4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Andrea Parri parri.andrea@gmail.com
[ Upstream commit 472e8c55cf6622d1c112dc2bc777f68bbd4189db ]
Successful RMW operations are supposed to be fully ordered, but Alpha's xchg() and cmpxchg() do not meet this requirement.
Will Deacon noticed the bug:
So MP using xchg:
WRITE_ONCE(x, 1) xchg(y, 1)
smp_load_acquire(y) == 1 READ_ONCE(x) == 0
would be allowed.
... which thus violates the above requirement.
Fix it by adding a leading smp_mb() to the xchg() and cmpxchg() implementations.
Reported-by: Will Deacon will.deacon@arm.com Signed-off-by: Andrea Parri parri.andrea@gmail.com Acked-by: Paul E. McKenney paulmck@linux.vnet.ibm.com Cc: Alan Stern stern@rowland.harvard.edu Cc: Andrew Morton akpm@linux-foundation.org Cc: Ivan Kokshaysky ink@jurassic.park.msu.ru Cc: Linus Torvalds torvalds@linux-foundation.org Cc: Matt Turner mattst88@gmail.com Cc: Peter Zijlstra peterz@infradead.org Cc: Richard Henderson rth@twiddle.net Cc: Thomas Gleixner tglx@linutronix.de Cc: linux-alpha@vger.kernel.org Link: http://lkml.kernel.org/r/1519291488-5752-1-git-send-email-parri.andrea@gmail... Signed-off-by: Ingo Molnar mingo@kernel.org Signed-off-by: Sasha Levin alexander.levin@microsoft.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- arch/alpha/include/asm/xchg.h | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-)
--- a/arch/alpha/include/asm/xchg.h +++ b/arch/alpha/include/asm/xchg.h @@ -11,6 +11,10 @@ * Atomic exchange. * Since it can be used to implement critical sections * it must clobber "memory" (also for interrupts in UP). + * + * The leading and the trailing memory barriers guarantee that these + * operations are fully ordered. + * */
static inline unsigned long @@ -18,6 +22,7 @@ ____xchg(_u8, volatile char *m, unsigned { unsigned long ret, tmp, addr64;
+ smp_mb(); __asm__ __volatile__( " andnot %4,7,%3\n" " insbl %1,%4,%1\n" @@ -42,6 +47,7 @@ ____xchg(_u16, volatile short *m, unsign { unsigned long ret, tmp, addr64;
+ smp_mb(); __asm__ __volatile__( " andnot %4,7,%3\n" " inswl %1,%4,%1\n" @@ -66,6 +72,7 @@ ____xchg(_u32, volatile int *m, unsigned { unsigned long dummy;
+ smp_mb(); __asm__ __volatile__( "1: ldl_l %0,%4\n" " bis $31,%3,%1\n" @@ -86,6 +93,7 @@ ____xchg(_u64, volatile long *m, unsigne { unsigned long dummy;
+ smp_mb(); __asm__ __volatile__( "1: ldq_l %0,%4\n" " bis $31,%3,%1\n" @@ -127,9 +135,12 @@ ____xchg(, volatile void *ptr, unsigned * store NEW in MEM. Return the initial value in MEM. Success is * indicated by comparing RETURN with OLD. * - * The memory barrier is placed in SMP unconditionally, in order to - * guarantee that dependency ordering is preserved when a dependency - * is headed by an unsuccessful operation. + * The leading and the trailing memory barriers guarantee that these + * operations are fully ordered. + * + * The trailing memory barrier is placed in SMP unconditionally, in + * order to guarantee that dependency ordering is preserved when a + * dependency is headed by an unsuccessful operation. */
static inline unsigned long @@ -137,6 +148,7 @@ ____cmpxchg(_u8, volatile char *m, unsig { unsigned long prev, tmp, cmp, addr64;
+ smp_mb(); __asm__ __volatile__( " andnot %5,7,%4\n" " insbl %1,%5,%1\n" @@ -164,6 +176,7 @@ ____cmpxchg(_u16, volatile short *m, uns { unsigned long prev, tmp, cmp, addr64;
+ smp_mb(); __asm__ __volatile__( " andnot %5,7,%4\n" " inswl %1,%5,%1\n" @@ -191,6 +204,7 @@ ____cmpxchg(_u32, volatile int *m, int o { unsigned long prev, cmp;
+ smp_mb(); __asm__ __volatile__( "1: ldl_l %0,%5\n" " cmpeq %0,%3,%1\n" @@ -214,6 +228,7 @@ ____cmpxchg(_u64, volatile long *m, unsi { unsigned long prev, cmp;
+ smp_mb(); __asm__ __volatile__( "1: ldq_l %0,%5\n" " cmpeq %0,%3,%1\n"