Paul E. McKenney [Mon, 29 Jun 2020 19:03:15 +0000 (12:03 -0700)]
Merge branches 'doc.2020.06.29a', 'fixes.2020.06.29a', 'kfree_rcu.2020.06.29a', 'rcu-tasks.2020.06.29a', 'scale.2020.06.29a', 'srcu.2020.06.29a' and 'torture.2020.06.29a' into HEAD
Paul E. McKenney [Wed, 17 Jun 2020 00:07:15 +0000 (17:07 -0700)]
torture: Remove obsolete "cd $KVM"
In the dim distant past, qemu commands needed to be run from the
rcutorture directory, but this is no longer the case. This commit
therefore removes the now-useless "cd $KVM" from the kvm-test-1-run.sh
script.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 16 Jun 2020 23:34:52 +0000 (16:34 -0700)]
torture: Avoid duplicate specification of qemu command
Currently, the qemu command is constructed twice, once to dump it
to the qemu-cmd file and again to execute it. This is of course an
accident waiting to happen, but is done to ensure that the remainder
of the script has an accurate idea of the running qemu command's PID.
This commit therefore places both the qemu command and the PID capture
into a new temporary file and sources that temporary file. Thus the
single construction of the qemu command into the qemu-cmd file suffices
for both purposes.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 16 Jun 2020 22:38:24 +0000 (15:38 -0700)]
torture: Dump ftrace at shutdown only if requested
If there is a large number of torture tests running concurrently,
all of which are dumping large ftrace buffers at shutdown time, the
resulting dumping can take a very long time, particularly on systems
with rotating-rust storage. This commit therefore adds a default-off
torture.ftrace_dump_at_shutdown module parameter that enables
shutdown-time ftrace-buffer dumping.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 16 Jun 2020 21:14:09 +0000 (14:14 -0700)]
torture: Add kvm-tranform.sh script for qemu-cmd files
This commit adds a script that transforms qemu-cmd files to allow them
and the corresponding kernels to be run in contexts other than the one
that they were created for, including on systems other than the one that
they were built on. For example, this allows the build products from a
--buildonly run to be transformed to allow distributed rcutorture testing.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 16 Jun 2020 17:38:57 +0000 (10:38 -0700)]
torture: Improve diagnostic for KCSAN-incapable compilers
Using --kcsan when the compiler does not support KCSAN results in this:
:CONFIG_KCSAN=y: improperly set
:CONFIG_KCSAN_REPORT_ONCE_IN_MS=100000: improperly set
:CONFIG_KCSAN_VERBOSE=y: improperly set
:CONFIG_KCSAN_INTERRUPT_WATCHER=y: improperly set
Clean KCSAN run in /home/git/linux-rcu/tools/testing/selftests/rcutorture/res/2020.06.16-09.53.16
This is a bit obtuse, so this commit adds checks resulting in this:
:CONFIG_KCSAN=y: improperly set
:CONFIG_KCSAN_REPORT_ONCE_IN_MS=100000: improperly set
:CONFIG_KCSAN_VERBOSE=y: improperly set
:CONFIG_KCSAN_INTERRUPT_WATCHER=y: improperly set
Compiler or architecture does not support KCSAN!
Did you forget to switch your compiler with --kmake-arg CC=<cc-that-supports-kcsan>?
Suggested-by: Marco Elver <elver@google.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org> Acked-by: Marco Elver <elver@google.com>
Paul E. McKenney [Tue, 16 Jun 2020 17:02:54 +0000 (10:02 -0700)]
torture: Correctly summarize build-only runs
Currently, kvm-recheck.sh complains that qemu failed for --buildonly
runs, which is sort of true given that qemu can hardly succeed if not
invoked in the first place. Nevertheless, this commit swaps the order
of checks in kvm-recheck.sh so that --buildonly runs will be summarized
more straightforwardly.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Marco Elver [Tue, 16 Jun 2020 09:49:24 +0000 (11:49 +0200)]
torture: Pass --kmake-arg to all make invocations
We need to pass the arguments provided to --kmake-arg to all make
invocations. In particular, the make invocations generating the configs
need to see the final make arguments, e.g. if config variables depend on
particular variables that are passed to make.
For example, when using '--kcsan --kmake-arg CC=clang-11', we would lose
CONFIG_KCSAN=y due to 'make oldconfig' not seeing that we want to use a
compiler that supports KCSAN.
Signed-off-by: Marco Elver <elver@google.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Thu, 11 Jun 2020 23:43:14 +0000 (16:43 -0700)]
rcutorture: Check for unwatched readers
RCU is supposed to be watching all non-idle kernel code and also all
softirq handlers. This commit adds some teeth to this statement by
adding a WARN_ON_ONCE().
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Wed, 10 Jun 2020 21:08:19 +0000 (14:08 -0700)]
torture: Abstract out console-log error detection
This commit pulls the simple pattern-based error detection from the
console log into a new console-badness.sh file. This will enable future
commits to end a run on the first error.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Wed, 10 Jun 2020 00:58:30 +0000 (17:58 -0700)]
torture: Add a stop-run capability
When bisecting RCU issues, it is often the case that the first error in
an unsuccessful run will happen quickly, but that a successful run must
go on for some time in order to obtain a sufficiently low false-negative
error rate. In many cases, a bisection requires multiple concurrent
runs, in which case the first failure in any run indicates failure,
pure and simple. In such cases, it would speed things up greatly if
the first failure terminated all runs.
This commit therefore adds scripting that checks for a file named "STOP"
in the top-level results directory, terminating the run when it appears.
Note that in-progress builds will continue until completion, but future
builds and all runs will be cut short.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Fri, 5 Jun 2020 17:29:28 +0000 (10:29 -0700)]
torture: Create qemu-cmd in --buildonly runs
One reason to do a --buildonly run is to use the build products elsewhere,
for example, to do the actual test on some other system. Part of doing
the test is the actual qemu command, which is not currently produced
by --buildonly runs. This commit therefore causes --buildonly runs to
create this file.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Mon, 4 May 2020 23:41:53 +0000 (16:41 -0700)]
torture: Add --allcpus argument to the kvm.sh script
Leaving off the kvm.sh script's --cpus argument results in the script
testing the scenarios sequentially, which can be quite slow. However,
having to specify the actual number of CPUs can be error-prone.
This commit therefore adds a --allcpus argument that causes kvm.sh to
use all available CPUs.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Mon, 4 May 2020 23:40:53 +0000 (16:40 -0700)]
torture: Remove whitespace from identify_qemu_vcpus output
The identify_qemu_vcpus bash function can return numbers including
whitespace characters, which can be a bit annoying in some bash
dollar-sign substitutions. This commit therefore strips all spaces and
tabs from the value that identify_qemu_vcpus outputs.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Mon, 27 Apr 2020 02:20:37 +0000 (19:20 -0700)]
rcutorture: NULL rcu_torture_current earlier in cleanup code
Currently, the rcu_torture_current variable remains non-NULL until after
all readers have stopped. During this time, rcu_torture_stats_print()
will think that the test is still ongoing, which can result in confusing
dmesg output. This commit therefore NULLs rcu_torture_current immediately
after the rcu_torture_writer() kthread has decided to stop, thus informing
rcu_torture_stats_print() much sooner.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The current console parsing assumes that console lines containing "!!!"
are statistics lines from which it can parse the number of rcutorture
too-short grace-period failures. This prints confusing output for
other problems, including memory exhaustion. This commit therefore
differentiates between these cases and prints an appropriate error string.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Sun, 26 Apr 2020 23:48:46 +0000 (16:48 -0700)]
torture: Set configfile variable to current scenario
The torture-test recheck logic fails to set the configfile variable to
the current scenario, so this commit properly initializes this variable.
This change isn't critical given that all errors for a given scenario
follow that scenario's heading, but it is easier on the eyes to repeat it.
And this repetition also prevents confusion as to whether a given message
goes with the previous heading or the next one.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Fri, 24 Apr 2020 18:21:40 +0000 (11:21 -0700)]
rcutorture: Add races with task-exit processing
Several variants of Linux-kernel RCU interact with task-exit processing,
including preemptible RCU, Tasks RCU, and Tasks Trace RCU. This commit
therefore adds testing of this interaction to rcutorture by adding
rcutorture.read_exit_burst and rcutorture.read_exit_delay kernel-boot
parameters. These kernel parameters control the frequency and spacing
of special read-then-exit kthreads that are spawned.
[ paulmck: Apply feedback from Dan Carpenter's static checker. ]
[ paulmck: Reduce latency to avoid false-positive shutdown hangs. ] Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Zou Wei [Mon, 13 Apr 2020 12:02:59 +0000 (20:02 +0800)]
locktorture: Use true and false to assign to bool variables
This commit fixes the following coccicheck warnings:
kernel/locking/locktorture.c:689:6-10: WARNING: Assignment of 0/1 to bool variable
kernel/locking/locktorture.c:907:2-20: WARNING: Assignment of 0/1 to bool variable
kernel/locking/locktorture.c:938:3-20: WARNING: Assignment of 0/1 to bool variable
kernel/locking/locktorture.c:668:2-19: WARNING: Assignment of 0/1 to bool variable
kernel/locking/locktorture.c:674:2-19: WARNING: Assignment of 0/1 to bool variable
kernel/locking/locktorture.c:634:2-20: WARNING: Assignment of 0/1 to bool variable
kernel/locking/locktorture.c:640:2-20: WARNING: Assignment of 0/1 to bool variable
Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Zou Wei <zou_wei@huawei.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Mon, 4 May 2020 22:55:47 +0000 (15:55 -0700)]
torture: Add script to smoke-test commits in a branch
This commit adds a kvm-check-branches.sh script that takes a list
of commits and commit ranges and runs a short rcutorture test on all
scenarios on each specified commit. A summary is printed at the end, and
the script returns success if all rcutorture runs completed without error.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 28 Apr 2020 21:41:48 +0000 (14:41 -0700)]
torture: Remove qemu dependency on EFI firmware
On some (probably misconfigured) systems, the torture-test scripting
will cause qemu to complain about missing EFI firmware, often because
qemu is trying to traverse broken symbolic links to find that firmware.
Which is a bit silly given that the default torture-test guest OS has
but a single binary for its userspace, and thus is unlikely to do much
in the way of networking in any case.
This commit therefore avoids such problems by specifying "-net none"
to qemu unless the TORTURE_QEMU_INTERACTIVE environment variable is set
(for example, by having specified "--interactive" to kvm.sh), in which
case "-net nic -net user" is specified to qemu instead. Either choice
may be overridden by specifying the "-net" argument of your choice to
the kvm.sh "--qemu-args" parameter.
Sebastian Andrzej Siewior [Tue, 26 May 2020 13:41:34 +0000 (15:41 +0200)]
srcu: Avoid local_irq_save() before acquiring spinlock_t
SRCU disables interrupts to get a stable per-CPU pointer and then
acquires the spinlock which is in the per-CPU data structure. The
release uses spin_unlock_irqrestore(). While this is correct on a non-RT
kernel, this conflicts with the RT semantics because the spinlock is
converted to a 'sleeping' spinlock. Sleeping locks can obviously not be
acquired with interrupts disabled.
Acquire the per-CPU pointer `ssp->sda' without disabling preemption and
then acquire the spinlock_t of the per-CPU data structure. The lock will
ensure that the data is consistent.
The added call to check_init_srcu_struct() is now needed because a
statically defined srcu_struct may remain uninitialized until this
point and the newly introduced locking operation requires an initialized
spinlock_t.
This change was tested for four hours with 8*SRCU-N and 8*SRCU-P without
causing any warnings.
Cc: Lai Jiangshan <jiangshanlai@gmail.com> Cc: "Paul E. McKenney" <paulmck@kernel.org> Cc: Josh Triplett <josh@joshtriplett.org> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Cc: rcu@vger.kernel.org Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Wed, 17 Jun 2020 18:53:53 +0000 (11:53 -0700)]
refperf: Rename refperf.c to refscale.c and change internal names
This commit further avoids conflation of refperf with the kernel's perf
feature by renaming kernel/rcu/refperf.c to kernel/rcu/refscale.c,
and also by similarly renaming the functions and variables inside
this file. This has the side effect of changing the names of the
kernel boot parameters, so kernel-parameters.txt and ver_functions.sh
are also updated.
The rcutorture --torture type remains refperf, and this will be
addressed in a separate commit.
Reported-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Wed, 17 Jun 2020 18:33:54 +0000 (11:33 -0700)]
refperf: Rename RCU_REF_PERF_TEST to RCU_REF_SCALE_TEST
The old Kconfig option name is all too easy to conflate with the
unrelated "perf" feature, so this commit renames RCU_REF_PERF_TEST to
RCU_REF_SCALE_TEST.
Reported-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The synchronize_rcu_tasks_trace() header comment incorrectly claims that
any number of things delimit RCU Tasks Trace read-side critical sections,
when in fact only rcu_read_lock_trace() and rcu_read_unlock_trace() do so.
This commit therefore fixes this comment, and, while in the area, fixes
a typo in the rcu_read_lock_trace() header comment.
Reported-by: Alexei Starovoitov <alexei.starovoitov@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Wed, 3 Jun 2020 18:56:34 +0000 (11:56 -0700)]
refperf: Add test for RCU Tasks readers
This commit adds testing for RCU Tasks readers to the refperf module.
This also applies to RCU Rude readers, as both flavors have empty
(as in non-existent) read-side markers.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Mon, 1 Jun 2020 01:14:57 +0000 (18:14 -0700)]
refperf: Change readdelay module parameter to nanoseconds
The current units of microseconds are too coarse, so this commit
changes the units to nanoseconds. However, ndelay is used only for the
nanoseconds with udelay being used for whole microseconds. For example,
setting refperf.readdelay=1500 results in a udelay(1) followed by an
ndelay(500).
Suggested-by: Akira Yokosawa <akiyks@gmail.com>
[ paulmck: Abstracted delay per Akira feedback and move from 80 to 100 lines. ]
[ paulmck: Fix names as suggested by kbuild test robot. ] Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Fri, 29 May 2020 20:11:26 +0000 (13:11 -0700)]
refperf: Adjust refperf.loop default value
With the various measurement optimizations, 10,000 loops normally
suffices. This commit therefore reduces the refperf.loops default value
from 10,000,000 to 10,000.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Thu, 28 May 2020 23:37:35 +0000 (16:37 -0700)]
refperf: Add read-side delay module parameter
This commit adds a refperf.readdelay module parameter that controls the
duration of each critical section. This parameter allows gathering data
showing how the performance differences between the various primitives
vary with critical-section length.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the reader-launch wait loop from ref_perf_init()
to main_func(), removing one layer of wakeup and allowing slightly
faster system boot.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 26 May 2020 22:30:09 +0000 (15:30 -0700)]
refperf: Output per-experiment data points
Currently, it is necessary to manually edit the console output to see
anything more than statistics, and sometimes the statistics can indicate
outliers that need more investigation. This commit therefore dumps out
the per-experiment measurements, sorted in ascending order, just before
dumping out the statistics.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 26 May 2020 21:26:25 +0000 (14:26 -0700)]
refperf: Label experiment-number column "Runs"
The experiment-number column is currently labeled "Threads", which is
misleading at best. This commit therefore relabels it as "Runs", and
adjusts the scripts accordingly.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 26 May 2020 19:34:57 +0000 (12:34 -0700)]
refperf: Add warmup and cooldown processing phases
This commit causes all the readers to start running unmeasured load
until all readers have done at least one such run (thus having warmed
up), then run the measured load, and then run unmeasured load until all
readers have completed their measured load. This approach avoids any
thread running measured load while other readers are idle.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 26 May 2020 18:40:52 +0000 (11:40 -0700)]
refperf: More closely synchronize reader start times
Currently, readers are awakened individually. On most systems, this
results in significant wakeup delay from one reader to the next, which
can result in the first and last reader having sole access to the
synchronization primitive in question. If that synchronization primitive
involves shared memory, those readers will rack up a huge number of
operations in a very short time, causing large perturbations in the
results.
This commit therefore has the readers busy-wait after being awakened,
and uses a new n_started variable to synchronize their start times.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 26 May 2020 18:22:03 +0000 (11:22 -0700)]
refperf: Convert reader_task structure's "start" field to int
This commit converts the reader_task structure's "start" field to int
in order to demote a full barrier to an smp_load_acquire() and also to
simplify the code a bit. While in the area, and to enlist the compiler's
help in ensuring that nothing was missed, the field's name was changed
to start_reader.
Also while in the area, change the main_func() store to use
smp_store_release() to further fortify against wait/wake races.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 26 May 2020 17:57:34 +0000 (10:57 -0700)]
refperf: Tune reader measurement interval
This commit moves a printk() out of the measurement interval, converts
a atomic_dec()/atomic_read() pair to atomic_dec_and_test(), and adds
a smp_mb__before_atomic() to avoid potential wake/wait hangs. These
changes have the added benefit of reducing the number of loops required
for amortizing loop overhead for CONFIG_PREEMPT=n RCU measurements from
1,000,000 to 10,000. This reduction in turn shortens the test, reducing
the probability of interference.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 26 May 2020 16:32:57 +0000 (09:32 -0700)]
refperf: Make functions static
Because the reset_readers() and process_durations() functions are used
only within kernel/rcu/refperf.c, this commit makes them static.
Reported-by: kbuild test robot <lkp@intel.com> Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, the buffer used to accumulate the thread-summary output is
fixed size, which will cause problems if someone decides to run on a large
number of PCUs. This commit therefore dynamically allocates this buffer.
[ paulmck: Fix memory allocation as suggested by KASAN. ] Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, the buffer used to accumulate the experiment-summary output
is fixed size, which will cause problems if someone decides to run
one hundred experiments. This commit therefore dynamically allocates
this buffer.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Tue, 26 May 2020 00:22:24 +0000 (17:22 -0700)]
refperf: Provide module parameter to specify number of experiments
The current code uses the number of threads both to limit the number
of threads and to specify the number of experiments, but also varies
the number of threads as the experiments progress. This commit takes
a different approach by adding an refperf.nruns module parameter that
specifies the number of experiments, and furthermore uses the same
number of threads for each experiment.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Mon, 25 May 2020 22:31:07 +0000 (15:31 -0700)]
refperf: Allow decimal nanoseconds
The CONFIG_PREEMPT=n rcu_read_lock()/rcu_read_unlock() pair's overhead,
even including loop overhead, is far less than one nanosecond.
Since logscale plots are not all that happy with zero values, provide
picoseconds as decimals.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Mon, 25 May 2020 21:59:06 +0000 (14:59 -0700)]
refperf: Hoist function-pointer calls out of the loop
Current runs show PREEMPT=n rcu_read_lock()/rcu_read_unlock() pairs
consuming between 20 and 30 nanoseconds, when in fact the actual value is
zero, give or take the barrier() asm's effect on compiler optimizations.
The additional overhead is caused by function calls through pointers
(especially in these days of Spectre mitigations) and perhaps also
needless argument passing, a non-const loop limit, and an upcounting loop.
This commit therefore combines the ->readlock() and ->readunlock()
function pointers into a single ->readsection() function pointer that
takes the loop count as a const parameter and keeps any data passed
from the read-lock to the read-unlock internal to this new function.
These changes reduce the measured overhead of the aforementioned
PREEMPT=n rcu_read_lock()/rcu_read_unlock() pairs from between 20 and
30 nanoseconds to somewhere south of 500 picoseconds.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Mon, 25 May 2020 21:16:44 +0000 (14:16 -0700)]
refperf: Add holdoff parameter to allow CPUs to come online
This commit adds an rcuperf module parameter named "holdoff" that
defaults to 10 seconds if refperf is built in and to zero otherwise.
The assumption is that all the CPUs are online by the time that the
modprobe and insmod commands are going to do anything, and that normal
systems will have all the CPUs online within ten seconds.
Larger systems may take many tens of seconds or even minutes to get
to this point, hence this being a module parameter instead of being a
hard-coded constant.
Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Mon, 25 May 2020 16:22:24 +0000 (09:22 -0700)]
rcuperf: Add comments explaining the high reader overhead
This commit adds comments explaining why the readers have otherwise insane
levels of measurement overhead, namely that they are intended as a test
load for update-side performance measurements, not as a straight-up
read-side performance test.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Joel Fernandes (Google) [Mon, 25 May 2020 04:36:48 +0000 (00:36 -0400)]
refperf: Add a test to measure performance of read-side synchronization
Add a test for comparing the performance of RCU with various read-side
synchronization mechanisms. The test has proved useful for collecting
data and performing these comparisons.
Currently RCU, SRCU, reader-writer lock, reader-writer semaphore and
reference counting can be measured using refperf.perf_type parameter.
Each invocation of the test runs measures performance of a specific
mechanism.
The maximum number of CPUs to concurrently run readers on is chosen by
the test itself and is 75% of the total number of CPUs. So if you had 24
CPUs, the test runs with a maximum of 18 parallel readers.
A number of experiments are conducted, and in each experiment, the
number of readers is increased by 1, upto the 75% of CPUs mark. During
each experiment, all readers execute an empty loop with refperf.loops
iterations and time the total loop duration. This is then averaged.
Example output:
Parameters "refperf.perf_type=srcu refperf.loops=2000000" looks like:
Paul E. McKenney [Tue, 9 Jun 2020 16:24:51 +0000 (09:24 -0700)]
rcu-tasks: Fix code-style issues
This commit declares trc_n_readers_need_end and trc_wait static and
replaced a "&" with "&&". The "&" happened to work because the values
are bool, but accidents waiting to happen and all that...
Reported-by: kbuild test robot <lkp@intel.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The show_rcu_tasks_gp_kthreads() function is not invoked by Tiny RCU,
but is nevertheless defined in Tiny RCU builds that enable Tasks Trace
RCU. This commit therefore conditionally compiles this function so
that it is defined only in builds that actually use it.
Reported-by: kbuild test robot <lkp@intel.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Fri, 29 May 2020 02:33:47 +0000 (19:33 -0700)]
rcu-tasks: Add #include of rcupdate_trace.h to update.c
Although this is in some strict sense unnecessary, it is good to allow
the compiler to compare the function declaration with its definition.
This commit therefore adds a #include of linux/rcupdate_trace.h to
kernel/rcu/update.c.
Reported-by: kbuild test robot <lkp@intel.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Thu, 7 May 2020 23:47:13 +0000 (16:47 -0700)]
rcu-tasks: Convert sleeps to idle priority
This commit converts the long-standing schedule_timeout_interruptible()
and schedule_timeout_uninterruptible() calls used by the various Tasks
RCU's grace-period kthreads to schedule_timeout_idle(). This conversion
avoids polluting the load-average with Tasks-RCU-related sleeping.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
lib/test_vmalloc.c: Add test cases for kvfree_rcu()
Introduce four new test cases for testing the kvfree_rcu()
interface. Two of them belong to single argument functionality
and another two for 2-argument functionality.
The aim is to stress and check how kvfree_rcu() behaves under
different load and memory conditions and analyze its performance
throughput.
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
rcu: Introduce single argument kvfree_rcu() interface
Make kvfree_rcu() capable of freeing objects that will not
embed an rcu_head within it. This saves storage overhead in
such objects. Reclaiming headless objects this way requires
only a single argument (pointer to the object).
After this patch, there are two ways to use kvfree_rcu():
a) kvfree_rcu(ptr, rhf);
struct X {
struct rcu_head rhf;
unsigned char data[100];
};
void *ptr = kvmalloc(sizeof(struct X), GFP_KERNEL);
if (ptr)
kvfree_rcu(ptr, rhf);
b) kvfree_rcu(ptr);
void *ptr = kvmalloc(some_bytes, GFP_KERNEL);
if (ptr)
kvfree_rcu(ptr);
Note that the headless usage (example b) can only be used in a code
that can sleep. This is enforced by the CONFIG_DEBUG_ATOMIC_SLEEP
option.
Co-developed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Update the kvfree_call_rcu() function with head-less support.
This allows RCU to reclaim objects without an embedded rcu_head.
tree-RCU:
We introduce two chains of arrays to store SLAB-backed and vmalloc
pointers, each. Storage in either of these arrays does not require
embedding an rcu_head within the object.
Maintaining the arrays may become impossible due to high memory
pressure. For such cases there is an emergency path. Objects with
rcu_head inside are just queued on a backup rcu_head list. Later on
that list is drained. As for the head-less variant, as the current
context can sleep, the following emergency measures are applied:
a) Synchronously wait until a grace period has elapsed.
b) Call kvfree().
tiny-RCU:
For double argument calls, there are no new changes in behavior. For
single argument call, kvfree() is directly inlined on the current
stack after a synchronize_rcu() call. Note that for tiny-RCU, any
call to synchronize_rcu() is actually a quiescent state, therefore
it does nothing.
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Co-developed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
kvmalloc() can allocate two types of objects: SLAB backed
and vmalloc backed. How it behaves depends on requested
object's size and memory pressure.
Add a kvfree_rcu() interface that can free memory allocated
via kvmalloc(). It is a simple alias to kfree_rcu() which
can now handle either type of object.
p = kvmalloc(10 * PAGE_SIZE);
if (p)
kvfree_rcu(p, rcu);
<snip>
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Co-developed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
mm/list_lru.c: Rename kvfree_rcu() to local variant
Rename kvfree_rcu() function to the kvfree_rcu_local() one.
The purpose is to prevent a conflict of two same function
declarations. The kvfree_rcu() will be globally visible
what would lead to a build error. No functional change.
Cc: linux-mm@kvack.org Cc: rcu@vger.kernel.org Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
1. Rename rcu_invoke_kfree_callback() to rcu_invoke_kvfree_callback(),
as well as the associated trace events, so the rcu_kfree_callback(),
becomes rcu_kvfree_callback(). The reason is to be aligned with kvfree()
notation.
2. Rename __is_kfree_rcu_offset to __is_kvfree_rcu_offset. All RCU
paths use kvfree() now instead of kfree(), thus rename it.
3. Rename kfree_call_rcu() to the kvfree_call_rcu(). The reason is,
it is capable of freeing vmalloc() memory now. Do the same with
__kfree_rcu() macro, it becomes __kvfree_rcu(), the goal is the
same.
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Co-developed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Replace kfree() with kvfree() in rcu_reclaim_tiny().
This makes it possible to release either SLAB or vmalloc
objects after a GP.
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
rcu/tree: Maintain separate array for vmalloc ptrs
To do so, we use an array of kvfree_rcu_bulk_data structures.
It consists of two elements:
- index number 0 corresponds to slab pointers.
- index number 1 corresponds to vmalloc pointers.
Keeping vmalloc pointers separated from slab pointers makes
it possible to invoke the right freeing API for the right
kind of pointer.
It also prepares us for future headless support for vmalloc
and SLAB objects. Such objects cannot be queued on a linked
list and are instead directly into an array.
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Co-developed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
In order to reduce the dynamic need for pages in kfree_rcu(),
pre-allocate a configurable number of pages per CPU and link
them in a list. When kfree_rcu() reclaims objects, the object's
container page is cached into a list instead of being released
to the low-level page allocator.
Such an approach provides O(1) access to free pages while also
reducing the number of requests to the page allocator. It also
makes the kfree_rcu() code to have free pages available during
a low memory condition.
A read-only sysfs parameter (rcu_min_cached_objs) reflects the
minimum number of allowed cached pages per CPU.
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Sebastian Andrzej Siewior [Mon, 25 May 2020 21:47:51 +0000 (23:47 +0200)]
rcu/tree: Use static initializer for krc.lock
The per-CPU variable is initialized at runtime in
kfree_rcu_batch_init(). This function is invoked before
'rcu_scheduler_active' is set to 'RCU_SCHEDULER_RUNNING'.
After the initialisation, '->initialized' is to true.
The raw_spin_lock is only acquired if '->initialized' is
set to true. The worqueue item is only used if 'rcu_scheduler_active'
set to RCU_SCHEDULER_RUNNING which happens after initialisation.
Use a static initializer for krc.lock and remove the runtime
initialisation of the lock. Since the lock can now be always
acquired, remove the '->initialized' check.
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
rcu/tree: Move kfree_rcu_cpu locking/unlocking to separate functions
Introduce helpers to lock and unlock per-cpu "kfree_rcu_cpu"
structures. That will make kfree_call_rcu() more readable
and prevent programming errors.
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Joel Fernandes (Google) [Mon, 25 May 2020 21:47:48 +0000 (23:47 +0200)]
rcu/tree: Make debug_objects logic independent of rcu_head
kfree_rcu()'s debug_objects logic uses the address of the object's
embedded rcu_head to queue/unqueue. Instead of this, make use of the
object's address itself as preparation for future headless kfree_rcu()
support.
Reviewed-by: Uladzislau Rezki <urezki@gmail.com> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
rcu/tree: Repeat the monitor if any free channel is busy
It is possible that one of the channels cannot be detached
because its free channel is busy and previously queued data
has not been processed yet. On the other hand, another
channel can be successfully detached causing the monitor
work to stop.
Prevent that by rescheduling the monitor work if there are
any channels in the pending state after a detach attempt.
Fixes: 34c881745549e ("rcu: Support kfree_bulk() interface in kfree_rcu()") Acked-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Joel Fernandes (Google) [Mon, 25 May 2020 21:47:46 +0000 (23:47 +0200)]
rcu/tree: Skip entry into the page allocator for PREEMPT_RT
To keep the kfree_rcu() code working in purely atomic sections on RT,
such as non-threaded IRQ handlers and raw spinlock sections, avoid
calling into the page allocator which uses sleeping locks on RT.
In fact, even if the caller is preemptible, the kfree_rcu() code is
not, as the krcp->lock is a raw spinlock.
Calling into the page allocator is optional and avoiding it should be
Ok, especially with the page pre-allocation support in future patches.
Such pre-allocation would further avoid the a need for a dynamically
allocated page in the first place.
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Reviewed-by: Uladzislau Rezki <urezki@gmail.com> Co-developed-by: Uladzislau Rezki <urezki@gmail.com> Signed-off-by: Uladzislau Rezki <urezki@gmail.com> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Joel Fernandes (Google) [Mon, 25 May 2020 21:47:45 +0000 (23:47 +0200)]
rcu/tree: Keep kfree_rcu() awake during lock contention
On PREEMPT_RT kernels, the krcp spinlock gets converted to an rt-mutex
and causes kfree_rcu() callers to sleep. This makes it unusable for
callers in purely atomic sections such as non-threaded IRQ handlers and
raw spinlock sections. Fix it by converting the spinlock to a raw
spinlock.
Vetting all code paths, there is no reason to believe that the raw
spinlock will hurt RT latencies as it is not held for a long time.
Cc: bigeasy@linutronix.de Cc: Uladzislau Rezki <urezki@gmail.com> Reviewed-by: Uladzislau Rezki <urezki@gmail.com> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The current code provokes some kernel-doc warnings:
./kernel/rcu/tree.c:2915: warning: Function parameter or member 'count' not described in 'kfree_rcu_cpu'
./include/linux/rculist.h:517: warning: bad line: [@right ][node2 ... ]
./include/linux/rculist.h:2: WARNING: Unexpected indentation.
This commit therefore moves the comment for "count" to the kernel-doc
markup and adds a missing "*" on one kernel-doc continuation line.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Randy Dunlap [Mon, 15 Jun 2020 19:28:05 +0000 (12:28 -0700)]
kernel/rcu/tree.c: Fix kernel-doc warnings
Fix kernel-doc warning:
../kernel/rcu/tree.c:959: warning: Excess function parameter 'irq' description in 'rcu_nmi_enter'
Fixes: cf7614e13c8f ("rcu: Refactor rcu_{nmi,irq}_{enter,exit}()") Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Cc: Byungchul Park <byungchul.park@lge.com> Cc: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Wei Yang [Fri, 12 Jun 2020 02:07:54 +0000 (10:07 +0800)]
rcu: grpnum just records group number
The ->grpnum field in the rcu_node structure contains the bit position
in this structure's parent's bitmasks, which is not the CPU number.
This commit therefore adjusts this field's comment accordingly.
Signed-off-by: Wei Yang <richard.weiyang@linux.alibaba.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Wei Yang [Fri, 12 Jun 2020 02:07:53 +0000 (10:07 +0800)]
rcu: grplo/grphi just records CPU number
The ->grplo and ->grphi fields store the lowest and highest CPU number
covered by to a rcu_node structure, which is not the group number.
This commit therefore adjusts these fields' comments to match reality.
Signed-off-by: Wei Yang <richard.weiyang@linux.alibaba.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Wei Yang [Fri, 12 Jun 2020 02:07:52 +0000 (10:07 +0800)]
rcu: gp_max is protected by root rcu_node's lock
Because gp_max is protected by root rcu_node's lock, this commit moves
the gp_max definition to the region of the rcu_node structure containing
fields protected by this lock.
Signed-off-by: Wei Yang <richard.weiyang@linux.alibaba.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Peter Enderborg [Thu, 4 Jun 2020 10:23:20 +0000 (12:23 +0200)]
rcu: Stop shrinker loop
The count and scan can be separated in time, and there is a fair chance
that all work is already done when the scan starts, which might in turn
result in a needless retry. This commit therefore avoids this retry by
returning SHRINK_STOP.
Reviewed-by: Uladzislau Rezki (Sony) <urezki@gmail.com> Signed-off-by: Peter Enderborg <peter.enderborg@sony.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Thu, 28 May 2020 15:49:29 +0000 (08:49 -0700)]
lockdep: Complain only once about RCU in extended quiescent state
Currently, lockdep_rcu_suspicious() complains twice about RCU read-side
critical sections being invoked from within extended quiescent states,
for example:
RCU used illegally from idle CPU!
rcu_scheduler_active = 2, debug_locks = 1
RCU used illegally from extended quiescent state!
This commit therefore saves a couple lines of code and one line of
console-log output by eliminating the first of these two complaints.
Paul E. McKenney [Wed, 20 May 2020 00:00:54 +0000 (17:00 -0700)]
rcu: Mark rcu_nmi_enter() call to rcu_cleanup_after_idle() noinstr
The objtool complains about the call to rcu_cleanup_after_idle() from
rcu_nmi_enter(), so this commit adds instrumentation_begin() before that
call and instrumentation_end() after it.
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Frederic Weisbecker [Fri, 15 May 2020 00:34:29 +0000 (02:34 +0200)]
tick/nohz: Narrow down noise while setting current task's tick dependency
Setting a tick dependency on any task, including the case where a task
sets that dependency on itself, triggers an IPI to all CPUs. That is
of course suboptimal but it had previously not been an issue because it
was only used by POSIX CPU timers on nohz_full, which apparently never
occurs in latency-sensitive workloads in production. (Or users of such
systems are suffering in silence on the one hand or venting their ire
on the wrong people on the other.)
But RCU now sets a task tick dependency on the current task in order
to fix stall issues that can occur during RCU callback processing.
Thus, RCU callback processing triggers frequent system-wide IPIs from
nohz_full CPUs. This is quite counter-productive, after all, avoiding
IPIs is what nohz_full is supposed to be all about.
This commit therefore optimizes tasks' self-setting of a task tick
dependency by using tick_nohz_full_kick() to avoid the system-wide IPI.
Instead, only the execution of the one task is disturbed, which is
acceptable given that this disturbance is well down into the noise
compared to the degree to which the RCU callback processing itself
disturbs execution.
Fixes: 6a949b7af82d (rcu: Force on tick when invoking lots of callbacks) Reported-by: Matt Fleming <matt@codeblueprint.co.uk> Signed-off-by: Frederic Weisbecker <frederic@kernel.org> Cc: stable@kernel.org Cc: Paul E. McKenney <paulmck@kernel.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ingo Molnar <mingo@kernel.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
On a PREEMPT=n kernel, the try_release_extent_mapping() function's
"while" loop might run for a very long time on a large I/O. This commit
therefore adds a cond_resched() to this loop, providing RCU any needed
quiescent states.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Thu, 7 May 2020 23:38:29 +0000 (16:38 -0700)]
rcu: Expedited grace-period sleeps to idle priority
This commit converts the schedule_timeout_uninterruptible() call used
by RCU's expedited grace-period processing to schedule_timeout_idle().
This conversion avoids polluting the load-average with RCU-related
sleeping.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Thu, 7 May 2020 23:36:10 +0000 (16:36 -0700)]
rcu: No-CBs-related sleeps to idle priority
This commit converts the schedule_timeout_interruptible() call used by
RCU's no-CBs grace-period kthreads to schedule_timeout_idle(). This
conversion avoids polluting the load-average with RCU-related sleeping.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Thu, 7 May 2020 23:34:38 +0000 (16:34 -0700)]
rcu: Priority-boost-related sleeps to idle priority
This commit converts the long-standing schedule_timeout_interruptible()
call used by RCU's priority-boosting kthreads to schedule_timeout_idle().
This conversion avoids polluting the load-average with RCU-related
sleeping.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Thu, 7 May 2020 22:44:46 +0000 (15:44 -0700)]
rcu: Grace-period-kthread related sleeps to idle priority
This commit converts the long-standing schedule_timeout_interruptible()
and schedule_timeout_uninterruptible() calls used by RCU's grace-period
kthread to schedule_timeout_idle(). This conversion avoids polluting
the load-average with RCU-related sleeping.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Madhuparna Bhowmik [Mon, 4 May 2020 12:05:05 +0000 (08:05 -0400)]
trace: events: rcu: Change description of rcu_dyntick trace event
The different strings used for describing the polarity are
Start, End and StillNonIdle. Since StillIdle is not used in any trace
point for rcu_dyntick, it can be removed and StillNonIdle can be added
in the description. Because StillNonIdle is used in a few tracepoints for
rcu_dyntick.
Similarly, USER, IDLE and IRQ are used for describing context in
the rcu_dyntick tracepoints. Since, "KERNEL" is not used for any
of the rcu_dyntick tracepoints, remove it from the description.
Signed-off-by: Madhuparna Bhowmik <madhuparnabhowmik10@gmail.com> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The rcu_callback_map lockdep_map structure was added back in 2013, but
its purpose has become obscure. This commit therefore documments that the
purpose of rcu_callback map is, in the words of commit 24ef659a857 ("rcu:
Provide better diagnostics for blocking in RCU callback functions"),
to help lockdep to tie an "inappropriate voluntary context switch back
to the fact that the function is being invoked from within a callback."
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Paul E. McKenney [Fri, 1 May 2020 23:49:48 +0000 (16:49 -0700)]
rcu: Add callbacks-invoked counters
This commit adds a count of the callbacks invoked to the per-CPU rcu_data
structure. This count is printed by the show_rcu_gp_kthreads() that
is invoked by rcutorture and the RCU CPU stall-warning code. It is also
intended for use by drgn.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Wei Yang [Sun, 19 Apr 2020 21:57:15 +0000 (21:57 +0000)]
rcu: Simplify the calculation of rcu_state.ncpus
There is only 1 bit set in mask, which means that the only difference
between oldmask and the new one will be at the position where the bit is
set in mask. This commit therefore updates rcu_state.ncpus by checking
whether the bit in mask is already set in rnp->expmaskinitnext.
Signed-off-by: Wei Yang <richard.weiyang@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
And on a PREEMPT=n kernel, the "while (vma)" loop in exit_mmap() can run
for a very long time given a large process. This commit therefore adds
a cond_resched() to this loop, providing RCU any needed quiescent states.
Cc: Andrew Morton <akpm@linux-foundation.org> Cc: <linux-mm@kvack.org> Reviewed-by: Shakeel Butt <shakeelb@google.com> Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Wei Yang [Wed, 15 Apr 2020 22:26:55 +0000 (22:26 +0000)]
rcu: Initialize and destroy rcu_synchronize only when necessary
The __wait_rcu_gp() function unconditionally initializes and cleans up
each element of rs_array[], whether used or not. This is slightly
wasteful and rather confusing, so this commit skips both initialization
and cleanup for duplicate callback functions.
Signed-off-by: Wei Yang <richard.weiyang@gmail.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
projects / linux.git / log