1.. SPDX-License-Identifier: GPL-2.0 2 3============================== 4Using RCU's CPU Stall Detector 5============================== 6 7This document first discusses what sorts of issues RCU's CPU stall 8detector can locate, and then discusses kernel parameters and Kconfig 9options that can be used to fine-tune the detector's operation. Finally, 10this document explains the stall detector's "splat" format. 11 12 13What Causes RCU CPU Stall Warnings? 14=================================== 15 16So your kernel printed an RCU CPU stall warning. The next question is 17"What caused it?" The following problems can result in RCU CPU stall 18warnings: 19 20- A CPU looping in an RCU read-side critical section. 21 22- A CPU looping with interrupts disabled. 23 24- A CPU looping with preemption disabled. 25 26- A CPU looping with bottom halves disabled. 27 28- For !CONFIG_PREEMPTION kernels, a CPU looping anywhere in the kernel 29 without invoking schedule(). If the looping in the kernel is 30 really expected and desirable behavior, you might need to add 31 some calls to cond_resched(). 32 33- Booting Linux using a console connection that is too slow to 34 keep up with the boot-time console-message rate. For example, 35 a 115Kbaud serial console can be *way* too slow to keep up 36 with boot-time message rates, and will frequently result in 37 RCU CPU stall warning messages. Especially if you have added 38 debug printk()s. 39 40- Anything that prevents RCU's grace-period kthreads from running. 41 This can result in the "All QSes seen" console-log message. 42 This message will include information on when the kthread last 43 ran and how often it should be expected to run. It can also 44 result in the ``rcu_.*kthread starved for`` console-log message, 45 which will include additional debugging information. 46 47- A CPU-bound real-time task in a CONFIG_PREEMPTION kernel, which might 48 happen to preempt a low-priority task in the middle of an RCU 49 read-side critical section. This is especially damaging if 50 that low-priority task is not permitted to run on any other CPU, 51 in which case the next RCU grace period can never complete, which 52 will eventually cause the system to run out of memory and hang. 53 While the system is in the process of running itself out of 54 memory, you might see stall-warning messages. 55 56- A CPU-bound real-time task in a CONFIG_PREEMPT_RT kernel that 57 is running at a higher priority than the RCU softirq threads. 58 This will prevent RCU callbacks from ever being invoked, 59 and in a CONFIG_PREEMPT_RCU kernel will further prevent 60 RCU grace periods from ever completing. Either way, the 61 system will eventually run out of memory and hang. In the 62 CONFIG_PREEMPT_RCU case, you might see stall-warning 63 messages. 64 65 You can use the rcutree.kthread_prio kernel boot parameter to 66 increase the scheduling priority of RCU's kthreads, which can 67 help avoid this problem. However, please note that doing this 68 can increase your system's context-switch rate and thus degrade 69 performance. 70 71- A periodic interrupt whose handler takes longer than the time 72 interval between successive pairs of interrupts. This can 73 prevent RCU's kthreads and softirq handlers from running. 74 Note that certain high-overhead debugging options, for example 75 the function_graph tracer, can result in interrupt handler taking 76 considerably longer than normal, which can in turn result in 77 RCU CPU stall warnings. 78 79- Testing a workload on a fast system, tuning the stall-warning 80 timeout down to just barely avoid RCU CPU stall warnings, and then 81 running the same workload with the same stall-warning timeout on a 82 slow system. Note that thermal throttling and on-demand governors 83 can cause a single system to be sometimes fast and sometimes slow! 84 85- A hardware or software issue shuts off the scheduler-clock 86 interrupt on a CPU that is not in dyntick-idle mode. This 87 problem really has happened, and seems to be most likely to 88 result in RCU CPU stall warnings for CONFIG_NO_HZ_COMMON=n kernels. 89 90- A hardware or software issue that prevents time-based wakeups 91 from occurring. These issues can range from misconfigured or 92 buggy timer hardware through bugs in the interrupt or exception 93 path (whether hardware, firmware, or software) through bugs 94 in Linux's timer subsystem through bugs in the scheduler, and, 95 yes, even including bugs in RCU itself. It can also result in 96 the ``rcu_.*timer wakeup didn't happen for`` console-log message, 97 which will include additional debugging information. 98 99- A low-level kernel issue that either fails to invoke one of the 100 variants of rcu_eqs_enter(true), rcu_eqs_exit(true), ct_idle_enter(), 101 ct_idle_exit(), ct_irq_enter(), or ct_irq_exit() on the one 102 hand, or that invokes one of them too many times on the other. 103 Historically, the most frequent issue has been an omission 104 of either irq_enter() or irq_exit(), which in turn invoke 105 ct_irq_enter() or ct_irq_exit(), respectively. Building your 106 kernel with CONFIG_RCU_EQS_DEBUG=y can help track down these types 107 of issues, which sometimes arise in architecture-specific code. 108 109- A bug in the RCU implementation. 110 111- A hardware failure. This is quite unlikely, but has occurred 112 at least once in real life. A CPU failed in a running system, 113 becoming unresponsive, but not causing an immediate crash. 114 This resulted in a series of RCU CPU stall warnings, eventually 115 leading the realization that the CPU had failed. 116 117The RCU, RCU-sched, and RCU-tasks implementations have CPU stall warning. 118Note that SRCU does *not* have CPU stall warnings. Please note that 119RCU only detects CPU stalls when there is a grace period in progress. 120No grace period, no CPU stall warnings. 121 122To diagnose the cause of the stall, inspect the stack traces. 123The offending function will usually be near the top of the stack. 124If you have a series of stall warnings from a single extended stall, 125comparing the stack traces can often help determine where the stall 126is occurring, which will usually be in the function nearest the top of 127that portion of the stack which remains the same from trace to trace. 128If you can reliably trigger the stall, ftrace can be quite helpful. 129 130RCU bugs can often be debugged with the help of CONFIG_RCU_TRACE 131and with RCU's event tracing. For information on RCU's event tracing, 132see include/trace/events/rcu.h. 133 134 135Fine-Tuning the RCU CPU Stall Detector 136====================================== 137 138The rcuupdate.rcu_cpu_stall_suppress module parameter disables RCU's 139CPU stall detector, which detects conditions that unduly delay RCU grace 140periods. This module parameter enables CPU stall detection by default, 141but may be overridden via boot-time parameter or at runtime via sysfs. 142The stall detector's idea of what constitutes "unduly delayed" is 143controlled by a set of kernel configuration variables and cpp macros: 144 145CONFIG_RCU_CPU_STALL_TIMEOUT 146---------------------------- 147 148 This kernel configuration parameter defines the period of time 149 that RCU will wait from the beginning of a grace period until it 150 issues an RCU CPU stall warning. This time period is normally 151 21 seconds. 152 153 This configuration parameter may be changed at runtime via the 154 /sys/module/rcupdate/parameters/rcu_cpu_stall_timeout, however 155 this parameter is checked only at the beginning of a cycle. 156 So if you are 10 seconds into a 40-second stall, setting this 157 sysfs parameter to (say) five will shorten the timeout for the 158 *next* stall, or the following warning for the current stall 159 (assuming the stall lasts long enough). It will not affect the 160 timing of the next warning for the current stall. 161 162 Stall-warning messages may be enabled and disabled completely via 163 /sys/module/rcupdate/parameters/rcu_cpu_stall_suppress. 164 165CONFIG_RCU_EXP_CPU_STALL_TIMEOUT 166-------------------------------- 167 168 Same as the CONFIG_RCU_CPU_STALL_TIMEOUT parameter but only for 169 the expedited grace period. This parameter defines the period 170 of time that RCU will wait from the beginning of an expedited 171 grace period until it issues an RCU CPU stall warning. This time 172 period is normally 20 milliseconds on Android devices. A zero 173 value causes the CONFIG_RCU_CPU_STALL_TIMEOUT value to be used, 174 after conversion to milliseconds. 175 176 This configuration parameter may be changed at runtime via the 177 /sys/module/rcupdate/parameters/rcu_exp_cpu_stall_timeout, however 178 this parameter is checked only at the beginning of a cycle. If you 179 are in a current stall cycle, setting it to a new value will change 180 the timeout for the -next- stall. 181 182 Stall-warning messages may be enabled and disabled completely via 183 /sys/module/rcupdate/parameters/rcu_cpu_stall_suppress. 184 185RCU_STALL_DELAY_DELTA 186--------------------- 187 188 Although the lockdep facility is extremely useful, it does add 189 some overhead. Therefore, under CONFIG_PROVE_RCU, the 190 RCU_STALL_DELAY_DELTA macro allows five extra seconds before 191 giving an RCU CPU stall warning message. (This is a cpp 192 macro, not a kernel configuration parameter.) 193 194RCU_STALL_RAT_DELAY 195------------------- 196 197 The CPU stall detector tries to make the offending CPU print its 198 own warnings, as this often gives better-quality stack traces. 199 However, if the offending CPU does not detect its own stall in 200 the number of jiffies specified by RCU_STALL_RAT_DELAY, then 201 some other CPU will complain. This delay is normally set to 202 two jiffies. (This is a cpp macro, not a kernel configuration 203 parameter.) 204 205rcupdate.rcu_task_stall_timeout 206------------------------------- 207 208 This boot/sysfs parameter controls the RCU-tasks stall warning 209 interval. A value of zero or less suppresses RCU-tasks stall 210 warnings. A positive value sets the stall-warning interval 211 in seconds. An RCU-tasks stall warning starts with the line: 212 213 INFO: rcu_tasks detected stalls on tasks: 214 215 And continues with the output of sched_show_task() for each 216 task stalling the current RCU-tasks grace period. 217 218 219Interpreting RCU's CPU Stall-Detector "Splats" 220============================================== 221 222For non-RCU-tasks flavors of RCU, when a CPU detects that some other 223CPU is stalling, it will print a message similar to the following:: 224 225 INFO: rcu_sched detected stalls on CPUs/tasks: 226 2-...: (3 GPs behind) idle=06c/0/0 softirq=1453/1455 fqs=0 227 16-...: (0 ticks this GP) idle=81c/0/0 softirq=764/764 fqs=0 228 (detected by 32, t=2603 jiffies, g=7075, q=625) 229 230This message indicates that CPU 32 detected that CPUs 2 and 16 were both 231causing stalls, and that the stall was affecting RCU-sched. This message 232will normally be followed by stack dumps for each CPU. Please note that 233PREEMPT_RCU builds can be stalled by tasks as well as by CPUs, and that 234the tasks will be indicated by PID, for example, "P3421". It is even 235possible for an rcu_state stall to be caused by both CPUs *and* tasks, 236in which case the offending CPUs and tasks will all be called out in the list. 237In some cases, CPUs will detect themselves stalling, which will result 238in a self-detected stall. 239 240CPU 2's "(3 GPs behind)" indicates that this CPU has not interacted with 241the RCU core for the past three grace periods. In contrast, CPU 16's "(0 242ticks this GP)" indicates that this CPU has not taken any scheduling-clock 243interrupts during the current stalled grace period. 244 245The "idle=" portion of the message prints the dyntick-idle state. 246The hex number before the first "/" is the low-order 12 bits of the 247dynticks counter, which will have an even-numbered value if the CPU 248is in dyntick-idle mode and an odd-numbered value otherwise. The hex 249number between the two "/"s is the value of the nesting, which will be 250a small non-negative number if in the idle loop (as shown above) and a 251very large positive number otherwise. 252 253The "softirq=" portion of the message tracks the number of RCU softirq 254handlers that the stalled CPU has executed. The number before the "/" 255is the number that had executed since boot at the time that this CPU 256last noted the beginning of a grace period, which might be the current 257(stalled) grace period, or it might be some earlier grace period (for 258example, if the CPU might have been in dyntick-idle mode for an extended 259time period). The number after the "/" is the number that have executed 260since boot until the current time. If this latter number stays constant 261across repeated stall-warning messages, it is possible that RCU's softirq 262handlers are no longer able to execute on this CPU. This can happen if 263the stalled CPU is spinning with interrupts are disabled, or, in -rt 264kernels, if a high-priority process is starving RCU's softirq handler. 265 266The "fqs=" shows the number of force-quiescent-state idle/offline 267detection passes that the grace-period kthread has made across this 268CPU since the last time that this CPU noted the beginning of a grace 269period. 270 271The "detected by" line indicates which CPU detected the stall (in this 272case, CPU 32), how many jiffies have elapsed since the start of the grace 273period (in this case 2603), the grace-period sequence number (7075), and 274an estimate of the total number of RCU callbacks queued across all CPUs 275(625 in this case). 276 277If the grace period ends just as the stall warning starts printing, 278there will be a spurious stall-warning message, which will include 279the following:: 280 281 INFO: Stall ended before state dump start 282 283This is rare, but does happen from time to time in real life. It is also 284possible for a zero-jiffy stall to be flagged in this case, depending 285on how the stall warning and the grace-period initialization happen to 286interact. Please note that it is not possible to entirely eliminate this 287sort of false positive without resorting to things like stop_machine(), 288which is overkill for this sort of problem. 289 290If all CPUs and tasks have passed through quiescent states, but the 291grace period has nevertheless failed to end, the stall-warning splat 292will include something like the following:: 293 294 All QSes seen, last rcu_preempt kthread activity 23807 (4297905177-4297881370), jiffies_till_next_fqs=3, root ->qsmask 0x0 295 296The "23807" indicates that it has been more than 23 thousand jiffies 297since the grace-period kthread ran. The "jiffies_till_next_fqs" 298indicates how frequently that kthread should run, giving the number 299of jiffies between force-quiescent-state scans, in this case three, 300which is way less than 23807. Finally, the root rcu_node structure's 301->qsmask field is printed, which will normally be zero. 302 303If the relevant grace-period kthread has been unable to run prior to 304the stall warning, as was the case in the "All QSes seen" line above, 305the following additional line is printed:: 306 307 rcu_sched kthread starved for 23807 jiffies! g7075 f0x0 RCU_GP_WAIT_FQS(3) ->state=0x1 ->cpu=5 308 Unless rcu_sched kthread gets sufficient CPU time, OOM is now expected behavior. 309 310Starving the grace-period kthreads of CPU time can of course result 311in RCU CPU stall warnings even when all CPUs and tasks have passed 312through the required quiescent states. The "g" number shows the current 313grace-period sequence number, the "f" precedes the ->gp_flags command 314to the grace-period kthread, the "RCU_GP_WAIT_FQS" indicates that the 315kthread is waiting for a short timeout, the "state" precedes value of the 316task_struct ->state field, and the "cpu" indicates that the grace-period 317kthread last ran on CPU 5. 318 319If the relevant grace-period kthread does not wake from FQS wait in a 320reasonable time, then the following additional line is printed:: 321 322 kthread timer wakeup didn't happen for 23804 jiffies! g7076 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x402 323 324The "23804" indicates that kthread's timer expired more than 23 thousand 325jiffies ago. The rest of the line has meaning similar to the kthread 326starvation case. 327 328Additionally, the following line is printed:: 329 330 Possible timer handling issue on cpu=4 timer-softirq=11142 331 332Here "cpu" indicates that the grace-period kthread last ran on CPU 4, 333where it queued the fqs timer. The number following the "timer-softirq" 334is the current ``TIMER_SOFTIRQ`` count on cpu 4. If this value does not 335change on successive RCU CPU stall warnings, there is further reason to 336suspect a timer problem. 337 338These messages are usually followed by stack dumps of the CPUs and tasks 339involved in the stall. These stack traces can help you locate the cause 340of the stall, keeping in mind that the CPU detecting the stall will have 341an interrupt frame that is mainly devoted to detecting the stall. 342 343 344Multiple Warnings From One Stall 345================================ 346 347If a stall lasts long enough, multiple stall-warning messages will 348be printed for it. The second and subsequent messages are printed at 349longer intervals, so that the time between (say) the first and second 350message will be about three times the interval between the beginning 351of the stall and the first message. It can be helpful to compare the 352stack dumps for the different messages for the same stalled grace period. 353 354 355Stall Warnings for Expedited Grace Periods 356========================================== 357 358If an expedited grace period detects a stall, it will place a message 359like the following in dmesg:: 360 361 INFO: rcu_sched detected expedited stalls on CPUs/tasks: { 7-... } 21119 jiffies s: 73 root: 0x2/. 362 363This indicates that CPU 7 has failed to respond to a reschedule IPI. 364The three periods (".") following the CPU number indicate that the CPU 365is online (otherwise the first period would instead have been "O"), 366that the CPU was online at the beginning of the expedited grace period 367(otherwise the second period would have instead been "o"), and that 368the CPU has been online at least once since boot (otherwise, the third 369period would instead have been "N"). The number before the "jiffies" 370indicates that the expedited grace period has been going on for 21,119 371jiffies. The number following the "s:" indicates that the expedited 372grace-period sequence counter is 73. The fact that this last value is 373odd indicates that an expedited grace period is in flight. The number 374following "root:" is a bitmask that indicates which children of the root 375rcu_node structure correspond to CPUs and/or tasks that are blocking the 376current expedited grace period. If the tree had more than one level, 377additional hex numbers would be printed for the states of the other 378rcu_node structures in the tree. 379 380As with normal grace periods, PREEMPT_RCU builds can be stalled by 381tasks as well as by CPUs, and that the tasks will be indicated by PID, 382for example, "P3421". 383 384It is entirely possible to see stall warnings from normal and from 385expedited grace periods at about the same time during the same run. 386