rcupdate: fix 2 bugs of rcu_barrier*()

current rcu_barrier_bh() is like this:

void rcu_barrier_bh(void)

{
	BUG_ON(in_interrupt());
	/* Take cpucontrol mutex to protect against CPU hotplug */
	mutex_lock(&rcu_barrier_mutex);
	init_completion(&rcu_barrier_completion);
	atomic_set(&rcu_barrier_cpu_count, 0);
	/*
	 * The queueing of callbacks in all CPUs must be atomic with
	 * respect to RCU, otherwise one CPU may queue a callback,
	 * wait for a grace period, decrement barrier count and call
	 * complete(), while other CPUs have not yet queued anything.
	 * So, we need to make sure that grace periods cannot complete
	 * until all the callbacks are queued.
	 */
	rcu_read_lock();
	on_each_cpu(rcu_barrier_func, (void *)RCU_BARRIER_BH, 1);
	rcu_read_unlock();
	wait_for_completion(&rcu_barrier_completion);
	mutex_unlock(&rcu_barrier_mutex);
}

this is bug, rcu_read_lock() cannot make sure that "grace periods for RCU_BH cannot complete until all the callbacks are queued".
it only make sure that race periods for RCU cannot complete until all the callbacks are queued.

so we must use rcu_read_lock_bh() for rcu_barrier_bh(). like this:

void rcu_barrier_bh(void)

{
	......
	rcu_read_lock_bh();
	on_each_cpu(rcu_barrier_func, (void *)RCU_BARRIER_BH, 1);
	rcu_read_unlock_bh();
	......
}

and also rcu_barrier() rcu_barrier_sched() are implemented like this. it will bring a lot of duplicate code. My patch uses another way to fix this bug, please see the comment of my patch.

Bug 2:
on_each_cpu() do not imply wmb, so we need a explicit wmb. I became a paranoid too.

Signed-off-by Lai Jiangshan <[email protected]>

---

kernel/rcupdate.c +0 -3

index 467d594..a667e21 100644
@@ -119,18 +119,23 @@ static void _rcu_barrier(enum rcu_barrier type)
 	/* Take cpucontrol mutex to protect against CPU hotplug */
 	mutex_lock(&rcu_barrier_mutex);
 	init_completion(&rcu_barrier_completion);
-	atomic_set(&rcu_barrier_cpu_count, 0);
 	/*
-	 * The queueing of callbacks in all CPUs must be atomic with
-	 * respect to RCU, otherwise one CPU may queue a callback,

kernel/rcupdate.c +2 -1

 	on_each_cpu(rcu_barrier_func, (void *)type, 1);
-	rcu_read_unlock();
+	if (atomic_dec_and_test(&rcu_barrier_cpu_count))
+		complete(&rcu_barrier_completion);
 	wait_for_completion(&rcu_barrier_completion);
 	mutex_unlock(&rcu_barrier_mutex);
 }

Quoted 41 lines · Lai Jiangshan

current rcu_barrier_bh() is like this:

void rcu_barrier_bh(void)

{
	BUG_ON(in_interrupt());
	/* Take cpucontrol mutex to protect against CPU hotplug */
	mutex_lock(&rcu_barrier_mutex);
	init_completion(&rcu_barrier_completion);
	atomic_set(&rcu_barrier_cpu_count, 0);
	/*
	 * The queueing of callbacks in all CPUs must be atomic with
	 * respect to RCU, otherwise one CPU may queue a callback,
	 * wait for a grace period, decrement barrier count and call
	 * complete(), while other CPUs have not yet queued anything.
	 * So, we need to make sure that grace periods cannot complete
	 * until all the callbacks are queued.
	 */
	rcu_read_lock();
	on_each_cpu(rcu_barrier_func, (void *)RCU_BARRIER_BH, 1);
	rcu_read_unlock();
	wait_for_completion(&rcu_barrier_completion);
	mutex_unlock(&rcu_barrier_mutex);
}

this is bug, rcu_read_lock() cannot make sure that "grace periods for RCU_BH cannot complete until all the callbacks are queued".
it only make sure that race periods for RCU cannot complete until all the callbacks are queued.

so we must use rcu_read_lock_bh() for rcu_barrier_bh(). like this:

void rcu_barrier_bh(void)

{
	......
	rcu_read_lock_bh();
	on_each_cpu(rcu_barrier_func, (void *)RCU_BARRIER_BH, 1);
	rcu_read_unlock_bh();
	......
}

Quoted 3 lines

and also rcu_barrier() rcu_barrier_sched() are implemented like this. it will bring a lot of duplicate code. My patch uses another way to fix this bug, please see the comment of my patch.

Excellent catch!!! I had incorrectly convinced myself that because RCU read-side implies an RCU_BH and RCU_SCHED that I could simply use an RCU read-side critical section. Thank you for finding this!

Just out of curiosity, did an actual oops/hang lead you to this bug, or did you find it by inspection?

Quoted 4 lines

Bug 2:
on_each_cpu() do not imply wmb, so we need a explicit wmb. I became a paranoid too.

Actually, there is a memory barrier in the list_add_tail_rcu() in the implementation of smp_call_function(), and furthermore, the way that atomic operations work on all architectures I am aware of removes the need for the memory barrier. Nevertheless, I have absolutely no objection to adding this memory barrier. This code path is used infrequently and has high overhead anyway, so I agree that making it easy to understand is the correct approach. If it were on the read side, I would argue. ;-)

In any case, I must agree that you are doing a good job of learning to be paranoid!

The only change I suggest is to rewrite the comments as shown below.

With that update, this change should be applied.

Reviewed-by Paul E. McKenney <[email protected]>

Quoted 14 lines

Signed-off-by Lai Jiangshan <[email protected]>

---

kernel/rcupdate.c +0 -3

index 467d594..a667e21 100644
@@ -119,18 +119,23 @@ static void _rcu_barrier(enum rcu_barrier type)
 	/* Take cpucontrol mutex to protect against CPU hotplug */
 	mutex_lock(&rcu_barrier_mutex);
 	init_completion(&rcu_barrier_completion);
-	atomic_set(&rcu_barrier_cpu_count, 0);
 	/*
-	 * The queueing of callbacks in all CPUs must be atomic with
-	 * respect to RCU, otherwise one CPU may queue a callback,

kernel/rcupdate.c static void _rcu_barrier(enum rcu_barrier type) · line 128

Quoted 10 lines

kernel/rcupdate.c +6 -4

-	 * wait for a grace period, decrement barrier count and call
-	 * complete(), while other CPUs have not yet queued anything.
-	 * So, we need to make sure that grace periods cannot complete
-	 * until all the callbacks are queued.
+	 * init and set rcu_barrier_cpu_count to 1, otherwise(set it to 0)
+	 * one CPU may queue a callback, wait for a grace period, decrement
+	 * barrier count and call complete(), while other CPUs have not yet
+	 * queued anything.
+	 * So, we need to make sure that rcu_barrier_cpu_count cannot become
+	 * 0 until all the callbacks are queued.

* Initialize rcu_barrier_cpu_count to 1, then invoke
* rcu_barrier_func() on each CPU, so that each CPU also has
* incremented rcu_barrier_cpu_count. Only then is it safe to
* decrement rcu_barrier_cpu_count -- otherwise the first CPU
* might complete its grace period before all of the other CPUs
* did their increment, causing this function to return too
* early.

kernel/rcupdate.c static void _rcu_barrier(enum rcu_barrier type) · line 135

Quoted 8 lines

kernel/rcupdate.c +6 -1

 	 */
-	rcu_read_lock();
+	atomic_set(&rcu_barrier_cpu_count, 1);
+	/*
+	 * rcu_barrier_cpu_count = 1 must be visible to cpus before
+	 * them call rcu_barrier_func().
+	 */
+	smp_wmb();

	smp_wmb(); /* atomic_set() must precede all rcu_barrier_func()s. */

Quoted 10 lines

kernel/rcupdate.c +2 -1

 	on_each_cpu(rcu_barrier_func, (void *)type, 1);
-	rcu_read_unlock();
+	if (atomic_dec_and_test(&rcu_barrier_cpu_count))
+		complete(&rcu_barrier_completion);
 	wait_for_completion(&rcu_barrier_completion);
 	mutex_unlock(&rcu_barrier_mutex);
 }

Quoted 50 lines · Paul E. McKenney

Quoted 41 lines · Lai Jiangshan

current rcu_barrier_bh() is like this:

void rcu_barrier_bh(void)

{
	BUG_ON(in_interrupt());
	/* Take cpucontrol mutex to protect against CPU hotplug */
	mutex_lock(&rcu_barrier_mutex);
	init_completion(&rcu_barrier_completion);
	atomic_set(&rcu_barrier_cpu_count, 0);
	/*
	 * The queueing of callbacks in all CPUs must be atomic with
	 * respect to RCU, otherwise one CPU may queue a callback,
	 * wait for a grace period, decrement barrier count and call
	 * complete(), while other CPUs have not yet queued anything.
	 * So, we need to make sure that grace periods cannot complete
	 * until all the callbacks are queued.
	 */
	rcu_read_lock();
	on_each_cpu(rcu_barrier_func, (void *)RCU_BARRIER_BH, 1);
	rcu_read_unlock();
	wait_for_completion(&rcu_barrier_completion);
	mutex_unlock(&rcu_barrier_mutex);
}

this is bug, rcu_read_lock() cannot make sure that "grace periods for RCU_BH cannot complete until all the callbacks are queued".
it only make sure that race periods for RCU cannot complete until all the callbacks are queued.

so we must use rcu_read_lock_bh() for rcu_barrier_bh(). like this:

void rcu_barrier_bh(void)

{
	......
	rcu_read_lock_bh();
	on_each_cpu(rcu_barrier_func, (void *)RCU_BARRIER_BH, 1);
	rcu_read_unlock_bh();
	......
}

Quoted 3 lines

and also rcu_barrier() rcu_barrier_sched() are implemented like this. it will bring a lot of duplicate code. My patch uses another way to fix this bug, please see the comment of my patch.

Excellent catch!!! I had incorrectly convinced myself that because RCU read-side implies an RCU_BH and RCU_SCHED that I could simply use an RCU read-side critical section. Thank you for finding this!

Quoted 2 lines

Just out of curiosity, did an actual oops/hang lead you to this bug, or did you find it by inspection?

by inspection. I was planning to put synchronize_rcu* back to kernel/rcupdate.c and I found the code and the comments are inconsistent suddenly when I was reviewing kernel/rcupdate.c.

Quoted 11 lines

Quoted 3 lines

Bug 2:
on_each_cpu() do not imply wmb, so we need a explicit wmb. I became a paranoid too.

Actually, there is a memory barrier in the list_add_tail_rcu() in the implementation of smp_call_function(), and furthermore, the way that atomic operations work on all architectures I am aware of removes the need for the memory barrier. Nevertheless, I have absolutely no objection to adding this memory barrier. This code path is used infrequently and has high overhead anyway, so I agree that making it easy to understand is the correct approach. If it were on the read side, I would argue. ;-)

I will remove this wmb.
Thank you a lot

Lai.

Quoted 67 lines

In any case, I must agree that you are doing a good job of learning to be paranoid!

The only change I suggest is to rewrite the comments as shown below.

With that update, this change should be applied.

Reviewed-by Paul E. McKenney <[email protected]>

Quoted 14 lines

Signed-off-by Lai Jiangshan <[email protected]>

---

kernel/rcupdate.c +0 -3

index 467d594..a667e21 100644
@@ -119,18 +119,23 @@ static void _rcu_barrier(enum rcu_barrier type)
 	/* Take cpucontrol mutex to protect against CPU hotplug */
 	mutex_lock(&rcu_barrier_mutex);
 	init_completion(&rcu_barrier_completion);
-	atomic_set(&rcu_barrier_cpu_count, 0);
 	/*
-	 * The queueing of callbacks in all CPUs must be atomic with
-	 * respect to RCU, otherwise one CPU may queue a callback,

kernel/rcupdate.c static void _rcu_barrier(enum rcu_barrier type) · line 128

Quoted 10 lines

kernel/rcupdate.c +6 -4

-	 * wait for a grace period, decrement barrier count and call
-	 * complete(), while other CPUs have not yet queued anything.
-	 * So, we need to make sure that grace periods cannot complete
-	 * until all the callbacks are queued.
+	 * init and set rcu_barrier_cpu_count to 1, otherwise(set it to 0)
+	 * one CPU may queue a callback, wait for a grace period, decrement
+	 * barrier count and call complete(), while other CPUs have not yet
+	 * queued anything.
+	 * So, we need to make sure that rcu_barrier_cpu_count cannot become
+	 * 0 until all the callbacks are queued.

* Initialize rcu_barrier_cpu_count to 1, then invoke
* rcu_barrier_func() on each CPU, so that each CPU also has
* incremented rcu_barrier_cpu_count. Only then is it safe to
* decrement rcu_barrier_cpu_count -- otherwise the first CPU
* might complete its grace period before all of the other CPUs
* did their increment, causing this function to return too
* early.

kernel/rcupdate.c static void _rcu_barrier(enum rcu_barrier type) · line 135

Quoted 8 lines

kernel/rcupdate.c +6 -1

 	 */
-	rcu_read_lock();
+	atomic_set(&rcu_barrier_cpu_count, 1);
+	/*
+	 * rcu_barrier_cpu_count = 1 must be visible to cpus before
+	 * them call rcu_barrier_func().
+	 */
+	smp_wmb();

	smp_wmb(); /* atomic_set() must precede all rcu_barrier_func()s. */

Quoted 10 lines

kernel/rcupdate.c +2 -1

 	on_each_cpu(rcu_barrier_func, (void *)type, 1);
-	rcu_read_unlock();
+	if (atomic_dec_and_test(&rcu_barrier_cpu_count))
+		complete(&rcu_barrier_completion);
 	wait_for_completion(&rcu_barrier_completion);
 	mutex_unlock(&rcu_barrier_mutex);
 }