Merge branch 'rcu/next' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-2.6-rcu into core/rcu

17 files changed, 1207 insertions, 285 deletions

diff --git a/Documentation/RCU/trace.txt b/Documentation/RCU/trace.txt
index a851118775d..6a8c73f55b8 100644
--- a/Documentation/RCU/trace.txt
+++ b/Documentation/RCU/trace.txt
@@ -1,18 +1,22 @@
 CONFIG_RCU_TRACE debugfs Files and Formats
 
 
-The rcutree implementation of RCU provides debugfs trace output that
-summarizes counters and state.  This information is useful for debugging
-RCU itself, and can sometimes also help to debug abuses of RCU.
-The following sections describe the debugfs files and formats.
+The rcutree and rcutiny implementations of RCU provide debugfs trace
+output that summarizes counters and state.  This information is useful for
+debugging RCU itself, and can sometimes also help to debug abuses of RCU.
+The following sections describe the debugfs files and formats, first
+for rcutree and next for rcutiny.
 
 
-Hierarchical RCU debugfs Files and Formats
+CONFIG_TREE_RCU and CONFIG_TREE_PREEMPT_RCU debugfs Files and Formats
 
-This implementation of RCU provides three debugfs files under the
+These implementations of RCU provides five debugfs files under the
 top-level directory RCU: rcu/rcudata (which displays fields in struct
-rcu_data), rcu/rcugp (which displays grace-period counters), and
-rcu/rcuhier (which displays the struct rcu_node hierarchy).
+rcu_data), rcu/rcudata.csv (which is a .csv spreadsheet version of
+rcu/rcudata), rcu/rcugp (which displays grace-period counters),
+rcu/rcuhier (which displays the struct rcu_node hierarchy), and
+rcu/rcu_pending (which displays counts of the reasons that the
+rcu_pending() function decided that there was core RCU work to do).
 
 The output of "cat rcu/rcudata" looks as follows:
 
@@ -130,7 +134,8 @@ o	"ci" is the number of RCU callbacks that have been invoked for
 	been registered in absence of CPU-hotplug activity.
 
 o	"co" is the number of RCU callbacks that have been orphaned due to
-	this CPU going offline.
+	this CPU going offline.  These orphaned callbacks have been moved
+	to an arbitrarily chosen online CPU.
 
 o	"ca" is the number of RCU callbacks that have been adopted due to
 	other CPUs going offline.  Note that ci+co-ca+ql is the number of
@@ -168,12 +173,12 @@ o	"gpnum" is the number of grace periods that have started.  It is
 
 The output of "cat rcu/rcuhier" looks as follows, with very long lines:
 
-c=6902 g=6903 s=2 jfq=3 j=72c7 nfqs=13142/nfqsng=0(13142) fqlh=6 oqlen=0
+c=6902 g=6903 s=2 jfq=3 j=72c7 nfqs=13142/nfqsng=0(13142) fqlh=6
 1/1 .>. 0:127 ^0    
 3/3 .>. 0:35 ^0    0/0 .>. 36:71 ^1    0/0 .>. 72:107 ^2    0/0 .>. 108:127 ^3    
 3/3f .>. 0:5 ^0    2/3 .>. 6:11 ^1    0/0 .>. 12:17 ^2    0/0 .>. 18:23 ^3    0/0 .>. 24:29 ^4    0/0 .>. 30:35 ^5    0/0 .>. 36:41 ^0    0/0 .>. 42:47 ^1    0/0 .>. 48:53 ^2    0/0 .>. 54:59 ^3    0/0 .>. 60:65 ^4    0/0 .>. 66:71 ^5    0/0 .>. 72:77 ^0    0/0 .>. 78:83 ^1    0/0 .>. 84:89 ^2    0/0 .>. 90:95 ^3    0/0 .>. 96:101 ^4    0/0 .>. 102:107 ^5    0/0 .>. 108:113 ^0    0/0 .>. 114:119 ^1    0/0 .>. 120:125 ^2    0/0 .>. 126:127 ^3    
 rcu_bh:
-c=-226 g=-226 s=1 jfq=-5701 j=72c7 nfqs=88/nfqsng=0(88) fqlh=0 oqlen=0
+c=-226 g=-226 s=1 jfq=-5701 j=72c7 nfqs=88/nfqsng=0(88) fqlh=0
 0/1 .>. 0:127 ^0    
 0/3 .>. 0:35 ^0    0/0 .>. 36:71 ^1    0/0 .>. 72:107 ^2    0/0 .>. 108:127 ^3    
 0/3f .>. 0:5 ^0    0/3 .>. 6:11 ^1    0/0 .>. 12:17 ^2    0/0 .>. 18:23 ^3    0/0 .>. 24:29 ^4    0/0 .>. 30:35 ^5    0/0 .>. 36:41 ^0    0/0 .>. 42:47 ^1    0/0 .>. 48:53 ^2    0/0 .>. 54:59 ^3    0/0 .>. 60:65 ^4    0/0 .>. 66:71 ^5    0/0 .>. 72:77 ^0    0/0 .>. 78:83 ^1    0/0 .>. 84:89 ^2    0/0 .>. 90:95 ^3    0/0 .>. 96:101 ^4    0/0 .>. 102:107 ^5    0/0 .>. 108:113 ^0    0/0 .>. 114:119 ^1    0/0 .>. 120:125 ^2    0/0 .>. 126:127 ^3
@@ -212,11 +217,6 @@ o	"fqlh" is the number of calls to force_quiescent_state() that
 	exited immediately (without even being counted in nfqs above)
 	due to contention on ->fqslock.
 
-o	"oqlen" is the number of callbacks on the "orphan" callback
-	list.  RCU callbacks are placed on this list by CPUs going
-	offline, and are "adopted" either by the CPU helping the outgoing
-	CPU or by the next rcu_barrier*() call, whichever comes first.
-
 o	Each element of the form "1/1 0:127 ^0" represents one struct
 	rcu_node.  Each line represents one level of the hierarchy, from
 	root to leaves.  It is best to think of the rcu_data structures
@@ -326,3 +326,115 @@ o	"nn" is the number of times that this CPU needed nothing.  Alert
 	readers will note that the rcu "nn" number for a given CPU very
 	closely matches the rcu_bh "np" number for that same CPU.  This
 	is due to short-circuit evaluation in rcu_pending().
+
+
+CONFIG_TINY_RCU and CONFIG_TINY_PREEMPT_RCU debugfs Files and Formats
+
+These implementations of RCU provides a single debugfs file under the
+top-level directory RCU, namely rcu/rcudata, which displays fields in
+rcu_bh_ctrlblk, rcu_sched_ctrlblk and, for CONFIG_TINY_PREEMPT_RCU,
+rcu_preempt_ctrlblk.
+
+The output of "cat rcu/rcudata" is as follows:
+
+rcu_preempt: qlen=24 gp=1097669 g197/p197/c197 tasks=...
+             ttb=. btg=no ntb=184 neb=0 nnb=183 j=01f7 bt=0274
+             normal balk: nt=1097669 gt=0 bt=371 b=0 ny=25073378 nos=0
+             exp balk: bt=0 nos=0
+rcu_sched: qlen: 0
+rcu_bh: qlen: 0
+
+This is split into rcu_preempt, rcu_sched, and rcu_bh sections, with the
+rcu_preempt section appearing only in CONFIG_TINY_PREEMPT_RCU builds.
+The last three lines of the rcu_preempt section appear only in
+CONFIG_RCU_BOOST kernel builds.  The fields are as follows:
+
+o	"qlen" is the number of RCU callbacks currently waiting either
+	for an RCU grace period or waiting to be invoked.  This is the
+	only field present for rcu_sched and rcu_bh, due to the
+	short-circuiting of grace period in those two cases.
+
+o	"gp" is the number of grace periods that have completed.
+
+o	"g197/p197/c197" displays the grace-period state, with the
+	"g" number being the number of grace periods that have started
+	(mod 256), the "p" number being the number of grace periods
+	that the CPU has responded to (also mod 256), and the "c"
+	number being the number of grace periods that have completed
+	(once again mode 256).
+
+	Why have both "gp" and "g"?  Because the data flowing into
+	"gp" is only present in a CONFIG_RCU_TRACE kernel.
+
+o	"tasks" is a set of bits.  The first bit is "T" if there are
+	currently tasks that have recently blocked within an RCU
+	read-side critical section, the second bit is "N" if any of the
+	aforementioned tasks are blocking the current RCU grace period,
+	and the third bit is "E" if any of the aforementioned tasks are
+	blocking the current expedited grace period.  Each bit is "."
+	if the corresponding condition does not hold.
+
+o	"ttb" is a single bit.  It is "B" if any of the blocked tasks
+	need to be priority boosted and "." otherwise.
+
+o	"btg" indicates whether boosting has been carried out during
+	the current grace period, with "exp" indicating that boosting
+	is in progress for an expedited grace period, "no" indicating
+	that boosting has not yet started for a normal grace period,
+	"begun" indicating that boosting has bebug for a normal grace
+	period, and "done" indicating that boosting has completed for
+	a normal grace period.
+
+o	"ntb" is the total number of tasks subjected to RCU priority boosting
+	periods since boot.
+
+o	"neb" is the number of expedited grace periods that have had
+	to resort to RCU priority boosting since boot.
+
+o	"nnb" is the number of normal grace periods that have had
+	to resort to RCU priority boosting since boot.
+
+o	"j" is the low-order 12 bits of the jiffies counter in hexadecimal.
+
+o	"bt" is the low-order 12 bits of the value that the jiffies counter
+	will have at the next time that boosting is scheduled to begin.
+
+o	In the line beginning with "normal balk", the fields are as follows:
+
+	o	"nt" is the number of times that the system balked from
+		boosting because there were no blocked tasks to boost.
+		Note that the system will balk from boosting even if the
+		grace period is overdue when the currently running task
+		is looping within an RCU read-side critical section.
+		There is no point in boosting in this case, because
+		boosting a running task won't make it run any faster.
+
+	o	"gt" is the number of times that the system balked
+		from boosting because, although there were blocked tasks,
+		none of them were preventing the current grace period
+		from completing.
+
+	o	"bt" is the number of times that the system balked
+		from boosting because boosting was already in progress.
+
+	o	"b" is the number of times that the system balked from
+		boosting because boosting had already completed for
+		the grace period in question.
+
+	o	"ny" is the number of times that the system balked from
+		boosting because it was not yet time to start boosting
+		the grace period in question.
+
+	o	"nos" is the number of times that the system balked from
+		boosting for inexplicable ("not otherwise specified")
+		reasons.  This can actually happen due to races involving
+		increments of the jiffies counter.
+
+o	In the line beginning with "exp balk", the fields are as follows:
+
+	o	"bt" is the number of times that the system balked from
+		boosting because there were no blocked tasks to boost.
+
+	o	"nos" is the number of times that the system balked from
+		 boosting for inexplicable ("not otherwise specified")
+		 reasons.
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 1f8c06ce0fa..6b281fae114 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -83,6 +83,12 @@ extern struct group_info init_groups;
  */
 # define CAP_INIT_BSET  CAP_FULL_SET
 
+#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,
@@ -94,7 +100,8 @@ 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_TREE_PREEMPT()					\
+	INIT_TASK_RCU_BOOST()
 #else
 #define INIT_TASK_RCU_PREEMPT(tsk)
 #endif
diff --git a/include/linux/rculist.h b/include/linux/rculist.h
index f31ef61f1c6..2dea94fc440 100644
--- a/include/linux/rculist.h
+++ b/include/linux/rculist.h
@@ -241,11 +241,6 @@ static inline void list_splice_init_rcu(struct list_head *list,
 #define list_first_entry_rcu(ptr, type, member) \
 	list_entry_rcu((ptr)->next, type, member)
 
-#define __list_for_each_rcu(pos, head) \
-	for (pos = rcu_dereference_raw(list_next_rcu(head)); \
-		pos != (head); \
-		pos = rcu_dereference_raw(list_next_rcu((pos)))
-
 /**
  * list_for_each_entry_rcu	-	iterate over rcu list of given type
  * @pos:	the type * to use as a loop cursor.
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 03cda7bed98..af561485628 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -47,6 +47,8 @@
 extern int rcutorture_runnable; /* for sysctl */
 #endif /* #ifdef CONFIG_RCU_TORTURE_TEST */
 
+#define UINT_CMP_GE(a, b)	(UINT_MAX / 2 >= (a) - (b))
+#define UINT_CMP_LT(a, b)	(UINT_MAX / 2 < (a) - (b))
 #define ULONG_CMP_GE(a, b)	(ULONG_MAX / 2 >= (a) - (b))
 #define ULONG_CMP_LT(a, b)	(ULONG_MAX / 2 < (a) - (b))
 
@@ -66,7 +68,6 @@ extern void call_rcu_sched(struct rcu_head *head,
 extern void synchronize_sched(void);
 extern void rcu_barrier_bh(void);
 extern void rcu_barrier_sched(void);
-extern void synchronize_sched_expedited(void);
 extern int sched_expedited_torture_stats(char *page);
 
 static inline void __rcu_read_lock_bh(void)
@@ -118,7 +119,6 @@ static inline int rcu_preempt_depth(void)
 #endif /* #else #ifdef CONFIG_PREEMPT_RCU */
 
 /* Internal to kernel */
-extern void rcu_init(void);
 extern void rcu_sched_qs(int cpu);
 extern void rcu_bh_qs(int cpu);
 extern void rcu_check_callbacks(int cpu, int user);
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index 13877cb93a6..30ebd7c8d87 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -27,7 +27,9 @@
 
 #include <linux/cache.h>
 
-#define rcu_init_sched()	do { } while (0)
+static inline void rcu_init(void)
+{
+}
 
 #ifdef CONFIG_TINY_RCU
 
@@ -58,6 +60,11 @@ static inline void synchronize_rcu_bh_expedited(void)
 	synchronize_sched();
 }
 
+static inline void synchronize_sched_expedited(void)
+{
+	synchronize_sched();
+}
+
 #ifdef CONFIG_TINY_RCU
 
 static inline void rcu_preempt_note_context_switch(void)
@@ -125,16 +132,12 @@ static inline void rcu_cpu_stall_reset(void)
 }
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
-
 extern int rcu_scheduler_active __read_mostly;
 extern void rcu_scheduler_starting(void);
-
 #else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
-
 static inline void rcu_scheduler_starting(void)
 {
 }
-
 #endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */
 
 #endif /* __LINUX_RCUTINY_H */
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index 95518e62879..3a933482734 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -30,6 +30,7 @@
 #ifndef __LINUX_RCUTREE_H
 #define __LINUX_RCUTREE_H
 
+extern void rcu_init(void);
 extern void rcu_note_context_switch(int cpu);
 extern int rcu_needs_cpu(int cpu);
 extern void rcu_cpu_stall_reset(void);
@@ -47,6 +48,7 @@ static inline void exit_rcu(void)
 #endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */
 
 extern void synchronize_rcu_bh(void);
+extern void synchronize_sched_expedited(void);
 extern void synchronize_rcu_expedited(void);
 
 static inline void synchronize_rcu_bh_expedited(void)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 223874538b3..d8005503cc6 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1229,6 +1229,9 @@ 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;
@@ -1759,7 +1762,8 @@ extern void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *
 #ifdef CONFIG_PREEMPT_RCU
 
 #define RCU_READ_UNLOCK_BLOCKED (1 << 0) /* blocked while in RCU read-side. */
-#define RCU_READ_UNLOCK_NEED_QS (1 << 1) /* RCU core needs CPU response. */
+#define RCU_READ_UNLOCK_BOOSTED (1 << 1) /* boosted while in RCU read-side. */
+#define RCU_READ_UNLOCK_NEED_QS (1 << 2) /* RCU core needs CPU response. */
 
 static inline void rcu_copy_process(struct task_struct *p)
 {
@@ -1767,7 +1771,10 @@ static inline void rcu_copy_process(struct task_struct *p)
 	p->rcu_read_unlock_special = 0;
 #ifdef CONFIG_TREE_PREEMPT_RCU
 	p->rcu_blocked_node = NULL;
-#endif
+#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);
 }
 
diff --git a/init/Kconfig b/init/Kconfig
index c9728992a77..526ec1c7456 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -393,7 +393,6 @@ config PREEMPT_RCU
 
 config RCU_TRACE
 	bool "Enable tracing for RCU"
-	depends on TREE_RCU || TREE_PREEMPT_RCU
 	help
 	  This option provides tracing in RCU which presents stats
 	  in debugfs for debugging RCU implementation.
@@ -459,6 +458,60 @@ config TREE_RCU_TRACE
 	  TREE_PREEMPT_RCU implementations, permitting Makefile to
 	  trivially select kernel/rcutree_trace.c.
 
+config RCU_BOOST
+	bool "Enable RCU priority boosting"
+	depends on RT_MUTEXES && TINY_PREEMPT_RCU
+	default n
+	help
+	  This option boosts the priority of preempted RCU readers that
+	  block the current preemptible RCU grace period for too long.
+	  This option also prevents heavy loads from blocking RCU
+	  callback invocation for all flavors of RCU.
+
+	  Say Y here if you are working with real-time apps or heavy loads
+	  Say N here if you are unsure.
+
+config RCU_BOOST_PRIO
+	int "Real-time priority to boost RCU readers to"
+	range 1 99
+	depends on RCU_BOOST
+	default 1
+	help
+	  This option specifies the real-time priority to which preempted
+	  RCU readers are to be boosted.  If you are working with CPU-bound
+	  real-time applications, you should specify a priority higher then
+	  the highest-priority CPU-bound application.
+
+	  Specify the real-time priority, or take the default if unsure.
+
+config RCU_BOOST_DELAY
+	int "Milliseconds to delay boosting after RCU grace-period start"
+	range 0 3000
+	depends on RCU_BOOST
+	default 500
+	help
+	  This option specifies the time to wait after the beginning of
+	  a given grace period before priority-boosting preempted RCU
+	  readers blocking that grace period.  Note that any RCU reader
+	  blocking an expedited RCU grace period is boosted immediately.
+
+	  Accept the default if unsure.
+
+config SRCU_SYNCHRONIZE_DELAY
+	int "Microseconds to delay before waiting for readers"
+	range 0 20
+	default 10
+	help
+	  This option controls how long SRCU delays before entering its
+	  loop waiting on SRCU readers.  The purpose of this loop is
+	  to avoid the unconditional context-switch penalty that would
+	  otherwise be incurred if there was an active SRCU reader,
+	  in a manner similar to adaptive locking schemes.  This should
+	  be set to be a bit longer than the common-case SRCU read-side
+	  critical-section overhead.
+
+	  Accept the default if unsure.
+
 endmenu # "RCU Subsystem"
 
 config IKCONFIG
diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c
index d806735342a..03449372474 100644
--- a/kernel/rcutiny.c
+++ b/kernel/rcutiny.c
@@ -36,31 +36,16 @@
 #include <linux/time.h>
 #include <linux/cpu.h>
 
-/* Global control variables for rcupdate callback mechanism. */
-struct rcu_ctrlblk {
-	struct rcu_head *rcucblist;	/* List of pending callbacks (CBs). */
-	struct rcu_head **donetail;	/* ->next pointer of last "done" CB. */
-	struct rcu_head **curtail;	/* ->next pointer of last CB. */
-};
-
-/* Definition for rcupdate control block. */
-static struct rcu_ctrlblk rcu_sched_ctrlblk = {
-	.donetail	= &rcu_sched_ctrlblk.rcucblist,
-	.curtail	= &rcu_sched_ctrlblk.rcucblist,
-};
-
-static struct rcu_ctrlblk rcu_bh_ctrlblk = {
-	.donetail	= &rcu_bh_ctrlblk.rcucblist,
-	.curtail	= &rcu_bh_ctrlblk.rcucblist,
-};
-
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-int rcu_scheduler_active __read_mostly;
-EXPORT_SYMBOL_GPL(rcu_scheduler_active);
-#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+/* Controls for rcu_kthread() kthread, replacing RCU_SOFTIRQ used previously. */
+static struct task_struct *rcu_kthread_task;
+static DECLARE_WAIT_QUEUE_HEAD(rcu_kthread_wq);
+static unsigned long have_rcu_kthread_work;
+static void invoke_rcu_kthread(void);
 
 /* Forward declarations for rcutiny_plugin.h. */
-static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp);
+struct rcu_ctrlblk;
+static void rcu_process_callbacks(struct rcu_ctrlblk *rcp);
+static int rcu_kthread(void *arg);
 static void __call_rcu(struct rcu_head *head,
 		       void (*func)(struct rcu_head *rcu),
 		       struct rcu_ctrlblk *rcp);
@@ -123,7 +108,7 @@ void rcu_sched_qs(int cpu)
 {
 	if (rcu_qsctr_help(&rcu_sched_ctrlblk) +
 	    rcu_qsctr_help(&rcu_bh_ctrlblk))
-		raise_softirq(RCU_SOFTIRQ);
+		invoke_rcu_kthread();
 }
 
 /*
@@ -132,7 +117,7 @@ void rcu_sched_qs(int cpu)
 void rcu_bh_qs(int cpu)
 {
 	if (rcu_qsctr_help(&rcu_bh_ctrlblk))
-		raise_softirq(RCU_SOFTIRQ);
+		invoke_rcu_kthread();
 }
 
 /*
@@ -152,13 +137,14 @@ void rcu_check_callbacks(int cpu, int user)
 }
 
 /*
- * Helper function for rcu_process_callbacks() that operates on the
- * specified rcu_ctrlkblk structure.
+ * Invoke the RCU callbacks on the specified rcu_ctrlkblk structure
+ * whose grace period has elapsed.
  */
-static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp)
+static void rcu_process_callbacks(struct rcu_ctrlblk *rcp)
 {
 	struct rcu_head *next, *list;
 	unsigned long flags;
+	RCU_TRACE(int cb_count = 0);
 
 	/* If no RCU callbacks ready to invoke, just return. */
 	if (&rcp->rcucblist == rcp->donetail)
@@ -180,19 +166,58 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp)
 		next = list->next;
 		prefetch(next);
 		debug_rcu_head_unqueue(list);
+		local_bh_disable();
 		list->func(list);
+		local_bh_enable();
 		list = next;
+		RCU_TRACE(cb_count++);
 	}
+	RCU_TRACE(rcu_trace_sub_qlen(rcp, cb_count));
 }
 
 /*
- * Invoke any callbacks whose grace period has completed.
+ * This kthread invokes RCU callbacks whose grace periods have
+ * elapsed.  It is awakened as needed, and takes the place of the
+ * RCU_SOFTIRQ that was used previously for this purpose.
+ * This is a kthread, but it is never stopped, at least not until
+ * the system goes down.
  */
-static void rcu_process_callbacks(struct softirq_action *unused)
+static int rcu_kthread(void *arg)
 {
-	__rcu_process_callbacks(&rcu_sched_ctrlblk);
-	__rcu_process_callbacks(&rcu_bh_ctrlblk);
-	rcu_preempt_process_callbacks();
+	unsigned long work;
+	unsigned long morework;
+	unsigned long flags;
+
+	for (;;) {
+		wait_event(rcu_kthread_wq, have_rcu_kthread_work != 0);
+		morework = rcu_boost();
+		local_irq_save(flags);
+		work = have_rcu_kthread_work;
+		have_rcu_kthread_work = morework;
+		local_irq_restore(flags);
+		if (work) {
+			rcu_process_callbacks(&rcu_sched_ctrlblk);
+			rcu_process_callbacks(&rcu_bh_ctrlblk);
+			rcu_preempt_process_callbacks();
+		}
+		schedule_timeout_interruptible(1); /* Leave CPU for others. */
+	}
+
+	return 0;  /* Not reached, but needed to shut gcc up. */
+}
+
+/*
+ * Wake up rcu_kthread() to process callbacks now eligible for invocation
+ * or to boost readers.
+ */
+static void invoke_rcu_kthread(void)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	have_rcu_kthread_work = 1;
+	wake_up(&rcu_kthread_wq);
+	local_irq_restore(flags);
 }
 
 /*
@@ -230,6 +255,7 @@ static void __call_rcu(struct rcu_head *head,
 	local_irq_save(flags);
 	*rcp->curtail = head;
 	rcp->curtail = &head->next;
+	RCU_TRACE(rcp->qlen++);
 	local_irq_restore(flags);
 }
 
@@ -282,7 +308,16 @@ void rcu_barrier_sched(void)
 }
 EXPORT_SYMBOL_GPL(rcu_barrier_sched);
 
-void __init rcu_init(void)
+/*
+ * Spawn the kthread that invokes RCU callbacks.
+ */
+static int __init rcu_spawn_kthreads(void)
 {
-	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
+	struct sched_param sp;
+
+	rcu_kthread_task = kthread_run(rcu_kthread, NULL, "rcu_kthread");
+	sp.sched_priority = RCU_BOOST_PRIO;
+	sched_setscheduler_nocheck(rcu_kthread_task, SCHED_FIFO, &sp);
+	return 0;
 }
+early_initcall(rcu_spawn_kthreads);
diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h
index 6ceca4f745f..015abaea962 100644
--- a/kernel/rcutiny_plugin.h
+++ b/kernel/rcutiny_plugin.h
@@ -22,6 +22,40 @@
  * Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
  */
 
+#include <linux/kthread.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+
+#ifdef CONFIG_RCU_TRACE
+#define RCU_TRACE(stmt)	stmt
+#else /* #ifdef CONFIG_RCU_TRACE */
+#define RCU_TRACE(stmt)
+#endif /* #else #ifdef CONFIG_RCU_TRACE */
+
+/* Global control variables for rcupdate callback mechanism. */
+struct rcu_ctrlblk {
+	struct rcu_head *rcucblist;	/* List of pending callbacks (CBs). */
+	struct rcu_head **donetail;	/* ->next pointer of last "done" CB. */
+	struct rcu_head **curtail;	/* ->next pointer of last CB. */
+	RCU_TRACE(long qlen);		/* Number of pending CBs. */
+};
+
+/* Definition for rcupdate control block. */
+static struct rcu_ctrlblk rcu_sched_ctrlblk = {
+	.donetail	= &rcu_sched_ctrlblk.rcucblist,
+	.curtail	= &rcu_sched_ctrlblk.rcucblist,
+};
+
+static struct rcu_ctrlblk rcu_bh_ctrlblk = {
+	.donetail	= &rcu_bh_ctrlblk.rcucblist,
+	.curtail	= &rcu_bh_ctrlblk.rcucblist,
+};
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+int rcu_scheduler_active __read_mostly;
+EXPORT_SYMBOL_GPL(rcu_scheduler_active);
+#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
+
 #ifdef CONFIG_TINY_PREEMPT_RCU
 
 #include <linux/delay.h>
@@ -46,17 +80,45 @@ struct rcu_preempt_ctrlblk {
 	struct list_head *gp_tasks;
 				/* Pointer to the first task blocking the */
 				/*  current grace period, or NULL if there */
-				/*  is not such task. */
+				/*  is no such task. */
 	struct list_head *exp_tasks;
 				/* Pointer to first task blocking the */
 				/*  current expedited grace period, or NULL */
 				/*  if there is no such task.  If there */
 				/*  is no current expedited grace period, */
 				/*  then there cannot be any such task. */
+#ifdef CONFIG_RCU_BOOST
+	struct list_head *boost_tasks;
+				/* Pointer to first task that needs to be */
+				/*  priority-boosted, or NULL if no priority */
+				/*  boosting is needed.  If there is no */
+				/*  current or expedited grace period, there */
+				/*  can be no such task. */
+#endif /* #ifdef CONFIG_RCU_BOOST */
 	u8 gpnum;		/* Current grace period. */
 	u8 gpcpu;		/* Last grace period blocked by the CPU. */
 	u8 completed;		/* Last grace period completed. */
 				/*  If all three are equal, RCU is idle. */
+#ifdef CONFIG_RCU_BOOST
+	s8 boosted_this_gp;	/* Has boosting already happened? */
+	unsigned long boost_time; /* When to start boosting (jiffies) */
+#endif /* #ifdef CONFIG_RCU_BOOST */
+#ifdef CONFIG_RCU_TRACE
+	unsigned long n_grace_periods;
+#ifdef CONFIG_RCU_BOOST
+	unsigned long n_tasks_boosted;
+	unsigned long n_exp_boosts;
+	unsigned long n_normal_boosts;
+	unsigned long n_normal_balk_blkd_tasks;
+	unsigned long n_normal_balk_gp_tasks;
+	unsigned long n_normal_balk_boost_tasks;
+	unsigned long n_normal_balk_boosted;
+	unsigned long n_normal_balk_notyet;
+	unsigned long n_normal_balk_nos;
+	unsigned long n_exp_balk_blkd_tasks;
+	unsigned long n_exp_balk_nos;
+#endif /* #ifdef CONFIG_RCU_BOOST */
+#endif /* #ifdef CONFIG_RCU_TRACE */
 };
 
 static struct rcu_preempt_ctrlblk rcu_preempt_ctrlblk = {
@@ -122,6 +184,210 @@ static int rcu_preempt_gp_in_progress(void)
 }
 
 /*
+ * Advance a ->blkd_tasks-list pointer to the next entry, instead
+ * returning NULL if at the end of the list.
+ */
+static struct list_head *rcu_next_node_entry(struct task_struct *t)
+{
+	struct list_head *np;
+
+	np = t->rcu_node_entry.next;
+	if (np == &rcu_preempt_ctrlblk.blkd_tasks)
+		np = NULL;
+	return np;
+}
+
+#ifdef CONFIG_RCU_TRACE
+
+#ifdef CONFIG_RCU_BOOST
+static void rcu_initiate_boost_trace(void);
+static void rcu_initiate_exp_boost_trace(void);
+#endif /* #ifdef CONFIG_RCU_BOOST */
+
+/*
+ * Dump additional statistice for TINY_PREEMPT_RCU.
+ */
+static void show_tiny_preempt_stats(struct seq_file *m)
+{
+	seq_printf(m, "rcu_preempt: qlen=%ld gp=%lu g%u/p%u/c%u tasks=%c%c%c\n",
+		   rcu_preempt_ctrlblk.rcb.qlen,
+		   rcu_preempt_ctrlblk.n_grace_periods,
+		   rcu_preempt_ctrlblk.gpnum,
+		   rcu_preempt_ctrlblk.gpcpu,
+		   rcu_preempt_ctrlblk.completed,
+		   "T."[list_empty(&rcu_preempt_ctrlblk.blkd_tasks)],
+		   "N."[!rcu_preempt_ctrlblk.gp_tasks],
+		   "E."[!rcu_preempt_ctrlblk.exp_tasks]);
+#ifdef CONFIG_RCU_BOOST
+	seq_printf(m, "             ttb=%c btg=",
+		   "B."[!rcu_preempt_ctrlblk.boost_tasks]);
+	switch (rcu_preempt_ctrlblk.boosted_this_gp) {
+	case -1:
+		seq_puts(m, "exp");
+		break;
+	case 0:
+		seq_puts(m, "no");
+		break;
+	case 1:
+		seq_puts(m, "begun");
+		break;
+	case 2:
+		seq_puts(m, "done");
+		break;
+	default:
+		seq_printf(m, "?%d?", rcu_preempt_ctrlblk.boosted_this_gp);
+	}
+	seq_printf(m, " ntb=%lu neb=%lu nnb=%lu j=%04x bt=%04x\n",
+		   rcu_preempt_ctrlblk.n_tasks_boosted,
+		   rcu_preempt_ctrlblk.n_exp_boosts,
+		   rcu_preempt_ctrlblk.n_normal_boosts,
+		   (int)(jiffies & 0xffff),
+		   (int)(rcu_preempt_ctrlblk.boost_time & 0xffff));
+	seq_printf(m, "             %s: nt=%lu gt=%lu bt=%lu b=%lu ny=%lu nos=%lu\n",
+		   "normal balk",
+		   rcu_preempt_ctrlblk.n_normal_balk_blkd_tasks,
+		   rcu_preempt_ctrlblk.n_normal_balk_gp_tasks,
+		   rcu_preempt_ctrlblk.n_normal_balk_boost_tasks,
+		   rcu_preempt_ctrlblk.n_normal_balk_boosted,
+		   rcu_preempt_ctrlblk.n_normal_balk_notyet,
+		   rcu_preempt_ctrlblk.n_normal_balk_nos);
+	seq_printf(m, "             exp balk: bt=%lu nos=%lu\n",
+		   rcu_preempt_ctrlblk.n_exp_balk_blkd_tasks,
+		   rcu_preempt_ctrlblk.n_exp_balk_nos);
+#endif /* #ifdef CONFIG_RCU_BOOST */
+}
+
+#endif /* #ifdef CONFIG_RCU_TRACE */
+
+#ifdef CONFIG_RCU_BOOST
+
+#include "rtmutex_common.h"
+
+/*
+ * Carry out RCU priority boosting on the task indicated by ->boost_tasks,
+ * and advance ->boost_tasks to the next task in the ->blkd_tasks list.
+ */
+static int rcu_boost(void)
+{
+	unsigned long flags;
+	struct rt_mutex mtx;
+	struct list_head *np;
+	struct task_struct *t;
+
+	if (rcu_preempt_ctrlblk.boost_tasks == NULL)
+		return 0;  /* Nothing to boost. */
+	raw_local_irq_save(flags);
+	rcu_preempt_ctrlblk.boosted_this_gp++;
+	t = container_of(rcu_preempt_ctrlblk.boost_tasks, struct task_struct,
+			 rcu_node_entry);
+	np = rcu_next_node_entry(t);
+	rt_mutex_init_proxy_locked(&mtx, t);
+	t->rcu_boost_mutex = &mtx;
+	t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BOOSTED;
+	raw_local_irq_restore(flags);
+	rt_mutex_lock(&mtx);
+	RCU_TRACE(rcu_preempt_ctrlblk.n_tasks_boosted++);
+	rcu_preempt_ctrlblk.boosted_this_gp++;
+	rt_mutex_unlock(&mtx);
+	return rcu_preempt_ctrlblk.boost_tasks != NULL;
+}
+
+/*
+ * Check to see if it is now time to start boosting RCU readers blocking
+ * the current grace period, and, if so, tell the rcu_kthread_task to
+ * start boosting them.  If there is an expedited boost in progress,
+ * we wait for it to complete.
+ *
+ * If there are no blocked readers blocking the current grace period,
+ * return 0 to let the caller know, otherwise return 1.  Note that this
+ * return value is independent of whether or not boosting was done.
+ */
+static int rcu_initiate_boost(void)
+{
+	if (!rcu_preempt_blocked_readers_cgp()) {
+		RCU_TRACE(rcu_preempt_ctrlblk.n_normal_balk_blkd_tasks++);
+		return 0;
+	}
+	if (rcu_preempt_ctrlblk.gp_tasks != NULL &&
+	    rcu_preempt_ctrlblk.boost_tasks == NULL &&
+	    rcu_preempt_ctrlblk.boosted_this_gp == 0 &&
+	    ULONG_CMP_GE(jiffies, rcu_preempt_ctrlblk.boost_time)) {
+		rcu_preempt_ctrlblk.boost_tasks = rcu_preempt_ctrlblk.gp_tasks;
+		invoke_rcu_kthread();
+		RCU_TRACE(rcu_preempt_ctrlblk.n_normal_boosts++);
+	} else
+		RCU_TRACE(rcu_initiate_boost_trace());
+	return 1;
+}
+
+/*
+ * Initiate boosting for an expedited grace period.
+ */
+static void rcu_initiate_expedited_boost(void)
+{
+	unsigned long flags;
+
+	raw_local_irq_save(flags);
+	if (!list_empty(&rcu_preempt_ctrlblk.blkd_tasks)) {
+		rcu_preempt_ctrlblk.boost_tasks =
+			rcu_preempt_ctrlblk.blkd_tasks.next;
+		rcu_preempt_ctrlblk.boosted_this_gp = -1;
+		invoke_rcu_kthread();
+		RCU_TRACE(rcu_preempt_ctrlblk.n_exp_boosts++);
+	} else
+		RCU_TRACE(rcu_initiate_exp_boost_trace());
+	raw_local_irq_restore(flags);
+}
+
+#define RCU_BOOST_DELAY_JIFFIES DIV_ROUND_UP(CONFIG_RCU_BOOST_DELAY * HZ, 1000);
+
+/*
+ * Do priority-boost accounting for the start of a new grace period.
+ */
+static void rcu_preempt_boost_start_gp(void)
+{
+	rcu_preempt_ctrlblk.boost_time = jiffies + RCU_BOOST_DELAY_JIFFIES;
+	if (rcu_preempt_ctrlblk.boosted_this_gp > 0)
+		rcu_preempt_ctrlblk.boosted_this_gp = 0;
+}
+
+#else /* #ifdef CONFIG_RCU_BOOST */
+
+/*