Merge branch 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull RCU changes from Ingo Molar:
 "The main changes:

   - torture-test updates
   - callback-offloading changes
   - maintainership changes
   - update RCU documentation
   - miscellaneous fixes"

* 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (32 commits)
  rcu: Allow for NULL tick_nohz_full_mask when nohz_full= missing
  rcu: Fix a sparse warning in rcu_report_unblock_qs_rnp()
  rcu: Fix a sparse warning in rcu_initiate_boost()
  rcu: Fix __rcu_reclaim() to use true/false for bool
  rcu: Remove CONFIG_PROVE_RCU_DELAY
  rcu: Use __this_cpu_read() instead of per_cpu_ptr()
  rcu: Don't use NMIs to dump other CPUs' stacks
  rcu: Bind grace-period kthreads to non-NO_HZ_FULL CPUs
  rcu: Simplify priority boosting by putting rt_mutex in rcu_node
  rcu: Check both root and current rcu_node when setting up future grace period
  rcu: Allow post-unlock reference for rt_mutex
  rcu: Loosen __call_rcu()'s rcu_head alignment constraint
  rcu: Eliminate read-modify-write ACCESS_ONCE() calls
  rcu: Remove redundant ACCESS_ONCE() from tick_do_timer_cpu
  rcu: Make rcu node arrays static const char * const
  signal: Explain local_irq_save() call
  rcu: Handle obsolete references to TINY_PREEMPT_RCU
  rcu: Document deadlock-avoidance information for rcu_read_unlock()
  scripts: Teach get_maintainer.pl about the new "R:" tag
  rcu: Update rcu torture maintainership filename patterns
  ...

Documentation/RCU/RTFP.txt

@@ -2451,8 +2451,8 @@ lot of {Linux} into your technology!!!"
 ,month="February"
 ,year="2010"
 ,note="Available:
-\url{http://kerneltrap.com/mailarchive/linux-netdev/2010/2/26/6270589}
-[Viewed March 20, 2011]"
+\url{http://thread.gmane.org/gmane.linux.network/153338}
+[Viewed June 9, 2014]"
 ,annotation={
 	Use a pair of list_head structures to support RCU-protected
 	resizable hash tables.

Documentation/RCU/rcuref.txt

 Reference-count design for elements of lists/arrays protected by RCU.
 
+
+Please note that the percpu-ref feature is likely your first
+stop if you need to combine reference counts and RCU.  Please see
+include/linux/percpu-refcount.h for more information.  However, in
+those unusual cases where percpu-ref would consume too much memory,
+please read on.
+
+------------------------------------------------------------------------
+
 Reference counting on elements of lists which are protected by traditional
 reader/writer spinlocks or semaphores are straightforward:
 

Documentation/kernel-parameters.txt

@@ -2813,6 +2813,13 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			quiescent states.  Units are jiffies, minimum
 			value is one, and maximum value is HZ.
 
+	rcutree.rcu_nocb_leader_stride= [KNL]
+			Set the number of NOCB kthread groups, which
+			defaults to the square root of the number of
+			CPUs.  Larger numbers reduces the wakeup overhead
+			on the per-CPU grace-period kthreads, but increases
+			that same overhead on each group's leader.
+
 	rcutree.qhimark= [KNL]
 			Set threshold of queued RCU callbacks beyond which
 			batch limiting is disabled.

Documentation/memory-barriers.txt

@@ -757,10 +757,14 @@ SMP BARRIER PAIRING
 When dealing with CPU-CPU interactions, certain types of memory barrier should
 always be paired.  A lack of appropriate pairing is almost certainly an error.
 
-A write barrier should always be paired with a data dependency barrier or read
-barrier, though a general barrier would also be viable.  Similarly a read
-barrier or a data dependency barrier should always be paired with at least an
-write barrier, though, again, a general barrier is viable:
+General barriers pair with each other, though they also pair with
+most other types of barriers, albeit without transitivity.  An acquire
+barrier pairs with a release barrier, but both may also pair with other
+barriers, including of course general barriers.  A write barrier pairs
+with a data dependency barrier, an acquire barrier, a release barrier,
+a read barrier, or a general barrier.  Similarly a read barrier or a
+data dependency barrier pairs with a write barrier, an acquire barrier,
+a release barrier, or a general barrier:
 
 	CPU 1		      CPU 2
 	===============	      ===============
@@ -1893,6 +1897,21 @@ between the STORE to indicate the event and the STORE to set TASK_RUNNING:
 	    <general barrier>		  STORE current->state
 	LOAD event_indicated
 
+To repeat, this write memory barrier is present if and only if something
+is actually awakened.  To see this, consider the following sequence of
+events, where X and Y are both initially zero:
+
+	CPU 1				CPU 2
+	===============================	===============================
+	X = 1;				STORE event_indicated
+	smp_mb();			wake_up();
+	Y = 1;				wait_event(wq, Y == 1);
+	wake_up();			  load from Y sees 1, no memory barrier
+					load from X might see 0
+
+In contrast, if a wakeup does occur, CPU 2's load from X would be guaranteed
+to see 1.
+
 The available waker functions include:
 
 	complete();

MAINTAINERS

@@ -70,6 +70,8 @@ Descriptions of section entries:
 
 	P: Person (obsolete)
 	M: Mail patches to: FullName <address@domain>
+	R: Designated reviewer: FullName <address@domain>
+	   These reviewers should be CCed on patches.
 	L: Mailing list that is relevant to this area
 	W: Web-page with status/info
 	Q: Patchwork web based patch tracking system site
@@ -7443,10 +7445,14 @@ L:	linux-kernel@vger.kernel.org
 S:	Supported
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git
 F:	Documentation/RCU/torture.txt
-F:	kernel/rcu/torture.c
+F:	kernel/rcu/rcutorture.c
 
 RCUTORTURE TEST FRAMEWORK
 M:	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
+M:	Josh Triplett <josh@joshtriplett.org>
+R:	Steven Rostedt <rostedt@goodmis.org>
+R:	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
+R:	Lai Jiangshan <laijs@cn.fujitsu.com>
 L:	linux-kernel@vger.kernel.org
 S:	Supported
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git
@@ -7469,8 +7475,11 @@ S:	Supported
 F:	net/rds/
 
 READ-COPY UPDATE (RCU)
-M:	Dipankar Sarma <dipankar@in.ibm.com>
 M:	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
+M:	Josh Triplett <josh@joshtriplett.org>
+R:	Steven Rostedt <rostedt@goodmis.org>
+R:	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
+R:	Lai Jiangshan <laijs@cn.fujitsu.com>
 L:	linux-kernel@vger.kernel.org
 W:	http://www.rdrop.com/users/paulmck/RCU/
 S:	Supported
@@ -7480,7 +7489,7 @@ X:	Documentation/RCU/torture.txt
 F:	include/linux/rcu*
 X:	include/linux/srcu.h
 F:	kernel/rcu/
-X:	kernel/rcu/torture.c
+X:	kernel/torture.c
 
 REAL TIME CLOCK (RTC) SUBSYSTEM
 M:	Alessandro Zummo <a.zummo@towertech.it>
@@ -8263,6 +8272,9 @@ F:	mm/sl?b*
 SLEEPABLE READ-COPY UPDATE (SRCU)
 M:	Lai Jiangshan <laijs@cn.fujitsu.com>
 M:	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
+M:	Josh Triplett <josh@joshtriplett.org>
+R:	Steven Rostedt <rostedt@goodmis.org>
+R:	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
 L:	linux-kernel@vger.kernel.org
 W:	http://www.rdrop.com/users/paulmck/RCU/
 S:	Supported

include/linux/init_task.h

@@ -102,12 +102,6 @@ extern struct group_info init_groups;
 #define INIT_IDS
 #endif
 
-#ifdef CONFIG_RCU_BOOST
-#define INIT_TASK_RCU_BOOST()						\
-	.rcu_boost_mutex = NULL,
-#else
-#define INIT_TASK_RCU_BOOST()
-#endif
 #ifdef CONFIG_TREE_PREEMPT_RCU
 #define INIT_TASK_RCU_TREE_PREEMPT()					\
 	.rcu_blocked_node = NULL,
@@ -119,8 +113,7 @@ extern struct group_info init_groups;
 	.rcu_read_lock_nesting = 0,					\
 	.rcu_read_unlock_special = 0,					\
 	.rcu_node_entry = LIST_HEAD_INIT(tsk.rcu_node_entry),		\
-	INIT_TASK_RCU_TREE_PREEMPT()					\
-	INIT_TASK_RCU_BOOST()
+	INIT_TASK_RCU_TREE_PREEMPT()
 #else
 #define INIT_TASK_RCU_PREEMPT(tsk)
 #endif

include/linux/rcupdate.h

@@ -826,15 +826,14 @@ static inline void rcu_preempt_sleep_check(void)
  * read-side critical section that would block in a !PREEMPT kernel.
  * But if you want the full story, read on!
  *
- * In non-preemptible RCU implementations (TREE_RCU and TINY_RCU), it
- * is illegal to block while in an RCU read-side critical section.  In
- * preemptible RCU implementations (TREE_PREEMPT_RCU and TINY_PREEMPT_RCU)
- * in CONFIG_PREEMPT kernel builds, RCU read-side critical sections may
- * be preempted, but explicit blocking is illegal.  Finally, in preemptible
- * RCU implementations in real-time (with -rt patchset) kernel builds,
- * RCU read-side critical sections may be preempted and they may also
- * block, but only when acquiring spinlocks that are subject to priority
- * inheritance.
+ * In non-preemptible RCU implementations (TREE_RCU and TINY_RCU),
+ * it is illegal to block while in an RCU read-side critical section.
+ * In preemptible RCU implementations (TREE_PREEMPT_RCU) in CONFIG_PREEMPT
+ * kernel builds, RCU read-side critical sections may be preempted,
+ * but explicit blocking is illegal.  Finally, in preemptible RCU
+ * implementations in real-time (with -rt patchset) kernel builds, RCU
+ * read-side critical sections may be preempted and they may also block, but
+ * only when acquiring spinlocks that are subject to priority inheritance.
  */
 static inline void rcu_read_lock(void)
 {
@@ -858,6 +857,34 @@ static inline void rcu_read_lock(void)
 /**
  * rcu_read_unlock() - marks the end of an RCU read-side critical section.
  *
+ * In most situations, rcu_read_unlock() is immune from deadlock.
+ * However, in kernels built with CONFIG_RCU_BOOST, rcu_read_unlock()
+ * is responsible for deboosting, which it does via rt_mutex_unlock().
+ * Unfortunately, this function acquires the scheduler's runqueue and
+ * priority-inheritance spinlocks.  This means that deadlock could result
+ * if the caller of rcu_read_unlock() already holds one of these locks or
+ * any lock that is ever acquired while holding them.
+ *
+ * That said, RCU readers are never priority boosted unless they were
+ * preempted.  Therefore, one way to avoid deadlock is to make sure
+ * that preemption never happens within any RCU read-side critical
+ * section whose outermost rcu_read_unlock() is called with one of
+ * rt_mutex_unlock()'s locks held.  Such preemption can be avoided in
+ * a number of ways, for example, by invoking preempt_disable() before
+ * critical section's outermost rcu_read_lock().
+ *
+ * Given that the set of locks acquired by rt_mutex_unlock() might change
+ * at any time, a somewhat more future-proofed approach is to make sure
+ * that that preemption never happens within any RCU read-side critical
+ * section whose outermost rcu_read_unlock() is called with irqs disabled.
+ * This approach relies on the fact that rt_mutex_unlock() currently only
+ * acquires irq-disabled locks.
+ *
+ * The second of these two approaches is best in most situations,
+ * however, the first approach can also be useful, at least to those
+ * developers willing to keep abreast of the set of locks acquired by
+ * rt_mutex_unlock().
+ *
  * See rcu_read_lock() for more information.
  */
 static inline void rcu_read_unlock(void)

include/linux/sched.h

@@ -1270,9 +1270,6 @@ struct task_struct {
 #ifdef CONFIG_TREE_PREEMPT_RCU
 	struct rcu_node *rcu_blocked_node;
 #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */
-#ifdef CONFIG_RCU_BOOST
-	struct rt_mutex *rcu_boost_mutex;
-#endif /* #ifdef CONFIG_RCU_BOOST */
 
 #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT)
 	struct sched_info sched_info;
@@ -2009,9 +2006,6 @@ static inline void rcu_copy_process(struct task_struct *p)
 #ifdef CONFIG_TREE_PREEMPT_RCU
 	p->rcu_blocked_node = NULL;
 #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */
-#ifdef CONFIG_RCU_BOOST
-	p->rcu_boost_mutex = NULL;
-#endif /* #ifdef CONFIG_RCU_BOOST */
 	INIT_LIST_HEAD(&p->rcu_node_entry);
 }
 

include/linux/tick.h

@@ -12,6 +12,7 @@
 #include <linux/hrtimer.h>
 #include <linux/context_tracking_state.h>
 #include <linux/cpumask.h>
+#include <linux/sched.h>
 
 #ifdef CONFIG_GENERIC_CLOCKEVENTS
 
@@ -162,6 +163,7 @@ static inline u64 get_cpu_iowait_time_us(int cpu, u64 *unused) { return -1; }
 #ifdef CONFIG_NO_HZ_FULL
 extern bool tick_nohz_full_running;
 extern cpumask_var_t tick_nohz_full_mask;
+extern cpumask_var_t housekeeping_mask;
 
 static inline bool tick_nohz_full_enabled(void)
 {
@@ -194,6 +196,24 @@ static inline void tick_nohz_full_kick_all(void) { }
 static inline void __tick_nohz_task_switch(struct task_struct *tsk) { }
 #endif
 
+static inline bool is_housekeeping_cpu(int cpu)
+{
+#ifdef CONFIG_NO_HZ_FULL
+	if (tick_nohz_full_enabled())
+		return cpumask_test_cpu(cpu, housekeeping_mask);
+#endif
+	return true;
+}
+
+static inline void housekeeping_affine(struct task_struct *t)
+{
+#ifdef CONFIG_NO_HZ_FULL
+	if (tick_nohz_full_enabled())
+		set_cpus_allowed_ptr(t, housekeeping_mask);
+
+#endif
+}
+
 static inline void tick_nohz_full_check(void)
 {
 	if (tick_nohz_full_enabled())

init/Kconfig

@@ -505,7 +505,7 @@ config PREEMPT_RCU
 	def_bool TREE_PREEMPT_RCU
 	help
 	  This option enables preemptible-RCU code that is common between
-	  the TREE_PREEMPT_RCU and TINY_PREEMPT_RCU implementations.
+	  TREE_PREEMPT_RCU and, in the old days, TINY_PREEMPT_RCU.
 
 config RCU_STALL_COMMON
 	def_bool ( TREE_RCU || TREE_PREEMPT_RCU || RCU_TRACE )
@@ -737,7 +737,7 @@ choice
 
 config RCU_NOCB_CPU_NONE
 	bool "No build_forced no-CBs CPUs"
-	depends on RCU_NOCB_CPU && !NO_HZ_FULL
+	depends on RCU_NOCB_CPU && !NO_HZ_FULL_ALL
 	help
 	  This option does not force any of the CPUs to be no-CBs CPUs.
 	  Only CPUs designated by the rcu_nocbs= boot parameter will be
@@ -751,7 +751,7 @@ config RCU_NOCB_CPU_NONE
 
 config RCU_NOCB_CPU_ZERO
 	bool "CPU 0 is a build_forced no-CBs CPU"
-	depends on RCU_NOCB_CPU && !NO_HZ_FULL
+	depends on RCU_NOCB_CPU && !NO_HZ_FULL_ALL
 	help
 	  This option forces CPU 0 to be a no-CBs CPU, so that its RCU
 	  callbacks are invoked by a per-CPU kthread whose name begins

kernel/rcu/rcu.h

@@ -99,6 +99,10 @@ static inline void debug_rcu_head_unqueue(struct rcu_head *head)
 
 void kfree(const void *);
 
+/*
+ * Reclaim the specified callback, either by invoking it (non-lazy case)
+ * or freeing it directly (lazy case).  Return true if lazy, false otherwise.
+ */
 static inline bool __rcu_reclaim(const char *rn, struct rcu_head *head)
 {
 	unsigned long offset = (unsigned long)head->func;
@@ -108,12 +112,12 @@ static inline bool __rcu_reclaim(const char *rn, struct rcu_head *head)
 		RCU_TRACE(trace_rcu_invoke_kfree_callback(rn, head, offset));
 		kfree((void *)head - offset);
 		rcu_lock_release(&rcu_callback_map);
-		return 1;
+		return true;
 	} else {
 		RCU_TRACE(trace_rcu_invoke_callback(rn, head));
 		head->func(head);
 		rcu_lock_release(&rcu_callback_map);
-		return 0;
+		return false;
 	}
 }
 

kernel/rcu/srcu.c

@@ -298,9 +298,9 @@ int __srcu_read_lock(struct srcu_struct *sp)
 
 	idx = ACCESS_ONCE(sp->completed) & 0x1;
 	preempt_disable();
-	ACCESS_ONCE(this_cpu_ptr(sp->per_cpu_ref)->c[idx]) += 1;
+	__this_cpu_inc(sp->per_cpu_ref->c[idx]);
 	smp_mb(); /* B */  /* Avoid leaking the critical section. */
-	ACCESS_ONCE(this_cpu_ptr(sp->per_cpu_ref)->seq[idx]) += 1;
+	__this_cpu_inc(sp->per_cpu_ref->seq[idx]);
 	preempt_enable();
 	return idx;
 }

kernel/rcu/tree.c

@@ -1013,10 +1013,7 @@ static void record_gp_stall_check_time(struct rcu_state *rsp)
 }
 
 /*
- * Dump stacks of all tasks running on stalled CPUs.  This is a fallback
- * for architectures that do not implement trigger_all_cpu_backtrace().
- * The NMI-triggered stack traces are more accurate because they are
- * printed by the target CPU.
+ * Dump stacks of all tasks running on stalled CPUs.
  */
 static void rcu_dump_cpu_stacks(struct rcu_state *rsp)
 {
@@ -1094,7 +1091,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp)
 	       (long)rsp->gpnum, (long)rsp->completed, totqlen);
 	if (ndetected == 0)
 		pr_err("INFO: Stall ended before state dump start\n");
-	else if (!trigger_all_cpu_backtrace())
+	else
 		rcu_dump_cpu_stacks(rsp);
 
 	/* Complain about tasks blocking the grace period. */
@@ -1125,8 +1122,7 @@ static void print_cpu_stall(struct rcu_state *rsp)
 	pr_cont(" (t=%lu jiffies g=%ld c=%ld q=%lu)\n",
 		jiffies - rsp->gp_start,
 		(long)rsp->gpnum, (long)rsp->completed, totqlen);
-	if (!trigger_all_cpu_backtrace())
-		dump_stack();
+	rcu_dump_cpu_stacks(rsp);
 
 	raw_spin_lock_irqsave(&rnp->lock, flags);
 	if (ULONG_CMP_GE(jiffies, ACCESS_ONCE(rsp->jiffies_stall)))
@@ -1305,10 +1301,16 @@ rcu_start_future_gp(struct rcu_node *rnp, struct rcu_data *rdp,
 	 * believe that a grace period is in progress, then we must wait
 	 * for the one following, which is in "c".  Because our request
 	 * will be noticed at the end of the current grace period, we don't
-	 * need to explicitly start one.
+	 * need to explicitly start one.  We only do the lockless check
+	 * of rnp_root's fields if the current rcu_node structure thinks
+	 * there is no grace period in flight, and because we hold rnp->lock,
+	 * the only possible change is when rnp_root's two fields are
+	 * equal, in which case rnp_root->gpnum might be concurrently
+	 * incremented.  But that is OK, as it will just result in our
+	 * doing some extra useless work.
 	 */
 	if (rnp->gpnum != rnp->completed ||
-	    ACCESS_ONCE(rnp->gpnum) != ACCESS_ONCE(rnp->completed)) {
+	    ACCESS_ONCE(rnp_root->gpnum) != ACCESS_ONCE(rnp_root->completed)) {
 		rnp->need_future_gp[c & 0x1]++;
 		trace_rcu_future_gp(rnp, rdp, c, TPS("Startedleaf"));
 		goto out;
@@ -1645,11 +1647,6 @@ static int rcu_gp_init(struct rcu_state *rsp)
 					    rnp->level, rnp->grplo,
 					    rnp->grphi, rnp->qsmask);
 		raw_spin_unlock_irq(&rnp->lock);
-#ifdef CONFIG_PROVE_RCU_DELAY
-		if ((prandom_u32() % (rcu_num_nodes + 1)) == 0 &&
-		    system_state == SYSTEM_RUNNING)
-			udelay(200);
-#endif /* #ifdef CONFIG_PROVE_RCU_DELAY */
 		cond_resched();
 	}
 
@@ -2347,7 +2344,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp)
 	}
 	smp_mb(); /* List handling before counting for rcu_barrier(). */
 	rdp->qlen_lazy -= count_lazy;
-	ACCESS_ONCE(rdp->qlen) -= count;
+	ACCESS_ONCE(rdp->qlen) = rdp->qlen - count;
 	rdp->n_cbs_invoked += count;
 
 	/* Reinstate batch limit if we have worked down the excess. */
@@ -2485,14 +2482,14 @@ static void force_quiescent_state(struct rcu_state *rsp)
 	struct rcu_node *rnp_old = NULL;
 
 	/* Funnel through hierarchy to reduce memory contention. */
-	rnp = per_cpu_ptr(rsp->rda, raw_smp_processor_id())->mynode;
+	rnp = __this_cpu_read(rsp->rda->mynode);
 	for (; rnp != NULL; rnp = rnp->parent) {
 		ret = (ACCESS_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) ||
 		      !raw_spin_trylock(&rnp->fqslock);
 		if (rnp_old != NULL)
 			raw_spin_unlock(&rnp_old->fqslock);
 		if (ret) {
-			ACCESS_ONCE(rsp->n_force_qs_lh)++;
+			rsp->n_force_qs_lh++;
 			return;
 		}
 		rnp_old = rnp;
@@ -2504,7 +2501,7 @@ static void force_quiescent_state(struct rcu_state *rsp)
 	smp_mb__after_unlock_lock();
 	raw_spin_unlock(&rnp_old->fqslock);
 	if (ACCESS_ONCE(rsp->gp_flags) & RCU_GP_FLAG_FQS) {
-		ACCESS_ONCE(rsp->n_force_qs_lh)++;
+		rsp->n_force_qs_lh++;
 		raw_spin_unlock_irqrestore(&rnp_old->lock, flags);
 		return;  /* Someone beat us to it. */
 	}
@@ -2662,7 +2659,7 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu),
 	unsigned long flags;
 	struct rcu_data *rdp;
 
-	WARN_ON_ONCE((unsigned long)head & 0x3); /* Misaligned rcu_head! */
+	WARN_ON_ONCE((unsigned long)head & 0x1); /* Misaligned rcu_head! */
 	if (debug_rcu_head_queue(head)) {
 		/* Probable double call_rcu(), so leak the callback. */
 		ACCESS_ONCE(head->func) = rcu_leak_callback;
@@ -2693,7 +2690,7 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu),
 		local_irq_restore(flags);
 		return;
 	}
-	ACCESS_ONCE(rdp->qlen)++;
+	ACCESS_ONCE(rdp->qlen) = rdp->qlen + 1;
 	if (lazy)
 		rdp->qlen_lazy++;
 	else
@@ -3257,7 +3254,7 @@ static void _rcu_barrier(struct rcu_state *rsp)
 	 * ACCESS_ONCE() to prevent the compiler from speculating
 	 * the increment to precede the early-exit check.
 	 */
-	ACCESS_ONCE(rsp->n_barrier_done)++;
+	ACCESS_ONCE(rsp->n_barrier_done) = rsp->n_barrier_done + 1;
 	WARN_ON_ONCE((rsp->n_barrier_done & 0x1) != 1);
 	_rcu_barrier_trace(rsp, "Inc1", -1, rsp->n_barrier_done);
 	smp_mb(); /* Order ->n_barrier_done increment with below mechanism. */
@@ -3307,7 +3304,7 @@ static void _rcu_barrier(struct rcu_state *rsp)
 
 	/* Increment ->n_barrier_done to prevent duplicate work. */
 	smp_mb(); /* Keep increment after above mechanism. */
-	ACCESS_ONCE(rsp->n_barrier_done)++;
+	ACCESS_ONCE(rsp->n_barrier_done) = rsp->n_barrier_done + 1;
 	WARN_ON_ONCE((rsp->n_barrier_done & 0x1) != 0);
 	_rcu_barrier_trace(rsp, "Inc2", -1, rsp->n_barrier_done);
 	smp_mb(); /* Keep increment before caller's subsequent code. */
@@ -3564,11 +3561,13 @@ static void __init rcu_init_levelspread(struct rcu_state *rsp)
 static void __init rcu_init_one(struct rcu_state *rsp,
 		struct rcu_data __percpu *rda)
 {
-	static char *buf[] = { "rcu_node_0",
+	static const char * const buf[] = {
+		"rcu_node_0",
 		"rcu_node_1",
 		"rcu_node_2",
 		"rcu_node_3" };  /* Match MAX_RCU_LVLS */
-	static char *fqs[] = { "rcu_node_fqs_0",
+	static const char * const fqs[] = {
+		"rcu_node_fqs_0",
 		"rcu_node_fqs_1",
 		"rcu_node_fqs_2",
 		"rcu_node_fqs_3" };  /* Match MAX_RCU_LVLS */

kernel/rcu/tree.h

@@ -172,6 +172,14 @@ struct rcu_node {
 				/*  queued on this rcu_node structure that */
 				/*  are blocking the current grace period, */
 				/*  there can be no such task. */
+	struct completion boost_completion;
+				/* Used to ensure that the rt_mutex used */
+				/*  to carry out the boosting is fully */
+				/*  released with no future boostee accesses */
+				/*  before that rt_mutex is re-initialized. */
+	struct rt_mutex boost_mtx;
+				/* Used only for the priority-boosting */
+				/*  side effect, not as a lock. */
 	unsigned long boost_time;
 				/* When to start boosting (jiffies). */
 	struct task_struct *boost_kthread_task;
@@ -334,11 +342,29 @@ struct rcu_data {
 	struct rcu_head **nocb_tail;
 	atomic_long_t nocb_q_count;	/* # CBs waiting for kthread */
 	atomic_long_t nocb_q_count_lazy; /*  (approximate). */
+	struct rcu_head *nocb_follower_head; /* CBs ready to invoke. */
+	struct rcu_head **nocb_follower_tail;
+	atomic_long_t nocb_follower_count; /* # CBs ready to invoke. */
+	atomic_long_t nocb_follower_count_lazy; /*  (approximate). */
 	int nocb_p_count;		/* # CBs being invoked by kthread */
 	int nocb_p_count_lazy;		/*  (approximate). */
 	wait_queue_head_t nocb_wq;	/* For nocb kthreads to sleep on. */
 	struct task_struct *nocb_kthread;
 	bool nocb_defer_wakeup;		/* Defer wakeup of nocb_kthread. */
+
+	/* The following fields are used by the leader, hence own cacheline. */
+	struct rcu_head *nocb_gp_head ____cacheline_internodealigned_in_smp;
+					/* CBs waiting for GP. */
+	struct rcu_head **nocb_gp_tail;
+	long nocb_gp_count;
+	long nocb_gp_count_lazy;
+	bool nocb_leader_wake;		/* Is the nocb leader thread awake? */
+	struct rcu_data *nocb_next_follower;
+					/* Next follower in wakeup chain. */
+
+	/* The following fields are used by the follower, hence new cachline. */
+	struct rcu_data *nocb_leader ____cacheline_internodealigned_in_smp;
+					/* Leader CPU takes GP-end wakeups. */
 #endif /* #ifdef CONFIG_RCU_NOCB_CPU */
 
 	/* 8) RCU CPU stall data. */
@@ -587,8 +613,14 @@ static bool rcu_nohz_full_cpu(struct rcu_state *rsp);
 /* Sum up queue lengths for tracing. */
 static inline void rcu_nocb_q_lengths(struct rcu_data *rdp, long *ql, long *qll)
 {
-	*ql = atomic_long_read(&rdp->nocb_q_count) + rdp->nocb_p_count;
-	*qll = atomic_long_read(&rdp->nocb_q_count_lazy) + rdp->nocb_p_count_lazy;
+	*ql = atomic_long_read(&rdp->nocb_q_count) +
+	      rdp->nocb_p_count +
+	      atomic_long_read(&rdp->nocb_follower_count) +
+	      rdp->nocb_p_count + rdp->nocb_gp_count;
+	*qll = atomic_long_read(&rdp->nocb_q_count_lazy) +
+	       rdp->nocb_p_count_lazy +
+	       atomic_long_read(&rdp->nocb_follower_count_lazy) +
+	       rdp->nocb_p_count_lazy + rdp->nocb_gp_count_lazy;
 }
 #else /* #ifdef CONFIG_RCU_NOCB_CPU */
 static inline void rcu_nocb_q_lengths(struct rcu_data *rdp, long *ql, long *qll)

kernel/rcu/tree_plugin.h

@@ -33,6 +33,7 @@
 #define RCU_KTHREAD_PRIO 1
 
 #ifdef CONFIG_RCU_BOOST
+#include "../locking/rtmutex_common.h"
 #define RCU_BOOST_PRIO CONFIG_RCU_BOOST_PRIO
 #else
 #define RCU_BOOST_PRIO RCU_KTHREAD_PRIO
@@ -336,7 +337,7 @@ void rcu_read_unlock_special(struct task_struct *t)
 	unsigned long flags;
 	struct list_head *np;
 #ifdef CONFIG_RCU_BOOST
-	struct rt_mutex *rbmp = NULL;
+	bool drop_boost_mutex = false;
 #endif /* #ifdef CONFIG_RCU_BOOST */
 	struct rcu_node *rnp;
 	int special;
@@ -398,11 +399,8 @@ void rcu_read_unlock_special(struct task_struct *t)
 #ifdef CONFIG_RCU_BOOST
 		if (&t->rcu_node_entry == rnp->boost_tasks)
 			rnp->boost_tasks = np;
-		/* Snapshot/clear ->rcu_boost_mutex with rcu_node lock held. */
-		if (t->rcu_boost_mutex) {
-			rbmp = t->rcu_boost_mutex;
-			t->rcu_boost_mutex = NULL;
-		}
+		/* Snapshot ->boost_mtx ownership with rcu_node lock held. */
+		drop_boost_mutex = rt_mutex_owner(&rnp->boost_mtx) == t;
 #endif /* #ifdef CONFIG_RCU_BOOST */
 
 		/*
@@ -427,8 +425,10 @@ void rcu_read_unlock_special(struct task_struct *t)
 
 #ifdef CONFIG_RCU_BOOST
 		/* Unboost if we were boosted. */
-		if (rbmp)
-			rt_mutex_unlock(rbmp);
+		if (drop_boost_mutex) {
+			rt_mutex_unlock(&rnp->boost_mtx);
+			complete(&rnp->boost_completion);
+		}
 #endif /* #ifdef CONFIG_RCU_BOOST */
 
 		/*
@@ -988,6 +988,7 @@ static int rcu_preempt_blocked_readers_cgp(struct rcu_node *rnp)
 
 /* Because preemptible RCU does not exist, no quieting of tasks. */
 static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, unsigned long flags)
+	__releases(rnp->lock)
 {
 	raw_spin_unlock_irqrestore(&rnp->lock, flags);
 }
@@ -1149,7 +1150,6 @@ static void rcu_wake_cond(struct task_struct *t, int status)
 static int rcu_boost(struct rcu_node *rnp)
 {
 	unsigned long flags;
-	struct rt_mutex mtx;
 	struct task_struct *t;
 	struct list_head *tb;
 
@@ -1200,11 +1200,15 @@ static int rcu_boost(struct rcu_node *rnp)
 	 * section.
 	 */
 	t = container_of(tb, struct task_struct, rcu_node_entry);
-	rt_mutex_init_proxy_locked(&mtx, t);
-	t->rcu_boost_mutex = &mtx;
+	rt_mutex_init_proxy_locked(&rnp->boost_mtx, t);
+	init_completion(&rnp->boost_completion);
 	raw_spin_unlock_irqrestore(&rnp->lock, flags);
-	rt_mutex_lock(&mtx);  /* Side effect: boosts task t's priority. */
-	rt_mutex_unlock(&mtx);  /* Keep lockdep happy. */
+	/* Lock only for side effect: boosts task t's priority. */
+	rt_mutex_lock(&rnp->boost_mtx);
+	rt_mutex_unlock(&rnp->boost_mtx);  /* Then keep lockdep happy. */
+
+	/* Wait for boostee to be done w/boost_mtx before reinitializing. */
+	wait_for_completion(&rnp->boost_completion);
 
 	return ACCESS_ONCE(rnp->exp_tasks) != NULL ||
 	       ACCESS_ONCE(rnp->boost_tasks) != NULL;
@@ -1256,6 +1260,7 @@ static int rcu_boost_kthread(void *arg)
  * about it going away.
  */
 static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags)
+	__releases(rnp->lock)
 {
 	struct task_struct *t;
 
@@ -1491,6 +1496,7 @@ static void rcu_prepare_kthreads(int cpu)
 #else /* #ifdef CONFIG_RCU_BOOST */
 
 static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags)
+	__releases(rnp->lock)
 {
 	raw_spin_unlock_irqrestore(&rnp->lock, flags);
 }
@@ -2059,6 +2065,22 @@ bool rcu_is_nocb_cpu(int cpu)
 }
 #endif /* #ifndef CONFIG_RCU_NOCB_CPU_ALL */
 
+/*
+ * Kick the leader kthread for this NOCB group.
+ */
+static void wake_nocb_leader(struct rcu_data *rdp, bool force)
+{
+	struct rcu_data *rdp_leader = rdp->nocb_leader;
+
+	if (!ACCESS_ONCE(rdp_leader->nocb_kthread))
+		return;
+	if (!ACCESS_ONCE(rdp_leader->nocb_leader_wake) || force) {
+		/* Prior xchg orders against prior callback enqueue. */
+		ACCESS_ONCE(rdp_leader->nocb_leader_wake) = true;
+		wake_up(&rdp_leader->nocb_wq);
+	}
+}
+
 /*
  * Enqueue the specified string of rcu_head structures onto the specified
  * CPU's no-CBs lists.  The CPU is specified by rdp, the head of the
@@ -2093,7 +2115,8 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp,
 	len = atomic_long_read(&rdp->nocb_q_count);
 	if (old_rhpp == &rdp->nocb_head) {
 		if (!irqs_disabled_flags(flags)) {
-			wake_up(&rdp->nocb_wq); /* ... if queue was empty ... */
+			/* ... if queue was empty ... */
+			wake_nocb_leader(rdp, false);
 			trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
 					    TPS("WakeEmpty"));
 		} else {
@@ -2103,7 +2126,8 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp,
 		}
 		rdp->qlen_last_fqs_check = 0;
 	} else if (len > rdp->qlen_last_fqs_check + qhimark) {
-		wake_up_process(t); /* ... or if many callbacks queued. */
+		/* ... or if many callbacks queued. */
+		wake_nocb_leader(rdp, true);
 		rdp->qlen_last_fqs_check = LONG_MAX / 2;
 		trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeOvf"));
 	} else {
@@ -2213,55 +2237,173 @@ static void rcu_nocb_wait_gp(struct rcu_data *rdp)
 }
 
 /*
- * Per-rcu_data kthread, but only for no-CBs CPUs.  Each kthread invokes
- * callbacks queued by the corresponding no-CBs CPU.
+ * Leaders come here to wait for additional callbacks to show up.
+ * This function does not return until callbacks appear.
  */
-static int rcu_nocb_kthread(void *arg)
+static void nocb_leader_wait(struct rcu_data *my_rdp)
 {
-	int c, cl;
-	bool firsttime = 1;
-	struct rcu_head *list;
-	struct rcu_head *next;
+	bool firsttime = true;
+	bool gotcbs;
+	struct rcu_data *rdp;
 	struct rcu_head **tail;
-	struct rcu_data *rdp = arg;
 
-	/* Each pass through this loop invokes one batch of callbacks */
+wait_again:
+
+	/* Wait for callbacks to appear. */
+	if (!rcu_nocb_poll) {
+		trace_rcu_nocb_wake(my_rdp->rsp->name, my_rdp->cpu, "Sleep");
+		wait_event_interruptible(my_rdp->nocb_wq,
+					 ACCESS_ONCE(my_rdp->nocb_leader_wake));
+		/* Memory barrier handled by smp_mb() calls below and repoll. */
+	} else if (firsttime) {
+		firsttime = false; /* Don't drown trace log with "Poll"! */
+		trace_rcu_nocb_wake(my_rdp->rsp->name, my_rdp->cpu, "Poll");
+	}
+
+	/*
+	 * Each pass through the following loop checks a follower for CBs.
+	 * We are our own first follower.  Any CBs found are moved to
+	 * nocb_gp_head, where they await a grace period.
+	 */
+	gotcbs = false;
+	for (rdp = my_rdp; rdp; rdp = rdp->nocb_next_follower) {
+		rdp->nocb_gp_head = ACCESS_ONCE(rdp->nocb_head);
+		if (!rdp->nocb_gp_head)
+			continue;  /* No CBs here, try next follower. */
+
+		/* Move callbacks to wait-for-GP list, which is empty. */
+		ACCESS_ONCE(rdp->nocb_head) = NULL;
+		rdp->nocb_gp_tail = xchg(&rdp->nocb_tail, &rdp->nocb_head);
+		rdp->nocb_gp_count = atomic_long_xchg(&rdp->nocb_q_count, 0);
+		rdp->nocb_gp_count_lazy =
+			atomic_long_xchg(&rdp->nocb_q_count_lazy, 0);
+		gotcbs = true;
+	}
+
+	/*
+	 * If there were no callbacks, sleep a bit, rescan after a
+	 * memory barrier, and go retry.
+	 */
+	if (unlikely(!gotcbs)) {
+		if (!rcu_nocb_poll)
+			trace_rcu_nocb_wake(my_rdp->rsp->name, my_rdp->cpu,
+					    "WokeEmpty");
+		flush_signals(current);
+		schedule_timeout_interruptible(1);
+
+		/* Rescan in case we were a victim of memory ordering. */
+		my_rdp->nocb_leader_wake = false;
+		smp_mb();  /* Ensure _wake false before scan. */
+		for (rdp = my_rdp; rdp; rdp = rdp->nocb_next_follower)
+			if (ACCESS_ONCE(rdp->nocb_head)) {
+				/* Found CB, so short-circuit next wait. */
+				my_rdp->nocb_leader_wake = true;
+				break;
+			}
+		goto wait_again;
+	}
+
+	/* Wait for one grace period. */
+	rcu_nocb_wait_gp(my_rdp);
+
+	/*
+	 * We left ->nocb_leader_wake set to reduce cache thrashing.
+	 * We clear it now, but recheck for new callbacks while
+	 * traversing our follower list.
+	 */
+	my_rdp->nocb_leader_wake = false;
+	smp_mb(); /* Ensure _wake false before scan of ->nocb_head. */
+
+	/* Each pass through the following loop wakes a follower, if needed. */
+	for (rdp = my_rdp; rdp; rdp = rdp->nocb_next_follower) {
+		if (ACCESS_ONCE(rdp->nocb_head))
+			my_rdp->nocb_leader_wake = true; /* No need to wait. */
+		if (!rdp->nocb_gp_head)
+			continue; /* No CBs, so no need to wake follower. */
+
+		/* Append callbacks to follower's "done" list. */
+		tail = xchg(&rdp->nocb_follower_tail, rdp->nocb_gp_tail);
+		*tail = rdp->nocb_gp_head;
+		atomic_long_add(rdp->nocb_gp_count, &rdp->nocb_follower_count);
+		atomic_long_add(rdp->nocb_gp_count_lazy,
+				&rdp->nocb_follower_count_lazy);
+		if (rdp != my_rdp && tail == &rdp->nocb_follower_head) {
+			/*
+			 * List was empty, wake up the follower.
+			 * Memory barriers supplied by atomic_long_add().
+			 */
+			wake_up(&rdp->nocb_wq);
+		}
+	}
+
+	/* If we (the leader) don't have CBs, go wait some more. */
+	if (!my_rdp->nocb_follower_head)
+		goto wait_again;
+}
+
+/*
+ * Followers come here to wait for additional callbacks to show up.
+ * This function does not return until callbacks appear.
+ */
+static void nocb_follower_wait(struct rcu_data *rdp)
+{
+	bool firsttime = true;
+
 	for (;;) {
-		/* If not polling, wait for next batch of callbacks. */
 		if (!rcu_nocb_poll) {
 			trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
-					    TPS("Sleep"));
-			wait_event_interruptible(rdp->nocb_wq, rdp->nocb_head);
-			/* Memory barrier provide by xchg() below. */
+					    "FollowerSleep");
+			wait_event_interruptible(rdp->nocb_wq,
+						 ACCESS_ONCE(rdp->nocb_follower_head));
 		} else if (firsttime) {
-			firsttime = 0;
-			trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
-					    TPS("Poll"));
+			/* Don't drown trace log with "Poll"! */
+			firsttime = false;
+			trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, "Poll");
+		}
+		if (smp_load_acquire(&rdp->nocb_follower_head)) {
+			/* ^^^ Ensure CB invocation follows _head test. */
+			return;
 		}
-		list = ACCESS_ONCE(rdp->nocb_head);
-		if (!list) {
 		if (!rcu_nocb_poll)
 			trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
-						    TPS("WokeEmpty"));
-			schedule_timeout_interruptible(1);
+					    "WokeEmpty");
 		flush_signals(current);
-			continue;
+		schedule_timeout_interruptible(1);
 	}
-		firsttime = 1;
-		trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
-				    TPS("WokeNonEmpty"));
+}
 
-		/*
-		 * Extract queued callbacks, update counts, and wait
-		 * for a grace period to elapse.
+/*
+ * Per-rcu_data kthread, but only for no-CBs CPUs.  Each kthread invokes
+ * callbacks queued by the corresponding no-CBs CPU, however, there is
+ * an optional leader-follower relationship so that the grace-period
+ * kthreads don't have to do quite so many wakeups.
  */
-		ACCESS_ONCE(rdp->nocb_head) = NULL;
-		tail = xchg(&rdp->nocb_tail, &rdp->nocb_head);
-		c = atomic_long_xchg(&rdp->nocb_q_count, 0);
-		cl = atomic_long_xchg(&rdp->nocb_q_count_lazy, 0);
-		ACCESS_ONCE(rdp->nocb_p_count) += c;
-		ACCESS_ONCE(rdp->nocb_p_count_lazy) += cl;
-		rcu_nocb_wait_gp(rdp);
+static int rcu_nocb_kthread(void *arg)
+{
+	int c, cl;
+	struct rcu_head *list;
+	struct rcu_head *next;
+	struct rcu_head **tail;
+	struct rcu_data *rdp = arg;
+
+	/* Each pass through this loop invokes one batch of callbacks */
+	for (;;) {
+		/* Wait for callbacks. */
+		if (rdp->nocb_leader == rdp)
+			nocb_leader_wait(rdp);
+		else
+			nocb_follower_wait(rdp);
+
+		/* Pull the ready-to-invoke callbacks onto local list. */
+		list = ACCESS_ONCE(rdp->nocb_follower_head);
+		BUG_ON(!list);
+		trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, "WokeNonEmpty");
+		ACCESS_ONCE(rdp->nocb_follower_head) = NULL;
+		tail = xchg(&rdp->nocb_follower_tail, &rdp->nocb_follower_head);
+		c = atomic_long_xchg(&rdp->nocb_follower_count, 0);
+		cl = atomic_long_xchg(&rdp->nocb_follower_count_lazy, 0);
+		rdp->nocb_p_count += c;
+		rdp->nocb_p_count_lazy += cl;
 
 		/* Each pass through the following loop invokes a callback. */
 		trace_rcu_batch_start(rdp->rsp->name, cl, c, -1);
@@ -2305,7 +2447,7 @@ static void do_nocb_deferred_wakeup(struct rcu_data *rdp)
 	if (!rcu_nocb_need_deferred_wakeup(rdp))
 		return;
 	ACCESS_ONCE(rdp->nocb_defer_wakeup) = false;
-	wake_up(&rdp->nocb_wq);
+	wake_nocb_leader(rdp, false);
 	trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("DeferredWakeEmpty"));
 }
 
@@ -2314,19 +2456,57 @@ static void __init rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp)
 {
 	rdp->nocb_tail = &rdp->nocb_head;
 	init_waitqueue_head(&rdp->nocb_wq);
+	rdp->nocb_follower_tail = &rdp->nocb_follower_head;
 }
 
-/* Create a kthread for each RCU flavor for each no-CBs CPU. */
+/* How many follower CPU IDs per leader?  Default of -1 for sqrt(nr_cpu_ids). */
+static int rcu_nocb_leader_stride = -1;
+module_param(rcu_nocb_leader_stride, int, 0444);
+
+/*
+ * Create a kthread for each RCU flavor for each no-CBs CPU.
+ * Also initialize leader-follower relationships.
+ */
 static void __init rcu_spawn_nocb_kthreads(struct rcu_state *rsp)
 {
 	int cpu;
+	int ls = rcu_nocb_leader_stride;
+	int nl = 0;  /* Next leader. */
 	struct rcu_data *rdp;
+	struct rcu_data *rdp_leader = NULL;  /* Suppress misguided gcc warn. */
+	struct rcu_data *rdp_prev = NULL;
 	struct task_struct *t;
 
 	if (rcu_nocb_mask == NULL)
 		return;
+#if defined(CONFIG_NO_HZ_FULL) && !defined(CONFIG_NO_HZ_FULL_ALL)
+	if (tick_nohz_full_running)
+		cpumask_or(rcu_nocb_mask, rcu_nocb_mask, tick_nohz_full_mask);
+#endif /* #if defined(CONFIG_NO_HZ_FULL) && !defined(CONFIG_NO_HZ_FULL_ALL) */
+	if (ls == -1) {
+		ls = int_sqrt(nr_cpu_ids);
+		rcu_nocb_leader_stride = ls;
+	}
+
+	/*
+	 * Each pass through this loop sets up one rcu_data structure and
+	 * spawns one rcu_nocb_kthread().
+	 */
 	for_each_cpu(cpu, rcu_nocb_mask) {
 		rdp = per_cpu_ptr(rsp->rda, cpu);
+		if (rdp->cpu >= nl) {
+			/* New leader, set up for followers & next leader. */
+			nl = DIV_ROUND_UP(rdp->cpu + 1, ls) * ls;
+			rdp->nocb_leader = rdp;
+			rdp_leader = rdp;
+		} else {
+			/* Another follower, link to previous leader. */
+			rdp->nocb_leader = rdp_leader;
+			rdp_prev->nocb_next_follower = rdp;
+		}
+		rdp_prev = rdp;
+
+		/* Spawn the kthread for this CPU. */
 		t = kthread_run(rcu_nocb_kthread, rdp,
 				"rcuo%c/%d", rsp->abbr, cpu);
 		BUG_ON(IS_ERR(t));
@@ -2843,12 +3023,16 @@ static bool rcu_nohz_full_cpu(struct rcu_state *rsp)
  */
 static void rcu_bind_gp_kthread(void)
 {
-#ifdef CONFIG_NO_HZ_FULL
-	int cpu = ACCESS_ONCE(tick_do_timer_cpu);
+	int __maybe_unused cpu;
 
-	if (cpu < 0 || cpu >= nr_cpu_ids)
+	if (!tick_nohz_full_enabled())
 		return;
-	if (raw_smp_processor_id() != cpu)
+#ifdef CONFIG_NO_HZ_FULL_SYSIDLE
+	cpu = tick_do_timer_cpu;
+	if (cpu >= 0 && cpu < nr_cpu_ids && raw_smp_processor_id() != cpu)
 		set_cpus_allowed_ptr(current, cpumask_of(cpu));
-#endif /* #ifdef CONFIG_NO_HZ_FULL */
+#else /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */
+	if (!is_housekeeping_cpu(raw_smp_processor_id()))
+		housekeeping_affine(current);
+#endif /* #else #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */
 }

kernel/rcu/update.c

@@ -90,9 +90,6 @@ void __rcu_read_unlock(void)
 	} else {
 		barrier();  /* critical section before exit code. */
 		t->rcu_read_lock_nesting = INT_MIN;
-#ifdef CONFIG_PROVE_RCU_DELAY
-		udelay(10); /* Make preemption more probable. */
-#endif /* #ifdef CONFIG_PROVE_RCU_DELAY */
 		barrier();  /* assign before ->rcu_read_unlock_special load */
 		if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special)))
 			rcu_read_unlock_special(t);

kernel/signal.c

@@ -1263,6 +1263,10 @@ struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
 	struct sighand_struct *sighand;
 
 	for (;;) {
+		/*
+		 * Disable interrupts early to avoid deadlocks.
+		 * See rcu_read_unlock() comment header for details.
+		 */
 		local_irq_save(*flags);
 		rcu_read_lock();
 		sighand = rcu_dereference(tsk->sighand);

kernel/time/tick-sched.c

@@ -154,6 +154,7 @@ static void tick_sched_handle(struct tick_sched *ts, struct pt_regs *regs)
 
 #ifdef CONFIG_NO_HZ_FULL
 cpumask_var_t tick_nohz_full_mask;
+cpumask_var_t housekeeping_mask;
 bool tick_nohz_full_running;
 
 static bool can_stop_full_tick(void)
@@ -281,6 +282,7 @@ static int __init tick_nohz_full_setup(char *str)
 	int cpu;
 
 	alloc_bootmem_cpumask_var(&tick_nohz_full_mask);
+	alloc_bootmem_cpumask_var(&housekeeping_mask);
 	if (cpulist_parse(str, tick_nohz_full_mask) < 0) {
 		pr_warning("NOHZ: Incorrect nohz_full cpumask\n");
 		return 1;
@@ -291,6 +293,8 @@ static int __init tick_nohz_full_setup(char *str)
 		pr_warning("NO_HZ: Clearing %d from nohz_full range for timekeeping\n", cpu);
 		cpumask_clear_cpu(cpu, tick_nohz_full_mask);
 	}
+	cpumask_andnot(housekeeping_mask,
+		       cpu_possible_mask, tick_nohz_full_mask);
 	tick_nohz_full_running = true;
 
 	return 1;
@@ -332,9 +336,15 @@ static int tick_nohz_init_all(void)
 		pr_err("NO_HZ: Can't allocate full dynticks cpumask\n");
 		return err;
 	}
+	if (!alloc_cpumask_var(&housekeeping_mask, GFP_KERNEL)) {
+		pr_err("NO_HZ: Can't allocate not-full dynticks cpumask\n");
+		return err;
+	}
 	err = 0;
 	cpumask_setall(tick_nohz_full_mask);
 	cpumask_clear_cpu(smp_processor_id(), tick_nohz_full_mask);
+	cpumask_clear(housekeeping_mask);
+	cpumask_set_cpu(smp_processor_id(), housekeeping_mask);
 	tick_nohz_full_running = true;
 #endif
 	return err;

kernel/torture.c

@@ -708,7 +708,7 @@ int _torture_create_kthread(int (*fn)(void *arg), void *arg, char *s, char *m,
 	int ret = 0;
 
 	VERBOSE_TOROUT_STRING(m);
-	*tp = kthread_run(fn, arg, s);
+	*tp = kthread_run(fn, arg, "%s", s);
 	if (IS_ERR(*tp)) {
 		ret = PTR_ERR(*tp);
 		VERBOSE_TOROUT_ERRSTRING(f);

lib/Kconfig.debug

@@ -1131,20 +1131,6 @@ config PROVE_RCU_REPEATEDLY
 
 	 Say N if you are unsure.
 
-config PROVE_RCU_DELAY
-	bool "RCU debugging: preemptible RCU race provocation"
-	depends on DEBUG_KERNEL && PREEMPT_RCU
-	default n
-	help
-	 There is a class of races that involve an unlikely preemption
-	 of __rcu_read_unlock() just after ->rcu_read_lock_nesting has
-	 been set to INT_MIN.  This feature inserts a delay at that
-	 point to increase the probability of these races.
-
-	 Say Y to increase probability of preemption of __rcu_read_unlock().
-
-	 Say N if you are unsure.
-
 config SPARSE_RCU_POINTER
 	bool "RCU debugging: sparse-based checks for pointer usage"
 	default n

scripts/get_maintainer.pl

@@ -21,6 +21,7 @@ my $lk_path = "./";
 my $email = 1;
 my $email_usename = 1;
 my $email_maintainer = 1;
+my $email_reviewer = 1;
 my $email_list = 1;
 my $email_subscriber_list = 0;
 my $email_git_penguin_chiefs = 0;
@@ -202,6 +203,7 @@ if (!GetOptions(
 		'remove-duplicates!' => \$email_remove_duplicates,
 		'mailmap!' => \$email_use_mailmap,
 		'm!' => \$email_maintainer,
+		'r!' => \$email_reviewer,
 		'n!' => \$email_usename,
 		'l!' => \$email_list,
 		's!' => \$email_subscriber_list,
@@ -260,7 +262,8 @@ if ($sections) {
 }
 
 if ($email &&
-    ($email_maintainer + $email_list + $email_subscriber_list +
+    ($email_maintainer + $email_reviewer +
+     $email_list + $email_subscriber_list +
      $email_git + $email_git_penguin_chiefs + $email_git_blame) == 0) {
     die "$P: Please select at least 1 email option\n";
 }
@@ -750,6 +753,7 @@ MAINTAINER field selection options:
     --hg-since => hg history to use (default: $email_hg_since)
     --interactive => display a menu (mostly useful if used with the --git option)
     --m => include maintainer(s) if any
+    --r => include reviewer(s) if any
     --n => include name 'Full Name <addr\@domain.tld>'
     --l => include list(s) if any
     --s => include subscriber only list(s) if any
@@ -1064,6 +1068,22 @@ sub add_categories {
 		    my $role = get_maintainer_role($i);
 		    push_email_addresses($pvalue, $role);
 		}
+	    } elsif ($ptype eq "R") {
+		my ($name, $address) = parse_email($pvalue);
+		if ($name eq "") {
+		    if ($i > 0) {
+			my $tv = $typevalue[$i - 1];
+			if ($tv =~ m/^(\C):\s*(.*)/) {
+			    if ($1 eq "P") {
+				$name = $2;
+				$pvalue = format_email($name, $address, $email_usename);
+			    }
+			}
+		    }
+		}
+		if ($email_reviewer) {
+		    push_email_addresses($pvalue, 'reviewer');
+		}
 	    } elsif ($ptype eq "T") {
 		push(@scm, $pvalue);
 	    } elsif ($ptype eq "W") {

tools/testing/selftests/rcutorture/bin/kvm-recheck.sh

@@ -54,10 +54,16 @@ do
 			if test -f "$i/qemu-cmd"
 			then
 				print_bug qemu failed
+				echo "   $i"
+			elif test -f "$i/buildonly"
+			then
+				echo Build-only run, no boot/test
+				configcheck.sh $i/.config $i/ConfigFragment
+				parse-build.sh $i/Make.out $configfile
 			else
 				print_bug Build failed
-			fi
 				echo "   $i"
 			fi
+		fi
 	done
 done

tools/testing/selftests/rcutorture/bin/kvm-test-1-run.sh

@@ -42,6 +42,7 @@ grace=120
 
 T=/tmp/kvm-test-1-run.sh.$$
 trap 'rm -rf $T' 0
+touch $T
 
 . $KVM/bin/functions.sh
 . $KVPATH/ver_functions.sh
@@ -131,7 +132,10 @@ boot_args=$6
 
 cd $KVM
 kstarttime=`awk 'BEGIN { print systime() }' < /dev/null`
-echo ' ---' `date`: Starting kernel
+if test -z "$TORTURE_BUILDONLY"
+then
+	echo ' ---' `date`: Starting kernel
+fi
 
 # Generate -smp qemu argument.
 qemu_args="-nographic $qemu_args"
@@ -157,12 +161,13 @@ boot_args="`configfrag_boot_params "$boot_args" "$config_template"`"
 # Generate kernel-version-specific boot parameters
 boot_args="`per_version_boot_params "$boot_args" $builddir/.config $seconds`"
 
-echo $QEMU $qemu_args -m 512 -kernel $builddir/$BOOT_IMAGE -append \"$qemu_append $boot_args\" > $resdir/qemu-cmd
 if test -n "$TORTURE_BUILDONLY"
 then
 	echo Build-only run specified, boot/test omitted.
+	touch $resdir/buildonly
 	exit 0
 fi
+echo $QEMU $qemu_args -m 512 -kernel $builddir/$BOOT_IMAGE -append \"$qemu_append $boot_args\" > $resdir/qemu-cmd
 ( $QEMU $qemu_args -m 512 -kernel $builddir/$BOOT_IMAGE -append "$qemu_append $boot_args"; echo $? > $resdir/qemu-retval ) &
 qemu_pid=$!
 commandcompleted=0

tools/testing/selftests/rcutorture/bin/kvm.sh

@@ -340,12 +340,18 @@ function dump(first, pastlast)
 	for (j = 1; j < jn; j++) {
 		builddir=KVM "/b" j
 		print "rm -f " builddir ".ready"
-		print "echo ----", cfr[j], cpusr[j] ovf ": Starting kernel. `date`";
-		print "echo ----", cfr[j], cpusr[j] ovf ": Starting kernel. `date` >> " rd "/log";
+		print "if test -z \"$TORTURE_BUILDONLY\""
+		print "then"
+		print "\techo ----", cfr[j], cpusr[j] ovf ": Starting kernel. `date`";
+		print "\techo ----", cfr[j], cpusr[j] ovf ": Starting kernel. `date` >> " rd "/log";
+		print "fi"
 	}
 	print "wait"
-	print "echo ---- All kernel runs complete. `date`";
-	print "echo ---- All kernel runs complete. `date` >> " rd "/log";
+	print "if test -z \"$TORTURE_BUILDONLY\""
+	print "then"
+	print "\techo ---- All kernel runs complete. `date`";
+	print "\techo ---- All kernel runs complete. `date` >> " rd "/log";
+	print "fi"
 	for (j = 1; j < jn; j++) {
 		builddir=KVM "/b" j
 		print "echo ----", cfr[j], cpusr[j] ovf ": Build/run results:";
@@ -385,10 +391,7 @@ echo
 echo
 echo " --- `date` Test summary:"
 echo Results directory: $resdir/$ds
-if test -z "$TORTURE_BUILDONLY"
-then
-	kvm-recheck.sh $resdir/$ds
-fi
+kvm-recheck.sh $resdir/$ds
 ___EOF___
 
 if test "$dryrun" = script
@@ -403,7 +406,7 @@ then
 		sed -e 's/:.*$//' -e 's/^echo //'
 	exit 0
 else
-	# Not a dryru, so run the script.
+	# Not a dryrun, so run the script.
 	sh $T/script
 fi
 

tools/testing/selftests/rcutorture/configs/rcu/TREE01

@@ -15,7 +15,6 @@ CONFIG_RCU_FANOUT_EXACT=n
 CONFIG_RCU_NOCB_CPU=y
 CONFIG_RCU_NOCB_CPU_ZERO=y
 CONFIG_DEBUG_LOCK_ALLOC=n
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=n
 CONFIG_RCU_CPU_STALL_VERBOSE=n
 CONFIG_RCU_BOOST=n

tools/testing/selftests/rcutorture/configs/rcu/TREE02

@@ -18,7 +18,6 @@ CONFIG_RCU_FANOUT_EXACT=n
 CONFIG_RCU_NOCB_CPU=n
 CONFIG_DEBUG_LOCK_ALLOC=y
 CONFIG_PROVE_LOCKING=n
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=n
 CONFIG_RCU_CPU_STALL_VERBOSE=y
 CONFIG_RCU_BOOST=n

tools/testing/selftests/rcutorture/configs/rcu/TREE02-T

@@ -18,7 +18,6 @@ CONFIG_RCU_FANOUT_EXACT=n
 CONFIG_RCU_NOCB_CPU=n
 CONFIG_DEBUG_LOCK_ALLOC=y
 CONFIG_PROVE_LOCKING=n
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=n
 CONFIG_RCU_CPU_STALL_VERBOSE=y
 CONFIG_RCU_BOOST=n

tools/testing/selftests/rcutorture/configs/rcu/TREE03

@@ -14,7 +14,6 @@ CONFIG_RCU_FANOUT_LEAF=4
 CONFIG_RCU_FANOUT_EXACT=n
 CONFIG_RCU_NOCB_CPU=n
 CONFIG_DEBUG_LOCK_ALLOC=n
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=n
 CONFIG_RCU_CPU_STALL_VERBOSE=n
 CONFIG_RCU_BOOST=y

tools/testing/selftests/rcutorture/configs/rcu/TREE04

@@ -18,7 +18,6 @@ CONFIG_RCU_FANOUT_LEAF=2
 CONFIG_RCU_FANOUT_EXACT=n
 CONFIG_RCU_NOCB_CPU=n
 CONFIG_DEBUG_LOCK_ALLOC=n
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=y
 CONFIG_RCU_CPU_STALL_VERBOSE=y
 CONFIG_DEBUG_OBJECTS_RCU_HEAD=n

tools/testing/selftests/rcutorture/configs/rcu/TREE05

@@ -18,7 +18,6 @@ CONFIG_RCU_NOCB_CPU_NONE=y
 CONFIG_DEBUG_LOCK_ALLOC=y
 CONFIG_PROVE_LOCKING=y
 CONFIG_PROVE_RCU=y
-CONFIG_PROVE_RCU_DELAY=y
 CONFIG_RCU_CPU_STALL_INFO=n
 CONFIG_RCU_CPU_STALL_VERBOSE=n
 CONFIG_DEBUG_OBJECTS_RCU_HEAD=n

tools/testing/selftests/rcutorture/configs/rcu/TREE06

@@ -19,7 +19,6 @@ CONFIG_RCU_NOCB_CPU=n
 CONFIG_DEBUG_LOCK_ALLOC=y
 CONFIG_PROVE_LOCKING=y
 CONFIG_PROVE_RCU=y
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=n
 CONFIG_RCU_CPU_STALL_VERBOSE=n
 CONFIG_DEBUG_OBJECTS_RCU_HEAD=y

tools/testing/selftests/rcutorture/configs/rcu/TREE07

@@ -17,7 +17,6 @@ CONFIG_RCU_FANOUT_LEAF=2
 CONFIG_RCU_FANOUT_EXACT=n
 CONFIG_RCU_NOCB_CPU=n
 CONFIG_DEBUG_LOCK_ALLOC=n
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=y
 CONFIG_RCU_CPU_STALL_VERBOSE=n
 CONFIG_DEBUG_OBJECTS_RCU_HEAD=n

tools/testing/selftests/rcutorture/configs/rcu/TREE08

@@ -18,7 +18,6 @@ CONFIG_RCU_FANOUT_LEAF=2
 CONFIG_RCU_NOCB_CPU=y
 CONFIG_RCU_NOCB_CPU_ALL=y
 CONFIG_DEBUG_LOCK_ALLOC=n
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=n
 CONFIG_RCU_CPU_STALL_VERBOSE=n
 CONFIG_RCU_BOOST=n

tools/testing/selftests/rcutorture/configs/rcu/TREE08-T

@@ -18,7 +18,6 @@ CONFIG_RCU_FANOUT_LEAF=2
 CONFIG_RCU_NOCB_CPU=y
 CONFIG_RCU_NOCB_CPU_ALL=y
 CONFIG_DEBUG_LOCK_ALLOC=n
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=n
 CONFIG_RCU_CPU_STALL_VERBOSE=n
 CONFIG_RCU_BOOST=n

tools/testing/selftests/rcutorture/configs/rcu/TREE09

@@ -13,7 +13,6 @@ CONFIG_SUSPEND=n
 CONFIG_HIBERNATION=n
 CONFIG_RCU_NOCB_CPU=n
 CONFIG_DEBUG_LOCK_ALLOC=n
-CONFIG_PROVE_RCU_DELAY=n
 CONFIG_RCU_CPU_STALL_INFO=n
 CONFIG_RCU_CPU_STALL_VERBOSE=n
 CONFIG_RCU_BOOST=n

tools/testing/selftests/rcutorture/configs/rcu/v0.0/P5-U-T-NH-sd-SMP-hp

@@ -13,7 +13,6 @@ CONFIG_PREEMPT_VOLUNTARY=n
 CONFIG_PREEMPT=y
 #CHECK#CONFIG_TREE_PREEMPT_RCU=y
 CONFIG_DEBUG_KERNEL=y
-CONFIG_PROVE_RCU_DELAY=y
 CONFIG_DEBUG_OBJECTS=y
 CONFIG_DEBUG_OBJECTS_RCU_HEAD=y
 CONFIG_RT_MUTEXES=y

tools/testing/selftests/rcutorture/configs/rcu/v3.12/P5-U-T-NH-sd-SMP-hp

@@ -13,7 +13,6 @@ CONFIG_PREEMPT_VOLUNTARY=n
 CONFIG_PREEMPT=y
 #CHECK#CONFIG_TREE_PREEMPT_RCU=y
 CONFIG_DEBUG_KERNEL=y
-CONFIG_PROVE_RCU_DELAY=y
 CONFIG_DEBUG_OBJECTS=y
 CONFIG_DEBUG_OBJECTS_RCU_HEAD=y
 CONFIG_RT_MUTEXES=y

tools/testing/selftests/rcutorture/configs/rcu/v3.3/P5-U-T-NH-sd-SMP-hp

@@ -13,7 +13,6 @@ CONFIG_PREEMPT_VOLUNTARY=n
 CONFIG_PREEMPT=y
 #CHECK#CONFIG_TREE_PREEMPT_RCU=y
 CONFIG_DEBUG_KERNEL=y
-CONFIG_PROVE_RCU_DELAY=y
 CONFIG_DEBUG_OBJECTS=y
 CONFIG_DEBUG_OBJECTS_RCU_HEAD=y
 CONFIG_RT_MUTEXES=y

tools/testing/selftests/rcutorture/configs/rcu/v3.5/P5-U-T-NH-sd-SMP-hp

@@ -13,7 +13,6 @@ CONFIG_PREEMPT_VOLUNTARY=n
 CONFIG_PREEMPT=y
 #CHECK#CONFIG_TREE_PREEMPT_RCU=y
 CONFIG_DEBUG_KERNEL=y
-CONFIG_PROVE_RCU_DELAY=y
 CONFIG_DEBUG_OBJECTS=y
 CONFIG_DEBUG_OBJECTS_RCU_HEAD=y
 CONFIG_RT_MUTEXES=y

tools/testing/selftests/rcutorture/doc/TREE_RCU-kconfig.txt

@@ -14,7 +14,6 @@ CONFIG_NO_HZ_FULL_SYSIDLE -- Do one.
 CONFIG_PREEMPT -- Do half.  (First three and #8.)
 CONFIG_PROVE_LOCKING -- Do all but two, covering CONFIG_PROVE_RCU and not.
 CONFIG_PROVE_RCU -- Do all but one under CONFIG_PROVE_LOCKING.
-CONFIG_PROVE_RCU_DELAY -- Do one.
 CONFIG_RCU_BOOST -- one of TREE_PREEMPT_RCU.
 CONFIG_RCU_BOOST_PRIO -- set to 2 for _BOOST testing.
 CONFIG_RCU_CPU_STALL_INFO -- do one with and without _VERBOSE.