diff --git a/include/asm-sh/atomic.h b/include/asm-sh/atomic.h index 8bdc1ba56f736570788c445bf5ef4f4dc1bde5a8..28305c3cbddf88d770f7c22020bec367121e74a2 100644 --- a/include/asm-sh/atomic.h +++ b/include/asm-sh/atomic.h @@ -28,11 +28,11 @@ static inline void atomic_add(int i, atomic_t *v) unsigned long tmp; __asm__ __volatile__ ( -"1: movli.l @%3, %0 ! atomic_add \n" -" add %2, %0 \n" -" movco.l %0, @%3 \n" +"1: movli.l @%2, %0 ! atomic_add \n" +" add %1, %0 \n" +" movco.l %0, @%2 \n" " bf 1b \n" - : "=&z" (tmp), "=r" (&v->counter) + : "=&z" (tmp) : "r" (i), "r" (&v->counter) : "t"); #else @@ -50,11 +50,11 @@ static inline void atomic_sub(int i, atomic_t *v) unsigned long tmp; __asm__ __volatile__ ( -"1: movli.l @%3, %0 ! atomic_sub \n" -" sub %2, %0 \n" -" movco.l %0, @%3 \n" +"1: movli.l @%2, %0 ! atomic_sub \n" +" sub %1, %0 \n" +" movco.l %0, @%2 \n" " bf 1b \n" - : "=&z" (tmp), "=r" (&v->counter) + : "=&z" (tmp) : "r" (i), "r" (&v->counter) : "t"); #else @@ -80,12 +80,12 @@ static inline int atomic_add_return(int i, atomic_t *v) #ifdef CONFIG_CPU_SH4A __asm__ __volatile__ ( -"1: movli.l @%3, %0 ! atomic_add_return \n" -" add %2, %0 \n" -" movco.l %0, @%3 \n" +"1: movli.l @%2, %0 ! atomic_add_return \n" +" add %1, %0 \n" +" movco.l %0, @%2 \n" " bf 1b \n" " synco \n" - : "=&z" (temp), "=r" (&v->counter) + : "=&z" (temp) : "r" (i), "r" (&v->counter) : "t"); #else @@ -109,12 +109,12 @@ static inline int atomic_sub_return(int i, atomic_t *v) #ifdef CONFIG_CPU_SH4A __asm__ __volatile__ ( -"1: movli.l @%3, %0 ! atomic_sub_return \n" -" sub %2, %0 \n" -" movco.l %0, @%3 \n" +"1: movli.l @%2, %0 ! atomic_sub_return \n" +" sub %1, %0 \n" +" movco.l %0, @%2 \n" " bf 1b \n" " synco \n" - : "=&z" (temp), "=r" (&v->counter) + : "=&z" (temp) : "r" (i), "r" (&v->counter) : "t"); #else @@ -186,11 +186,11 @@ static inline void atomic_clear_mask(unsigned int mask, atomic_t *v) unsigned long tmp; __asm__ __volatile__ ( -"1: movli.l @%3, %0 ! atomic_clear_mask \n" -" and %2, %0 \n" -" movco.l %0, @%3 \n" +"1: movli.l @%2, %0 ! atomic_clear_mask \n" +" and %1, %0 \n" +" movco.l %0, @%2 \n" " bf 1b \n" - : "=&z" (tmp), "=r" (&v->counter) + : "=&z" (tmp) : "r" (~mask), "r" (&v->counter) : "t"); #else @@ -208,11 +208,11 @@ static inline void atomic_set_mask(unsigned int mask, atomic_t *v) unsigned long tmp; __asm__ __volatile__ ( -"1: movli.l @%3, %0 ! atomic_set_mask \n" -" or %2, %0 \n" -" movco.l %0, @%3 \n" +"1: movli.l @%2, %0 ! atomic_set_mask \n" +" or %1, %0 \n" +" movco.l %0, @%2 \n" " bf 1b \n" - : "=&z" (tmp), "=r" (&v->counter) + : "=&z" (tmp) : "r" (mask), "r" (&v->counter) : "t"); #else