Commit graph

18 commits

Author SHA1 Message Date
Paul E. McKenney
f3ea978b71 scftorture: Add debug output for wrong-CPU warning
This commit adds the desired CPU, the actual CPU, and nr_cpu_ids to
the wrong-CPU warning in scftorture_invoker(), the better to help with
debugging.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2021-01-04 13:53:41 -08:00
Paul E. McKenney
85558182d5 scftorture: Add full-test stutter capability
In virtual environments on systems with hardware assist, inter-processor
interrupts must do very different things based on whether the target
vCPU is running or not.  This commit therefore enables torture-test
stuttering to better test these running/not-running transitions.

Suggested-by: Chris Mason <clm@fb.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-11-06 17:13:56 -08:00
Paul E. McKenney
1ac78b49d6 scftorture: Add an alternative IPI vector
The scftorture tests currently use only smp_call_function() and
friends, which means that these tests cannot locate bugs caused by
interactions between different IPI vectors.  This commit therefore adds
the rescheduling IPI to the mix.

Note that this commit permits resched_cpus() only when scftorture is
built in.  This is a workaround.  Longer term, this will use real wakeups
rather than resched_cpu().

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-11-06 17:13:49 -08:00
Paul E. McKenney
65bd77f554 scftorture: Add cond_resched() to test loop
Although the test loop does randomly delay, which would provide quiescent
states and so forth, it is possible for there to be a series of long
smp_call_function*() handler runtimes with no delays, which results in
softlockup and RCU CPU stall warning messages.  This commit therefore
inserts a cond_resched() into the main test loop.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:38 -07:00
Paul E. McKenney
9e66bf03f9 scftorture: Adapt memory-ordering test to UP operation
On uniprocessor systems, smp_call_function() does nothing.  This commit
therefore avoids complaining about the lack of handler accesses in the
single-CPU case where there is no handler.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:37 -07:00
Paul E. McKenney
a7c072ef26 scftorture: Block scftorture_invoker() kthreads for offline CPUs
Currently, CPU-hotplug operations might result in all but two
of (say) 100 CPUs being offline, which in turn might result in
false-positive diagnostics due to overload.  This commit therefore
causes scftorture_invoker() kthreads for offline CPUs to loop blocking
for 200 milliseconds at a time, thus continuously adjusting the number
of threads to match the number of online CPUs.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:37 -07:00
Paul E. McKenney
de77d4da54 scftorture: Check unexpected "switch" statement value
This commit adds a "default" case to the switch statement in
scftorture_invoke_one() which contains a WARN_ON_ONCE() and an assignment
to ->scfc_out to suppress knock-on warnings.  These knock-on warnings
could otherwise cause the user to think that there was a memory-ordering
problem in smp_call_function() instead of a bug in scftorture.c itself.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:37 -07:00
Wei Yongjun
9a52a57467 scftorture: Make symbol 'scf_torture_rand' static
The sparse tool complains as follows

kernel/scftorture.c:124:1: warning:
 symbol '__pcpu_scope_scf_torture_rand' was not declared. Should it be static?

And this per-CPU variable is not used outside of scftorture.c,
so this commit marks it static.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:36 -07:00
Paul E. McKenney
ee7035d295 scftorture: Prevent compiler from reducing race probabilities
Detecting smp_call_function() memory misordering requires close timing,
so it is necessary to have the checks immediately before and after
the call to the smp_call_function*() function under test.  This commit
therefore inserts barrier() calls to prevent the compiler from optimizing
memory-misordering detection down into the zone of extreme improbability.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:36 -07:00
Paul E. McKenney
dbf83b655a scftorture: Flag errors in torture-compatible manner
This commit prints error counts on the statistics line and also adds a
"!!!" if any of the counters are non-zero.  Allocation failures are
(somewhat) forgiven, but all other errors result in a "FAILURE" print
at the end of the test.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:35 -07:00
Paul E. McKenney
4df55bddc1 scftorture: Consolidate scftorture_invoke_one() scf_check initialization
This commit hoists much of the initialization of the scf_check
structure out of the switch statement, thus saving a few lines of code.
The initialization of the ->scfc_in field remains in each leg of the
switch statement in order to more heavily stress memory ordering.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:35 -07:00
Paul E. McKenney
676e546964 scftorture: Consolidate scftorture_invoke_one() check and kfree()
This commit moves checking of the ->scfc_out field and the freeing of
the scf_check structure down below the end of switch statement, thus
saving a few lines of code.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:35 -07:00
Paul E. McKenney
34e8c4837a scftorture: Add smp_call_function() memory-ordering checks
This commit adds checks for memory misordering across calls to and
returns from smp_call_function() in the case where the caller waits.
Misordering results in a splat.

Note that in contrast to smp_call_function_single(), this code does not
test memory ordering into the handler in the no-wait case because none
of the handlers would be able to free the scf_check structure without
introducing heavy synchronization to work out which was last.

[ paulmck: s/GFP_KERNEL/GFP_ATOMIC/ per kernel test robot feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:34 -07:00
Paul E. McKenney
980205ee84 scftorture: Add smp_call_function_many() memory-ordering checks
This commit adds checks for memory misordering across calls to and
returns from smp_call_function_many() in the case where the caller waits.
Misordering results in a splat.

Note that in contrast to smp_call_function_single(), this code does not
test memory ordering into the handler in the no-wait case because none
of the handlers would be able to free the scf_check structure without
introducing heavy synchronization to work out which was last.

[ paulmck: s/GFP_KERNEL/GFP_ATOMIC/ per kernel test robot feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:34 -07:00
Paul E. McKenney
b93e21a51e scftorture: Add smp_call_function_single() memory-ordering checks
This commit adds checks for memory misordering across calls to
smp_call_function_single() and also across returns in the case where
the caller waits.  Misordering results in a splat.

[ paulmck: s/GFP_KERNEL/GFP_ATOMIC/ per kernel test robot feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:33 -07:00
Paul E. McKenney
dba3142b37 scftorture: Summarize per-thread statistics
This commit summarizes the per-thread statistics, providing counts of
the number of single, many, and all calls, both no-wait and wait, and,
for the single case, the number where the target CPU was offline.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:33 -07:00
Paul E. McKenney
5022b8ac60 scftorture: Implement weighted primitive selection
This commit uses the scftorture.weight* kernel parameters to randomly
chooses between smp_call_function_single(), smp_call_function_many(),
and smp_call_function().  For each variant, it also randomly chooses
whether to invoke it synchronously (wait=1) or asynchronously (wait=0).
The percentage weighting for each option are dumped to the console log
(search for "scf_sel_dump").

This accumulates statistics, which a later commit will dump out at the
end of the run.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:32 -07:00
Paul E. McKenney
e9d338a0b1 scftorture: Add smp_call_function() torture test
This commit adds an smp_call_function() torture test that repeatedly
invokes this function and complains if things go badly awry.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
2020-08-24 18:38:31 -07:00