Subject: Re: First attempt at a VAX -current build
To: Greg Oster <oster@cs.usask.ca>
From: Brian Chase <vaxzilla@jarai.org>
List: port-vax
Date: 12/09/2003 23:03:35
On Tue, 9 Dec 2003, Greg Oster wrote:

> edit the Makefile for xxboot, and move "NOMAN= #define" to *before*
> the line: .include <bsd.own.mk>
> Do the same thing for xxboot_ustarfs in src/sys/arch/vax/boot/
> xxboot_ustarfs.  (Thanks to Matt Thomas for the suggestion)
>
> With those changes I've managed to get a successfull cross-build to
> complete.  (Full build command was:  ./build.sh -U -u -o -m vax  \\
> -V CHECKFLIST_FLAGS="-e -m" -D /u1/builds/build18/install \\
> -R /u1/builds/build18/release -T /u1/builds/build18/tools release
> )

I guess I'll have to wait until tomorrow morning to play with the
results.

> Now to test the binaries :-}

I'd be curious to know what your results are (and what type of system(s)
you're running the binaries on).  The binaries from my first build seem
to work just fine on my VLC, but when I run the same on a VAXstation
3100/M76 and my MicroVAX 3100/10e, the kernel panics and drops into the
debugger.  This happens during the boot process just as the filesystems
are checked.  I've also tried booting up to single user mode, which
works, but if I manually run an fsck the system panics in the same way.

If I then boot the systems from a secondary 1.6.1 drive, I find that the
superblock on the -current drive's filesystem is corrupt.  The fsck
easily fixes it with one of the other superblock copies, but this is a
bit weird.

I'm not quite sure what to make of it yet, and I figured I'd wait until
I got this second pass compiled before I started categorizing the
problem further.

-brian.