Subject: Re: Further observations...
To: David A. Gatwood <marsmail@globegate.utm.edu>
From: Bill Studenmund <wrstuden@loki.stanford.edu>
List: port-mac68k
Date: 02/07/1998 14:17:40
On Sat, 7 Feb 1998, David A. Gatwood wrote:

> On Sat, 7 Feb 1998, Bill Studenmund wrote:
> > Paul Goyette and I worked on hfsfs. We got it to where it'd work
> > single-user, but there was something wrong with how it handled buffers,
> > and it'd panic the kernel if anything else was using a filesystem at the
> > same time.
> 
> Was that the one that could only read, or could it write, too?

Only read.

> > If anyone else wants to work on this, go for it. Though I'd recomend
> > instead that you port macfs, the linux HFS-reading FS. It seems to write
> > safely, and Paul Hargrove, its author, would be happy for there to be a
> > NetBSD port. Also, mkLinux seems to use it, so there'll be more long-term
> > support from Apple. It will still be GPL'd though. He received some
> > assistance from a mac-emulator company on the understanding that the code
> > stay GPL'd, and he is respecting that agreement.
> 
> By any chance are the most recent versions of your hfsfs work available in
> source form?  If not, could you mail me a copy?  I'd love to have a look. 
> Not certain that I could figure anything out, but I'd certainly give it a
> shot....

The stuff on puma is, AFAIK, the latest. Paul and I haven't touched it in
years.

That code base would be a good educational reference, but I think Paul
Hargrove's macfs would be a better port. It'll write, and someone is
adding HFS+ support to it. It's only problem is it was not at all designed
to be used in anything other than Linux; there's no MacFS/UnixFS boundry
layer.

Take care,

Bill