"Jesper Louis Andersen" wrote:

> Quoting Chuck Yerkes (chuck+nbsd@2003.snew.com):
> 
> > Yeah, and that's a big issue.
> > 1) it's for linux
> > 2) the license, afair, prevents its inclusion
> > 3) it runs on 1 of the 500 some platforms that netbsd runs on ;)
> 
> In the vain hope its extended warnings could ligthen up some dark
> spots in the code, it might be worth a try to use it on the tree. The
> question here would not be inclusion but rather tree-pruning.

I've used an older version of it to fix some errors in the NetBSD
kernel.  Unfortunately the resulting kernel didn't run for reasons
that I didn't track down at the time.

Newer versions of icc should be able to build NetBSD userland.  I
did get md5 working once, and it was something in the order of 25-30%
faster than an md5 built with gcc 2.95.3 for it's speed check mode
(it's -t option) on a 700MHz PIII that I was using at the time.

Simon.
--
Simon Burge                            <simonb@wasabisystems.com>
NetBSD Support and Service:         http://www.wasabisystems.com/