Microoptimization to allow the compiler to evaluate ntohl() etc on

known constants at compile time rather than at run time.  We have a number
of nasty hacks around the place to cache ntohl() of constants (eg: nfs).
This change allows the compiler to compile-time evaluate ntohl(1) as
0x01000000 rather than having to emit assembler code to do it.  This
has other smaller flow-on effects because the compiler can see that
ntohl(constant) itself has a constant value now and can propagate the
compile time evaluation.

Obtained from:  Ideas from NetBSD and Linux, and some code from NetBSD

1 files changed, 49 insertions, 4 deletions

diff --git a/sys/i386/include/endian.h b/sys/i386/include/endian.h
index 3ee77f5..dc924b1 100644
--- a/sys/i386/include/endian.h
+++ b/sys/i386/include/endian.h
@@ -69,31 +69,76 @@
 
 #ifdef __GNUC__
 
-#define __word_swap_int(x) \
+#define __word_swap_int_var(x) \
 __extension__ ({ register __uint32_t __X = (x); \
    __asm ("rorl $16, %0" : "+r" (__X)); \
    __X; })
 
+#ifdef __OPTIMIZE__
+
+#define	__word_swap_int_const(x) \
+	((((x) & 0xffff0000) >> 16) | \
+	 (((x) & 0x0000ffff) << 16))
+#define	__word_swap_int(x) (__builtin_constant_p(x) ? \
+	__word_swap_int_const(x) : __word_swap_int_var(x))
+
+#else	/* __OPTIMIZE__ */
+
+#define	__word_swap_int(x) __word_swap_int_var(x)
+
+#endif	/* __OPTIMIZE__ */
+
 #if defined(_KERNEL) && (defined(I486_CPU) || defined(I586_CPU) || defined(I686_CPU)) && !defined(I386_CPU)
 
-#define __byte_swap_int(x) \
+#define __byte_swap_int_var(x) \
 __extension__ ({ register __uint32_t __X = (x); \
    __asm ("bswap %0" : "+r" (__X)); \
    __X; })
 #else
 
-#define __byte_swap_int(x) \
+#define __byte_swap_int_var(x) \
 __extension__ ({ register __uint32_t __X = (x); \
    __asm ("xchgb %h0, %b0\n\trorl $16, %0\n\txchgb %h0, %b0" \
        : "+q" (__X)); \
    __X; })
 #endif
 
-#define __byte_swap_word(x) \
+#ifdef __OPTIMIZE__
+
+#define	__byte_swap_int_const(x) \
+	((((x) & 0xff000000) >> 24) | \
+	 (((x) & 0x00ff0000) >>  8) | \
+	 (((x) & 0x0000ff00) <<  8) | \
+	 (((x) & 0x000000ff) << 24))
+#define	__byte_swap_int(x) (__builtin_constant_p(x) ? \
+	__byte_swap_int_const(x) : __byte_swap_int_var(x))
+
+#else	/* __OPTIMIZE__ */
+
+#define	__byte_swap_int(x) __byte_swap_int_var(x)
+
+#endif	/* __OPTIMIZE__ */
+
+#define __byte_swap_word_var(x) \
 __extension__ ({ register __uint16_t __X = (x); \
    __asm ("xchgb %h0, %b0" : "+q" (__X)); \
    __X; })
 
+#ifdef __OPTIMIZE__
+
+#define	__byte_swap_word_const(x) \
+	((((x) & 0xff00) >> 8) | \
+	 (((x) & 0x00ff) << 8))
+
+#define	__byte_swap_word(x) (__builtin_constant_p(x) ? \
+	__byte_swap_word_const(x) : __byte_swap_word_var(x))
+
+#else	/* __OPTIMIZE__ */
+
+#define	__byte_swap_word(x) __byte_swap_word_var(x)
+
+#endif	/* __OPTIMIZE__ */
+
 static __inline __uint64_t
 __bswap64(__uint64_t _x)
 {