Documentation/RCU/stallwarn.txt

   1 Using RCU's CPU Stall Detector
   2
   3 This document first discusses what sorts of issues RCU's CPU stall
   4 detector can locate, and then discusses kernel parameters and Kconfig
   5 options that can be used to fine-tune the detector's operation.  Finally,
   6 this document explains the stall detector's "splat" format.
   7
   8
   9 What Causes RCU CPU Stall Warnings?
  10
  11 So your kernel printed an RCU CPU stall warning.  The next question is
  12 "What caused it?"  The following problems can result in RCU CPU stall
  13 warnings:
  14
  15 o       A CPU looping in an RCU read-side critical section.
  16
  17 o       A CPU looping with interrupts disabled.
  18
  19 o       A CPU looping with preemption disabled.  This condition can
  20         result in RCU-sched stalls and, if ksoftirqd is in use, RCU-bh
  21         stalls.
  22
  23 o       A CPU looping with bottom halves disabled.  This condition can
  24         result in RCU-sched and RCU-bh stalls.
  25
  26 o       For !CONFIG_PREEMPT kernels, a CPU looping anywhere in the
  27         kernel without invoking schedule().  Note that cond_resched()
  28         does not necessarily prevent RCU CPU stall warnings.  Therefore,
  29         if the looping in the kernel is really expected and desirable
  30         behavior, you might need to replace some of the cond_resched()
  31         calls with calls to cond_resched_rcu_qs().
  32
  33 o       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 o       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.
  44
  45 o       A CPU-bound real-time task in a CONFIG_PREEMPT kernel, which might
  46         happen to preempt a low-priority task in the middle of an RCU
  47         read-side critical section.   This is especially damaging if
  48         that low-priority task is not permitted to run on any other CPU,
  49         in which case the next RCU grace period can never complete, which
  50         will eventually cause the system to run out of memory and hang.
  51         While the system is in the process of running itself out of
  52         memory, you might see stall-warning messages.
  53
  54 o       A CPU-bound real-time task in a CONFIG_PREEMPT_RT kernel that
  55         is running at a higher priority than the RCU softirq threads.
  56         This will prevent RCU callbacks from ever being invoked,
  57         and in a CONFIG_PREEMPT_RCU kernel will further prevent
  58         RCU grace periods from ever completing.  Either way, the
  59         system will eventually run out of memory and hang.  In the
  60         CONFIG_PREEMPT_RCU case, you might see stall-warning
  61         messages.
  62
  63 o       A hardware or software issue shuts off the scheduler-clock
  64         interrupt on a CPU that is not in dyntick-idle mode.  This
  65         problem really has happened, and seems to be most likely to
  66         result in RCU CPU stall warnings for CONFIG_NO_HZ_COMMON=n kernels.
  67
  68 o       A bug in the RCU implementation.
  69
  70 o       A hardware failure.  This is quite unlikely, but has occurred
  71         at least once in real life.  A CPU failed in a running system,
  72         becoming unresponsive, but not causing an immediate crash.
  73         This resulted in a series of RCU CPU stall warnings, eventually
  74         leading the realization that the CPU had failed.
  75
  76 The RCU, RCU-sched, RCU-bh, and RCU-tasks implementations have CPU stall
  77 warning.  Note that SRCU does -not- have CPU stall warnings.  Please note
  78 that RCU only detects CPU stalls when there is a grace period in progress.
  79 No grace period, no CPU stall warnings.
  80
  81 To diagnose the cause of the stall, inspect the stack traces.
  82 The offending function will usually be near the top of the stack.
  83 If you have a series of stall warnings from a single extended stall,
  84 comparing the stack traces can often help determine where the stall
  85 is occurring, which will usually be in the function nearest the top of
  86 that portion of the stack which remains the same from trace to trace.
  87 If you can reliably trigger the stall, ftrace can be quite helpful.
  88
  89 RCU bugs can often be debugged with the help of CONFIG_RCU_TRACE
  90 and with RCU's event tracing.  For information on RCU's event tracing,
  91 see include/trace/events/rcu.h.
  92
  93
  94 Fine-Tuning the RCU CPU Stall Detector
  95
  96 The rcuupdate.rcu_cpu_stall_suppress module parameter disables RCU's
  97 CPU stall detector, which detects conditions that unduly delay RCU grace
  98 periods.  This module parameter enables CPU stall detection by default,
  99 but may be overridden via boot-time parameter or at runtime via sysfs.
 100 The stall detector's idea of what constitutes "unduly delayed" is
 101 controlled by a set of kernel configuration variables and cpp macros:
 102
 103 CONFIG_RCU_CPU_STALL_TIMEOUT
 104
 105         This kernel configuration parameter defines the period of time
 106         that RCU will wait from the beginning of a grace period until it
 107         issues an RCU CPU stall warning.  This time period is normally
 108         21 seconds.
 109
 110         This configuration parameter may be changed at runtime via the
 111         /sys/module/rcupdate/parameters/rcu_cpu_stall_timeout, however
 112         this parameter is checked only at the beginning of a cycle.
 113         So if you are 10 seconds into a 40-second stall, setting this
 114         sysfs parameter to (say) five will shorten the timeout for the
 115         -next- stall, or the following warning for the current stall
 116         (assuming the stall lasts long enough).  It will not affect the
 117         timing of the next warning for the current stall.
 118
 119         Stall-warning messages may be enabled and disabled completely via
 120         /sys/module/rcupdate/parameters/rcu_cpu_stall_suppress.
 121
 122 RCU_STALL_DELAY_DELTA
 123
 124         Although the lockdep facility is extremely useful, it does add
 125         some overhead.  Therefore, under CONFIG_PROVE_RCU, the
 126         RCU_STALL_DELAY_DELTA macro allows five extra seconds before
 127         giving an RCU CPU stall warning message.  (This is a cpp
 128         macro, not a kernel configuration parameter.)
 129
 130 RCU_STALL_RAT_DELAY
 131
 132         The CPU stall detector tries to make the offending CPU print its
 133         own warnings, as this often gives better-quality stack traces.
 134         However, if the offending CPU does not detect its own stall in
 135         the number of jiffies specified by RCU_STALL_RAT_DELAY, then
 136         some other CPU will complain.  This delay is normally set to
 137         two jiffies.  (This is a cpp macro, not a kernel configuration
 138         parameter.)
 139
 140 rcupdate.rcu_task_stall_timeout
 141
 142         This boot/sysfs parameter controls the RCU-tasks stall warning
 143         interval.  A value of zero or less suppresses RCU-tasks stall
 144         warnings.  A positive value sets the stall-warning interval
 145         in jiffies.  An RCU-tasks stall warning starts with the line:
 146
 147                 INFO: rcu_tasks detected stalls on tasks:
 148
 149         And continues with the output of sched_show_task() for each
 150         task stalling the current RCU-tasks grace period.
 151
 152
 153 Interpreting RCU's CPU Stall-Detector "Splats"
 154
 155 For non-RCU-tasks flavors of RCU, when a CPU detects that it is stalling,
 156 it will print a message similar to the following:
 157
 158 INFO: rcu_sched_state detected stall on CPU 5 (t=2500 jiffies)
 159
 160 This message indicates that CPU 5 detected that it was causing a stall,
 161 and that the stall was affecting RCU-sched.  This message will normally be
 162 followed by a stack dump of the offending CPU.  On TREE_RCU kernel builds,
 163 RCU and RCU-sched are implemented by the same underlying mechanism,
 164 while on PREEMPT_RCU kernel builds, RCU is instead implemented
 165 by rcu_preempt_state.
 166
 167 On the other hand, if the offending CPU fails to print out a stall-warning
 168 message quickly enough, some other CPU will print a message similar to
 169 the following:
 170
 171 INFO: rcu_bh_state detected stalls on CPUs/tasks: { 3 5 } (detected by 2, 2502 jiffies)
 172
 173 This message indicates that CPU 2 detected that CPUs 3 and 5 were both
 174 causing stalls, and that the stall was affecting RCU-bh.  This message
 175 will normally be followed by stack dumps for each CPU.  Please note that
 176 PREEMPT_RCU builds can be stalled by tasks as well as by CPUs,
 177 and that the tasks will be indicated by PID, for example, "P3421".
 178 It is even possible for a rcu_preempt_state stall to be caused by both
 179 CPUs -and- tasks, in which case the offending CPUs and tasks will all
 180 be called out in the list.
 181
 182 Finally, if the grace period ends just as the stall warning starts
 183 printing, there will be a spurious stall-warning message:
 184
 185 INFO: rcu_bh_state detected stalls on CPUs/tasks: { } (detected by 4, 2502 jiffies)
 186
 187 This is rare, but does happen from time to time in real life.  It is also
 188 possible for a zero-jiffy stall to be flagged in this case, depending
 189 on how the stall warning and the grace-period initialization happen to
 190 interact.  Please note that it is not possible to entirely eliminate this
 191 sort of false positive without resorting to things like stop_machine(),
 192 which is overkill for this sort of problem.
 193
 194 Recent kernels will print a long form of the stall-warning message:
 195
 196         INFO: rcu_preempt detected stall on CPU
 197         0: (63959 ticks this GP) idle=241/3fffffffffffffff/0 softirq=82/543
 198            (t=65000 jiffies)
 199
 200 In kernels with CONFIG_RCU_FAST_NO_HZ, more information is printed:
 201
 202         INFO: rcu_preempt detected stall on CPU
 203         0: (64628 ticks this GP) idle=dd5/3fffffffffffffff/0 softirq=82/543 last_accelerate: a345/d342 nonlazy_posted: 25 .D
 204            (t=65000 jiffies)
 205
 206 The "(64628 ticks this GP)" indicates that this CPU has taken more
 207 than 64,000 scheduling-clock interrupts during the current stalled
 208 grace period.  If the CPU was not yet aware of the current grace
 209 period (for example, if it was offline), then this part of the message
 210 indicates how many grace periods behind the CPU is.
 211
 212 The "idle=" portion of the message prints the dyntick-idle state.
 213 The hex number before the first "/" is the low-order 12 bits of the
 214 dynticks counter, which will have an even-numbered value if the CPU is
 215 in dyntick-idle mode and an odd-numbered value otherwise.  The hex
 216 number between the two "/"s is the value of the nesting, which will
 217 be a small positive number if in the idle loop and a very large positive
 218 number (as shown above) otherwise.
 219
 220 The "softirq=" portion of the message tracks the number of RCU softirq
 221 handlers that the stalled CPU has executed.  The number before the "/"
 222 is the number that had executed since boot at the time that this CPU
 223 last noted the beginning of a grace period, which might be the current
 224 (stalled) grace period, or it might be some earlier grace period (for
 225 example, if the CPU might have been in dyntick-idle mode for an extended
 226 time period.  The number after the "/" is the number that have executed
 227 since boot until the current time.  If this latter number stays constant
 228 across repeated stall-warning messages, it is possible that RCU's softirq
 229 handlers are no longer able to execute on this CPU.  This can happen if
 230 the stalled CPU is spinning with interrupts are disabled, or, in -rt
 231 kernels, if a high-priority process is starving RCU's softirq handler.
 232
 233 For CONFIG_RCU_FAST_NO_HZ kernels, the "last_accelerate:" prints the
 234 low-order 16 bits (in hex) of the jiffies counter when this CPU last
 235 invoked rcu_try_advance_all_cbs() from rcu_needs_cpu() or last invoked
 236 rcu_accelerate_cbs() from rcu_prepare_for_idle().  The "nonlazy_posted:"
 237 prints the number of non-lazy callbacks posted since the last call to
 238 rcu_needs_cpu().  Finally, an "L" indicates that there are currently
 239 no non-lazy callbacks ("." is printed otherwise, as shown above) and
 240 "D" indicates that dyntick-idle processing is enabled ("." is printed
 241 otherwise, for example, if disabled via the "nohz=" kernel boot parameter).
 242
 243 If the relevant grace-period kthread has been unable to run prior to
 244 the stall warning, the following additional line is printed:
 245
 246         rcu_preempt kthread starved for 2023 jiffies!
 247
 248 Starving the grace-period kthreads of CPU time can of course result in
 249 RCU CPU stall warnings even when all CPUs and tasks have passed through
 250 the required quiescent states.
 251
 252
 253 Multiple Warnings From One Stall
 254
 255 If a stall lasts long enough, multiple stall-warning messages will be
 256 printed for it.  The second and subsequent messages are printed at
 257 longer intervals, so that the time between (say) the first and second
 258 message will be about three times the interval between the beginning
 259 of the stall and the first message.
 260
 261
 262 Stall Warnings for Expedited Grace Periods
 263
 264 If an expedited grace period detects a stall, it will place a message
 265 like the following in dmesg:
 266
 267         INFO: rcu_sched detected expedited stalls on CPUs: { 1 2 6 } 26009 jiffies s: 1043
 268
 269 This indicates that CPUs 1, 2, and 6 have failed to respond to a
 270 reschedule IPI, that the expedited grace period has been going on for
 271 26,009 jiffies, and that the expedited grace-period sequence counter is
 272 1043.  The fact that this last value is odd indicates that an expedited
 273 grace period is in flight.
 274
 275 It is entirely possible to see stall warnings from normal and from
 276 expedited grace periods at about the same time from the same run.