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>

6 files changed, 108 insertions, 45 deletions

diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index d5b387669da..1f4517d55b1 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -139,7 +139,7 @@ static inline void account_system_vtime(struct task_struct *tsk)
 #endif
 
 #if defined(CONFIG_NO_HZ)
-#if defined(CONFIG_TINY_RCU)
+#if defined(CONFIG_TINY_RCU) || defined(CONFIG_TINY_PREEMPT_RCU)
 extern void rcu_enter_nohz(void);
 extern void rcu_exit_nohz(void);
 
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 6460fc65ed6..2fea6c8ef6b 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -82,11 +82,17 @@ extern struct group_info init_groups;
 # define CAP_INIT_BSET  CAP_FULL_SET
 
 #ifdef CONFIG_TREE_PREEMPT_RCU
+#define INIT_TASK_RCU_TREE_PREEMPT()					\
+	.rcu_blocked_node = NULL,
+#else
+#define INIT_TASK_RCU_TREE_PREEMPT(tsk)
+#endif
+#ifdef CONFIG_PREEMPT_RCU
 #define INIT_TASK_RCU_PREEMPT(tsk)					\
 	.rcu_read_lock_nesting = 0,					\
 	.rcu_read_unlock_special = 0,					\
-	.rcu_blocked_node = NULL,					\
-	.rcu_node_entry = LIST_HEAD_INIT(tsk.rcu_node_entry),
+	.rcu_node_entry = LIST_HEAD_INIT(tsk.rcu_node_entry),		\
+	INIT_TASK_RCU_TREE_PREEMPT()
 #else
 #define INIT_TASK_RCU_PREEMPT(tsk)
 #endif
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 27b44b3e302..24b89664938 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -58,7 +58,6 @@ struct rcu_head {
 };
 
 /* Exported common interfaces */
-extern void rcu_barrier(void);
 extern void rcu_barrier_bh(void);
 extern void rcu_barrier_sched(void);
 extern void synchronize_sched_expedited(void);
@@ -69,7 +68,7 @@ extern void rcu_init(void);
 
 #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU)
 #include <linux/rcutree.h>
-#elif defined(CONFIG_TINY_RCU)
+#elif defined(CONFIG_TINY_RCU) || defined(CONFIG_TINY_PREEMPT_RCU)
 #include <linux/rcutiny.h>
 #else
 #error "Unknown RCU implementation specified to kernel configuration"
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index e2e893144a8..4cc5eba4161 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -29,66 +29,51 @@
 
 void rcu_sched_qs(int cpu);
 void rcu_bh_qs(int cpu);
-static inline void rcu_note_context_switch(int cpu)
-{
-	rcu_sched_qs(cpu);
-}
 
+#ifdef CONFIG_TINY_RCU
 #define __rcu_read_lock()	preempt_disable()
 #define __rcu_read_unlock()	preempt_enable()
+#else /* #ifdef CONFIG_TINY_RCU */
+void __rcu_read_lock(void);
+void __rcu_read_unlock(void);
+#endif /* #else #ifdef CONFIG_TINY_RCU */
 #define __rcu_read_lock_bh()	local_bh_disable()
 #define __rcu_read_unlock_bh()	local_bh_enable()
-#define call_rcu_sched		call_rcu
+extern void call_rcu_sched(struct rcu_head *head,
+			   void (*func)(struct rcu_head *rcu));
 
 #define rcu_init_sched()	do { } while (0)
-extern void rcu_check_callbacks(int cpu, int user);
 
-static inline int rcu_needs_cpu(int cpu)
-{
-	return 0;
-}
+extern void synchronize_sched(void);
 
-/*
- * Return the number of grace periods.
- */
-static inline long rcu_batches_completed(void)
-{
-	return 0;
-}
+#ifdef CONFIG_TINY_RCU
 
-/*
- * Return the number of bottom-half grace periods.
- */
-static inline long rcu_batches_completed_bh(void)
-{
-	return 0;
-}
+#define call_rcu		call_rcu_sched
 
-static inline void rcu_force_quiescent_state(void)
+static inline void synchronize_rcu(void)
 {
+	synchronize_sched();
 }
 
-static inline void rcu_bh_force_quiescent_state(void)
+static inline void synchronize_rcu_expedited(void)
 {
+	synchronize_sched();	/* Only one CPU, so pretty fast anyway!!! */
 }
 
-static inline void rcu_sched_force_quiescent_state(void)
+static inline void rcu_barrier(void)
 {
+	rcu_barrier_sched();  /* Only one CPU, so only one list of callbacks! */
 }
 
-extern void synchronize_sched(void);
+#else /* #ifdef CONFIG_TINY_RCU */
 
-static inline void synchronize_rcu(void)
-{
-	synchronize_sched();
-}
+void synchronize_rcu(void);
+void rcu_barrier(void);
+void synchronize_rcu_expedited(void);
 
-static inline void synchronize_rcu_bh(void)
-{
-	synchronize_sched();
-}
+#endif /* #else #ifdef CONFIG_TINY_RCU */
 
-static inline void synchronize_rcu_expedited(void)
+static inline void synchronize_rcu_bh(void)
 {
 	synchronize_sched();
 }
@@ -117,15 +102,82 @@ static inline void rcu_exit_nohz(void)
 
 #endif /* #else #ifdef CONFIG_NO_HZ */
 
+#ifdef CONFIG_TINY_RCU
+
+static inline void rcu_preempt_note_context_switch(void)
+{
+}
+
 static inline void exit_rcu(void)
 {
 }
 
+static inline int rcu_needs_cpu(int cpu)
+{
+	return 0;
+}
+
 static inline int rcu_preempt_depth(void)
 {
 	return 0;
 }
 
+#else /* #ifdef CONFIG_TINY_RCU */
+
+void rcu_preempt_note_context_switch(void);
+extern void exit_rcu(void);
+int rcu_preempt_needs_cpu(void);
+
+static inline int rcu_needs_cpu(int cpu)
+{
+	return rcu_preempt_needs_cpu();
+}
+
+/*
+ * Defined as macro as it is a very low level header
+ * included from areas that don't even know about current
+ * FIXME: combine with include/linux/rcutree.h into rcupdate.h.
+ */
+#define rcu_preempt_depth() (current->rcu_read_lock_nesting)
+
+#endif /* #else #ifdef CONFIG_TINY_RCU */
+
+static inline void rcu_note_context_switch(int cpu)
+{
+	rcu_sched_qs(cpu);
+	rcu_preempt_note_context_switch();
+}
+
+extern void rcu_check_callbacks(int cpu, int user);
+
+/*
+ * Return the number of grace periods.
+ */
+static inline long rcu_batches_completed(void)
+{
+	return 0;
+}
+
+/*
+ * Return the number of bottom-half grace periods.
+ */
+static inline long rcu_batches_completed_bh(void)
+{
+	return 0;
+}
+
+static inline void rcu_force_quiescent_state(void)
+{
+}
+
+static inline void rcu_bh_force_quiescent_state(void)
+{
+}
+
+static inline void rcu_sched_force_quiescent_state(void)
+{
+}
+
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
 
 extern int rcu_scheduler_active __read_mostly;
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index c0ed1c056f2..c13b85dd22b 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -95,6 +95,8 @@ static inline void synchronize_rcu_bh_expedited(void)
 	synchronize_sched_expedited();
 }
 
+extern void rcu_barrier(void);
+
 extern void rcu_check_callbacks(int cpu, int user);
 
 extern long rcu_batches_completed(void);
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2c756666c11..e18473f0eb7 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1202,11 +1202,13 @@ struct task_struct {
 	unsigned int policy;
 	cpumask_t cpus_allowed;
 
-#ifdef CONFIG_TREE_PREEMPT_RCU
+#ifdef CONFIG_PREEMPT_RCU
 	int rcu_read_lock_nesting;
 	char rcu_read_unlock_special;
-	struct rcu_node *rcu_blocked_node;
 	struct list_head rcu_node_entry;
+#endif /* #ifdef CONFIG_PREEMPT_RCU */
+#ifdef CONFIG_TREE_PREEMPT_RCU
+	struct rcu_node *rcu_blocked_node;
 #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */
 
 #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT)
@@ -1740,7 +1742,7 @@ extern void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *
 #define tsk_used_math(p) ((p)->flags & PF_USED_MATH)
 #define used_math() tsk_used_math(current)
 
-#ifdef CONFIG_TREE_PREEMPT_RCU
+#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. */
@@ -1749,7 +1751,9 @@ static inline void rcu_copy_process(struct task_struct *p)
 {
 	p->rcu_read_lock_nesting = 0;
 	p->rcu_read_unlock_special = 0;
+#ifdef CONFIG_TREE_PREEMPT_RCU
 	p->rcu_blocked_node = NULL;
+#endif
 	INIT_LIST_HEAD(&p->rcu_node_entry);
 }