LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
@ 2008-11-06 6:47 Lai Jiangshan
2008-11-06 6:57 ` Ingo Molnar
2008-11-09 0:51 ` Paul E. McKenney
0 siblings, 2 replies; 16+ messages in thread
From: Lai Jiangshan @ 2008-11-06 6:47 UTC (permalink / raw)
To: Ingo Molnar
Cc: Paul E. McKenney, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List
this fix remove ugly macro, and increase readability for rcupdate codes
changed from v1:
use HAVE_SPECIAL_RCU_BH/SCHED instead of define duplicate version of
synchronize_sched().
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
---
diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h
index 5f89b62..68e84ff 100644
--- a/include/linux/rcuclassic.h
+++ b/include/linux/rcuclassic.h
@@ -32,6 +32,7 @@
#ifndef __LINUX_RCUCLASSIC_H
#define __LINUX_RCUCLASSIC_H
+#define HAVE_SPECIAL_RCU_BH
#include <linux/cache.h>
#include <linux/spinlock.h>
@@ -166,12 +167,7 @@ extern struct lockdep_map rcu_lock_map;
local_bh_enable(); \
} while (0)
-#define __synchronize_sched() synchronize_rcu()
-
-#define call_rcu_sched(head, func) call_rcu(head, func)
-
extern void __rcu_init(void);
-#define rcu_init_sched() do { } while (0)
extern void rcu_check_callbacks(int cpu, int user);
extern void rcu_restart_cpu(int cpu);
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 86f1f5e..7861bee 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -112,6 +112,7 @@ struct rcu_head {
*/
#define rcu_read_unlock() __rcu_read_unlock()
+#ifdef HAVE_SPECIAL_RCU_BH
/**
* rcu_read_lock_bh - mark the beginning of a softirq-only RCU critical section
*
@@ -125,13 +126,19 @@ struct rcu_head {
*/
#define rcu_read_lock_bh() __rcu_read_lock_bh()
-/*
+/**
* rcu_read_unlock_bh - marks the end of a softirq-only RCU critical section
*
* See rcu_read_lock_bh() for more information.
*/
#define rcu_read_unlock_bh() __rcu_read_unlock_bh()
+#else
+#define rcu_bh_qsctr_inc(cpu)
+#define rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
+#define rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
+#endif /* HAVE_SPECIAL_RCU_BH */
+
/**
* rcu_read_lock_sched - mark the beginning of a RCU-classic critical section
*
@@ -189,45 +196,6 @@ struct rcu_head {
(p) = (v); \
})
-/* Infrastructure to implement the synchronize_() primitives. */
-
-struct rcu_synchronize {
- struct rcu_head head;
- struct completion completion;
-};
-
-extern void wakeme_after_rcu(struct rcu_head *head);
-
-#define synchronize_rcu_xxx(name, func) \
-void name(void) \
-{ \
- struct rcu_synchronize rcu; \
- \
- init_completion(&rcu.completion); \
- /* Will wake me after RCU finished. */ \
- func(&rcu.head, wakeme_after_rcu); \
- /* Wait for it. */ \
- wait_for_completion(&rcu.completion); \
-}
-
-/**
- * synchronize_sched - block until all CPUs have exited any non-preemptive
- * kernel code sequences.
- *
- * This means that all preempt_disable code sequences, including NMI and
- * hardware-interrupt handlers, in progress on entry will have completed
- * before this primitive returns. However, this does not guarantee that
- * softirq handlers will have completed, since in some kernels, these
- * handlers can run in process context, and can block.
- *
- * This primitive provides the guarantees made by the (now removed)
- * synchronize_kernel() API. In contrast, synchronize_rcu() only
- * guarantees that rcu_read_lock() sections will have completed.
- * In "classic RCU", these two guarantees happen to be one and
- * the same, but can differ in realtime RCU implementations.
- */
-#define synchronize_sched() __synchronize_sched()
-
/**
* call_rcu - Queue an RCU callback for invocation after a grace period.
* @head: structure to be used for queueing the RCU updates.
@@ -242,6 +210,11 @@ void name(void) \
extern void call_rcu(struct rcu_head *head,
void (*func)(struct rcu_head *head));
+
+extern void synchronize_rcu(void);
+extern void rcu_barrier(void);
+
+#ifdef HAVE_SPECIAL_RCU_BH
/**
* call_rcu_bh - Queue an RCU for invocation after a quicker grace period.
* @head: structure to be used for queueing the RCU updates.
@@ -263,11 +236,46 @@ extern void call_rcu(struct rcu_head *head,
extern void call_rcu_bh(struct rcu_head *head,
void (*func)(struct rcu_head *head));
-/* Exported common interfaces */
-extern void synchronize_rcu(void);
-extern void rcu_barrier(void);
extern void rcu_barrier_bh(void);
+#else
+
+#define call_rcu_bh call_rcu
+#define rcu_barrier_bh rcu_barrier
+
+/*
+ * Return the number of RCU batches processed thus far. Useful for debug
+ * and statistic. The _bh variant is identifcal to straight RCU
+ */
+static inline long rcu_batches_completed_bh(void)
+{
+ return rcu_batches_completed();
+}
+#endif /* HAVE_SPECIAL_RCU_BH */
+
+#ifdef HAVE_SPECIAL_RCU_SCHED
+/**
+ * call_rcu_sched - Queue RCU callback for invocation after sched grace period.
+ * @head: structure to be used for queueing the RCU updates.
+ * @func: actual update function to be invoked after the grace period
+ *
+ * The update function will be invoked some time after a full
+ * synchronize_sched()-style grace period elapses, in other words after
+ * all currently executing preempt-disabled sections of code (including
+ * hardirq handlers, NMI handlers, and local_irq_save() blocks) have
+ * completed.
+ */
+extern void call_rcu_sched(struct rcu_head *head,
+ void (*func)(struct rcu_head *head));
+
+extern void synchronize_sched(void);
extern void rcu_barrier_sched(void);
+#else
+#define call_rcu_sched call_rcu
+#define synchronize_sched synchronize_rcu
+#define rcu_barrier_sched rcu_barrier
+
+#define rcu_init_sched() do { } while (0)
+#endif /* HAVE_SPECIAL_RCU_SCHED */
/* Internal to kernel */
extern void rcu_init(void);
diff --git a/include/linux/rcupreempt.h b/include/linux/rcupreempt.h
index 3e05c09..f30d1c8 100644
--- a/include/linux/rcupreempt.h
+++ b/include/linux/rcupreempt.h
@@ -32,6 +32,7 @@
#ifndef __LINUX_RCUPREEMPT_H
#define __LINUX_RCUPREEMPT_H
+#define HAVE_SPECIAL_RCU_SCHED
#include <linux/cache.h>
#include <linux/spinlock.h>
@@ -56,54 +57,18 @@ static inline void rcu_qsctr_inc(int cpu)
rdssp->sched_qs++;
}
-#define rcu_bh_qsctr_inc(cpu)
-
-/*
- * Someone might want to pass call_rcu_bh as a function pointer.
- * So this needs to just be a rename and not a macro function.
- * (no parentheses)
- */
-#define call_rcu_bh call_rcu
-
-/**
- * call_rcu_sched - Queue RCU callback for invocation after sched grace period.
- * @head: structure to be used for queueing the RCU updates.
- * @func: actual update function to be invoked after the grace period
- *
- * The update function will be invoked some time after a full
- * synchronize_sched()-style grace period elapses, in other words after
- * all currently executing preempt-disabled sections of code (including
- * hardirq handlers, NMI handlers, and local_irq_save() blocks) have
- * completed.
- */
-extern void call_rcu_sched(struct rcu_head *head,
- void (*func)(struct rcu_head *head));
extern void __rcu_read_lock(void) __acquires(RCU);
extern void __rcu_read_unlock(void) __releases(RCU);
extern int rcu_pending(int cpu);
extern int rcu_needs_cpu(int cpu);
-#define __rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
-#define __rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
-
-extern void __synchronize_sched(void);
-
extern void __rcu_init(void);
extern void rcu_init_sched(void);
extern void rcu_check_callbacks(int cpu, int user);
extern void rcu_restart_cpu(int cpu);
extern long rcu_batches_completed(void);
-/*
- * Return the number of RCU batches processed thus far. Useful for debug
- * and statistic. The _bh variant is identifcal to straight RCU
- */
-static inline long rcu_batches_completed_bh(void)
-{
- return rcu_batches_completed();
-}
-
#ifdef CONFIG_RCU_TRACE
struct rcupreempt_trace;
extern long *rcupreempt_flipctr(int cpu);
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index ad63af8..70cff32 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -56,11 +56,16 @@ static atomic_t rcu_barrier_cpu_count;
static DEFINE_MUTEX(rcu_barrier_mutex);
static struct completion rcu_barrier_completion;
+struct rcu_synchronize {
+ struct rcu_head head;
+ struct completion completion;
+};
+
/*
* Awaken the corresponding synchronize_rcu() instance now that a
* grace period has elapsed.
*/
-void wakeme_after_rcu(struct rcu_head *head)
+static void wakeme_after_rcu(struct rcu_head *head)
{
struct rcu_synchronize *rcu;
@@ -77,10 +82,48 @@ void wakeme_after_rcu(struct rcu_head *head)
* sections are delimited by rcu_read_lock() and rcu_read_unlock(),
* and may be nested.
*/
-void synchronize_rcu(void); /* Makes kernel-doc tools happy */
-synchronize_rcu_xxx(synchronize_rcu, call_rcu)
+void synchronize_rcu(void)
+{
+ struct rcu_synchronize rcu;
+
+ init_completion(&rcu.completion);
+ /* Will wake me after RCU finished. */
+ call_rcu(&rcu.head, wakeme_after_rcu);
+ /* Wait for it. */
+ wait_for_completion(&rcu.completion);
+}
EXPORT_SYMBOL_GPL(synchronize_rcu);
+#ifdef HAVE_SPECIAL_RCU_SCHED
+/**
+ * synchronize_sched - block until all CPUs have exited any non-preemptive
+ * kernel code sequences.
+ *
+ * This means that all preempt_disable code sequences, including NMI and
+ * hardware-interrupt handlers, in progress on entry will have completed
+ * before this primitive returns. However, this does not guarantee that
+ * softirq handlers will have completed, since in some kernels, these
+ * handlers can run in process context, and can block.
+ *
+ * This primitive provides the guarantees made by the (now removed)
+ * synchronize_kernel() API. In contrast, synchronize_rcu() only
+ * guarantees that rcu_read_lock() sections will have completed.
+ * In "classic RCU", these two guarantees happen to be one and
+ * the same, but can differ in realtime RCU implementations.
+ */
+void synchronize_sched(void)
+{
+ struct rcu_synchronize rcu;
+
+ init_completion(&rcu.completion);
+ /* Will wake me after RCU finished. */
+ call_rcu_sched(&rcu.head, wakeme_after_rcu);
+ /* Wait for it. */
+ wait_for_completion(&rcu.completion);
+}
+EXPORT_SYMBOL_GPL(synchronize_sched);
+#endif /* HAVE_SPECIAL_RCU_SCHED */
+
static void rcu_barrier_callback(struct rcu_head *notused)
{
if (atomic_dec_and_test(&rcu_barrier_cpu_count))
@@ -145,6 +188,7 @@ void rcu_barrier(void)
}
EXPORT_SYMBOL_GPL(rcu_barrier);
+#ifdef HAVE_SPECIAL_RCU_BH
/**
* rcu_barrier_bh - Wait until all in-flight call_rcu_bh() callbacks complete.
*/
@@ -153,7 +197,9 @@ void rcu_barrier_bh(void)
_rcu_barrier(RCU_BARRIER_BH);
}
EXPORT_SYMBOL_GPL(rcu_barrier_bh);
+#endif /* HAVE_SPECIAL_RCU_BH */
+#ifdef HAVE_SPECIAL_RCU_SCHED
/**
* rcu_barrier_sched - Wait for in-flight call_rcu_sched() callbacks.
*/
@@ -162,6 +208,7 @@ void rcu_barrier_sched(void)
_rcu_barrier(RCU_BARRIER_SCHED);
}
EXPORT_SYMBOL_GPL(rcu_barrier_sched);
+#endif /* HAVE_SPECIAL_RCU_SCHED */
void __init rcu_init(void)
{
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
index 59236e8..2068ad9 100644
--- a/kernel/rcupreempt.c
+++ b/kernel/rcupreempt.c
@@ -1161,15 +1161,6 @@ void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
EXPORT_SYMBOL_GPL(call_rcu_sched);
/*
- * Wait until all currently running preempt_disable() code segments
- * (including hardware-irq-disable segments) complete. Note that
- * in -rt this does -not- necessarily result in all currently executing
- * interrupt -handlers- having completed.
- */
-synchronize_rcu_xxx(__synchronize_sched, call_rcu_sched)
-EXPORT_SYMBOL_GPL(__synchronize_sched);
-
-/*
* kthread function that manages call_rcu_sched grace periods.
*/
static int rcu_sched_grace_period(void *arg)
^ permalink raw reply related [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-06 6:47 [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2 Lai Jiangshan
@ 2008-11-06 6:57 ` Ingo Molnar
2008-11-09 0:51 ` Paul E. McKenney
1 sibling, 0 replies; 16+ messages in thread
From: Ingo Molnar @ 2008-11-06 6:57 UTC (permalink / raw)
To: Lai Jiangshan
Cc: Paul E. McKenney, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List
* Lai Jiangshan <laijs@cn.fujitsu.com> wrote:
> this fix remove ugly macro, and increase readability for rcupdate codes
looks good to me, if Paul acks the concept too.
Two small details:
> +++ b/include/linux/rcuclassic.h
> @@ -32,6 +32,7 @@
>
> #ifndef __LINUX_RCUCLASSIC_H
> #define __LINUX_RCUCLASSIC_H
> +#define HAVE_SPECIAL_RCU_BH
please use def_bool to define CONFIG_RCU_HAVE_SPECIAL_RCU_BH
and:
> +#else
> +#define rcu_bh_qsctr_inc(cpu)
> +#define rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
> +#define rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
> +#endif /* HAVE_SPECIAL_RCU_BH */
use inline functions please. CPP defines should never be used in new
code. (use inlines instead of macros and enums/const instead of
constant #define's)
Ingo
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-06 6:47 [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2 Lai Jiangshan
2008-11-06 6:57 ` Ingo Molnar
@ 2008-11-09 0:51 ` Paul E. McKenney
2008-11-10 3:22 ` Lai Jiangshan
1 sibling, 1 reply; 16+ messages in thread
From: Paul E. McKenney @ 2008-11-09 0:51 UTC (permalink / raw)
To: Lai Jiangshan
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton, Linux Kernel Mailing List
On Thu, Nov 06, 2008 at 02:47:44PM +0800, Lai Jiangshan wrote:
>
> this fix remove ugly macro, and increase readability for rcupdate codes
>
> changed from v1:
> use HAVE_SPECIAL_RCU_BH/SCHED instead of define duplicate version of
> synchronize_sched().
Hello, Jiangshan!
I very much like getting rid of the ugly macro. I of course like the
kernel-doc fixes. ;-)
I am not yet convinced of the HAVE_SPECIAL_RCU_BH and
HAVE_SPECIAL_RCU_SCHED pieces. It is not clear to me that this approach
is simpler than the current approach of simply providing the appropriate
definitions for the symbols in the implementation-specific rcuxxx.h
file.
Am I missing something?
Thanx, Paul
> Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
> ---
> diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h
> index 5f89b62..68e84ff 100644
> --- a/include/linux/rcuclassic.h
> +++ b/include/linux/rcuclassic.h
> @@ -32,6 +32,7 @@
>
> #ifndef __LINUX_RCUCLASSIC_H
> #define __LINUX_RCUCLASSIC_H
> +#define HAVE_SPECIAL_RCU_BH
>
> #include <linux/cache.h>
> #include <linux/spinlock.h>
> @@ -166,12 +167,7 @@ extern struct lockdep_map rcu_lock_map;
> local_bh_enable(); \
> } while (0)
>
> -#define __synchronize_sched() synchronize_rcu()
> -
> -#define call_rcu_sched(head, func) call_rcu(head, func)
> -
> extern void __rcu_init(void);
> -#define rcu_init_sched() do { } while (0)
> extern void rcu_check_callbacks(int cpu, int user);
> extern void rcu_restart_cpu(int cpu);
>
> diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
> index 86f1f5e..7861bee 100644
> --- a/include/linux/rcupdate.h
> +++ b/include/linux/rcupdate.h
> @@ -112,6 +112,7 @@ struct rcu_head {
> */
> #define rcu_read_unlock() __rcu_read_unlock()
>
> +#ifdef HAVE_SPECIAL_RCU_BH
> /**
> * rcu_read_lock_bh - mark the beginning of a softirq-only RCU critical section
> *
> @@ -125,13 +126,19 @@ struct rcu_head {
> */
> #define rcu_read_lock_bh() __rcu_read_lock_bh()
>
> -/*
> +/**
> * rcu_read_unlock_bh - marks the end of a softirq-only RCU critical section
> *
> * See rcu_read_lock_bh() for more information.
> */
> #define rcu_read_unlock_bh() __rcu_read_unlock_bh()
>
> +#else
> +#define rcu_bh_qsctr_inc(cpu)
> +#define rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
> +#define rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
> +#endif /* HAVE_SPECIAL_RCU_BH */
> +
> /**
> * rcu_read_lock_sched - mark the beginning of a RCU-classic critical section
> *
> @@ -189,45 +196,6 @@ struct rcu_head {
> (p) = (v); \
> })
>
> -/* Infrastructure to implement the synchronize_() primitives. */
> -
> -struct rcu_synchronize {
> - struct rcu_head head;
> - struct completion completion;
> -};
> -
> -extern void wakeme_after_rcu(struct rcu_head *head);
> -
> -#define synchronize_rcu_xxx(name, func) \
> -void name(void) \
> -{ \
> - struct rcu_synchronize rcu; \
> - \
> - init_completion(&rcu.completion); \
> - /* Will wake me after RCU finished. */ \
> - func(&rcu.head, wakeme_after_rcu); \
> - /* Wait for it. */ \
> - wait_for_completion(&rcu.completion); \
> -}
> -
> -/**
> - * synchronize_sched - block until all CPUs have exited any non-preemptive
> - * kernel code sequences.
> - *
> - * This means that all preempt_disable code sequences, including NMI and
> - * hardware-interrupt handlers, in progress on entry will have completed
> - * before this primitive returns. However, this does not guarantee that
> - * softirq handlers will have completed, since in some kernels, these
> - * handlers can run in process context, and can block.
> - *
> - * This primitive provides the guarantees made by the (now removed)
> - * synchronize_kernel() API. In contrast, synchronize_rcu() only
> - * guarantees that rcu_read_lock() sections will have completed.
> - * In "classic RCU", these two guarantees happen to be one and
> - * the same, but can differ in realtime RCU implementations.
> - */
> -#define synchronize_sched() __synchronize_sched()
> -
> /**
> * call_rcu - Queue an RCU callback for invocation after a grace period.
> * @head: structure to be used for queueing the RCU updates.
> @@ -242,6 +210,11 @@ void name(void) \
> extern void call_rcu(struct rcu_head *head,
> void (*func)(struct rcu_head *head));
>
> +
> +extern void synchronize_rcu(void);
> +extern void rcu_barrier(void);
> +
> +#ifdef HAVE_SPECIAL_RCU_BH
> /**
> * call_rcu_bh - Queue an RCU for invocation after a quicker grace period.
> * @head: structure to be used for queueing the RCU updates.
> @@ -263,11 +236,46 @@ extern void call_rcu(struct rcu_head *head,
> extern void call_rcu_bh(struct rcu_head *head,
> void (*func)(struct rcu_head *head));
>
> -/* Exported common interfaces */
> -extern void synchronize_rcu(void);
> -extern void rcu_barrier(void);
> extern void rcu_barrier_bh(void);
> +#else
> +
> +#define call_rcu_bh call_rcu
> +#define rcu_barrier_bh rcu_barrier
> +
> +/*
> + * Return the number of RCU batches processed thus far. Useful for debug
> + * and statistic. The _bh variant is identifcal to straight RCU
> + */
> +static inline long rcu_batches_completed_bh(void)
> +{
> + return rcu_batches_completed();
> +}
> +#endif /* HAVE_SPECIAL_RCU_BH */
> +
> +#ifdef HAVE_SPECIAL_RCU_SCHED
> +/**
> + * call_rcu_sched - Queue RCU callback for invocation after sched grace period.
> + * @head: structure to be used for queueing the RCU updates.
> + * @func: actual update function to be invoked after the grace period
> + *
> + * The update function will be invoked some time after a full
> + * synchronize_sched()-style grace period elapses, in other words after
> + * all currently executing preempt-disabled sections of code (including
> + * hardirq handlers, NMI handlers, and local_irq_save() blocks) have
> + * completed.
> + */
> +extern void call_rcu_sched(struct rcu_head *head,
> + void (*func)(struct rcu_head *head));
> +
> +extern void synchronize_sched(void);
> extern void rcu_barrier_sched(void);
> +#else
> +#define call_rcu_sched call_rcu
> +#define synchronize_sched synchronize_rcu
> +#define rcu_barrier_sched rcu_barrier
> +
> +#define rcu_init_sched() do { } while (0)
> +#endif /* HAVE_SPECIAL_RCU_SCHED */
>
> /* Internal to kernel */
> extern void rcu_init(void);
> diff --git a/include/linux/rcupreempt.h b/include/linux/rcupreempt.h
> index 3e05c09..f30d1c8 100644
> --- a/include/linux/rcupreempt.h
> +++ b/include/linux/rcupreempt.h
> @@ -32,6 +32,7 @@
>
> #ifndef __LINUX_RCUPREEMPT_H
> #define __LINUX_RCUPREEMPT_H
> +#define HAVE_SPECIAL_RCU_SCHED
>
> #include <linux/cache.h>
> #include <linux/spinlock.h>
> @@ -56,54 +57,18 @@ static inline void rcu_qsctr_inc(int cpu)
>
> rdssp->sched_qs++;
> }
> -#define rcu_bh_qsctr_inc(cpu)
> -
> -/*
> - * Someone might want to pass call_rcu_bh as a function pointer.
> - * So this needs to just be a rename and not a macro function.
> - * (no parentheses)
> - */
> -#define call_rcu_bh call_rcu
> -
> -/**
> - * call_rcu_sched - Queue RCU callback for invocation after sched grace period.
> - * @head: structure to be used for queueing the RCU updates.
> - * @func: actual update function to be invoked after the grace period
> - *
> - * The update function will be invoked some time after a full
> - * synchronize_sched()-style grace period elapses, in other words after
> - * all currently executing preempt-disabled sections of code (including
> - * hardirq handlers, NMI handlers, and local_irq_save() blocks) have
> - * completed.
> - */
> -extern void call_rcu_sched(struct rcu_head *head,
> - void (*func)(struct rcu_head *head));
>
> extern void __rcu_read_lock(void) __acquires(RCU);
> extern void __rcu_read_unlock(void) __releases(RCU);
> extern int rcu_pending(int cpu);
> extern int rcu_needs_cpu(int cpu);
>
> -#define __rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
> -#define __rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
> -
> -extern void __synchronize_sched(void);
> -
> extern void __rcu_init(void);
> extern void rcu_init_sched(void);
> extern void rcu_check_callbacks(int cpu, int user);
> extern void rcu_restart_cpu(int cpu);
> extern long rcu_batches_completed(void);
>
> -/*
> - * Return the number of RCU batches processed thus far. Useful for debug
> - * and statistic. The _bh variant is identifcal to straight RCU
> - */
> -static inline long rcu_batches_completed_bh(void)
> -{
> - return rcu_batches_completed();
> -}
> -
> #ifdef CONFIG_RCU_TRACE
> struct rcupreempt_trace;
> extern long *rcupreempt_flipctr(int cpu);
> diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
> index ad63af8..70cff32 100644
> --- a/kernel/rcupdate.c
> +++ b/kernel/rcupdate.c
> @@ -56,11 +56,16 @@ static atomic_t rcu_barrier_cpu_count;
> static DEFINE_MUTEX(rcu_barrier_mutex);
> static struct completion rcu_barrier_completion;
>
> +struct rcu_synchronize {
> + struct rcu_head head;
> + struct completion completion;
> +};
> +
> /*
> * Awaken the corresponding synchronize_rcu() instance now that a
> * grace period has elapsed.
> */
> -void wakeme_after_rcu(struct rcu_head *head)
> +static void wakeme_after_rcu(struct rcu_head *head)
> {
> struct rcu_synchronize *rcu;
>
> @@ -77,10 +82,48 @@ void wakeme_after_rcu(struct rcu_head *head)
> * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
> * and may be nested.
> */
> -void synchronize_rcu(void); /* Makes kernel-doc tools happy */
> -synchronize_rcu_xxx(synchronize_rcu, call_rcu)
> +void synchronize_rcu(void)
> +{
> + struct rcu_synchronize rcu;
> +
> + init_completion(&rcu.completion);
> + /* Will wake me after RCU finished. */
> + call_rcu(&rcu.head, wakeme_after_rcu);
> + /* Wait for it. */
> + wait_for_completion(&rcu.completion);
> +}
> EXPORT_SYMBOL_GPL(synchronize_rcu);
>
> +#ifdef HAVE_SPECIAL_RCU_SCHED
> +/**
> + * synchronize_sched - block until all CPUs have exited any non-preemptive
> + * kernel code sequences.
> + *
> + * This means that all preempt_disable code sequences, including NMI and
> + * hardware-interrupt handlers, in progress on entry will have completed
> + * before this primitive returns. However, this does not guarantee that
> + * softirq handlers will have completed, since in some kernels, these
> + * handlers can run in process context, and can block.
> + *
> + * This primitive provides the guarantees made by the (now removed)
> + * synchronize_kernel() API. In contrast, synchronize_rcu() only
> + * guarantees that rcu_read_lock() sections will have completed.
> + * In "classic RCU", these two guarantees happen to be one and
> + * the same, but can differ in realtime RCU implementations.
> + */
> +void synchronize_sched(void)
> +{
> + struct rcu_synchronize rcu;
> +
> + init_completion(&rcu.completion);
> + /* Will wake me after RCU finished. */
> + call_rcu_sched(&rcu.head, wakeme_after_rcu);
> + /* Wait for it. */
> + wait_for_completion(&rcu.completion);
> +}
> +EXPORT_SYMBOL_GPL(synchronize_sched);
> +#endif /* HAVE_SPECIAL_RCU_SCHED */
> +
> static void rcu_barrier_callback(struct rcu_head *notused)
> {
> if (atomic_dec_and_test(&rcu_barrier_cpu_count))
> @@ -145,6 +188,7 @@ void rcu_barrier(void)
> }
> EXPORT_SYMBOL_GPL(rcu_barrier);
>
> +#ifdef HAVE_SPECIAL_RCU_BH
> /**
> * rcu_barrier_bh - Wait until all in-flight call_rcu_bh() callbacks complete.
> */
> @@ -153,7 +197,9 @@ void rcu_barrier_bh(void)
> _rcu_barrier(RCU_BARRIER_BH);
> }
> EXPORT_SYMBOL_GPL(rcu_barrier_bh);
> +#endif /* HAVE_SPECIAL_RCU_BH */
>
> +#ifdef HAVE_SPECIAL_RCU_SCHED
> /**
> * rcu_barrier_sched - Wait for in-flight call_rcu_sched() callbacks.
> */
> @@ -162,6 +208,7 @@ void rcu_barrier_sched(void)
> _rcu_barrier(RCU_BARRIER_SCHED);
> }
> EXPORT_SYMBOL_GPL(rcu_barrier_sched);
> +#endif /* HAVE_SPECIAL_RCU_SCHED */
>
> void __init rcu_init(void)
> {
> diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
> index 59236e8..2068ad9 100644
> --- a/kernel/rcupreempt.c
> +++ b/kernel/rcupreempt.c
> @@ -1161,15 +1161,6 @@ void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
> EXPORT_SYMBOL_GPL(call_rcu_sched);
>
> /*
> - * Wait until all currently running preempt_disable() code segments
> - * (including hardware-irq-disable segments) complete. Note that
> - * in -rt this does -not- necessarily result in all currently executing
> - * interrupt -handlers- having completed.
> - */
> -synchronize_rcu_xxx(__synchronize_sched, call_rcu_sched)
> -EXPORT_SYMBOL_GPL(__synchronize_sched);
> -
> -/*
> * kthread function that manages call_rcu_sched grace periods.
> */
> static int rcu_sched_grace_period(void *arg)
>
>
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-09 0:51 ` Paul E. McKenney
@ 2008-11-10 3:22 ` Lai Jiangshan
2008-11-10 18:45 ` Paul E. McKenney
0 siblings, 1 reply; 16+ messages in thread
From: Lai Jiangshan @ 2008-11-10 3:22 UTC (permalink / raw)
To: paulmck
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton, Linux Kernel Mailing List
Paul E. McKenney wrote:
> On Thu, Nov 06, 2008 at 02:47:44PM +0800, Lai Jiangshan wrote:
>> this fix remove ugly macro, and increase readability for rcupdate codes
>>
>> changed from v1:
>> use HAVE_SPECIAL_RCU_BH/SCHED instead of define duplicate version of
>> synchronize_sched().
>
> Hello, Jiangshan!
>
> I very much like getting rid of the ugly macro. I of course like the
> kernel-doc fixes. ;-)
>
> I am not yet convinced of the HAVE_SPECIAL_RCU_BH and
> HAVE_SPECIAL_RCU_SCHED pieces. It is not clear to me that this approach
> is simpler than the current approach of simply providing the appropriate
> definitions for the symbols in the implementation-specific rcuxxx.h
> file.
>
> Am I missing something?
>
> Thanx, Paul
>
I think:
RCU_BH is not required, we can used RCU instead. so HAVE_SPECIAL_RCU_BH
will help for implementation which has not RCU_BH.
HAVE_SPECIAL_RCU_SCHED is a little different, RCU and RCU_SCHED are both
required for the kernel. But I think, in an implementation,
if rcu_read_lock_sched() implies rcu_read_lock(), we may not need implement
RCU_SCHED too(sometimes we may implement RCU_SCHED for performance).
so HAVE_SPECIAL_RCU_SCHED will help.
Thanx, Lai.
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-10 3:22 ` Lai Jiangshan
@ 2008-11-10 18:45 ` Paul E. McKenney
2008-11-11 0:55 ` Lai Jiangshan
0 siblings, 1 reply; 16+ messages in thread
From: Paul E. McKenney @ 2008-11-10 18:45 UTC (permalink / raw)
To: Lai Jiangshan
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton, Linux Kernel Mailing List
On Mon, Nov 10, 2008 at 11:22:15AM +0800, Lai Jiangshan wrote:
> Paul E. McKenney wrote:
> > On Thu, Nov 06, 2008 at 02:47:44PM +0800, Lai Jiangshan wrote:
> >> this fix remove ugly macro, and increase readability for rcupdate codes
> >>
> >> changed from v1:
> >> use HAVE_SPECIAL_RCU_BH/SCHED instead of define duplicate version of
> >> synchronize_sched().
> >
> > Hello, Jiangshan!
> >
> > I very much like getting rid of the ugly macro. I of course like the
> > kernel-doc fixes. ;-)
> >
> > I am not yet convinced of the HAVE_SPECIAL_RCU_BH and
> > HAVE_SPECIAL_RCU_SCHED pieces. It is not clear to me that this approach
> > is simpler than the current approach of simply providing the appropriate
> > definitions for the symbols in the implementation-specific rcuxxx.h
> > file.
> >
> > Am I missing something?
> >
> > Thanx, Paul
> >
>
> I think:
>
> RCU_BH is not required, we can used RCU instead. so HAVE_SPECIAL_RCU_BH
> will help for implementation which has not RCU_BH.
>
> HAVE_SPECIAL_RCU_SCHED is a little different, RCU and RCU_SCHED are both
> required for the kernel. But I think, in an implementation,
> if rcu_read_lock_sched() implies rcu_read_lock(), we may not need implement
> RCU_SCHED too(sometimes we may implement RCU_SCHED for performance).
> so HAVE_SPECIAL_RCU_SCHED will help.
If I understand correctly, this is the "old way":
------------------------------------------------------------------------
rcupdate.h:
#define rcu_read_lock_bh() __rcu_read_lock_bh()
#define rcu_read_unlock_bh() __rcu_read_unlock_bh()
rcupreempt.h:
#define __rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
#define __rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
------------------------------------------------------------------------
And then this is the "new way":
------------------------------------------------------------------------
rcupdate.h:
#ifdef HAVE_SPECIAL_RCU_BH
#define rcu_read_lock_bh() __rcu_read_lock_bh()
#define rcu_read_unlock_bh() __rcu_read_unlock_bh()
#else
#define __rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
#define __rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
#endif /* HAVE_SPECIAL_RCU_BH */
rcupreempt.h:
#define HAVE_SPECIAL_RCU_BH
------------------------------------------------------------------------
If we had ten different RCU implementations, then the "new way" would save
a little bit of code. But the "old way" is a bit easier to figure out.
So I am in favor of getting rid of the ugly macro, and also in favor
of fixing the kerneldoc, but opposed to the HAVE_SPECIAL_RCU_BH and
HAVE_SPECIAL_RCU_SCHED changes.
Or am I missing something?
Thanx, Paul
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-10 18:45 ` Paul E. McKenney
@ 2008-11-11 0:55 ` Lai Jiangshan
2008-11-11 1:03 ` Paul E. McKenney
0 siblings, 1 reply; 16+ messages in thread
From: Lai Jiangshan @ 2008-11-11 0:55 UTC (permalink / raw)
To: paulmck
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton, Linux Kernel Mailing List
Paul E. McKenney wrote:
> On Mon, Nov 10, 2008 at 11:22:15AM +0800, Lai Jiangshan wrote:
>> Paul E. McKenney wrote:
>>> On Thu, Nov 06, 2008 at 02:47:44PM +0800, Lai Jiangshan wrote:
>>>> this fix remove ugly macro, and increase readability for rcupdate codes
>>>>
>>>> changed from v1:
>>>> use HAVE_SPECIAL_RCU_BH/SCHED instead of define duplicate version of
>>>> synchronize_sched().
>>> Hello, Jiangshan!
>>>
>>> I very much like getting rid of the ugly macro. I of course like the
>>> kernel-doc fixes. ;-)
>>>
>>> I am not yet convinced of the HAVE_SPECIAL_RCU_BH and
>>> HAVE_SPECIAL_RCU_SCHED pieces. It is not clear to me that this approach
>>> is simpler than the current approach of simply providing the appropriate
>>> definitions for the symbols in the implementation-specific rcuxxx.h
>>> file.
>>>
>>> Am I missing something?
>>>
>>> Thanx, Paul
>>>
>> I think:
>>
>> RCU_BH is not required, we can used RCU instead. so HAVE_SPECIAL_RCU_BH
>> will help for implementation which has not RCU_BH.
>>
>> HAVE_SPECIAL_RCU_SCHED is a little different, RCU and RCU_SCHED are both
>> required for the kernel. But I think, in an implementation,
>> if rcu_read_lock_sched() implies rcu_read_lock(), we may not need implement
>> RCU_SCHED too(sometimes we may implement RCU_SCHED for performance).
>> so HAVE_SPECIAL_RCU_SCHED will help.
>
> If I understand correctly, this is the "old way":
>
> ------------------------------------------------------------------------
>
> rcupdate.h:
>
> #define rcu_read_lock_bh() __rcu_read_lock_bh()
> #define rcu_read_unlock_bh() __rcu_read_unlock_bh()
>
> rcupreempt.h:
>
> #define __rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
> #define __rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
>
> ------------------------------------------------------------------------
>
> And then this is the "new way":
>
> ------------------------------------------------------------------------
>
> rcupdate.h:
>
> #ifdef HAVE_SPECIAL_RCU_BH
> #define rcu_read_lock_bh() __rcu_read_lock_bh()
> #define rcu_read_unlock_bh() __rcu_read_unlock_bh()
> #else
> #define __rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
> #define __rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
> #endif /* HAVE_SPECIAL_RCU_BH */
>
> rcupreempt.h:
>
> #define HAVE_SPECIAL_RCU_BH
>
> ------------------------------------------------------------------------
>
> If we had ten different RCU implementations, then the "new way" would save
> a little bit of code. But the "old way" is a bit easier to figure out.
>
> So I am in favor of getting rid of the ugly macro, and also in favor
> of fixing the kerneldoc, but opposed to the HAVE_SPECIAL_RCU_BH and
> HAVE_SPECIAL_RCU_SCHED changes.
I apprehended and agree with you. Thanx.
Lai.
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-11 0:55 ` Lai Jiangshan
@ 2008-11-11 1:03 ` Paul E. McKenney
2008-11-13 2:48 ` Lai Jiangshan
0 siblings, 1 reply; 16+ messages in thread
From: Paul E. McKenney @ 2008-11-11 1:03 UTC (permalink / raw)
To: Lai Jiangshan
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton, Linux Kernel Mailing List
On Tue, Nov 11, 2008 at 08:55:00AM +0800, Lai Jiangshan wrote:
> Paul E. McKenney wrote:
> > On Mon, Nov 10, 2008 at 11:22:15AM +0800, Lai Jiangshan wrote:
> >> Paul E. McKenney wrote:
> >>> On Thu, Nov 06, 2008 at 02:47:44PM +0800, Lai Jiangshan wrote:
> >>>> this fix remove ugly macro, and increase readability for rcupdate codes
> >>>>
> >>>> changed from v1:
> >>>> use HAVE_SPECIAL_RCU_BH/SCHED instead of define duplicate version of
> >>>> synchronize_sched().
> >>> Hello, Jiangshan!
> >>>
> >>> I very much like getting rid of the ugly macro. I of course like the
> >>> kernel-doc fixes. ;-)
> >>>
> >>> I am not yet convinced of the HAVE_SPECIAL_RCU_BH and
> >>> HAVE_SPECIAL_RCU_SCHED pieces. It is not clear to me that this approach
> >>> is simpler than the current approach of simply providing the appropriate
> >>> definitions for the symbols in the implementation-specific rcuxxx.h
> >>> file.
> >>>
> >>> Am I missing something?
> >>>
> >>> Thanx, Paul
> >>>
> >> I think:
> >>
> >> RCU_BH is not required, we can used RCU instead. so HAVE_SPECIAL_RCU_BH
> >> will help for implementation which has not RCU_BH.
> >>
> >> HAVE_SPECIAL_RCU_SCHED is a little different, RCU and RCU_SCHED are both
> >> required for the kernel. But I think, in an implementation,
> >> if rcu_read_lock_sched() implies rcu_read_lock(), we may not need implement
> >> RCU_SCHED too(sometimes we may implement RCU_SCHED for performance).
> >> so HAVE_SPECIAL_RCU_SCHED will help.
> >
> > If I understand correctly, this is the "old way":
> >
> > ------------------------------------------------------------------------
> >
> > rcupdate.h:
> >
> > #define rcu_read_lock_bh() __rcu_read_lock_bh()
> > #define rcu_read_unlock_bh() __rcu_read_unlock_bh()
> >
> > rcupreempt.h:
> >
> > #define __rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
> > #define __rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
> >
> > ------------------------------------------------------------------------
> >
> > And then this is the "new way":
> >
> > ------------------------------------------------------------------------
> >
> > rcupdate.h:
> >
> > #ifdef HAVE_SPECIAL_RCU_BH
> > #define rcu_read_lock_bh() __rcu_read_lock_bh()
> > #define rcu_read_unlock_bh() __rcu_read_unlock_bh()
> > #else
> > #define __rcu_read_lock_bh() { rcu_read_lock(); local_bh_disable(); }
> > #define __rcu_read_unlock_bh() { local_bh_enable(); rcu_read_unlock(); }
> > #endif /* HAVE_SPECIAL_RCU_BH */
> >
> > rcupreempt.h:
> >
> > #define HAVE_SPECIAL_RCU_BH
> >
> > ------------------------------------------------------------------------
> >
> > If we had ten different RCU implementations, then the "new way" would save
> > a little bit of code. But the "old way" is a bit easier to figure out.
> >
> > So I am in favor of getting rid of the ugly macro, and also in favor
> > of fixing the kerneldoc, but opposed to the HAVE_SPECIAL_RCU_BH and
> > HAVE_SPECIAL_RCU_SCHED changes.
>
> I apprehended and agree with you. Thanx.
Sounds good -- and thank you for your much-needed efforts to improve
the RCU implementation!
Thanx, Paul
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-11 1:03 ` Paul E. McKenney
@ 2008-11-13 2:48 ` Lai Jiangshan
2008-11-13 17:31 ` Paul E. McKenney
0 siblings, 1 reply; 16+ messages in thread
From: Lai Jiangshan @ 2008-11-13 2:48 UTC (permalink / raw)
To: paulmck
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton, Linux Kernel Mailing List
Hi, Paul,
Could you add a RCU document about unloadable modules for kernel?
Thanx, Lai.
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-13 2:48 ` Lai Jiangshan
@ 2008-11-13 17:31 ` Paul E. McKenney
2008-11-14 1:03 ` Lai Jiangshan
0 siblings, 1 reply; 16+ messages in thread
From: Paul E. McKenney @ 2008-11-13 17:31 UTC (permalink / raw)
To: Lai Jiangshan
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List, corbet
On Thu, Nov 13, 2008 at 10:48:33AM +0800, Lai Jiangshan wrote:
>
> Hi, Paul,
>
> Could you add a RCU document about unloadable modules for kernel?
You thinking in terms of an ASCII version of
http://lwn.net/Articles/217484/?
If so, please see attached patch and let me know what you think.
Being too lazy to convert the cartoon to ASCII graphics, I simply
left a URL to the .jpg on the LWN website. Thus we need an ack/nack
from Jon Corbet (CCed).
Of course, an alternative is to simply include the URL of the original
LWN article in 00-INDEX. Thoughts?
Thanx, Paul
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
00-INDEX | 2
rcubarrier.txt | 309 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 311 insertions(+)
diff -urpNa -X dontdiff linux-2.6.27/Documentation/RCU/00-INDEX linux-2.6.27-rcu_barrierdoc/Documentation/RCU/00-INDEX
--- linux-2.6.27/Documentation/RCU/00-INDEX 2008-10-09 15:13:53.000000000 -0700
+++ linux-2.6.27-rcu_barrierdoc/Documentation/RCU/00-INDEX 2008-11-13 08:46:17.000000000 -0800
@@ -12,6 +12,8 @@ rcuref.txt
- Reference-count design for elements of lists/arrays protected by RCU
rcu.txt
- RCU Concepts
+rcubarrier.txt
+ - Unloading modules that use RCU callbacks
RTFP.txt
- List of RCU papers (bibliography) going back to 1980.
torture.txt
diff -urpNa -X dontdiff linux-2.6.27/Documentation/RCU/rcubarrier.txt linux-2.6.27-rcu_barrierdoc/Documentation/RCU/rcubarrier.txt
--- linux-2.6.27/Documentation/RCU/rcubarrier.txt 1969-12-31 16:00:00.000000000 -0800
+++ linux-2.6.27-rcu_barrierdoc/Documentation/RCU/rcubarrier.txt 2008-11-13 09:26:16.000000000 -0800
@@ -0,0 +1,309 @@
+RCU and Unloadable Modules
+
+[Originally published in LWN Jan. 14, 2007: http://lwn.net/Articles/217484/]
+
+RCU (read-copy update) is a synchronization mechanism that can be thought
+of as a replacement for read-writer locking (among other things), but with
+very low-overhead readers that are immune to deadlock, priority inversion,
+and unbounded latency. RCU read-side critical sections are delimited
+by rcu_read_lock() and rcu_read_unlock(), which, in non-CONFIG_PREEMPT
+kernels, generate no code whatsoever.
+
+This means that RCU writers are unaware of the presence of concurrent
+readers, so that RCU updates to shared data must be undertaken quite
+carefully, leaving an old version of the data structure in place until all
+pre-existing readers have finished. These old versions are needed because
+such readers might hold a reference to them. RCU updates can therefore be
+rather expensive, and RCU is thus best suited for read-mostly situations.
+
+How can an RCU writer possibly determine when all readers are finished,
+given that readers might well leave absolutely no trace of their
+presence? There is a synchronize_rcu() primitive that blocks until all
+pre-existing readers have completed. An updater wishing to delete an
+element p from a linked list might do the following, while holding an
+appropriate lock, of course:
+
+ list_del_rcu(p);
+ synchronize_rcu();
+ kfree(p);
+
+But the above code cannot be used in IRQ context -- the call_rcu()
+primitive must be used instead. This primitive takes a pointer to an
+rcu_head struct placed within the RCU-protected data structure and
+another pointer to a function that may be invoked later to free that
+structure. Code to delete an element p from the linked list from IRQ
+context might then be as follows:
+
+ list_del_rcu(p);
+ call_rcu(&p->rcu, p_callback);
+
+Since call_rcu() never blocks, this code can safely be used from within
+IRQ context. The function p_callback() might be defined as follows:
+
+ static void p_callback(struct rcu_head *rp)
+ {
+ struct pstruct *p = container_of(rp, struct pstruct, rcu);
+
+ kfree(p);
+ }
+
+
+Unloading Modules That Use call_rcu()
+
+But what if p_callback is defined in an unloadable module?
+
+If we unload the module while some RCU callbacks are pending,
+the CPUs executing these callbacks are going to be severely
+disappointed when they are later invoked, as fancifully depicted at
+http://lwn.net/images/ns/kernel/rcu-drop.jpg.
+
+We could try placing a synchronize_rcu() in the module-exit code path,
+but this is not sufficient. Although synchronize_rcu() does wait for a
+grace period to elapse, it does not wait for the callbacks to complete.
+
+One might be tempted to try several back-to-back synchronize_rcu()
+calls, but this is still not guaranteed to work. If there is a very
+heavy RCU-callback load, then some of the callbacks might be deferred
+in order to allow other processing to proceed. Such deferral is required
+in realtime kernels in order to avoid excessive scheduling latencies.
+
+
+rcu_barrier()
+
+We instead need the rcu_barrier() primitive. This primitive is similar
+to synchronize_rcu(), but instead of waiting solely for a grace
+period to elapse, it also waits for all outstanding RCU callbacks to
+complete. Pseudo-code using rcu_barrier() is as follows:
+
+ 1. Prevent any new RCU callbacks from being posted.
+ 2. Execute rcu_barrier().
+ 3. Allow the module to be unloaded.
+
+Quick Quiz #1: Why is there no srcu_barrier()?
+
+Quick Quiz #2: Why is there no rcu_barrier_bh()?
+
+The rcutorture module makes use of rcu_barrier in its exit function
+as follows:
+
+ 1 static void
+ 2 rcu_torture_cleanup(void)
+ 3 {
+ 4 int i;
+ 5
+ 6 fullstop = 1;
+ 7 if (shuffler_task != NULL) {
+ 8 VERBOSE_PRINTK_STRING("Stopping rcu_torture_shuffle task");
+ 9 kthread_stop(shuffler_task);
+10 }
+11 shuffler_task = NULL;
+12
+13 if (writer_task != NULL) {
+14 VERBOSE_PRINTK_STRING("Stopping rcu_torture_writer task");
+15 kthread_stop(writer_task);
+16 }
+17 writer_task = NULL;
+18
+19 if (reader_tasks != NULL) {
+20 for (i = 0; i < nrealreaders; i++) {
+21 if (reader_tasks[i] != NULL) {
+22 VERBOSE_PRINTK_STRING(
+23 "Stopping rcu_torture_reader task");
+24 kthread_stop(reader_tasks[i]);
+25 }
+26 reader_tasks[i] = NULL;
+27 }
+28 kfree(reader_tasks);
+29 reader_tasks = NULL;
+30 }
+31 rcu_torture_current = NULL;
+32
+33 if (fakewriter_tasks != NULL) {
+34 for (i = 0; i < nfakewriters; i++) {
+35 if (fakewriter_tasks[i] != NULL) {
+36 VERBOSE_PRINTK_STRING(
+37 "Stopping rcu_torture_fakewriter task");
+38 kthread_stop(fakewriter_tasks[i]);
+39 }
+40 fakewriter_tasks[i] = NULL;
+41 }
+42 kfree(fakewriter_tasks);
+43 fakewriter_tasks = NULL;
+44 }
+45
+46 if (stats_task != NULL) {
+47 VERBOSE_PRINTK_STRING("Stopping rcu_torture_stats task");
+48 kthread_stop(stats_task);
+49 }
+50 stats_task = NULL;
+51
+52 /* Wait for all RCU callbacks to fire. */
+53 rcu_barrier();
+54
+55 rcu_torture_stats_print(); /* -After- the stats thread is stopped! */
+56
+57 if (cur_ops->cleanup != NULL)
+58 cur_ops->cleanup();
+59 if (atomic_read(&n_rcu_torture_error))
+60 rcu_torture_print_module_parms("End of test: FAILURE");
+61 else
+62 rcu_torture_print_module_parms("End of test: SUCCESS");
+63 }
+
+Line 6 sets a global variable that prevents any RCU callbacks from
+re-posting themselves. This will not be necessary in most cases, since
+RCU callbacks rarely include calls to call_rcu(). However, the rcutorture
+module is an exception to this rule, and therefore needs to set this
+global variable.
+
+Lines 7-50 stop all the kernel tasks associated with the rcutorture
+module. Therefore, once execution reaches line 53, no more rcutorture
+RCU callbacks will be posted. The rcu_barrier() call on line 53 waits
+for any pre-existing callbacks to complete.
+
+Then lines 55-62 print status and do operation-specific cleanup, and
+then return, permitting the module-unload operation to be completed.
+
+Quick Quiz #3: Is there any other situation where rcu_barrier() might
+ be required?
+
+Your module might have additional complications. For example, if your
+module invokes call_rcu() from timers, you will need to first cancel all
+the timers, and only then invoke rcu_barrier() to wait for any remaining
+RCU callbacks to complete.
+
+
+Implementing rcu_barrier()
+
+Dipankar Sarma's implementation of rcu_barrier() makes use of the fact
+that RCU callbacks are never reordered once queued on one of the per-CPU
+queues. His implementation queues an RCU callback on each of the per-CPU
+callback queues, and then waits until they have all started executing, at
+which point, all earlier RCU callbacks are guaranteed to have completed.
+
+The code for rcu_barrier() is as follows:
+
+ 1 void rcu_barrier(void)
+ 2 {
+ 3 BUG_ON(in_interrupt());
+ 4 /* Take cpucontrol mutex to protect against CPU hotplug */
+ 5 mutex_lock(&rcu_barrier_mutex);
+ 6 init_completion(&rcu_barrier_completion);
+ 7 atomic_set(&rcu_barrier_cpu_count, 0);
+ 8 on_each_cpu(rcu_barrier_func, NULL, 0, 1);
+ 9 wait_for_completion(&rcu_barrier_completion);
+10 mutex_unlock(&rcu_barrier_mutex);
+11 }
+
+Line 3 verifies that the caller is in process context, and lines 5 and 10
+use rcu_barrier_mutex to ensure that only one rcu_barrier() is using the
+global completion and counters at a time, which are initialized on lines
+6 and 7. Line 8 causes each CPU to invoke rcu_barrier_func(), which is
+shown below. Note that the final "1" in on_each_cpu()'s argument list
+ensures that all the calls to rcu_barrier_func() will have completed
+before on_each_cpu() returns. Line 9 then waits for the completion.
+
+The rcu_barrier_func() runs on each CPU, where it invokes call_rcu()
+to post an RCU callback, as follows:
+
+ 1 static void rcu_barrier_func(void *notused)
+ 2 {
+ 3 int cpu = smp_processor_id();
+ 4 struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
+ 5 struct rcu_head *head;
+ 6
+ 7 head = &rdp->barrier;
+ 8 atomic_inc(&rcu_barrier_cpu_count);
+ 9 call_rcu(head, rcu_barrier_callback);
+10 }
+
+Lines 3 and 4 locate RCU's internal per-CPU rcu_data structure,
+which contains the struct rcu_head that needed for the later call to
+call_rcu(). Line 7 picks up a pointer to this struct rcu_head, and line
+8 increments a global counter. This counter will later be decremented
+by the callback. Line 9 then registers the rcu_barrier_callback() on
+the current CPU's queue.
+
+The rcu_barrier_callback() function simply atomically decrements the
+rcu_barrier_cpu_count variable and finalizes the completion when it
+reaches zero, as follows:
+
+ 1 static void rcu_barrier_callback(struct rcu_head *notused)
+ 2 {
+ 3 if (atomic_dec_and_test(&rcu_barrier_cpu_count))
+ 4 complete(&rcu_barrier_completion);
+ 5 }
+
+Quick Quiz #4: What happens if CPU 0's rcu_barrier_func() executes
+ immediately (thus incrementing rcu_barrier_cpu_count to the
+ value one), but the other CPU's rcu_barrier_func() invocations
+ are delayed for a full grace period? Couldn't this result in
+ rcu_barrier() returning prematurely?
+
+
+rcu_barrier() Summary
+
+The rcu_barrier() primitive has seen relatively little use, since most
+code using RCU is in the core kernel rather than in modules. However, if
+you are using RCU from an unloadable module, you need to use rcu_barrier()
+so that your module may be safely unloaded.
+
+
+Answers to Quick Quizzes
+
+Quick Quiz #1: Why is there no srcu_barrier()?
+
+Answer: Since there is no call_srcu(), there can be no outstanding SRCU
+ callbacks. Therefore, there is no need to wait for them.
+
+Quick Quiz #2: Why is there no rcu_barrier_bh()?
+
+Answer: Because no one has needed it yet. As soon as someone needs to
+ use call_rcu_bh() from within an unloadable module, they will
+ need an rcu_barrier_bh().
+
+Quick Quiz #3: Is there any other situation where rcu_barrier() might
+ be required?
+
+Answer: Interestingly enough, rcu_barrier() was not originally
+ implemented for module unloading. Nikita Danilov was using
+ RCU in a filesystem, which resulted in a similar situation at
+ filesystem-unmount time. Dipankar Sarma coded up rcu_barrier()
+ in response, so that Nikita could invoke it during the
+ filesystem-unmount process.
+
+ Much later, yours truly hit the RCU module-unload problem when
+ implementing rcutorture, and found that rcu_barrier() solves
+ this problem as well.
+
+Quick Quiz #4: What happens if CPU 0's rcu_barrier_func() executes
+ immediately (thus incrementing rcu_barrier_cpu_count to the
+ value one), but the other CPU's rcu_barrier_func() invocations
+ are delayed for a full grace period? Couldn't this result in
+ rcu_barrier() returning prematurely?
+
+Answer: This cannot happen. The reason is that on_each_cpu() has its last
+ argument, the wait flag, set to "1". This flag is passed through
+ to smp_call_function() and further to smp_call_function_on_cpu(),
+ causing this latter to spin until the cross-CPU invocation of
+ rcu_barrier_func() has completed. This by itself would prevent
+ a grace period from completing on non-CONFIG_PREEMPT kernels,
+ since each CPU must undergo a context switch (or other quiescent
+ state) before the grace period can complete. However, this is
+ of no use in CONFIG_PREEMPT kernels.
+
+ Therefore, on_each_cpu() disables preemption across its call
+ to smp_call_function() and also across the local call to
+ rcu_barrier_func(). This prevents the local CPU from context
+ switching, again preventing grace periods from completing. This
+ means that all CPUs have executed rcu_barrier_func() before
+ the first rcu_barrier_callback() can possibly execute, in turn
+ preventing rcu_barrier_cpu_count from prematurely reaching zero.
+
+ Currently, -rt implementations of RCU keep but a single global
+ queue for RCU callbacks, and thus do not suffer from this
+ problem. However, when the -rt RCU eventually does have per-CPU
+ callback queues, things will have to change. One simple change
+ is to add an rcu_read_lock() before line 8 of rcu_barrier()
+ and an rcu_read_unlock() after line 8 of this same function. If
+ you can think of a better change, please let me know!
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-13 17:31 ` Paul E. McKenney
@ 2008-11-14 1:03 ` Lai Jiangshan
2008-11-14 2:11 ` Paul E. McKenney
0 siblings, 1 reply; 16+ messages in thread
From: Lai Jiangshan @ 2008-11-14 1:03 UTC (permalink / raw)
To: paulmck
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List, corbet
Paul E. McKenney wrote:
> On Thu, Nov 13, 2008 at 10:48:33AM +0800, Lai Jiangshan wrote:
>> Hi, Paul,
>>
>> Could you add a RCU document about unloadable modules for kernel?
>
> You thinking in terms of an ASCII version of
> http://lwn.net/Articles/217484/?
>
> If so, please see attached patch and let me know what you think.
> Being too lazy to convert the cartoon to ASCII graphics, I simply
> left a URL to the .jpg on the LWN website. Thus we need an ack/nack
> from Jon Corbet (CCed).
Hi, Paul
Thank you. it's a very good document.
I found several modules which need rcu_barrier(). So I'm going to
do some cleanup for them. A document for rcu_barrier() will help
these cleanup patches be accepted easily by maintainers.
Lai.
>
> Of course, an alternative is to simply include the URL of the original
> LWN article in 00-INDEX. Thoughts?
>
> Thanx, Paul
>
> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> ---
> +The code for rcu_barrier() is as follows:
> +
> + 1 void rcu_barrier(void)
> + 2 {
> + 3 BUG_ON(in_interrupt());
> + 4 /* Take cpucontrol mutex to protect against CPU hotplug */
> + 5 mutex_lock(&rcu_barrier_mutex);
> + 6 init_completion(&rcu_barrier_completion);
> + 7 atomic_set(&rcu_barrier_cpu_count, 0);
> + 8 on_each_cpu(rcu_barrier_func, NULL, 0, 1);
> + 9 wait_for_completion(&rcu_barrier_completion);
> +10 mutex_unlock(&rcu_barrier_mutex);
> +11 }
> +
this is a little old.
> +
> +Quick Quiz #2: Why is there no rcu_barrier_bh()?
> +
> +Answer: Because no one has needed it yet. As soon as someone needs to
> + use call_rcu_bh() from within an unloadable module, they will
> + need an rcu_barrier_bh().
> +
add here.
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-14 1:03 ` Lai Jiangshan
@ 2008-11-14 2:11 ` Paul E. McKenney
2008-11-14 7:39 ` Lai Jiangshan
0 siblings, 1 reply; 16+ messages in thread
From: Paul E. McKenney @ 2008-11-14 2:11 UTC (permalink / raw)
To: Lai Jiangshan
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List, corbet
On Fri, Nov 14, 2008 at 09:03:20AM +0800, Lai Jiangshan wrote:
> Paul E. McKenney wrote:
> > On Thu, Nov 13, 2008 at 10:48:33AM +0800, Lai Jiangshan wrote:
> >> Hi, Paul,
> >>
> >> Could you add a RCU document about unloadable modules for kernel?
> >
> > You thinking in terms of an ASCII version of
> > http://lwn.net/Articles/217484/?
> >
> > If so, please see attached patch and let me know what you think.
> > Being too lazy to convert the cartoon to ASCII graphics, I simply
> > left a URL to the .jpg on the LWN website. Thus we need an ack/nack
> > from Jon Corbet (CCed).
>
> Hi, Paul
>
> Thank you. it's a very good document.
Updated version attached.
> I found several modules which need rcu_barrier(). So I'm going to
> do some cleanup for them. A document for rcu_barrier() will help
> these cleanup patches be accepted easily by maintainers.
Sounds very good!!!
> Lai.
>
> >
> > Of course, an alternative is to simply include the URL of the original
> > LWN article in 00-INDEX. Thoughts?
> >
> > Thanx, Paul
> >
> > Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> > ---
>
> > +The code for rcu_barrier() is as follows:
> > +
> > + 1 void rcu_barrier(void)
> > + 2 {
> > + 3 BUG_ON(in_interrupt());
> > + 4 /* Take cpucontrol mutex to protect against CPU hotplug */
> > + 5 mutex_lock(&rcu_barrier_mutex);
> > + 6 init_completion(&rcu_barrier_completion);
> > + 7 atomic_set(&rcu_barrier_cpu_count, 0);
> > + 8 on_each_cpu(rcu_barrier_func, NULL, 0, 1);
> > + 9 wait_for_completion(&rcu_barrier_completion);
> > +10 mutex_unlock(&rcu_barrier_mutex);
> > +11 }
> > +
>
> this is a little old.
Indeed!!! Good catch! I left this code, but added words saying that
it was the original and that it has since been rewritten to add
support for rcu_barrier_bh() and rcu_barrier_sched().
> > +
> > +Quick Quiz #2: Why is there no rcu_barrier_bh()?
> > +
> > +Answer: Because no one has needed it yet. As soon as someone needs to
> > + use call_rcu_bh() from within an unloadable module, they will
> > + need an rcu_barrier_bh().
> > +
>
> add here.
Good point, I deleted this Quick Quiz.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
diff -urpNa -X dontdiff linux-2.6.27/Documentation/RCU/00-INDEX linux-2.6.27-rcu_barrierdoc/Documentation/RCU/00-INDEX
--- linux-2.6.27/Documentation/RCU/00-INDEX 2008-10-09 15:13:53.000000000 -0700
+++ linux-2.6.27-rcu_barrierdoc/Documentation/RCU/00-INDEX 2008-11-13 08:46:17.000000000 -0800
@@ -12,6 +12,8 @@ rcuref.txt
- Reference-count design for elements of lists/arrays protected by RCU
rcu.txt
- RCU Concepts
+rcubarrier.txt
+ - Unloading modules that use RCU callbacks
RTFP.txt
- List of RCU papers (bibliography) going back to 1980.
torture.txt
diff -urpNa -X dontdiff linux-2.6.27/Documentation/RCU/rcubarrier.txt linux-2.6.27-rcu_barrierdoc/Documentation/RCU/rcubarrier.txt
--- linux-2.6.27/Documentation/RCU/rcubarrier.txt 1969-12-31 16:00:00.000000000 -0800
+++ linux-2.6.27-rcu_barrierdoc/Documentation/RCU/rcubarrier.txt 2008-11-13 18:08:35.000000000 -0800
@@ -0,0 +1,304 @@
+RCU and Unloadable Modules
+
+[Originally published in LWN Jan. 14, 2007: http://lwn.net/Articles/217484/]
+
+RCU (read-copy update) is a synchronization mechanism that can be thought
+of as a replacement for read-writer locking (among other things), but with
+very low-overhead readers that are immune to deadlock, priority inversion,
+and unbounded latency. RCU read-side critical sections are delimited
+by rcu_read_lock() and rcu_read_unlock(), which, in non-CONFIG_PREEMPT
+kernels, generate no code whatsoever.
+
+This means that RCU writers are unaware of the presence of concurrent
+readers, so that RCU updates to shared data must be undertaken quite
+carefully, leaving an old version of the data structure in place until all
+pre-existing readers have finished. These old versions are needed because
+such readers might hold a reference to them. RCU updates can therefore be
+rather expensive, and RCU is thus best suited for read-mostly situations.
+
+How can an RCU writer possibly determine when all readers are finished,
+given that readers might well leave absolutely no trace of their
+presence? There is a synchronize_rcu() primitive that blocks until all
+pre-existing readers have completed. An updater wishing to delete an
+element p from a linked list might do the following, while holding an
+appropriate lock, of course:
+
+ list_del_rcu(p);
+ synchronize_rcu();
+ kfree(p);
+
+But the above code cannot be used in IRQ context -- the call_rcu()
+primitive must be used instead. This primitive takes a pointer to an
+rcu_head struct placed within the RCU-protected data structure and
+another pointer to a function that may be invoked later to free that
+structure. Code to delete an element p from the linked list from IRQ
+context might then be as follows:
+
+ list_del_rcu(p);
+ call_rcu(&p->rcu, p_callback);
+
+Since call_rcu() never blocks, this code can safely be used from within
+IRQ context. The function p_callback() might be defined as follows:
+
+ static void p_callback(struct rcu_head *rp)
+ {
+ struct pstruct *p = container_of(rp, struct pstruct, rcu);
+
+ kfree(p);
+ }
+
+
+Unloading Modules That Use call_rcu()
+
+But what if p_callback is defined in an unloadable module?
+
+If we unload the module while some RCU callbacks are pending,
+the CPUs executing these callbacks are going to be severely
+disappointed when they are later invoked, as fancifully depicted at
+http://lwn.net/images/ns/kernel/rcu-drop.jpg.
+
+We could try placing a synchronize_rcu() in the module-exit code path,
+but this is not sufficient. Although synchronize_rcu() does wait for a
+grace period to elapse, it does not wait for the callbacks to complete.
+
+One might be tempted to try several back-to-back synchronize_rcu()
+calls, but this is still not guaranteed to work. If there is a very
+heavy RCU-callback load, then some of the callbacks might be deferred
+in order to allow other processing to proceed. Such deferral is required
+in realtime kernels in order to avoid excessive scheduling latencies.
+
+
+rcu_barrier()
+
+We instead need the rcu_barrier() primitive. This primitive is similar
+to synchronize_rcu(), but instead of waiting solely for a grace
+period to elapse, it also waits for all outstanding RCU callbacks to
+complete. Pseudo-code using rcu_barrier() is as follows:
+
+ 1. Prevent any new RCU callbacks from being posted.
+ 2. Execute rcu_barrier().
+ 3. Allow the module to be unloaded.
+
+Quick Quiz #1: Why is there no srcu_barrier()?
+
+The rcutorture module makes use of rcu_barrier in its exit function
+as follows:
+
+ 1 static void
+ 2 rcu_torture_cleanup(void)
+ 3 {
+ 4 int i;
+ 5
+ 6 fullstop = 1;
+ 7 if (shuffler_task != NULL) {
+ 8 VERBOSE_PRINTK_STRING("Stopping rcu_torture_shuffle task");
+ 9 kthread_stop(shuffler_task);
+10 }
+11 shuffler_task = NULL;
+12
+13 if (writer_task != NULL) {
+14 VERBOSE_PRINTK_STRING("Stopping rcu_torture_writer task");
+15 kthread_stop(writer_task);
+16 }
+17 writer_task = NULL;
+18
+19 if (reader_tasks != NULL) {
+20 for (i = 0; i < nrealreaders; i++) {
+21 if (reader_tasks[i] != NULL) {
+22 VERBOSE_PRINTK_STRING(
+23 "Stopping rcu_torture_reader task");
+24 kthread_stop(reader_tasks[i]);
+25 }
+26 reader_tasks[i] = NULL;
+27 }
+28 kfree(reader_tasks);
+29 reader_tasks = NULL;
+30 }
+31 rcu_torture_current = NULL;
+32
+33 if (fakewriter_tasks != NULL) {
+34 for (i = 0; i < nfakewriters; i++) {
+35 if (fakewriter_tasks[i] != NULL) {
+36 VERBOSE_PRINTK_STRING(
+37 "Stopping rcu_torture_fakewriter task");
+38 kthread_stop(fakewriter_tasks[i]);
+39 }
+40 fakewriter_tasks[i] = NULL;
+41 }
+42 kfree(fakewriter_tasks);
+43 fakewriter_tasks = NULL;
+44 }
+45
+46 if (stats_task != NULL) {
+47 VERBOSE_PRINTK_STRING("Stopping rcu_torture_stats task");
+48 kthread_stop(stats_task);
+49 }
+50 stats_task = NULL;
+51
+52 /* Wait for all RCU callbacks to fire. */
+53 rcu_barrier();
+54
+55 rcu_torture_stats_print(); /* -After- the stats thread is stopped! */
+56
+57 if (cur_ops->cleanup != NULL)
+58 cur_ops->cleanup();
+59 if (atomic_read(&n_rcu_torture_error))
+60 rcu_torture_print_module_parms("End of test: FAILURE");
+61 else
+62 rcu_torture_print_module_parms("End of test: SUCCESS");
+63 }
+
+Line 6 sets a global variable that prevents any RCU callbacks from
+re-posting themselves. This will not be necessary in most cases, since
+RCU callbacks rarely include calls to call_rcu(). However, the rcutorture
+module is an exception to this rule, and therefore needs to set this
+global variable.
+
+Lines 7-50 stop all the kernel tasks associated with the rcutorture
+module. Therefore, once execution reaches line 53, no more rcutorture
+RCU callbacks will be posted. The rcu_barrier() call on line 53 waits
+for any pre-existing callbacks to complete.
+
+Then lines 55-62 print status and do operation-specific cleanup, and
+then return, permitting the module-unload operation to be completed.
+
+Quick Quiz #2: Is there any other situation where rcu_barrier() might
+ be required?
+
+Your module might have additional complications. For example, if your
+module invokes call_rcu() from timers, you will need to first cancel all
+the timers, and only then invoke rcu_barrier() to wait for any remaining
+RCU callbacks to complete.
+
+
+Implementing rcu_barrier()
+
+Dipankar Sarma's implementation of rcu_barrier() makes use of the fact
+that RCU callbacks are never reordered once queued on one of the per-CPU
+queues. His implementation queues an RCU callback on each of the per-CPU
+callback queues, and then waits until they have all started executing, at
+which point, all earlier RCU callbacks are guaranteed to have completed.
+
+The original code for rcu_barrier() was as follows:
+
+ 1 void rcu_barrier(void)
+ 2 {
+ 3 BUG_ON(in_interrupt());
+ 4 /* Take cpucontrol mutex to protect against CPU hotplug */
+ 5 mutex_lock(&rcu_barrier_mutex);
+ 6 init_completion(&rcu_barrier_completion);
+ 7 atomic_set(&rcu_barrier_cpu_count, 0);
+ 8 on_each_cpu(rcu_barrier_func, NULL, 0, 1);
+ 9 wait_for_completion(&rcu_barrier_completion);
+10 mutex_unlock(&rcu_barrier_mutex);
+11 }
+
+Line 3 verifies that the caller is in process context, and lines 5 and 10
+use rcu_barrier_mutex to ensure that only one rcu_barrier() is using the
+global completion and counters at a time, which are initialized on lines
+6 and 7. Line 8 causes each CPU to invoke rcu_barrier_func(), which is
+shown below. Note that the final "1" in on_each_cpu()'s argument list
+ensures that all the calls to rcu_barrier_func() will have completed
+before on_each_cpu() returns. Line 9 then waits for the completion.
+
+This code was rewritten in 2008 to support rcu_barrier_bh() and
+rcu_barrier_sched() in addition to the original rcu_barrier().
+
+The rcu_barrier_func() runs on each CPU, where it invokes call_rcu()
+to post an RCU callback, as follows:
+
+ 1 static void rcu_barrier_func(void *notused)
+ 2 {
+ 3 int cpu = smp_processor_id();
+ 4 struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
+ 5 struct rcu_head *head;
+ 6
+ 7 head = &rdp->barrier;
+ 8 atomic_inc(&rcu_barrier_cpu_count);
+ 9 call_rcu(head, rcu_barrier_callback);
+10 }
+
+Lines 3 and 4 locate RCU's internal per-CPU rcu_data structure,
+which contains the struct rcu_head that needed for the later call to
+call_rcu(). Line 7 picks up a pointer to this struct rcu_head, and line
+8 increments a global counter. This counter will later be decremented
+by the callback. Line 9 then registers the rcu_barrier_callback() on
+the current CPU's queue.
+
+The rcu_barrier_callback() function simply atomically decrements the
+rcu_barrier_cpu_count variable and finalizes the completion when it
+reaches zero, as follows:
+
+ 1 static void rcu_barrier_callback(struct rcu_head *notused)
+ 2 {
+ 3 if (atomic_dec_and_test(&rcu_barrier_cpu_count))
+ 4 complete(&rcu_barrier_completion);
+ 5 }
+
+Quick Quiz #3: What happens if CPU 0's rcu_barrier_func() executes
+ immediately (thus incrementing rcu_barrier_cpu_count to the
+ value one), but the other CPU's rcu_barrier_func() invocations
+ are delayed for a full grace period? Couldn't this result in
+ rcu_barrier() returning prematurely?
+
+
+rcu_barrier() Summary
+
+The rcu_barrier() primitive has seen relatively little use, since most
+code using RCU is in the core kernel rather than in modules. However, if
+you are using RCU from an unloadable module, you need to use rcu_barrier()
+so that your module may be safely unloaded.
+
+
+Answers to Quick Quizzes
+
+Quick Quiz #1: Why is there no srcu_barrier()?
+
+Answer: Since there is no call_srcu(), there can be no outstanding SRCU
+ callbacks. Therefore, there is no need to wait for them.
+
+Quick Quiz #2: Is there any other situation where rcu_barrier() might
+ be required?
+
+Answer: Interestingly enough, rcu_barrier() was not originally
+ implemented for module unloading. Nikita Danilov was using
+ RCU in a filesystem, which resulted in a similar situation at
+ filesystem-unmount time. Dipankar Sarma coded up rcu_barrier()
+ in response, so that Nikita could invoke it during the
+ filesystem-unmount process.
+
+ Much later, yours truly hit the RCU module-unload problem when
+ implementing rcutorture, and found that rcu_barrier() solves
+ this problem as well.
+
+Quick Quiz #3: What happens if CPU 0's rcu_barrier_func() executes
+ immediately (thus incrementing rcu_barrier_cpu_count to the
+ value one), but the other CPU's rcu_barrier_func() invocations
+ are delayed for a full grace period? Couldn't this result in
+ rcu_barrier() returning prematurely?
+
+Answer: This cannot happen. The reason is that on_each_cpu() has its last
+ argument, the wait flag, set to "1". This flag is passed through
+ to smp_call_function() and further to smp_call_function_on_cpu(),
+ causing this latter to spin until the cross-CPU invocation of
+ rcu_barrier_func() has completed. This by itself would prevent
+ a grace period from completing on non-CONFIG_PREEMPT kernels,
+ since each CPU must undergo a context switch (or other quiescent
+ state) before the grace period can complete. However, this is
+ of no use in CONFIG_PREEMPT kernels.
+
+ Therefore, on_each_cpu() disables preemption across its call
+ to smp_call_function() and also across the local call to
+ rcu_barrier_func(). This prevents the local CPU from context
+ switching, again preventing grace periods from completing. This
+ means that all CPUs have executed rcu_barrier_func() before
+ the first rcu_barrier_callback() can possibly execute, in turn
+ preventing rcu_barrier_cpu_count from prematurely reaching zero.
+
+ Currently, -rt implementations of RCU keep but a single global
+ queue for RCU callbacks, and thus do not suffer from this
+ problem. However, when the -rt RCU eventually does have per-CPU
+ callback queues, things will have to change. One simple change
+ is to add an rcu_read_lock() before line 8 of rcu_barrier()
+ and an rcu_read_unlock() after line 8 of this same function. If
+ you can think of a better change, please let me know!
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-14 2:11 ` Paul E. McKenney
@ 2008-11-14 7:39 ` Lai Jiangshan
2008-11-14 19:25 ` Jonathan Corbet
0 siblings, 1 reply; 16+ messages in thread
From: Lai Jiangshan @ 2008-11-14 7:39 UTC (permalink / raw)
To: paulmck
Cc: Ingo Molnar, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List, corbet
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Paul E. McKenney wrote:
[...]
> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> ---
>
> diff -urpNa -X dontdiff linux-2.6.27/Documentation/RCU/00-INDEX linux-2.6.27-rcu_barrierdoc/Documentation/RCU/00-INDEX
> --- linux-2.6.27/Documentation/RCU/00-INDEX 2008-10-09 15:13:53.000000000 -0700
> +++ linux-2.6.27-rcu_barrierdoc/Documentation/RCU/00-INDEX 2008-11-13 08:46:17.000000000 -0800
> @@ -12,6 +12,8 @@ rcuref.txt
> - Reference-count design for elements of lists/arrays protected by RCU
> rcu.txt
> - RCU Concepts
> +rcubarrier.txt
> + - Unloading modules that use RCU callbacks
> RTFP.txt
> - List of RCU papers (bibliography) going back to 1980.
> torture.txt
> diff -urpNa -X dontdiff linux-2.6.27/Documentation/RCU/rcubarrier.txt linux-2.6.27-rcu_barrierdoc/Documentation/RCU/rcubarrier.txt
> --- linux-2.6.27/Documentation/RCU/rcubarrier.txt 1969-12-31 16:00:00.000000000 -0800
> +++ linux-2.6.27-rcu_barrierdoc/Documentation/RCU/rcubarrier.txt 2008-11-13 18:08:35.000000000 -0800
[...]
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-14 7:39 ` Lai Jiangshan
@ 2008-11-14 19:25 ` Jonathan Corbet
2008-11-15 20:39 ` Paul E. McKenney
0 siblings, 1 reply; 16+ messages in thread
From: Jonathan Corbet @ 2008-11-14 19:25 UTC (permalink / raw)
To: Lai Jiangshan
Cc: paulmck, Ingo Molnar, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List
On Fri, 14 Nov 2008 15:39:02 +0800
Lai Jiangshan <laijs@cn.fujitsu.com> wrote:
> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
>
> Paul E. McKenney wrote:
All seems good. To answer Paul's original question:
> If so, please see attached patch and let me know what you think.
> Being too lazy to convert the cartoon to ASCII graphics, I simply
> left a URL to the .jpg on the LWN website. Thus we need an ack/nack
> from Jon Corbet (CCed).
Of course I have no problem with the LWN links; they will remain stable.
If nobody objects, I'll drop this into my documentation tree. Gotta
have *something* there, after all, or it might start to feel
neglected...
jon
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-14 19:25 ` Jonathan Corbet
@ 2008-11-15 20:39 ` Paul E. McKenney
2008-11-17 12:57 ` Lai Jiangshan
0 siblings, 1 reply; 16+ messages in thread
From: Paul E. McKenney @ 2008-11-15 20:39 UTC (permalink / raw)
To: Jonathan Corbet
Cc: Lai Jiangshan, Ingo Molnar, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List
On Fri, Nov 14, 2008 at 12:25:43PM -0700, Jonathan Corbet wrote:
> On Fri, 14 Nov 2008 15:39:02 +0800
> Lai Jiangshan <laijs@cn.fujitsu.com> wrote:
>
> > Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
> >
> > Paul E. McKenney wrote:
>
> All seems good. To answer Paul's original question:
>
> > If so, please see attached patch and let me know what you think.
> > Being too lazy to convert the cartoon to ASCII graphics, I simply
> > left a URL to the .jpg on the LWN website. Thus we need an ack/nack
> > from Jon Corbet (CCed).
>
> Of course I have no problem with the LWN links; they will remain stable.
>
> If nobody objects, I'll drop this into my documentation tree. Gotta
> have *something* there, after all, or it might start to feel
> neglected...
Works for me!
Thanx, Paul
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-15 20:39 ` Paul E. McKenney
@ 2008-11-17 12:57 ` Lai Jiangshan
2008-11-17 21:28 ` Jonathan Corbet
0 siblings, 1 reply; 16+ messages in thread
From: Lai Jiangshan @ 2008-11-17 12:57 UTC (permalink / raw)
To: Ingo Molnar
Cc: paulmck, Jonathan Corbet, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List
Hi, Ingo,
Is there any problem for this document added to kernel?
Thanx, Lai.
Paul E. McKenney wrote:
> On Fri, Nov 14, 2008 at 12:25:43PM -0700, Jonathan Corbet wrote:
>> On Fri, 14 Nov 2008 15:39:02 +0800
>> Lai Jiangshan <laijs@cn.fujitsu.com> wrote:
>>
>>> Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
>>>
>>> Paul E. McKenney wrote:
>> All seems good. To answer Paul's original question:
>>
>>> If so, please see attached patch and let me know what you think.
>>> Being too lazy to convert the cartoon to ASCII graphics, I simply
>>> left a URL to the .jpg on the LWN website. Thus we need an ack/nack
>>> from Jon Corbet (CCed).
>> Of course I have no problem with the LWN links; they will remain stable.
>>
>> If nobody objects, I'll drop this into my documentation tree. Gotta
>> have *something* there, after all, or it might start to feel
>> neglected...
>
> Works for me!
>
> Thanx, Paul
>
>
>
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2
2008-11-17 12:57 ` Lai Jiangshan
@ 2008-11-17 21:28 ` Jonathan Corbet
0 siblings, 0 replies; 16+ messages in thread
From: Jonathan Corbet @ 2008-11-17 21:28 UTC (permalink / raw)
To: Lai Jiangshan
Cc: Ingo Molnar, paulmck, Peter Zijlstra, Andrew Morton,
Linux Kernel Mailing List
On Mon, 17 Nov 2008 20:57:40 +0800
Lai Jiangshan <laijs@cn.fujitsu.com> wrote:
> Is there any problem for this document added to kernel?
No problem; I'll get it queued up for 2.6.29 unless somebody else gets
it there first.
jon
^ permalink raw reply [flat|nested] 16+ messages in thread
end of thread, other threads:[~2008-11-17 21:29 UTC | newest]
Thread overview: 16+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-11-06 6:47 [PATCH] rcupdate: move synchronize_sched() back to rcupdate.c V2 Lai Jiangshan
2008-11-06 6:57 ` Ingo Molnar
2008-11-09 0:51 ` Paul E. McKenney
2008-11-10 3:22 ` Lai Jiangshan
2008-11-10 18:45 ` Paul E. McKenney
2008-11-11 0:55 ` Lai Jiangshan
2008-11-11 1:03 ` Paul E. McKenney
2008-11-13 2:48 ` Lai Jiangshan
2008-11-13 17:31 ` Paul E. McKenney
2008-11-14 1:03 ` Lai Jiangshan
2008-11-14 2:11 ` Paul E. McKenney
2008-11-14 7:39 ` Lai Jiangshan
2008-11-14 19:25 ` Jonathan Corbet
2008-11-15 20:39 ` Paul E. McKenney
2008-11-17 12:57 ` Lai Jiangshan
2008-11-17 21:28 ` Jonathan Corbet
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).