Commit graph

24 commits

Author SHA1 Message Date
Paul E. McKenney
1aa03f1188 rcu: Simplify unboosting checks
This is a port of commit #82e78d80 from TREE_PREEMPT_RCU to
TINY_PREEMPT_RCU.

This commit uses the fact that current->rcu_boost_mutex is set
any time that the RCU_READ_UNLOCK_BOOSTED flag is set in the
current->rcu_read_unlock_special bitmask.  This allows tests of
the bit to be changed to tests of the pointer, which in turn allows
the RCU_READ_UNLOCK_BOOSTED flag to be eliminated.

Please note that the check of current->rcu_read_unlock_special need not
change because any time that RCU_READ_UNLOCK_BOOSTED was set, so was
RCU_READ_UNLOCK_BLOCKED.  Therefore, __rcu_read_unlock() can continue
testing current->rcu_read_unlock_special for non-zero, as before.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2012-02-21 09:03:43 -08:00
Paul E. McKenney
8762705ad4 rcu: Inform RCU of irq_exit() activity
This is a port to TINY_RCU of Peter Zijlstra's commit #ec433f0c5

The rcu_read_unlock_special() function relies on in_irq() to exclude
scheduler activity from interrupt level.  This fails because exit_irq()
can invoke the scheduler after clearing the preempt_count() bits that
in_irq() uses to determine that it is at interrupt level.  This situation
can result in failures as follows:

     $task			IRQ		SoftIRQ

     rcu_read_lock()

     /* do stuff */

     <preempt> |= UNLOCK_BLOCKED

     rcu_read_unlock()
       --t->rcu_read_lock_nesting

    			irq_enter();
    			/* do stuff, don't use RCU */
    			irq_exit();
    			  sub_preempt_count(IRQ_EXIT_OFFSET);
    			  invoke_softirq()

    					ttwu();
    					  spin_lock_irq(&pi->lock)
    					  rcu_read_lock();
    					  /* do stuff */
    					  rcu_read_unlock();
    					    rcu_read_unlock_special()
    					      rcu_report_exp_rnp()
    					        ttwu()
    					          spin_lock_irq(&pi->lock) /* deadlock */

       rcu_read_unlock_special(t);

This can be triggered 'easily' because invoke_softirq() immediately does
a ttwu() of ksoftirqd/# instead of doing the in-place softirq stuff first,
but even without that the above happens.

Cure this by also excluding softirqs from the rcu_read_unlock_special()
handler and ensuring the force_irqthreads ksoftirqd/# wakeup is done
from full softirq context.

It is also necessary to delay the ->rcu_read_lock_nesting decrement until
after rcu_read_unlock_special().  This delay is handled by the commit
"Protect __rcu_read_unlock() against scheduler-using irq handlers".

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2012-02-21 09:03:42 -08:00
Paul E. McKenney
768dfffdff rcu: Prevent RCU callbacks from executing before scheduler initialized
This is a port of commit #b0d3041 from TREE_RCU to TREE_PREEMPT_RCU.

Under some rare but real combinations of configuration parameters, RCU
callbacks are posted during early boot that use kernel facilities that are
not yet initialized.  Therefore, when these callbacks are invoked, hard
hangs and crashes ensue.  This commit therefore prevents RCU callbacks
from being invoked until after the scheduler is fully up and running,
as in after multiple tasks have been spawned.

It might well turn out that a better approach is to identify the specific
RCU callbacks that are causing this problem, but that discussion will
wait until such time as someone really needs an RCU callback to be invoked
(as opposed to merely registered) during early boot.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2012-02-21 09:03:41 -08:00
Paul E. McKenney
afef20540f rcu: Streamline code produced by __rcu_read_unlock()
This is a port of commit #be0e1e21 to TINY_PREEMPT_RCU.  This uses
noinline to prevent rcu_read_unlock_special() from being inlined into
__rcu_read_unlock().

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2012-02-21 09:03:40 -08:00
Paul E. McKenney
26861faf89 rcu: Protect __rcu_read_unlock() against scheduler-using irq handlers
This commit ports commit #10f39bb1b2 (rcu: protect __rcu_read_unlock()
against scheduler-using irq handlers) from TREE_PREEMPT_RCU to
TINY_PREEMPT_RCU.  The following is a corresponding port of that
commit message.

The addition of RCU read-side critical sections within runqueue and
priority-inheritance critical sections introduced some deadlocks,
for example, involving interrupts from __rcu_read_unlock() where the
interrupt handlers call wake_up().  This situation can cause the
instance of __rcu_read_unlock() invoked from interrupt to do some
of the processing that would otherwise have been carried out by the
task-level instance of __rcu_read_unlock().  When the interrupt-level
instance of __rcu_read_unlock() is called with a scheduler lock held from
interrupt-entry/exit situations where in_irq() returns false, deadlock can
result.  Of course, in a UP kernel, there are not really any deadlocks,
but the upper-level critical section can still be be fatally confused
by the lower-level critical section changing things out from under it.

This commit resolves these deadlocks by using negative values of the
per-task ->rcu_read_lock_nesting counter to indicate that an instance of
__rcu_read_unlock() is in flight, which in turn prevents instances from
interrupt handlers from doing any special processing.  Note that nested
rcu_read_lock()/rcu_read_unlock() pairs are still permitted, but they will
never see ->rcu_read_lock_nesting go to zero, and will therefore never
invoke rcu_read_unlock_special(), thus preventing them from seeing the
RCU_READ_UNLOCK_BLOCKED bit should it be set in ->rcu_read_unlock_special.
This patch also adds a check for ->rcu_read_unlock_special being negative
in rcu_check_callbacks(), thus preventing the RCU_READ_UNLOCK_NEED_QS
bit from being set should a scheduling-clock interrupt occur while
__rcu_read_unlock() is exiting from an outermost RCU read-side critical
section.

Of course, __rcu_read_unlock() can be preempted during the time that
->rcu_read_lock_nesting is negative.  This could result in the setting
of the RCU_READ_UNLOCK_BLOCKED bit after __rcu_read_unlock() checks it,
and would also result it this task being queued on the corresponding
rcu_node structure's blkd_tasks list.  Therefore, some later RCU read-side
critical section would enter rcu_read_unlock_special() to clean up --
which could result in deadlock (OK, OK, fatal confusion) if that RCU
read-side critical section happened to be in the scheduler where the
runqueue or priority-inheritance locks were held.

To prevent the possibility of fatal confusion that might result from
preemption during the time that ->rcu_read_lock_nesting is negative,
this commit also makes rcu_preempt_note_context_switch() check for
negative ->rcu_read_lock_nesting, thus refraining from queuing the task
(and from setting RCU_READ_UNLOCK_BLOCKED) if we are already exiting
from the outermost RCU read-side critical section (in other words,
we really are no longer actually in that RCU read-side critical
section).  In addition, rcu_preempt_note_context_switch() invokes
rcu_read_unlock_special() to carry out the cleanup in this case, which
clears out the ->rcu_read_unlock_special bits and dequeues the task
(if necessary), in turn avoiding needless delay of the current RCU grace
period and needless RCU priority boosting.

It is still illegal to call rcu_read_unlock() while holding a scheduler
lock if the prior RCU read-side critical section has ever had both
preemption and irqs enabled.  However, the common use case is legal,
namely where then entire RCU read-side critical section executes with
irqs disabled, for example, when the scheduler lock is held across the
entire lifetime of the RCU read-side critical section.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2012-02-21 09:03:39 -08:00
Paul E. McKenney
fe15d706cf rcu: Add lockdep-RCU checks for simple self-deadlock
It is illegal to have a grace period within a same-flavor RCU read-side
critical section, so this commit adds lockdep-RCU checks to splat when
such abuse is encountered.  This commit does not detect more elaborate
RCU deadlock situations.  These situations might be a job for lockdep
enhancements.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2012-02-21 09:03:23 -08:00
Paul E. McKenney
4f89b336fd rcu: Apply ACCESS_ONCE() to rcu_boost() return value
Both TINY_RCU's and TREE_RCU's implementations of rcu_boost() access
the ->boost_tasks and ->exp_tasks fields without preventing concurrent
changes to these fields.  This commit therefore applies ACCESS_ONCE in
order to prevent compiler mischief.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2011-12-11 10:33:19 -08:00
Paul E. McKenney
4968c300e1 rcu: Augment rcu_batch_end tracing for idle and callback state
The current rcu_batch_end event trace records only the name of the RCU
flavor and the total number of callbacks that remain queued on the
current CPU.  This is insufficient for testing and tuning the new
dyntick-idle RCU_FAST_NO_HZ code, so this commit adds idle state along
with whether or not any of the callbacks that were ready to invoke
at the beginning of rcu_do_batch() are still queued.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2011-12-11 10:32:22 -08:00
Paul Gortmaker
bdfa97bf72 kernel: fix up module header handling in rcutiny files
The file rcutiny.c does not need moduleparam.h header, as
there are no modparams in this file.

However rcutiny_plugin.h does define a module_init() and
a module_exit() and it uses the various MODULE_ macros, so
it really does need module.h included.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-31 09:20:13 -04:00
Paul E. McKenney
965a002b4f rcu: Make TINY_RCU also use softirq for RCU_BOOST=n
This patch #ifdefs TINY_RCU kthreads out of the kernel unless RCU_BOOST=y,
thus eliminating context-switch overhead if RCU priority boosting has
not been configured.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2011-09-28 21:38:20 -07:00
Paul E. McKenney
e99033c5c1 rcu: Put names into TINY_RCU structures under RCU_TRACE
In order to allow event tracing to distinguish between flavors of
RCU, we need those names in the relevant RCU data structures.  TINY_RCU
has avoided them for memory-footprint reasons, so add them only if
CONFIG_RCU_TRACE=y.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2011-09-28 21:38:14 -07:00
Paul E. McKenney
2c42818e96 rcu: Abstract common code for RCU grace-period-wait primitives
Pull the code that waits for an RCU grace period into a single function,
which is then called by synchronize_rcu() and friends in the case of
TREE_RCU and TREE_PREEMPT_RCU, and from rcu_barrier() and friends in
the case of TINY_RCU and TINY_PREEMPT_RCU.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2011-09-28 21:36:42 -07:00
Paul E. McKenney
7e8b4c7234 rcu: Converge TINY_RCU expedited and normal boosting
This applies a trick from TREE_RCU boosting to TINY_RCU, eliminating
code and adding comments.  The key point is that it is possible for
the booster thread itself to work out whether there is a normal or
expedited boost required based solely on local information.  There
is therefore no need for boost initiation to know or care what type
of boosting is required.  In addition, when boosting is complete for
a given grace period, then by definition there cannot be any more
boosting for that grace period.  This allows eliminating yet more
state and statistics.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
2011-05-05 23:16:58 -07:00
Paul E. McKenney
203373c81b rcu: remove useless ->boosted_this_gp field
The ->boosted_this_gp field is a holdover from an earlier design that
was to carry out multiple boost operations in parallel.  It is not required
by the current design, which boosts one task at a time.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2011-05-05 23:16:58 -07:00
Paul E. McKenney
ddeb75814f rcu: code cleanups in TINY_RCU priority boosting.
Extraneous semicolon, bad comment, and fold INIT_LIST_HEAD() into
list_del() to get list_del_init().

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
2011-05-05 23:16:58 -07:00
Lai Jiangshan
ba74f4d7e5 rcu: call __rcu_read_unlock() in exit_rcu for tiny RCU
Using __rcu_read_lock() in place of rcu_read_lock() leaves any debug
state as it really should be, namely with the lock still held.

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2011-03-04 08:05:08 -08:00
Paul E. McKenney
e940cc804e rcu: Distinguish between boosting and boosted
RCU priority boosting's tracing did not distinguish between ongoing
boosting and completion of boosting.  This commit therefore adds this
capability.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2010-11-29 22:01:56 -08:00
Paul E. McKenney
9e571a82f0 rcu: add tracing for TINY_RCU and TINY_PREEMPT_RCU
Add tracing for the tiny RCU implementations, including statistics on
boosting in the case of TINY_PREEMPT_RCU and RCU_BOOST.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2010-11-29 22:01:55 -08:00
Paul E. McKenney
24278d1483 rcu: priority boosting for TINY_PREEMPT_RCU
Add priority boosting, but only for TINY_PREEMPT_RCU.  This is enabled
by the default-off RCU_BOOST kernel parameter.  The priority to which to
boost preempted RCU readers is controlled by the RCU_BOOST_PRIO kernel
parameter (defaulting to real-time priority 1) and the time to wait
before boosting the readers blocking a given grace period is controlled
by the RCU_BOOST_DELAY kernel parameter (defaulting to 500 milliseconds).

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2010-11-29 22:01:54 -08:00
Paul E. McKenney
b2c0710c46 rcu: move TINY_RCU from softirq to kthread
If RCU priority boosting is to be meaningful, callback invocation must
be boosted in addition to preempted RCU readers.  Otherwise, in presence
of CPU real-time threads, the grace period ends, but the callbacks don't
get invoked.  If the callbacks don't get invoked, the associated memory
doesn't get freed, so the system is still subject to OOM.

But it is not reasonable to priority-boost RCU_SOFTIRQ, so this commit
moves the callback invocations to a kthread, which can be boosted easily.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2010-11-17 15:45:09 -08:00
Paul E. McKenney
dd7c4d8973 rcu: performance fixes to TINY_PREEMPT_RCU callback checking
This commit tightens up checks in rcu_preempt_check_callbacks() to avoid
unnecessary special handling at rcu_read_unlock() time.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2010-08-27 10:51:17 -07:00
Paul E. McKenney
a3dc3fb161 rcu: repair code-duplication FIXMEs
Combine the duplicate definitions of ULONG_CMP_GE(), ULONG_CMP_LT(),
and rcu_preempt_depth() into include/linux/rcupdate.h.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2010-08-20 09:00:13 -07:00
Paul E. McKenney
a57eb940d1 rcu: Add a TINY_PREEMPT_RCU
Implement a small-memory-footprint uniprocessor-only implementation of
preemptible RCU.  This implementation uses but a single blocked-tasks
list rather than the combinatorial number used per leaf rcu_node by
TREE_PREEMPT_RCU, which reduces memory consumption and greatly simplifies
processing.  This version also takes advantage of uniprocessor execution
to accelerate grace periods in the case where there are no readers.

The general design is otherwise broadly similar to that of TREE_PREEMPT_RCU.

This implementation is a step towards having RCU implementation driven
off of the SMP and PREEMPT kernel configuration variables, which can
happen once this implementation has accumulated sufficient experience.

Removed ACCESS_ONCE() from __rcu_read_unlock() and added barrier() as
suggested by Steve Rostedt in order to avoid the compiler-reordering
issue noted by Mathieu Desnoyers (http://lkml.org/lkml/2010/8/16/183).

As can be seen below, CONFIG_TINY_PREEMPT_RCU represents almost 5Kbyte
savings compared to CONFIG_TREE_PREEMPT_RCU.  Of course, for non-real-time
workloads, CONFIG_TINY_RCU is even better.

	CONFIG_TREE_PREEMPT_RCU

	   text	   data	    bss	    dec	   filename
	     13	      0	      0	     13	   kernel/rcupdate.o
	   6170	    825	     28	   7023	   kernel/rcutree.o
				   ----
				   7026    Total

	CONFIG_TINY_PREEMPT_RCU

	   text	   data	    bss	    dec	   filename
	     13	      0	      0	     13	   kernel/rcupdate.o
	   2081	     81	      8	   2170	   kernel/rcutiny.o
				   ----
				   2183    Total

	CONFIG_TINY_RCU (non-preemptible)

	   text	   data	    bss	    dec	   filename
	     13	      0	      0	     13	   kernel/rcupdate.o
	    719	     25	      0	    744	   kernel/rcutiny.o
				    ---
				    757    Total

Requested-by: Loïc Minier <loic.minier@canonical.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2010-08-20 08:55:00 -07:00
Paul E. McKenney
bbad937983 rcu: slim down rcutiny by removing rcu_scheduler_active and friends
TINY_RCU does not need rcu_scheduler_active unless CONFIG_DEBUG_LOCK_ALLOC.
So conditionally compile rcu_scheduler_active in order to slim down
rcutiny a bit more.  Also gets rid of an EXPORT_SYMBOL_GPL, which is
responsible for most of the slimming.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
2010-05-10 11:08:34 -07:00