[RFC PATCH 3/5] asm-generic/bitops/atomic.h: Rewrite using atomic_fetch_*

Peter Zijlstra peterz at infradead.org
Fri Feb 16 02:21:00 PST 2018


On Thu, Feb 15, 2018 at 06:20:49PM +0000, Will Deacon wrote:
> On Thu, Feb 15, 2018 at 06:08:47PM +0100, Peter Zijlstra wrote:
> > On Thu, Feb 15, 2018 at 03:29:33PM +0000, Will Deacon wrote:
> > > +static inline void __clear_bit_unlock(unsigned int nr,
> > > +				      volatile unsigned long *p)
> > > +{
> > > +	unsigned long old;
> > >  
> > > +	p += BIT_WORD(nr);
> > > +	old = READ_ONCE(*p);
> > > +	old &= ~BIT_MASK(nr);
> > > +	smp_store_release(p, old);
> > 
> > This should be atomic_set_release() I think, for the special case where
> > atomics are implemented with spinlocks, see the 'fun' case in
> > Documentation/atomic_t.txt.
> 
> My understanding of __clear_bit_unlock is that there is guaranteed to be
> no concurrent accesses to the same word, so why would it matter whether
> locks are used to implement atomics?


commit f75d48644c56a31731d17fa693c8175328957e1d
Author: Peter Zijlstra <peterz at infradead.org>
Date:   Wed Mar 9 12:40:54 2016 +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 at synopsys.com>
    Tested-by: Vineet Gupta <Vineet.Gupta1 at synopsys.com>
    Signed-off-by: Peter Zijlstra (Intel) <peterz at infradead.org>
    Cc: Andrew Morton <akpm at linux-foundation.org>
    Cc: Christoph Lameter <cl at linux.com>
    Cc: David Rientjes <rientjes at google.com>
    Cc: Helge Deller <deller at gmx.de>
    Cc: James E.J. Bottomley <jejb at parisc-linux.org>
    Cc: Joonsoo Kim <iamjoonsoo.kim at lge.com>
    Cc: Linus Torvalds <torvalds at linux-foundation.org>
    Cc: Noam Camus <noamc at ezchip.com>
    Cc: Paul E. McKenney <paulmck at linux.vnet.ibm.com>
    Cc: Pekka Enberg <penberg at kernel.org>
    Cc: Peter Zijlstra <peterz at infradead.org>
    Cc: Thomas Gleixner <tglx at linutronix.de>
    Cc: stable at vger.kernel.org
    Link: http://lkml.kernel.org/r/20160309114054.GJ6356@twins.programming.kicks-ass.net
    Signed-off-by: Ingo Molnar <mingo at kernel.org>

diff --git a/include/asm-generic/bitops/lock.h b/include/asm-generic/bitops/lock.h
index c30266e94806..8ef0ccbf8167 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)
 
 #endif /* _ASM_GENERIC_BITOPS_LOCK_H_ */



More information about the linux-arm-kernel mailing list