The current approach to RCU priority boosting uses an rt_mutex strictly
for its priority-boosting side effects.  The rt_mutex_init_proxy_locked()
function is used by the booster to initialize the lock as held by the
boostee.  The booster then uses rt_mutex_lock() to acquire this rt_mutex,
which priority-boosts the boostee.  When the boostee reaches the end
of its outermost RCU read-side critical section, it checks a field in
its task structure to see whether it has been boosted, and, if so, uses
rt_mutex_unlock() to release the rt_mutex.  The booster can then go on
to boost the next task that is blocking the current RCU grace period.

But reasonable implementations of rt_mutex_unlock() might result in the
boostee referencing the rt_mutex's data after releasing it.  But the
booster might have re-initialized the rt_mutex between the time that the
boostee released it and the time that it later referenced it.  This is
clearly asking for trouble, so this commit introduces a completion that
forces the booster to wait until the boostee has completely finished with
the rt_mutex, thus avoiding the case where the booster is re-initializing
the rt_mutex before the last boostee's last reference to that rt_mutex.

This of course does introduce some overhead, but the priority-boosting
code paths are miles from any possible fastpath, and the overhead of
executing the completion will normally be quite small compared to the
overhead of priority boosting and deboosting, so this should be OK.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>

In kernels built with CONFIG_NO_HZ_FULL, tick_do_timer_cpu is constant
once boot completes.  Thus, there is no need to wrap it in ACCESS_ONCE()
in code that is built only when CONFIG_NO_HZ_FULL.  This commit therefore
removes the redundant ACCESS_ONCE().
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: NFrederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: NLai Jiangshan <laijs@cn.fujitsu.com>

Commit ac1bea85 (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: NDave Hansen <dave.hansen@intel.com>
Signed-off-by: NPaul 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: NJosh 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: NPaul 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: NUma Sharma <uma.sharma523@gmail.com>
[ paulmck: Changed the declaration and several additional uses. ]
Signed-off-by: NPaul 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: NChristoph Lameter <cl@linux.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh 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: NFrederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh 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: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh 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: NAndreea-Cristina Bernat <bernat.ada@gmail.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh 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: NAndreea-Cristina Bernat <bernat.ada@gmail.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh 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: NPaul 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: NJosh 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: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Signed-off-by: NLiu Ping Fan <kernelfans@gmail.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Mostly scripted conversion of the smp_mb__* barriers.
Signed-off-by: NPeter Zijlstra <peterz@infradead.org>
Acked-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/n/tip-55dhyhocezdw1dg7u19hmh1u@git.kernel.org
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: linux-arch@vger.kernel.org
Signed-off-by: NIngo Molnar <mingo@kernel.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: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh 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: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh 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: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh 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: NPaul Bolle <pebolle@tiscali.nl>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh 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: NGreg KH <gregkh@linuxfoundation.org>
Reported-by: NSteven Rostedt <rostedt@goodmis.org>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.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: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NPeter 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.comSigned-off-by: NIngo 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: NMike Galbraith <bitbucket@online.de>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: NMike Galbraith <bitbucket@online.de>
Toasted-by: NFrederic Weisbecker <fweisbec@gmail.com>

After commit #10f39bb1 (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: NLai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: NPaul 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: NDave Jones <davej@redhat.com>
Signed-off-by: NPaul 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: NPaul 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: NMatthew Whitehead <tedheadster@gmail.com>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Reviewed-by: NSteven Rostedt <rostedt@goodmis.org>
Reviewed-by: NPaul 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

Suggested-by: NIngo Molnar <mingo@kernel.org>
Signed-off-by: NPeter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/n/tip-p9ijt8div0hwldexwfm4nlhj@git.kernel.org
[ Fixed build failure in kernel/rcu/tree_plugin.h. ]
Signed-off-by: NIngo Molnar <mingo@kernel.org>

Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: NIngo Molnar <mingo@kernel.org>

If a non-lazy callback arrives on a CPU that has previously gone idle
with no non-lazy callbacks, invoke_rcu_core() forces the RCU core to
run.  However, it does not update the conditions, which could result
in several closely spaced invocations of the RCU core, which in turn
could result in an excessively high context-switch rate and resulting
high overhead.

This commit therefore updates the ->all_lazy and ->nonlazy_posted_snap
fields to prevent closely spaced invocations.
Reported-by: NTibor Billes <tbilles@gmx.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: NTibor Billes <tbilles@gmx.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

The rcu_try_advance_all_cbs() function is invoked on each attempted
entry to and every exit from idle.  If this function determines that
there are callbacks ready to invoke, the caller will invoke the RCU
core, which in turn will result in a pair of context switches.  If a
CPU enters and exits idle extremely frequently, this can result in
an excessive number of context switches and high CPU overhead.

This commit therefore causes rcu_try_advance_all_cbs() to throttle
itself, refusing to do work more than once per jiffy.
Reported-by: NTibor Billes <tbilles@gmx.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: NTibor Billes <tbilles@gmx.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

The rcu_try_advance_all_cbs() function returns a bool saying whether or
not there are callbacks ready to invoke, but rcu_cleanup_after_idle()
rechecks this regardless.  This commit therefore uses the value returned
by rcu_try_advance_all_cbs() instead of making rcu_cleanup_after_idle()
do this recheck.
Reported-by: NTibor Billes <tbilles@gmx.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: NTibor Billes <tbilles@gmx.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

Some architectures have sparse cpu mask. UltraSparc's cpuinfo for example:

CPU0: online
CPU2: online

So, set only possible CPUs when CONFIG_RCU_NOCB_CPU_ALL is enabled.

Also, check that user passes right 'rcu_nocbs=' option.
Signed-off-by: NKirill Tkhai <tkhai@yandex.ru>
CC: Dipankar Sarma <dipankar@in.ibm.com>
[ paulmck: Fix pr_info() issue noted by scripts/checkpatch.pl. ]
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>

This commit adds event traces to track all of rcu_nocb_kthread()'s
blocking and awakening.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>

One way to distinguish between NOCB and non-NOCB rcu_callback trace
events is that the former always print zero for the lazy and non-lazy
queue lengths.  Unfortunately, this also means that we cannot see the NOCB
queue lengths.  This commit therefore accesses the NOCB queue lengths,
but negates them.  NOCB rcu_callback trace events should therefore have
negative queue lengths.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
[ paulmck: Match operand size per kbuild test robot's advice. ]

Lost wakeups from call_rcu() to the rcuo kthreads can result in hangs
that are difficult to diagnose.  This commit therefore adds tracing to
help pin down the cause of these hangs.
Reported-by: NClark Williams <williams@redhat.com>
Reported-by: NCarsten Emde <C.Emde@osadl.org>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
[ paulmck: Add const per kbuild test robot's advice. ]

__get_cpu_var() is used for multiple purposes in the kernel source. One
of them is address calculation via the form &__get_cpu_var(x). This
calculates the address for the instance of the percpu variable of the
current processor based on an offset.

Other use cases are for storing and retrieving data from the current
processors percpu area.  __get_cpu_var() can be used as an lvalue when
writing data or on the right side of an assignment.

__get_cpu_var() is defined as :

__get_cpu_var() always only does an address determination. However,
store and retrieve operations could use a segment prefix (or global
register on other platforms) to avoid the address calculation.

this_cpu_write() and this_cpu_read() can directly take an offset into
a percpu area and use optimized assembly code to read and write per
cpu variables.

This patch converts __get_cpu_var into either an explicit address
calculation using this_cpu_ptr() or into a use of this_cpu operations
that use the offset. Thereby address calcualtions are avoided and less
registers are used when code is generated.

At the end of the patchset all uses of __get_cpu_var have been removed
so the macro is removed too.

The patchset includes passes over all arches as well. Once these
operations are used throughout then specialized macros can be defined in
non -x86 arches as well in order to optimize per cpu access by f.e. using
a global register that may be set to the per cpu base.

Transformations done to __get_cpu_var()

1. Determine the address of the percpu instance of the current processor.

	DEFINE_PER_CPU(int, y);
	int *x = &__get_cpu_var(y);

    Converts to

	int *x = this_cpu_ptr(&y);

2. Same as #1 but this time an array structure is involved.

	DEFINE_PER_CPU(int, y[20]);
	int *x = __get_cpu_var(y);

    Converts to

	int *x = this_cpu_ptr(y);

3. Retrieve the content of the current processors instance of a per cpu
   variable.

	DEFINE_PER_CPU(int, u);
	int x = __get_cpu_var(y)

   Converts to

	int x = __this_cpu_read(y);

4. Retrieve the content of a percpu struct

	DEFINE_PER_CPU(struct mystruct, y);
	struct mystruct x = __get_cpu_var(y);

   Converts to

	memcpy(this_cpu_ptr(&x), y, sizeof(x));

5. Assignment to a per cpu variable

	DEFINE_PER_CPU(int, y)
	__get_cpu_var(y) = x;

   Converts to

	this_cpu_write(y, x);

6. Increment/Decrement etc of a per cpu variable

	DEFINE_PER_CPU(int, y);
	__get_cpu_var(y)++

   Converts to

	this_cpu_inc(y)
Signed-off-by: NChristoph Lameter <cl@linux.com>
[ paulmck: Address conflicts. ]
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>

This commit replaces an incorrect (but fortunately functional)
bitwise OR ("|") operator with the correct logical OR ("||").
Reported-by: Nkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>

Because RCU's quiescent-state-forcing mechanism is used to drive the
full-system-idle state machine, and because this mechanism is executed
by RCU's grace-period kthreads, this commit forces these kthreads to
run on the timekeeping CPU (tick_do_timer_cpu).  To do otherwise would
mean that the RCU grace-period kthreads would force the system into
non-idle state every time they drove the state machine, which would
be just a bit on the futile side.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

This commit adds the state machine that takes the per-CPU idle data
as input and produces a full-system-idle indication as output.  This
state machine is driven out of RCU's quiescent-state-forcing
mechanism, which invokes rcu_sysidle_check_cpu() to collect per-CPU
idle state and then rcu_sysidle_report() to drive the state machine.

The full-system-idle state is sampled using rcu_sys_is_idle(), which
also drives the state machine if RCU is idle (and does so by forcing
RCU to become non-idle).  This function returns true if all but the
timekeeping CPU (tick_do_timer_cpu) are idle and have been idle long
enough to avoid memory contention on the full_sysidle_state state
variable.  The rcu_sysidle_force_exit() may be called externally
to reset the state machine back into non-idle state.

For large systems the state machine is driven out of RCU's
force-quiescent-state logic, which provides good scalability at the price
of millisecond-scale latencies on the transition to full-system-idle
state.  This is not so good for battery-powered systems, which are usually
small enough that they don't need to care about scalability, but which
do care deeply about energy efficiency.  Small systems therefore drive
the state machine directly out of the idle-entry code.  The number of
CPUs in a "small" system is defined by a new NO_HZ_FULL_SYSIDLE_SMALL
Kconfig parameter, which defaults to 8.  Note that this is a build-time
definition.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
[ paulmck: Use true and false for boolean constants per Lai Jiangshan. ]
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>
[ paulmck: Simplify logic and provide better comments for memory barriers,
  based on review comments and questions by Lai Jiangshan. ]

This commit adds control variables and states for full-system idle.
The system will progress through the states in numerical order when
the system is fully idle (other than the timekeeping CPU), and reset
down to the initial state if any non-timekeeping CPU goes non-idle.
The current state is kept in full_sysidle_state.

One flavor of RCU will be in charge of driving the state machine,
defined by rcu_sysidle_state.  This should be the busiest flavor of RCU.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

This commit adds the code that updates the rcu_dyntick structure's
new fields to track the per-CPU idle state based on interrupts and
transitions into and out of the idle loop (NMIs are ignored because NMI
handlers cannot cleanly read out the time anyway).  This code is similar
to the code that maintains RCU's idea of per-CPU idleness, but differs
in that RCU treats CPUs running in user mode as idle, where this new
code does not.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: NFrederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>

This commit adds fields to the rcu_dyntick structure that are used to
detect idle CPUs.  These new fields differ from the existing ones in
that the existing ones consider a CPU executing in user mode to be idle,
where the new ones consider CPUs executing in user mode to be busy.
The handling of these new fields is otherwise quite similar to that for
the exiting fields.  This commit also adds the initialization required
for these fields.

So, why is usermode execution treated differently, with RCU considering
it a quiescent state equivalent to idle, while in contrast the new
full-system idle state detection considers usermode execution to be
non-idle?

It turns out that although one of RCU's quiescent states is usermode
execution, it is not a full-system idle state.  This is because the
purpose of the full-system idle state is not RCU, but rather determining
when accurate timekeeping can safely be disabled.  Whenever accurate
timekeeping is required in a CONFIG_NO_HZ_FULL kernel, at least one
CPU must keep the scheduling-clock tick going.  If even one CPU is
executing in user mode, accurate timekeeping is requires, particularly for
architectures where gettimeofday() and friends do not enter the kernel.
Only when all CPUs are really and truly idle can accurate timekeeping be
disabled, allowing all CPUs to turn off the scheduling clock interrupt,
thus greatly improving energy efficiency.

This naturally raises the question "Why is this code in RCU rather than in
timekeeping?", and the answer is that RCU has the data and infrastructure
to efficiently make this determination.
Signed-off-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: NFrederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>