Re: [PATCH] x86-64: memset optimization

[Stephen Hemminger <shemminger@linux-foundation.org>]

On Sat, 18 Aug 2007 20:55:11 +0200
Andi Kleen <ak@suse.de> wrote:

> 
> > The problem is that on x86-64 you are overriding memset() 
> 
> I don't.  You must be looking at old source
> 
> asm-x86_64/string.h 2.6.23rc3:
> 
> #define __HAVE_ARCH_MEMSET
> void *memset(void *s, int c, size_t n);
> 
> I wanted to do the same on i386 too, but there were some minor obstacles.
> The problem is that the out of line fallback i386 memset is currently
> quite dumb and needs to be rewritten to expand the fill char on its
> own like the x86-64 version. Probably best would be just to port
> the x86-64 version. I just hadn't had time for that.
> 
> [Patches welcome, but if you do ask me for my old memset test harness]
> 
> -Andi

I am looking at current source, built with current (non-experimental) GCC
from Fedora Core 7. If I dissassemble ether_setup, which is

void ether_setup(struct net_device *dev)
{
...

	memset(dev->broadcast, 0xFF, ETH_ALEN);
}

I see a tail recursion (jmp) to memset which is the code in arch/x86_64/lib/memset.S