doctor@fruitbat.org
Thu, 26 Aug 1999 14:34:18 -0700 (PDT)
David McWherter said ...
>
> > > > They might look more favorably on the updates if they were submitted as
> > > > diff patches instead of straight file replacements.
> > >
> > > Right...I asked Cort what format he would prefer.
> > > I haven't heard anything yet...
> >
> > Hope he responds soon. Anyone else we could route the updates through?
>
> Cort indicated that either format would be hunkey dorey the last time
> I submitted them to him a long while ago. One thing to note is that
> Linux is undergoing lots of revisions - it may be hard to keep integrating
> patches into the official distibution as well as making sure the newer
> versions work in PPC-land simultaneously.
Supposedly once the patches make it into the stardard kernel there won't
be any need for ANS specific patches to be releases separately. I had
originally hoped that the PPC branch would be unified under the standard
kernel as well, but perhaps that's asking a bit much. If so, then having
the patches integrated into the PPC branch by Cort would be sufficient.
The point is to make the ANS an officially supported part of the PPC port.
> > > > What's missing is Doc for the ANS changes (ie: Configure.help for the
> > > > ans-lcd and cirrus54m30fb drivers). Also, the configure prompt for the
> > > > ans-lcd probably should be moved to the Character Devices Config.in
> > > > instead of the ppc main config.in. Heck, perhaps the ans-lcd code should
> > > > be moved to the drivers/char directory (it is a char device, after all).
> > > > Also, an official major/minor device number needs to be allocate to the
> > > > ans-lcd driver. 160 isn't an official number, if I recall.
> > >
> > > The LCD is not as important...
> >
> > Perhaps not, but I still think it should be moved...
>
> The LCD driver is really a hack. I have a feeling that putting it
> into the real Linux kernel almost mandates that we should update
> it so that it's more compatible with the "right" way of doing things
> (ie: choosing an appropriate device minor number, probing the system
> correctly to make sure that the device is there before writing to
> it, etc, etc). It shouldn't be a lot of work, but nonzero in magnitude.
I'd already assumed that the driver would be updated as needed (official
major number and all). Perhaps even adding a but of logic to make it
behave more like a terminal instead of a raw interleaved display device.
maybe, say, something that understood VT100 controls?
> > > any volunteers to do docs for the video and MACE changes?
> >
> > If no one else volunteers, I'll do it.
>
> Are you looking for Configure-script messages, or something more in
> depth? I've a feeling that what's there doesn't require much in
Yes, and yes. The configure scripts are minor, IMHO, and mostly done
(except for the ans-lcd placement). However, Configure.help needs
updating and a short ANS specific doc file would be helpful those who
have questions on picking the right kernel options for the ANS.
> the way of extensive documentation - the MACE change is really just
> a small 1-line patch or so (unless it's been extended since I
Driver changes go in README and Changes files (which, too, need updating).
> last looked at it), and the video patch is fairly vanilla - no
> configuratable options, etc.
The Cirrus driver could perhaps use a little hacking so it can activate
power-save mode ;)
> -David
-- Peter A. Castro (doctor@fruitbat.org) or (pcastro@us.oracle.com)
This archive was generated by hypermail 2.0b3 on Thu Aug 26 1999 - 21:37:07 GMT