Re: Re: musl libc, memcpy

[Rich Felker <dalias@aerifal.cx>]

[-- Attachment #1: Type: text/plain, Size: 417 bytes --]

On Wed, Aug 01, 2012 at 01:40:11AM -0400, Rich Felker wrote:
> On Wed, Aug 01, 2012 at 12:27:22AM -0400, Rich Felker wrote:
> > I'm attaching a (possibly buggy; not heavily tested) rep-movsd-based
> > version. I'd be interested in hearing how it performs.
> 
> And here is the attachment...

And here's a version that might be faster; reportedly, rep movsd works
better when the destination address is aligned.

Rich

[-- Attachment #2: memcpy.s --]
[-- Type: text/plain, Size: 341 bytes --]

.global xmemcpy
.type xmemcpy,@function
xmemcpy:
	push %esi
	push %edi
	mov 12(%esp),%edi
	mov 16(%esp),%esi
	mov 20(%esp),%ecx
	mov %edi,%eax
	cmp $4,%ecx
	jc 1f
	test $3,%edi
	jz 1f
2:	movsb
	dec %ecx
	test $3,%edi
	jnz 2b
1:	mov %ecx,%edx
	shr $2,%ecx
	rep
	movsl
	and $3,%edx
	jz 1f
2:	movsb
	dec %edx
	jnz 2b
1:	pop %edi
	pop %esi
	ret