Mark Brinicombe said:
> On Sun, 31 May 1998, Chris G. Demetriou wrote:
> 
> > > The arm32/arm32/disksubr.c file does not appear to have any support for
> > > DOS partition tables, despite the comments in the file. I seem to
> > > remember a mention of this on the mailing list before, and had thought
> > > that the latest kernels would support DOS partition tables.
> > 
> > I sent diffs that included support for them to Mark Brinicombe a while
> > (probably a month or two) ago.  I think he ran into a bug or two with
> > them, and it was right before he was going on a trip or to a show or
> > something, and i've been too busy to follow up on it...
> [ ... ]
> 
> I have kernels here with your patches in. I have not had a chance to
> retest them with my NT disk again. 
> I was not sure if these were ready to commit. You mentioned something
> about ideally the code being MI (I've already found needs to do the
> reverse i.e. mount Acorn disks on a i386 box).
> Was the plan to make this MI then commit it or rather commit for the
> arm32 and add a MI version to the wish list ;-)

I'd say "commit it for the arm32 (when it's known to work) and add an
MI version to the wish list."

Actually, i think the code as-is can be used by multiple ports (I
think i have it running on my i386 laptop, after minor mods required
by the renaming of some of the constants), and would expect that after
the ARM port starts using that it that at least some other ports would
pick it up before too long.  At that point, the question becomes one
of "where does the code go, and how is it best used," and that's not a
problem I want to solve."



chris