1 CONFIG_RCU_TRACE debugfs Files and Formats
4 The rcutree and rcutiny implementations of RCU provide debugfs trace
5 output that summarizes counters and state. This information is useful for
6 debugging RCU itself, and can sometimes also help to debug abuses of RCU.
7 The following sections describe the debugfs files and formats, first
8 for rcutree and next for rcutiny.
11 CONFIG_TREE_RCU and CONFIG_TREE_PREEMPT_RCU debugfs Files and Formats
13 These implementations of RCU provides several debugfs files under the
14 top-level directory "rcu":
17 Displays fields in struct rcu_data.
19 Comma-separated values spreadsheet version of rcudata.
21 Displays grace-period counters.
23 Displays the struct rcu_node hierarchy.
25 Displays counts of the reasons rcu_pending() decided that RCU had
28 Displays rcutorture test progress.
30 Displays RCU boosting statistics. Only present if
33 The output of "cat rcu/rcudata" looks as follows:
36 0 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=545/1/0 df=50 of=0 ri=0 ql=163 qs=NRW. kt=0/W/0 ktl=ebc3 b=10 ci=153737 co=0 ca=0
37 1 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=967/1/0 df=58 of=0 ri=0 ql=634 qs=NRW. kt=0/W/1 ktl=58c b=10 ci=191037 co=0 ca=0
38 2 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=1081/1/0 df=175 of=0 ri=0 ql=74 qs=N.W. kt=0/W/2 ktl=da94 b=10 ci=75991 co=0 ca=0
39 3 c=20942 g=20943 pq=1 pqc=20942 qp=1 dt=1846/0/0 df=404 of=0 ri=0 ql=0 qs=.... kt=0/W/3 ktl=d1cd b=10 ci=72261 co=0 ca=0
40 4 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=369/1/0 df=83 of=0 ri=0 ql=48 qs=N.W. kt=0/W/4 ktl=e0e7 b=10 ci=128365 co=0 ca=0
41 5 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=381/1/0 df=64 of=0 ri=0 ql=169 qs=NRW. kt=0/W/5 ktl=fb2f b=10 ci=164360 co=0 ca=0
42 6 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=1037/1/0 df=183 of=0 ri=0 ql=62 qs=N.W. kt=0/W/6 ktl=d2ad b=10 ci=65663 co=0 ca=0
43 7 c=20897 g=20897 pq=1 pqc=20896 qp=0 dt=1572/0/0 df=382 of=0 ri=0 ql=0 qs=.... kt=0/W/7 ktl=cf15 b=10 ci=75006 co=0 ca=0
45 0 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=545/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/0 ktl=ebc3 b=10 ci=0 co=0 ca=0
46 1 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=967/1/0 df=3 of=0 ri=1 ql=0 qs=.... kt=0/W/1 ktl=58c b=10 ci=151 co=0 ca=0
47 2 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=1081/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/2 ktl=da94 b=10 ci=0 co=0 ca=0
48 3 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=1846/0/0 df=8 of=0 ri=1 ql=0 qs=.... kt=0/W/3 ktl=d1cd b=10 ci=0 co=0 ca=0
49 4 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=369/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/4 ktl=e0e7 b=10 ci=0 co=0 ca=0
50 5 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=381/1/0 df=4 of=0 ri=1 ql=0 qs=.... kt=0/W/5 ktl=fb2f b=10 ci=0 co=0 ca=0
51 6 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=1037/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/6 ktl=d2ad b=10 ci=0 co=0 ca=0
52 7 c=1474 g=1474 pq=1 pqc=1473 qp=0 dt=1572/0/0 df=8 of=0 ri=1 ql=0 qs=.... kt=0/W/7 ktl=cf15 b=10 ci=0 co=0 ca=0
54 The first section lists the rcu_data structures for rcu_sched, the second
55 for rcu_bh. Note that CONFIG_TREE_PREEMPT_RCU kernels will have an
56 additional section for rcu_preempt. Each section has one line per CPU,
57 or eight for this 8-CPU system. The fields are as follows:
59 o The number at the beginning of each line is the CPU number.
60 CPUs numbers followed by an exclamation mark are offline,
61 but have been online at least once since boot. There will be
62 no output for CPUs that have never been online, which can be
63 a good thing in the surprisingly common case where NR_CPUS is
64 substantially larger than the number of actual CPUs.
66 o "c" is the count of grace periods that this CPU believes have
67 completed. Offlined CPUs and CPUs in dynticks idle mode may
68 lag quite a ways behind, for example, CPU 6 under "rcu_sched"
69 above, which has been offline through not quite 40,000 RCU grace
70 periods. It is not unusual to see CPUs lagging by thousands of
73 o "g" is the count of grace periods that this CPU believes have
74 started. Again, offlined CPUs and CPUs in dynticks idle mode
75 may lag behind. If the "c" and "g" values are equal, this CPU
76 has already reported a quiescent state for the last RCU grace
77 period that it is aware of, otherwise, the CPU believes that it
78 owes RCU a quiescent state.
80 o "pq" indicates that this CPU has passed through a quiescent state
81 for the current grace period. It is possible for "pq" to be
82 "1" and "c" different than "g", which indicates that although
83 the CPU has passed through a quiescent state, either (1) this
84 CPU has not yet reported that fact, (2) some other CPU has not
85 yet reported for this grace period, or (3) both.
87 o "pqc" indicates which grace period the last-observed quiescent
88 state for this CPU corresponds to. This is important for handling
89 the race between CPU 0 reporting an extended dynticks-idle
90 quiescent state for CPU 1 and CPU 1 suddenly waking up and
91 reporting its own quiescent state. If CPU 1 was the last CPU
92 for the current grace period, then the CPU that loses this race
93 will attempt to incorrectly mark CPU 1 as having checked in for
94 the next grace period!
96 o "qp" indicates that RCU still expects a quiescent state from
97 this CPU. Offlined CPUs and CPUs in dyntick idle mode might
98 well have qp=1, which is OK: RCU is still ignoring them.
100 o "dt" is the current value of the dyntick counter that is incremented
101 when entering or leaving dynticks idle state, either by the
102 scheduler or by irq. This number is even if the CPU is in
103 dyntick idle mode and odd otherwise. The number after the first
104 "/" is the interrupt nesting depth when in dyntick-idle state,
105 or one greater than the interrupt-nesting depth otherwise.
106 The number after the second "/" is the NMI nesting depth.
108 This field is displayed only for CONFIG_NO_HZ kernels.
110 o "df" is the number of times that some other CPU has forced a
111 quiescent state on behalf of this CPU due to this CPU being in
114 This field is displayed only for CONFIG_NO_HZ kernels.
116 o "of" is the number of times that some other CPU has forced a
117 quiescent state on behalf of this CPU due to this CPU being
118 offline. In a perfect world, this might never happen, but it
119 turns out that offlining and onlining a CPU can take several grace
120 periods, and so there is likely to be an extended period of time
121 when RCU believes that the CPU is online when it really is not.
122 Please note that erring in the other direction (RCU believing a
123 CPU is offline when it is really alive and kicking) is a fatal
124 error, so it makes sense to err conservatively.
126 o "ri" is the number of times that RCU has seen fit to send a
127 reschedule IPI to this CPU in order to get it to report a
130 o "ql" is the number of RCU callbacks currently residing on
131 this CPU. This is the total number of callbacks, regardless
132 of what state they are in (new, waiting for grace period to
133 start, waiting for grace period to end, ready to invoke).
135 o "qs" gives an indication of the state of the callback queue
136 with four characters:
138 "N" Indicates that there are callbacks queued that are not
139 ready to be handled by the next grace period, and thus
140 will be handled by the grace period following the next
143 "R" Indicates that there are callbacks queued that are
144 ready to be handled by the next grace period.
146 "W" Indicates that there are callbacks queued that are
147 waiting on the current grace period.
149 "D" Indicates that there are callbacks queued that have
150 already been handled by a prior grace period, and are
151 thus waiting to be invoked. Note that callbacks in
152 the process of being invoked are not counted here.
153 Callbacks in the process of being invoked are those
154 that have been removed from the rcu_data structures
155 queues by rcu_do_batch(), but which have not yet been
158 If there are no callbacks in a given one of the above states,
159 the corresponding character is replaced by ".".
161 o "kt" is the per-CPU kernel-thread state. The digit preceding
162 the first slash is zero if there is no work pending and 1
163 otherwise. The character between the first pair of slashes is
166 "S" The kernel thread is stopped, in other words, all
167 CPUs corresponding to this rcu_node structure are
170 "R" The kernel thread is running.
172 "W" The kernel thread is waiting because there is no work
175 "O" The kernel thread is waiting because it has been
176 forced off of its designated CPU or because its
177 ->cpus_allowed mask permits it to run on other than
180 "Y" The kernel thread is yielding to avoid hogging CPU.
182 "?" Unknown value, indicates a bug.
184 The number after the final slash is the CPU that the kthread
185 is actually running on.
187 o "ktl" is the low-order 16 bits (in hexadecimal) of the count of
188 the number of times that this CPU's per-CPU kthread has gone
189 through its loop servicing invoke_rcu_cpu_kthread() requests.
191 o "b" is the batch limit for this CPU. If more than this number
192 of RCU callbacks is ready to invoke, then the remainder will
195 o "ci" is the number of RCU callbacks that have been invoked for
196 this CPU. Note that ci+ql is the number of callbacks that have
197 been registered in absence of CPU-hotplug activity.
199 o "co" is the number of RCU callbacks that have been orphaned due to
200 this CPU going offline. These orphaned callbacks have been moved
201 to an arbitrarily chosen online CPU.
203 o "ca" is the number of RCU callbacks that have been adopted due to
204 other CPUs going offline. Note that ci+co-ca+ql is the number of
205 RCU callbacks registered on this CPU.
207 There is also an rcu/rcudata.csv file with the same information in
208 comma-separated-variable spreadsheet format.
211 The output of "cat rcu/rcugp" looks as follows:
213 rcu_sched: completed=33062 gpnum=33063
214 rcu_bh: completed=464 gpnum=464
216 Again, this output is for both "rcu_sched" and "rcu_bh". Note that
217 kernels built with CONFIG_TREE_PREEMPT_RCU will have an additional
218 "rcu_preempt" line. The fields are taken from the rcu_state structure,
221 o "completed" is the number of grace periods that have completed.
222 It is comparable to the "c" field from rcu/rcudata in that a
223 CPU whose "c" field matches the value of "completed" is aware
224 that the corresponding RCU grace period has completed.
226 o "gpnum" is the number of grace periods that have started. It is
227 comparable to the "g" field from rcu/rcudata in that a CPU
228 whose "g" field matches the value of "gpnum" is aware that the
229 corresponding RCU grace period has started.
231 If these two fields are equal (as they are for "rcu_bh" above),
232 then there is no grace period in progress, in other words, RCU
233 is idle. On the other hand, if the two fields differ (as they
234 do for "rcu_sched" above), then an RCU grace period is in progress.
237 The output of "cat rcu/rcuhier" looks as follows, with very long lines:
239 c=6902 g=6903 s=2 jfq=3 j=72c7 nfqs=13142/nfqsng=0(13142) fqlh=6
241 3/3 ..>. 0:35 ^0 0/0 ..>. 36:71 ^1 0/0 ..>. 72:107 ^2 0/0 ..>. 108:127 ^3
242 3/3f ..>. 0:5 ^0 2/3 ..>. 6:11 ^1 0/0 ..>. 12:17 ^2 0/0 ..>. 18:23 ^3 0/0 ..>. 24:29 ^4 0/0 ..>. 30:35 ^5 0/0 ..>. 36:41 ^0 0/0 ..>. 42:47 ^1 0/0 ..>. 48:53 ^2 0/0 ..>. 54:59 ^3 0/0 ..>. 60:65 ^4 0/0 ..>. 66:71 ^5 0/0 ..>. 72:77 ^0 0/0 ..>. 78:83 ^1 0/0 ..>. 84:89 ^2 0/0 ..>. 90:95 ^3 0/0 ..>. 96:101 ^4 0/0 ..>. 102:107 ^5 0/0 ..>. 108:113 ^0 0/0 ..>. 114:119 ^1 0/0 ..>. 120:125 ^2 0/0 ..>. 126:127 ^3
244 c=-226 g=-226 s=1 jfq=-5701 j=72c7 nfqs=88/nfqsng=0(88) fqlh=0
246 0/3 ..>. 0:35 ^0 0/0 ..>. 36:71 ^1 0/0 ..>. 72:107 ^2 0/0 ..>. 108:127 ^3
247 0/3f ..>. 0:5 ^0 0/3 ..>. 6:11 ^1 0/0 ..>. 12:17 ^2 0/0 ..>. 18:23 ^3 0/0 ..>. 24:29 ^4 0/0 ..>. 30:35 ^5 0/0 ..>. 36:41 ^0 0/0 ..>. 42:47 ^1 0/0 ..>. 48:53 ^2 0/0 ..>. 54:59 ^3 0/0 ..>. 60:65 ^4 0/0 ..>. 66:71 ^5 0/0 ..>. 72:77 ^0 0/0 ..>. 78:83 ^1 0/0 ..>. 84:89 ^2 0/0 ..>. 90:95 ^3 0/0 ..>. 96:101 ^4 0/0 ..>. 102:107 ^5 0/0 ..>. 108:113 ^0 0/0 ..>. 114:119 ^1 0/0 ..>. 120:125 ^2 0/0 ..>. 126:127 ^3
249 This is once again split into "rcu_sched" and "rcu_bh" portions,
250 and CONFIG_TREE_PREEMPT_RCU kernels will again have an additional
251 "rcu_preempt" section. The fields are as follows:
253 o "c" is exactly the same as "completed" under rcu/rcugp.
255 o "g" is exactly the same as "gpnum" under rcu/rcugp.
257 o "s" is the "signaled" state that drives force_quiescent_state()'s
260 o "jfq" is the number of jiffies remaining for this grace period
261 before force_quiescent_state() is invoked to help push things
262 along. Note that CPUs in dyntick-idle mode throughout the grace
263 period will not report on their own, but rather must be check by
264 some other CPU via force_quiescent_state().
266 o "j" is the low-order four hex digits of the jiffies counter.
267 Yes, Paul did run into a number of problems that turned out to
268 be due to the jiffies counter no longer counting. Why do you ask?
270 o "nfqs" is the number of calls to force_quiescent_state() since
273 o "nfqsng" is the number of useless calls to force_quiescent_state(),
274 where there wasn't actually a grace period active. This can
275 happen due to races. The number in parentheses is the difference
276 between "nfqs" and "nfqsng", or the number of times that
277 force_quiescent_state() actually did some real work.
279 o "fqlh" is the number of calls to force_quiescent_state() that
280 exited immediately (without even being counted in nfqs above)
281 due to contention on ->fqslock.
283 o Each element of the form "1/1 0:127 ^0" represents one struct
284 rcu_node. Each line represents one level of the hierarchy, from
285 root to leaves. It is best to think of the rcu_data structures
286 as forming yet another level after the leaves. Note that there
287 might be either one, two, or three levels of rcu_node structures,
288 depending on the relationship between CONFIG_RCU_FANOUT and
291 o The numbers separated by the "/" are the qsmask followed
292 by the qsmaskinit. The qsmask will have one bit
293 set for each entity in the next lower level that
294 has not yet checked in for the current grace period.
295 The qsmaskinit will have one bit for each entity that is
296 currently expected to check in during each grace period.
297 The value of qsmaskinit is assigned to that of qsmask
298 at the beginning of each grace period.
300 For example, for "rcu_sched", the qsmask of the first
301 entry of the lowest level is 0x14, meaning that we
302 are still waiting for CPUs 2 and 4 to check in for the
303 current grace period.
305 o The characters separated by the ">" indicate the state
306 of the blocked-tasks lists. A "G" preceding the ">"
307 indicates that at least one task blocked in an RCU
308 read-side critical section blocks the current grace
309 period, while a "E" preceding the ">" indicates that
310 at least one task blocked in an RCU read-side critical
311 section blocks the current expedited grace period.
312 A "T" character following the ">" indicates that at
313 least one task is blocked within an RCU read-side
314 critical section, regardless of whether any current
315 grace period (expedited or normal) is inconvenienced.
316 A "." character appears if the corresponding condition
317 does not hold, so that "..>." indicates that no tasks
318 are blocked. In contrast, "GE>T" indicates maximal
319 inconvenience from blocked tasks.
321 o The numbers separated by the ":" are the range of CPUs
322 served by this struct rcu_node. This can be helpful
323 in working out how the hierarchy is wired together.
325 For example, the first entry at the lowest level shows
326 "0:5", indicating that it covers CPUs 0 through 5.
328 o The number after the "^" indicates the bit in the
329 next higher level rcu_node structure that this
330 rcu_node structure corresponds to.
332 For example, the first entry at the lowest level shows
333 "^0", indicating that it corresponds to bit zero in
334 the first entry at the middle level.
337 The output of "cat rcu/rcu_pending" looks as follows:
340 0 np=255892 qsp=53936 rpq=85 cbr=0 cng=14417 gpc=10033 gps=24320 nf=6445 nn=146741
341 1 np=261224 qsp=54638 rpq=33 cbr=0 cng=25723 gpc=16310 gps=2849 nf=5912 nn=155792
342 2 np=237496 qsp=49664 rpq=23 cbr=0 cng=2762 gpc=45478 gps=1762 nf=1201 nn=136629
343 3 np=236249 qsp=48766 rpq=98 cbr=0 cng=286 gpc=48049 gps=1218 nf=207 nn=137723
344 4 np=221310 qsp=46850 rpq=7 cbr=0 cng=26 gpc=43161 gps=4634 nf=3529 nn=123110
345 5 np=237332 qsp=48449 rpq=9 cbr=0 cng=54 gpc=47920 gps=3252 nf=201 nn=137456
346 6 np=219995 qsp=46718 rpq=12 cbr=0 cng=50 gpc=42098 gps=6093 nf=4202 nn=120834
347 7 np=249893 qsp=49390 rpq=42 cbr=0 cng=72 gpc=38400 gps=17102 nf=41 nn=144888
349 0 np=146741 qsp=1419 rpq=6 cbr=0 cng=6 gpc=0 gps=0 nf=2 nn=145314
350 1 np=155792 qsp=12597 rpq=3 cbr=0 cng=0 gpc=4 gps=8 nf=3 nn=143180
351 2 np=136629 qsp=18680 rpq=1 cbr=0 cng=0 gpc=7 gps=6 nf=0 nn=117936
352 3 np=137723 qsp=2843 rpq=0 cbr=0 cng=0 gpc=10 gps=7 nf=0 nn=134863
353 4 np=123110 qsp=12433 rpq=0 cbr=0 cng=0 gpc=4 gps=2 nf=0 nn=110671
354 5 np=137456 qsp=4210 rpq=1 cbr=0 cng=0 gpc=6 gps=5 nf=0 nn=133235
355 6 np=120834 qsp=9902 rpq=2 cbr=0 cng=0 gpc=6 gps=3 nf=2 nn=110921
356 7 np=144888 qsp=26336 rpq=0 cbr=0 cng=0 gpc=8 gps=2 nf=0 nn=118542
358 As always, this is once again split into "rcu_sched" and "rcu_bh"
359 portions, with CONFIG_TREE_PREEMPT_RCU kernels having an additional
360 "rcu_preempt" section. The fields are as follows:
362 o "np" is the number of times that __rcu_pending() has been invoked
363 for the corresponding flavor of RCU.
365 o "qsp" is the number of times that the RCU was waiting for a
366 quiescent state from this CPU.
368 o "rpq" is the number of times that the CPU had passed through
369 a quiescent state, but not yet reported it to RCU.
371 o "cbr" is the number of times that this CPU had RCU callbacks
372 that had passed through a grace period, and were thus ready
375 o "cng" is the number of times that this CPU needed another
376 grace period while RCU was idle.
378 o "gpc" is the number of times that an old grace period had
379 completed, but this CPU was not yet aware of it.
381 o "gps" is the number of times that a new grace period had started,
382 but this CPU was not yet aware of it.
384 o "nf" is the number of times that this CPU suspected that the
385 current grace period had run for too long, and thus needed to
388 Please note that "forcing" consists of sending resched IPIs
389 to holdout CPUs. If that CPU really still is in an old RCU
390 read-side critical section, then we really do have to wait for it.
391 The assumption behing "forcing" is that the CPU is not still in
392 an old RCU read-side critical section, but has not yet responded
393 for some other reason.
395 o "nn" is the number of times that this CPU needed nothing. Alert
396 readers will note that the rcu "nn" number for a given CPU very
397 closely matches the rcu_bh "np" number for that same CPU. This
398 is due to short-circuit evaluation in rcu_pending().
401 The output of "cat rcu/rcutorture" looks as follows:
403 rcutorture test sequence: 0 (test in progress)
404 rcutorture update version number: 615
406 The first line shows the number of rcutorture tests that have completed
407 since boot. If a test is currently running, the "(test in progress)"
408 string will appear as shown above. The second line shows the number of
409 update cycles that the current test has started, or zero if there is
413 The output of "cat rcu/rcuboost" looks as follows:
415 0:5 tasks=.... kt=W ntb=0 neb=0 nnb=0 j=2f95 bt=300f
416 balk: nt=0 egt=989 bt=0 nb=0 ny=0 nos=16
417 6:7 tasks=.... kt=W ntb=0 neb=0 nnb=0 j=2f95 bt=300f
418 balk: nt=0 egt=225 bt=0 nb=0 ny=0 nos=6
420 This information is output only for rcu_preempt. Each two-line entry
421 corresponds to a leaf rcu_node strcuture. The fields are as follows:
423 o "n:m" is the CPU-number range for the corresponding two-line
424 entry. In the sample output above, the first entry covers
425 CPUs zero through five and the second entry covers CPUs 6
428 o "tasks=TNEB" gives the state of the various segments of the
429 rnp->blocked_tasks list:
431 "T" This indicates that there are some tasks that blocked
432 while running on one of the corresponding CPUs while
433 in an RCU read-side critical section.
435 "N" This indicates that some of the blocked tasks are preventing
436 the current normal (non-expedited) grace period from
439 "E" This indicates that some of the blocked tasks are preventing
440 the current expedited grace period from completing.
442 "B" This indicates that some of the blocked tasks are in
443 need of RCU priority boosting.
445 Each character is replaced with "." if the corresponding
446 condition does not hold.
448 o "kt" is the state of the RCU priority-boosting kernel
449 thread associated with the corresponding rcu_node structure.
450 The state can be one of the following:
452 "S" The kernel thread is stopped, in other words, all
453 CPUs corresponding to this rcu_node structure are
456 "R" The kernel thread is running.
458 "W" The kernel thread is waiting because there is no work
461 "Y" The kernel thread is yielding to avoid hogging CPU.
463 "?" Unknown value, indicates a bug.
465 o "ntb" is the number of tasks boosted.
467 o "neb" is the number of tasks boosted in order to complete an
468 expedited grace period.
470 o "nnb" is the number of tasks boosted in order to complete a
471 normal (non-expedited) grace period. When boosting a task
472 that was blocking both an expedited and a normal grace period,
473 it is counted against the expedited total above.
475 o "j" is the low-order 16 bits of the jiffies counter in
478 o "bt" is the low-order 16 bits of the value that the jiffies
479 counter will have when we next start boosting, assuming that
480 the current grace period does not end beforehand. This is
483 o "balk: nt" counts the number of times we didn't boost (in
484 other words, we balked) even though it was time to boost because
485 there were no blocked tasks to boost. This situation occurs
486 when there is one blocked task on one rcu_node structure and
487 none on some other rcu_node structure.
489 o "egt" counts the number of times we balked because although
490 there were blocked tasks, none of them were blocking the
491 current grace period, whether expedited or otherwise.
493 o "bt" counts the number of times we balked because boosting
494 had already been initiated for the current grace period.
496 o "nb" counts the number of times we balked because there
497 was at least one task blocking the current non-expedited grace
498 period that never had blocked. If it is already running, it
499 just won't help to boost its priority!
501 o "ny" counts the number of times we balked because it was
502 not yet time to start boosting.
504 o "nos" counts the number of times we balked for other
505 reasons, e.g., the grace period ended first.
508 CONFIG_TINY_RCU and CONFIG_TINY_PREEMPT_RCU debugfs Files and Formats
510 These implementations of RCU provides a single debugfs file under the
511 top-level directory RCU, namely rcu/rcudata, which displays fields in
512 rcu_bh_ctrlblk, rcu_sched_ctrlblk and, for CONFIG_TINY_PREEMPT_RCU,
515 The output of "cat rcu/rcudata" is as follows:
517 rcu_preempt: qlen=24 gp=1097669 g197/p197/c197 tasks=...
518 ttb=. btg=no ntb=184 neb=0 nnb=183 j=01f7 bt=0274
519 normal balk: nt=1097669 gt=0 bt=371 b=0 ny=25073378 nos=0
524 This is split into rcu_preempt, rcu_sched, and rcu_bh sections, with the
525 rcu_preempt section appearing only in CONFIG_TINY_PREEMPT_RCU builds.
526 The last three lines of the rcu_preempt section appear only in
527 CONFIG_RCU_BOOST kernel builds. The fields are as follows:
529 o "qlen" is the number of RCU callbacks currently waiting either
530 for an RCU grace period or waiting to be invoked. This is the
531 only field present for rcu_sched and rcu_bh, due to the
532 short-circuiting of grace period in those two cases.
534 o "gp" is the number of grace periods that have completed.
536 o "g197/p197/c197" displays the grace-period state, with the
537 "g" number being the number of grace periods that have started
538 (mod 256), the "p" number being the number of grace periods
539 that the CPU has responded to (also mod 256), and the "c"
540 number being the number of grace periods that have completed
541 (once again mode 256).
543 Why have both "gp" and "g"? Because the data flowing into
544 "gp" is only present in a CONFIG_RCU_TRACE kernel.
546 o "tasks" is a set of bits. The first bit is "T" if there are
547 currently tasks that have recently blocked within an RCU
548 read-side critical section, the second bit is "N" if any of the
549 aforementioned tasks are blocking the current RCU grace period,
550 and the third bit is "E" if any of the aforementioned tasks are
551 blocking the current expedited grace period. Each bit is "."
552 if the corresponding condition does not hold.
554 o "ttb" is a single bit. It is "B" if any of the blocked tasks
555 need to be priority boosted and "." otherwise.
557 o "btg" indicates whether boosting has been carried out during
558 the current grace period, with "exp" indicating that boosting
559 is in progress for an expedited grace period, "no" indicating
560 that boosting has not yet started for a normal grace period,
561 "begun" indicating that boosting has bebug for a normal grace
562 period, and "done" indicating that boosting has completed for
563 a normal grace period.
565 o "ntb" is the total number of tasks subjected to RCU priority boosting
568 o "neb" is the number of expedited grace periods that have had
569 to resort to RCU priority boosting since boot.
571 o "nnb" is the number of normal grace periods that have had
572 to resort to RCU priority boosting since boot.
574 o "j" is the low-order 16 bits of the jiffies counter in hexadecimal.
576 o "bt" is the low-order 16 bits of the value that the jiffies counter
577 will have at the next time that boosting is scheduled to begin.
579 o In the line beginning with "normal balk", the fields are as follows:
581 o "nt" is the number of times that the system balked from
582 boosting because there were no blocked tasks to boost.
583 Note that the system will balk from boosting even if the
584 grace period is overdue when the currently running task
585 is looping within an RCU read-side critical section.
586 There is no point in boosting in this case, because
587 boosting a running task won't make it run any faster.
589 o "gt" is the number of times that the system balked
590 from boosting because, although there were blocked tasks,
591 none of them were preventing the current grace period
594 o "bt" is the number of times that the system balked
595 from boosting because boosting was already in progress.
597 o "b" is the number of times that the system balked from
598 boosting because boosting had already completed for
599 the grace period in question.
601 o "ny" is the number of times that the system balked from
602 boosting because it was not yet time to start boosting
603 the grace period in question.
605 o "nos" is the number of times that the system balked from
606 boosting for inexplicable ("not otherwise specified")
607 reasons. This can actually happen due to races involving
608 increments of the jiffies counter.
610 o In the line beginning with "exp balk", the fields are as follows:
612 o "bt" is the number of times that the system balked from
613 boosting because there were no blocked tasks to boost.
615 o "nos" is the number of times that the system balked from
616 boosting for inexplicable ("not otherwise specified")