From: Andrew Apted <ajapted@netspace.net.au>
  To  : ggi-develop@eskimo.com
  Date: Sat, 29 Aug 1998 15:32:23 +1000

Re: Problems with ATI Mach64 Driver...

Jon writes:

>  	IMHO it should be changed and all cfbx options should be required
>  to be compiled into the kernel.  I was thinking about this the last time I
>  looked at fbcon-kgi.c.  Since kgicon has to be able to handle a variety 
>  of driver types with a variety of capabilities, it needs to be able to 
>  handle all modes.

The only advantage I can see of having the cfb* as options in the kernel
config is reducing kernel code size.  But... if/when we integrate kgicon
into the kernel makefile system, then we either (a) force all the cfb*
options to be set whenever `[ ] Enable KGICON support' is set (otherwise
it don't compile), or (b) live with it.  Since users have to enable the
`[ ] Advanced low level driver options' to disable the cfb* options,
then I say (b) keep the #ifdefs and let the users RTFM.

If kgicon is staying out of the kernel makefile system, then removing
those #ifdefs would be reasonable (we'll get `why doesn't it compile ?'
instead of `why doesn't it work ?') although not essential IMHO.

So the question is: do we integrate kgicon into the kernel makefile
system (i.e. right now) ?  What worries me is the big question mark
floating above Steffen's new KGI API (and the other KGI APIs like the
evstack and suidkgi changes).

Cheers,
_____________________________________________  ____
                                               \  /
  Andrew Apted   <andrew@ggi-project.org>       \/