Re: memove performance of NetBSD

[Thor Lancelot Simon <tls%panix.com@localhost>]

On Thu, Jan 28, 2010 at 11:56:09AM +0530, Channa wrote:
> 
> Yes the performance improves when the memmove is done as below:
> 
> memmove(dst, src+1, sizeof(src));
> memmove(dst, src+2, sizeof(src));
> memmove(dst, src+3, sizeof(src));

That's a pretty weird special case, since the start addresses overlap.
What if you use more realistic start addresses?

> If I remove the hand unrolling of the loop (as done in the patch) the
> performance does not improve.

It might not be a win if the addresses don't all fall in the same cache
line...

Thor