 some period of time, so the exact wait period is a judgment call.
1493 One of our pair of veterinarians might wait 30 seconds before pronouncing
1494 the cat dead, while the other might insist on waiting a full minute.
1495 The two veterinarians would then disagree on the state of the cat during
1496 the final 30 seconds of the minute following the last heartbeat.
1499 Interestingly enough, this same situation applies to hardware.
1500 When push comes to shove, how do we tell whether or not some
1501 external server has failed?
1502 We send messages to it periodically, and declare it failed if we
1503 don't receive a response within a given period of time.
1504 Policy decisions can usually tolerate short
1505 periods of inconsistency.
1506 The policy was decided some time ago, and is only now being put into
1507 effect, so a few milliseconds of delay is normally inconsequential.
1510 However, there are algorithms that absolutely must see consistent data.
1511 For example, the translation between a user-level SystemV semaphore
1512 ID to the corresponding in-kernel data structure is protected by RCU,
1513 but it is absolutely forbidden to update a semaphore that has just been
1515 In the Linux kernel, this need for consistency is accommodated by acquiring
1516 spinlocks located in the in-kernel data structure from within
1517 the RCU read-side critical section, and this is indicated by the
1518 green box in the figure above.
1519 Many other techniques may be used, and are in fact used within the
1523 In short, RCU is not required to maintain consistency, and other
1524 mechanisms may be used in concert with RCU when consistency is required.
1525 RCU's specialization allows it to do its job extremely well, and its
1526 ability to interoperate with other synchronization mechanisms allows
1527 the right mix of synchronization tools to be used for a given job.
1529 <h3><a name="Performance and Scalability">Performance and Scalability</a></h3>
1532 Energy efficiency is a critical component of performance today,
1533 and Linux-kernel RCU implementations must therefore avoid unnecessarily
1534 awakening idle CPUs.
1535 I cannot claim that this requirement was premeditated.
1536 In fact, I learned of it during a telephone conversation in which I
1537 was given “frank and open” feedback on the importance
1538 of energy efficiency in battery-powered systems and on specific
1539 energy-efficiency shortcomings of the Linux-kernel RCU implementation.
1540 In my experience, the battery-powered embedded community will consider
1541 any unnecessary wakeups to be extremely unfriendly acts.
1542 So much so that mere Linux-kernel-mailing-list posts are
1543 insufficient to vent their ire.
1546 Memory consumption is not particularly important for in most
1547 situations, and has become decreasingly
1548 so as memory sizes have expanded and memory
1549 costs have plummeted.
1550 However, as I learned from Matt Mackall's
1551 <a href="http://elinux.org/Linux_Tiny-FAQ">bloatwatch</a>
1552 efforts, memory footprint is critically important on single-CPU systems with
1553 non-preemptible (<tt>CONFIG_PREEMPT=n</tt>) kernels, and thus
1554 <a href="https://lkml.kernel.org/g/20090113221724.GA15307@linux.vnet.ibm.com">tiny RCU</a>
1556 Josh Triplett has since taken over the small-memory banner with his
1557 <a href="https://tiny.wiki.kernel.org/">Linux kernel tinification</a>
1558 project, which resulted in
1559 <a href="#Sleepable RCU">SRCU</a>
1560 becoming optional for those kernels not needing it.
1563 The remaining performance requirements are, for the most part,
1565 For example, in keeping with RCU's read-side specialization,
1566 <tt>rcu_dereference()</tt> should have negligible overhead (for
1567 example, suppression of a few minor compiler optimizations).
1568 Similarly, in non-preemptible environments, <tt>rcu_read_lock()</tt> and
1569 <tt>rcu_read_unlock()</tt> should have exactly zero overhead.
1572 In preemptible environments, in the case where the RCU read-side
1573 critical section was not preempted (as will be the case for the
1574 highest-priority real-time process), <tt>rcu_read_lock()</tt> and
1575 <tt>rcu_read_unlock()</tt> should have minimal overhead.
1576 In particular, they should not contain atomic read-modify-write
1577 operations, memory-barrier instructions, preemption disabling,
1578 interrupt disabling, or backwards branches.
1579 However, in the case where the RCU read-side critical section was preempted,
1580 <tt>rcu_read_unlock()</tt> may acquire spinlocks and disable interrupts.
1581 This is why it is better to nest an RCU read-side critical section
1582 within a preempt-disable region than vice versa, at least in cases
1583 where that critical section is short enough to avoid unduly degrading
1584 real-time latencies.
1587 The <tt>synchronize_rcu()</tt> grace-period-wait primitive is
1588 optimized for throughput.
1589 It may therefore incur several milliseconds of latency in addition to
1590 the duration of the longest RCU read-side critical section.
1591 On the other hand, multiple concurrent invocations of
1592 <tt>synchronize_rcu()</tt> are required to use batching optimizations
1593 so that they can be satisfied by a single underlying grace-period-wait
1595 For example, in the Linux kernel, it is not unusual for a single
1596 grace-period-wait operation to serve more than
1597 <a href="https://www.usenix.org/conference/2004-usenix-annual-technical-conference/making-rcu-safe-deep-sub-millisecond-response">1,000 separate invocations</a>
1598 of <tt>synchronize_rcu()</tt>, thus amortizing the per-invocation
1599 overhead down to nearly zero.
1600 However, the grace-period optimization is also required to avoid
1601 measurable degradation of real-time scheduling and interrupt latencies.
1604 In some cases, the multi-millisecond <tt>synchronize_rcu()</tt>
1605 latencies are unacceptable.
1606 In these cases, <tt>synchronize_rcu_expedited()</tt> may be used
1607 instead, reducing the grace-period latency down to a few tens of
1608 microseconds on small systems, at least in cases where the RCU read-side
1609 critical sections are short.
1610 There are currently no special latency requirements for
1611 <tt>synchronize_rcu_expedited()</tt> on large systems, but,
1612 consistent with the empirical nature of the RCU specification,
1613 that is subject to change.
1614 However, there most definitely are scalability requirements:
1615 A storm of <tt>synchronize_rcu_expedited()</tt> invocations on 4096
1616 CPUs should at least make reasonable forward progress.
1617 In return for its shorter latencies, <tt>synchronize_rcu_expedited()</tt>
1618 is permitted to impose modest degradation of real-time latency
1619 on non-idle online CPUs.
1620 Here, “modest” means roughly the same latency
1621 degradation as a scheduling-clock interrupt.
1624 There are a number of situations where even
1625 <tt>synchronize_rcu_expedited()</tt>'s reduced grace-period
1626 latency is unacceptable.
1627 In these situations, the asynchronous <tt>call_rcu()</tt> can be
1628 used in place of <tt>synchronize_rcu()</tt> as follows:
1635 4 struct rcu_head rh;
1638 7 static void remove_gp_cb(struct rcu_head *rhp)
1640 9 struct foo *p = container_of(rhp, struct foo, rh);
1645 14 bool remove_gp_asynchronous(void)
1649 18 spin_lock(&gp_lock);
1650 19 p = rcu_dereference(gp);
1652 21 spin_unlock(&gp_lock);
1655 24 rcu_assign_pointer(gp, NULL);
1656 25 call_rcu(&p->rh, remove_gp_cb);
1657 26 spin_unlock(&gp_lock);
1664 A definition of <tt>struct foo</tt> is finally needed, and appears
1666 The function <tt>remove_gp_cb()</tt> is passed to <tt>call_rcu()</tt>
1667 on line 25, and will be invoked after the end of a subsequent
1669 This gets the same effect as <tt>remove_gp_synchronous()</tt>,
1670 but without forcing the updater to wait for a grace period to elapse.
1671 The <tt>call_rcu()</tt> function may be used in a number of
1672 situations where neither <tt>synchronize_rcu()</tt> nor
1673 <tt>synchronize_rcu_expedited()</tt> would be legal,
1674 including within preempt-disable code, <tt>local_bh_disable()</tt> code,
1675 interrupt-disable code, and interrupt handlers.
1676 However, even <tt>call_rcu()</tt> is illegal within NMI handlers
1677 and from idle and offline CPUs.
1678 The callback function (<tt>remove_gp_cb()</tt> in this case) will be
1679 executed within softirq (software interrupt) environment within the
1681 either within a real softirq handler or under the protection
1682 of <tt>local_bh_disable()</tt>.
1683 In both the Linux kernel and in userspace, it is bad practice to
1684 write an RCU callback function that takes too long.
1685 Long-running operations should be relegated to separate threads or
1686 (in the Linux kernel) workqueues.
1689 <tr><th> </th></tr>
1690 <tr><th align="left">Quick Quiz:</th></tr>
1692 Why does line 19 use <tt>rcu_access_pointer()</tt>?
1693 After all, <tt>call_rcu()</tt> on line 25 stores into the
1694 structure, which would interact badly with concurrent insertions.
1695 Doesn't this mean that <tt>rcu_dereference()</tt> is required?
1697 <tr><th align="left">Answer:</th></tr>
1698 <tr><td bgcolor="#ffffff"><font color="ffffff">
1699 Presumably the <tt>->gp_lock</tt> acquired on line 18 excludes
1700 any changes, including any insertions that <tt>rcu_dereference()</tt>
1701 would protect against.
1702 Therefore, any insertions will be delayed until after
1703 <tt>->gp_lock</tt>
1704 is released on line 25, which in turn means that
1705 <tt>rcu_access_pointer()</tt> suffices.
1707 <tr><td> </td></tr>
1711 However, all that <tt>remove_gp_cb()</tt> is doing is
1712 invoking <tt>kfree()</tt> on the data element.
1713 This is a common idiom, and is supported by <tt>kfree_rcu()</tt>,
1714 which allows “fire and forget” operation as shown below:
1721 4 struct rcu_head rh;
1724 7 bool remove_gp_faf(void)
1728 11 spin_lock(&gp_lock);
1729 12 p = rcu_dereference(gp);
1731 14 spin_unlock(&gp_lock);
1734 17 rcu_assign_pointer(gp, NULL);
1735 18 kfree_rcu(p, rh);
1736 19 spin_unlock(&gp_lock);
1743 Note that <tt>remove_gp_faf()</tt> simply invokes
1744 <tt>kfree_rcu()</tt> and proceeds, without any need to pay any
1745 further attention to the subsequent grace period and <tt>kfree()</tt>.
1746 It is permissible to invoke <tt>kfree_rcu()</tt> from the same
1747 environments as for <tt>call_rcu()</tt>.
1748 Interestingly enough, DYNIX/ptx had the equivalents of
1749 <tt>call_rcu()</tt> and <tt>kfree_rcu()</tt>, but not
1750 <tt>synchronize_rcu()</tt>.
1751 This was due to the fact that RCU was not heavily used within DYNIX/ptx,
1752 so the very few places that needed something like
1753 <tt>synchronize_rcu()</tt> simply open-coded it.
1756 <tr><th> </th></tr>
1757 <tr><th align="left">Quick Quiz:</th></tr>
1759 Earlier it was claimed that <tt>call_rcu()</tt> and
1760 <tt>kfree_rcu()</tt> allowed updaters to avoid being blocked
1762 But how can that be correct, given that the invocation of the callback
1763 and the freeing of the memory (respectively) must still wait for
1764 a grace period to elapse?
1766 <tr><th align="left">Answer:</th></tr>
1767 <tr><td bgcolor="#ffffff"><font color="ffffff">
1768 We could define things this way, but keep in mind that this sort of
1769 definition would say that updates in garbage-collected languages
1770 cannot complete until the next time the garbage collector runs,
1771 which does not seem at all reasonable.
1772 The key point is that in most cases, an updater using either
1773 <tt>call_rcu()</tt> or <tt>kfree_rcu()</tt> can proceed to the
1774 next update as soon as it has invoked <tt>call_rcu()</tt> or
1775 <tt>kfree_rcu()</tt>, without having to wait for a subsequent
1778 <tr><td> </td></tr>
1782 But what if the updater must wait for the completion of code to be
1783 executed after the end of the grace period, but has other tasks
1784 that can be carried out in the meantime?
1785 The polling-style <tt>get_state_synchronize_rcu()</tt> and
1786 <tt>cond_synchronize_rcu()</tt> functions may be used for this
1787 purpose, as shown below:
1791 1 bool remove_gp_poll(void)
1796 6 spin_lock(&gp_lock);
1797 7 p = rcu_access_pointer(gp);
1799 9 spin_unlock(&gp_lock);
1802 12 rcu_assign_pointer(gp, NULL);
1803 13 spin_unlock(&gp_lock);
1804 14 s = get_state_synchronize_rcu();
1805 15 do_something_while_waiting();
1806 16 cond_synchronize_rcu(s);
1814 On line 14, <tt>get_state_synchronize_rcu()</tt> obtains a
1815 “cookie” from RCU,
1816 then line 15 carries out other tasks,
1817 and finally, line 16 returns immediately if a grace period has
1818 elapsed in the meantime, but otherwise waits as required.
1819 The need for <tt>get_state_synchronize_rcu</tt> and
1820 <tt>cond_synchronize_rcu()</tt> has appeared quite recently,
1821 so it is too early to tell whether they will stand the test of time.
1824 RCU thus provides a range of tools to allow updaters to strike the
1825 required tradeoff between latency, flexibility and CPU overhead.
1827 <h3><a name="Composability">Composability</a></h3>
1830 Composability has received much attention in recent years, perhaps in part
1831 due to the collision of multicore hardware with object-oriented techniques
1832 designed in single-threaded environments for single-threaded use.
1833 And in theory, RCU read-side critical sections may be composed, and in
1834 fact may be nested arbitrarily deeply.
1835 In practice, as with all real-world implementations of composable
1836 constructs, there are limitations.
1839 Implementations of RCU for which <tt>rcu_read_lock()</tt>
1840 and <tt>rcu_read_unlock()</tt> generate no code, such as
1841 Linux-kernel RCU when <tt>CONFIG_PREEMPT=n</tt>, can be
1842 nested arbitrarily deeply.
1843 After all, there is no overhead.
1844 Except that if all these instances of <tt>rcu_read_lock()</tt>
1845 and <tt>rcu_read_unlock()</tt> are visible to the compiler,
1846 compilation will eventually fail due to exhausting memory,
1847 mass storage, or user patience, whichever comes first.
1848 If the nesting is not visible to the compiler, as is the case with
1849 mutually recursive functions each in its own translation unit,
1850 stack overflow will result.
1851 If the nesting takes the form of loops, perhaps in the guise of tail
1852 recursion, either the control variable
1853 will overflow or (in the Linux kernel) you will get an RCU CPU stall warning.
1854 Nevertheless, this class of RCU implementations is one
1855 of the most composable constructs in existence.
1858 RCU implementations that explicitly track nesting depth
1859 are limited by the nesting-depth counter.
1860 For example, the Linux kernel's preemptible RCU limits nesting to
1862 This should suffice for almost all practical purposes.
1863 That said, a consecutive pair of RCU read-side critical sections
1864 between which there is an operation that waits for a grace period
1865 cannot be enclosed in another RCU read-side critical section.
1866 This is because it is not legal to wait for a grace period within
1867 an RCU read-side critical section: To do so would result either
1869 in RCU implicitly splitting the enclosing RCU read-side critical
1870 section, neither of which is conducive to a long-lived and prosperous
1874 It is worth noting that RCU is not alone in limiting composability.
1875 For example, many transactional-memory implementations prohibit
1876 composing a pair of transactions separated by an irrevocable
1877 operation (for example, a network receive operation).
1878 For another example, lock-based critical sections can be composed
1879 surprisingly freely, but only if deadlock is avoided.
1882 In short, although RCU read-side critical sections are highly composable,
1883 care is required in some situations, just as is the case for any other
1884 composable synchronization mechanism.
1886 <h3><a name="Corner Cases">Corner Cases</a></h3>
1889 A given RCU workload might have an endless and intense stream of
1890 RCU read-side critical sections, perhaps even so intense that there
1891 was never a point in time during which there was not at least one
1892 RCU read-side critical section in flight.
1893 RCU cannot allow this situation to block grace periods: As long as
1894 all the RCU read-side critical sections are finite, grace periods
1895 must also be finite.
1898 That said, preemptible RCU implementations could potentially result
1899 in RCU read-side critical sections being preempted for long durations,
1900 which has the effect of creating a long-duration RCU read-side
1902 This situation can arise only in heavily loaded systems, but systems using
1903 real-time priorities are of course more vulnerable.
1904 Therefore, RCU priority boosting is provided to help deal with this
1906 That said, the exact requirements on RCU priority boosting will likely
1907 evolve as more experience accumulates.
1910 Other workloads might have very high update rates.
1911 Although one can argue that such workloads should instead use
1912 something other than RCU, the fact remains that RCU must
1913 handle such workloads gracefully.
1914 This requirement is another factor driving batching of grace periods,
1915 but it is also the driving force behind the checks for large numbers
1916 of queued RCU callbacks in the <tt>call_rcu()</tt> code path.
1917 Finally, high update rates should not delay RCU read-side critical
1918 sections, although some small read-side delays can occur when using
1919 <tt>synchronize_rcu_expedited()</tt>, courtesy of this function's use
1920 of <tt>smp_call_function_single()</tt>.
1923 Although all three of these corner cases were understood in the early
1924 1990s, a simple user-level test consisting of <tt>close(open(path))</tt>
1926 in the early 2000s suddenly provided a much deeper appreciation of the
1927 high-update-rate corner case.
1928 This test also motivated addition of some RCU code to react to high update
1929 rates, for example, if a given CPU finds itself with more than 10,000
1930 RCU callbacks queued, it will cause RCU to take evasive action by
1931 more aggressively starting grace periods and more aggressively forcing
1932 completion of grace-period processing.
1933 This evasive action causes the grace period to complete more quickly,
1934 but at the cost of restricting RCU's batching optimizations, thus
1935 increasing the CPU overhead incurred by that grace period.
1937 <h2><a name="Software-Engineering Requirements">
1938 Software-Engineering Requirements</a></h2>
1941 Between Murphy's Law and “To err is human”, it is necessary to
1942 guard against mishaps and misuse:
1945 <li> It is all too easy to forget to use <tt>rcu_read_lock()</tt>
1946 everywhere that it is needed, so kernels built with
1947 <tt>CONFIG_PROVE_RCU=y</tt> will splat if
1948 <tt>rcu_dereference()</tt> is used outside of an
1949 RCU read-side critical section.
1950 Update-side code can use <tt>rcu_dereference_protected()</tt>,
1952 <a href="https://lwn.net/Articles/371986/">lockdep expression</a>
1953 to indicate what is providing the protection.
1954 If the indicated protection is not provided, a lockdep splat
1958 Code shared between readers and updaters can use
1959 <tt>rcu_dereference_check()</tt>, which also takes a
1960 lockdep expression, and emits a lockdep splat if neither
1961 <tt>rcu_read_lock()</tt> nor the indicated protection
1963 In addition, <tt>rcu_dereference_raw()</tt> is used in those
1964 (hopefully rare) cases where the required protection cannot
1965 be easily described.
1966 Finally, <tt>rcu_read_lock_held()</tt> is provided to
1967 allow a function to verify that it has been invoked within
1968 an RCU read-side critical section.
1969 I was made aware of this set of requirements shortly after Thomas
1970 Gleixner audited a number of RCU uses.
1971 <li> A given function might wish to check for RCU-related preconditions
1972 upon entry, before using any other RCU API.
1973 The <tt>rcu_lockdep_assert()</tt> does this job,
1974 asserting the expression in kernels having lockdep enabled
1975 and doing nothing otherwise.
1976 <li> It is also easy to forget to use <tt>rcu_assign_pointer()</tt>
1977 and <tt>rcu_dereference()</tt>, perhaps (incorrectly)
1978 substituting a simple assignment.
1979 To catch this sort of error, a given RCU-protected pointer may be
1980 tagged with <tt>__rcu</tt>, after which sparse
1981 will complain about simple-assignment accesses to that pointer.
1982 Arnd Bergmann made me aware of this requirement, and also
1984 <a href="https://lwn.net/Articles/376011/">patch series</a>.
1985 <li> Kernels built with <tt>CONFIG_DEBUG_OBJECTS_RCU_HEAD=y</tt>
1986 will splat if a data element is passed to <tt>call_rcu()</tt>
1987 twice in a row, without a grace period in between.
1988 (This error is similar to a double free.)
1989 The corresponding <tt>rcu_head</tt> structures that are
1990 dynamically allocated are automatically tracked, but
1991 <tt>rcu_head</tt> structures allocated on the stack
1992 must be initialized with <tt>init_rcu_head_on_stack()</tt>
1993 and cleaned up with <tt>destroy_rcu_head_on_stack()</tt>.
1994 Similarly, statically allocated non-stack <tt>rcu_head</tt>
1995 structures must be initialized with <tt>init_rcu_head()</tt>
1996 and cleaned up with <tt>destroy_rcu_head()</tt>.
1997 Mathieu Desnoyers made me aware of this requirement, and also
1999 <a href="https://lkml.kernel.org/g/20100319013024.GA28456@Krystal">patch</a>.
2000 <li> An infinite loop in an RCU read-side critical section will
2001 eventually trigger an RCU CPU stall warning splat, with
2002 the duration of “eventually” being controlled by the
2003 <tt>RCU_CPU_STALL_TIMEOUT</tt> <tt>Kconfig</tt> option, or,
2004 alternatively, by the
2005 <tt>rcupdate.rcu_cpu_stall_timeout</tt> boot/sysfs
2007 However, RCU is not obligated to produce this splat
2008 unless there is a grace period waiting on that particular
2009 RCU read-side critical section.
2011 Some extreme workloads might intentionally delay
2012 RCU grace periods, and systems running those workloads can
2013 be booted with <tt>rcupdate.rcu_cpu_stall_suppress</tt>
2014 to suppress the splats.
2015 This kernel parameter may also be set via <tt>sysfs</tt>.
2016 Furthermore, RCU CPU stall warnings are counter-productive
2017 during sysrq dumps and during panics.
2018 RCU therefore supplies the <tt>rcu_sysrq_start()</tt> and
2019 <tt>rcu_sysrq_end()</tt> API members to be called before
2020 and after long sysrq dumps.
2021 RCU also supplies the <tt>rcu_panic()</tt> notifier that is
2022 automatically invoked at the beginning of a panic to suppress
2023 further RCU CPU stall warnings.
2026 This requirement made itself known in the early 1990s, pretty
2027 much the first time that it was necessary to debug a CPU stall.
2028 That said, the initial implementation in DYNIX/ptx was quite
2029 generic in comparison with that of Linux.
2030 <li> Although it would be very good to detect pointers leaking out
2031 of RCU read-side critical sections, there is currently no
2032 good way of doing this.
2033 One complication is the need to distinguish between pointers
2034 leaking and pointers that have been handed off from RCU to
2035 some other synchronization mechanism, for example, reference
2037 <li> In kernels built with <tt>CONFIG_RCU_TRACE=y</tt>, RCU-related
2038 information is provided via event tracing.
2039 <li> Open-coded use of <tt>rcu_assign_pointer()</tt> and
2040 <tt>rcu_dereference()</tt> to create typical linked
2041 data structures can be surprisingly error-prone.
2042 Therefore, RCU-protected
2043 <a href="https://lwn.net/Articles/609973/#RCU List APIs">linked lists</a>
2044 and, more recently, RCU-protected
2045 <a href="https://lwn.net/Articles/612100/">hash tables</a>
2047 Many other special-purpose RCU-protected data structures are
2048 available in the Linux kernel and the userspace RCU library.
2049 <li> Some linked structures are created at compile time, but still
2050 require <tt>__rcu</tt> checking.
2051 The <tt>RCU_POINTER_INITIALIZER()</tt> macro serves this
2053 <li> It is not necessary to use <tt>rcu_assign_pointer()</tt>
2054 when creating linked structures that are to be published via
2055 a single external pointer.
2056 The <tt>RCU_INIT_POINTER()</tt> macro is provided for
2057 this task and also for assigning <tt>NULL</tt> pointers
2062 This not a hard-and-fast list: RCU's diagnostic capabilities will
2063 continue to be guided by the number and type of usage bugs found
2064 in real-world RCU usage.
2066 <h2><a name="Linux Kernel Complications">Linux Kernel Complications</a></h2>
2069 The Linux kernel provides an interesting environment for all kinds of
2070 software, including RCU.
2071 Some of the relevant points of interest are as follows:
2074 <li> <a href="#Configuration">Configuration</a>.
2075 <li> <a href="#Firmware Interface">Firmware Interface</a>.
2076 <li> <a href="#Early Boot">Early Boot</a>.
2077 <li> <a href="#Interrupts and NMIs">
2078 Interrupts and non-maskable interrupts (NMIs)</a>.
2079 <li> <a href="#Loadable Modules">Loadable Modules</a>.
2080 <li> <a href="#Hotplug CPU">Hotplug CPU</a>.
2081 <li> <a href="#Scheduler and RCU">Scheduler and RCU</a>.
2082 <li> <a href="#Tracing and RCU">Tracing and RCU</a>.
2083 <li> <a href="#Energy Efficiency">Energy Efficiency</a>.
2084 <li> <a href="#Scheduling-Clock Interrupts and RCU">
2085 Scheduling-Clock Interrupts and RCU</a>.
2086 <li> <a href="#Memory Efficiency">Memory Efficiency</a>.
2087 <li> <a href="#Performance, Scalability, Response Time, and Reliability">
2088 Performance, Scalability, Response Time, and Reliability</a>.
2092 This list is probably incomplete, but it does give a feel for the
2093 most notable Linux-kernel complications.
2094 Each of the following sections covers one of the above topics.
2096 <h3><a name="Configuration">Configuration</a></h3>
2099 RCU's goal is automatic configuration, so that almost nobody
2100 needs to worry about RCU's <tt>Kconfig</tt> options.
2101 And for almost all users, RCU does in fact work well
2102 “out of the box.”
2105 However, there are specialized use cases that are handled by
2106 kernel boot parameters and <tt>Kconfig</tt> options.
2107 Unfortunately, the <tt>Kconfig</tt> system will explicitly ask users
2108 about new <tt>Kconfig</tt> options, which requires almost all of them
2109 be hidden behind a <tt>CONFIG_RCU_EXPERT</tt> <tt>Kconfig</tt> option.
2112 This all should be quite obvious, but the fact remains that
2113 Linus Torvalds recently had to
2114 <a href="https://lkml.kernel.org/g/CA+55aFy4wcCwaL4okTs8wXhGZ5h-ibecy_Meg9C4MNQrUnwMcg@mail.gmail.com">remind</a>
2115 me of this requirement.
2117 <h3><a name="Firmware Interface">Firmware Interface</a></h3>
2120 In many cases, kernel obtains information about the system from the
2121 firmware, and sometimes things are lost in translation.
2122 Or the translation is accurate, but the original message is bogus.
2125 For example, some systems' firmware overreports the number of CPUs,
2126 sometimes by a large factor.
2127 If RCU naively believed the firmware, as it used to do,
2128 it would create too many per-CPU kthreads.
2129 Although the resulting system will still run correctly, the extra
2130 kthreads needlessly consume memory and can cause confusion
2131 when they show up in <tt>ps</tt> listings.
2134 RCU must therefore wait for a given CPU to actually come online before
2135 it can allow itself to believe that the CPU actually exists.
2136 The resulting “ghost CPUs” (which are never going to
2137 come online) cause a number of
2138 <a href="https://paulmck.livejournal.com/37494.html">interesting complications</a>.
2140 <h3><a name="Early Boot">Early Boot</a></h3>
2143 The Linux kernel's boot sequence is an interesting process,
2144 and RCU is used early, even before <tt>rcu_init()</tt>
2146 In fact, a number of RCU's primitives can be used as soon as the
2147 initial task's <tt>task_struct</tt> is available and the
2148 boot CPU's per-CPU variables are set up.
2149 The read-side primitives (<tt>rcu_read_lock()</tt>,
2150 <tt>rcu_read_unlock()</tt>, <tt>rcu_dereference()</tt>,
2151 and <tt>rcu_access_pointer()</tt>) will operate normally very early on,
2152 as will <tt>rcu_assign_pointer()</tt>.
2155 Although <tt>call_rcu()</tt> may be invoked at any
2156 time during boot, callbacks are not guaranteed to be invoked until after
2157 all of RCU's kthreads have been spawned, which occurs at
2158 <tt>early_initcall()</tt> time.
2159 This delay in callback invocation is due to the fact that RCU does not
2160 invoke callbacks until it is fully initialized, and this full initialization
2161 cannot occur until after the scheduler has initialized itself to the
2162 point where RCU can spawn and run its kthreads.
2163 In theory, it would be possible to invoke callbacks earlier,
2164 however, this is not a panacea because there would be severe restrictions
2165 on what operations those callbacks could invoke.
2168 Perhaps surprisingly, <tt>synchronize_rcu()</tt>,
2169 <a href="#Bottom-Half Flavor"><tt>synchronize_rcu_bh()</tt></a>
2170 (<a href="#Bottom-Half Flavor">discussed below</a>),
2171 <a href="#Sched Flavor"><tt>synchronize_sched()</tt></a>,
2172 <tt>synchronize_rcu_expedited()</tt>,
2173 <tt>synchronize_rcu_bh_expedited()</tt>, and
2174 <tt>synchronize_sched_expedited()</tt>
2175 will all operate normally
2176 during very early boot, the reason being that there is only one CPU
2177 and preemption is disabled.
2178 This means that the call <tt>synchronize_rcu()</tt> (or friends)
2179 itself is a quiescent
2180 state and thus a grace period, so the early-boot implementation can
2184 However, once the scheduler has spawned its first kthread, this early
2185 boot trick fails for <tt>synchronize_rcu()</tt> (as well as for
2186 <tt>synchronize_rcu_expedited()</tt>) in <tt>CONFIG_PREEMPT=y</tt>
2188 The reason is that an RCU read-side critical section might be preempted,
2189 which means that a subsequent <tt>synchronize_rcu()</tt> really does have
2190 to wait for something, as opposed to simply returning immediately.
2191 Unfortunately, <tt>synchronize_rcu()</tt> can't do this until all of
2192 its kthreads are spawned, which doesn't happen until some time during
2193 <tt>early_initcalls()</tt> time.
2194 But this is no excuse: RCU is nevertheless required to correctly handle
2195 synchronous grace periods during this time period.
2196 Once all of its kthreads are up and running, RCU starts running
2200 <tr><th> </th></tr>
2201 <tr><th align="left">Quick Quiz:</th></tr>
2203 How can RCU possibly handle grace periods before all of its
2204 kthreads have been spawned???
2206 <tr><th align="left">Answer:</th></tr>
2207 <tr><td bgcolor="#ffffff"><font color="ffffff">
2211 <p><font color="ffffff">
2212 During the “dead zone” between the time that the
2213 scheduler spawns the first task and the time that all of RCU's
2214 kthreads have been spawned, all synchronous grace periods are
2215 handled by the expedited grace-period mechanism.
2216 At runtime, this expedited mechanism relies on workqueues, but
2217 during the dead zone the requesting task itself drives the
2218 desired expedited grace period.
2219 Because dead-zone execution takes place within task context,
2221 Once the dead zone ends, expedited grace periods go back to
2222 using workqueues, as is required to avoid problems that would
2223 otherwise occur when a user task received a POSIX signal while
2224 driving an expedited grace period.
2227 <p><font color="ffffff">
2228 And yes, this does mean that it is unhelpful to send POSIX
2229 signals to random tasks between the time that the scheduler
2230 spawns its first kthread and the time that RCU's kthreads
2231 have all been spawned.
2232 If there ever turns out to be a good reason for sending POSIX
2233 signals during that time, appropriate adjustments will be made.
2234 (If it turns out that POSIX signals are sent during this time for
2235 no good reason, other adjustments will be made, appropriate
2238 <tr><td> </td></tr>
2242 I learned of these boot-time requirements as a result of a series of
2245 <h3><a name="Interrupts and NMIs">Interrupts and NMIs</a></h3>
2248 The Linux kernel has interrupts, and RCU read-side critical sections are
2249 legal within interrupt handlers and within interrupt-disabled regions
2250 of code, as are invocations of <tt>call_rcu()</tt>.
2253 Some Linux-kernel architectures can enter an interrupt handler from
2254 non-idle process context, and then just never leave it, instead stealthily
2255 transitioning back to process context.
2256 This trick is sometimes used to invoke system calls from inside the kernel.
2257 These “half-interrupts” mean that RCU has to be very careful
2258 about how it counts interrupt nesting levels.
2259 I learned of this requirement the hard way during a rewrite
2260 of RCU's dyntick-idle code.
2263 The Linux kernel has non-maskable interrupts (NMIs), and
2264 RCU read-side critical sections are legal within NMI handlers.
2265 Thankfully, RCU update-side primitives, including
2266 <tt>call_rcu()</tt>, are prohibited within NMI handlers.
2269 The name notwithstanding, some Linux-kernel architectures
2270 can have nested NMIs, which RCU must handle correctly.
2272 <a href="https://lkml.kernel.org/g/CALCETrXLq1y7e_dKFPgou-FKHB6Pu-r8+t-6Ds+8=va7anBWDA@mail.gmail.com">surprised me</a>
2273 with this requirement;
2274 he also kindly surprised me with
2275 <a href="https://lkml.kernel.org/g/CALCETrXSY9JpW3uE6H8WYk81sg56qasA2aqmjMPsq5dOtzso=g@mail.gmail.com">an algorithm</a>
2276 that meets this requirement.
2278 <h3><a name="Loadable Modules">Loadable Modules</a></h3>
2281 The Linux kernel has loadable modules, and these modules can
2283 After a given module has been unloaded, any attempt to call
2284 one of its functions results in a segmentation fault.
2285 The module-unload functions must therefore cancel any
2286 delayed calls to loadable-module functions, for example,
2287 any outstanding <tt>mod_timer()</tt> must be dealt with
2288 via <tt>del_timer_sync()</tt> or similar.
2291 Unfortunately, there is no way to cancel an RCU callback;
2292 once you invoke <tt>call_rcu()</tt>, the callback function is
2293 going to eventually be invoked, unless the system goes down first.
2294 Because it is normally considered socially irresponsible to crash the system
2295 in response to a module unload request, we need some other way
2296 to deal with in-flight RCU callbacks.
2299 RCU therefore provides
2300 <tt><a href="https://lwn.net/Articles/217484/">rcu_barrier()</a></tt>,
2301 which waits until all in-flight RCU callbacks have been invoked.
2302 If a module uses <tt>call_rcu()</tt>, its exit function should therefore
2303 prevent any future invocation of <tt>call_rcu()</tt>, then invoke
2304 <tt>rcu_barrier()</tt>.
2305 In theory, the underlying module-unload code could invoke
2306 <tt>rcu_barrier()</tt> unconditionally, but in practice this would
2307 incur unacceptable latencies.
2310 Nikita Danilov noted this requirement for an analogous filesystem-unmount
2311 situation, and Dipankar Sarma incorporated <tt>rcu_barrier()</tt> into RCU.
2312 The need for <tt>rcu_barrier()</tt> for module unloading became
2316 <b>Important note</b>: The <tt>rcu_barrier()</tt> function is not,
2317 repeat, <i>not</i>, obligated to wait for a grace period.
2318 It is instead only required to wait for RCU callbacks that have
2319 already been posted.
2320 Therefore, if there are no RCU callbacks posted anywhere in the system,
2321 <tt>rcu_barrier()</tt> is within its rights to return immediately.
2322 Even if there are callbacks posted, <tt>rcu_barrier()</tt> does not
2323 necessarily need to wait for a grace period.
2326 <tr><th> </th></tr>
2327 <tr><th align="left">Quick Quiz:</th></tr>
2330 Each RCU callbacks must wait for a grace period to complete,
2331 and <tt>rcu_barrier()</tt> must wait for each pre-existing
2332 callback to be invoked.
2333 Doesn't <tt>rcu_barrier()</tt> therefore need to wait for
2334 a full grace period if there is even one callback posted anywhere
2337 <tr><th align="left">Answer:</th></tr>
2338 <tr><td bgcolor="#ffffff"><font color="ffffff">
2342 <p><font color="ffffff">
2343 Yes, each RCU callbacks must wait for a grace period to complete,
2344 but it might well be partly (or even completely) finished waiting
2345 by the time <tt>rcu_barrier()</tt> is invoked.
2346 In that case, <tt>rcu_barrier()</tt> need only wait for the
2347 remaining portion of the grace period to elapse.
2348 So even if there are quite a few callbacks posted,
2349 <tt>rcu_barrier()</tt> might well return quite quickly.
2352 <p><font color="ffffff">
2353 So if you need to wait for a grace period as well as for all
2354 pre-existing callbacks, you will need to invoke both
2355 <tt>synchronize_rcu()</tt> and <tt>rcu_barrier()</tt>.
2356 If latency is a concern, you can always use workqueues
2357 to invoke them concurrently.
2359 <tr><td> </td></tr>
2362 <h3><a name="Hotplug CPU">Hotplug CPU</a></h3>
2365 The Linux kernel supports CPU hotplug, which means that CPUs
2367 It is of course illegal to use any RCU API member from an offline CPU,
2368 with the exception of <a href="#Sleepable RCU">SRCU</a> read-side
2370 This requirement was present from day one in DYNIX/ptx, but
2371 on the other hand, the Linux kernel's CPU-hotplug implementation
2372 is “interesting.”
2375 The Linux-kernel CPU-hotplug implementation has notifiers that
2376 are used to allow the various kernel subsystems (including RCU)
2377 to respond appropriately to a given CPU-hotplug operation.
2378 Most RCU operations may be invoked from CPU-hotplug notifiers,
2379 including even synchronous grace-period operations such as
2380 <tt>synchronize_rcu()</tt> and <tt>synchronize_rcu_expedited()</tt>.
2383 However, all-callback-wait operations such as
2384 <tt>rcu_barrier()</tt> are also not supported, due to the
2385 fact that there are phases of CPU-hotplug operations where
2386 the outgoing CPU's callbacks will not be invoked until after
2387 the CPU-hotplug operation ends, which could also result in deadlock.
2388 Furthermore, <tt>rcu_barrier()</tt> blocks CPU-hotplug operations
2389 during its execution, which results in another type of deadlock
2390 when invoked from a CPU-hotplug notifier.
2392 <h3><a name="Scheduler and RCU">Scheduler and RCU</a></h3>
2395 RCU depends on the scheduler, and the scheduler uses RCU to
2396 protect some of its data structures.
2397 This means the scheduler is forbidden from acquiring
2398 the runqueue locks and the priority-inheritance locks
2399 in the middle of an outermost RCU read-side critical section unless either
2400 (1) it releases them before exiting that same
2401 RCU read-side critical section, or
2402 (2) interrupts are disabled across
2403 that entire RCU read-side critical section.
2404 This same prohibition also applies (recursively!) to any lock that is acquired
2405 while holding any lock to which this prohibition applies.
2406 Adhering to this rule prevents preemptible RCU from invoking
2407 <tt>rcu_read_unlock_special()</tt> while either runqueue or
2408 priority-inheritance locks are held, thus avoiding deadlock.
2411 Prior to v4.4, it was only necessary to disable preemption across
2412 RCU read-side critical sections that acquired scheduler locks.
2413 In v4.4, expedited grace periods started using IPIs, and these
2414 IPIs could force a <tt>rcu_read_unlock()</tt> to take the slowpath.
2415 Therefore, this expedited-grace-period change required disabling of
2416 interrupts, not just preemption.
2419 For RCU's part, the preemptible-RCU <tt>rcu_read_unlock()</tt>
2420 implementation must be written carefully to avoid similar deadlocks.
2421 In particular, <tt>rcu_read_unlock()</tt> must tolerate an
2422 interrupt where the interrupt handler invokes both
2423 <tt>rcu_read_lock()</tt> and <tt>rcu_read_unlock()</tt>.
2424 This possibility requires <tt>rcu_read_unlock()</tt> to use
2425 negative nesting levels to avoid destructive recursion via
2426 interrupt handler's use of RCU.
2429 This pair of mutual scheduler-RCU requirements came as a
2430 <a href="https://lwn.net/Articles/453002/">complete surprise</a>.
2433 As noted above, RCU makes use of kthreads, and it is necessary to
2434 avoid excessive CPU-time accumulation by these kthreads.
2435 This requirement was no surprise, but RCU's violation of it
2436 when running context-switch-heavy workloads when built with
2437 <tt>CONFIG_NO_HZ_FULL=y</tt>
2438 <a href="http://www.rdrop.com/users/paulmck/scalability/paper/BareMetal.2015.01.15b.pdf">did come as a surprise [PDF]</a>.
2439 RCU has made good progress towards meeting this requirement, even
2440 for context-switch-have <tt>CONFIG_NO_HZ_FULL=y</tt> workloads,
2441 but there is room for further improvement.
2443 <h3><a name="Tracing and RCU">Tracing and RCU</a></h3>
2446 It is possible to use tracing on RCU code, but tracing itself
2448 For this reason, <tt>rcu_dereference_raw_notrace()</tt>
2449 is provided for use by tracing, which avoids the destructive
2450 recursion that could otherwise ensue.
2451 This API is also used by virtualization in some architectures,
2452 where RCU readers execute in environments in which tracing
2454 The tracing folks both located the requirement and provided the
2455 needed fix, so this surprise requirement was relatively painless.
2457 <h3><a name="Energy Efficiency">Energy Efficiency</a></h3>
2460 Interrupting idle CPUs is considered socially unacceptable,
2461 especially by people with battery-powered embedded systems.
2462 RCU therefore conserves energy by detecting which CPUs are
2463 idle, including tracking CPUs that have been interrupted from idle.
2464 This is a large part of the energy-efficiency requirement,
2465 so I learned of this via an irate phone call.
2468 Because RCU avoids interrupting idle CPUs, it is illegal to
2469 execute an RCU read-side critical section on an idle CPU.
2470 (Kernels built with <tt>CONFIG_PROVE_RCU=y</tt> will splat
2472 The <tt>RCU_NONIDLE()</tt> macro and <tt>_rcuidle</tt>
2473 event tracing is provided to work around this restriction.
2474 In addition, <tt>rcu_is_watching()</tt> may be used to
2475 test whether or not it is currently legal to run RCU read-side
2476 critical sections on this CPU.
2477 I learned of the need for diagnostics on the one hand
2478 and <tt>RCU_NONIDLE()</tt> on the other while inspecting
2480 Steven Rostedt supplied <tt>_rcuidle</tt> event tracing,
2481 which is used quite heavily in the idle loop.
2482 However, there are some restrictions on the code placed within
2483 <tt>RCU_NONIDLE()</tt>:
2486 <li> Blocking is prohibited.
2487 In practice, this is not a serious restriction given that idle
2488 tasks are prohibited from blocking to begin with.
2489 <li> Although nesting <tt>RCU_NONIDLE()</tt> is permitted, they cannot
2490 nest indefinitely deeply.
2491 However, given that they can be nested on the order of a million
2492 deep, even on 32-bit systems, this should not be a serious
2494 This nesting limit would probably be reached long after the
2495 compiler OOMed or the stack overflowed.
2496 <li> Any code path that enters <tt>RCU_NONIDLE()</tt> must sequence
2497 out of that same <tt>RCU_NONIDLE()</tt>.
2498 For example, the following is grossly illegal:
2504 3 goto bad_idea; /* BUG!!! */
2505 4 do_something_else();});
2511 It is just as illegal to transfer control into the middle of
2512 <tt>RCU_NONIDLE()</tt>'s argument.
2513 Yes, in theory, you could transfer in as long as you also
2514 transferred out, but in practice you could also expect to get sharply
2515 worded review comments.
2519 It is similarly socially unacceptable to interrupt an
2520 <tt>nohz_full</tt> CPU running in userspace.
2521 RCU must therefore track <tt>nohz_full</tt> userspace
2523 RCU must therefore be able to sample state at two points in
2524 time, and be able to determine whether or not some other CPU spent
2525 any time idle and/or executing in userspace.
2528 These energy-efficiency requirements have proven quite difficult to
2529 understand and to meet, for example, there have been more than five
2530 clean-sheet rewrites of RCU's energy-efficiency code, the last of
2531 which was finally able to demonstrate
2532 <a href="http://www.rdrop.com/users/paulmck/realtime/paper/AMPenergy.2013.04.19a.pdf">real energy savings running on real hardware [PDF]</a>.
2534 I learned of many of these requirements via angry phone calls:
2535 Flaming me on the Linux-kernel mailing list was apparently not
2536 sufficient to fully vent their ire at RCU's energy-efficiency bugs!
2538 <h3><a name="Scheduling-Clock Interrupts and RCU">
2539 Scheduling-Clock Interrupts and RCU</a></h3>
2542 The kernel transitions between in-kernel non-idle execution, userspace
2543 execution, and the idle loop.
2544 Depending on kernel configuration, RCU handles these states differently:
2547 <tr><th><tt>HZ</tt> Kconfig</th>
2551 <tr><th align="left"><tt>HZ_PERIODIC</tt></th>
2552 <td>Can rely on scheduling-clock interrupt.</td>
2553 <td>Can rely on scheduling-clock interrupt and its
2554 detection of interrupt from usermode.</td>
2555 <td>Can rely on RCU's dyntick-idle detection.</td></tr>
2556 <tr><th align="left"><tt>NO_HZ_IDLE</tt></th>
2557 <td>Can rely on scheduling-clock interrupt.</td>
2558 <td>Can rely on scheduling-clock interrupt and its
2559 detection of interrupt from usermode.</td>
2560 <td>Can rely on RCU's dyntick-idle detection.</td></tr>
2561 <tr><th align="left"><tt>NO_HZ_FULL</tt></th>
2562 <td>Can only sometimes rely on scheduling-clock interrupt.
2563 In other cases, it is necessary to bound kernel execution
2564 times and/or use IPIs.</td>
2565 <td>Can rely on RCU's dyntick-idle detection.</td>
2566 <td>Can rely on RCU's dyntick-idle detection.</td></tr>
2570 <tr><th> </th></tr>
2571 <tr><th align="left">Quick Quiz:</th></tr>
2573 Why can't <tt>NO_HZ_FULL</tt> in-kernel execution rely on the
2574 scheduling-clock interrupt, just like <tt>HZ_PERIODIC</tt>
2575 and <tt>NO_HZ_IDLE</tt> do?
2577 <tr><th align="left">Answer:</th></tr>
2578 <tr><td bgcolor="#ffffff"><font color="ffffff">
2579 Because, as a performance optimization, <tt>NO_HZ_FULL</tt>
2580 does not necessarily re-enable the scheduling-clock interrupt
2581 on entry to each and every system call.
2583 <tr><td> </td></tr>
2587 However, RCU must be reliably informed as to whether any given
2588 CPU is currently in the idle loop, and, for <tt>NO_HZ_FULL</tt>,
2589 also whether that CPU is executing in usermode, as discussed
2590 <a href="#Energy Efficiency">earlier</a>.
2591 It also requires that the scheduling-clock interrupt be enabled when
2595 <li> If a CPU is either idle or executing in usermode, and RCU believes
2596 it is non-idle, the scheduling-clock tick had better be running.
2597 Otherwise, you will get RCU CPU stall warnings. Or at best,
2598 very long (11-second) grace periods, with a pointless IPI waking
2599 the CPU from time to time.
2600 <li> If a CPU is in a portion of the kernel that executes RCU read-side
2601 critical sections, and RCU believes this CPU to be idle, you will get
2602 random memory corruption. <b>DON'T DO THIS!!!</b>
2604 <br>This is one reason to test with lockdep, which will complain
2605 about this sort of thing.
2606 <li> If a CPU is in a portion of the kernel that is absolutely
2607 positively no-joking guaranteed to never execute any RCU read-side
2608 critical sections, and RCU believes this CPU to to be idle,
2609 no problem. This sort of thing is used by some architectures
2610 for light-weight exception handlers, which can then avoid the
2611 overhead of <tt>rcu_irq_enter()</tt> and <tt>rcu_irq_exit()</tt>
2612 at exception entry and exit, respectively.
2613 Some go further and avoid the entireties of <tt>irq_enter()</tt>
2614 and <tt>irq_exit()</tt>.
2616 <br>Just make very sure you are running some of your tests with
2617 <tt>CONFIG_PROVE_RCU=y</tt>, just in case one of your code paths
2618 was in fact joking about not doing RCU read-side critical sections.
2619 <li> If a CPU is executing in the kernel with the scheduling-clock
2620 interrupt disabled and RCU believes this CPU to be non-idle,
2621 and if the CPU goes idle (from an RCU perspective) every few
2622 jiffies, no problem. It is usually OK for there to be the
2623 occasional gap between idle periods of up to a second or so.
2625 <br>If the gap grows too long, you get RCU CPU stall warnings.
2626 <li> If a CPU is either idle or executing in usermode, and RCU believes
2627 it to be idle, of course no problem.
2628 <li> If a CPU is executing in the kernel, the kernel code
2629 path is passing through quiescent states at a reasonable
2630 frequency (preferably about once per few jiffies, but the
2631 occasional excursion to a second or so is usually OK) and the
2632 scheduling-clock interrupt is enabled, of course no problem.
2634 <br>If the gap between a successive pair of quiescent states grows
2635 too long, you get RCU CPU stall warnings.
2639 <tr><th> </th></tr>
2640 <tr><th align="left">Quick Quiz:</th></tr>
2642 But what if my driver has a hardware interrupt handler
2643 that can run for many seconds?
2644 I cannot invoke <tt>schedule()</tt> from an hardware
2645 interrupt handler, after all!
2647 <tr><th align="left">Answer:</th></tr>
2648 <tr><td bgcolor="#ffffff"><font color="ffffff">
2649 One approach is to do <tt>rcu_irq_exit();rcu_irq_enter();</tt>
2651 But given that long-running interrupt handlers can cause
2652 other problems, not least for response time, shouldn't you
2653 work to keep your interrupt handler's runtime within reasonable
2656 <tr><td> </td></tr>
2660 But as long as RCU is properly informed of kernel state transitions between
2661 in-kernel execution, usermode execution, and idle, and as long as the
2662 scheduling-clock interrupt is enabled when RCU needs it to be, you
2663 can rest assured that the bugs you encounter will be in some other
2664 part of RCU or some other part of the kernel!
2666 <h3><a name="Memory Efficiency">Memory Efficiency</a></h3>
2669 Although small-memory non-realtime systems can simply use Tiny RCU,
2670 code size is only one aspect of memory efficiency.
2671 Another aspect is the size of the <tt>rcu_head</tt> structure
2672 used by <tt>call_rcu()</tt> and <tt>kfree_rcu()</tt>.
2673 Although this structure contains nothing more than a pair of pointers,
2674 it does appear in many RCU-protected data structures, including
2675 some that are size critical.
2676 The <tt>page</tt> structure is a case in point, as evidenced by
2677 the many occurrences of the <tt>union</tt> keyword within that structure.
2680 This need for memory efficiency is one reason that RCU uses hand-crafted
2681 singly linked lists to track the <tt>rcu_head</tt> structures that
2682 are waiting for a grace period to elapse.
2683 It is also the reason why <tt>rcu_head</tt> structures do not contain
2684 debug information, such as fields tracking the file and line of the
2685 <tt>call_rcu()</tt> or <tt>kfree_rcu()</tt> that posted them.
2686 Although this information might appear in debug-only kernel builds at some
2687 point, in the meantime, the <tt>->func</tt> field will often provide
2688 the needed debug information.
2691 However, in some cases, the need for memory efficiency leads to even
2692 more extreme measures.
2693 Returning to the <tt>page</tt> structure, the <tt>rcu_head</tt> field
2694 shares storage with a great many other structures that are used at
2695 various points in the corresponding page's lifetime.
2696 In order to correctly resolve certain
2697 <a href="https://lkml.kernel.org/g/1439976106-137226-1-git-send-email-kirill.shutemov@linux.intel.com">race conditions</a>,
2698 the Linux kernel's memory-management subsystem needs a particular bit
2699 to remain zero during all phases of grace-period processing,
2700 and that bit happens to map to the bottom bit of the
2701 <tt>rcu_head</tt> structure's <tt>->next</tt> field.
2702 RCU makes this guarantee as long as <tt>call_rcu()</tt>
2703 is used to post the callback, as opposed to <tt>kfree_rcu()</tt>
2704 or some future “lazy”
2705 variant of <tt>call_rcu()</tt> that might one day be created for
2706 energy-efficiency purposes.
2709 That said, there are limits.
2710 RCU requires that the <tt>rcu_head</tt> structure be aligned to a
2711 two-byte boundary, and passing a misaligned <tt>rcu_head</tt>
2712 structure to one of the <tt>call_rcu()</tt> family of functions
2713 will result in a splat.
2714 It is therefore necessary to exercise caution when packing
2715 structures containing fields of type <tt>rcu_head</tt>.
2716 Why not a four-byte or even eight-byte alignment requirement?
2717 Because the m68k architecture provides only two-byte alignment,
2718 and thus acts as alignment's least common denominator.
2721 The reason for reserving the bottom bit of pointers to
2722 <tt>rcu_head</tt> structures is to leave the door open to
2723 “lazy” callbacks whose invocations can safely be deferred.
2724 Deferring invocation could potentially have energy-efficiency
2725 benefits, but only if the rate of non-lazy callbacks decreases
2726 significantly for some important workload.
2727 In the meantime, reserving the bottom bit keeps this option open
2728 in case it one day becomes useful.
2730 <h3><a name="Performance, Scalability, Response Time, and Reliability">
2731 Performance, Scalability, Response Time, and Reliability</a></h3>
2735 <a href="#Performance and Scalability">earlier discussion</a>,
2736 RCU is used heavily by hot code paths in performance-critical
2737 portions of the Linux kernel's networking, security, virtualization,
2738 and scheduling code paths.
2739 RCU must therefore use efficient implementations, especially in its
2740 read-side primitives.
2741 To that end, it would be good if preemptible RCU's implementation
2742 of <tt>rcu_read_lock()</tt> could be inlined, however, doing
2743 this requires resolving <tt>#include</tt> issues with the
2744 <tt>task_struct</tt> structure.
2747 The Linux kernel supports hardware configurations with up to
2748 4096 CPUs, which means that RCU must be extremely scalable.
2749 Algorithms that involve frequent acquisitions of global locks or
2750 frequent atomic operations on global variables simply cannot be
2751 tolerated within the RCU implementation.
2752 RCU therefore makes heavy use of a combining tree based on the
2753 <tt>rcu_node</tt> structure.
2754 RCU is required to tolerate all CPUs continuously invoking any
2755 combination of RCU's runtime primitives with minimal per-operation
2757 In fact, in many cases, increasing load must <i>decrease</i> the
2758 per-operation overhead, witness the batching optimizations for
2759 <tt>synchronize_rcu()</tt>, <tt>call_rcu()</tt>,
2760 <tt>synchronize_rcu_expedited()</tt>, and <tt>rcu_barrier()</tt>.
2761 As a general rule, RCU must cheerfully accept whatever the
2762 rest of the Linux kernel decides to throw at it.
2765 The Linux kernel is used for real-time workloads, especially
2766 in conjunction with the
2767 <a href="https://rt.wiki.kernel.org/index.php/Main_Page">-rt patchset</a>.
2768 The real-time-latency response requirements are such that the
2769 traditional approach of disabling preemption across RCU
2770 read-side critical sections is inappropriate.
2771 Kernels built with <tt>CONFIG_PREEMPT=y</tt> therefore
2772 use an RCU implementation that allows RCU read-side critical
2773 sections to be preempted.
2774 This requirement made its presence known after users made it
2775 clear that an earlier
2776 <a href="https://lwn.net/Articles/107930/">real-time patch</a>
2777 did not meet their needs, in conjunction with some
2778 <a href="https://lkml.kernel.org/g/20050318002026.GA2693@us.ibm.com">RCU issues</a>
2779 encountered by a very early version of the -rt patchset.
2782 In addition, RCU must make do with a sub-100-microsecond real-time latency
2784 In fact, on smaller systems with the -rt patchset, the Linux kernel
2785 provides sub-20-microsecond real-time latencies for the whole kernel,
2787 RCU's scalability and latency must therefore be sufficient for
2788 these sorts of configurations.
2789 To my surprise, the sub-100-microsecond real-time latency budget
2790 <a href="http://www.rdrop.com/users/paulmck/realtime/paper/bigrt.2013.01.31a.LCA.pdf">
2791 applies to even the largest systems [PDF]</a>,
2792 up to and including systems with 4096 CPUs.
2793 This real-time requirement motivated the grace-period kthread, which
2794 also simplified handling of a number of race conditions.
2797 RCU must avoid degrading real-time response for CPU-bound threads, whether
2798 executing in usermode (which is one use case for
2799 <tt>CONFIG_NO_HZ_FULL=y</tt>) or in the kernel.
2800 That said, CPU-bound loops in the kernel must execute
2801 <tt>cond_resched()</tt> at least once per few tens of milliseconds
2802 in order to avoid receiving an IPI from RCU.
2805 Finally, RCU's status as a synchronization primitive means that
2806 any RCU failure can result in arbitrary memory corruption that can be
2807 extremely difficult to debug.
2808 This means that RCU must be extremely reliable, which in
2809 practice also means that RCU must have an aggressive stress-test
2811 This stress-test suite is called <tt>rcutorture</tt>.
2814 Although the need for <tt>rcutorture</tt> was no surprise,
2815 the current immense popularity of the Linux kernel is posing
2816 interesting—and perhaps unprecedented—validation
2818 To see this, keep in mind that there are well over one billion
2819 instances of the Linux kernel running today, given Android
2820 smartphones, Linux-powered televisions, and servers.
2821 This number can be expected to increase sharply with the advent of
2822 the celebrated Internet of Things.
2825 Suppose that RCU contains a race condition that manifests on average
2826 once per million years of runtime.
2827 This bug will be occurring about three times per <i>day</i> across
2829 RCU could simply hide behind hardware error rates, given that no one
2830 should really expect their smartphone to last for a million years.
2831 However, anyone taking too much comfort from this thought should
2832 consider the fact that in most jurisdictions, a successful multi-year
2833 test of a given mechanism, which might include a Linux kernel,
2834 suffices for a number of types of safety-critical certifications.
2835 In fact, rumor has it that the Linux kernel is already being used
2836 in production for safety-critical applications.
2837 I don't know about you, but I would feel quite bad if a bug in RCU
2839 Which might explain my recent focus on validation and verification.
2841 <h2><a name="Other RCU Flavors">Other RCU Flavors</a></h2>
2844 One of the more surprising things about RCU is that there are now
2845 no fewer than five <i>flavors</i>, or API families.
2846 In addition, the primary flavor that has been the sole focus up to
2847 this point has two different implementations, non-preemptible and
2849 The other four flavors are listed below, with requirements for each
2850 described in a separate section.
2853 <li> <a href="#Bottom-Half Flavor">Bottom-Half Flavor</a>
2854 <li> <a href="#Sched Flavor">Sched Flavor</a>
2855 <li> <a href="#Sleepable RCU">Sleepable RCU</a>
2856 <li> <a href="#Tasks RCU">Tasks RCU</a>
2857 <li> <a href="#Waiting for Multiple Grace Periods">
2858 Waiting for Multiple Grace Periods</a>
2861 <h3><a name="Bottom-Half Flavor">Bottom-Half Flavor</a></h3>
2864 The softirq-disable (AKA “bottom-half”,
2865 hence the “_bh” abbreviations)
2866 flavor of RCU, or <i>RCU-bh</i>, was developed by
2867 Dipankar Sarma to provide a flavor of RCU that could withstand the
2868 network-based denial-of-service attacks researched by Robert
2870 These attacks placed so much networking load on the system
2871 that some of the CPUs never exited softirq execution,
2872 which in turn prevented those CPUs from ever executing a context switch,
2873 which, in the RCU implementation of that time, prevented grace periods
2875 The result was an out-of-memory condition and a system hang.
2878 The solution was the creation of RCU-bh, which does
2879 <tt>local_bh_disable()</tt>
2880 across its read-side critical sections, and which uses the transition
2881 from one type of softirq processing to another as a quiescent state
2882 in addition to context switch, idle, user mode, and offline.
2883 This means that RCU-bh grace periods can complete even when some of
2884 the CPUs execute in softirq indefinitely, thus allowing algorithms
2885 based on RCU-bh to withstand network-based denial-of-service attacks.
2889 <tt>rcu_read_lock_bh()</tt> and <tt>rcu_read_unlock_bh()</tt>
2890 disable and re-enable softirq handlers, any attempt to start a softirq
2892 RCU-bh read-side critical section will be deferred.
2893 In this case, <tt>rcu_read_unlock_bh()</tt>
2894 will invoke softirq processing, which can take considerable time.
2895 One can of course argue that this softirq overhead should be associated
2896 with the code following the RCU-bh read-side critical section rather
2897 than <tt>rcu_read_unlock_bh()</tt>, but the fact
2898 is that most profiling tools cannot be expected to make this sort
2899 of fine distinction.
2900 For example, suppose that a three-millisecond-long RCU-bh read-side
2901 critical section executes during a time of heavy networking load.
2902 There will very likely be an attempt to invoke at least one softirq
2903 handler during that three milliseconds, but any such invocation will
2904 be delayed until the time of the <tt>rcu_read_unlock_bh()</tt>.
2905 This can of course make it appear at first glance as if
2906 <tt>rcu_read_unlock_bh()</tt> was executing very slowly.
2910 <a href="https://lwn.net/Articles/609973/#RCU Per-Flavor API Table">RCU-bh API</a>
2912 <tt>rcu_read_lock_bh()</tt>,
2913 <tt>rcu_read_unlock_bh()</tt>,
2914 <tt>rcu_dereference_bh()</tt>,
2915 <tt>rcu_dereference_bh_check()</tt>,
2916 <tt>synchronize_rcu_bh()</tt>,
2917 <tt>synchronize_rcu_bh_expedited()</tt>,
2918 <tt>call_rcu_bh()</tt>,
2919 <tt>rcu_barrier_bh()</tt>, and
2920 <tt>rcu_read_lock_bh_held()</tt>.
2922 <h3><a name="Sched Flavor">Sched Flavor</a></h3>
2925 Before preemptible RCU, waiting for an RCU grace period had the
2926 side effect of also waiting for all pre-existing interrupt
2928 However, there are legitimate preemptible-RCU implementations that
2929 do not have this property, given that any point in the code outside
2930 of an RCU read-side critical section can be a quiescent state.
2931 Therefore, <i>RCU-sched</i> was created, which follows “classic”
2932 RCU in that an RCU-sched grace period waits for for pre-existing
2933 interrupt and NMI handlers.
2934 In kernels built with <tt>CONFIG_PREEMPT=n</tt>, the RCU and RCU-sched
2935 APIs have identical implementations, while kernels built with
2936 <tt>CONFIG_PREEMPT=y</tt> provide a separate implementation for each.
2939 Note well that in <tt>CONFIG_PREEMPT=y</tt> kernels,
2940 <tt>rcu_read_lock_sched()</tt> and <tt>rcu_read_unlock_sched()</tt>
2941 disable and re-enable preemption, respectively.
2942 This means that if there was a preemption attempt during the
2943 RCU-sched read-side critical section, <tt>rcu_read_unlock_sched()</tt>
2944 will enter the scheduler, with all the latency and overhead entailed.
2945 Just as with <tt>rcu_read_unlock_bh()</tt>, this can make it look
2946 as if <tt>rcu_read_unlock_sched()</tt> was executing very slowly.
2947 However, the highest-priority task won't be preempted, so that task
2948 will enjoy low-overhead <tt>rcu_read_unlock_sched()</tt> invocations.
2952 <a href="https://lwn.net/Articles/609973/#RCU Per-Flavor API Table">RCU-sched API</a>
2954 <tt>rcu_read_lock_sched()</tt>,
2955 <tt>rcu_read_unlock_sched()</tt>,
2956 <tt>rcu_read_lock_sched_notrace()</tt>,
2957 <tt>rcu_read_unlock_sched_notrace()</tt>,
2958 <tt>rcu_dereference_sched()</tt>,
2959 <tt>rcu_dereference_sched_check()</tt>,
2960 <tt>synchronize_sched()</tt>,
2961 <tt>synchronize_rcu_sched_expedited()</tt>,
2962 <tt>call_rcu_sched()</tt>,
2963 <tt>rcu_barrier_sched()</tt>, and
2964 <tt>rcu_read_lock_sched_held()</tt>.
2965 However, anything that disables preemption also marks an RCU-sched
2966 read-side critical section, including
2967 <tt>preempt_disable()</tt> and <tt>preempt_enable()</tt>,
2968 <tt>local_irq_save()</tt> and <tt>local_irq_restore()</tt>,
2971 <h3><a name="Sleepable RCU">Sleepable RCU</a></h3>
2974 For well over a decade, someone saying “I need to block within
2975 an RCU read-side critical section” was a reliable indication
2976 that this someone did not understand RCU.
2977 After all, if you are always blocking in an RCU read-side critical
2978 section, you can probably afford to use a higher-overhead synchronization
2980 However, that changed with the advent of the Linux kernel's notifiers,
2981 whose RCU read-side critical
2982 sections almost never sleep, but sometimes need to.
2983 This resulted in the introduction of
2984 <a href="https://lwn.net/Articles/202847/">sleepable RCU</a>,
2988 SRCU allows different domains to be defined, with each such domain
2989 defined by an instance of an <tt>srcu_struct</tt> structure.
2990 A pointer to this structure must be passed in to each SRCU function,
2991 for example, <tt>synchronize_srcu(&ss)</tt>, where
2992 <tt>ss</tt> is the <tt>srcu_struct</tt> structure.
2993 The key benefit of these domains is that a slow SRCU reader in one
2994 domain does not delay an SRCU grace period in some other domain.
2995 That said, one consequence of these domains is that read-side code
2996 must pass a “cookie” from <tt>srcu_read_lock()</tt>
2997 to <tt>srcu_read_unlock()</tt>, for example, as follows:
3003 3 idx = srcu_read_lock(&ss);
3005 5 srcu_read_unlock(&ss, idx);
3010 As noted above, it is legal to block within SRCU read-side critical sections,
3011 however, with great power comes great responsibility.
3012 If you block forever in one of a given domain's SRCU read-side critical
3013 sections, then that domain's grace periods will also be blocked forever.
3014 Of course, one good way to block forever is to deadlock, which can
3015 happen if any operation in a given domain's SRCU read-side critical
3016 section can block waiting, either directly or indirectly, for that domain's
3017 grace period to elapse.
3018 For example, this results in a self-deadlock:
3024 3 idx = srcu_read_lock(&ss);
3026 5 synchronize_srcu(&ss);
3027 6 srcu_read_unlock(&ss, idx);
3032 However, if line 5 acquired a mutex that was held across
3033 a <tt>synchronize_srcu()</tt> for domain <tt>ss</tt>,
3034 deadlock would still be possible.
3035 Furthermore, if line 5 acquired a mutex that was held across
3036 a <tt>synchronize_srcu()</tt> for some other domain <tt>ss1</tt>,
3037 and if an <tt>ss1</tt>-domain SRCU read-side critical section
3038 acquired another mutex that was held across as <tt>ss</tt>-domain
3039 <tt>synchronize_srcu()</tt>,
3040 deadlock would again be possible.
3041 Such a deadlock cycle could extend across an arbitrarily large number
3042 of different SRCU domains.
3043 Again, with great power comes great responsibility.
3046 Unlike the other RCU flavors, SRCU read-side critical sections can
3047 run on idle and even offline CPUs.
3048 This ability requires that <tt>srcu_read_lock()</tt> and
3049 <tt>srcu_read_unlock()</tt> contain memory barriers, which means
3050 that SRCU readers will run a bit slower than would RCU readers.
3051 It also motivates the <tt>smp_mb__after_srcu_read_unlock()</tt>
3052 API, which, in combination with <tt>srcu_read_unlock()</tt>,
3053 guarantees a full memory barrier.
3056 Also unlike other RCU flavors, SRCU's callbacks-wait function
3057 <tt>srcu_barrier()</tt> may be invoked from CPU-hotplug notifiers,
3058 though this is not necessarily a good idea.
3059 The reason that this is possible is that SRCU is insensitive
3060 to whether or not a CPU is online, which means that <tt>srcu_barrier()</tt>
3061 need not exclude CPU-hotplug operations.
3064 SRCU also differs from other RCU flavors in that SRCU's expedited and
3065 non-expedited grace periods are implemented by the same mechanism.
3066 This means that in the current SRCU implementation, expediting a
3067 future grace period has the side effect of expediting all prior
3068 grace periods that have not yet completed.
3069 (But please note that this is a property of the current implementation,
3070 not necessarily of future implementations.)
3071 In addition, if SRCU has been idle for longer than the interval
3072 specified by the <tt>srcutree.exp_holdoff</tt> kernel boot parameter
3073 (25 microseconds by default),
3074 and if a <tt>synchronize_srcu()</tt> invocation ends this idle period,
3075 that invocation will be automatically expedited.
3078 As of v4.12, SRCU's callbacks are maintained per-CPU, eliminating
3079 a locking bottleneck present in prior kernel versions.
3080 Although this will allow users to put much heavier stress on
3081 <tt>call_srcu()</tt>, it is important to note that SRCU does not
3082 yet take any special steps to deal with callback flooding.
3083 So if you are posting (say) 10,000 SRCU callbacks per second per CPU,
3084 you are probably totally OK, but if you intend to post (say) 1,000,000
3085 SRCU callbacks per second per CPU, please run some tests first.
3086 SRCU just might need a few adjustment to deal with that sort of load.
3087 Of course, your mileage may vary based on the speed of your CPUs and
3088 the size of your memory.
3092 <a href="https://lwn.net/Articles/609973/#RCU Per-Flavor API Table">SRCU API</a>
3094 <tt>srcu_read_lock()</tt>,
3095 <tt>srcu_read_unlock()</tt>,
3096 <tt>srcu_dereference()</tt>,
3097 <tt>srcu_dereference_check()</tt>,
3098 <tt>synchronize_srcu()</tt>,
3099 <tt>synchronize_srcu_expedited()</tt>,
3100 <tt>call_srcu()</tt>,
3101 <tt>srcu_barrier()</tt>, and
3102 <tt>srcu_read_lock_held()</tt>.
3104 <tt>DEFINE_SRCU()</tt>,
3105 <tt>DEFINE_STATIC_SRCU()</tt>, and
3106 <tt>init_srcu_struct()</tt>
3107 APIs for defining and initializing <tt>srcu_struct</tt> structures.
3109 <h3><a name="Tasks RCU">Tasks RCU</a></h3>
3112 Some forms of tracing use “trampolines” to handle the
3113 binary rewriting required to install different types of probes.
3114 It would be good to be able to free old trampolines, which sounds
3115 like a job for some form of RCU.
3116 However, because it is necessary to be able to install a trace
3117 anywhere in the code, it is not possible to use read-side markers
3118 such as <tt>rcu_read_lock()</tt> and <tt>rcu_read_unlock()</tt>.
3119 In addition, it does not work to have these markers in the trampoline
3120 itself, because there would need to be instructions following
3121 <tt>rcu_read_unlock()</tt>.
3122 Although <tt>synchronize_rcu()</tt> would guarantee that execution
3123 reached the <tt>rcu_read_unlock()</tt>, it would not be able to
3124 guarantee that execution had completely left the trampoline.
3127 The solution, in the form of
3128 <a href="https://lwn.net/Articles/607117/"><i>Tasks RCU</i></a>,
3130 read-side critical sections that are delimited by voluntary context
3131 switches, that is, calls to <tt>schedule()</tt>,
3132 <tt>cond_resched()</tt>, and
3133 <tt>synchronize_rcu_tasks()</tt>.
3134 In addition, transitions to and from userspace execution also delimit
3135 tasks-RCU read-side critical sections.
3138 The tasks-RCU API is quite compact, consisting only of
3139 <tt>call_rcu_tasks()</tt>,
3140 <tt>synchronize_rcu_tasks()</tt>, and
3141 <tt>rcu_barrier_tasks()</tt>.
3143 <h3><a name="Waiting for Multiple Grace Periods">
3144 Waiting for Multiple Grace Periods</a></h3>
3147 Perhaps you have an RCU protected data structure that is accessed from
3148 RCU read-side critical sections, from softirq handlers, and from
3149 hardware interrupt handlers.
3150 That is three flavors of RCU, the normal flavor, the bottom-half flavor,
3151 and the sched flavor.
3152 How to wait for a compound grace period?
3155 The best approach is usually to “just say no!” and
3156 insert <tt>rcu_read_lock()</tt> and <tt>rcu_read_unlock()</tt>
3157 around each RCU read-side critical section, regardless of what
3158 environment it happens to be in.
3159 But suppose that some of the RCU read-side critical sections are
3160 on extremely hot code paths, and that use of <tt>CONFIG_PREEMPT=n</tt>
3161 is not a viable option, so that <tt>rcu_read_lock()</tt> and
3162 <tt>rcu_read_unlock()</tt> are not free.
3166 You <i>could</i> wait on all three grace periods in succession, as follows:
3170 1 synchronize_rcu();
3171 2 synchronize_rcu_bh();
3172 3 synchronize_sched();
3177 This works, but triples the update-side latency penalty.
3178 In cases where this is not acceptable, <tt>synchronize_rcu_mult()</tt>
3179 may be used to wait on all three flavors of grace period concurrently:
3183 1 synchronize_rcu_mult(call_rcu, call_rcu_bh, call_rcu_sched);
3188 But what if it is necessary to also wait on SRCU?
3189 This can be done as follows:
3193 1 static void call_my_srcu(struct rcu_head *head,
3194 2 void (*func)(struct rcu_head *head))
3196 4 call_srcu(&my_srcu, head, func);
3199 7 synchronize_rcu_mult(call_rcu, call_rcu_bh, call_rcu_sched, call_my_srcu);
3204 If you needed to wait on multiple different flavors of SRCU
3205 (but why???), you would need to create a wrapper function resembling
3206 <tt>call_my_srcu()</tt> for each SRCU flavor.
3209 <tr><th> </th></tr>
3210 <tr><th align="left">Quick Quiz:</th></tr>
3212 But what if I need to wait for multiple RCU flavors, but I also need
3213 the grace periods to be expedited?
3215 <tr><th align="left">Answer:</th></tr>
3216 <tr><td bgcolor="#ffffff"><font color="ffffff">
3217 If you are using expedited grace periods, there should be less penalty
3218 for waiting on them in succession.
3219 But if that is nevertheless a problem, you can use workqueues
3220 or multiple kthreads to wait on the various expedited grace
3221 periods concurrently.
3223 <tr><td> </td></tr>
3227 Again, it is usually better to adjust the RCU read-side critical sections
3228 to use a single flavor of RCU, but when this is not feasible, you can use
3229 <tt>synchronize_rcu_mult()</tt>.
3231 <h2><a name="Possible Future Changes">Possible Future Changes</a></h2>
3234 One of the tricks that RCU uses to attain update-side scalability is
3235 to increase grace-period latency with increasing numbers of CPUs.
3236 If this becomes a serious problem, it will be necessary to rework the
3237 grace-period state machine so as to avoid the need for the additional
3241 Expedited grace periods scan the CPUs, so their latency and overhead
3242 increases with increasing numbers of CPUs.
3243 If this becomes a serious problem on large systems, it will be necessary
3244 to do some redesign to avoid this scalability problem.
3247 RCU disables CPU hotplug in a few places, perhaps most notably in the
3248 <tt>rcu_barrier()</tt> operations.
3249 If there is a strong reason to use <tt>rcu_barrier()</tt> in CPU-hotplug
3250 notifiers, it will be necessary to avoid disabling CPU hotplug.
3251 This would introduce some complexity, so there had better be a <i>very</i>
3255 The tradeoff between grace-period latency on the one hand and interruptions
3256 of other CPUs on the other hand may need to be re-examined.
3257 The desire is of course for zero grace-period latency as well as zero
3258 interprocessor interrupts undertaken during an expedited grace period
3260 While this ideal is unlikely to be achievable, it is quite possible that
3261 further improvements can be made.
3264 The multiprocessor implementations of RCU use a combining tree that
3265 groups CPUs so as to reduce lock contention and increase cache locality.
3266 However, this combining tree does not spread its memory across NUMA
3267 nodes nor does it align the CPU groups with hardware features such
3268 as sockets or cores.
3269 Such spreading and alignment is currently believed to be unnecessary
3270 because the hotpath read-side primitives do not access the combining
3271 tree, nor does <tt>call_rcu()</tt> in the common case.
3272 If you believe that your architecture needs such spreading and alignment,
3273 then your architecture should also benefit from the
3274 <tt>rcutree.rcu_fanout_leaf</tt> boot parameter, which can be set
3275 to the number of CPUs in a socket, NUMA node, or whatever.
3276 If the number of CPUs is too large, use a fraction of the number of
3278 If the number of CPUs is a large prime number, well, that certainly
3279 is an “interesting” architectural choice!
3280 More flexible arrangements might be considered, but only if
3281 <tt>rcutree.rcu_fanout_leaf</tt> has proven inadequate, and only
3282 if the inadequacy has been demonstrated by a carefully run and
3283 realistic system-level workload.
3286 Please note that arrangements that require RCU to remap CPU numbers will
3287 require extremely good demonstration of need and full exploration of
3291 There is an embarrassingly large number of flavors of RCU, and this
3292 number has been increasing over time.
3293 Perhaps it will be possible to combine some at some future date.
3296 RCU's various kthreads are reasonably recent additions.
3297 It is quite likely that adjustments will be required to more gracefully
3298 handle extreme loads.
3299 It might also be necessary to be able to relate CPU utilization by
3300 RCU's kthreads and softirq handlers to the code that instigated this
3302 For example, RCU callback overhead might be charged back to the
3303 originating <tt>call_rcu()</tt> instance, though probably not
3304 in production kernels.
3306 <h2><a name="Summary">Summary</a></h2>
3309 This document has presented more than two decade's worth of RCU
3311 Given that the requirements keep changing, this will not be the last
3312 word on this subject, but at least it serves to get an important
3313 subset of the requirements set forth.
3315 <h2><a name="Acknowledgments">Acknowledgments</a></h2>
3317 I am grateful to Steven Rostedt, Lai Jiangshan, Ingo Molnar,
3318 Oleg Nesterov, Borislav Petkov, Peter Zijlstra, Boqun Feng, and
3319 Andy Lutomirski for their help in rendering
3320 this article human readable, and to Michelle Rankin for her support
3322 Other contributions are acknowledged in the Linux kernel's git archive.
git.samba.org / sfrench / cifs-2.6.git / blob