Skip to contentSkip to conversation

osq_lock: fix osq_lock queue corruption

PATCH In review Prateek Sood 4 participants 6 messages Jul 14, 2017 – Sep 1, 2017

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 osq_lock: fix osq_lock queue corruption

Fix ordering of link creation between node->prev and prev->next in osq_lock(). A case in which the status of optimistic spin queue is CPU6->CPU2 in which CPU6 has acquired the lock. At this point if CPU0 comes in to acquire osq_lock, it will update the tail count. After tail count update if CPU2 starts to unqueue itself from optimistic spin queue, it will find updated tail count with CPU0 and update CPU2 node->next to NULL in osq_wait_next(). If reordering of following stores happen then prev->next where prev being CPU2 would be updated to point to CPU0 node:

	node->prev = prev;
	WRITE_ONCE(prev->next, node);
At this point if next instruction
	WRITE_ONCE(next->prev, prev);

in CPU2 path is committed before the update of CPU0 node->prev = prev then CPU0 node->prev will point to CPU6 node. At this point if CPU0 path's node->prev = prev is committed resulting in change of CPU0 prev back to CPU2 node. CPU2 node->next is NULL currently, so if CPU0 gets into unqueue path of osq_lock it will keep spinning in infinite loop as condition prev->next == node will never be true.

---
 kernel/locking/osq_lock.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)
kernel/locking/osq_lock.c +26 -0
index 05a3785..1d371a0 100644
@@ -104,6 +104,32 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 
 	prev = decode_cpu(old);
 	node->prev = prev;
+
+	/*
+	 * We need to avoid reordering of link updation sequence of osq.
+	 * A case in which the status of optimistic spin queue is
+	 * CPU6->CPU2 in which CPU6 has acquired the lock. At this point
+	 * if CPU0 comes in to acquire osq_lock, it will update the tail
+	 * count. After tail count update if CPU2 starts to unqueue itself
+	 * from optimistic spin queue, it will find updated tail count with
+	 * CPU0 and update CPU2 node->next to NULL in osq_wait_next(). If
+	 * reordering of following stores happen then prev->next where prev
+	 * being CPU2 would be updated to point to CPU0 node:
+	 *      node->prev = prev;
+	 *      WRITE_ONCE(prev->next, node);
+	 *
+	 * At this point if next instruction
+	 *      WRITE_ONCE(next->prev, prev);
+	 * in CPU2 path is committed before the update of CPU0 node->prev =
+	 * prev then CPU0 node->prev will point to CPU6 node. At this point
+	 * if CPU0 path's node->prev = prev is committed resulting in change
+	 * of CPU0 prev back to CPU2 node. CPU2 node->next is NULL, so if
+	 * CPU0 gets into unqueue path of osq_lock it will keep spinning
+	 * in infinite loop as condition prev->next == node will never be
+	 * true.
+	 */
+	smp_wmb();
+
 	WRITE_ONCE(prev->next, node);
 
 	/*
Signature · 2 lines
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc., 
is a member of Code Aurora Forum, a Linux Foundation Collaborative Project.
maintainer
#1 lore raw

I added a few pictures, just the text didn't want to make sense to me.

Quoted 3 lines · Prateek Sood

Fix ordering of link creation between node->prev and prev->next in osq_lock(). A case in which the status of optimistic spin queue is CPU6->CPU2 in which CPU6 has acquired the lock.

tail
v
,-. <- ,-.
|6| |2|
-' -> -'

Quoted 2 lines

At this point if CPU0 comes in to acquire osq_lock, it will update the tail count.

CPU2 CPU0
----------------------------------

				       tail
				         v
			  ,-. <- ,-.    ,-.
			  |6|    |2|    |0|
			  `-' -> `-'    `-'
Quoted 3 lines

After tail count update if CPU2 starts to unqueue itself from optimistic spin queue, it will find updated tail count with CPU0 and update CPU2 node->next to NULL in osq_wait_next().

unqueue-A

tail
v

  ,-. <- ,-.    ,-.
  |6|    |2|    |0|
  `-'    `-'    `-'

unqueue-B

  ->tail != curr && !node->next
Quoted 2 lines

If reordering of following stores happen then
prev->next where prev being CPU2 would be updated to point to CPU0 node:

				       tail
				         v
			  ,-. <- ,-.    ,-.
			  |6|    |2|    |0|
			  `-' -> `-' -> `-'

osq_wait_next()

    node->next <- 0
    xchg(node->next, NULL)

	       tail
	         v
  ,-. <- ,-.    ,-.
  |6|    |2|    |0|
  `-'    `-'    `-'

unqueue-C

Quoted 4 lines

At this point if next instruction

	WRITE_ONCE(next->prev, prev);

in CPU2 path is committed before the update of CPU0 node->prev = prev then CPU0 node->prev will point to CPU6 node.

	       tail
    v----------. v
  ,-. <- ,-.    ,-.
  |6|    |2|    |0|
  `-'    `-'    `-'
     `----------^
Quoted 3 lines

At this point if CPU0 path's node->prev = prev is committed resulting in change of CPU0 prev back to CPU2 node. CPU2 node->next is NULL currently,

				       tail
			                 v
			  ,-. <- ,-. <- ,-.
			  |6|    |2|    |0|
			  `-'    `-'    `-'
			     `----------^
Quoted 2 lines

so if CPU0 gets into unqueue path of osq_lock it will keep spinning in infinite loop as condition prev->next == node will never be true.

Also updated the comment..

---
 kernel/locking/osq_lock.c |   13 +++++++++++++
 1 file changed, 13 insertions(+)
kernel/locking/osq_lock.c +13 -0
@@ -109,6 +109,19 @@ bool osq_lock(struct optimistic_spin_que
 
 	prev = decode_cpu(old);
 	node->prev = prev;
+
+	/*
+	 * osq_lock()			unqueue
+	 *
+	 * node->prev = prev		osq_wait_next()
+	 * WMB				MB
+	 * prev->next = node		next->prev = prev // unqueue-C
+	 *
+	 * Here 'node->prev' and 'next->prev' are the same variable and we need
+	 * to ensure these stores happen in-order to avoid corrupting the list.
+	 */
+	smp_wmb();
+
 	WRITE_ONCE(prev->next, node);
 
 	/*
#2 lore raw

[tip:locking/core] locking/osq_lock: Fix osq_lock queue corruption

In reply to Prateek Sood
locking/osq_lock: Fix osq_lock queue corruption
Commit-ID:  50972fe78f24f1cd0b9d7bbf1f87d2be9e4f412e
Gitweb:     http://git.kernel.org/tip/50972fe78f24f1cd0b9d7bbf1f87d2be9e4f412e
Author:     Prateek Sood <[email protected]>
AuthorDate: Fri, 14 Jul 2017 19:17:56 +0530
Committer:  Ingo Molnar <[email protected]>
CommitDate: Thu, 10 Aug 2017 12:28:54 +0200

locking/osq_lock: Fix osq_lock queue corruption

Fix ordering of link creation between node->prev and prev->next in osq_lock(). A case in which the status of optimistic spin queue is CPU6->CPU2 in which CPU6 has acquired the lock.

tail
v
,-. <- ,-.
|6| |2|
-' -> -'

At this point if CPU0 comes in to acquire osq_lock, it will update the tail count.

CPU2 CPU0
----------------------------------

				       tail
				         v
			  ,-. <- ,-.    ,-.
			  |6|    |2|    |0|
			  `-' -> `-'    `-'

After tail count update if CPU2 starts to unqueue itself from optimistic spin queue, it will find an updated tail count with CPU0 and update CPU2 node->next to NULL in osq_wait_next().

unqueue-A

tail
v

  ,-. <- ,-.    ,-.
  |6|    |2|    |0|
  `-'    `-'    `-'

unqueue-B

  ->tail != curr && !node->next

If reordering of following stores happen then prev->next where prev being CPU2 would be updated to point to CPU0 node:

				       tail
				         v
			  ,-. <- ,-.    ,-.
			  |6|    |2|    |0|
			  `-'    `-' -> `-'

osq_wait_next()

    node->next <- 0
    xchg(node->next, NULL)

	       tail
	         v
  ,-. <- ,-.    ,-.
  |6|    |2|    |0|
  `-'    `-'    `-'

unqueue-C

At this point if next instruction

	WRITE_ONCE(next->prev, prev);

in CPU2 path is committed before the update of CPU0 node->prev = prev then CPU0 node->prev will point to CPU6 node.

	       tail
    v----------. v
  ,-. <- ,-.    ,-.
  |6|    |2|    |0|
  `-'    `-'    `-'
     `----------^

At this point if CPU0 path's node->prev = prev is committed resulting in change of CPU0 prev back to CPU2 node. CPU2 node->next is NULL currently,

				       tail
			                 v
			  ,-. <- ,-. <- ,-.
			  |6|    |2|    |0|
			  `-'    `-'    `-'
			     `----------^

so if CPU0 gets into unqueue path of osq_lock it will keep spinning in infinite loop as condition prev->next == node will never be true.

---
 kernel/locking/osq_lock.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)
kernel/locking/osq_lock.c +13 -0
index a316794..a74ee6a 100644
@@ -109,6 +109,19 @@ bool osq_lock(struct optimistic_spin_queue *lock)
 
 	prev = decode_cpu(old);
 	node->prev = prev;
+
+	/*
+	 * osq_lock()			unqueue
+	 *
+	 * node->prev = prev		osq_wait_next()
+	 * WMB				MB
+	 * prev->next = node		next->prev = prev // unqueue-C
+	 *
+	 * Here 'node->prev' and 'next->prev' are the same variable and we need
+	 * to ensure these stores happen in-order to avoid corrupting the list.
+	 */
+	smp_wmb();
+
 	WRITE_ONCE(prev->next, node);
 
 	/*
#3 lore raw

Re: [tip:locking/core] locking/osq_lock: Fix osq_lock queue corruption

Quoted 9 lines · tip-bot for Prateek Sood

Any reason -stable wasn't Cc'ed here? This goes back to 3.15:

fb0527bd5ea (locking/mutexes: Introduce cancelable MCS lock for adaptive spinning)

Thanks,
Davidlohr

maintainer
#4 lore raw

Re: [tip:locking/core] locking/osq_lock: Fix osq_lock queue corruption

Quoted 12 lines · Davidlohr Bueso
Quoted 9 lines · tip-bot for Prateek Sood
Quoted 3 lines

Any reason -stable wasn't Cc'ed here? This goes back to 3.15:

fb0527bd5ea (locking/mutexes: Introduce cancelable MCS lock for adaptive spinning)

No real reason, and yes, I should have done a Fixes tag I suppose :/

#5 lore raw

Re: [tip:locking/core] locking/osq_lock: Fix osq_lock queue corruption

Quoted 13 lines · Peter Zijlstra
Quoted 12 lines · Davidlohr Bueso
Quoted 9 lines · tip-bot for Prateek Sood
Quoted 5 lines
Quoted 3 lines

Any reason -stable wasn't Cc'ed here? This goes back to 3.15:

fb0527bd5ea (locking/mutexes: Introduce cancelable MCS lock for adaptive spinning)

No real reason, and yes, I should have done a Fixes tag I suppose :/

Ok, Cc'ed now.

Thanks,
Davidlohr