Re: [PATCH] Modernize i386 string.h

[Andi Kleen]

On Mon, Jan 19, 2004 at 04:49:54PM -0800, Richard Henderson wrote:
> On Sun, Jan 18, 2004 at 09:09:19PM +0100, Andi Kleen wrote:
> > +#define __HAVE_ARCH_MEMCPY 1
> > +extern void *__memcpy(void *to, const void *from, size_t len); 
> > +#define memcpy(dst,src,len) \
> > +	({ size_t __len = (len);				\
> > +	   void *__ret;						\
> > +	   if (__builtin_constant_p(len) && __len >= 128)	\
> > +		 __ret = __memcpy((dst),(src),__len);		\
> > +	   else							\
> > +		 __ret = __builtin_memcpy((dst),(src),__len);	\
> > +	   __ret; }) 
> 
> Why not just __builtin_memcpy?  Or indeed, why bother defining
> anything at all, since the compiler will infer __builtin_memcpy
> from the external symbol memcpy.

I was considering that when I moved the code over from x86-64.

On x86-64 (in gcc 3.1 timeframe) I did it originally this way 
because memcpy didn't have this "use external for big copies" logic.
I wasn't sure if gcc 3.3 handles it now correctly. If you can
confirm that it does I will happily remove it.

-Andi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/