Merge branches 'doc.2013.12.03a', 'fixes.2013.12.12a', 'rcutorture.2013.12.03a' and 'sparse.2013.12.12a' into HEAD
doc.2013.12.03a: Topic branch for documentation changes.
fixes.2013.12.12a: Topic branch for miscellaneous fixes.
rcutorture.2013.12.03a: Topic branch for new rcutorture/KVM scripting.
sparse.2013.12.12a: Topic branch for sparse-RCU changes.
diff --git a/Documentation/RCU/trace.txt b/Documentation/RCU/trace.txt
index f3778f8..910870b 100644
--- a/Documentation/RCU/trace.txt
+++ b/Documentation/RCU/trace.txt
@@ -396,14 +396,14 @@
The output of "cat rcu/rcu_sched/rcu_pending" looks as follows:
- 0!np=26111 qsp=29 rpq=5386 cbr=1 cng=570 gpc=3674 gps=577 nn=15903
- 1!np=28913 qsp=35 rpq=6097 cbr=1 cng=448 gpc=3700 gps=554 nn=18113
- 2!np=32740 qsp=37 rpq=6202 cbr=0 cng=476 gpc=4627 gps=546 nn=20889
- 3 np=23679 qsp=22 rpq=5044 cbr=1 cng=415 gpc=3403 gps=347 nn=14469
- 4!np=30714 qsp=4 rpq=5574 cbr=0 cng=528 gpc=3931 gps=639 nn=20042
- 5 np=28910 qsp=2 rpq=5246 cbr=0 cng=428 gpc=4105 gps=709 nn=18422
- 6!np=38648 qsp=5 rpq=7076 cbr=0 cng=840 gpc=4072 gps=961 nn=25699
- 7 np=37275 qsp=2 rpq=6873 cbr=0 cng=868 gpc=3416 gps=971 nn=25147
+ 0!np=26111 qsp=29 rpq=5386 cbr=1 cng=570 gpc=3674 gps=577 nn=15903 ndw=0
+ 1!np=28913 qsp=35 rpq=6097 cbr=1 cng=448 gpc=3700 gps=554 nn=18113 ndw=0
+ 2!np=32740 qsp=37 rpq=6202 cbr=0 cng=476 gpc=4627 gps=546 nn=20889 ndw=0
+ 3 np=23679 qsp=22 rpq=5044 cbr=1 cng=415 gpc=3403 gps=347 nn=14469 ndw=0
+ 4!np=30714 qsp=4 rpq=5574 cbr=0 cng=528 gpc=3931 gps=639 nn=20042 ndw=0
+ 5 np=28910 qsp=2 rpq=5246 cbr=0 cng=428 gpc=4105 gps=709 nn=18422 ndw=0
+ 6!np=38648 qsp=5 rpq=7076 cbr=0 cng=840 gpc=4072 gps=961 nn=25699 ndw=0
+ 7 np=37275 qsp=2 rpq=6873 cbr=0 cng=868 gpc=3416 gps=971 nn=25147 ndw=0
The fields are as follows:
@@ -432,6 +432,10 @@
o "gps" is the number of times that a new grace period had started,
but this CPU was not yet aware of it.
+o "ndw" is the number of times that a wakeup of an rcuo
+ callback-offload kthread had to be deferred in order to avoid
+ deadlock.
+
o "nn" is the number of times that this CPU needed nothing.
@@ -443,7 +447,7 @@
balk: nt=0 egt=6541 bt=0 nb=0 ny=126 nos=0
This information is output only for rcu_preempt. Each two-line entry
-corresponds to a leaf rcu_node strcuture. The fields are as follows:
+corresponds to a leaf rcu_node structure. The fields are as follows:
o "n:m" is the CPU-number range for the corresponding two-line
entry. In the sample output above, the first entry covers
diff --git a/Documentation/circular-buffers.txt b/Documentation/circular-buffers.txt
index 8117e5b..88951b1 100644
--- a/Documentation/circular-buffers.txt
+++ b/Documentation/circular-buffers.txt
@@ -160,6 +160,7 @@
spin_lock(&producer_lock);
unsigned long head = buffer->head;
+ /* The spin_unlock() and next spin_lock() provide needed ordering. */
unsigned long tail = ACCESS_ONCE(buffer->tail);
if (CIRC_SPACE(head, tail, buffer->size) >= 1) {
@@ -168,9 +169,8 @@
produce_item(item);
- smp_wmb(); /* commit the item before incrementing the head */
-
- buffer->head = (head + 1) & (buffer->size - 1);
+ smp_store_release(buffer->head,
+ (head + 1) & (buffer->size - 1));
/* wake_up() will make sure that the head is committed before
* waking anyone up */
@@ -183,9 +183,14 @@
before the head index makes it available to the consumer and then instructs the
CPU that the revised head index must be written before the consumer is woken.
-Note that wake_up() doesn't have to be the exact mechanism used, but whatever
-is used must guarantee a (write) memory barrier between the update of the head
-index and the change of state of the consumer, if a change of state occurs.
+Note that wake_up() does not guarantee any sort of barrier unless something
+is actually awakened. We therefore cannot rely on it for ordering. However,
+there is always one element of the array left empty. Therefore, the
+producer must produce two elements before it could possibly corrupt the
+element currently being read by the consumer. Therefore, the unlock-lock
+pair between consecutive invocations of the consumer provides the necessary
+ordering between the read of the index indicating that the consumer has
+vacated a given element and the write by the producer to that same element.
THE CONSUMER
@@ -195,21 +200,20 @@
spin_lock(&consumer_lock);
- unsigned long head = ACCESS_ONCE(buffer->head);
+ /* Read index before reading contents at that index. */
+ unsigned long head = smp_load_acquire(buffer->head);
unsigned long tail = buffer->tail;
if (CIRC_CNT(head, tail, buffer->size) >= 1) {
- /* read index before reading contents at that index */
- smp_read_barrier_depends();
/* extract one item from the buffer */
struct item *item = buffer[tail];
consume_item(item);
- smp_mb(); /* finish reading descriptor before incrementing tail */
-
- buffer->tail = (tail + 1) & (buffer->size - 1);
+ /* Finish reading descriptor before incrementing tail. */
+ smp_store_release(buffer->tail,
+ (tail + 1) & (buffer->size - 1));
}
spin_unlock(&consumer_lock);
@@ -218,12 +222,17 @@
the new item, and then it shall make sure the CPU has finished reading the item
before it writes the new tail pointer, which will erase the item.
-
-Note the use of ACCESS_ONCE() in both algorithms to read the opposition index.
-This prevents the compiler from discarding and reloading its cached value -
-which some compilers will do across smp_read_barrier_depends(). This isn't
-strictly needed if you can be sure that the opposition index will _only_ be
-used the once.
+Note the use of ACCESS_ONCE() and smp_load_acquire() to read the
+opposition index. This prevents the compiler from discarding and
+reloading its cached value - which some compilers will do across
+smp_read_barrier_depends(). This isn't strictly needed if you can
+be sure that the opposition index will _only_ be used the once.
+The smp_load_acquire() additionally forces the CPU to order against
+subsequent memory references. Similarly, smp_store_release() is used
+in both algorithms to write the thread's index. This documents the
+fact that we are writing to something that can be read concurrently,
+prevents the compiler from tearing the store, and enforces ordering
+against previous accesses.
===============
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 50680a5..9063e74 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -2625,7 +2625,6 @@
for RCU-preempt, and "s" for RCU-sched, and "N"
is the CPU number. This reduces OS jitter on the
offloaded CPUs, which can be useful for HPC and
-
real-time workloads. It can also improve energy
efficiency for asymmetric multiprocessors.
@@ -2641,8 +2640,8 @@
periodically wake up to do the polling.
rcutree.blimit= [KNL]
- Set maximum number of finished RCU callbacks to process
- in one batch.
+ Set maximum number of finished RCU callbacks to
+ process in one batch.
rcutree.rcu_fanout_leaf= [KNL]
Increase the number of CPUs assigned to each
@@ -2661,8 +2660,8 @@
value is one, and maximum value is HZ.
rcutree.qhimark= [KNL]
- Set threshold of queued
- RCU callbacks over which batch limiting is disabled.
+ Set threshold of queued RCU callbacks beyond which
+ batch limiting is disabled.
rcutree.qlowmark= [KNL]
Set threshold of queued RCU callbacks below which
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 4dd5ee2..a0b97c4 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1763,7 +1763,7 @@
}
if (all) {
- rcu_assign_pointer(bond->curr_active_slave, NULL);
+ RCU_INIT_POINTER(bond->curr_active_slave, NULL);
} else if (oldcurrent == slave) {
/*
* Note that we hold RTNL over this sequence, so there
diff --git a/include/linux/rculist.h b/include/linux/rculist.h
index 45a0a9e..dbaf990 100644
--- a/include/linux/rculist.h
+++ b/include/linux/rculist.h
@@ -55,8 +55,8 @@
next->prev = new;
}
#else
-extern void __list_add_rcu(struct list_head *new,
- struct list_head *prev, struct list_head *next);
+void __list_add_rcu(struct list_head *new,
+ struct list_head *prev, struct list_head *next);
#endif
/**
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 39cbb88..3e355c6 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -50,13 +50,13 @@
#endif /* #ifdef CONFIG_RCU_TORTURE_TEST */
#if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU)
-extern void rcutorture_record_test_transition(void);
-extern void rcutorture_record_progress(unsigned long vernum);
-extern void do_trace_rcu_torture_read(const char *rcutorturename,
- struct rcu_head *rhp,
- unsigned long secs,
- unsigned long c_old,
- unsigned long c);
+void rcutorture_record_test_transition(void);
+void rcutorture_record_progress(unsigned long vernum);
+void do_trace_rcu_torture_read(const char *rcutorturename,
+ struct rcu_head *rhp,
+ unsigned long secs,
+ unsigned long c_old,
+ unsigned long c);
#else
static inline void rcutorture_record_test_transition(void)
{
@@ -65,11 +65,11 @@
{
}
#ifdef CONFIG_RCU_TRACE
-extern void do_trace_rcu_torture_read(const char *rcutorturename,
- struct rcu_head *rhp,
- unsigned long secs,
- unsigned long c_old,
- unsigned long c);
+void do_trace_rcu_torture_read(const char *rcutorturename,
+ struct rcu_head *rhp,
+ unsigned long secs,
+ unsigned long c_old,
+ unsigned long c);
#else
#define do_trace_rcu_torture_read(rcutorturename, rhp, secs, c_old, c) \
do { } while (0)
@@ -118,8 +118,8 @@
* if CPU A and CPU B are the same CPU (but again only if the system has
* more than one CPU).
*/
-extern void call_rcu(struct rcu_head *head,
- void (*func)(struct rcu_head *head));
+void call_rcu(struct rcu_head *head,
+ void (*func)(struct rcu_head *head));
#else /* #ifdef CONFIG_PREEMPT_RCU */
@@ -149,8 +149,8 @@
* See the description of call_rcu() for more detailed information on
* memory ordering guarantees.
*/
-extern void call_rcu_bh(struct rcu_head *head,
- void (*func)(struct rcu_head *head));
+void call_rcu_bh(struct rcu_head *head,
+ void (*func)(struct rcu_head *head));
/**
* call_rcu_sched() - Queue an RCU for invocation after sched grace period.
@@ -171,16 +171,16 @@
* See the description of call_rcu() for more detailed information on
* memory ordering guarantees.
*/
-extern void call_rcu_sched(struct rcu_head *head,
- void (*func)(struct rcu_head *rcu));
+void call_rcu_sched(struct rcu_head *head,
+ void (*func)(struct rcu_head *rcu));
-extern void synchronize_sched(void);
+void synchronize_sched(void);
#ifdef CONFIG_PREEMPT_RCU
-extern void __rcu_read_lock(void);
-extern void __rcu_read_unlock(void);
-extern void rcu_read_unlock_special(struct task_struct *t);
+void __rcu_read_lock(void);
+void __rcu_read_unlock(void);
+void rcu_read_unlock_special(struct task_struct *t);
void synchronize_rcu(void);
/*
@@ -216,19 +216,19 @@
#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);
+void rcu_init(void);
+void rcu_sched_qs(int cpu);
+void rcu_bh_qs(int cpu);
+void rcu_check_callbacks(int cpu, int user);
struct notifier_block;
-extern void rcu_idle_enter(void);
-extern void rcu_idle_exit(void);
-extern void rcu_irq_enter(void);
-extern void rcu_irq_exit(void);
+void rcu_idle_enter(void);
+void rcu_idle_exit(void);
+void rcu_irq_enter(void);
+void rcu_irq_exit(void);
#ifdef CONFIG_RCU_USER_QS
-extern void rcu_user_enter(void);
-extern void rcu_user_exit(void);
+void rcu_user_enter(void);
+void rcu_user_exit(void);
#else
static inline void rcu_user_enter(void) { }
static inline void rcu_user_exit(void) { }
@@ -262,7 +262,7 @@
} while (0)
#if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) || defined(CONFIG_SMP)
-extern bool __rcu_is_watching(void);
+bool __rcu_is_watching(void);
#endif /* #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) || defined(CONFIG_SMP) */
/*
@@ -289,8 +289,8 @@
* initialization.
*/
#ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD
-extern void init_rcu_head_on_stack(struct rcu_head *head);
-extern void destroy_rcu_head_on_stack(struct rcu_head *head);
+void init_rcu_head_on_stack(struct rcu_head *head);
+void destroy_rcu_head_on_stack(struct rcu_head *head);
#else /* !CONFIG_DEBUG_OBJECTS_RCU_HEAD */
static inline void init_rcu_head_on_stack(struct rcu_head *head)
{
@@ -325,6 +325,7 @@
extern struct lockdep_map rcu_lock_map;
extern struct lockdep_map rcu_bh_lock_map;
extern struct lockdep_map rcu_sched_lock_map;
+extern struct lockdep_map rcu_callback_map;
extern int debug_lockdep_rcu_enabled(void);
/**
@@ -362,7 +363,7 @@
* rcu_read_lock_bh_held() is defined out of line to avoid #include-file
* hell.
*/
-extern int rcu_read_lock_bh_held(void);
+int rcu_read_lock_bh_held(void);
/**
* rcu_read_lock_sched_held() - might we be in RCU-sched read-side critical section?
@@ -448,7 +449,7 @@
#ifdef CONFIG_PROVE_RCU
-extern int rcu_my_thread_group_empty(void);
+int rcu_my_thread_group_empty(void);
/**
* rcu_lockdep_assert - emit lockdep splat if specified condition not met
@@ -548,10 +549,48 @@
smp_read_barrier_depends(); \
(_________p1); \
})
-#define __rcu_assign_pointer(p, v, space) \
+
+/**
+ * RCU_INITIALIZER() - statically initialize an RCU-protected global variable
+ * @v: The value to statically initialize with.
+ */
+#define RCU_INITIALIZER(v) (typeof(*(v)) __force __rcu *)(v)
+
+/**
+ * rcu_assign_pointer() - assign to RCU-protected pointer
+ * @p: pointer to assign to
+ * @v: value to assign (publish)
+ *
+ * Assigns the specified value to the specified RCU-protected
+ * pointer, ensuring that any concurrent RCU readers will see
+ * any prior initialization.
+ *
+ * Inserts memory barriers on architectures that require them
+ * (which is most of them), and also prevents the compiler from
+ * reordering the code that initializes the structure after the pointer
+ * assignment. More importantly, this call documents which pointers
+ * will be dereferenced by RCU read-side code.
+ *
+ * In some special cases, you may use RCU_INIT_POINTER() instead
+ * of rcu_assign_pointer(). RCU_INIT_POINTER() is a bit faster due
+ * to the fact that it does not constrain either the CPU or the compiler.
+ * That said, using RCU_INIT_POINTER() when you should have used
+ * rcu_assign_pointer() is a very bad thing that results in
+ * impossible-to-diagnose memory corruption. So please be careful.
+ * See the RCU_INIT_POINTER() comment header for details.
+ *
+ * Note that rcu_assign_pointer() evaluates each of its arguments only
+ * once, appearances notwithstanding. One of the "extra" evaluations
+ * is in typeof() and the other visible only to sparse (__CHECKER__),
+ * neither of which actually execute the argument. As with most cpp
+ * macros, this execute-arguments-only-once property is important, so
+ * please be careful when making changes to rcu_assign_pointer() and the
+ * other macros that it invokes.
+ */
+#define rcu_assign_pointer(p, v) \
do { \
smp_wmb(); \
- (p) = (typeof(*v) __force space *)(v); \
+ ACCESS_ONCE(p) = RCU_INITIALIZER(v); \
} while (0)
@@ -890,32 +929,6 @@
}
/**
- * rcu_assign_pointer() - assign to RCU-protected pointer
- * @p: pointer to assign to
- * @v: value to assign (publish)
- *
- * Assigns the specified value to the specified RCU-protected
- * pointer, ensuring that any concurrent RCU readers will see
- * any prior initialization.
- *
- * Inserts memory barriers on architectures that require them
- * (which is most of them), and also prevents the compiler from
- * reordering the code that initializes the structure after the pointer
- * assignment. More importantly, this call documents which pointers
- * will be dereferenced by RCU read-side code.
- *
- * In some special cases, you may use RCU_INIT_POINTER() instead
- * of rcu_assign_pointer(). RCU_INIT_POINTER() is a bit faster due
- * to the fact that it does not constrain either the CPU or the compiler.
- * That said, using RCU_INIT_POINTER() when you should have used
- * rcu_assign_pointer() is a very bad thing that results in
- * impossible-to-diagnose memory corruption. So please be careful.
- * See the RCU_INIT_POINTER() comment header for details.
- */
-#define rcu_assign_pointer(p, v) \
- __rcu_assign_pointer((p), (v), __rcu)
-
-/**
* RCU_INIT_POINTER() - initialize an RCU protected pointer
*
* Initialize an RCU-protected pointer in special cases where readers
@@ -949,7 +962,7 @@
*/
#define RCU_INIT_POINTER(p, v) \
do { \
- p = (typeof(*v) __force __rcu *)(v); \
+ p = RCU_INITIALIZER(v); \
} while (0)
/**
@@ -958,7 +971,7 @@
* GCC-style initialization for an RCU-protected pointer in a structure field.
*/
#define RCU_POINTER_INITIALIZER(p, v) \
- .p = (typeof(*v) __force __rcu *)(v)
+ .p = RCU_INITIALIZER(v)
/*
* Does the specified offset indicate that the corresponding rcu_head
@@ -1005,7 +1018,7 @@
__kfree_rcu(&((ptr)->rcu_head), offsetof(typeof(*(ptr)), rcu_head))
#ifdef CONFIG_RCU_NOCB_CPU
-extern bool rcu_is_nocb_cpu(int cpu);
+bool rcu_is_nocb_cpu(int cpu);
#else
static inline bool rcu_is_nocb_cpu(int cpu) { return false; }
#endif /* #else #ifdef CONFIG_RCU_NOCB_CPU */
@@ -1013,8 +1026,8 @@
/* Only for use by adaptive-ticks code. */
#ifdef CONFIG_NO_HZ_FULL_SYSIDLE
-extern bool rcu_sys_is_idle(void);
-extern void rcu_sysidle_force_exit(void);
+bool rcu_sys_is_idle(void);
+void rcu_sysidle_force_exit(void);
#else /* #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */
static inline bool rcu_sys_is_idle(void)
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index 09ebcbe..6f01771 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -125,7 +125,7 @@
#ifdef CONFIG_DEBUG_LOCK_ALLOC
extern int rcu_scheduler_active __read_mostly;
-extern void rcu_scheduler_starting(void);
+void rcu_scheduler_starting(void);
#else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
static inline void rcu_scheduler_starting(void)
{
diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
index 4b9c815..72137ee 100644
--- a/include/linux/rcutree.h
+++ b/include/linux/rcutree.h
@@ -30,9 +30,9 @@
#ifndef __LINUX_RCUTREE_H
#define __LINUX_RCUTREE_H
-extern void rcu_note_context_switch(int cpu);
-extern int rcu_needs_cpu(int cpu, unsigned long *delta_jiffies);
-extern void rcu_cpu_stall_reset(void);
+void rcu_note_context_switch(int cpu);
+int rcu_needs_cpu(int cpu, unsigned long *delta_jiffies);
+void rcu_cpu_stall_reset(void);
/*
* Note a virtualization-based context switch. This is simply a
@@ -44,9 +44,9 @@
rcu_note_context_switch(cpu);
}
-extern void synchronize_rcu_bh(void);
-extern void synchronize_sched_expedited(void);
-extern void synchronize_rcu_expedited(void);
+void synchronize_rcu_bh(void);
+void synchronize_sched_expedited(void);
+void synchronize_rcu_expedited(void);
void kfree_call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu));
@@ -71,25 +71,25 @@
synchronize_sched_expedited();
}
-extern void rcu_barrier(void);
-extern void rcu_barrier_bh(void);
-extern void rcu_barrier_sched(void);
+void rcu_barrier(void);
+void rcu_barrier_bh(void);
+void rcu_barrier_sched(void);
extern unsigned long rcutorture_testseq;
extern unsigned long rcutorture_vernum;
-extern long rcu_batches_completed(void);
-extern long rcu_batches_completed_bh(void);
-extern long rcu_batches_completed_sched(void);
+long rcu_batches_completed(void);
+long rcu_batches_completed_bh(void);
+long rcu_batches_completed_sched(void);
-extern void rcu_force_quiescent_state(void);
-extern void rcu_bh_force_quiescent_state(void);
-extern void rcu_sched_force_quiescent_state(void);
+void rcu_force_quiescent_state(void);
+void rcu_bh_force_quiescent_state(void);
+void rcu_sched_force_quiescent_state(void);
-extern void exit_rcu(void);
+void exit_rcu(void);
-extern void rcu_scheduler_starting(void);
+void rcu_scheduler_starting(void);
extern int rcu_scheduler_active __read_mostly;
-extern bool rcu_is_watching(void);
+bool rcu_is_watching(void);
#endif /* __LINUX_RCUTREE_H */
diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h
index 7859a0a..79c3877 100644
--- a/kernel/rcu/rcu.h
+++ b/kernel/rcu/rcu.h
@@ -96,19 +96,22 @@
}
#endif /* #else !CONFIG_DEBUG_OBJECTS_RCU_HEAD */
-extern void kfree(const void *);
+void kfree(const void *);
static inline bool __rcu_reclaim(const char *rn, struct rcu_head *head)
{
unsigned long offset = (unsigned long)head->func;
+ rcu_lock_acquire(&rcu_callback_map);
if (__is_kfree_rcu_offset(offset)) {
RCU_TRACE(trace_rcu_invoke_kfree_callback(rn, head, offset));
kfree((void *)head - offset);
+ rcu_lock_release(&rcu_callback_map);
return 1;
} else {
RCU_TRACE(trace_rcu_invoke_callback(rn, head));
head->func(head);
+ rcu_lock_release(&rcu_callback_map);
return 0;
}
}
diff --git a/kernel/rcu/srcu.c b/kernel/rcu/srcu.c
index 01d5ccb..3318d82 100644
--- a/kernel/rcu/srcu.c
+++ b/kernel/rcu/srcu.c
@@ -363,6 +363,29 @@
/*
* Enqueue an SRCU callback on the specified srcu_struct structure,
* initiating grace-period processing if it is not already running.
+ *
+ * Note that all CPUs must agree that the grace period extended beyond
+ * all pre-existing SRCU read-side critical section. On systems with
+ * more than one CPU, this means that when "func()" is invoked, each CPU
+ * is guaranteed to have executed a full memory barrier since the end of
+ * its last corresponding SRCU read-side critical section whose beginning
+ * preceded the call to call_rcu(). It also means that each CPU executing
+ * an SRCU read-side critical section that continues beyond the start of
+ * "func()" must have executed a memory barrier after the call_rcu()
+ * but before the beginning of that SRCU read-side critical section.
+ * Note that these guarantees include CPUs that are offline, idle, or
+ * executing in user mode, as well as CPUs that are executing in the kernel.
+ *
+ * Furthermore, if CPU A invoked call_rcu() and CPU B invoked the
+ * resulting SRCU callback function "func()", then both CPU A and CPU
+ * B are guaranteed to execute a full memory barrier during the time
+ * interval between the call to call_rcu() and the invocation of "func()".
+ * This guarantee applies even if CPU A and CPU B are the same CPU (but
+ * again only if the system has more than one CPU).
+ *
+ * Of course, these guarantees apply only for invocations of call_srcu(),
+ * srcu_read_lock(), and srcu_read_unlock() that are all passed the same
+ * srcu_struct structure.
*/
void call_srcu(struct srcu_struct *sp, struct rcu_head *head,
void (*func)(struct rcu_head *head))
@@ -459,7 +482,30 @@
* Note that it is illegal to call synchronize_srcu() from the corresponding
* SRCU read-side critical section; doing so will result in deadlock.
* However, it is perfectly legal to call synchronize_srcu() on one
- * srcu_struct from some other srcu_struct's read-side critical section.
+ * srcu_struct from some other srcu_struct's read-side critical section,
+ * as long as the resulting graph of srcu_structs is acyclic.
+ *
+ * There are memory-ordering constraints implied by synchronize_srcu().
+ * On systems with more than one CPU, when synchronize_srcu() returns,
+ * each CPU is guaranteed to have executed a full memory barrier since
+ * the end of its last corresponding SRCU-sched read-side critical section
+ * whose beginning preceded the call to synchronize_srcu(). In addition,
+ * each CPU having an SRCU read-side critical section that extends beyond
+ * the return from synchronize_srcu() is guaranteed to have executed a
+ * full memory barrier after the beginning of synchronize_srcu() and before
+ * the beginning of that SRCU read-side critical section. Note that these
+ * guarantees include CPUs that are offline, idle, or executing in user mode,
+ * as well as CPUs that are executing in the kernel.
+ *
+ * Furthermore, if CPU A invoked synchronize_srcu(), which returned
+ * to its caller on CPU B, then both CPU A and CPU B are guaranteed
+ * to have executed a full memory barrier during the execution of
+ * synchronize_srcu(). This guarantee applies even if CPU A and CPU B
+ * are the same CPU, but again only if the system has more than one CPU.
+ *
+ * Of course, these memory-ordering guarantees apply only when
+ * synchronize_srcu(), srcu_read_lock(), and srcu_read_unlock() are
+ * passed the same srcu_struct structure.
*/
void synchronize_srcu(struct srcu_struct *sp)
{
@@ -476,12 +522,8 @@
* Wait for an SRCU grace period to elapse, but be more aggressive about
* spinning rather than blocking when waiting.
*
- * Note that it is also illegal to call synchronize_srcu_expedited()
- * from the corresponding SRCU read-side critical section;
- * doing so will result in deadlock. However, it is perfectly legal
- * to call synchronize_srcu_expedited() on one srcu_struct from some
- * other srcu_struct's read-side critical section, as long as
- * the resulting graph of srcu_structs is acyclic.
+ * Note that synchronize_srcu_expedited() has the same deadlock and
+ * memory-ordering properties as does synchronize_srcu().
*/
void synchronize_srcu_expedited(struct srcu_struct *sp)
{
@@ -491,6 +533,7 @@
/**
* srcu_barrier - Wait until all in-flight call_srcu() callbacks complete.
+ * @sp: srcu_struct on which to wait for in-flight callbacks.
*/
void srcu_barrier(struct srcu_struct *sp)
{
diff --git a/kernel/rcu/torture.c b/kernel/rcu/torture.c
index 3929cd4..732f8ae 100644
--- a/kernel/rcu/torture.c
+++ b/kernel/rcu/torture.c
@@ -139,8 +139,6 @@
#define VERBOSE_PRINTK_ERRSTRING(s) \
do { if (verbose) pr_alert("%s" TORTURE_FLAG "!!! " s "\n", torture_type); } while (0)
-static char printk_buf[4096];
-
static int nrealreaders;
static struct task_struct *writer_task;
static struct task_struct **fakewriter_tasks;
@@ -376,7 +374,7 @@
void (*call)(struct rcu_head *head, void (*func)(struct rcu_head *rcu));
void (*cb_barrier)(void);
void (*fqs)(void);
- int (*stats)(char *page);
+ void (*stats)(char *page);
int irq_capable;
int can_boost;
const char *name;
@@ -578,21 +576,19 @@
srcu_barrier(&srcu_ctl);
}
-static int srcu_torture_stats(char *page)
+static void srcu_torture_stats(char *page)
{
- int cnt = 0;
int cpu;
int idx = srcu_ctl.completed & 0x1;
- cnt += sprintf(&page[cnt], "%s%s per-CPU(idx=%d):",
+ page += sprintf(page, "%s%s per-CPU(idx=%d):",
torture_type, TORTURE_FLAG, idx);
for_each_possible_cpu(cpu) {
- cnt += sprintf(&page[cnt], " %d(%lu,%lu)", cpu,
+ page += sprintf(page, " %d(%lu,%lu)", cpu,
per_cpu_ptr(srcu_ctl.per_cpu_ref, cpu)->c[!idx],
per_cpu_ptr(srcu_ctl.per_cpu_ref, cpu)->c[idx]);
}
- cnt += sprintf(&page[cnt], "\n");
- return cnt;
+ sprintf(page, "\n");
}
static void srcu_torture_synchronize_expedited(void)
@@ -1052,10 +1048,9 @@
/*
* Create an RCU-torture statistics message in the specified buffer.
*/
-static int
+static void
rcu_torture_printk(char *page)
{
- int cnt = 0;
int cpu;
int i;
long pipesummary[RCU_TORTURE_PIPE_LEN + 1] = { 0 };
@@ -1071,8 +1066,8 @@
if (pipesummary[i] != 0)
break;
}
- cnt += sprintf(&page[cnt], "%s%s ", torture_type, TORTURE_FLAG);
- cnt += sprintf(&page[cnt],
+ page += sprintf(page, "%s%s ", torture_type, TORTURE_FLAG);
+ page += sprintf(page,
"rtc: %p ver: %lu tfle: %d rta: %d rtaf: %d rtf: %d ",
rcu_torture_current,
rcu_torture_current_version,
@@ -1080,53 +1075,52 @@
atomic_read(&n_rcu_torture_alloc),
atomic_read(&n_rcu_torture_alloc_fail),
atomic_read(&n_rcu_torture_free));
- cnt += sprintf(&page[cnt], "rtmbe: %d rtbke: %ld rtbre: %ld ",
+ page += sprintf(page, "rtmbe: %d rtbke: %ld rtbre: %ld ",
atomic_read(&n_rcu_torture_mberror),
n_rcu_torture_boost_ktrerror,
n_rcu_torture_boost_rterror);
- cnt += sprintf(&page[cnt], "rtbf: %ld rtb: %ld nt: %ld ",
+ page += sprintf(page, "rtbf: %ld rtb: %ld nt: %ld ",
n_rcu_torture_boost_failure,
n_rcu_torture_boosts,
n_rcu_torture_timers);
- cnt += sprintf(&page[cnt],
+ page += sprintf(page,
"onoff: %ld/%ld:%ld/%ld %d,%d:%d,%d %lu:%lu (HZ=%d) ",
n_online_successes, n_online_attempts,
n_offline_successes, n_offline_attempts,
min_online, max_online,
min_offline, max_offline,
sum_online, sum_offline, HZ);
- cnt += sprintf(&page[cnt], "barrier: %ld/%ld:%ld",
+ page += sprintf(page, "barrier: %ld/%ld:%ld",
n_barrier_successes,
n_barrier_attempts,
n_rcu_torture_barrier_error);
- cnt += sprintf(&page[cnt], "\n%s%s ", torture_type, TORTURE_FLAG);
+ page += sprintf(page, "\n%s%s ", torture_type, TORTURE_FLAG);
if (atomic_read(&n_rcu_torture_mberror) != 0 ||
n_rcu_torture_barrier_error != 0 ||
n_rcu_torture_boost_ktrerror != 0 ||
n_rcu_torture_boost_rterror != 0 ||
n_rcu_torture_boost_failure != 0 ||
i > 1) {
- cnt += sprintf(&page[cnt], "!!! ");
+ page += sprintf(page, "!!! ");
atomic_inc(&n_rcu_torture_error);
WARN_ON_ONCE(1);
}
- cnt += sprintf(&page[cnt], "Reader Pipe: ");
+ page += sprintf(page, "Reader Pipe: ");
for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++)
- cnt += sprintf(&page[cnt], " %ld", pipesummary[i]);
- cnt += sprintf(&page[cnt], "\n%s%s ", torture_type, TORTURE_FLAG);
- cnt += sprintf(&page[cnt], "Reader Batch: ");
+ page += sprintf(page, " %ld", pipesummary[i]);
+ page += sprintf(page, "\n%s%s ", torture_type, TORTURE_FLAG);
+ page += sprintf(page, "Reader Batch: ");
for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++)
- cnt += sprintf(&page[cnt], " %ld", batchsummary[i]);
- cnt += sprintf(&page[cnt], "\n%s%s ", torture_type, TORTURE_FLAG);
- cnt += sprintf(&page[cnt], "Free-Block Circulation: ");
+ page += sprintf(page, " %ld", batchsummary[i]);
+ page += sprintf(page, "\n%s%s ", torture_type, TORTURE_FLAG);
+ page += sprintf(page, "Free-Block Circulation: ");
for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++) {
- cnt += sprintf(&page[cnt], " %d",
+ page += sprintf(page, " %d",
atomic_read(&rcu_torture_wcount[i]));
}
- cnt += sprintf(&page[cnt], "\n");
+ page += sprintf(page, "\n");
if (cur_ops->stats)
- cnt += cur_ops->stats(&page[cnt]);
- return cnt;
+ cur_ops->stats(page);
}
/*
@@ -1140,10 +1134,17 @@
static void
rcu_torture_stats_print(void)
{
- int cnt;
+ int size = nr_cpu_ids * 200 + 8192;
+ char *buf;
- cnt = rcu_torture_printk(printk_buf);
- pr_alert("%s", printk_buf);
+ buf = kmalloc(size, GFP_KERNEL);
+ if (!buf) {
+ pr_err("rcu-torture: Out of memory, need: %d", size);
+ return;
+ }
+ rcu_torture_printk(buf);
+ pr_alert("%s", buf);
+ kfree(buf);
}
/*
@@ -1578,6 +1579,7 @@
{
long myid = (long)arg;
bool lastphase = 0;
+ bool newphase;
struct rcu_head rcu;
init_rcu_head_on_stack(&rcu);
@@ -1585,10 +1587,11 @@
set_user_nice(current, 19);
do {
wait_event(barrier_cbs_wq[myid],
- barrier_phase != lastphase ||
+ (newphase =
+ ACCESS_ONCE(barrier_phase)) != lastphase ||
kthread_should_stop() ||
fullstop != FULLSTOP_DONTSTOP);
- lastphase = barrier_phase;
+ lastphase = newphase;
smp_mb(); /* ensure barrier_phase load before ->call(). */
if (kthread_should_stop() || fullstop != FULLSTOP_DONTSTOP)
break;
@@ -1625,7 +1628,7 @@
if (kthread_should_stop() || fullstop != FULLSTOP_DONTSTOP)
break;
n_barrier_attempts++;
- cur_ops->cb_barrier();
+ cur_ops->cb_barrier(); /* Implies smp_mb() for wait_event(). */
if (atomic_read(&barrier_cbs_invoked) != n_barrier_cbs) {
n_rcu_torture_barrier_error++;
WARN_ON_ONCE(1);
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index dd08198..e37bd56 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -369,6 +369,9 @@
static void rcu_eqs_enter_common(struct rcu_dynticks *rdtp, long long oldval,
bool user)
{
+ struct rcu_state *rsp;
+ struct rcu_data *rdp;
+
trace_rcu_dyntick(TPS("Start"), oldval, rdtp->dynticks_nesting);
if (!user && !is_idle_task(current)) {
struct task_struct *idle __maybe_unused =
@@ -380,6 +383,10 @@
current->pid, current->comm,
idle->pid, idle->comm); /* must be idle task! */
}
+ for_each_rcu_flavor(rsp) {
+ rdp = this_cpu_ptr(rsp->rda);
+ do_nocb_deferred_wakeup(rdp);
+ }
rcu_prepare_for_idle(smp_processor_id());
/* CPUs seeing atomic_inc() must see prior RCU read-side crit sects */
smp_mb__before_atomic_inc(); /* See above. */
@@ -411,11 +418,12 @@
rdtp = this_cpu_ptr(&rcu_dynticks);
oldval = rdtp->dynticks_nesting;
WARN_ON_ONCE((oldval & DYNTICK_TASK_NEST_MASK) == 0);
- if ((oldval & DYNTICK_TASK_NEST_MASK) == DYNTICK_TASK_NEST_VALUE)
+ if ((oldval & DYNTICK_TASK_NEST_MASK) == DYNTICK_TASK_NEST_VALUE) {
rdtp->dynticks_nesting = 0;
- else
+ rcu_eqs_enter_common(rdtp, oldval, user);
+ } else {
rdtp->dynticks_nesting -= DYNTICK_TASK_NEST_VALUE;
- rcu_eqs_enter_common(rdtp, oldval, user);
+ }
}
/**
@@ -533,11 +541,12 @@
rdtp = this_cpu_ptr(&rcu_dynticks);
oldval = rdtp->dynticks_nesting;
WARN_ON_ONCE(oldval < 0);
- if (oldval & DYNTICK_TASK_NEST_MASK)
+ if (oldval & DYNTICK_TASK_NEST_MASK) {
rdtp->dynticks_nesting += DYNTICK_TASK_NEST_VALUE;
- else
+ } else {
rdtp->dynticks_nesting = DYNTICK_TASK_EXIT_IDLE;
- rcu_eqs_exit_common(rdtp, oldval, user);
+ rcu_eqs_exit_common(rdtp, oldval, user);
+ }
}
/**
@@ -716,7 +725,7 @@
bool ret;
if (in_nmi())
- return 1;
+ return true;
preempt_disable();
rdp = this_cpu_ptr(&rcu_sched_data);
rnp = rdp->mynode;
@@ -755,6 +764,12 @@
}
/*
+ * This function really isn't for public consumption, but RCU is special in
+ * that context switches can allow the state machine to make progress.
+ */
+extern void resched_cpu(int cpu);
+
+/*
* Return true if the specified CPU has passed through a quiescent
* state by virtue of being in or having passed through an dynticks
* idle state since the last call to dyntick_save_progress_counter()
@@ -812,16 +827,34 @@
*/
rcu_kick_nohz_cpu(rdp->cpu);
+ /*
+ * Alternatively, the CPU might be running in the kernel
+ * for an extended period of time without a quiescent state.
+ * Attempt to force the CPU through the scheduler to gain the
+ * needed quiescent state, but only if the grace period has gone
+ * on for an uncommonly long time. If there are many stuck CPUs,
+ * we will beat on the first one until it gets unstuck, then move
+ * to the next. Only do this for the primary flavor of RCU.
+ */
+ if (rdp->rsp == rcu_state &&
+ ULONG_CMP_GE(ACCESS_ONCE(jiffies), rdp->rsp->jiffies_resched)) {
+ rdp->rsp->jiffies_resched += 5;
+ resched_cpu(rdp->cpu);
+ }
+
return 0;
}
static void record_gp_stall_check_time(struct rcu_state *rsp)
{
unsigned long j = ACCESS_ONCE(jiffies);
+ unsigned long j1;
rsp->gp_start = j;
smp_wmb(); /* Record start time before stall time. */
- rsp->jiffies_stall = j + rcu_jiffies_till_stall_check();
+ j1 = rcu_jiffies_till_stall_check();
+ rsp->jiffies_stall = j + j1;
+ rsp->jiffies_resched = j + j1 / 2;
}
/*
@@ -1509,6 +1542,7 @@
rdp = this_cpu_ptr(rsp->rda);
if (rnp == rdp->mynode)
__note_gp_changes(rsp, rnp, rdp);
+ /* smp_mb() provided by prior unlock-lock pair. */
nocb += rcu_future_gp_cleanup(rsp, rnp);
raw_spin_unlock_irq(&rnp->lock);
cond_resched();
@@ -1553,6 +1587,7 @@
wait_event_interruptible(rsp->gp_wq,
ACCESS_ONCE(rsp->gp_flags) &
RCU_GP_FLAG_INIT);
+ /* Locking provides needed memory barrier. */
if (rcu_gp_init(rsp))
break;
cond_resched();
@@ -1582,6 +1617,7 @@
(!ACCESS_ONCE(rnp->qsmask) &&
!rcu_preempt_blocked_readers_cgp(rnp)),
j);
+ /* Locking provides needed memory barriers. */
/* If grace period done, leave loop. */
if (!ACCESS_ONCE(rnp->qsmask) &&
!rcu_preempt_blocked_readers_cgp(rnp))
@@ -1901,13 +1937,13 @@
* Adopt the RCU callbacks from the specified rcu_state structure's
* orphanage. The caller must hold the ->orphan_lock.
*/
-static void rcu_adopt_orphan_cbs(struct rcu_state *rsp)
+static void rcu_adopt_orphan_cbs(struct rcu_state *rsp, unsigned long flags)
{
int i;
struct rcu_data *rdp = __this_cpu_ptr(rsp->rda);
/* No-CBs CPUs are handled specially. */
- if (rcu_nocb_adopt_orphan_cbs(rsp, rdp))
+ if (rcu_nocb_adopt_orphan_cbs(rsp, rdp, flags))
return;
/* Do the accounting first. */
@@ -1986,7 +2022,7 @@
/* Orphan the dead CPU's callbacks, and adopt them if appropriate. */
rcu_send_cbs_to_orphanage(cpu, rsp, rnp, rdp);
- rcu_adopt_orphan_cbs(rsp);
+ rcu_adopt_orphan_cbs(rsp, flags);
/* Remove the outgoing CPU from the masks in the rcu_node hierarchy. */
mask = rdp->grpmask; /* rnp->grplo is constant. */
@@ -2303,6 +2339,9 @@
/* If there are callbacks ready, invoke them. */
if (cpu_has_callbacks_ready_to_invoke(rdp))
invoke_rcu_callbacks(rsp, rdp);
+
+ /* Do any needed deferred wakeups of rcuo kthreads. */
+ do_nocb_deferred_wakeup(rdp);
}
/*
@@ -2437,7 +2476,7 @@
if (cpu != -1)
rdp = per_cpu_ptr(rsp->rda, cpu);
- offline = !__call_rcu_nocb(rdp, head, lazy);
+ offline = !__call_rcu_nocb(rdp, head, lazy, flags);
WARN_ON_ONCE(offline);
/* _call_rcu() is illegal on offline CPU; leak the callback. */
local_irq_restore(flags);
@@ -2757,6 +2796,10 @@
/* Check for CPU stalls, if enabled. */
check_cpu_stall(rsp, rdp);
+ /* Is this CPU a NO_HZ_FULL CPU that should ignore RCU? */
+ if (rcu_nohz_full_cpu(rsp))
+ return 0;
+
/* Is the RCU core waiting for a quiescent state from this CPU? */
if (rcu_scheduler_fully_active &&
rdp->qs_pending && !rdp->passed_quiesce) {
@@ -2790,6 +2833,12 @@
return 1;
}
+ /* Does this CPU need a deferred NOCB wakeup? */
+ if (rcu_nocb_need_deferred_wakeup(rdp)) {
+ rdp->n_rp_nocb_defer_wakeup++;
+ return 1;
+ }
+
/* nothing to do */
rdp->n_rp_need_nothing++;
return 0;
@@ -3214,9 +3263,9 @@
{
int i;
- for (i = rcu_num_lvls - 1; i > 0; i--)
+ rsp->levelspread[rcu_num_lvls - 1] = rcu_fanout_leaf;
+ for (i = rcu_num_lvls - 2; i >= 0; i--)
rsp->levelspread[i] = CONFIG_RCU_FANOUT;
- rsp->levelspread[0] = rcu_fanout_leaf;
}
#else /* #ifdef CONFIG_RCU_FANOUT_EXACT */
static void __init rcu_init_levelspread(struct rcu_state *rsp)
@@ -3346,6 +3395,8 @@
if (rcu_fanout_leaf == CONFIG_RCU_FANOUT_LEAF &&
nr_cpu_ids == NR_CPUS)
return;
+ pr_info("RCU: Adjusting geometry for rcu_fanout_leaf=%d, nr_cpu_ids=%d\n",
+ rcu_fanout_leaf, nr_cpu_ids);
/*
* Compute number of nodes that can be handled an rcu_node tree
diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h
index 52be957..8c19873 100644
--- a/kernel/rcu/tree.h
+++ b/kernel/rcu/tree.h
@@ -317,6 +317,7 @@
unsigned long n_rp_cpu_needs_gp;
unsigned long n_rp_gp_completed;
unsigned long n_rp_gp_started;
+ unsigned long n_rp_nocb_defer_wakeup;
unsigned long n_rp_need_nothing;
/* 6) _rcu_barrier() and OOM callbacks. */
@@ -335,6 +336,7 @@
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. */
#endif /* #ifdef CONFIG_RCU_NOCB_CPU */
/* 8) RCU CPU stall data. */
@@ -453,6 +455,8 @@
/* but in jiffies. */
unsigned long jiffies_stall; /* Time at which to check */
/* for CPU stalls. */
+ unsigned long jiffies_resched; /* Time at which to resched */
+ /* a reluctant CPU. */
unsigned long gp_max; /* Maximum GP duration in */
/* jiffies. */
const char *name; /* Name of structure. */
@@ -548,9 +552,12 @@
static void rcu_nocb_gp_cleanup(struct rcu_state *rsp, struct rcu_node *rnp);
static void rcu_init_one_nocb(struct rcu_node *rnp);
static bool __call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *rhp,
- bool lazy);
+ bool lazy, unsigned long flags);
static bool rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp,
- struct rcu_data *rdp);
+ struct rcu_data *rdp,
+ unsigned long flags);
+static bool rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp);
+static void do_nocb_deferred_wakeup(struct rcu_data *rdp);
static void rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp);
static void rcu_spawn_nocb_kthreads(struct rcu_state *rsp);
static void rcu_kick_nohz_cpu(int cpu);
@@ -564,6 +571,7 @@
unsigned long maxj);
static void rcu_bind_gp_kthread(void);
static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp);
+static bool rcu_nohz_full_cpu(struct rcu_state *rsp);
#endif /* #ifndef RCU_TREE_NONCORE */
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index 6abb03d..e0885cb 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -361,10 +361,14 @@
special = t->rcu_read_unlock_special;
if (special & RCU_READ_UNLOCK_NEED_QS) {
rcu_preempt_qs(smp_processor_id());
+ if (!t->rcu_read_unlock_special) {
+ local_irq_restore(flags);
+ return;
+ }
}
- /* Hardware IRQ handlers cannot block. */
- if (in_irq() || in_serving_softirq()) {
+ /* Hardware IRQ handlers cannot block, complain if they get here. */
+ if (WARN_ON_ONCE(in_irq() || in_serving_softirq())) {
local_irq_restore(flags);
return;
}
@@ -779,8 +783,10 @@
}
if (rnp->parent == NULL) {
raw_spin_unlock_irqrestore(&rnp->lock, flags);
- if (wake)
+ if (wake) {
+ smp_mb(); /* EGP done before wake_up(). */
wake_up(&sync_rcu_preempt_exp_wq);
+ }
break;
}
mask = rnp->grpmask;
@@ -1852,6 +1858,7 @@
/* Wait for callbacks from earlier instance to complete. */
wait_event(oom_callback_wq, atomic_read(&oom_callback_count) == 0);
+ smp_mb(); /* Ensure callback reuse happens after callback invocation. */
/*
* Prevent premature wakeup: ensure that all increments happen
@@ -2101,7 +2108,8 @@
static void __call_rcu_nocb_enqueue(struct rcu_data *rdp,
struct rcu_head *rhp,
struct rcu_head **rhtp,
- int rhcount, int rhcount_lazy)
+ int rhcount, int rhcount_lazy,
+ unsigned long flags)
{
int len;
struct rcu_head **old_rhpp;
@@ -2122,9 +2130,16 @@
}
len = atomic_long_read(&rdp->nocb_q_count);
if (old_rhpp == &rdp->nocb_head) {
- wake_up(&rdp->nocb_wq); /* ... only if queue was empty ... */
+ if (!irqs_disabled_flags(flags)) {
+ wake_up(&rdp->nocb_wq); /* ... if queue was empty ... */
+ trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
+ TPS("WakeEmpty"));
+ } else {
+ rdp->nocb_defer_wakeup = true;
+ trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
+ TPS("WakeEmptyIsDeferred"));
+ }
rdp->qlen_last_fqs_check = 0;
- trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeEmpty"));
} else if (len > rdp->qlen_last_fqs_check + qhimark) {
wake_up_process(t); /* ... or if many callbacks queued. */
rdp->qlen_last_fqs_check = LONG_MAX / 2;
@@ -2145,12 +2160,12 @@
* "rcuo" kthread can find it.
*/
static bool __call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *rhp,
- bool lazy)
+ bool lazy, unsigned long flags)
{
if (!rcu_is_nocb_cpu(rdp->cpu))
return 0;
- __call_rcu_nocb_enqueue(rdp, rhp, &rhp->next, 1, lazy);
+ __call_rcu_nocb_enqueue(rdp, rhp, &rhp->next, 1, lazy, flags);
if (__is_kfree_rcu_offset((unsigned long)rhp->func))
trace_rcu_kfree_callback(rdp->rsp->name, rhp,
(unsigned long)rhp->func,
@@ -2168,7 +2183,8 @@
* not a no-CBs CPU.
*/
static bool __maybe_unused rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp,
- struct rcu_data *rdp)
+ struct rcu_data *rdp,
+ unsigned long flags)
{
long ql = rsp->qlen;
long qll = rsp->qlen_lazy;
@@ -2182,14 +2198,14 @@
/* First, enqueue the donelist, if any. This preserves CB ordering. */
if (rsp->orphan_donelist != NULL) {
__call_rcu_nocb_enqueue(rdp, rsp->orphan_donelist,
- rsp->orphan_donetail, ql, qll);
+ rsp->orphan_donetail, ql, qll, flags);
ql = qll = 0;
rsp->orphan_donelist = NULL;
rsp->orphan_donetail = &rsp->orphan_donelist;
}
if (rsp->orphan_nxtlist != NULL) {
__call_rcu_nocb_enqueue(rdp, rsp->orphan_nxtlist,
- rsp->orphan_nxttail, ql, qll);
+ rsp->orphan_nxttail, ql, qll, flags);
ql = qll = 0;
rsp->orphan_nxtlist = NULL;
rsp->orphan_nxttail = &rsp->orphan_nxtlist;
@@ -2250,6 +2266,7 @@
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. */
} else if (firsttime) {
firsttime = 0;
trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu,
@@ -2310,6 +2327,22 @@
return 0;
}
+/* Is a deferred wakeup of rcu_nocb_kthread() required? */
+static bool rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp)
+{
+ return ACCESS_ONCE(rdp->nocb_defer_wakeup);
+}
+
+/* Do a deferred wakeup of rcu_nocb_kthread(). */
+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);
+ trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("DeferredWakeEmpty"));
+}
+
/* Initialize per-rcu_data variables for no-CBs CPUs. */
static void __init rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp)
{
@@ -2365,13 +2398,14 @@
}
static bool __call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *rhp,
- bool lazy)
+ bool lazy, unsigned long flags)
{
return 0;
}
static bool __maybe_unused rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp,
- struct rcu_data *rdp)
+ struct rcu_data *rdp,
+ unsigned long flags)
{
return 0;
}
@@ -2380,6 +2414,15 @@
{
}
+static bool rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp)
+{
+ return false;
+}
+
+static void do_nocb_deferred_wakeup(struct rcu_data *rdp)
+{
+}
+
static void __init rcu_spawn_nocb_kthreads(struct rcu_state *rsp)
{
}
@@ -2829,3 +2872,23 @@
}
#endif /* #else #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */
+
+/*
+ * Is this CPU a NO_HZ_FULL CPU that should ignore RCU so that the
+ * grace-period kthread will do force_quiescent_state() processing?
+ * The idea is to avoid waking up RCU core processing on such a
+ * CPU unless the grace period has extended for too long.
+ *
+ * This code relies on the fact that all NO_HZ_FULL CPUs are also
+ * CONFIG_RCU_NOCB_CPUs.
+ */
+static bool rcu_nohz_full_cpu(struct rcu_state *rsp)
+{
+#ifdef CONFIG_NO_HZ_FULL
+ if (tick_nohz_full_cpu(smp_processor_id()) &&
+ (!rcu_gp_in_progress(rsp) ||
+ ULONG_CMP_LT(jiffies, ACCESS_ONCE(rsp->gp_start) + HZ)))
+ return 1;
+#endif /* #ifdef CONFIG_NO_HZ_FULL */
+ return 0;
+}
diff --git a/kernel/rcu/tree_trace.c b/kernel/rcu/tree_trace.c
index 3596797..4def475 100644
--- a/kernel/rcu/tree_trace.c
+++ b/kernel/rcu/tree_trace.c
@@ -364,9 +364,10 @@
rdp->n_rp_report_qs,
rdp->n_rp_cb_ready,
rdp->n_rp_cpu_needs_gp);
- seq_printf(m, "gpc=%ld gps=%ld nn=%ld\n",
+ seq_printf(m, "gpc=%ld gps=%ld nn=%ld ndw%ld\n",
rdp->n_rp_gp_completed,
rdp->n_rp_gp_started,
+ rdp->n_rp_nocb_defer_wakeup,
rdp->n_rp_need_nothing);
}
diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c
index 6cb3dff..802365c 100644
--- a/kernel/rcu/update.c
+++ b/kernel/rcu/update.c
@@ -128,6 +128,11 @@
STATIC_LOCKDEP_MAP_INIT("rcu_read_lock_sched", &rcu_sched_lock_key);
EXPORT_SYMBOL_GPL(rcu_sched_lock_map);
+static struct lock_class_key rcu_callback_key;
+struct lockdep_map rcu_callback_map =
+ STATIC_LOCKDEP_MAP_INIT("rcu_callback", &rcu_callback_key);
+EXPORT_SYMBOL_GPL(rcu_callback_map);
+
int notrace debug_lockdep_rcu_enabled(void)
{
return rcu_scheduler_active && debug_locks &&
