Index:
[thread]
[date]
[subject]
[author]
From: teunis <teunis@computersupportcentre.com>
To : ggi-develop@eskimo.com
Date: Thu, 17 Sep 1998 23:12:06 -0700 (MST)
Re: kgicon -> linux-2.2 ideas (was Re: Makefile vote II)
On Thu, 17 Sep 1998, Sengan Baring-Gould wrote:
> teunis wrote:
>
[clip]
> > Why not pack up a KGI-install of a (working) set of MediaGX KGI/fbcon
> > drivers as a patch-set and send that in? Why send the whole driver tree
> > in?
> >
> > (picking an example from a currently unsupported-in-linux (afaik) card :)
>
> Thanks ;-)Actually Suse's Xfree is on the attack vector for this one.
>
> > (It might not work. From the list afaik the MediaGX drivers may still
> > need work I don't know as I don't have one)
>
> A bit. But I think we need the KGI API to be set in stone before bothering
> Linus.
Point made...
But actually this is prolly a good vector to introduce KGIcon:
1. Get KGI api ironed-out (and soon!)
2. Make it so kgicon can be patched into kernel tree easily
3. Find a good driver and send that and -only- that driver as
a kernel-patch to Linus.
3 is the important bit. Taking, say, the MediaGX driver - which should
have non-GGI-backed support and sending it in as a new fbcon driver should
sorta pave the way. I suspect this will be a lot easier accepted than a
whole raft (driver-tree) of drivers that could be overwelming. Then send
a new driver (or several) as soon as that one driver's both considered
tested and accepted :)
Be even more cool if the entire KGIcon package could be accepted into the
linux CVS project.... That's prolly a lot easier as Linus -does- have it
in against GGI (or used to anyways). Mostly 'cause of all the religious
wars about graphics the GGI people useta have on linux-kernel before GGI
was formed elsewhere :) [and a little after too - for, like, a year or so
:] (I was one of those *sigh*. Shoulda bought in right away :)
I shut up now :) [think I've described this point to death anyways]
G'day, eh? :)
- Teunis
Index:
[thread]
[date]
[subject]
[author]