Subject: Re: Compiler timings on varous MVII NetBSDs etc.
To: emanuel stiebler <>
From: NetBSD Bob <>
List: port-vax
Date: 01/23/2001 13:05:54
> > Then again, I have thought it might be fun to resurrect a 32V suite
> > on an MVII critter.... hmmmmm..... that suite flies like the wind,
> > I have read, but it is only patched for ONE particular VAX configuration.
> > But, it is is our historical BSD ancestor.
> Which machine was it ?

I am reunrolling the tarballs now, to confirm that.  What I thought I read
somewhere was that the 32V came out of London/Reiser's lab in AT&T(?), and
was the first VAX unix port.  I thought I read somewhere that it was heavily
hand hacked, and was supposed to be quite fast at the time on the VAX.
Berkeley got hold of it, and it formed the basis for the original BSD
ports.  All this is third hand or what I have read here and there.

Anyway, get this.... the kernel is 59528 bytes!  Talk about a lean
and mean VAX UNIX kernel!  The date stamp is March 22, 1979.

From the 32V setup manual:

     ``The distribution tape can be used only on a DEC
       VAX-11/780 with RP06 or RM03 disks and with TE16
       tape drives.''

A grep on VAX in the kernel code only pulls up 4 references to VAX.
A grep on VAX in the includes only pulls up 17 references to VAX.
A grep on VAX in the standalone code only pulls up 25 refereces to VAX.
A grep on VAX in the mdec code only pulls up 3 references to VAX.

It looks more like a straight PDP-11 V7 port than VAX.  Interesting.
Maybe the hacking was not all that substantial, after all!

That might not make it directly usable on many machines.  But, with
some hacks here and there, it might be workable on an MVII critter.
It would make a fun comparative system against say 4.3BSD and
NetBSD-1.2 and NetBSD-1.5.