[PATCH] m68knommu: optimize atomic macros

For some reason the asm versions of atomic_dec_and_test and friends were
masked by inefficient and big macros. With it I spare up to 8k in the text
size of my coldfire linux and 5k on my cpu32 version.

Patch originally submitted by Philippe De Muyter <phdm@macqel.be>
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

include/asm-m68knommu/atomic.h

@@ -20,36 +20,79 @@ typedef struct { int counter; } atomic_t;
 
 static __inline__ void atomic_add(int i, atomic_t *v)
 {
-	__asm__ __volatile__("addl %1,%0" : "=m" (*v) : "d" (i), "0" (*v));
+#ifdef CONFIG_COLDFIRE
+	__asm__ __volatile__("addl %1,%0" : "+m" (*v) : "d" (i));
+#else
+	__asm__ __volatile__("addl %1,%0" : "+m" (*v) : "di" (i));
+#endif
 }
 
 static __inline__ void atomic_sub(int i, atomic_t *v)
 {
-	__asm__ __volatile__("subl %1,%0" : "=m" (*v) : "d" (i), "0" (*v));
+#ifdef CONFIG_COLDFIRE
+	__asm__ __volatile__("subl %1,%0" : "+m" (*v) : "d" (i));
+#else
+	__asm__ __volatile__("subl %1,%0" : "+m" (*v) : "di" (i));
+#endif
+}
+
+static __inline__ int atomic_sub_and_test(int i, atomic_t * v)
+{
+	char c;
+#ifdef CONFIG_COLDFIRE
+	__asm__ __volatile__("subl %2,%1; seq %0"
+			     : "=d" (c), "+m" (*v)
+			     : "d" (i));
+#else
+	__asm__ __volatile__("subl %2,%1; seq %0"
+			     : "=d" (c), "+m" (*v)
+			     : "di" (i));
+#endif
+	return c != 0;
 }
 
 static __inline__ void atomic_inc(volatile atomic_t *v)
 {
-	__asm__ __volatile__("addql #1,%0" : "=m" (*v): "0" (*v));
+	__asm__ __volatile__("addql #1,%0" : "+m" (*v));
+}
+
+/*
+ * atomic_inc_and_test - increment and test
+ * @v: pointer of type atomic_t
+ *
+ * Atomically increments @v by 1
+ * and returns true if the result is zero, or false for all
+ * other cases.
+ */
+
+static __inline__ int atomic_inc_and_test(volatile atomic_t *v)
+{
+	char c;
+	__asm__ __volatile__("addql #1,%1; seq %0" : "=d" (c), "+m" (*v));
+	return c != 0;
 }
 
 static __inline__ void atomic_dec(volatile atomic_t *v)
 {
-	__asm__ __volatile__("subql #1,%0" : "=m" (*v): "0" (*v));
+	__asm__ __volatile__("subql #1,%0" : "+m" (*v));
 }
 
 static __inline__ int atomic_dec_and_test(volatile atomic_t *v)
 {
 	char c;
-	__asm__ __volatile__("subql #1,%1; seq %0" : "=d" (c), "=m" (*v): "1" (*v));
+	__asm__ __volatile__("subql #1,%1; seq %0" : "=d" (c), "+m" (*v));
 	return c != 0;
 }
 
-#define atomic_clear_mask(mask, v) \
-	__asm__ __volatile__("andl %1,%0" : "=m" (*v) : "id" (~(mask)),"0"(*v))
+static __inline__ void atomic_clear_mask(unsigned long mask, unsigned long *v)
+{
+	__asm__ __volatile__("andl %1,%0" : "+m" (*v) : "id" (~(mask)));
+}
 
-#define atomic_set_mask(mask, v) \
-	__asm__ __volatile__("orl %1,%0" : "=m" (*v) : "id" (mask),"0"(*v))
+static __inline__ void atomic_set_mask(unsigned long mask, unsigned long *v)
+{
+	__asm__ __volatile__("orl %1,%0" : "+m" (*v) : "id" (mask));
+}
 
 /* Atomic operations are already serializing */
 #define smp_mb__before_atomic_dec()    barrier()
@@ -88,17 +131,4 @@ extern __inline__ int atomic_sub_return(int i, atomic_t * v)
 #define atomic_dec_return(v) atomic_sub_return(1,(v))
 #define atomic_inc_return(v) atomic_add_return(1,(v))
 
-/*
- * atomic_inc_and_test - increment and test
- * @v: pointer of type atomic_t
- *
- * Atomically increments @v by 1
- * and returns true if the result is zero, or false for all
- * other cases.
- */
-#define atomic_inc_and_test(v) (atomic_inc_return(v) == 0)
-
-#define atomic_sub_and_test(i,v) (atomic_sub_return((i), (v)) == 0)
-#define atomic_dec_and_test(v) (atomic_sub_return(1, (v)) == 0)
-
 #endif /* __ARCH_M68KNOMMU_ATOMIC __ */