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 provide several debugfs directories under the
14 top-level directory "rcu":
20 Each directory contains files for the corresponding flavor of RCU.
21 Note that rcu/rcu_preempt is only present for CONFIG_TREE_PREEMPT_RCU.
22 For CONFIG_TREE_RCU, the RCU flavor maps onto the RCU-sched flavor,
23 so that activity for both appears in rcu/rcu_sched.
25 In addition, the following file appears in the top-level directory:
26 rcu/rcutorture. This file displays rcutorture test progress. The output
27 of "cat rcu/rcutorture" looks as follows:
29 rcutorture test sequence: 0 (test in progress)
30 rcutorture update version number: 615
32 The first line shows the number of rcutorture tests that have completed
33 since boot. If a test is currently running, the "(test in progress)"
34 string will appear as shown above. The second line shows the number of
35 update cycles that the current test has started, or zero if there is
39 Within each flavor directory (rcu/rcu_bh, rcu/rcu_sched, and possibly
40 also rcu/rcu_preempt) the following files will be present:
43 Displays fields in struct rcu_data.
45 Displays grace-period counters.
47 Displays the struct rcu_node hierarchy.
49 Displays counts of the reasons rcu_pending() decided that RCU had
52 Displays RCU boosting statistics. Only present if
55 The output of "cat rcu/rcu_preempt/rcudata" looks as follows:
57 0!c=30455 g=30456 pq=1 qp=1 dt=126535/140000000000000/0 df=2002 of=4 ql=0/0 qs=N... b=10 ci=74572 nci=0 co=1131 ca=716
58 1!c=30719 g=30720 pq=1 qp=0 dt=132007/140000000000000/0 df=1874 of=10 ql=0/0 qs=N... b=10 ci=123209 nci=0 co=685 ca=982
59 2!c=30150 g=30151 pq=1 qp=1 dt=138537/140000000000000/0 df=1707 of=8 ql=0/0 qs=N... b=10 ci=80132 nci=0 co=1328 ca=1458
60 3 c=31249 g=31250 pq=1 qp=0 dt=107255/140000000000000/0 df=1749 of=6 ql=0/450 qs=NRW. b=10 ci=151700 nci=0 co=509 ca=622
61 4!c=29502 g=29503 pq=1 qp=1 dt=83647/140000000000000/0 df=965 of=5 ql=0/0 qs=N... b=10 ci=65643 nci=0 co=1373 ca=1521
62 5 c=31201 g=31202 pq=1 qp=1 dt=70422/0/0 df=535 of=7 ql=0/0 qs=.... b=10 ci=58500 nci=0 co=764 ca=698
63 6!c=30253 g=30254 pq=1 qp=1 dt=95363/140000000000000/0 df=780 of=5 ql=0/0 qs=N... b=10 ci=100607 nci=0 co=1414 ca=1353
64 7 c=31178 g=31178 pq=1 qp=0 dt=91536/0/0 df=547 of=4 ql=0/0 qs=.... b=10 ci=109819 nci=0 co=1115 ca=969
66 This file has one line per CPU, or eight for this 8-CPU system.
67 The fields are as follows:
69 o The number at the beginning of each line is the CPU number.
70 CPUs numbers followed by an exclamation mark are offline,
71 but have been online at least once since boot. There will be
72 no output for CPUs that have never been online, which can be
73 a good thing in the surprisingly common case where NR_CPUS is
74 substantially larger than the number of actual CPUs.
76 o "c" is the count of grace periods that this CPU believes have
77 completed. Offlined CPUs and CPUs in dynticks idle mode may lag
78 quite a ways behind, for example, CPU 4 under "rcu_sched" above,
79 which has been offline through 16 RCU grace periods. It is not
80 unusual to see offline CPUs lagging by thousands of grace periods.
81 Note that although the grace-period number is an unsigned long,
82 it is printed out as a signed long to allow more human-friendly
83 representation near boot time.
85 o "g" is the count of grace periods that this CPU believes have
86 started. Again, offlined CPUs and CPUs in dynticks idle mode
87 may lag behind. If the "c" and "g" values are equal, this CPU
88 has already reported a quiescent state for the last RCU grace
89 period that it is aware of, otherwise, the CPU believes that it
90 owes RCU a quiescent state.
92 o "pq" indicates that this CPU has passed through a quiescent state
93 for the current grace period. It is possible for "pq" to be
94 "1" and "c" different than "g", which indicates that although
95 the CPU has passed through a quiescent state, either (1) this
96 CPU has not yet reported that fact, (2) some other CPU has not
97 yet reported for this grace period, or (3) both.
99 o "qp" indicates that RCU still expects a quiescent state from
100 this CPU. Offlined CPUs and CPUs in dyntick idle mode might
101 well have qp=1, which is OK: RCU is still ignoring them.
103 o "dt" is the current value of the dyntick counter that is incremented
104 when entering or leaving idle, either due to a context switch or
105 due to an interrupt. This number is even if the CPU is in idle
106 from RCU's viewpoint and odd otherwise. The number after the
107 first "/" is the interrupt nesting depth when in idle state,
108 or a large number added to the interrupt-nesting depth when
109 running a non-idle task. Some architectures do not accurately
110 count interrupt nesting when running in non-idle kernel context,
111 which can result in interesting anomalies such as negative
112 interrupt-nesting levels. The number after the second "/"
113 is the NMI nesting depth.
115 o "df" is the number of times that some other CPU has forced a
116 quiescent state on behalf of this CPU due to this CPU being in
119 o "of" is the number of times that some other CPU has forced a
120 quiescent state on behalf of this CPU due to this CPU being
121 offline. In a perfect world, this might never happen, but it
122 turns out that offlining and onlining a CPU can take several grace
123 periods, and so there is likely to be an extended period of time
124 when RCU believes that the CPU is online when it really is not.
125 Please note that erring in the other direction (RCU believing a
126 CPU is offline when it is really alive and kicking) is a fatal
127 error, so it makes sense to err conservatively.
129 o "ql" is the number of RCU callbacks currently residing on
130 this CPU. The first number is the number of "lazy" callbacks
131 that are known to RCU to only be freeing memory, and the number
132 after the "/" is the total number of callbacks, lazy or not.
133 These counters count callbacks regardless of what phase of
134 grace-period processing that they are in (new, waiting for
135 grace period to start, waiting for grace period to end, ready
138 o "qs" gives an indication of the state of the callback queue
139 with four characters:
141 "N" Indicates that there are callbacks queued that are not
142 ready to be handled by the next grace period, and thus
143 will be handled by the grace period following the next
146 "R" Indicates that there are callbacks queued that are
147 ready to be handled by the next grace period.
149 "W" Indicates that there are callbacks queued that are
150 waiting on the current grace period.
152 "D" Indicates that there are callbacks queued that have
153 already been handled by a prior grace period, and are
154 thus waiting to be invoked. Note that callbacks in
155 the process of being invoked are not counted here.
156 Callbacks in the process of being invoked are those
157 that have been removed from the rcu_data structures
158 queues by rcu_do_batch(), but which have not yet been
161 If there are no callbacks in a given one of the above states,
162 the corresponding character is replaced by ".".
164 o "b" is the batch limit for this CPU. If more than this number
165 of RCU callbacks is ready to invoke, then the remainder will
168 o "ci" is the number of RCU callbacks that have been invoked for
169 this CPU. Note that ci+nci+ql is the number of callbacks that have
170 been registered in absence of CPU-hotplug activity.
172 o "nci" is the number of RCU callbacks that have been offloaded from
173 this CPU. This will always be zero unless the kernel was built
174 with CONFIG_RCU_NOCB_CPU=y and the "rcu_nocbs=" kernel boot
175 parameter was specified.
177 o "co" is the number of RCU callbacks that have been orphaned due to
178 this CPU going offline. These orphaned callbacks have been moved
179 to an arbitrarily chosen online CPU.
181 o "ca" is the number of RCU callbacks that have been adopted by this
182 CPU due to other CPUs going offline. Note that ci+co-ca+ql is
183 the number of RCU callbacks registered on this CPU.
186 Kernels compiled with CONFIG_RCU_BOOST=y display the following from
187 /debug/rcu/rcu_preempt/rcudata:
189 0!c=12865 g=12866 pq=1 qp=1 dt=83113/140000000000000/0 df=288 of=11 ql=0/0 qs=N... kt=0/O ktl=944 b=10 ci=60709 nci=0 co=748 ca=871
190 1 c=14407 g=14408 pq=1 qp=0 dt=100679/140000000000000/0 df=378 of=7 ql=0/119 qs=NRW. kt=0/W ktl=9b6 b=10 ci=109740 nci=0 co=589 ca=485
191 2 c=14407 g=14408 pq=1 qp=0 dt=105486/0/0 df=90 of=9 ql=0/89 qs=NRW. kt=0/W ktl=c0c b=10 ci=83113 nci=0 co=533 ca=490
192 3 c=14407 g=14408 pq=1 qp=0 dt=107138/0/0 df=142 of=8 ql=0/188 qs=NRW. kt=0/W ktl=b96 b=10 ci=121114 nci=0 co=426 ca=290
193 4 c=14405 g=14406 pq=1 qp=1 dt=50238/0/0 df=706 of=7 ql=0/0 qs=.... kt=0/W ktl=812 b=10 ci=34929 nci=0 co=643 ca=114
194 5!c=14168 g=14169 pq=1 qp=0 dt=45465/140000000000000/0 df=161 of=11 ql=0/0 qs=N... kt=0/O ktl=b4d b=10 ci=47712 nci=0 co=677 ca=722
195 6 c=14404 g=14405 pq=1 qp=0 dt=59454/0/0 df=94 of=6 ql=0/0 qs=.... kt=0/W ktl=e57 b=10 ci=55597 nci=0 co=701 ca=811
196 7 c=14407 g=14408 pq=1 qp=1 dt=68850/0/0 df=31 of=8 ql=0/0 qs=.... kt=0/W ktl=14bd b=10 ci=77475 nci=0 co=508 ca=1042
198 This is similar to the output discussed above, but contains the following
201 o "kt" is the per-CPU kernel-thread state. The digit preceding
202 the first slash is zero if there is no work pending and 1
203 otherwise. The character between the first pair of slashes is
206 "S" The kernel thread is stopped, in other words, all
207 CPUs corresponding to this rcu_node structure are
210 "R" The kernel thread is running.
212 "W" The kernel thread is waiting because there is no work
215 "O" The kernel thread is waiting because it has been
216 forced off of its designated CPU or because its
217 ->cpus_allowed mask permits it to run on other than
220 "Y" The kernel thread is yielding to avoid hogging CPU.
222 "?" Unknown value, indicates a bug.
224 The number after the final slash is the CPU that the kthread
225 is actually running on.
227 This field is displayed only for CONFIG_RCU_BOOST kernels.
229 o "ktl" is the low-order 16 bits (in hexadecimal) of the count of
230 the number of times that this CPU's per-CPU kthread has gone
231 through its loop servicing invoke_rcu_cpu_kthread() requests.
233 This field is displayed only for CONFIG_RCU_BOOST kernels.
236 The output of "cat rcu/rcu_preempt/rcugp" looks as follows:
238 completed=31249 gpnum=31250 age=1 max=18
240 These fields are taken from the rcu_state structure, and are as follows:
242 o "completed" is the number of grace periods that have completed.
243 It is comparable to the "c" field from rcu/rcudata in that a
244 CPU whose "c" field matches the value of "completed" is aware
245 that the corresponding RCU grace period has completed.
247 o "gpnum" is the number of grace periods that have started. It is
248 similarly comparable to the "g" field from rcu/rcudata in that
249 a CPU whose "g" field matches the value of "gpnum" is aware that
250 the corresponding RCU grace period has started.
252 If these two fields are equal, then there is no grace period
253 in progress, in other words, RCU is idle. On the other hand,
254 if the two fields differ (as they are above), then an RCU grace
255 period is in progress.
257 o "age" is the number of jiffies that the current grace period
258 has extended for, or zero if there is no grace period currently
261 o "max" is the age in jiffies of the longest-duration grace period
264 The output of "cat rcu/rcu_preempt/rcuhier" looks as follows:
266 c=14407 g=14408 s=0 jfq=2 j=c863 nfqs=12040/nfqsng=0(12040) fqlh=1051 oqlen=0/0
268 e/e ..>. 0:3 ^0 d/d ..>. 4:7 ^1
270 The fields are as follows:
272 o "c" is exactly the same as "completed" under rcu/rcu_preempt/rcugp.
274 o "g" is exactly the same as "gpnum" under rcu/rcu_preempt/rcugp.
276 o "s" is the current state of the force_quiescent_state()
279 o "jfq" is the number of jiffies remaining for this grace period
280 before force_quiescent_state() is invoked to help push things
281 along. Note that CPUs in idle mode throughout the grace period
282 will not report on their own, but rather must be check by some
283 other CPU via force_quiescent_state().
285 o "j" is the low-order four hex digits of the jiffies counter.
286 Yes, Paul did run into a number of problems that turned out to
287 be due to the jiffies counter no longer counting. Why do you ask?
289 o "nfqs" is the number of calls to force_quiescent_state() since
292 o "nfqsng" is the number of useless calls to force_quiescent_state(),
293 where there wasn't actually a grace period active. This can
294 no longer happen due to grace-period processing being pushed
295 into a kthread. The number in parentheses is the difference
296 between "nfqs" and "nfqsng", or the number of times that
297 force_quiescent_state() actually did some real work.
299 o "fqlh" is the number of calls to force_quiescent_state() that
300 exited immediately (without even being counted in nfqs above)
301 due to contention on ->fqslock.
303 o Each element of the form "3/3 ..>. 0:7 ^0" represents one rcu_node
304 structure. Each line represents one level of the hierarchy,
305 from root to leaves. It is best to think of the rcu_data
306 structures as forming yet another level after the leaves.
307 Note that there might be either one, two, three, or even four
308 levels of rcu_node structures, depending on the relationship
309 between CONFIG_RCU_FANOUT, CONFIG_RCU_FANOUT_LEAF (possibly
310 adjusted using the rcu_fanout_leaf kernel boot parameter), and
311 CONFIG_NR_CPUS (possibly adjusted using the nr_cpu_ids count of
312 possible CPUs for the booting hardware).
314 o The numbers separated by the "/" are the qsmask followed
315 by the qsmaskinit. The qsmask will have one bit
316 set for each entity in the next lower level that has
317 not yet checked in for the current grace period ("e"
318 indicating CPUs 5, 6, and 7 in the example above).
319 The qsmaskinit will have one bit for each entity that is
320 currently expected to check in during each grace period.
321 The value of qsmaskinit is assigned to that of qsmask
322 at the beginning of each grace period.
324 o The characters separated by the ">" indicate the state
325 of the blocked-tasks lists. A "G" preceding the ">"
326 indicates that at least one task blocked in an RCU
327 read-side critical section blocks the current grace
328 period, while a "E" preceding the ">" indicates that
329 at least one task blocked in an RCU read-side critical
330 section blocks the current expedited grace period.
331 A "T" character following the ">" indicates that at
332 least one task is blocked within an RCU read-side
333 critical section, regardless of whether any current
334 grace period (expedited or normal) is inconvenienced.
335 A "." character appears if the corresponding condition
336 does not hold, so that "..>." indicates that no tasks
337 are blocked. In contrast, "GE>T" indicates maximal
338 inconvenience from blocked tasks. CONFIG_TREE_RCU
339 builds of the kernel will always show "..>.".
341 o The numbers separated by the ":" are the range of CPUs
342 served by this struct rcu_node. This can be helpful
343 in working out how the hierarchy is wired together.
345 For example, the example rcu_node structure shown above
346 has "0:7", indicating that it covers CPUs 0 through 7.
348 o The number after the "^" indicates the bit in the
349 next higher level rcu_node structure that this rcu_node
350 structure corresponds to. For example, the "d/d ..>. 4:7
351 ^1" has a "1" in this position, indicating that it
352 corresponds to the "1" bit in the "3" shown in the
353 "3/3 ..>. 0:7 ^0" entry on the next level up.
356 The output of "cat rcu/rcu_sched/rcu_pending" looks as follows:
358 0!np=26111 qsp=29 rpq=5386 cbr=1 cng=570 gpc=3674 gps=577 nn=15903
359 1!np=28913 qsp=35 rpq=6097 cbr=1 cng=448 gpc=3700 gps=554 nn=18113
360 2!np=32740 qsp=37 rpq=6202 cbr=0 cng=476 gpc=4627 gps=546 nn=20889
361 3 np=23679 qsp=22 rpq=5044 cbr=1 cng=415 gpc=3403 gps=347 nn=14469
362 4!np=30714 qsp=4 rpq=5574 cbr=0 cng=528 gpc=3931 gps=639 nn=20042
363 5 np=28910 qsp=2 rpq=5246 cbr=0 cng=428 gpc=4105 gps=709 nn=18422
364 6!np=38648 qsp=5 rpq=7076 cbr=0 cng=840 gpc=4072 gps=961 nn=25699
365 7 np=37275 qsp=2 rpq=6873 cbr=0 cng=868 gpc=3416 gps=971 nn=25147
367 The fields are as follows:
369 o The leading number is the CPU number, with "!" indicating
372 o "np" is the number of times that __rcu_pending() has been invoked
373 for the corresponding flavor of RCU.
375 o "qsp" is the number of times that the RCU was waiting for a
376 quiescent state from this CPU.
378 o "rpq" is the number of times that the CPU had passed through
379 a quiescent state, but not yet reported it to RCU.
381 o "cbr" is the number of times that this CPU had RCU callbacks
382 that had passed through a grace period, and were thus ready
385 o "cng" is the number of times that this CPU needed another
386 grace period while RCU was idle.
388 o "gpc" is the number of times that an old grace period had
389 completed, but this CPU was not yet aware of it.
391 o "gps" is the number of times that a new grace period had started,
392 but this CPU was not yet aware of it.
394 o "nn" is the number of times that this CPU needed nothing.
397 The output of "cat rcu/rcuboost" looks as follows:
399 0:3 tasks=.... kt=W ntb=0 neb=0 nnb=0 j=c864 bt=c894
400 balk: nt=0 egt=4695 bt=0 nb=0 ny=56 nos=0
401 4:7 tasks=.... kt=W ntb=0 neb=0 nnb=0 j=c864 bt=c894
402 balk: nt=0 egt=6541 bt=0 nb=0 ny=126 nos=0
404 This information is output only for rcu_preempt. Each two-line entry
405 corresponds to a leaf rcu_node strcuture. The fields are as follows:
407 o "n:m" is the CPU-number range for the corresponding two-line
408 entry. In the sample output above, the first entry covers
409 CPUs zero through three and the second entry covers CPUs four
412 o "tasks=TNEB" gives the state of the various segments of the
413 rnp->blocked_tasks list:
415 "T" This indicates that there are some tasks that blocked
416 while running on one of the corresponding CPUs while
417 in an RCU read-side critical section.
419 "N" This indicates that some of the blocked tasks are preventing
420 the current normal (non-expedited) grace period from
423 "E" This indicates that some of the blocked tasks are preventing
424 the current expedited grace period from completing.
426 "B" This indicates that some of the blocked tasks are in
427 need of RCU priority boosting.
429 Each character is replaced with "." if the corresponding
430 condition does not hold.
432 o "kt" is the state of the RCU priority-boosting kernel
433 thread associated with the corresponding rcu_node structure.
434 The state can be one of the following:
436 "S" The kernel thread is stopped, in other words, all
437 CPUs corresponding to this rcu_node structure are
440 "R" The kernel thread is running.
442 "W" The kernel thread is waiting because there is no work
445 "Y" The kernel thread is yielding to avoid hogging CPU.
447 "?" Unknown value, indicates a bug.
449 o "ntb" is the number of tasks boosted.
451 o "neb" is the number of tasks boosted in order to complete an
452 expedited grace period.
454 o "nnb" is the number of tasks boosted in order to complete a
455 normal (non-expedited) grace period. When boosting a task
456 that was blocking both an expedited and a normal grace period,
457 it is counted against the expedited total above.
459 o "j" is the low-order 16 bits of the jiffies counter in
462 o "bt" is the low-order 16 bits of the value that the jiffies
463 counter will have when we next start boosting, assuming that
464 the current grace period does not end beforehand. This is
467 o "balk: nt" counts the number of times we didn't boost (in
468 other words, we balked) even though it was time to boost because
469 there were no blocked tasks to boost. This situation occurs
470 when there is one blocked task on one rcu_node structure and
471 none on some other rcu_node structure.
473 o "egt" counts the number of times we balked because although
474 there were blocked tasks, none of them were blocking the
475 current grace period, whether expedited or otherwise.
477 o "bt" counts the number of times we balked because boosting
478 had already been initiated for the current grace period.
480 o "nb" counts the number of times we balked because there
481 was at least one task blocking the current non-expedited grace
482 period that never had blocked. If it is already running, it
483 just won't help to boost its priority!
485 o "ny" counts the number of times we balked because it was
486 not yet time to start boosting.
488 o "nos" counts the number of times we balked for other
489 reasons, e.g., the grace period ended first.
492 CONFIG_TINY_RCU and CONFIG_TINY_PREEMPT_RCU debugfs Files and Formats
494 These implementations of RCU provides a single debugfs file under the
495 top-level directory RCU, namely rcu/rcudata, which displays fields in
496 rcu_bh_ctrlblk, rcu_sched_ctrlblk and, for CONFIG_TINY_PREEMPT_RCU,
499 The output of "cat rcu/rcudata" is as follows:
501 rcu_preempt: qlen=24 gp=1097669 g197/p197/c197 tasks=...
502 ttb=. btg=no ntb=184 neb=0 nnb=183 j=01f7 bt=0274
503 normal balk: nt=1097669 gt=0 bt=371 b=0 ny=25073378 nos=0
508 This is split into rcu_preempt, rcu_sched, and rcu_bh sections, with the
509 rcu_preempt section appearing only in CONFIG_TINY_PREEMPT_RCU builds.
510 The last three lines of the rcu_preempt section appear only in
511 CONFIG_RCU_BOOST kernel builds. The fields are as follows:
513 o "qlen" is the number of RCU callbacks currently waiting either
514 for an RCU grace period or waiting to be invoked. This is the
515 only field present for rcu_sched and rcu_bh, due to the
516 short-circuiting of grace period in those two cases.
518 o "gp" is the number of grace periods that have completed.
520 o "g197/p197/c197" displays the grace-period state, with the
521 "g" number being the number of grace periods that have started
522 (mod 256), the "p" number being the number of grace periods
523 that the CPU has responded to (also mod 256), and the "c"
524 number being the number of grace periods that have completed
525 (once again mode 256).
527 Why have both "gp" and "g"? Because the data flowing into
528 "gp" is only present in a CONFIG_RCU_TRACE kernel.
530 o "tasks" is a set of bits. The first bit is "T" if there are
531 currently tasks that have recently blocked within an RCU
532 read-side critical section, the second bit is "N" if any of the
533 aforementioned tasks are blocking the current RCU grace period,
534 and the third bit is "E" if any of the aforementioned tasks are
535 blocking the current expedited grace period. Each bit is "."
536 if the corresponding condition does not hold.
538 o "ttb" is a single bit. It is "B" if any of the blocked tasks
539 need to be priority boosted and "." otherwise.
541 o "btg" indicates whether boosting has been carried out during
542 the current grace period, with "exp" indicating that boosting
543 is in progress for an expedited grace period, "no" indicating
544 that boosting has not yet started for a normal grace period,
545 "begun" indicating that boosting has bebug for a normal grace
546 period, and "done" indicating that boosting has completed for
547 a normal grace period.
549 o "ntb" is the total number of tasks subjected to RCU priority boosting
552 o "neb" is the number of expedited grace periods that have had
553 to resort to RCU priority boosting since boot.
555 o "nnb" is the number of normal grace periods that have had
556 to resort to RCU priority boosting since boot.
558 o "j" is the low-order 16 bits of the jiffies counter in hexadecimal.
560 o "bt" is the low-order 16 bits of the value that the jiffies counter
561 will have at the next time that boosting is scheduled to begin.
563 o In the line beginning with "normal balk", the fields are as follows:
565 o "nt" is the number of times that the system balked from
566 boosting because there were no blocked tasks to boost.
567 Note that the system will balk from boosting even if the
568 grace period is overdue when the currently running task
569 is looping within an RCU read-side critical section.
570 There is no point in boosting in this case, because
571 boosting a running task won't make it run any faster.
573 o "gt" is the number of times that the system balked
574 from boosting because, although there were blocked tasks,
575 none of them were preventing the current grace period
578 o "bt" is the number of times that the system balked
579 from boosting because boosting was already in progress.
581 o "b" is the number of times that the system balked from
582 boosting because boosting had already completed for
583 the grace period in question.
585 o "ny" is the number of times that the system balked from
586 boosting because it was not yet time to start boosting
587 the grace period in question.
589 o "nos" is the number of times that the system balked from
590 boosting for inexplicable ("not otherwise specified")
591 reasons. This can actually happen due to races involving
592 increments of the jiffies counter.
594 o In the line beginning with "exp balk", the fields are as follows:
596 o "bt" is the number of times that the system balked from
597 boosting because there were no blocked tasks to boost.
599 o "nos" is the number of times that the system balked from
600 boosting for inexplicable ("not otherwise specified")