sh: Fix sh4a llsc-based cmpxchg()

This fixes up a typo in the ll/sc based cmpxchg code which apparently
wasn't getting a lot of testing due to the swapped old/new pair. With
that fixed up, the ll/sc code also starts using it and provides its own
atomic_add_unless().

Signed-off-by: Aoi Shinkai <shinkoi2005@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
diff --git a/arch/sh/include/asm/spinlock.h b/arch/sh/include/asm/spinlock.h
index 6028356..a28c9f0 100644
--- a/arch/sh/include/asm/spinlock.h
+++ b/arch/sh/include/asm/spinlock.h
@@ -26,7 +26,7 @@
 #define __raw_spin_is_locked(x)		((x)->lock <= 0)
 #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
 #define __raw_spin_unlock_wait(x) \
-	do { cpu_relax(); } while ((x)->lock)
+	do { while (__raw_spin_is_locked(x)) cpu_relax(); } while (0)
 
 /*
  * Simple spin lock operations.  There are two variants, one clears IRQ's