x86: unify prefetch operations

This patch moves the prefetch[w]? functions to processor.h
Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

include/asm-x86/processor.h

@@ -596,6 +596,36 @@ extern char ignore_fpu_irq;
 #define ARCH_HAS_PREFETCHW
 #define ARCH_HAS_SPINLOCK_PREFETCH
 
+#ifdef CONFIG_X86_32
+#define BASE_PREFETCH	ASM_NOP4
+#define ARCH_HAS_PREFETCH
+#else
+#define BASE_PREFETCH	"prefetcht0 (%1)"
+#endif
+
+/* Prefetch instructions for Pentium III and AMD Athlon */
+/* It's not worth to care about 3dnow! prefetches for the K6
+   because they are microcoded there and very slow.
+   However we don't do prefetches for pre XP Athlons currently
+   That should be fixed. */
+static inline void prefetch(const void *x)
+{
+	alternative_input(BASE_PREFETCH,
+			  "prefetchnta (%1)",
+			  X86_FEATURE_XMM,
+			  "r" (x));
+}
+
+/* 3dnow! prefetch to get an exclusive cache line. Useful for
+   spinlocks to avoid one state transition in the cache coherency protocol. */
+static inline void prefetchw(const void *x)
+{
+	alternative_input(BASE_PREFETCH,
+			  "prefetchw (%1)",
+			  X86_FEATURE_3DNOW,
+			  "r" (x));
+}
+
 #define spin_lock_prefetch(x)	prefetchw(x)
 /* This decides where the kernel will search for a free chunk of vm
  * space during mmap's.

include/asm-x86/processor_32.h

@@ -228,29 +228,4 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
 
 #define ASM_NOP_MAX 8
 
-/* Prefetch instructions for Pentium III and AMD Athlon */
-/* It's not worth to care about 3dnow! prefetches for the K6
-   because they are microcoded there and very slow.
-   However we don't do prefetches for pre XP Athlons currently
-   That should be fixed. */
-static inline void prefetch(const void *x)
-{
-	alternative_input(ASM_NOP4,
-			  "prefetchnta (%1)",
-			  X86_FEATURE_XMM,
-			  "r" (x));
-}
-
-#define ARCH_HAS_PREFETCH
-
-/* 3dnow! prefetch to get an exclusive cache line. Useful for 
-   spinlocks to avoid one state transition in the cache coherency protocol. */
-static inline void prefetchw(const void *x)
-{
-	alternative_input(ASM_NOP4,
-			  "prefetchw (%1)",
-			  X86_FEATURE_3DNOW,
-			  "r" (x));
-}
-
 #endif /* __ASM_I386_PROCESSOR_H */

include/asm-x86/processor_64.h

@@ -124,12 +124,4 @@ DECLARE_PER_CPU(struct orig_ist, orig_ist);
 
 #define ASM_NOP_MAX 8
 
-static inline void prefetchw(void *x) 
-{ 
-	alternative_input("prefetcht0 (%1)",
-			  "prefetchw (%1)",
-			  X86_FEATURE_3DNOW,
-			  "r" (x));
-} 
-
 #endif /* __ASM_X86_64_PROCESSOR_H */