[Dri-devel] Re: [Xpert]VSYNC/VBLANK reporting

[Vladimir D. <vo...@mi...>]

On 3 Jun 2002, Michel [ISO-8859-1] D=E4nzer wrote:

> On Mon, 2002-06-03 at 03:39, Vladimir Dergachev wrote:
> >
> > For people who were interested in observing VSYNC/VBLANK interrupts fro=
m
> > userspace there is an experimental implementation for radeons on
> > http://gatos.sf.net/ in CVS - module "km". Support for rage128 and mach=
64
> > chips would be easy to add if interested developers with these cards wi=
ll
> > ask - I did not do it yet so as to limit places I need to update if any
> > api changes come about.
>
> Maybe at least some of this stuff could be integrated into the DRM?
> Let's discuss that on dri-devel.

Could, and probably should, and, in fact, last time I discussed this on
IRC meeting (around February) people agreed this is a good idea.

But, I have not had as much time as I liked to keep up with dri
development. Also the km_api API is nowhere near stable yet.

The good news is that km_api would be extremely easy to integrate (it is
just a symbolic way to pass information from kernel to user space).

                         Vladimir Dergachev

PS Here are some links to km_api docs:

[Design]
http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/gatos/km/km.rfc.txt

[Actual]
http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/gatos/km/km.actual.rfc.txt

>
>
> --
> Earthling Michel D=E4nzer (MrCooper)/ Debian GNU/Linux (powerpc) develope=
r
> XFree86 and DRI project member   /  CS student, Free Software enthusiast
> _______________________________________________
> Xpert mailing list
> Xpert@XFree86.Org
> http://XFree86.Org/mailman/listinfo/xpert
>