Is anyone planning to update the apparently obsolete(*)
DRM drivers currently in 2.4.22-pre/rc for 2.4.23?
I noticed that the RH9 kernel applies a drm-4.3 patch,
so the code must be there, just not in Marcelo's tree :-(
(*) XFree86-4.3.0 tells me my radeon module from 2.4.22-pre10
is out of date, and glxgears runs like molasses :-(
With 2.6.0-test2, the X server doesn't complain, and rendering
performance is an order of magnitude better.
/Mikael
On Wednesday 06 August 2003 16:51, Mikael Pettersson wrote:
Hi Mikael,
> Is anyone planning to update the apparently obsolete(*)
> DRM drivers currently in 2.4.22-pre/rc for 2.4.23?
I have a pending DRM 4.3 update for .23-pre1. Marcelo did not accept it for
.22 ( I sent it first while -pre9 time or so. )
It's a complete DRM-4.3 tree. He has to decide between an update of existing
4.2 code or an addition of a new subdirectory drm-4.3 + proper config.in
entry.
ciao, Marc
> It's a complete DRM-4.3 tree. He has to decide between an update of existing
> 4.2 code or an addition of a new subdirectory drm-4.3 + proper config.in
> entry.
did you clean the tree up like in -ac's tree or did you take it as is
from some cvs repo ?
On Wed, 2003-08-06 at 17:16, Marc-Christian Petersen wrote:
> On Wednesday 06 August 2003 16:51, Mikael Pettersson wrote:
>
> > Is anyone planning to update the apparently obsolete(*)
> > DRM drivers currently in 2.4.22-pre/rc for 2.4.23?
>
> I have a pending DRM 4.3 update for .23-pre1. Marcelo did not accept it for
> .22 ( I sent it first while -pre9 time or so. )
>
> It's a complete DRM-4.3 tree.
Literally from XFree86 4.3, or from DRI CVS? The latter would be better,
as the former is no longer actively maintained.
> He has to decide between an update of existing 4.2 code or an
> addition of a new subdirectory drm-4.3 + proper config.in entry.
There's no need for a separate directory, the DRM has been backwards
compatible since 4.2 at least, and countless bugs have been fixed since
then.
It would also be great if you could isolate DRM fixes in the kernel and
post patches here - some kernel developers keep complaining how broken
the DRM code is, but never submit patches...
--
Earthling Michel Dänzer \ Debian (powerpc), XFree86 and DRI developer
Software libre enthusiast \ http://svcs.affero.net/rm.php?r=daenzer
On Wednesday 06 August 2003 17:37, Arjan van de Ven wrote:
Hi Arjan,
> > It's a complete DRM-4.3 tree. He has to decide between an update of
> > existing 4.2 code or an addition of a new subdirectory drm-4.3 + proper
> > config.in entry.
> did you clean the tree up like in -ac's tree or did you take it as is
> from some cvs repo ?
nope, cvs. If Alan will be so kind to send me the fixes he made and I don't
have to do the double-work, I'll integrate and test them up.
Or another choice would be that Alan will send his drm 4.3 code to Marcelo
once .22 final is out for .23-pre1 inclusion.
Decide yourself :)
ciao, Marc
On Wed, 6 Aug 2003, Marc-Christian Petersen wrote:
> On Wednesday 06 August 2003 16:51, Mikael Pettersson wrote:
>
> Hi Mikael,
>
> > Is anyone planning to update the apparently obsolete(*)
> > DRM drivers currently in 2.4.22-pre/rc for 2.4.23?
>
> I have a pending DRM 4.3 update for .23-pre1. Marcelo did not accept it for
> .22 ( I sent it first while -pre9 time or so. )
>
> It's a complete DRM-4.3 tree. He has to decide between an update of existing
> 4.2 code or an addition of a new subdirectory drm-4.3 + proper config.in
> entry.
Does DRM 4.3 work with both XFree 4.2 and 4.3 ?
I dont so, right?
On Wednesday 06 August 2003 18:58, Marcelo Tosatti wrote:
Hi Marcelo,
> Does DRM 4.3 work with both XFree 4.2 and 4.3 ?
> I dont so, right?
it will work for both.
ciao, Marc
On Mer, 2003-08-06 at 16:44, Marc-Christian Petersen wrote:
> > did you clean the tree up like in -ac's tree or did you take it as is
> > from some cvs repo ?
> nope, cvs. If Alan will be so kind to send me the fixes he made and I don't
> have to do the double-work, I'll integrate and test them up.
Your code won't work with 4.1 users i810 at least then, and has some
other problems that were fixed over time.
> Or another choice would be that Alan will send his drm 4.3 code to Marcelo
> once .22 final is out for .23-pre1 inclusion.
I can send him the -ac one minus the small change for the memory accounting
stuff easily enough. Thats a tree which lots of people have run (both vendor
and non vendor) so its probably a safer pick.
On Mer, 2003-08-06 at 17:58, Marcelo Tosatti wrote:
> > It's a complete DRM-4.3 tree. He has to decide between an update of existing
> > 4.2 code or an addition of a new subdirectory drm-4.3 + proper config.in
> > entry.
>
> Does DRM 4.3 work with both XFree 4.2 and 4.3 ?
>
> I dont so, right?
It doesn't. As discussed on the kernel list and DRI list a while ago.
The -ac tree / Red Hat one does because it has some additional magic to
spot i810 problems.
On Thursday 07 August 2003 13:21, Alan Cox wrote:
Hi Alan,
> I can send him the -ac one minus the small change for the memory accounting
> stuff easily enough. Thats a tree which lots of people have run (both
> vendor and non vendor) so its probably a safer pick.
please so so after .22 final :) ... Thanks.
ciao, Marc
On Thu, 2003-08-07 at 13:20, Alan Cox wrote:
> On Mer, 2003-08-06 at 17:58, Marcelo Tosatti wrote:
> > > It's a complete DRM-4.3 tree. He has to decide between an update of existing
> > > 4.2 code or an addition of a new subdirectory drm-4.3 + proper config.in
> > > entry.
> >
> > Does DRM 4.3 work with both XFree 4.2 and 4.3 ?
> >
> > I dont so, right?
>
> It doesn't. As discussed on the kernel list and DRI list a while ago.
> The -ac tree / Red Hat one does because it has some additional magic to
> spot i810 problems.
That's a bug which can be fixed then, doesn't warrant separate copies in
the kernel. I'm sure Dave would happily integrate the fix in DRI CVS.
--
Earthling Michel Dänzer \ Debian (powerpc), XFree86 and DRI developer
Software libre enthusiast \ http://svcs.affero.net/rm.php?r=daenzer
On Iau, 2003-08-07 at 14:47, Michel Dänzer wrote:
> > It doesn't. As discussed on the kernel list and DRI list a while ago.
> > The -ac tree / Red Hat one does because it has some additional magic to
> > spot i810 problems.
>
> That's a bug which can be fixed then, doesn't warrant separate copies in
> the kernel. I'm sure Dave would happily integrate the fix in DRI CVS.
I'd have to take a look what DRI CVS does and doesn't currently contain. That's
something I can't realistically expect time for until October