> 
> >A lot of people have been asking for this recently, and I must admit I'm
> >a bit puzzled as to why we need our own disklabels.
> 
> As of about a year ago the way disklabels were constructed from the MacOS
> partition information had bugs in it.  If we could construct our own native
> disklabels we could bypass those bugs (if Allan hasn't fixed them already).
> In any case the translation between the formats was fixed, and I can
> immagine that someone might want it done differently from the default.

But aren't these OUR bugs? Since we're going to have to be able to read MacOS
partition tables, wouldn't it be better to fix the translation, rather than
leave it broken and try and make somethig else work right?

As I seem to recall, the problem had to do with someone having a LOT
of partitions on a drive, and the disklabel algorythm not being
reliable for many partitions (it stuffed in lots of MacOS partitions, and
ran out of partition slots before it got the desired partition). Or something
similar.

If we want to support lots of partitions, we should probably do something like
the Atari (or Amiga?) port did and go up to 16 partitons per drive. :-)

Take care,

Bill