[tip:locking/urgent] bitops: Do not default to __clear_bit() for __clear_bit_unlock()

From: tip-bot for Peter Zijlstra
Date: Mon Mar 21 2016 - 07:17:58 EST

Commit-ID: f75d48644c56a31731d17fa693c8175328957e1d
Gitweb: http://git.kernel.org/tip/f75d48644c56a31731d17fa693c8175328957e1d
Author: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
AuthorDate: Wed, 9 Mar 2016 12:40:54 +0100
Committer: Ingo Molnar <mingo@xxxxxxxxxx>
CommitDate: Mon, 21 Mar 2016 10:50:48 +0100

bitops: Do not default to __clear_bit() for __clear_bit_unlock()

__clear_bit_unlock() is a special little snowflake. While it carries the
non-atomic '__' prefix, it is specifically documented to pair with
test_and_set_bit() and therefore should be 'somewhat' atomic.

Therefore the generic implementation of __clear_bit_unlock() cannot use
the fully non-atomic __clear_bit() as a default.

If an arch is able to do better; is must provide an implementation of
__clear_bit_unlock() itself.

Specifically, this came up as a result of hackbench livelock'ing in
slab_lock() on ARC with SMP + SLUB + !LLSC.

The issue was incorrect pairing of atomic ops.

slab_lock() -> bit_spin_lock() -> test_and_set_bit()
slab_unlock() -> __bit_spin_unlock() -> __clear_bit()

The non serializing __clear_bit() was getting "lost"

80543b8e: ld_s r2,[r13,0] <--- (A) Finds PG_locked is set
80543b90: or r3,r2,1 <--- (B) other core unlocks right here
80543b94: st_s r3,[r13,0] <--- (C) sets PG_locked (overwrites unlock)

Fixes ARC STAR 9000817404 (and probably more).

Reported-by: Vineet Gupta <Vineet.Gupta1@xxxxxxxxxxxx>
Tested-by: Vineet Gupta <Vineet.Gupta1@xxxxxxxxxxxx>
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Cc: Christoph Lameter <cl@xxxxxxxxx>
Cc: David Rientjes <rientjes@xxxxxxxxxx>
Cc: Helge Deller <deller@xxxxxx>
Cc: James E.J. Bottomley <jejb@xxxxxxxxxxxxxxxx>
Cc: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Noam Camus <noamc@xxxxxxxxxx>
Cc: Paul E. McKenney <paulmck@xxxxxxxxxxxxxxxxxx>
Cc: Pekka Enberg <penberg@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
Link: http://lkml.kernel.org/r/20160309114054.GJ6356@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
include/asm-generic/bitops/lock.h | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/include/asm-generic/bitops/lock.h b/include/asm-generic/bitops/lock.h
index c30266e..8ef0ccb 100644
--- a/include/asm-generic/bitops/lock.h
+++ b/include/asm-generic/bitops/lock.h
@@ -29,16 +29,16 @@ do { \
* @nr: the bit to set
* @addr: the address to start counting from
- * This operation is like clear_bit_unlock, however it is not atomic.
- * It does provide release barrier semantics so it can be used to unlock
- * a bit lock, however it would only be used if no other CPU can modify
- * any bits in the memory until the lock is released (a good example is
- * if the bit lock itself protects access to the other bits in the word).
+ * A weaker form of clear_bit_unlock() as used by __bit_lock_unlock(). If all
+ * the bits in the word are protected by this lock some archs can use weaker
+ * ops to safely unlock.
+ *
+ * See for example x86's implementation.
#define __clear_bit_unlock(nr, addr) \
do { \
- smp_mb(); \
- __clear_bit(nr, addr); \
+ smp_mb__before_atomic(); \
+ clear_bit(nr, addr); \
} while (0)