[forum] Re: communication

[Sven Luther]

On Mon, Mar 24, 2003 at 02:48:29PM -0500, David Dawes wrote:
> On Mon, Mar 24, 2003 at 01:50:47PM -0500, Nicholas Wourms wrote:
> >David Dawes wrote:
> >> On Sun, Mar 23, 2003 at 01:39:45PM -0600, Billy Biggs wrote:
> >[SNIP]
> >> I have no mechanism for maintaining the linux fbdev driver I've
> >> written within the context of the Linux kernel, so I provide it on
> >> my home page.  If I wanted to create a mailing list for it, I'd go
> >> to SourceForge.  Lots of Linux kernel sub-lists are hosted on
> >> SourceForge.  Does that make the work they do any less a part of
> >> the Linux kernel?
> >
> >It already exists.  What about the fbdev sourceforge project 
> >that James Simmons heads up?  He is, after all, the defacto 
> >maintainer of the linux fbdev/fbcon drivers.  I haven't seen 
> 
> I don't want to get lost in the details, but it depends on how
> finely you want to maintain things.  If James Simmons is the
> maintainer of linux fbdev/fbcon drivers, does that preclude
> maintainers for individual fbdev drivers?  I'm not angling to get

Yes, James is the maintainer of the fbdev/fbcon part of the linux
kernel. Altough a he went on job search he has had less time recently.
Antonino Daplas is also heavily involved these past times. That said,
all linux fbdev driver are supposed to at least follow what is going on,
especially if they work on the devel tree, so that they would be aware
of big changes being made and so.

> some sort of Linux kernel maintainer status, but it does seem like
> I'm not able to maintain my driver directly within the Linux kernel,
> so my remaining option is to maintain it elsewhere, and submit
> patches (I think using BK and saying "pull this" is essentially
> equivalent to submitting patches when you compare the steps).

Mmm, i don't know, i am not all that familiar with BK, but it is more
like you maintaina separate cvs branch, and when you say "pull this" it
gets merged back in. I don't think everyone has a BK tree, and James
mostly acts as a buffer between the fbdev developpers and linus. You can
think of him as a cvs commiter if you want.

> >you on the list, perhaps you might join and submit patches 
> >there so that your changes can be merged into the fbdev bk 
> >tree before Linus pulls from it.  There are also technical 
> >discussions concerning the present and future of the core 
> >fbdev api.  This way you could avoid the possibility of 
> >breakage like what probably happened when Linus pulled the 
> >new set of changes from the fbdev repo 1 day after your 
> >intel patch went in.
> 
> I have been intermittently following the fbdev discussions.  The
> driver I wrote was targetted primarily for the 2.4.x fbdev interfaces,
> and I haven't yet had the bandwidth to port it to the latest fbdev
> interfaces in the development kernel.  That's precisely why I didn't

The new interface is much simpler to write driver for. 

Friendly,

Sven Luther