Date: Sun, 27 Aug 1995 22:27:12 -0700 (PDT) From: sos@freebsd.org To: jdl@chromatic.com Cc: freebsd-current@freebsd.org, joerg_wunsch@uriah.heep.sax.de Subject: Re: That IDE CD ROM thang again... Message-ID: <199508280527.WAA01708@freefall.FreeBSD.org> In-Reply-To: <199508261347.IAA25153@chrome.onramp.net> from "Jon Loeliger" at Aug 26, 95 08:47:50 am
next in thread | previous in thread | raw e-mail | index | archive | help
In reply to Jon Loeliger who wrote:
>
> > I think it's in error to byte-shuffle only the ID string. Everything
> > from the drive must be byte-swapped, including the ID word and all
> > data.
Ahem, my own obseravtions suggests that it is only "some" fields
that require swapping, depending on model...
> Oh my! OK, I'll give this a hack attack and see what I get. Any chance
> you can give me a one minute pointer to *where* that might take place?
> Like, where the buffers coming back from the drive are filled? I see
> the one in _probe() that's obtained with the insw(). Will they all
> look like that?
>
> > (Of course, you have to make a distinction whether to swap or
> > not before.)
>
> Yea, this is curious. I get the feeling that some people are able
> to get the IDE CD drives to work and others aren't. Is there some
> reasonable distinction separating the two camps? Do we need to start
> a collection of working/not-working drives and form the basis of a
> boot time lookup table (ick) to determine byte-swapping needs?
Yep, we need a table of known drives...
(Actually this is a case where the DOS'mers have an advantage -
they get a driver with the drive, that knows its little secrets :)
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Soren Schmidt (sos@FreeBSD.org | sos@login.dknet.dk) FreeBSD Core Team
So much code to hack -- so little time
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199508280527.WAA01708>