> > Over the weekend I did some simple compiler tests on the NetBSDs
> > and other OS's I have up on my MicroVAX II critters.
> >
> > [...]
> >
> > OS                           Gkermit compile time (m:s)
> > --------------------------   --------------------------
> > 4.3BSD-Tahoe PCC             3:01
> > Ultrix 4.2 VAX-C             2.50
> > NetBSD-1.2 gcc 2.72          5:05
> > NetBSD-1.3 gcc 2.72.1        6:07
> > NetBSD-1.4.3 gcc 2.91        7:08
> > NetBSD-1.5 gcc 2.91          7:40
> > --------------------------   --------------------------
> 
> This comparison info is very handy.  It is at least good to see (well, in
> this trival case) that NetBSD-1.4.3 and NetBSD-1.5 w/gcc 2.91 seem to be
> about on par with eachother.

Well, yes and no.  They are supposedly the same compiler.  Yet,
there is a 15% slowdown in 1.5 relative to 1.4.3.  If we assume
that the compiler is the same, and the kernels are the same, then
that points to some bloat in the 1.5, and it being a system problem
and not just a compiler problem.

.....

> > Recompiling stripped kernels did not help anything speedwise.
> > All kernel configs were identical from 4.3BSD through NetBSD-1.5.
> > They were had stripped down to the bare minimum needed to run,
> > with no extra features enabled.
> 
> IIRC, the earlier non-stripped kernel compile benchmarks you posted showed
> a much greater performance discrepancy between NetBSD 1.4.3 and 1.5.
> Maybe that's still the case when compiling something intensive like the
> kernel where the compile processes will go into swap more often.

Well, I chose something simple like gkermit.  Are there any other
things that one might in particular compile that would show more
effects of small ram or OS bottlenecks?

It is not that any one thing is the culprit, but, collectively,
it all is adding up to a slowdown on the MVII class VAXencritters.

Thanks

Bob