Re: Quik and atyfb

[Simon Stapleton <simon.stapleton@gmail.com>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


On 19 Aug 2006, at 13:36, Holger Levsen wrote:

> Some more information from a private disucssion on IRC, p2-mate is  
> the quik
> maintainer:
>
> Aug 11 10:12:16 <h01ger>        p2-mate, have you seen
> http://lists.debian.org/debian-powerpc/2006/08/msg00073.html ?
> Aug 11 10:13:43 <h01ger>        p2-mate, re: that mail: i'll ask  
> the guy to
> file a bug against quik?
> Aug 11 10:14:08 <p2-mate>       h01ger: no
> Aug 11 10:14:13 <p2-mate>       it's not a quik bug
> Aug 11 10:14:33 <p2-mate>       it's probably because macos inits  
> the aty card
> in a different way then OF does
> Aug 11 10:14:52 <p2-mate>       there is very little I can do  
> against that I'm
> afraid
> Aug 11 10:16:05 <p2-mate>       I think the proper solution is to  
> fix atyfb
> Aug 11 10:16:09 <h01ger>        ah
> Aug 11 10:16:55 <p2-mate>       both miboot and bootx cause macos  
> to init the
> gfx hardware
> Aug 11 10:18:05 <h01ger>        so there are circumstances where  
> bootx is a
> better choice then quik?
> Aug 11 10:19:40 <p2-mate>       it causes more hw to be initialized  
> by macos
> yes
> Aug 11 10:19:51 <p2-mate>       personally I don't like bootx at all
> Aug 11 10:20:16 <p2-mate>       it makes for horrible boottimes

Hi Holger.

To an extent, I agree with p2-mate, the solution might be to cause  
atyfb to properly initialise the hardware, it would seem to be a  
cleaner way to do things if it's possible.

A quick hack in atyfb to add the same initialisation that bootx does  
(resetting bus mastering and flushing the buffer) doesn't seem to  
work thus far, though.  Meh.

Simon
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (Darwin)

iEYEARECAAYFAkTrHOMACgkQRKxFDFDQaY1A0wCfbUHKgdENo6maE9BK4bovrmAQ
ZYAAn0JNf+qhSGHQRII1tDMFvsLdNsGG
=CV81
-----END PGP SIGNATURE-----