Both the CONFIG_TASKS_RCU and CONFIG_TASKS_RUDE_RCU options
are broken when RCU_TINY is enabled as well, as some functions
are missing a declaration.
In file included from kernel/rcu/update.c:649:
kernel/rcu/tasks.h:1271:21: error: no previous prototype for 'get_rcu_tasks_rude_gp_kthread' [-Werror=missing-prototypes]
1271 | struct task_struct *get_rcu_tasks_rude_gp_kthread(void)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
kernel/rcu/rcuscale.c:330:27: error: 'get_rcu_tasks_rude_gp_kthread' undeclared here (not in a function); did you mean 'get_rcu_tasks_trace_gp_kthread'?
330 | .rso_gp_kthread = get_rcu_tasks_rude_gp_kthread,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
| get_rcu_tasks_trace_gp_kthread
In file included from /home/arnd/arm-soc/kernel/rcu/update.c:649:
kernel/rcu/tasks.h:1113:21: error: no previous prototype for 'get_rcu_tasks_gp_kthread' [-Werror=missing-prototypes]
1113 | struct task_struct *get_rcu_tasks_gp_kthread(void)
| ^~~~~~~~~~~~~~~~~~~~~~~~
Also, building with CONFIG_TASKS_RUDE_RCU but not CONFIG_TASKS_RCU is
broken because of some missing stub functions:
kernel/rcu/rcuscale.c:322:27: error: 'tasks_scale_read_lock' undeclared here (not in a function); did you mean 'srcu_scale_read_lock'?
322 | .readlock = tasks_scale_read_lock,
| ^~~~~~~~~~~~~~~~~~~~~
| srcu_scale_read_lock
kernel/rcu/rcuscale.c:323:27: error: 'tasks_scale_read_unlock' undeclared here (not in a function); did you mean 'srcu_scale_read_unlock'?
323 | .readunlock = tasks_scale_read_unlock,
| ^~~~~~~~~~~~~~~~~~~~~~~
| srcu_scale_read_unlock
Move the declarations outside of the RCU_TINY #ifdef and duplicate the
shared stub functions to address all of the above.
Fixes: 88d7ff818f0ce ("rcuscale: Add RCU Tasks Rude testing")
Fixes: 755f1c5eb416b ("rcuscale: Measure RCU Tasks Trace grace-period kthread CPU time")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit causes RCU Tasks Trace to output the CPU time consumed by
its grace-period kthread. The CPU time is whatever is in the designated
task's current->stime field, and thus is controlled by whatever CPU-time
accounting scheme is in effect.
This output appears in microseconds as follows on the console:
rcu_scale: Grace-period kthread CPU time: 42367.037
[ paulmck: Apply Willy Tarreau feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit adds the ability to output the CPU time consumed by the
grace-period kthread for the RCU variant under test. The CPU time is
whatever is in the designated task's current->stime field, and thus is
controlled by whatever CPU-time accounting scheme is in effect.
This output appears in microseconds as follows on the console:
rcu_scale: Grace-period kthread CPU time: 42367.037
[ paulmck: Apply feedback from Stephen Rothwell and kernel test robot. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Tested-by: Yujie Liu <yujie.liu@intel.com>
By default, rcuscale collects only 100 points of data per writer, but
arranging for all kthreads to be actively collecting (if not recording)
data during the time that any kthread might be recording. This works
well, but does not allow much time to bring external performance tools
to bear. This commit therefore adds a minruntime module parameter
that specifies a minimum data-collection interval in seconds.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Some workloads do isolated RCU work, and this can affect operation
latencies. This commit therefore adds a writer_holdoff_jiffies module
parameter that causes writers to block for the specified number of
jiffies between each pair of consecutive write-side operations.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit adds a "jiffies" test to refscale, allowing use of jiffies
to be compared to ktime_get_real_fast_ns(). On my x86 laptop, jiffies
is more than 20x faster. (Though for many uses, the tens-of-nanoseconds
overhead of ktime_get_real_fast_ns() will be just fine.)
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Running the refscale test occasionally crashes the kernel with the
following error:
[ 8569.952896] BUG: unable to handle page fault for address: ffffffffffffffe8
[ 8569.952900] #PF: supervisor read access in kernel mode
[ 8569.952902] #PF: error_code(0x0000) - not-present page
[ 8569.952904] PGD c4b048067 P4D c4b049067 PUD c4b04b067 PMD 0
[ 8569.952910] Oops: 0000 [#1] PREEMPT_RT SMP NOPTI
[ 8569.952916] Hardware name: Dell Inc. PowerEdge R750/0WMWCR, BIOS 1.2.4 05/28/2021
[ 8569.952917] RIP: 0010:prepare_to_wait_event+0x101/0x190
:
[ 8569.952940] Call Trace:
[ 8569.952941] <TASK>
[ 8569.952944] ref_scale_reader+0x380/0x4a0 [refscale]
[ 8569.952959] kthread+0x10e/0x130
[ 8569.952966] ret_from_fork+0x1f/0x30
[ 8569.952973] </TASK>
The likely cause is that init_waitqueue_head() is called after the call to
the torture_create_kthread() function that creates the ref_scale_reader
kthread. Although this init_waitqueue_head() call will very likely
complete before this kthread is created and starts running, it is
possible that the calling kthread will be delayed between the calls to
torture_create_kthread() and init_waitqueue_head(). In this case, the
new kthread will use the waitqueue head before it is properly initialized,
which is not good for the kernel's health and well-being.
The above crash happened here:
static inline void __add_wait_queue(...)
{
:
if (!(wq->flags & WQ_FLAG_PRIORITY)) <=== Crash here
The offset of flags from list_head entry in wait_queue_entry is
-0x18. If reader_tasks[i].wq.head.next is NULL as allocated reader_task
structure is zero initialized, the instruction will try to access address
0xffffffffffffffe8, which is exactly the fault address listed above.
This commit therefore invokes init_waitqueue_head() before creating
the kthread.
Fixes: 653ed64b01 ("refperf: Add a test to measure performance of read-side synchronization")
Signed-off-by: Waiman Long <longman@redhat.com>
Reviewed-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Reviewed-by: Davidlohr Bueso <dave@stgolabs.net>
Acked-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The various RCU Tasks flavors now do lazy grace periods when there are
only asynchronous grace period requests. By default, the system will let
250 milliseconds elapse after the first call_rcu_tasks*() callbacki is
queued before starting a grace period. In contrast, synchronous grace
period requests such as synchronize_rcu_tasks*() will start a grace
period immediately.
However, invoking one of the call_rcu_tasks*() functions in a too-tight
loop can result in a callback flood, which in turn can exhaust memory
if grace periods are delayed for too long.
This commit therefore sets a limit so that the grace-period kthread
will be awakened when any CPU's callback list expands to contain
rcupdate.rcu_task_lazy_lim callbacks elements (defaulting to 32, set to -1
to disable), the grace-period kthread will be awakened, thus cancelling
any ongoing laziness and getting out in front of the potential callback
flood.
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit adds kernel boot parameters for callback laziness, allowing
the RCU Tasks flavors to be individually adjusted.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The performance requirements on RCU Tasks, and in particular on RCU
Tasks Trace, have evolved over time as the workloads have evolved.
The current implementation is designed to provide low grace-period
latencies, and also to accommodate short-duration floods of callbacks.
However, current workloads can also provide a constant background
callback-queuing rate of a few hundred call_rcu_tasks_trace() invocations
per second. This results in continuous back-to-back RCU Tasks Trace
grace periods, which in turn can consume the better part of 10% of a CPU.
One could take the attitude that there are several tens of other CPUs on
the systems running such workloads, but energy efficiency is a thing.
On these systems, although asynchronous grace-period requests happen
every few milliseconds, synchronous grace-period requests are quite rare.
This commit therefore arrranges for grace periods to be initiated
immediately in response to calls to synchronize_rcu_tasks*() and
also to calls to synchronize_rcu_mult() that are passed one of the
call_rcu_tasks*() functions. These are recognized by the tell-tale
wakeme_after_rcu callback function.
In other cases, callbacks are gathered up for up to about 250 milliseconds
before a grace period is initiated. This results in more than an order of
magnitude reduction in RCU Tasks Trace grace periods, with corresponding
reduction in consumption of CPU time.
Reported-by: Alexei Starovoitov <ast@kernel.org>
Reported-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
If a CPU is executing a long series of non-sleeping system calls,
RCU grace periods can be delayed for on the order of a couple hundred
milliseconds. This is normally not a problem, but if each system call
does a call_rcu(), those callbacks can stack up. RCU will eventually
notice this callback storm, but use of rcu_request_urgent_qs_task()
allows the code invoking call_rcu() to give RCU a heads up.
This function is not for general use, not yet, anyway.
Reported-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20230706033447.54696-11-alexei.starovoitov@gmail.com
This code-movement-only commit moves the rcu_scale_cleanup() and
rcu_scale_shutdown() functions to follow kfree_scale_cleanup().
This is code movement is in preparation for a bug-fix patch that invokes
kfree_scale_cleanup() from rcu_scale_cleanup().
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Callbacks can only be queued as lazy on NOCB CPUs, therefore iterating
over the NOCB mask is enough for both counting and scanning. Just lock
the mostly uncontended barrier mutex on counting as well in order to
keep rcu_nocb_mask stable.
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The rcu_tasks_invoke_cbs() function relies on queue_work_on() to silently
fall back to WORK_CPU_UNBOUND when the specified CPU is offline. However,
the queue_work_on() function's silent fallback mechanism relies on that
CPU having been online at some time in the past. When queue_work_on()
is passed a CPU that has never been online, workqueue lockups ensue,
which can be bad for your kernel's general health and well-being.
This commit therefore checks whether a given CPU has ever been online,
and, if not substitutes WORK_CPU_UNBOUND in the subsequent call to
queue_work_on(). Why not simply omit the queue_work_on() call entirely?
Because this function is flooding callback-invocation notifications
to all CPUs, and must deal with possibilities that include a sparse
cpu_possible_mask.
This commit also moves the setting of the rcu_data structure's
->beenonline field to rcu_cpu_starting(), which executes on the
incoming CPU before that CPU has ever enabled interrupts. This ensures
that the required workqueues are present. In addition, because the
incoming CPU has not yet enabled its interrupts, there cannot yet have
been any softirq handlers running on this CPU, which means that the
WARN_ON_ONCE(!rdp->beenonline) within the RCU_SOFTIRQ handler cannot
have triggered yet.
Fixes: d363f833c6 ("rcu-tasks: Use workqueues for multiple rcu_tasks_invoke_cbs() invocations")
Reported-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, rcu_cpu_starting() is written so that it might be invoked
with interrupts enabled. However, it is always called when interrupts
are disabled, either by rcu_init(), notify_cpu_starting(), or from a
call point prior to the call to notify_cpu_starting().
But why bother requiring that interrupts be disabled? The purpose is
to allow the rcu_data structure's ->beenonline flag to be set after all
early processing has completed for the incoming CPU, thus allowing this
flag to be used to determine when workqueues have been set up for the
incoming CPU, while still allowing this flag to be used as a diagnostic
within rcu_core().
This commit therefore makes rcu_cpu_starting() rely on interrupts being
disabled.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The rcu_data structure's ->rcu_cpu_has_work field can be modified by
any CPU attempting to wake up the rcuc kthread. Therefore, this commit
marks accesses to this field from the rcu_cpu_kthread() function.
This data race was reported by KCSAN. Not appropriate for backporting
due to failure being unlikely.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The per-CPU rcu_data structure's ->cpu_no_qs.b.exp field is updated
only on the instance corresponding to the current CPU, but can be read
more widely. Unmarked accesses are OK from the corresponding CPU, but
only if interrupts are disabled, given that interrupt handlers can and
do modify this field.
Unfortunately, although the load from rcu_preempt_deferred_qs() is always
carried out from the corresponding CPU, interrupts are not necessarily
disabled. This commit therefore upgrades this load to READ_ONCE.
Similarly, the diagnostic access from synchronize_rcu_expedited_wait()
might run with interrupts disabled and from some other CPU. This commit
therefore marks this load with data_race().
Finally, the C-language access in rcu_preempt_ctxt_queue() is OK as
is because interrupts are disabled and this load is always from the
corresponding CPU. This commit adds a comment giving the rationale for
this access being safe.
This data race was reported by KCSAN. Not appropriate for backporting
due to failure being unlikely.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, if there are more than 100 ready-to-invoke RCU callbacks queued
on a given CPU, the rcu_do_batch() function sets a timeout for invocation
of the series. This timeout defaulting to three milliseconds, and may
be adjusted using the rcutree.rcu_resched_ns kernel boot parameter.
This timeout is checked using local_clock(), but the overhead of this
function combined with the common-case very small callback-invocation
overhead means that local_clock() is checked every 32nd invocation.
This works well except for longer-than average callbacks. For example,
a series of 500-microsecond-duration callbacks means that local_clock()
is checked only once every 16 milliseconds, which makes it difficult to
enforce a three-millisecond timeout.
This commit therefore adds a Kconfig option RCU_DOUBLE_CHECK_CB_TIME
that enables backup timeout checking using the coarser grained but
lighter weight jiffies. If the jiffies counter detects a timeout,
then local_clock() is consulted even if this is not the 32nd callback.
This prevents the aforementioned 16-millisecond latency blow.
Reported-by: Domas Mituzas <dmituzas@meta.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, a callback-invocation time limit is enforced only for
callbacks invoked from the softirq environment, the rationale being
that when callbacks are instead invoked from rcuc and rcuoc kthreads,
these callbacks cannot be holding up other softirq vectors.
Which is in fact true. However, if an rcuc kthread spends too much time
invoking callbacks, it can delay quiescent-state reports from its CPU,
which can also be a problem.
This commit therefore applies the callback-invocation time limit to
callback invocation from the rcuc kthreads as well as from softirq.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit uses rtp->name instead of __func__ and outputs the value
of rcu_task_cb_adjust, thus reducing console-log output.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The ->lazy_len is only checked locklessly. Recheck again under the
->nocb_lock to avoid spending more time on flushing/waking if not
necessary. The ->lazy_len can still increment concurrently (from 1 to
infinity) but under the ->nocb_lock we at least know for sure if there
are lazy callbacks at all (->lazy_len > 0).
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The shrinker resets the lazy callbacks counter in order to trigger the
pending lazy queue flush though the rcuog kthread. The counter reset is
protected by the ->nocb_lock against concurrent accesses...except
for one of them. Here is a list of existing synchronized readers/writer:
1) The first lazy enqueuer (incrementing ->lazy_len to 1) does so under
->nocb_lock and ->nocb_bypass_lock.
2) The further lazy enqueuers (incrementing ->lazy_len above 1) do so
under ->nocb_bypass_lock _only_.
3) The lazy flush checks and resets to 0 under ->nocb_lock and
->nocb_bypass_lock.
The shrinker protects its ->lazy_len reset against cases 1) and 3) but
not against 2). As such, setting ->lazy_len to 0 under the ->nocb_lock
may be cancelled right away by an overwrite from an enqueuer, leading
rcuog to ignore the flush.
To avoid that, use the proper bypass flush API which takes care of all
those details.
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The shrinker may run concurrently with callbacks (de-)offloading. As
such, calling rcu_nocb_lock() is very dangerous because it does a
conditional locking. The worst outcome is that rcu_nocb_lock() doesn't
lock but rcu_nocb_unlock() eventually unlocks, or the reverse, creating
an imbalance.
Fix this with protecting against (de-)offloading using the barrier mutex.
Although if the barrier mutex is contended, which should be rare, then
step aside so as not to trigger a mutex VS allocation
dependency chain.
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
If the rcutree.rcu_min_cached_objs kernel boot parameter is set to zero,
then krcp->page_cache_work will never be triggered to fill page cache.
In addition, the put_cached_bnode() will not fill page cache. As a
result krcp->bkvcache will always be empty, so there is no need to acquire
krcp->lock to get page from krcp->bkvcache. This commit therefore makes
drain_page_cache() return immediately if the rcu_min_cached_objs is zero.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Reviewed-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
When the fill_page_cache_func() function is invoked, it assumes that
the cache of pages is completely empty. However, there can be some time
between triggering execution of this function and its actual invocation.
During this time, kfree_rcu_work() might run, and might fill in part or
all of this cache of pages, thus invalidating the fill_page_cache_func()
function's assumption.
This will not overfill the cache because put_cached_bnode() will reject
the extra page. However, it will result in a needless allocation and
freeing of one extra page, which might not be helpful under lowish-memory
conditions.
This commit therefore causes the fill_page_cache_func() to explicitly
account for pages that have been placed into the cache shortly before
it starts running.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Reviewed-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
By default the cache size is 5 pages per CPU, but it can be disabled at
boot time by setting the rcu_min_cached_objs to zero. When that happens,
the current code will uselessly set an hrtimer to schedule refilling this
cache with zero pages. This commit therefore streamlines this process
by simply refusing the set the hrtimer when rcu_min_cached_objs is zero.
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The add_ptr_to_bulk_krc_lock() function is invoked to allocate a new
kfree_rcu() page, also known as a kvfree_rcu_bulk_data structure.
The kfree_rcu_cpu structure's lock is used to protect this operation,
except that this lock must be momentarily dropped when allocating memory.
It is clearly important that the lock that is reacquired be the same
lock that was acquired initially via krc_this_cpu_lock().
Unfortunately, this same krc_this_cpu_lock() function is used to
re-acquire this lock, and if the task migrated to some other CPU during
the memory allocation, this will result in the kvfree_rcu_bulk_data
structure being added to the wrong CPU's kfree_rcu_cpu structure.
This commit therefore replaces that second call to krc_this_cpu_lock()
with raw_spin_lock_irqsave() in order to explicitly acquire the lock on
the correct kfree_rcu_cpu structure, thus keeping things straight even
when the task migrates.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
If kvfree_rcu_bulk() sees that the required grace period has failed to
elapse, it leaks the memory because readers might still be using it.
But in that case, the debug-objects subsystem still marks the relevant
structures as having been freed, even though they are instead being
leaked.
This commit fixes this mismatch by invoking debug_rcu_bhead_unqueue()
only when we are actually going to free the objects.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Under low-memory conditions, kvfree_rcu() will use each object's
rcu_head structure to queue objects in a singly linked list headed by
the kfree_rcu_cpu structure's ->head field. This list is passed to
call_rcu() as a unit, but there is no indication of which grace period
this list needs to wait for. This in turn prevents adding debug checks
in the kfree_rcu_work() as was done for the two page-of-pointers channels
in the kfree_rcu_cpu structure.
This commit therefore adds a ->head_free_gp_snap field to the
kfree_rcu_cpu_work structure to record this grace-period number. It also
adds a WARN_ON_ONCE() to kfree_rcu_monitor() that checks to make sure
that the required grace period has in fact elapsed.
[ paulmck: Fix kerneldoc issue raised by Stephen Rothwell. ]
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit adds debugging checks to verify that the required RCU
grace period has elapsed for each kvfree_rcu_bulk_data structure that
arrives at the kvfree_rcu_bulk() function. These checks make use
of that structure's ->gp_snap field, which has been upgraded from an
unsigned long to an rcu_gp_oldstate structure. This upgrade reduces
the chances of false positives to nearly zero, even on 32-bit systems,
for which this structure carries 64 bits of state.
Cc: Ziwei Dai <ziwei.dai@unisoc.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
o MAINTAINERS files additions and changes.
o Fix hotplug warning in nohz code.
o Tick dependency changes by Zqiang.
o Lazy-RCU shrinker fixes by Zqiang.
o rcu-tasks stall reporting improvements by Neeraj.
o Initial changes for renaming of k[v]free_rcu() to its new k[v]free_rcu_mightsleep()
name for robustness.
o Documentation Updates:
o Significant changes to srcu_struct size.
o Deadlock detection for srcu_read_lock() vs synchronize_srcu() from Boqun.
o rcutorture and rcu-related tool, which are targeted for v6.4 from Boqun's tree.
o Other misc changes.
-----BEGIN PGP SIGNATURE-----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=Rgbj
-----END PGP SIGNATURE-----
Merge tag 'rcu.6.4.april5.2023.3' of git://git.kernel.org/pub/scm/linux/kernel/git/jfern/linux
Pull RCU updates from Joel Fernandes:
- Updates and additions to MAINTAINERS files, with Boqun being added to
the RCU entry and Zqiang being added as an RCU reviewer.
I have also transitioned from reviewer to maintainer; however, Paul
will be taking over sending RCU pull-requests for the next merge
window.
- Resolution of hotplug warning in nohz code, achieved by fixing
cpu_is_hotpluggable() through interaction with the nohz subsystem.
Tick dependency modifications by Zqiang, focusing on fixing usage of
the TICK_DEP_BIT_RCU_EXP bitmask.
- Avoid needless calls to the rcu-lazy shrinker for CONFIG_RCU_LAZY=n
kernels, fixed by Zqiang.
- Improvements to rcu-tasks stall reporting by Neeraj.
- Initial renaming of k[v]free_rcu() to k[v]free_rcu_mightsleep() for
increased robustness, affecting several components like mac802154,
drbd, vmw_vmci, tracing, and more.
A report by Eric Dumazet showed that the API could be unknowingly
used in an atomic context, so we'd rather make sure they know what
they're asking for by being explicit:
https://lore.kernel.org/all/20221202052847.2623997-1-edumazet@google.com/
- Documentation updates, including corrections to spelling,
clarifications in comments, and improvements to the srcu_size_state
comments.
- Better srcu_struct cache locality for readers, by adjusting the size
of srcu_struct in support of SRCU usage by Christoph Hellwig.
- Teach lockdep to detect deadlocks between srcu_read_lock() vs
synchronize_srcu() contributed by Boqun.
Previously lockdep could not detect such deadlocks, now it can.
- Integration of rcutorture and rcu-related tools, targeted for v6.4
from Boqun's tree, featuring new SRCU deadlock scenarios, test_nmis
module parameter, and more
- Miscellaneous changes, various code cleanups and comment improvements
* tag 'rcu.6.4.april5.2023.3' of git://git.kernel.org/pub/scm/linux/kernel/git/jfern/linux: (71 commits)
checkpatch: Error out if deprecated RCU API used
mac802154: Rename kfree_rcu() to kvfree_rcu_mightsleep()
rcuscale: Rename kfree_rcu() to kfree_rcu_mightsleep()
ext4/super: Rename kfree_rcu() to kfree_rcu_mightsleep()
net/mlx5: Rename kfree_rcu() to kfree_rcu_mightsleep()
net/sysctl: Rename kvfree_rcu() to kvfree_rcu_mightsleep()
lib/test_vmalloc.c: Rename kvfree_rcu() to kvfree_rcu_mightsleep()
tracing: Rename kvfree_rcu() to kvfree_rcu_mightsleep()
misc: vmw_vmci: Rename kvfree_rcu() to kvfree_rcu_mightsleep()
drbd: Rename kvfree_rcu() to kvfree_rcu_mightsleep()
rcu: Protect rcu_print_task_exp_stall() ->exp_tasks access
rcu: Avoid stack overflow due to __rcu_irq_enter_check_tick() being kprobe-ed
rcu-tasks: Report stalls during synchronize_srcu() in rcu_tasks_postscan()
rcu: Permit start_poll_synchronize_rcu_expedited() to be invoked early
rcu: Remove never-set needwake assignment from rcu_report_qs_rdp()
rcu: Register rcu-lazy shrinker only for CONFIG_RCU_LAZY=y kernels
rcu: Fix missing TICK_DEP_MASK_RCU_EXP dependency check
rcu: Fix set/clear TICK_DEP_BIT_RCU_EXP bitmask race
rcu/trace: use strscpy() to instead of strncpy()
tick/nohz: Fix cpu_is_hotpluggable() by checking with nohz subsystem
...
Memory passed to kvfree_rcu() that is to be freed is tracked by a
per-CPU kfree_rcu_cpu structure, which in turn contains pointers
to kvfree_rcu_bulk_data structures that contain pointers to memory
that has not yet been handed to RCU, along with an kfree_rcu_cpu_work
structure that tracks the memory that has already been handed to RCU.
These structures track three categories of memory: (1) Memory for
kfree(), (2) Memory for kvfree(), and (3) Memory for both that arrived
during an OOM episode. The first two categories are tracked in a
cache-friendly manner involving a dynamically allocated page of pointers
(the aforementioned kvfree_rcu_bulk_data structures), while the third
uses a simple (but decidedly cache-unfriendly) linked list through the
rcu_head structures in each block of memory.
On a given CPU, these three categories are handled as a unit, with that
CPU's kfree_rcu_cpu_work structure having one pointer for each of the
three categories. Clearly, new memory for a given category cannot be
placed in the corresponding kfree_rcu_cpu_work structure until any old
memory has had its grace period elapse and thus has been removed. And
the kfree_rcu_monitor() function does in fact check for this.
Except that the kfree_rcu_monitor() function checks these pointers one
at a time. This means that if the previous kfree_rcu() memory passed
to RCU had only category 1 and the current one has only category 2, the
kfree_rcu_monitor() function will send that current category-2 memory
along immediately. This can result in memory being freed too soon,
that is, out from under unsuspecting RCU readers.
To see this, consider the following sequence of events, in which:
o Task A on CPU 0 calls rcu_read_lock(), then uses "from_cset",
then is preempted.
o CPU 1 calls kfree_rcu(cset, rcu_head) in order to free "from_cset"
after a later grace period. Except that "from_cset" is freed
right after the previous grace period ended, so that "from_cset"
is immediately freed. Task A resumes and references "from_cset"'s
member, after which nothing good happens.
In full detail:
CPU 0 CPU 1
---------------------- ----------------------
count_memcg_event_mm()
|rcu_read_lock() <---
|mem_cgroup_from_task()
|// css_set_ptr is the "from_cset" mentioned on CPU 1
|css_set_ptr = rcu_dereference((task)->cgroups)
|// Hard irq comes, current task is scheduled out.
cgroup_attach_task()
|cgroup_migrate()
|cgroup_migrate_execute()
|css_set_move_task(task, from_cset, to_cset, true)
|cgroup_move_task(task, to_cset)
|rcu_assign_pointer(.., to_cset)
|...
|cgroup_migrate_finish()
|put_css_set_locked(from_cset)
|from_cset->refcount return 0
|kfree_rcu(cset, rcu_head) // free from_cset after new gp
|add_ptr_to_bulk_krc_lock()
|schedule_delayed_work(&krcp->monitor_work, ..)
kfree_rcu_monitor()
|krcp->bulk_head[0]'s work attached to krwp->bulk_head_free[]
|queue_rcu_work(system_wq, &krwp->rcu_work)
|if rwork->rcu.work is not in WORK_STRUCT_PENDING_BIT state,
|call_rcu(&rwork->rcu, rcu_work_rcufn) <--- request new gp
// There is a perious call_rcu(.., rcu_work_rcufn)
// gp end, rcu_work_rcufn() is called.
rcu_work_rcufn()
|__queue_work(.., rwork->wq, &rwork->work);
|kfree_rcu_work()
|krwp->bulk_head_free[0] bulk is freed before new gp end!!!
|The "from_cset" is freed before new gp end.
// the task resumes some time later.
|css_set_ptr->subsys[(subsys_id) <--- Caused kernel crash, because css_set_ptr is freed.
This commit therefore causes kfree_rcu_monitor() to refrain from moving
kfree_rcu() memory to the kfree_rcu_cpu_work structure until the RCU
grace period has completed for all three categories.
v2: Use helper function instead of inserted code block at kfree_rcu_monitor().
Fixes: 34c8817455 ("rcu: Support kfree_bulk() interface in kfree_rcu()")
Fixes: 5f3c8d6204 ("rcu/tree: Maintain separate array for vmalloc ptrs")
Reported-by: Mukesh Ojha <quic_mojha@quicinc.com>
Signed-off-by: Ziwei Dai <ziwei.dai@unisoc.com>
Reviewed-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Tested-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The kfree_rcu() and kvfree_rcu() macros' single-argument forms are
deprecated. Therefore switch to the new kfree_rcu_mightsleep() and
kvfree_rcu_mightsleep() variants. The goal is to avoid accidental use
of the single-argument forms, which can introduce functionality bugs in
atomic contexts and latency bugs in non-atomic contexts.
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>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
For kernels built with CONFIG_PREEMPT_RCU=y, the following scenario can
result in a NULL-pointer dereference:
CPU1 CPU2
rcu_preempt_deferred_qs_irqrestore rcu_print_task_exp_stall
if (special.b.blocked) READ_ONCE(rnp->exp_tasks) != NULL
raw_spin_lock_rcu_node
np = rcu_next_node_entry(t, rnp)
if (&t->rcu_node_entry == rnp->exp_tasks)
WRITE_ONCE(rnp->exp_tasks, np)
....
raw_spin_unlock_irqrestore_rcu_node
raw_spin_lock_irqsave_rcu_node
t = list_entry(rnp->exp_tasks->prev,
struct task_struct, rcu_node_entry)
(if rnp->exp_tasks is NULL, this
will dereference a NULL pointer)
The problem is that CPU2 accesses the rcu_node structure's->exp_tasks
field without holding the rcu_node structure's ->lock and CPU2 did
not observe CPU1's change to rcu_node structure's ->exp_tasks in time.
Therefore, if CPU1 sets rcu_node structure's->exp_tasks pointer to NULL,
then CPU2 might dereference that NULL pointer.
This commit therefore holds the rcu_node structure's ->lock while
accessing that structure's->exp_tasks field.
[ paulmck: Apply Frederic Weisbecker feedback. ]
Acked-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
The call to synchronize_srcu() from rcu_tasks_postscan() can be stalled
by a task getting stuck in do_exit() between that function's calls to
exit_tasks_rcu_start() and exit_tasks_rcu_finish(). To ease diagnosis
of this situation, print a stall warning message every rcu_task_stall_info
period when rcu_tasks_postscan() is stalled.
[ paulmck: Adjust to handle CONFIG_SMP=n. ]
Acked-by: Frederic Weisbecker <frederic@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Reported-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/rcu/20230111212736.GA1062057@paulmck-ThinkPad-P17-Gen-1/
Signed-off-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
According to the commit log of the patch that added it to the kernel,
start_poll_synchronize_rcu_expedited() can be invoked very early, as
in long before rcu_init() has been invoked. But before rcu_init(),
the rcu_data structure's ->mynode field has not yet been initialized.
This means that the start_poll_synchronize_rcu_expedited() function's
attempt to set the CPU's leaf rcu_node structure's ->exp_seq_poll_rq
field will result in a segmentation fault.
This commit therefore causes start_poll_synchronize_rcu_expedited() to
set ->exp_seq_poll_rq only after rcu_init() has initialized all CPUs'
rcu_data structures' ->mynode fields. It also removes the check from
the rcu_init() function so that start_poll_synchronize_rcu_expedited(
is unconditionally invoked. Yes, this might result in an unnecessary
boot-time grace period, but this is down in the noise.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
The rcu_accelerate_cbs() function is invoked by rcu_report_qs_rdp()
only if there is a grace period in progress that is still blocked
by at least one CPU on this rcu_node structure. This means that
rcu_accelerate_cbs() should never return the value true, and thus that
this function should never set the needwake variable and in turn never
invoke rcu_gp_kthread_wake().
This commit therefore removes the needwake variable and the invocation
of rcu_gp_kthread_wake() in favor of a WARN_ON_ONCE() on the call to
rcu_accelerate_cbs(). The purpose of this new WARN_ON_ONCE() is to
detect situations where the system's opinion differs from ours.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
The lazy_rcu_shrink_count() shrinker function is registered even in
kernels built with CONFIG_RCU_LAZY=n, in which case this function
uselessly consumes cycles learning that no CPU has any lazy callbacks
queued.
This commit therefore registers this shrinker function only in the kernels
built with CONFIG_RCU_LAZY=y, where it might actually do something useful.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
For kernels built with CONFIG_NO_HZ_FULL=y, the following scenario can result
in the scheduling-clock interrupt remaining enabled on a holdout CPU after
its quiescent state has been reported:
CPU1 CPU2
rcu_report_exp_cpu_mult synchronize_rcu_expedited_wait
acquires rnp->lock mask = rnp->expmask;
for_each_leaf_node_cpu_mask(rnp, cpu, mask)
rnp->expmask = rnp->expmask & ~mask; rdp = per_cpu_ptr(&rcu_data, cpu1);
for_each_leaf_node_cpu_mask(rnp, cpu, mask)
rdp = per_cpu_ptr(&rcu_data, cpu1);
if (!rdp->rcu_forced_tick_exp)
continue; rdp->rcu_forced_tick_exp = true;
tick_dep_set_cpu(cpu1, TICK_DEP_BIT_RCU_EXP);
The problem is that CPU2's sampling of rnp->expmask is obsolete by the
time it invokes tick_dep_set_cpu(), and CPU1 is not guaranteed to see
CPU2's store to ->rcu_forced_tick_exp in time to clear it. And even if
CPU1 does see that store, it might invoke tick_dep_clear_cpu() before
CPU2 got around to executing its tick_dep_set_cpu(), which would still
leave the victim CPU with its scheduler-clock tick running.
Either way, an nohz_full real-time application running on the victim
CPU would have its latency needlessly degraded.
Note that expedited RCU grace periods look at context-tracking
information, and so if the CPU is executing in nohz_full usermode
throughout, that CPU cannot be victimized in this manner.
This commit therefore causes synchronize_rcu_expedited_wait to hold
the rcu_node structure's ->lock when checking for holdout CPUs, setting
TICK_DEP_BIT_RCU_EXP, and invoking tick_dep_set_cpu(), thus preventing
this race.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Now that all references to CONFIG_SRCU have been removed, it is time to
remove CONFIG_SRCU itself.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: John Ogness <john.ogness@linutronix.de>
Cc: Petr Mladek <pmladek@suse.com>
Reviewed-by: John Ogness <john.ogness@linutronix.de>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
This commit adds a comment to help explain why the "else" clause of the
in_serving_softirq() "if" statement does not need to enforce a time limit.
The reason is that this "else" clause handles rcuoc kthreads that do not
block handlers for other softirq vectors.
Acked-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
There is an smp_mb() named "E" in srcu_flip() immediately before the
increment (flip) of the srcu_struct structure's ->srcu_idx.
The purpose of E is to order the preceding scan's read of lock counters
against the flipping of the ->srcu_idx, in order to prevent new readers
from continuing to use the old ->srcu_idx value, which might needlessly
extend the grace period.
However, this ordering is already enforced because of the control
dependency between the preceding scan and the ->srcu_idx flip.
This control dependency exists because atomic_long_read() is used
to scan the counts, because WRITE_ONCE() is used to flip ->srcu_idx,
and because ->srcu_idx is not flipped until the ->srcu_lock_count[] and
->srcu_unlock_count[] counts match. And such a match cannot happen when
there is an in-flight reader that started before the flip (observation
courtesy Mathieu Desnoyers).
The litmus test below (courtesy of Frederic Weisbecker, with changes
for ctrldep by Boqun and Joel) shows this:
C srcu
(*
* bad condition: P0's first scan (SCAN1) saw P1's idx=0 LOCK count inc, though P1 saw flip.
*
* So basically, the ->po ordering on both P0 and P1 is enforced via ->ppo
* (control deps) on both sides, and both P0 and P1 are interconnected by ->rf
* relations. Combining the ->ppo with ->rf, a cycle is impossible.
*)
{}
// updater
P0(int *IDX, int *LOCK0, int *UNLOCK0, int *LOCK1, int *UNLOCK1)
{
int lock1;
int unlock1;
int lock0;
int unlock0;
// SCAN1
unlock1 = READ_ONCE(*UNLOCK1);
smp_mb(); // A
lock1 = READ_ONCE(*LOCK1);
// FLIP
if (lock1 == unlock1) { // Control dep
smp_mb(); // E // Remove E and still passes.
WRITE_ONCE(*IDX, 1);
smp_mb(); // D
// SCAN2
unlock0 = READ_ONCE(*UNLOCK0);
smp_mb(); // A
lock0 = READ_ONCE(*LOCK0);
}
}
// reader
P1(int *IDX, int *LOCK0, int *UNLOCK0, int *LOCK1, int *UNLOCK1)
{
int tmp;
int idx1;
int idx2;
// 1st reader
idx1 = READ_ONCE(*IDX);
if (idx1 == 0) { // Control dep
tmp = READ_ONCE(*LOCK0);
WRITE_ONCE(*LOCK0, tmp + 1);
smp_mb(); /* B and C */
tmp = READ_ONCE(*UNLOCK0);
WRITE_ONCE(*UNLOCK0, tmp + 1);
} else {
tmp = READ_ONCE(*LOCK1);
WRITE_ONCE(*LOCK1, tmp + 1);
smp_mb(); /* B and C */
tmp = READ_ONCE(*UNLOCK1);
WRITE_ONCE(*UNLOCK1, tmp + 1);
}
}
exists (0:lock1=1 /\ 1:idx1=1)
More complicated litmus tests with multiple SRCU readers also show that
memory barrier E is not needed.
This commit therefore clarifies the comment on memory barrier E.
Why not also remove that redundant smp_mb()?
Because control dependencies are quite fragile due to their not being
recognized by most compilers and tools. Control dependencies therefore
exact an ongoing maintenance burden, and such a burden cannot be justified
in this slowpath. Therefore, that smp_mb() stays until such time as
its overhead becomes a measurable problem in a real workload running on
a real production system, or until such time as compilers start paying
attention to this sort of control dependency.
Co-developed-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Co-developed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Co-developed-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
The state space of the GP sequence number isn't documented and the
definitions of its special values are scattered. This commit therefore
gathers some common knowledge near the grace-period sequence-number
definitions.
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
This commit creates an srcu_usage pointer named "sup" as a shorter
synonym for the "ssp->srcu_sup" that was bloating several lines of code.
Cc: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit creates an srcu_usage pointer named "sup" as a shorter
synonym for the "ssp->srcu_sup" that was bloating several lines of code.
Cc: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit creates an srcu_usage pointer named "sup" as a shorter
synonym for the "ssp->srcu_sup" that was bloating several lines of code.
Cc: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit creates an srcu_usage pointer named "sup" as a shorter
synonym for the "ssp->srcu_sup" that was bloating several lines of code.
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Cc: Christoph Hellwig <hch@lst.de>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
If a given statically allocated in-module srcu_struct structure was ever
used for updates, srcu_module_going() will invoke cleanup_srcu_struct()
at module-exit time. This will check for the error case of SRCU readers
persisting past module-exit time. On the other hand, if this srcu_struct
structure never went through a grace period, srcu_module_going() only
invokes free_percpu(), which would result in strange failures if SRCU
readers persisted past module-exit time.
This commit therefore adds a srcu_readers_active() check to
srcu_module_going(), splatting if readers have persisted and refraining
from invoking free_percpu() in that case. Better to leak memory than
to suffer silent memory corruption!
[ paulmck: Apply Zhang, Qiang1 feedback on memory leak. ]
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->reschedule_jiffies, ->reschedule_count, and
->work fields from the srcu_struct structure to the srcu_usage structure
to reduce the size of the former in order to improve cache locality.
However, this means that the container_of() calls cannot get a pointer
to the srcu_struct because they are no longer in the srcu_struct.
This issue is addressed by adding a ->srcu_ssp field in the srcu_usage
structure that references the corresponding srcu_struct structure.
And given the presence of the sup pointer to the srcu_usage structure,
replace some ssp->srcu_usage-> instances with sup->.
[ paulmck Apply feedback from kernel test robot. ]
Link: https://lore.kernel.org/oe-kbuild-all/202303191400.iO5BOqka-lkp@intel.com/
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->srcu_barrier_seq, ->srcu_barrier_mutex,
->srcu_barrier_completion, and ->srcu_barrier_cpu_cnt fields from the
srcu_struct structure to the srcu_usage structure to reduce the size of
the former in order to improve cache locality.
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->sda_is_static field from the srcu_struct structure
to the srcu_usage structure to reduce the size of the former in order
to improve cache locality.
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->srcu_size_jiffies, ->srcu_n_lock_retries,
and ->srcu_n_exp_nodelay fields from the srcu_struct structure to the
srcu_usage structure to reduce the size of the former in order to improve
cache locality.
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->srcu_gp_seq, ->srcu_gp_seq_needed,
->srcu_gp_seq_needed_exp, ->srcu_gp_start, and ->srcu_last_gp_end fields
from the srcu_struct structure to the srcu_usage structure to reduce
the size of the former in order to improve cache locality.
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->srcu_gp_mutex field from the srcu_struct structure
to the srcu_usage structure to reduce the size of the former in order
to improve cache locality.
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->lock field from the srcu_struct structure to
the srcu_usage structure to reduce the size of the former in order to
improve cache locality.
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, both __init_srcu_struct() in CONFIG_DEBUG_LOCK_ALLOC=y kernels
and init_srcu_struct() in CONFIG_DEBUG_LOCK_ALLOC=n kernel initialize
the srcu_struct structure's ->lock before the srcu_usage structure has
been allocated. This of course prevents the ->lock from being moved
to the srcu_usage structure, so this commit moves the initialization
into the init_srcu_struct_fields() after the srcu_usage structure has
been allocated.
Cc: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->srcu_cb_mutex field from the srcu_struct structure
to the srcu_usage structure to reduce the size of the former in order
to improve cache locality.
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->srcu_size_state field from the srcu_struct
structure to the srcu_usage structure to reduce the size of the former
in order to improve cache locality.
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit moves the ->level[] array from the srcu_struct structure to
the srcu_usage structure to reduce the size of the former in order to
improve cache locality.
Suggested-by: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The current srcu_struct structure is on the order of 200 bytes in size
(depending on architecture and .config), which is much better than the
old-style 26K bytes, but still all too inconvenient when one is trying
to achieve good cache locality on a fastpath involving SRCU readers.
However, only a few fields in srcu_struct are used by SRCU readers.
The remaining fields could be offloaded to a new srcu_update
structure, thus shrinking the srcu_struct structure down to a few
tens of bytes. This commit begins this noble quest, a quest that is
complicated by open-coded initialization of the srcu_struct within the
srcu_notifier_head structure. This complication is addressed by updating
the srcu_notifier_head structure's open coding, given that there does
not appear to be a straightforward way of abstracting that initialization.
This commit moves only the ->node pointer to srcu_update. Later commits
will move additional fields.
[ paulmck: Fold in qiang1.zhang@intel.com's memory-leak fix. ]
Link: https://lore.kernel.org/all/20230320055751.4120251-1-qiang1.zhang@intel.com/
Suggested-by: Christoph Hellwig <hch@lst.de>
Cc: "Rafael J. Wysocki" <rafael.j.wysocki@intel.com>
Cc: "Michał Mirosław" <mirq-linux@rere.qmqm.pl>
Cc: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Further shrinking the srcu_struct structure is eased by requiring
that in-module srcu_struct structures rely more heavily on static
initialization. In particular, this preserves the property that
a module-load-time srcu_struct initialization can fail only due
to memory-allocation failure of the per-CPU srcu_data structures.
It might also slightly improve robustness by keeping the number of memory
allocations that must succeed down percpu_alloc() call.
This is in preparation for splitting an srcu_usage structure out
of the srcu_struct structure.
[ paulmck: Fold in qiang1.zhang@intel.com feedback. ]
Cc: Christoph Hellwig <hch@lst.de>
Tested-by: Sachin Sant <sachinp@linux.ibm.com>
Tested-by: "Zhang, Qiang1" <qiang1.zhang@intel.com>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The tasks_rcu_exit_srcu variable is used only by kernels built
with CONFIG_TASKS_RCU=y, but is defined for all kernesl with
CONFIG_TASKS_RCU_GENERIC=y. Therefore, in kernels built with
CONFIG_TASKS_RCU_GENERIC=y but CONFIG_TASKS_RCU=n, this gives
a "defined but not used" warning.
This commit therefore moves this variable under CONFIG_TASKS_RCU.
Link: https://lore.kernel.org/oe-kbuild-all/202303191536.XzMSyzTl-lkp@intel.com/
Reported-by: kernel test robot <lkp@intel.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Tested-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Add a test number 3 that creates deadlock cycles involving one RCU
Tasks Trace step and L-1 SRCU steps. Please note that lockdep will not
detect these deadlocks until synchronize_rcu_tasks_trace() is marked
with lockdep's new "sync" annotation, which will probably not happen
until some time after these markings prove their worth on SRCU.
Please note that these tests are available only in kernels built with
CONFIG_TASKS_TRACE_RCU=y.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
In order to test the new SRCU-lockdep functionality, this commit adds
an rcutorture.test_srcu_lockdep module parameter that, when non-zero,
selects an SRCU deadlock scenario to execute. This parameter is a
five-digit number formatted as DNNL, where "D" is 1 to force a deadlock
and 0 to avoid doing so; "NN" is the test number, 0 for SRCU-based, 1
for SRCU/mutex-based, and 2 for SRCU/rwsem-based; and "L" is the number
of steps in the deadlock cycle.
Note that rcutorture.test_srcu_lockdep=1 will also force a hard hang.
If a non-zero value of rcutorture.test_srcu_lockdep does not select a
deadlock scenario, a console message is printed and testing continues.
[ paulmck: Apply kernel test robot feedback, add rwsem support. ]
[ paulmck: Apply Dan Carpenter feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Although all flavors of RCU readers are annotated correctly with
lockdep as recursive read locks, they do not set the lock_acquire
'check' parameter. This means that RCU read locks are not added to
the lockdep dependency graph, which in turn means that lockdep cannot
detect RCU-based deadlocks. This is not a problem for RCU flavors having
atomic read-side critical sections because context-based annotations can
catch these deadlocks, see for example the RCU_LOCKDEP_WARN() statement
in synchronize_rcu(). But context-based annotations are not helpful
for sleepable RCU, especially given that it is perfectly legal to do
synchronize_srcu(&srcu1) within an srcu_read_lock(&srcu2).
However, we can detect SRCU-based by: (1) Making srcu_read_lock() a
'check'ed recursive read lock and (2) Making synchronize_srcu() a empty
write lock critical section. Even better, with the newly introduced
lock_sync(), we can avoid false positives about irq-unsafe/safe.
This commit therefore makes it so.
Note that NMI-safe SRCU read side critical sections are currently not
annotated, but might be annotated in the future.
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
[ boqun: Add comments for annotation per Waiman's suggestion ]
[ boqun: Fix comment warning reported by Stephen Rothwell ]
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
The ref_scale_shutdown() kthread/function uses wait_event() to wait for
the refscale test to complete. However, although the read-side tests
are normally extremely fast, there is no law against specifying a very
large value for the refscale.loops module parameter or against having
a slow read-side primitive. Either way, this might well trigger the
hung-task timeout.
This commit therefore replaces those wait_event() calls with calls to
wait_event_idle(), which do not trigger the hung-task timeout.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
The rcu_scale_shutdown() and kfree_scale_shutdown() kthreads/functions
use wait_event() to wait for the rcuscale test to complete. However,
each updater thread in such a test waits for at least 100 grace periods.
If each grace period takes more than 1.2 seconds, which is long, but
not insanely so, this can trigger the hung-task timeout.
This commit therefore replaces those wait_event() calls with calls to
wait_event_idle(), which do not trigger the hung-task timeout.
Reported-by: kernel test robot <yujie.liu@intel.com>
Reported-by: Liam Howlett <liam.howlett@oracle.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Tested-by: Yujie Liu <yujie.liu@intel.com>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Given a non-zero rcutorture.nocbs_nthreads module parameter, the specified
number of nocb kthreads will be created, regardless of whether or not
the RCU implementation under test is capable of offloading callbacks.
Please note that even vanilla RCU is incapable of offloading in kernels
built with CONFIG_RCU_NOCB_CPU=n. And when the RCU implementation is
incapable of offloading callbacks, there is no point in creating those
kthreads.
This commit therefore checks the cur_ops.torture_type module parameter and
CONFIG_RCU_NOCB_CPU Kconfig option in order to avoid creating unnecessary
nocb tasks.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
[ boqun: Fix checkpatch warning ]
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
After commit 8b700983de ("sched: Remove sched_set_*() return value"),
this variable is not used anymore. So eliminate it entirely.
Signed-off-by: Yue Hu <huyue2@coolpad.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
This commit adds a test_nmis module parameter to generate the
specified number of NMI stack backtraces 15 seconds apart. This module
parameter can be used to test NMI delivery and accompanying diagnostics.
Note that this parameter is ignored when rcutorture is a module rather
than built into the kernel. This could be changed with the addition of
an EXPORT_SYMBOL_GPL().
[ paulmck: Apply kernel test robot feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
During suspend, we see failures to suspend 1 in 300-500 suspends.
Looking closer, it appears that asynchronous RCU callbacks are being
queued as lazy even though synchronous callbacks are expedited. These
delays appear to not be very welcome by the suspend/resume code as
evidenced by these occasional suspend failures.
This commit modifies call_rcu() to check if rcu_async_should_hurry(),
which will return true if we are in suspend or in-kernel boot.
[ paulmck: Alphabetize local variables. ]
Ignoring the lazy hint makes the 3000 suspend/resume cycles pass
reliably on a 12th gen 12-core Intel CPU, and there is some evidence
that it also slightly speeds up boot performance.
Fixes: 3cb278e73b ("rcu: Make call_rcu() lazy to save power")
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Boot and suspend/resume should not be slowed down in kernels built with
CONFIG_RCU_LAZY=y. In particular, suspend can sometimes fail in such
kernels.
This commit therefore adds rcu_async_hurry(), rcu_async_relax(), and
rcu_async_should_hurry() functions that track whether or not either
a boot or a suspend/resume operation is in progress. This will
enable a later commit to refrain from laziness during those times.
Export rcu_async_should_hurry(), rcu_async_hurry(), and rcu_async_relax()
for later use by rcutorture.
[ paulmck: Apply feedback from Steve Rostedt. ]
Fixes: 3cb278e73b ("rcu: Make call_rcu() lazy to save power")
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The rcu_boost_kthread_setaffinity() function is invoked at
rcutree_online_cpu() and rcutree_offline_cpu() time, early in the online
timeline and late in the offline timeline, respectively. It is also
invoked from rcutree_dead_cpu(), however, in the absence of userspace
manipulations (for which userspace must take responsibility), this call
is redundant with that from rcutree_offline_cpu(). This redundancy can
be demonstrated by printing out the relevant cpumasks
This commit therefore removes the call to rcu_boost_kthread_setaffinity()
from rcutree_dead_cpu().
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
The maximum value of RCU CPU stall-warning timeouts has historically been
five minutes (300 seconds). However, the recently introduced expedited
RCU CPU stall-warning timeout is instead limited to 21 seconds. This
causes problems for CI/fuzzing services such as syzkaller by obscuring
the issue in question with expedited RCU CPU stall-warning timeout splats.
This commit therefore sets the RCU_EXP_CPU_STALL_TIMEOUT Kconfig options
upper bound to 300000 milliseconds, which is 300 seconds (AKA 5 minutes).
[ paulmck: Apply feedback from Hillf Danton. ]
[ paulmck: Apply feedback from Geert Uytterhoeven. ]
Reported-by: Dave Chinner <david@fromorbit.com>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Tested-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Time stamps are added to the output in kernels built with
CONFIG_PRINTK_TIME=y, which causes misaligned output. Therefore,
replace pr_cont() with pr_err(), which fixes alignment and gets
rid of a couple of despised pr_cont() calls.
Before:
[ 37.567343] rcu: INFO: rcu_preempt self-detected stall on CPU
[ 37.567839] rcu: 0-....: (1500 ticks this GP) idle=***
[ 37.568270] (t=1501 jiffies g=4717 q=28 ncpus=4)
[ 37.568668] CPU: 0 PID: 313 Comm: test0 Not tainted 6.1.0-rc4 #8
After:
[ 36.762074] rcu: INFO: rcu_preempt self-detected stall on CPU
[ 36.762543] rcu: 0-....: (1499 ticks this GP) idle=***
[ 36.763003] rcu: (t=1500 jiffies g=5097 q=27 ncpus=4)
[ 36.763522] CPU: 0 PID: 313 Comm: test0 Not tainted 6.1.0-rc4 #9
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Because RCU CPU stall warnings are driven from the scheduling-clock
interrupt handler, a workload consisting of a very large number of
short-duration hardware interrupts can result in misleading stall-warning
messages. On systems supporting only a single level of interrupts,
that is, where interrupts handlers cannot be interrupted, this can
produce misleading diagnostics. The stack traces will show the
innocent-bystander interrupted task, not the interrupts that are
at the very least exacerbating the stall.
This situation can be improved by displaying the number of interrupts
and the CPU time that they have consumed. Diagnosing other types
of stalls can be eased by also providing the count of softirqs and
the CPU time that they consumed as well as the number of context
switches and the task-level CPU time consumed.
Consider the following output given this change:
rcu: INFO: rcu_preempt self-detected stall on CPU
rcu: 0-....: (1250 ticks this GP) <omitted>
rcu: hardirqs softirqs csw/system
rcu: number: 624 45 0
rcu: cputime: 69 1 2425 ==> 2500(ms)
This output shows that the number of hard and soft interrupts is small,
there are no context switches, and the system takes up a lot of time. This
indicates that the current task is looping with preemption disabled.
The impact on system performance is negligible because snapshot is
recorded only once for all continuous RCU stalls.
This added debugging information is suppressed by default and can be
enabled by building the kernel with CONFIG_RCU_CPU_STALL_CPUTIME=y or
by booting with rcupdate.rcu_cpu_stall_cputime=1.
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Reviewed-by: Mukesh Ojha <quic_mojha@quicinc.com>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The sparse __acquires() and __releases() annotations provide very
little value. The argument is ignored, so sparse cannot tell the
differences between acquiring one lock and releasing another on the one
hand and acquiring and releasing a given lock on the other. In addition,
lockdep annotations provide much more precision, for but one example,
actually knowing which lock is held.
This commit therefore removes the __acquires() and __releases()
annotations from rcutorture.
Reported-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit adds three read-side-only tests of three use cases featuring
SLAB_TYPESAFE_BY_RCU: One using per-object reference counting, one using
per-object locking, and one using per-object sequence locking.
[ paulmck: Apply feedback from kernel test robot. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Current tests all have init() functions that are guaranteed to succeed.
But upcoming tests will need to allocate memory, thus possibly failing.
This commit therefore handles init() function failure.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The rcu_tasks_need_gpcb() determines whether or not: (1) There are
callbacks needing another grace period, (2) There are callbacks ready
to be invoked, and (3) It would be a good time to shrink back down to a
single-CPU callback list. This third case is interesting because some
other CPU might be adding new callbacks, which might suddenly make this
a very bad time to be shrinking.
This is currently handled by requiring call_rcu_tasks_generic() to
enqueue callbacks under the protection of rcu_read_lock() and requiring
rcu_tasks_need_gpcb() to wait for an RCU grace period to elapse before
finalizing the transition. This works well in practice.
Unfortunately, the current code assumes that a grace period whose end is
detected by the poll_state_synchronize_rcu() in the second "if" condition
actually ended before the earlier code counted the callbacks queued on
CPUs other than CPU 0 (local variable "ncbsnz"). Given the current code,
it is possible that a long-delayed call_rcu_tasks_generic() invocation
will queue a callback on a non-zero CPU after these CPUs have had their
callbacks counted and zero has been stored to ncbsnz. Such a callback
would trigger the WARN_ON_ONCE() in the second "if" statement.
To see this, consider the following sequence of events:
o CPU 0 invokes rcu_tasks_one_gp(), and counts fewer than
rcu_task_collapse_lim callbacks. It sees at least one
callback queued on some other CPU, thus setting ncbsnz
to a non-zero value.
o CPU 1 invokes call_rcu_tasks_generic() and loads 42 from
->percpu_enqueue_lim. It therefore decides to enqueue its
callback onto CPU 1's callback list, but is delayed.
o CPU 0 sees the rcu_task_cb_adjust is non-zero and that the number
of callbacks does not exceed rcu_task_collapse_lim. It therefore
checks percpu_enqueue_lim, and sees that its value is greater
than the value one. CPU 0 therefore starts the shift back
to a single callback list. It sets ->percpu_enqueue_lim to 1,
but CPU 1 has already read the old value of 42. It also gets
a grace-period state value from get_state_synchronize_rcu().
o CPU 0 sees that ncbsnz is non-zero in its second "if" statement,
so it declines to finalize the shrink operation.
o CPU 0 again invokes rcu_tasks_one_gp(), and counts fewer than
rcu_task_collapse_lim callbacks. It also sees that there are
no callback queued on any other CPU, and thus sets ncbsnz to zero.
o CPU 1 resumes execution and enqueues its callback onto its own
list. This invalidates the value of ncbsnz.
o CPU 0 sees the rcu_task_cb_adjust is non-zero and that the number
of callbacks does not exceed rcu_task_collapse_lim. It therefore
checks percpu_enqueue_lim, but sees that its value is already
unity. It therefore does not get a new grace-period state value.
o CPU 0 sees that rcu_task_cb_adjust is non-zero, ncbsnz is zero,
and that poll_state_synchronize_rcu() says that the grace period
has completed. it therefore finalizes the shrink operation,
setting ->percpu_dequeue_lim to the value one.
o CPU 0 does a debug check, scanning the other CPUs' callback lists.
It sees that CPU 1's list has a callback, so it (rightly)
triggers the WARN_ON_ONCE(). After all, the new value of
->percpu_dequeue_lim says to not bother looking at CPU 1's
callback list, which means that this callback will never be
invoked. This can result in hangs and maybe even OOMs.
Based on long experience with rcutorture, this is an extremely
low-probability race condition, but it really can happen, especially in
preemptible kernels or within guest OSes.
This commit therefore checks for completion of the grace period
before counting callbacks. With this change, in the above failure
scenario CPU 0 would know not to prematurely end the shrink operation
because the grace period would not have completed before the count
operation started.
[ paulmck: Adjust grace-period end rather than adding RCU reader. ]
[ paulmck: Avoid spurious WARN_ON_ONCE() with ->percpu_dequeue_lim check. ]
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The synchronize_rcu_tasks_rude() function invokes rcu_tasks_rude_wait_gp()
to wait one rude RCU-tasks grace period. The rcu_tasks_rude_wait_gp()
function in turn checks if there is only a single online CPU. If so, it
will immediately return, because a call to synchronize_rcu_tasks_rude()
is by definition a grace period on a single-CPU system. (We could
have blocked!)
Unfortunately, this check uses num_online_cpus() without synchronization,
which can result in too-short grace periods. To see this, consider the
following scenario:
CPU0 CPU1 (going offline)
migration/1 task:
cpu_stopper_thread
-> take_cpu_down
-> _cpu_disable
(dec __num_online_cpus)
->cpuhp_invoke_callback
preempt_disable
access old_data0
task1
del old_data0 .....
synchronize_rcu_tasks_rude()
task1 schedule out
....
task2 schedule in
rcu_tasks_rude_wait_gp()
->__num_online_cpus == 1
->return
....
task1 schedule in
->free old_data0
preempt_enable
When CPU1 decrements __num_online_cpus, its value becomes 1. However,
CPU1 has not finished going offline, and will take one last trip through
the scheduler and the idle loop before it actually stops executing
instructions. Because synchronize_rcu_tasks_rude() is mostly used for
tracing, and because both the scheduler and the idle loop can be traced,
this means that CPU0's prematurely ended grace period might disrupt the
tracing on CPU1. Given that this disruption might include CPU1 executing
instructions in memory that was just now freed (and maybe reallocated),
this is a matter of some concern.
This commit therefore removes that problematic single-CPU check from the
rcu_tasks_rude_wait_gp() function. This dispenses with the single-CPU
optimization, but there is no evidence indicating that this optimization
is important. In addition, synchronize_rcu_tasks_generic() contains a
similar optimization (albeit only for early boot), which also splats.
(As in exactly why are you invoking synchronize_rcu_tasks_rude() so
early in boot, anyway???)
It is OK for the synchronize_rcu_tasks_rude() function's check to be
unsynchronized because the only times that this check can evaluate to
true is when there is only a single CPU running with preemption
disabled.
While in the area, this commit also fixes a minor bug in which a
call to synchronize_rcu_tasks_rude() would instead be attributed to
synchronize_rcu_tasks().
[ paulmck: Add "synchronize_" prefix and "()" suffix. ]
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
RCU Tasks and PID-namespace unshare can interact in do_exit() in a
complicated circular dependency:
1) TASK A calls unshare(CLONE_NEWPID), this creates a new PID namespace
that every subsequent child of TASK A will belong to. But TASK A
doesn't itself belong to that new PID namespace.
2) TASK A forks() and creates TASK B. TASK A stays attached to its PID
namespace (let's say PID_NS1) and TASK B is the first task belonging
to the new PID namespace created by unshare() (let's call it PID_NS2).
3) Since TASK B is the first task attached to PID_NS2, it becomes the
PID_NS2 child reaper.
4) TASK A forks() again and creates TASK C which get attached to PID_NS2.
Note how TASK C has TASK A as a parent (belonging to PID_NS1) but has
TASK B (belonging to PID_NS2) as a pid_namespace child_reaper.
5) TASK B exits and since it is the child reaper for PID_NS2, it has to
kill all other tasks attached to PID_NS2, and wait for all of them to
die before getting reaped itself (zap_pid_ns_process()).
6) TASK A calls synchronize_rcu_tasks() which leads to
synchronize_srcu(&tasks_rcu_exit_srcu).
7) TASK B is waiting for TASK C to get reaped. But TASK B is under a
tasks_rcu_exit_srcu SRCU critical section (exit_notify() is between
exit_tasks_rcu_start() and exit_tasks_rcu_finish()), blocking TASK A.
8) TASK C exits and since TASK A is its parent, it waits for it to reap
TASK C, but it can't because TASK A waits for TASK B that waits for
TASK C.
Pid_namespace semantics can hardly be changed at this point. But the
coverage of tasks_rcu_exit_srcu can be reduced instead.
The current task is assumed not to be concurrently reapable at this
stage of exit_notify() and therefore tasks_rcu_exit_srcu can be
temporarily relaxed without breaking its constraints, providing a way
out of the deadlock scenario.
[ paulmck: Fix build failure by adding additional declaration. ]
Fixes: 3f95aa81d2 ("rcu: Make TASKS_RCU handle tasks that are almost done exiting")
Reported-by: Pengfei Xu <pengfei.xu@intel.com>
Suggested-by: Boqun Feng <boqun.feng@gmail.com>
Suggested-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Suggested-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Cc: Eric W . Biederman <ebiederm@xmission.com>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Ever since the following commit:
5a41344a3d ("srcu: Simplify __srcu_read_unlock() via this_cpu_dec()")
SRCU doesn't rely anymore on preemption to be disabled in order to
modify the per-CPU counter. And even then it used to be done from the API
itself.
Therefore and after checking further, it appears to be safe to remove
the preemption disablement around __srcu_read_[un]lock() in
exit_tasks_rcu_start() and exit_tasks_rcu_finish()
Suggested-by: Boqun Feng <boqun.feng@gmail.com>
Suggested-by: Paul E. McKenney <paulmck@kernel.org>
Suggested-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Make sure we don't need to look again into the depths of git blame in
order not to miss a subtle part about how rcu-tasks is dealing with
exiting tasks.
Suggested-by: Boqun Feng <boqun.feng@gmail.com>
Suggested-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Suggested-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, test_rcu_tasks_callback() reads from the jiffies counter only
once when this function is invoked. This introduces inaccuracies because
of the latencies induced by the synchronize_rcu_tasks*() invocations.
This commit therefore re-reads the jiffies counter at the beginning
of each test, thus avoiding penalizing later tests for the latencies
induced by earlier tests.
Therefore, this commit at the start of each RCU Tasks test, re-fetch the
jiffies time as the runstart time.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Because there is not guaranteed to be a full memory barrier between
the ->srcu_unlock_count increment of an srcu_read_unlock() and the
->srcu_lock_count increment of the next srcu_read_lock(), this next
srcu_read_lock() is not guaranteed to see the effect of the index flip
just prior to this comment. However, this next srcu_read_lock() will
execute a full memory barrier, so the srcu_read_lock() after that is
guaranteed to see that index flip.
This guarantee is illustrated by the following diagram of events and
the litmus test following that.
------------------------------------------------------------------------
READER UPDATER
------------- ----------
// idx is initially 0.
srcu_flip() {
smp_mb();
// RSCS
srcu_read_unlock() {
smp_mb();
idx++; // P
smp_mb(); // QQ
}
srcu_readers_unlock_idx(0) {
,--counted------------ count all unlock[0]; // Q
|
unlock[0]++; // X
}
smp_mb();
srcu_read_lock() {
READ(idx) = 0; ,---- count all lock[0]; // contributes imbalance of 1.
lock[0]++; ----counted |
smp_mb(); // PP } |
} |
|
// RSCS not going to effect above scan
|
srcu_read_unlock() { |
smp_mb(); |
unlock[0]++; |
} |
/
/
srcu_read_lock() { |
READ(idx); // Y -----cannot be counted because of P (has to sample idx as 1)
lock[1]++;
...
}
------------------------------------------------------------------------
This makes it similar to the store buffer pattern. Using X, Y, P and Q
annotated above, we get:
------------------------------------------------------------------------
READER UPDATER
X (write) P (write)
smp_mb(); //PP smp_mb(); //QQ
Y (read) Q (read)
------------------------------------------------------------------------
ASCII art courtesy of Joel Fernandes.
Reported-by: Joel Fernandes <joel@joelfernandes.org>
Reported-by: Boqun Feng <boqun.feng@gmail.com>
Reported-by: Frederic Weisbecker <frederic@kernel.org>
Reported-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The comment in srcu_readers_active_idx_check() following the smp_mb()
is out of date, hailing from a simpler time when preemption was disabled
across the bulk of __srcu_read_lock(). The fact that preemption was
disabled meant that the number of tasks that had fetched the old index
but not yet incremented counters was limited by the number of CPUs.
In our more complex modern times, the number of CPUs is no longer a limit.
This commit therefore updates this comment, additionally giving more
memory-ordering detail.
[ paulmck: Apply Nt->Nc feedback from Joel Fernandes. ]
Reported-by: Boqun Feng <boqun.feng@gmail.com>
Reported-by: Frederic Weisbecker <frederic@kernel.org>
Reported-by: "Joel Fernandes (Google)" <joel@joelfernandes.org>
Reported-by: Neeraj Upadhyay <neeraj.iitr10@gmail.com>
Reported-by: Uladzislau Rezki <urezki@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
The srcu_gp_start_if_needed() function now read-holds the srcu_struct
whose grace period is being started, which means that the corresponding
SRCU grace period cannot end. This in turn means that the SRCU
grace-period sequence number returned by rcu_seq_snap() cannot expire
during this time. And that means that the calls to rcu_seq_done() in
srcu_funnel_exp_start() and srcu_funnel_gp_start() can never return true.
This commit therefore removes these rcu_seq_done() checks, but adds checks
in kernels built with CONFIG_PROVE_RCU=y that splats if rcu_seq_done()
does somehow return true.
[ paulmck: Rearrange checks to handle kernels built with lockdep. ]
Signed-off-by: Pingfan Liu <kernelfans@gmail.com>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Cc: Frederic Weisbecker <frederic@kernel.org>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: rcu@vger.kernel.org
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Arnd Bergmann