Subject: Re: Native disklabels, was: Partitioning&formatting...
To: Frederick Bruckman <fb@enteract.com>
From: Bill Studenmund <wrstuden@nas.nasa.gov>
List: port-mac68k
Date: 08/31/1999 15:51:33
On Wed, 25 Aug 1999, Frederick Bruckman wrote:

> On Wed, 25 Aug 1999, Hauke Fath wrote:
> 
> > ISTR that the kernel's disklabel write routines are #ifdef'd out. Once upon
> > a long ago when I started work on the iwm driver, I had patches that
> > enabled reading and writing BSD disklabels on mac68k. I even submitted them
> > to Scott but they either were not approved or got lost in his mailbox, who
> > knows.

Reading is fine.

> > Anyway: What is the general opinion on enabling native disklabels for
> > mac68k? We could always step back from writing if we found a  magic number
> > indicating a MacOS partition table.

I'm actually working on this for macppc. mac68k should be able to steal
the changes back easily.

> IIRC, the unresolved issue was exactly where to put them. Presently,
> "disklabel -r" writes to the upper half of block 0, which interferes
> with the LaCie driver. It seems to only put private settings there for
> use of the partitioning utility--the driver works fine as long as you
> don't try to launch the utility, which then demands that you
> re-install the driver.

Offset 64 seems ok. That's where macppc puts them..

> I'd consider putting it inside a proper Apple Partititon Map
> partition. That way gives you enough rope to hang yourself. :-) 
> In any case, the general consensus was that the i386 way, of looking
> for it in several different places, kind of blows.

That'd wrk too.

Take care,

Bill