An 80-CPU system with a context-switch-heavy workload can require so
many NOCB kthread wakeups that the RCU grace-period kthreads spend several
tens of percent of a CPU just awakening things. This clearly will not
scale well: If you add enough CPUs, the RCU grace-period kthreads would
get behind, increasing grace-period latency.
To avoid this problem, this commit divides the NOCB kthreads into leaders
and followers, where the grace-period kthreads awaken the leaders each of
whom in turn awakens its followers. By default, the number of groups of
kthreads is the square root of the number of CPUs, but this default may
be overridden using the rcutree.rcu_nocb_leader_stride boot parameter.
This reduces the number of wakeups done per grace period by the RCU
grace-period kthread by the square root of the number of CPUs, but of
course by shifting those wakeups to the leaders. In addition, because
the leaders do grace periods on behalf of their respective followers,
the number of wakeups of the followers decreases by up to a factor of two.
Instead of being awakened once when new callbacks arrive and again
at the end of the grace period, the followers are awakened only at
the end of the grace period.
For a numerical example, in a 4096-CPU system, the grace-period kthread
would awaken 64 leaders, each of which would awaken its 63 followers
at the end of the grace period. This compares favorably with the 79
wakeups for the grace-period kthread on an 80-CPU system.
Reported-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Commit ac1bea8578 (Make cond_resched() report RCU quiescent states)
fixed a problem where a CPU looping in the kernel with but one runnable
task would give RCU CPU stall warnings, even if the in-kernel loop
contained cond_resched() calls. Unfortunately, in so doing, it introduced
performance regressions in Anton Blanchard's will-it-scale "open1" test.
The problem appears to be not so much the increased cond_resched() path
length as an increase in the rate at which grace periods complete, which
increased per-update grace-period overhead.
This commit takes a different approach to fixing this bug, mainly by
moving the RCU-visible quiescent state from cond_resched() to
rcu_note_context_switch(), and by further reducing the check to a
simple non-zero test of a single per-CPU variable. However, this
approach requires that the force-quiescent-state processing send
resched IPIs to the offending CPUs. These will be sent only once
the grace period has reached an age specified by the boot/sysfs
parameter rcutree.jiffies_till_sched_qs, or once the grace period
reaches an age halfway to the point at which RCU CPU stall warnings
will be emitted, whichever comes first.
Reported-by: Dave Hansen <dave.hansen@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Lameter <cl@gentwo.org>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
[ paulmck: Made rcu_momentary_dyntick_idle() as suggested by the
ktest build robot. Also fixed smp_mb() comment as noted by
Oleg Nesterov. ]
Merge with e552592e (Reduce overhead of cond_resched() checks for RCU)
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The variable and struct both having the name "rcu_state" confuses
sparse in some situations, so this commit changes the variable to
"rcu_state_p" in order to avoid this confusion. This also makes
things easier for human readers.
Signed-off-by: Uma Sharma <uma.sharma523@gmail.com>
[ paulmck: Changed the declaration and several additional uses. ]
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
__this_cpu_ptr is being phased out.
One special case is increment_cpu_stall_ticks().
A per cpu variable is incremented so use raw_cpu_inc().
Cc: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Currently, small systems move back into RCU_SYSIDLE_NOT from
RCU_SYSIDLE_SHORT and large systems do not. This works because moving
aggressively to RCU_SYSIDLE_NOT affects only performance, not correctness,
and on small systems, the performance impact should be negligible. That
said, this difference does make RCU a bit more complex, and RCU does not
seem to be suffering from any lack of complexity. This commit therefore
adjusts small-system operation to match that of large systems, so that
the state never moves back to RCU_SYSIDLE_NOT from RCU_SYSIDLE_SHORT.
Reported-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Currently, RCU binds the grace-period kthreads to the timekeeping
CPU only if CONFIG_NO_HZ_FULL_SYSIDLE=y. This means that these
kthreads must be bound manually when CONFIG_NO_HZ_FULL_SYSIDLE=n and
CONFIG_NO_HZ_FULL=y: Otherwise, these kthreads will induce OS jitter on
random CPUs. Given that we are trying to reduce the amount of manual
tweaking required to make CONFIG_NO_HZ_FULL=y work nicely, this commit
makes this binding happen when CONFIG_NO_HZ_FULL=y, even in cases where
CONFIG_NO_HZ_FULL_SYSIDLE=n.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
This patch merges the function rcu_force_quiescent_state() with
rcu_sched_force_quiescent_state(), using the rcu_state pointer. Firstly,
the rcu_sched_force_quiescent_state() function is deleted from the file
kernel/rcu/tree.c. Also, the rcu_force_quiescent_state() function that was
calling force_quiescent_state with the argument rcu_preempt_state pointer
was deleted as well. The new function that combines the old ones uses
the rcu_state pointer and is located after rcu_batches_completed_bh()
in kernel/rcu/tree.c.
Signed-off-by: Andreea-Cristina Bernat <bernat.ada@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
kfree_call_rcu is defined two times. When defined under CONFIG_TREE_PREEMPT_RCU,
it uses rcu_preempt_state. Otherwise, it uses rcu_sched_state.
This patch uses the rcu_state_pointer to combine the two definitions into one.
The resulting function is placed after the closing of the preprocessor
conditional CONFIG_TREE_PREEMPT_RCU.
Signed-off-by: Andreea-Cristina Bernat <bernat.ada@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
The rcu_start_gp_advanced() function currently uses irq_work_queue()
to defer wakeups of the RCU grace-period kthread. This deferring
is necessary to avoid RCU-scheduler deadlocks involving the rcu_node
structure's lock, meaning that RCU cannot call any of the scheduler's
wake-up functions while holding one of these locks.
Unfortunately, the second and subsequent calls to irq_work_queue() are
ignored, and the first call will be ignored (aside from queuing the work
item) if the scheduler-clock tick is turned off. This is OK for many
uses, especially those where irq_work_queue() is called from an interrupt
or softirq handler, because in those cases the scheduler-clock-tick state
will be re-evaluated, which will turn the scheduler-clock tick back on.
On the next tick, any deferred work will then be processed.
However, this strategy does not always work for RCU, which can be invoked
at process level from idle CPUs. In this case, the tick might never
be turned back on, indefinitely defering a grace-period start request.
Note that the RCU CPU stall detector cannot see this condition, because
there is no RCU grace period in progress. Therefore, we can (and do!)
see long tens-of-seconds stalls in grace-period handling. In theory,
we could see a full grace-period hang, but rcutorture testing to date
has seen only the tens-of-seconds stalls. Event tracing demonstrates
that irq_work_queue() is being called repeatedly to no effect during
these stalls: The "newreq" event appears repeatedly from a task that is
not one of the grace-period kthreads.
In theory, irq_work_queue() might be fixed to avoid this sort of issue,
but RCU's requirements are unusual and it is quite straightforward to pass
wake-up responsibility up through RCU's call chain, so that the wakeup
happens when the offending locks are released.
This commit therefore makes this change. The rcu_start_gp_advanced(),
rcu_start_future_gp(), rcu_accelerate_cbs(), rcu_advance_cbs(),
__note_gp_changes(), and rcu_start_gp() functions now return a boolean
which indicates when a wake-up is needed. A new rcu_gp_kthread_wake()
does the wakeup when it is necessary and safe to do so: No self-wakes,
no wake-ups if the ->gp_flags field indicates there is no need (as in
someone else did the wake-up before we got around to it), and no wake-ups
before the grace-period kthread has been created.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
In the old days, the only source of requests for future grace periods
was NOCB CPUs. This has changed: CPUs routinely post requests for
future grace periods in order to promote power efficiency and reduce
OS jitter with minimal impact on grace-period latency. This commit
therefore updates cpu_needs_another_gp() to invoke rcu_future_needs_gp()
instead of rcu_nocb_needs_gp(). The latter is no longer used, so is
now removed. This commit also adds tracing for the irq_work_queue()
wakeup case.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Liu Ping Fan <kernelfans@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
If CONFIG_RCU_NOCB_CPU_ALL=y, then no CPU will ever have RCU callbacks
because these callbacks will instead be handled by the rcuo kthreads.
However, the current version of RCU_FAST_NO_HZ nevertheless checks for RCU
callbacks. This commit therefore creates static inline implementations
of rcu_prepare_for_idle() and rcu_cleanup_after_idle() that are no-ops
when CONFIG_RCU_NOCB_CPU_ALL=y.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
If CONFIG_RCU_NOCB_CPU_ALL=y, then rcu_needs_cpu() will always
return false, however, the current version nevertheless checks
for RCU callbacks. This commit therefore creates a static inline
implementation of rcu_needs_cpu() that unconditionally returns false
when CONFIG_RCU_NOCB_CPU_ALL=y.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
If CONFIG_RCU_NOCB_CPU_ALL=y, then rcu_is_nocb_cpu() will always
return true, however, the current version nevertheless checks
rcu_nocb_mask. This commit therefore creates a static inline
implementation of rcu_is_nocb_cpu() that unconditionally returns
true when CONFIG_RCU_NOCB_CPU_ALL=y.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
This commit fixes a grammar issue in the rcu_nohz_full_cpu() comment
header, so that it is clear that the plural is CPUs not Kconfig options.
Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
All of the RCU source files have the usual GPL header, which contains a
long-obsolete postal address for FSF. To avoid the need to track the
FSF office's movements, this commit substitutes the URL where GPL may
be found.
Reported-by: Greg KH <gregkh@linuxfoundation.org>
Reported-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Pull RCU updates from Ingo Molnar:
- add RCU torture scripts/tooling
- static analysis improvements
- update RCU documentation
- miscellaneous fixes
* 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (52 commits)
rcu: Remove "extern" from function declarations in kernel/rcu/rcu.h
rcu: Remove "extern" from function declarations in include/linux/*rcu*.h
rcu/torture: Dynamically allocate SRCU output buffer to avoid overflow
rcu: Don't activate RCU core on NO_HZ_FULL CPUs
rcu: Warn on allegedly impossible rcu_read_unlock_special() from irq
rcu: Add an RCU_INITIALIZER for global RCU-protected pointers
rcu: Make rcu_assign_pointer's assignment volatile and type-safe
bonding: Use RCU_INIT_POINTER() for better overhead and for sparse
rcu: Add comment on evaluate-once properties of rcu_assign_pointer().
rcu: Provide better diagnostics for blocking in RCU callback functions
rcu: Improve SRCU's grace-period comments
rcu: Fix CONFIG_RCU_FANOUT_EXACT for odd fanout/leaf values
rcu: Fix coccinelle warnings
rcutorture: Stop tracking FSF's postal address
rcutorture: Move checkarg to functions.sh
rcutorture: Flag errors and warnings with color coding
rcutorture: Record results from repeated runs of the same test scenario
rcutorture: Test summary at end of run with less chattiness
rcutorture: Update comment in kvm.sh listing typical RCU trace events
rcutorture: Add tracing-enabled version of TREE08
...
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.15 (GNU/Linux)
iQEcBAABAgAGBQJSrhGrAAoJEHm+PkMAQRiGsNoH/jIK3CsQ2lbW7yRLXmfgtbzz
i2Kep6D4SDvmaLpLYOVC8xNYTiE8jtTbSXHomwP5wMZ63MQDhBfnEWsEWqeZ9+D9
3Q46p0QWuoBgYu2VGkoxTfygkT6hhSpwWIi3SeImbY4fg57OHiUil/+YGhORM4Qc
K4549OCTY3sIrgmWL77gzqjRUo+pQ4C73NKqZ3+5nlOmYBZC1yugk8mFwEpQkwhK
4NRNU760Fo+XIht/bINqRiPMddzC15p0mxvJy3cDW8bZa1tFSS9SB7AQUULBbcHL
+2dFlFOEb5SV1sNiNPrJ0W+h2qUh2e7kPB0F8epaBppgbwVdyQoC2u4uuLV2ZN0=
=lI2r
-----END PGP SIGNATURE-----
Merge tag 'v3.13-rc4' into core/locking
Merge Linux 3.13-rc4, to refresh this rather old tree with the latest fixes.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
RCU must ensure that there is the equivalent of a full memory
barrier between any memory access preceding grace period and any
memory access following that same grace period, regardless of
which CPU(s) happen to execute the two memory accesses.
Therefore, downgrading UNLOCK+LOCK to no longer imply a full
memory barrier requires some adjustments to RCU.
This commit therefore adds smp_mb__after_unlock_lock()
invocations as needed after the RCU lock acquisitions that need
to be part of a full-memory-barrier UNLOCK+LOCK.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: <linux-arch@vger.kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/1386799151-2219-7-git-send-email-paulmck@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Whenever a CPU receives a scheduling-clock interrupt, RCU checks to see
if the RCU core needs anything from this CPU. If so, RCU raises
RCU_SOFTIRQ to carry out any needed processing.
This approach has worked well historically, but it is undesirable on
NO_HZ_FULL CPUs. Such CPUs are expected to spend almost all of their time
in userspace, so that scheduling-clock interrupts can be disabled while
there is only one runnable task on the CPU in question. Unfortunately,
raising any softirq has the potential to wake up ksoftirqd, which would
provide the second runnable task on that CPU, preventing disabling of
scheduling-clock interrupts.
What is needed instead is for RCU to leave NO_HZ_FULL CPUs alone,
relying on the grace-period kthreads' quiescent-state forcing to
do any needed RCU work on behalf of those CPUs.
This commit therefore refrains from raising RCU_SOFTIRQ on any
NO_HZ_FULL CPUs during any grace periods that have been in effect
for less than one second. The one-second limit handles the case
where an inappropriate workload is running on a NO_HZ_FULL CPU
that features lots of scheduling-clock interrupts, but no idle
or userspace time.
Reported-by: Mike Galbraith <bitbucket@online.de>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Mike Galbraith <bitbucket@online.de>
Toasted-by: Frederic Weisbecker <fweisbec@gmail.com>
After commit #10f39bb1b2c1 (rcu: protect __rcu_read_unlock() against
scheduler-using irq handlers), it is no longer possible to enter
the main body of rcu_read_lock_special() from an NMI, interrupt, or
softirq handler. In theory, this implies that the check for "in_irq()
|| in_serving_softirq()" must always fail, so that in theory this check
could be removed entirely.
In practice, this commit wraps this condition with a WARN_ON_ONCE().
If this warning never triggers, then the condition will be removed
entirely.
[ paulmck: And one way of triggering the WARN_ON() is if a scheduling
clock interrupt occurs in an RCU read-side critical section, setting
RCU_READ_UNLOCK_NEED_QS, which is handled by rcu_read_unlock_special().
Updated this commit to return if only that bit was set. ]
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Dave Jones got the following lockdep splat:
> ======================================================
> [ INFO: possible circular locking dependency detected ]
> 3.12.0-rc3+ #92 Not tainted
> -------------------------------------------------------
> trinity-child2/15191 is trying to acquire lock:
> (&rdp->nocb_wq){......}, at: [<ffffffff8108ff43>] __wake_up+0x23/0x50
>
> but task is already holding lock:
> (&ctx->lock){-.-...}, at: [<ffffffff81154c19>] perf_event_exit_task+0x109/0x230
>
> which lock already depends on the new lock.
>
>
> the existing dependency chain (in reverse order) is:
>
> -> #3 (&ctx->lock){-.-...}:
> [<ffffffff810cc243>] lock_acquire+0x93/0x200
> [<ffffffff81733f90>] _raw_spin_lock+0x40/0x80
> [<ffffffff811500ff>] __perf_event_task_sched_out+0x2df/0x5e0
> [<ffffffff81091b83>] perf_event_task_sched_out+0x93/0xa0
> [<ffffffff81732052>] __schedule+0x1d2/0xa20
> [<ffffffff81732f30>] preempt_schedule_irq+0x50/0xb0
> [<ffffffff817352b6>] retint_kernel+0x26/0x30
> [<ffffffff813eed04>] tty_flip_buffer_push+0x34/0x50
> [<ffffffff813f0504>] pty_write+0x54/0x60
> [<ffffffff813e900d>] n_tty_write+0x32d/0x4e0
> [<ffffffff813e5838>] tty_write+0x158/0x2d0
> [<ffffffff811c4850>] vfs_write+0xc0/0x1f0
> [<ffffffff811c52cc>] SyS_write+0x4c/0xa0
> [<ffffffff8173d4e4>] tracesys+0xdd/0xe2
>
> -> #2 (&rq->lock){-.-.-.}:
> [<ffffffff810cc243>] lock_acquire+0x93/0x200
> [<ffffffff81733f90>] _raw_spin_lock+0x40/0x80
> [<ffffffff810980b2>] wake_up_new_task+0xc2/0x2e0
> [<ffffffff81054336>] do_fork+0x126/0x460
> [<ffffffff81054696>] kernel_thread+0x26/0x30
> [<ffffffff8171ff93>] rest_init+0x23/0x140
> [<ffffffff81ee1e4b>] start_kernel+0x3f6/0x403
> [<ffffffff81ee1571>] x86_64_start_reservations+0x2a/0x2c
> [<ffffffff81ee1664>] x86_64_start_kernel+0xf1/0xf4
>
> -> #1 (&p->pi_lock){-.-.-.}:
> [<ffffffff810cc243>] lock_acquire+0x93/0x200
> [<ffffffff8173419b>] _raw_spin_lock_irqsave+0x4b/0x90
> [<ffffffff810979d1>] try_to_wake_up+0x31/0x350
> [<ffffffff81097d62>] default_wake_function+0x12/0x20
> [<ffffffff81084af8>] autoremove_wake_function+0x18/0x40
> [<ffffffff8108ea38>] __wake_up_common+0x58/0x90
> [<ffffffff8108ff59>] __wake_up+0x39/0x50
> [<ffffffff8110d4f8>] __call_rcu_nocb_enqueue+0xa8/0xc0
> [<ffffffff81111450>] __call_rcu+0x140/0x820
> [<ffffffff81111b8d>] call_rcu+0x1d/0x20
> [<ffffffff81093697>] cpu_attach_domain+0x287/0x360
> [<ffffffff81099d7e>] build_sched_domains+0xe5e/0x10a0
> [<ffffffff81efa7fc>] sched_init_smp+0x3b7/0x47a
> [<ffffffff81ee1f4e>] kernel_init_freeable+0xf6/0x202
> [<ffffffff817200be>] kernel_init+0xe/0x190
> [<ffffffff8173d22c>] ret_from_fork+0x7c/0xb0
>
> -> #0 (&rdp->nocb_wq){......}:
> [<ffffffff810cb7ca>] __lock_acquire+0x191a/0x1be0
> [<ffffffff810cc243>] lock_acquire+0x93/0x200
> [<ffffffff8173419b>] _raw_spin_lock_irqsave+0x4b/0x90
> [<ffffffff8108ff43>] __wake_up+0x23/0x50
> [<ffffffff8110d4f8>] __call_rcu_nocb_enqueue+0xa8/0xc0
> [<ffffffff81111450>] __call_rcu+0x140/0x820
> [<ffffffff81111bb0>] kfree_call_rcu+0x20/0x30
> [<ffffffff81149abf>] put_ctx+0x4f/0x70
> [<ffffffff81154c3e>] perf_event_exit_task+0x12e/0x230
> [<ffffffff81056b8d>] do_exit+0x30d/0xcc0
> [<ffffffff8105893c>] do_group_exit+0x4c/0xc0
> [<ffffffff810589c4>] SyS_exit_group+0x14/0x20
> [<ffffffff8173d4e4>] tracesys+0xdd/0xe2
>
> other info that might help us debug this:
>
> Chain exists of:
> &rdp->nocb_wq --> &rq->lock --> &ctx->lock
>
> Possible unsafe locking scenario:
>
> CPU0 CPU1
> ---- ----
> lock(&ctx->lock);
> lock(&rq->lock);
> lock(&ctx->lock);
> lock(&rdp->nocb_wq);
>
> *** DEADLOCK ***
>
> 1 lock held by trinity-child2/15191:
> #0: (&ctx->lock){-.-...}, at: [<ffffffff81154c19>] perf_event_exit_task+0x109/0x230
>
> stack backtrace:
> CPU: 2 PID: 15191 Comm: trinity-child2 Not tainted 3.12.0-rc3+ #92
> ffffffff82565b70 ffff880070c2dbf8 ffffffff8172a363 ffffffff824edf40
> ffff880070c2dc38 ffffffff81726741 ffff880070c2dc90 ffff88022383b1c0
> ffff88022383aac0 0000000000000000 ffff88022383b188 ffff88022383b1c0
> Call Trace:
> [<ffffffff8172a363>] dump_stack+0x4e/0x82
> [<ffffffff81726741>] print_circular_bug+0x200/0x20f
> [<ffffffff810cb7ca>] __lock_acquire+0x191a/0x1be0
> [<ffffffff810c6439>] ? get_lock_stats+0x19/0x60
> [<ffffffff8100b2f4>] ? native_sched_clock+0x24/0x80
> [<ffffffff810cc243>] lock_acquire+0x93/0x200
> [<ffffffff8108ff43>] ? __wake_up+0x23/0x50
> [<ffffffff8173419b>] _raw_spin_lock_irqsave+0x4b/0x90
> [<ffffffff8108ff43>] ? __wake_up+0x23/0x50
> [<ffffffff8108ff43>] __wake_up+0x23/0x50
> [<ffffffff8110d4f8>] __call_rcu_nocb_enqueue+0xa8/0xc0
> [<ffffffff81111450>] __call_rcu+0x140/0x820
> [<ffffffff8109bc8f>] ? local_clock+0x3f/0x50
> [<ffffffff81111bb0>] kfree_call_rcu+0x20/0x30
> [<ffffffff81149abf>] put_ctx+0x4f/0x70
> [<ffffffff81154c3e>] perf_event_exit_task+0x12e/0x230
> [<ffffffff81056b8d>] do_exit+0x30d/0xcc0
> [<ffffffff810c9af5>] ? trace_hardirqs_on_caller+0x115/0x1e0
> [<ffffffff810c9bcd>] ? trace_hardirqs_on+0xd/0x10
> [<ffffffff8105893c>] do_group_exit+0x4c/0xc0
> [<ffffffff810589c4>] SyS_exit_group+0x14/0x20
> [<ffffffff8173d4e4>] tracesys+0xdd/0xe2
The underlying problem is that perf is invoking call_rcu() with the
scheduler locks held, but in NOCB mode, call_rcu() will with high
probability invoke the scheduler -- which just might want to use its
locks. The reason that call_rcu() needs to invoke the scheduler is
to wake up the corresponding rcuo callback-offload kthread, which
does the job of starting up a grace period and invoking the callbacks
afterwards.
One solution (championed on a related problem by Lai Jiangshan) is to
simply defer the wakeup to some point where scheduler locks are no longer
held. Since we don't want to unnecessarily incur the cost of such
deferral, the task before us is threefold:
1. Determine when it is likely that a relevant scheduler lock is held.
2. Defer the wakeup in such cases.
3. Ensure that all deferred wakeups eventually happen, preferably
sooner rather than later.
We use irqs_disabled_flags() as a proxy for relevant scheduler locks
being held. This works because the relevant locks are always acquired
with interrupts disabled. We may defer more often than needed, but that
is at least safe.
The wakeup deferral is tracked via a new field in the per-CPU and
per-RCU-flavor rcu_data structure, namely ->nocb_defer_wakeup.
This flag is checked by the RCU core processing. The __rcu_pending()
function now checks this flag, which causes rcu_check_callbacks()
to initiate RCU core processing at each scheduling-clock interrupt
where this flag is set. Of course this is not sufficient because
scheduling-clock interrupts are often turned off (the things we used to
be able to count on!). So the flags are also checked on entry to any
state that RCU considers to be idle, which includes both NO_HZ_IDLE idle
state and NO_HZ_FULL user-mode-execution state.
This approach should allow call_rcu() to be invoked regardless of what
locks you might be holding, the key word being "should".
Reported-by: Dave Jones <davej@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
It is all too easy to forget that wait_event() does not necessarily
imply a full memory barrier. The case where it does not is where the
condition transitions to true just as wait_event() starts execution.
This is actually a feature: The standard use of wait_event() involves
locking, in which case the locks provide the needed ordering (you hold a
lock across the wake_up() and acquire that same lock after wait_event()
returns).
Given that I did forget that wait_event() does not necessarily imply a
full memory barrier in one case, this commit fixes that case. This commit
also adds comments calling out the placement of existing memory barriers
relied on by wait_event() calls.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
RCU and the fine grained idle time accounting functions check
tick_nohz_enabled. But that variable is merily telling that NOHZ has
been enabled in the config and not been disabled on the command line.
But it does not tell anything about nohz being active. That's what all
this should check for.
Matthew reported, that the idle accounting on his old P1 machine
showed bogus values, when he enabled NOHZ in the config and did not
disable it on the kernel command line. The reason is that his machine
uses (refined) jiffies as a clocksource which explains why the "fine"
grained accounting went into lala land, because it depends on when the
system goes and leaves idle relative to the jiffies increment.
Provide a tick_nohz_active indicator and let RCU and the accounting
code use this instead of tick_nohz_enable.
Reported-and-tested-by: Matthew Whitehead <tedheadster@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Steven Rostedt <rostedt@goodmis.org>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: john.stultz@linaro.org
Cc: mwhitehe@redhat.com
Link: http://lkml.kernel.org/r/alpine.DEB.2.02.1311132052240.30673@ionos.tec.linutronix.de
Paul E. McKenney