matthew green wrote:

>    Actually, in -current, we have page coloring now, so depending on how well
>    (if at all) the other kernel did that, and on a random factor depending on
>    the program's access pattern, NetBSD could conceivably run it _faster_.
> 
> 
> didn't simonb say that the fastest way to run a program on a decstation
> was to compile it under ultrix and run it under netbsd?  of course, we
> might have bloated beyond that now... :-)

Of course now that I don't have Ultrix boxes handy I just run the
Ultrix C compiler (really the MIPS C compiler, and an ancestor to
the DEC/Compaq Alpha C compiler I think) under Ultrix emulation on
my pmaxen.  :-)

Even compute bound programs were a percent or two faster running under
NetBSD than Ultrix - maybe less time spent in the kernel during clock
interupts or something?  And this was back in the 1.4ish days (well
before page coloring, which doesn't affect MIPS CPUs anyway).

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