There is no agreed-upon definition of spin_unlock_wait()'s semantics,
and it appears that all callers could do just as well with a lock/unlock
pair. This commit therefore removes the underlying arch-specific
arch_spin_unlock_wait() for all architectures providing them.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: <linux-arch@vger.kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Andrea Parri <parri.andrea@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Will Deacon <will.deacon@arm.com>
Acked-by: Boqun Feng <boqun.feng@gmail.com>
There is no agreed-upon definition of spin_unlock_wait()'s semantics,
and it appears that all callers could do just as well with a lock/unlock
pair. This commit therefore removes spin_unlock_wait() and related
definitions from core code.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Andrea Parri <parri.andrea@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
There is no agreed-upon definition of spin_unlock_wait()'s semantics,
and it appears that all callers could do just as well with a lock/unlock
pair. This commit therefore eliminates the spin_unlock_wait() call and
associated else-clause and hoists the then-clause's lock and unlock out of
the "if" statement. This should be safe from a performance perspective
because according to Tejun there should be few if any drivers that don't
set their own error handler.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: <linux-ide@vger.kernel.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Andrea Parri <parri.andrea@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
There is no agreed-upon definition of spin_unlock_wait()'s semantics,
and it appears that all callers could do just as well with a lock/unlock
pair. This commit therefore replaces the spin_unlock_wait() call in
exit_sem() with spin_lock() followed immediately by spin_unlock().
This should be safe from a performance perspective because exit_sem()
is rarely invoked in production.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Andrea Parri <parri.andrea@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Manfred Spraul <manfred@colorfullife.com>
There is no agreed-upon definition of spin_unlock_wait()'s semantics, and
it appears that all callers could do just as well with a lock/unlock pair.
This commit therefore replaces the spin_unlock_wait() call in do_exit()
with spin_lock() followed immediately by spin_unlock(). This should be
safe from a performance perspective because the lock is a per-task lock,
and this is happening only at task-exit time.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Andrea Parri <parri.andrea@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
There is no agreed-upon definition of spin_unlock_wait()'s semantics,
and it appears that all callers could do just as well with a lock/unlock
pair. This commit therefore replaces the spin_unlock_wait() call in
completion_done() with spin_lock() followed immediately by spin_unlock().
This should be safe from a performance perspective because the lock
will be held only the wakeup happens really quickly.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Andrea Parri <parri.andrea@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
This commit documents the situations in which RCU needs the
scheduling-clock interrupt to be enabled, along with the consequences
of failing to meet RCU's needs in this area.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
There are too many ways for the compiler to optimize (that is, break)
dependencies carried via integer values, so it is now permissible to
carry dependencies only via pointers. This commit catches up some of
the documentation on this point.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The memory-barriers.txt document contains an obsolete passage stating that
smp_read_barrier_depends() is required to force ordering for read-to-write
dependencies. We now know that this is not required, even for DEC Alpha.
This commit therefore updates this passage to state that read-to-write
dependencies are respected even without smp_read_barrier_depends().
Reported-by: Lance Roy <ldr709@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Andrea Parri <parri.andrea@gmail.com>
Cc: Jade Alglave <j.alglave@ucl.ac.uk>
Cc: Luc Maranget <luc.maranget@inria.fr>
[ paulmck: Reference control-dependencies sections and use WRITE_ONCE()
per Will Deacon. Correctly place split-cache paragraph while there. ]
Acked-by: Will Deacon <will.deacon@arm.com>
Implement MEMBARRIER_CMD_PRIVATE_EXPEDITED with IPIs using cpumask built
from all runqueues for which current thread's mm is the same as the
thread calling sys_membarrier. It executes faster than the non-expedited
variant (no blocking). It also works on NOHZ_FULL configurations.
Scheduler-wise, it requires a memory barrier before and after context
switching between processes (which have different mm). The memory
barrier before context switch is already present. For the barrier after
context switch:
* Our TSO archs can do RELEASE without being a full barrier. Look at
x86 spin_unlock() being a regular STORE for example. But for those
archs, all atomics imply smp_mb and all of them have atomic ops in
switch_mm() for mm_cpumask(), and on x86 the CR3 load acts as a full
barrier.
* From all weakly ordered machines, only ARM64 and PPC can do RELEASE,
the rest does indeed do smp_mb(), so there the spin_unlock() is a full
barrier and we're good.
* ARM64 has a very heavy barrier in switch_to(), which suffices.
* PPC just removed its barrier from switch_to(), but appears to be
talking about adding something to switch_mm(). So add a
smp_mb__after_unlock_lock() for now, until this is settled on the PPC
side.
Changes since v3:
- Properly document the memory barriers provided by each architecture.
Changes since v2:
- Address comments from Peter Zijlstra,
- Add smp_mb__after_unlock_lock() after finish_lock_switch() in
finish_task_switch() to add the memory barrier we need after storing
to rq->curr. This is much simpler than the previous approach relying
on atomic_dec_and_test() in mmdrop(), which actually added a memory
barrier in the common case of switching between userspace processes.
- Return -EINVAL when MEMBARRIER_CMD_SHARED is used on a nohz_full
kernel, rather than having the whole membarrier system call returning
-ENOSYS. Indeed, CMD_PRIVATE_EXPEDITED is compatible with nohz_full.
Adapt the CMD_QUERY mask accordingly.
Changes since v1:
- move membarrier code under kernel/sched/ because it uses the
scheduler runqueue,
- only add the barrier when we switch from a kernel thread. The case
where we switch from a user-space thread is already handled by
the atomic_dec_and_test() in mmdrop().
- add a comment to mmdrop() documenting the requirement on the implicit
memory barrier.
CC: Peter Zijlstra <peterz@infradead.org>
CC: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
CC: Boqun Feng <boqun.feng@gmail.com>
CC: Andrew Hunter <ahh@google.com>
CC: Maged Michael <maged.michael@gmail.com>
CC: gromer@google.com
CC: Avi Kivity <avi@scylladb.com>
CC: Benjamin Herrenschmidt <benh@kernel.crashing.org>
CC: Paul Mackerras <paulus@samba.org>
CC: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Dave Watson <davejwatson@fb.com>
The rcu_idle_exit() and rcu_idle_enter() functions are exported because
they were originally used by RCU_NONIDLE(), which was intended to
be usable from modules. However, RCU_NONIDLE() now instead uses
rcu_irq_enter_irqson() and rcu_irq_exit_irqson(), which are not
exported, and there have been no complaints.
This commit therefore removes the exports from rcu_idle_exit() and
rcu_idle_enter().
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
All current callers of rcu_idle_enter() have irqs disabled, and
rcu_idle_enter() relies on this, but doesn't check. This commit
therefore adds a RCU_LOCKDEP_WARN() to add some verification to the trust.
While we are there, pass "true" rather than "1" to rcu_eqs_enter().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
All callers to rcu_idle_enter() have irqs disabled, so there is no
point in rcu_idle_enter disabling them again. This commit therefore
replaces the irq disabling with a RCU_LOCKDEP_WARN().
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
This commit adds assertions verifying the consistency of the rcu_node
structure's ->blkd_tasks list and its ->gp_tasks, ->exp_tasks, and
->boost_tasks pointers. In particular, the ->blkd_tasks lists must be
empty except for leaf rcu_node structures.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Set disable_rcu_irq_enter on not only rcu_eqs_enter_common() but also
rcu_eqs_exit(), since rcu_eqs_exit() suffers from the same issue as was
fixed for rcu_eqs_enter_common() by commit 03ecd3f48e ("rcu/tracing:
Add rcu_disabled to denote when rcu_irq_enter() will not work").
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The _rcu_barrier_trace() function is a wrapper for trace_rcu_barrier(),
which needs TPS() protection for strings passed through the second
argument. However, it has escaped prior TPS()-ification efforts because
it _rcu_barrier_trace() does not start with "trace_". This commit
therefore adds the needed TPS() protection
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
These RCU waits were set to use interruptible waits to avoid the kthreads
contributing to system load average, even though they are not interruptible
as they are spawned from a kthread. Use the new TASK_IDLE swaits which makes
our goal clear, and removes confusion about these paths possibly being
interruptible -- they are not.
When the system is idle the RCU grace-period kthread will spend all its time
blocked inside the swait_event_interruptible(). If the interruptible() was
not used, then this kthread would contribute to the load average. This means
that an idle system would have a load average of 2 (or 3 if PREEMPT=y),
rather than the load average of 0 that almost fifty years of UNIX has
conditioned sysadmins to expect.
The same argument applies to swait_event_interruptible_timeout() use. The
RCU grace-period kthread spends its time blocked inside this call while
waiting for grace periods to complete. In particular, if there was only one
busy CPU, but that CPU was frequently invoking call_rcu(), then the RCU
grace-period kthread would spend almost all its time blocked inside the
swait_event_interruptible_timeout(). This would mean that the load average
would be 2 rather than the expected 1 for the single busy CPU.
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Tested-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
There are cases where folks are using an interruptible swait when
using kthreads. This is rather confusing given you'd expect
interruptible waits to be -- interruptible, but kthreads are not
interruptible ! The reason for such practice though is to avoid
having these kthreads contribute to the system load average.
When systems are idle some kthreads may spend a lot of time blocking if
using swait_event_timeout(). This would contribute to the system load
average. On systems without preemption this would mean the load average
of an idle system is bumped to 2 instead of 0. On systems with PREEMPT=y
this would mean the load average of an idle system is bumped to 3
instead of 0.
This adds proper API using TASK_IDLE to make such goals explicit and
avoid confusion.
Suggested-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Tested-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Luis R. Rodriguez <mcgrof@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
There is currently event tracing to track when a task is preempted
within a preemptible RCU read-side critical section, and also when that
task subsequently reaches its outermost rcu_read_unlock(), but none
indicating when a new grace period starts when that grace period must
wait on pre-existing readers that have been been preempted at least once
since the beginning of their current RCU read-side critical sections.
This commit therefore adds an event trace at grace-period start in
the case where there are such readers. Note that only the first
reader in the list is traced.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
This commit saves a few lines in kernel/rcu/rcu.h by moving to single-line
definitions for trivial functions, instead of the old style where the
two curly braces each get their own line.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Strings used in event tracing need to be specially handled, for example,
using the TPS() macro. Without the TPS() macro, although output looks
fine from within a running kernel, extracting traces from a crash dump
produces garbage instead of strings. This commit therefore adds the TPS()
macro to some unadorned strings that were passed to event-tracing macros.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Currently, the exit-time support for TASKS_RCU is open-coded in do_exit().
This commit creates exit_tasks_rcu_start() and exit_tasks_rcu_finish()
APIs for do_exit() use. This has the benefit of confining the use of the
tasks_rcu_exit_srcu variable to one file, allowing it to become static.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The actual use of TASKS_RCU is only when PREEMPT, otherwise RCU-sched
is used instead. This commit therefore makes synchronize_rcu_tasks()
and call_rcu_tasks() available always, but mapped to synchronize_sched()
and call_rcu_sched(), respectively, when !PREEMPT. This approach also
allows some #ifdefs to be removed from rcutorture.
Reported-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
Acked-by: Ingo Molnar <mingo@kernel.org>
There is no agreed-upon definition of spin_unlock_wait()'s semantics,
and it appears that all callers could do just as well with a lock/unlock
pair. This commit therefore replaces the spin_unlock_wait() call in
do_task_dead() with spin_lock() followed immediately by spin_unlock().
This should be safe from a performance perspective because the lock is
this tasks ->pi_lock, and this is called only after the task exits.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Andrea Parri <parri.andrea@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
[ paulmck: Drop smp_mb() based on Peter Zijlstra's analysis:
http://lkml.kernel.org/r/20170811144150.26gowhxte7ri5fpk@hirez.programming.kicks-ass.net ]
Per-cpu workqueues have been tripping CPU affinity sanity checks while
a CPU is being offlined. A per-cpu kworker ends up running on a CPU
which isn't its target CPU while the CPU is online but inactive.
While the scheduler allows kthreads to wake up on an online but
inactive CPU, it doesn't allow a running kthread to be migrated to
such a CPU, which leads to an odd situation where setting affinity on
a sleeping and running kthread leads to different results.
Each mem-reclaim workqueue has one rescuer which guarantees forward
progress and the rescuer needs to bind itself to the CPU which needs
help in making forward progress; however, due to the above issue,
while set_cpus_allowed_ptr() succeeds, the rescuer doesn't end up on
the correct CPU if the CPU is in the process of going offline,
tripping the sanity check and executing the work item on the wrong
CPU.
This patch updates __migrate_task() so that kthreads can be migrated
into an inactive but online CPU.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Reported-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tree RCU guarantees that every online CPU has a memory barrier between
any given grace period and any of that CPU's RCU read-side sections that
must be ordered against that grace period. Since RCU doesn't always
know where read-side critical sections are, the actual implementation
guarantees order against prior and subsequent non-idle non-offline code,
whether in an RCU read-side critical section or not. As a result, there
does not need to be a memory barrier at the end of synchronize_rcu()
and friends because the ordering internal to the grace period has
ordered every CPU's post-grace-period execution against each CPU's
pre-grace-period execution, again for all non-idle online CPUs.
In contrast, SRCU can have non-idle online CPUs that are completely
uninvolved in a given SRCU grace period, for example, a CPU that
never runs any SRCU read-side critical sections and took no part in
the grace-period processing. It is in theory possible for a given
synchronize_srcu()'s wakeup to be delivered to a CPU that was completely
uninvolved in the prior SRCU grace period, which could mean that the
code following that synchronize_srcu() would end up being unordered with
respect to both the grace period and any pre-existing SRCU read-side
critical sections.
This commit therefore adds an smp_mb() to the end of __synchronize_srcu(),
which prevents this scenario from occurring.
Reported-by: Lance Roy <ldr709@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Lance Roy <ldr709@gmail.com>
Cc: <stable@vger.kernel.org> # 4.12.x
After adopting callbacks from a newly offlined CPU, the adopting CPU
checks to make sure that its callback list's count is zero only if the
list has no callbacks and vice versa. Unfortunately, it does so after
enabling interrupts, which means that false positives are possible due to
interrupt handlers invoking call_rcu(). Although these false positives
are improbable, rcutorture did make it happen once.
This commit therefore moves this check to an irq-disabled region of code,
thus suppressing the false positive.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Given changes to callback migration, rcu_cblist_head(),
rcu_cblist_tail(), rcu_cblist_count_cbs(), rcu_segcblist_segempty(),
rcu_segcblist_dequeued_lazy(), and rcu_segcblist_new_cbs() are
no longer used. This commit therefore removes them.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Given that the rcu_state structure's >orphan_pend and ->orphan_done
fields are used only during migration of callbacks from the recently
offlined CPU to a surviving CPU, if rcu_send_cbs_to_orphanage() and
rcu_adopt_orphan_cbs() are combined, these fields can become local
variables in the combined function. This commit therefore combines
rcu_send_cbs_to_orphanage() and rcu_adopt_orphan_cbs() into a new
rcu_segcblist_merge() function and removes the ->orphan_pend and
->orphan_done fields.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
When migrating callbacks from a newly offlined CPU, we are already
holding the root rcu_node structure's lock, so it costs almost nothing
to advance and accelerate the newly migrated callbacks. This patch
therefore makes this advancing and acceleration happen.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The ->orphan_lock is acquired and released only within the
rcu_migrate_callbacks() function, which now acquires the root rcu_node
structure's ->lock. This commit therefore eliminates the ->orphan_lock
in favor of the root rcu_node structure's ->lock.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
It is possible that the outgoing CPU is unaware of recent grace periods,
and so it is also possible that some of its pending callbacks are actually
ready to be invoked. The current callback-migration code would needlessly
force these callbacks to pass through another grace period. This commit
therefore invokes rcu_advance_cbs() on the outgoing CPU's callbacks in
order to give them full credit for having passed through any recent
grace periods.
This also fixes an odd theoretical bug where there are no callbacks in
the system except for those on the outgoing CPU, none of those callbacks
have yet been associated with a grace-period number, there is never again
another callback registered, and the surviving CPU never again takes a
scheduling-clock interrupt, never goes idle, and never enters nohz_full
userspace execution. Yes, this is (just barely) possible. It requires
that the surviving CPU be a nohz_full CPU, that its scheduler-clock
interrupt be shut off, and that it loop forever in the kernel. You get
bonus points if you can make this one happen! ;-)
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
RCU's CPU-hotplug callback-migration code first moves the outgoing
CPU's callbacks to ->orphan_done and ->orphan_pend, and only then
moves them to the NOCB callback list. This commit avoids the
extra step (and simplifies the code) by moving the callbacks directly
from the outgoing CPU's callback list to the NOCB callback list.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The current CPU-hotplug RCU-callback-migration code checks
for the source (newly offlined) CPU being a NOCBs CPU down in
rcu_send_cbs_to_orphanage(). This commit simplifies callback migration a
bit by moving this check up to rcu_migrate_callbacks(). This commit also
adds a check for the source CPU having no callbacks, which eases analysis
of the rcu_send_cbs_to_orphanage() and rcu_adopt_orphan_cbs() functions.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The rcu_node structure's ->n_cbs_orphaned and ->n_cbs_adopted fields
are updated, but never read. This commit therefore removes them.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The torture status line contains a series of values preceded by "onoff:".
The last value in that line, the one preceding the "HZ=" string, is
always zero. The reason that it is always zero is that torture_offline()
was incrementing the sum_offl pointer instead of the value that this
pointer referenced. This commit therefore makes this increment operate
on the statistic rather than the pointer to the statistic.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The update of the ->expmaskinitnext and of ->ncpus are unsynchronized,
with the value of ->ncpus being incremented long before the corresponding
->expmaskinitnext mask is updated. If an RCU expedited grace period
sees ->ncpus change, it will update the ->expmaskinit masks from the new
->expmaskinitnext masks. But it is possible that ->ncpus has already
been updated, but the ->expmaskinitnext masks still have their old values.
For the current expedited grace period, no harm done. The CPU could not
have been online before the grace period started, so there is no need to
wait for its non-existent pre-existing readers.
But the next RCU expedited grace period is in a world of hurt. The value
of ->ncpus has already been updated, so this grace period will assume
that the ->expmaskinitnext masks have not changed. But they have, and
they won't be taken into account until the next never-been-online CPU
comes online. This means that RCU will be ignoring some CPUs that it
should be paying attention to.
The solution is to update ->ncpus and ->expmaskinitnext while holding
the ->lock for the rcu_node structure containing the ->expmaskinitnext
mask. Because smp_store_release() is now used to update ->ncpus and
smp_load_acquire() is now used to locklessly read it, if the expedited
grace period sees ->ncpus change, then the updating CPU has to
already be holding the corresponding ->lock. Therefore, when the
expedited grace period later acquires that ->lock, it is guaranteed
to see the new value of ->expmaskinitnext.
On the other hand, if the expedited grace period loads ->ncpus just
before an update, earlier full memory barriers guarantee that
the incoming CPU isn't far enough along to be running any RCU readers.
This commit therefore makes the required change.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
RCU callbacks must be migrated away from an outgoing CPU, and this is
done near the end of the CPU-hotplug operation, after the outgoing CPU is
long gone. Unfortunately, this means that other CPU-hotplug callbacks
can execute while the outgoing CPU's callbacks are still immobilized
on the long-gone CPU's callback lists. If any of these CPU-hotplug
callbacks must wait, either directly or indirectly, for the invocation
of any of the immobilized RCU callbacks, the system will hang.
This commit avoids such hangs by migrating the callbacks away from the
outgoing CPU immediately upon its departure, shortly after the return
from __cpu_die() in takedown_cpu(). Thus, RCU is able to advance these
callbacks and invoke them, which allows all the after-the-fact CPU-hotplug
callbacks to wait on these RCU callbacks without risk of a hang.
While in the neighborhood, this commit also moves rcu_send_cbs_to_orphanage()
and rcu_adopt_orphan_cbs() under a pre-existing #ifdef to avoid including
dead code on the one hand and to avoid define-without-use warnings on the
other hand.
Reported-by: Jeffrey Hugo <jhugo@codeaurora.org>
Link: http://lkml.kernel.org/r/db9c91f6-1b17-6136-84f0-03c3c2581ab4@codeaurora.org
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Anna-Maria Gleixner <anna-maria@linutronix.de>
Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Richard Weinberger <richard@nod.at>
There is no agreed-upon definition of spin_unlock_wait()'s semantics,
and it appears that all callers could do just as well with a lock/unlock
pair. This commit therefore replaces the spin_unlock_wait() call in
task_work_run() with a spin_lock_irq() and a spin_unlock_irq() aruond
the cmpxchg() dequeue loop. This should be safe from a performance
perspective because ->pi_lock is local to the task and because calls to
the other side of the race, task_work_cancel(), should be rare.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
As we want to remove spin_unlock_wait() and replace it with explicit
spin_lock()/spin_unlock() calls, we can use this to simplify the
locking.
In addition:
- Reading nf_conntrack_locks_all needs ACQUIRE memory ordering.
- The new code avoids the backwards loop.
Only slightly tested, I did not manage to trigger calls to
nf_conntrack_all_lock().
V2: With improved comments, to clearly show how the barriers
pair.
Fixes: b16c29191d ("netfilter: nf_conntrack: use safer way to lock all buckets")
Signed-off-by: Manfred Spraul <manfred@colorfullife.com>
Cc: <stable@vger.kernel.org>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: Pablo Neira Ayuso <pablo@netfilter.org>
Cc: netfilter-devel@vger.kernel.org
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
There is still considerable confusion as to the semantics of
spin_unlock_wait(), but there seems to be universal agreement that
it is not that of a lock/unlock pair. This commit therefore removes
the comment added by 6016ffc387 ("atomics: Add header comment so
spin_unlock_wait()") in order to prevent at least that flavor of
confusion.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The handling of RCU's no-CBs CPUs has a maintenance headache, namely
that if call_rcu() is invoked with interrupts disabled, the rcuo kthread
wakeup must be defered to a point where we can be sure that scheduler
locks are not held. Of course, there are a lot of code paths leading
from an interrupts-disabled invocation of call_rcu(), and missing any
one of these can result in excessive callback-invocation latency, and
potentially even system hangs.
This commit therefore uses a timer to guarantee that the wakeup will
eventually occur. If one of the deferred-wakeup points kicks in, then
the timer is simply cancelled.
This commit also fixes up an incomplete removal of commits that were
intended to plug remaining exit paths, which should have the added
benefit of reducing the overhead of RCU's context-switch hooks. In
addition, it simplifies leader-to-follower callback-list handoff by
introducing locking. The call_rcu()-to-leader handoff continues to
use atomic operations in order to maintain good real-time latency for
common-case use of call_rcu().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
[ paulmck: Dan Carpenter fix for mod_timer() usage bug found by smatch. ]
The Linux kernel invokes call_rcu() from various interrupt/softirq
handlers, but rcutorture does not. This commit therefore adds this
behavior to rcutorture's repertoire.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
This commit augments the grace-period-kthread starvation debugging
messages by adding the last CPU that ran the kthread.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
This commit removes an unused local variable named ts_rem that is
marked __maybe_unused. Yes, the variable was assigned to, but it
was never used beyond that point, hence not needed.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
It appears that at least some of the rcutorture writer stall messages
coincide with unusually long CPU-online operations, for example, no
fewer than 205 seconds in a recent test. It is of course possible that
the writer stall is not unrelated to this unusually long CPU-hotplug
operation, and so this commit adds the rcutorture writer task's CPU to
the stall message to gain more information about this possible connection.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The maxcpus= kernel boot parameter limits the number of CPUs brought
online at boot time, but it does nothing to prevent additional CPUs
from being brought up later. Placing a hard cap on the total number
of CPUs is instead the job of the nr_cpus= boot parameter. This commit
therefore switches the configfrag_boot_cpus() shell function from maxcpus=
to nr_cpus=. This commit also adds a nr_cpus=43 kernel parameter to RCU's
TREE01 test scenario, but retains the maxcpus=8 kernel parameter in order
to test the ability of RCU expedited grace periods to handle new CPUs
coming online for the first time during grace-period initialization.
Finally, this commit makes the torture scheduling allow maxcpus= to
override other means of specifying the number of CPUs to allow for.
This last works because the torture kernel modules size their workloads
based on the number of CPUs present at the start of the test, not the
ultimate number of CPUs.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
