Subject: Re: grfmmap bug?
To: None <leo@wau.mis.ah.nl>
From: maximum entropy <entropy@zippy.bernstein.com>
List: port-atari
Date: 12/17/1998 13:55:19
>Date: Thu, 17 Dec 1998 15:10:23 +0100
>From: Leo Weppelman <leo@wau.mis.ah.nl>
>
>Mmmm.... iff (gd_regsize == 64), you call the mmap function like:
>     mmap(NULL, <fbsize>, protections & flags, fd, 64)
>
>to map de frame buffer. In my patched expression, this _is_ aligned to
>a page boundary (off == 0 before the return!).... am I missing something???

No, but I was missing something.  I've re-read the code and now it's
clear to me that one of my preconceptions about how this is supposed
to work was mistaken.  Sorry for the confusion.

>I'm not sure if much time should still be spend on the grf/ite interface...
>Over the last 2 weeks or so, I have been looking at the wscons-interface.
>This is the interface that is targetted as _the_ generic (== MI) console

OK.  Then I'll probably wait until wscons is ready to begin my
efforts.  It's a real pain to work on the silly thing anyway, now that
my TT's VME slot is taken up by the ethernet card.

Since as far as I know I'm the only person with a Leonardo graphics
card running NetBSD/atari (I believe only three of the cards were ever
produced), my true goal in working on the driver at all is to learn
more about NetBSD internals -- so that I can apply that knowledge to
more useful things in the future.  I think I've managed to accomplish
that goal this time, without writing a single line of code :-)

Cheers,
entropy

--
entropy -- it's not just a good idea, it's the second law.