memcpy(a,b,CONST) is not inlined by gcc 3.4.1 in Linux kernel

[Richard Guenther]

On Tue, 29 Mar 2005 17:37:06 +0300, Denis Vlasenko <vda@ilport.com.ua> wrote:
> Try testcase below the sig.
> 
> This causes nearly one thousand calls to memcpy in my kernel
> (not an allyesconfig one):

> 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);
> }

The question is, what reason does -Winline give for this inlining
decision?  And then
of course, how is the size estimate counted for the above.  What kind
of tree node do
we get for the ASM expression?

Richard.