[patch, so_7] : search_n optimization

[Paolo Carlini]

Paolo Carlini wrote:

>Agreed. And we already have a performance testcase, I will double check
>the numbers I get on a couple of modern machines.
>  
>
Chris, everything looks fine here and I'm seeing very good performance
numbers both on x86-linux and ia64-linux, almost 10x for the specific
testcase that we have got! In any case, I believe there is a large
evidence that the basic idea is sound.

I'm adding the patch to v7-branch, and I think a suited version can also
go in mainline. At your ease, please take care of that, thanks!

Paolo.

P.S. Unfortunately, only your name can go in the ChangeLog, because, to
my best knowledge, Dimitris never actually started the Copyright
assignment process. Anyway, you independently developed and submitted to
the list your own version of the algorithm, a lot of time ago, so...