Files
linux/include/linux
Peter Zijlstra 55eed755c6 locking/seqcount: Re-fix raw_read_seqcount_latch()
Commit 50755bc1c3 ("seqlock: fix raw_read_seqcount_latch()") broke
raw_read_seqcount_latch().

If you look at the comment that was modified; the thing that changes is
the seq count, not the latch pointer.

 * void latch_modify(struct latch_struct *latch, ...)
 * {
 *	smp_wmb();	<- Ensure that the last data[1] update is visible
 *	latch->seq++;
 *	smp_wmb();	<- Ensure that the seqcount update is visible
 *
 *	modify(latch->data[0], ...);
 *
 *	smp_wmb();	<- Ensure that the data[0] update is visible
 *	latch->seq++;
 *	smp_wmb();	<- Ensure that the seqcount update is visible
 *
 *	modify(latch->data[1], ...);
 * }
 *
 * The query will have a form like:
 *
 * struct entry *latch_query(struct latch_struct *latch, ...)
 * {
 *	struct entry *entry;
 *	unsigned seq, idx;
 *
 *	do {
 *		seq = lockless_dereference(latch->seq);

So here we have:

		seq = READ_ONCE(latch->seq);
		smp_read_barrier_depends();

Which is exactly what we want; the new code:

		seq = ({ p = READ_ONCE(latch);
			 smp_read_barrier_depends(); p })->seq;

is just wrong; because it looses the volatile read on seq, which can now
be torn or worse 'optimized'. And the read_depend barrier is also placed
wrong, we want it after the load of seq, to match the above data[]
up-to-date wmb()s.

Such that when we dereference latch->data[] below, we're guaranteed to
observe the right data.

 *
 *		idx = seq & 0x01;
 *		entry = data_query(latch->data[idx], ...);
 *
 *		smp_rmb();
 *	} while (seq != latch->seq);
 *
 *	return entry;
 * }

So yes, not passing a pointer is not pretty, but the code was correct,
and isn't anymore now.

Change to explicit READ_ONCE()+smp_read_barrier_depends() to avoid
confusion and allow strict lockless_dereference() checking.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Paul McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 50755bc1c3 ("seqlock: fix raw_read_seqcount_latch()")
Link: http://lkml.kernel.org/r/20160527111117.GL3192@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2016-06-03 08:37:25 +02:00
..
2016-03-04 23:50:27 -06:00
2016-05-17 12:29:34 -04:00
2016-03-15 16:55:16 -07:00
2016-02-16 13:04:58 -05:00
2016-05-31 22:12:43 +05:30
2016-05-17 15:48:12 -04:00
2016-04-06 13:18:19 +02:00
2016-03-11 14:58:58 -05:00
2016-02-11 09:59:22 -05:00
2016-03-22 15:36:02 -07:00
2016-05-11 19:31:40 -04:00
2016-05-11 22:37:54 +02:00
2016-05-02 09:32:04 -07:00
2016-03-22 15:36:02 -07:00
2016-03-22 15:36:02 -07:00
2016-03-09 15:43:42 -08:00
2016-05-03 15:07:35 -07:00
2016-05-19 19:12:14 -07:00
2016-05-26 15:35:44 -07:00
2016-05-27 15:57:31 -07:00
2016-05-10 23:56:28 -04:00
2016-05-17 15:48:12 -04:00
2016-05-17 15:47:55 -04:00
2016-05-23 17:04:14 -07:00
2016-04-25 15:09:11 -04:00
2016-02-16 13:04:58 -05:00
2016-05-02 09:00:56 -05:00
2016-05-23 17:04:14 -07:00
2016-05-19 19:12:14 -07:00
2016-02-11 18:35:48 -08:00
2016-02-11 18:35:48 -08:00
2016-03-14 15:43:11 -04:00
2016-02-16 13:04:58 -05:00
2016-03-17 15:09:34 -07:00
2016-03-22 15:36:02 -07:00
2016-03-22 15:36:02 -07:00
2016-03-22 15:36:02 -07:00
2016-05-26 00:04:58 -04:00
2016-05-26 15:35:44 -07:00
2016-05-19 19:12:14 -07:00
2016-05-19 19:12:14 -07:00
2016-05-08 23:46:14 -04:00
2016-05-20 17:58:30 -07:00
2016-05-19 19:12:14 -07:00
2016-05-23 17:04:14 -07:00
2016-04-07 16:53:29 -04:00
2016-05-20 17:58:30 -07:00
2016-02-22 16:10:08 -07:00