In comp.sys.hp.hpux Lars Bausch <lars.bausch@dotsch.de> wrote:

I had some problems that some additional headers must included in
some files, some problems in the code. 

Sounds like some run of the mill porting stuff.

But the developer from seti and boinc gave me great help. And so I
have the binaries for PARISC and the test workunit for HPUX is
running.

Good.

Btw. do you know some performance optimizing hints ? - I had
compiled it with gcc 3.3.1 and -O2. Is -O3 much better performant ?

YMMV :) I'd go ahead and try it and see what it changes for the test
workunit.

Is the HP Ansi C better in performance like gcc ?

Well, HP uses HP ANSI C and not gcc for SPECint :) I would strongly
suspect that code generation for PA-RISC is better in the HP compilers
than in gcc, although I suppose that since the PA-RISC linux stuff gcc
is probably better than it was before.

They way to see of course is to compile with the HP ANSI C compiler
and make the comparison.  Have the seti and boinc folks turned the
changes around?  I could probably try to find a couple spare cycles to
compile with the HP ANSI C compiler on 11.11.  Doubt that I have the
same machine, and I've no plans to put gcc on my PA-RISC system, but I
could email you a binary.

rick jones

BTW, at some point we should pick a single group for the followups.
My strawman would be comp.sys.hp.hpux since that is the one I read
regularly :)
-- 
oxymoron n, Hummer H2 with California Save Our Coasts and Oceans plates
these opinions are mine, all mine; HP might not want them anyway... :)
feel free to post, OR email to raj in cup.hp.com  but NOT BOTH...