OSDev.org

Hi,

From ManRix OS I've found someting interesting:

viki wrote:Hi,

From ManRix OS I've found someting interesting:

Code: Select all

static inline void * memcpy(void * to, const void * from, size_t n)
{
int d0, d1, d2;
__asm__ __volatile__(
	"rep ; movsl\n\t"
	"testb $2,%b4\n\t"
	"je 1f\n\t"
	"movsw\n"
	"1:\ttestb $1,%b4\n\t"
	"je 2f\n\t"
	"movsb\n"
	"2:"
	: "=&c" (d0), "=&D" (d1), "=&S" (d2)
	:"0" (n/4), "q" (n),"1" ((long) to),"2" ((long) from)
	: "memory");
return (to);
} 

Have you read anything at all of this thread? These optimizations are BAD. As in the big ol', Bad Thing™, you know?

I know that optimalization is wrong. But. If you dont know how the result should looks like after compilation then you are not able to set correct flags for your compiler. GCC for example has not only -O2 option that improve our code speed.