> On Wed, 3 Nov 1999, NetBSD Bob wrote:
> 
> > What else can be discarded for kernel only building?  I suspect the
> > include and lib trees may be needed, for sys to compile, but the rest
> > can probably go?  What if any changes need to be made to the makefile?
> 
> NFS mounted filesystems are really handy for this sort of problem. :-) 
> If all you're doing is recompiling the kernel, I'm fairly certain you can
> get by with just the /usr/src/sys subtree.  It's about 68MB in size.
> You don't need to alter any makefiles to recompile the kernel unless
> you're going in and added new source files to kernel.

.....

OK, that means I will have enough room without scrounging up more
drivespace.

I was expecting that generic sort of compile but thought I would
check to be sure, just in case there were some strange bits that
needed to be pulled in from elsewhere in the tree.

.....

> > What is the drive size limit on the boot drive in NetBSD on a MicroVAX
> > 3100?  Can it handle the RZ26 or is that pushing the 1 gig limit too
> > closely?  The RZ26 may work, but I was thinking the RZ26-L was over
> > the edge?  Can someone clarify the limit exactly?
> 
> I'm using a 2GB Micropolis at the boot drive on one of my VAXstation 3100s
> and I've had no problems booting from it.  I know the VMS people have long
> mumbled about the 1GB disk limit for boot drives on the older SCSI VAXen
> like the early MicroVAX and VAXstation 3100 users, but I've yet to run
> into it under NetBSD/vax.

GREAT!  I kept hearing a 1 gig limit here and there, but if that is myth,
anymore, I will be a happy camper.  I was getting  ready to go scrounging
for RZ25's, but now I will hunt for bigger fodder.  Are there any
limitations going the other way at 8 gigs or such?

Now, tonight, it is off to see the Vizards, in the Vonderfull
Vorld of Voz.....(:+}}.....and make VAXbaby do her compiling kernel
thing.... (egads, all-night kernel compiles, again.... oh, well).
Vun, Vun, Vun.....(:+}}...

.....

> I think the VMSers are suffering from a bad case of tunnel vision imposed
> upon them by their OS :-)

VAXbaby had a VMS 5sometingorotherbiggieverbosityderivedkindofhtingie
installed on her, but I did not what to get mixed up with that tunnel
visioned verbosity whale.  BSD's are so much lighter and airy.....(:+}}...

Thanks All....

Bob

...postscript...

One thought for Ragge or the gurus in the list....

Last night I was unloading the source tarballs on VAXbaby and
noticed that periodically, maybe every 200-500 or so bytes of
directory listing out of the tarball, or about the same ascii
out of du, the SCSI I/O would come to a screeching halt to maybe
110 baud from 9600 baud.  It seemed to be the disk I/O, but I
am not sure.  Is there buffering or something that might be
enlarged a bit in the disk I/O to make the I/O flow a little
more transition free, speedwise?  It is very noticeable on
the VT100 serial output stream.  I have not checked it on a
monitor on the VS3100, but will try that later.   

Suggestions or comments are appreciated.

Thanks

Bob