Documentation/RCU/torture.txt

   1 RCU Torture Test Operation
   2
   3
   4 CONFIG_RCU_TORTURE_TEST
   5
   6 The CONFIG_RCU_TORTURE_TEST config option is available for all RCU
   7 implementations.  It creates an rcutorture kernel module that can
   8 be loaded to run a torture test.  The test periodically outputs
   9 status messages via printk(), which can be examined via the dmesg
  10 command (perhaps grepping for "torture").  The test is started
  11 when the module is loaded, and stops when the module is unloaded.
  12
  13 CONFIG_RCU_TORTURE_TEST_RUNNABLE
  14
  15 It is also possible to specify CONFIG_RCU_TORTURE_TEST=y, which will
  16 result in the tests being loaded into the base kernel.  In this case,
  17 the CONFIG_RCU_TORTURE_TEST_RUNNABLE config option is used to specify
  18 whether the RCU torture tests are to be started immediately during
  19 boot or whether the /proc/sys/kernel/rcutorture_runnable file is used
  20 to enable them.  This /proc file can be used to repeatedly pause and
  21 restart the tests, regardless of the initial state specified by the
  22 CONFIG_RCU_TORTURE_TEST_RUNNABLE config option.
  23
  24 You will normally -not- want to start the RCU torture tests during boot
  25 (and thus the default is CONFIG_RCU_TORTURE_TEST_RUNNABLE=n), but doing
  26 this can sometimes be useful in finding boot-time bugs.
  27
  28
  29 MODULE PARAMETERS
  30
  31 This module has the following parameters:
  32
  33 fqs_duration    Duration (in microseconds) of artificially induced bursts
  34                 of force_quiescent_state() invocations.  In RCU
  35                 implementations having force_quiescent_state(), these
  36                 bursts help force races between forcing a given grace
  37                 period and that grace period ending on its own.
  38
  39 fqs_holdoff     Holdoff time (in microseconds) between consecutive calls
  40                 to force_quiescent_state() within a burst.
  41
  42 fqs_stutter     Wait time (in seconds) between consecutive bursts
  43                 of calls to force_quiescent_state().
  44
  45 gp_normal       Make the fake writers use normal synchronous grace-period
  46                 primitives.
  47
  48 gp_exp          Make the fake writers use expedited synchronous grace-period
  49                 primitives.  If both gp_normal and gp_exp are set, or
  50                 if neither gp_normal nor gp_exp are set, then randomly
  51                 choose the primitive so that about 50% are normal and
  52                 50% expedited.  By default, neither are set, which
  53                 gives best overall test coverage.
  54
  55 irqreader       Says to invoke RCU readers from irq level.  This is currently
  56                 done via timers.  Defaults to "1" for variants of RCU that
  57                 permit this.  (Or, more accurately, variants of RCU that do
  58                 -not- permit this know to ignore this variable.)
  59
  60 n_barrier_cbs   If this is nonzero, RCU barrier testing will be conducted,
  61                 in which case n_barrier_cbs specifies the number of
  62                 RCU callbacks (and corresponding kthreads) to use for
  63                 this testing.  The value cannot be negative.  If you
  64                 specify this to be non-zero when torture_type indicates a
  65                 synchronous RCU implementation (one for which a member of
  66                 the synchronize_rcu() rather than the call_rcu() family is
  67                 used -- see the documentation for torture_type below), an
  68                 error will be reported and no testing will be carried out.
  69
  70 nfakewriters    This is the number of RCU fake writer threads to run.  Fake
  71                 writer threads repeatedly use the synchronous "wait for
  72                 current readers" function of the interface selected by
  73                 torture_type, with a delay between calls to allow for various
  74                 different numbers of writers running in parallel.
  75                 nfakewriters defaults to 4, which provides enough parallelism
  76                 to trigger special cases caused by multiple writers, such as
  77                 the synchronize_srcu() early return optimization.
  78
  79 nreaders        This is the number of RCU reading threads supported.
  80                 The default is twice the number of CPUs.  Why twice?
  81                 To properly exercise RCU implementations with preemptible
  82                 read-side critical sections.
  83
  84 onoff_interval
  85                 The number of seconds between each attempt to execute a
  86                 randomly selected CPU-hotplug operation.  Defaults to
  87                 zero, which disables CPU hotplugging.  In HOTPLUG_CPU=n
  88                 kernels, rcutorture will silently refuse to do any
  89                 CPU-hotplug operations regardless of what value is
  90                 specified for onoff_interval.
  91
  92 onoff_holdoff   The number of seconds to wait until starting CPU-hotplug
  93                 operations.  This would normally only be used when
  94                 rcutorture was built into the kernel and started
  95                 automatically at boot time, in which case it is useful
  96                 in order to avoid confusing boot-time code with CPUs
  97                 coming and going.
  98
  99 shuffle_interval
 100                 The number of seconds to keep the test threads affinitied
 101                 to a particular subset of the CPUs, defaults to 3 seconds.
 102                 Used in conjunction with test_no_idle_hz.
 103
 104 shutdown_secs   The number of seconds to run the test before terminating
 105                 the test and powering off the system.  The default is
 106                 zero, which disables test termination and system shutdown.
 107                 This capability is useful for automated testing.
 108
 109 stall_cpu       The number of seconds that a CPU should be stalled while
 110                 within both an rcu_read_lock() and a preempt_disable().
 111                 This stall happens only once per rcutorture run.
 112                 If you need multiple stalls, use modprobe and rmmod to
 113                 repeatedly run rcutorture.  The default for stall_cpu
 114                 is zero, which prevents rcutorture from stalling a CPU.
 115
 116                 Note that attempts to rmmod rcutorture while the stall
 117                 is ongoing will hang, so be careful what value you
 118                 choose for this module parameter!  In addition, too-large
 119                 values for stall_cpu might well induce failures and
 120                 warnings in other parts of the kernel.  You have been
 121                 warned!
 122
 123 stall_cpu_holdoff
 124                 The number of seconds to wait after rcutorture starts
 125                 before stalling a CPU.  Defaults to 10 seconds.
 126
 127 stat_interval   The number of seconds between output of torture
 128                 statistics (via printk()).  Regardless of the interval,
 129                 statistics are printed when the module is unloaded.
 130                 Setting the interval to zero causes the statistics to
 131                 be printed -only- when the module is unloaded, and this
 132                 is the default.
 133
 134 stutter         The length of time to run the test before pausing for this
 135                 same period of time.  Defaults to "stutter=5", so as
 136                 to run and pause for (roughly) five-second intervals.
 137                 Specifying "stutter=0" causes the test to run continuously
 138                 without pausing, which is the old default behavior.
 139
 140 test_boost      Whether or not to test the ability of RCU to do priority
 141                 boosting.  Defaults to "test_boost=1", which performs
 142                 RCU priority-inversion testing only if the selected
 143                 RCU implementation supports priority boosting.  Specifying
 144                 "test_boost=0" never performs RCU priority-inversion
 145                 testing.  Specifying "test_boost=2" performs RCU
 146                 priority-inversion testing even if the selected RCU
 147                 implementation does not support RCU priority boosting,
 148                 which can be used to test rcutorture's ability to
 149                 carry out RCU priority-inversion testing.
 150
 151 test_boost_interval
 152                 The number of seconds in an RCU priority-inversion test
 153                 cycle.  Defaults to "test_boost_interval=7".  It is
 154                 usually wise for this value to be relatively prime to
 155                 the value selected for "stutter".
 156
 157 test_boost_duration
 158                 The number of seconds to do RCU priority-inversion testing
 159                 within any given "test_boost_interval".  Defaults to
 160                 "test_boost_duration=4".
 161
 162 test_no_idle_hz Whether or not to test the ability of RCU to operate in
 163                 a kernel that disables the scheduling-clock interrupt to
 164                 idle CPUs.  Boolean parameter, "1" to test, "0" otherwise.
 165                 Defaults to omitting this test.
 166
 167 torture_type    The type of RCU to test, with string values as follows:
 168
 169                 "rcu":  rcu_read_lock(), rcu_read_unlock() and call_rcu().
 170
 171                 "rcu_sync":  rcu_read_lock(), rcu_read_unlock(), and
 172                         synchronize_rcu().
 173
 174                 "rcu_expedited": rcu_read_lock(), rcu_read_unlock(), and
 175                         synchronize_rcu_expedited().
 176
 177                 "rcu_bh": rcu_read_lock_bh(), rcu_read_unlock_bh(), and
 178                         call_rcu_bh().
 179
 180                 "rcu_bh_sync": rcu_read_lock_bh(), rcu_read_unlock_bh(),
 181                         and synchronize_rcu_bh().
 182
 183                 "rcu_bh_expedited": rcu_read_lock_bh(), rcu_read_unlock_bh(),
 184                         and synchronize_rcu_bh_expedited().
 185
 186                 "srcu": srcu_read_lock(), srcu_read_unlock() and
 187                         call_srcu().
 188
 189                 "srcu_sync": srcu_read_lock(), srcu_read_unlock() and
 190                         synchronize_srcu().
 191
 192                 "srcu_expedited": srcu_read_lock(), srcu_read_unlock() and
 193                         synchronize_srcu_expedited().
 194
 195                 "sched": preempt_disable(), preempt_enable(), and
 196                         call_rcu_sched().
 197
 198                 "sched_sync": preempt_disable(), preempt_enable(), and
 199                         synchronize_sched().
 200
 201                 "sched_expedited": preempt_disable(), preempt_enable(), and
 202                         synchronize_sched_expedited().
 203
 204                 Defaults to "rcu".
 205
 206 verbose         Enable debug printk()s.  Default is disabled.
 207
 208
 209 OUTPUT
 210
 211 The statistics output is as follows:
 212
 213         rcu-torture:--- Start of test: nreaders=16 nfakewriters=4 stat_interval=30 verbose=0 test_no_idle_hz=1 shuffle_interval=3 stutter=5 irqreader=1 fqs_duration=0 fqs_holdoff=0 fqs_stutter=3 test_boost=1/0 test_boost_interval=7 test_boost_duration=4
 214         rcu-torture: rtc:           (null) ver: 155441 tfle: 0 rta: 155441 rtaf: 8884 rtf: 155440 rtmbe: 0 rtbe: 0 rtbke: 0 rtbre: 0 rtbf: 0 rtb: 0 nt: 3055767
 215         rcu-torture: Reader Pipe:  727860534 34213 0 0 0 0 0 0 0 0 0
 216         rcu-torture: Reader Batch:  727877838 17003 0 0 0 0 0 0 0 0 0
 217         rcu-torture: Free-Block Circulation:  155440 155440 155440 155440 155440 155440 155440 155440 155440 155440 0
 218         rcu-torture:--- End of test: SUCCESS: nreaders=16 nfakewriters=4 stat_interval=30 verbose=0 test_no_idle_hz=1 shuffle_interval=3 stutter=5 irqreader=1 fqs_duration=0 fqs_holdoff=0 fqs_stutter=3 test_boost=1/0 test_boost_interval=7 test_boost_duration=4
 219
 220 The command "dmesg | grep torture:" will extract this information on
 221 most systems.  On more esoteric configurations, it may be necessary to
 222 use other commands to access the output of the printk()s used by
 223 the RCU torture test.  The printk()s use KERN_ALERT, so they should
 224 be evident.  ;-)
 225
 226 The first and last lines show the rcutorture module parameters, and the
 227 last line shows either "SUCCESS" or "FAILURE", based on rcutorture's
 228 automatic determination as to whether RCU operated correctly.
 229
 230 The entries are as follows:
 231
 232 o       "rtc": The hexadecimal address of the structure currently visible
 233         to readers.
 234
 235 o       "ver": The number of times since boot that the RCU writer task
 236         has changed the structure visible to readers.
 237
 238 o       "tfle": If non-zero, indicates that the "torture freelist"
 239         containing structures to be placed into the "rtc" area is empty.
 240         This condition is important, since it can fool you into thinking
 241         that RCU is working when it is not.  :-/
 242
 243 o       "rta": Number of structures allocated from the torture freelist.
 244
 245 o       "rtaf": Number of allocations from the torture freelist that have
 246         failed due to the list being empty.  It is not unusual for this
 247         to be non-zero, but it is bad for it to be a large fraction of
 248         the value indicated by "rta".
 249
 250 o       "rtf": Number of frees into the torture freelist.
 251
 252 o       "rtmbe": A non-zero value indicates that rcutorture believes that
 253         rcu_assign_pointer() and rcu_dereference() are not working
 254         correctly.  This value should be zero.
 255
 256 o       "rtbe": A non-zero value indicates that one of the rcu_barrier()
 257         family of functions is not working correctly.
 258
 259 o       "rtbke": rcutorture was unable to create the real-time kthreads
 260         used to force RCU priority inversion.  This value should be zero.
 261
 262 o       "rtbre": Although rcutorture successfully created the kthreads
 263         used to force RCU priority inversion, it was unable to set them
 264         to the real-time priority level of 1.  This value should be zero.
 265
 266 o       "rtbf": The number of times that RCU priority boosting failed
 267         to resolve RCU priority inversion.
 268
 269 o       "rtb": The number of times that rcutorture attempted to force
 270         an RCU priority inversion condition.  If you are testing RCU
 271         priority boosting via the "test_boost" module parameter, this
 272         value should be non-zero.
 273
 274 o       "nt": The number of times rcutorture ran RCU read-side code from
 275         within a timer handler.  This value should be non-zero only
 276         if you specified the "irqreader" module parameter.
 277
 278 o       "Reader Pipe": Histogram of "ages" of structures seen by readers.
 279         If any entries past the first two are non-zero, RCU is broken.
 280         And rcutorture prints the error flag string "!!!" to make sure
 281         you notice.  The age of a newly allocated structure is zero,
 282         it becomes one when removed from reader visibility, and is
 283         incremented once per grace period subsequently -- and is freed
 284         after passing through (RCU_TORTURE_PIPE_LEN-2) grace periods.
 285
 286         The output displayed above was taken from a correctly working
 287         RCU.  If you want to see what it looks like when broken, break
 288         it yourself.  ;-)
 289
 290 o       "Reader Batch": Another histogram of "ages" of structures seen
 291         by readers, but in terms of counter flips (or batches) rather
 292         than in terms of grace periods.  The legal number of non-zero
 293         entries is again two.  The reason for this separate view is that
 294         it is sometimes easier to get the third entry to show up in the
 295         "Reader Batch" list than in the "Reader Pipe" list.
 296
 297 o       "Free-Block Circulation": Shows the number of torture structures
 298         that have reached a given point in the pipeline.  The first element
 299         should closely correspond to the number of structures allocated,
 300         the second to the number that have been removed from reader view,
 301         and all but the last remaining to the corresponding number of
 302         passes through a grace period.  The last entry should be zero,
 303         as it is only incremented if a torture structure's counter
 304         somehow gets incremented farther than it should.
 305
 306 Different implementations of RCU can provide implementation-specific
 307 additional information.  For example, SRCU provides the following
 308 additional line:
 309
 310         srcu-torture: per-CPU(idx=1): 0(0,1) 1(0,1) 2(0,0) 3(0,1)
 311
 312 This line shows the per-CPU counter state.  The numbers in parentheses are
 313 the values of the "old" and "current" counters for the corresponding CPU.
 314 The "idx" value maps the "old" and "current" values to the underlying
 315 array, and is useful for debugging.
 316
 317
 318 USAGE
 319
 320 The following script may be used to torture RCU:
 321
 322         #!/bin/sh
 323
 324         modprobe rcutorture
 325         sleep 3600
 326         rmmod rcutorture
 327         dmesg | grep torture:
 328
 329 The output can be manually inspected for the error flag of "!!!".
 330 One could of course create a more elaborate script that automatically
 331 checked for such errors.  The "rmmod" command forces a "SUCCESS",
 332 "FAILURE", or "RCU_HOTPLUG" indication to be printk()ed.  The first
 333 two are self-explanatory, while the last indicates that while there
 334 were no RCU failures, CPU-hotplug problems were detected.