xf86-video-radeonhd:master: 1 commit(s)
Reporting from xf86-video-radeonhd: Radeon HD video driver
Please visit:
http://gitweb.freedesktop.org/?p=xorg/driver/xf86-video-radeonhd
Or get your own copy by using:
git-clone git://anongit.freedesktop.org/xorg/driver/xf86-video-radeonhd
Commit against master at 69d4dabe...:
commit 3ac30d19e0b6da36ea17519b181ddf8cf2c96561
Author: Christian Koenig
On Mon, Oct 27, 2008 at 08:14:10AM -0700, Luc Verhaegen wrote:
Author: Christian Koenig
Date: Mon Oct 27 16:08:59 2008 +0100 HDMI: Add HDMI support.
Currently tested and working with RV620, RV630, RV635, RV670, RV770 and RS780.
Thanks to all who were involved here, and many thanks to Christian. No-one so far has made such a big code contribution in a way that completely fits in with the radeonhd driver design/phylosophy. This really is amazing code and an amazing effort. I can state that rv670 works, if you use TMDSA (and if you use a proper display, something for which the dell 3008wfp doesn't qualify in any way). I will continue to look into why it is not working with LVDS, but then the Sony TV is not showing any mode with LVDS atm. I'll see what i can do to get other hw, as far as is available to us, tested as well :) About the options: three things there... * we might want to get rid of the audio option altogether. The other option takes care of this already. And we might want to load the basic infrastructure anyway. * we might want to match the connector names with the connector names (without analog/digital), as reported by xrandr. * we might want to make audio on/off output specific attributes that can also be set/unset through xrandr. Luc Verhaegen. SUSE X Driver Developer. -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org
Am Montag, den 27.10.2008, 16:38 +0100 schrieb Luc Verhaegen:
Thanks to all who were involved here, and many thanks to Christian.
No-one so far has made such a big code contribution in a way that completely fits in with the radeonhd driver design/phylosophy. This really is amazing code and an amazing effort. Good to know that i could help out with this and it was really fun to do so :)
I can state that rv670 works, if you use TMDSA (and if you use a proper display, something for which the dell 3008wfp doesn't qualify in any way). I will continue to look into why it is not working with LVDS, but then the Sony TV is not showing any mode with LVDS atm. Cool, the fact that this wasn't tested on an rv670 before (AFAIK), but still works out of the box shows that the assumptions i made about the hardware are correct.
Is the LVDS problem HDMI related, or does it make problems on the Sony TV anyway?
About the options: three things there...
* we might want to get rid of the audio option altogether. The other option takes care of this already. And we might want to load the basic infrastructure anyway. These option wasn't used for so much anyway, i just wanted to have a global switch to turn everything off in case of a problem. I think we can remove this if there are no regressions related to this patch.
* we might want to match the connector names with the connector names (without analog/digital), as reported by xrandr. That was my intention anyway, but after the discussion about xrandr connector names started i didn't want do stick my nose into code properly changing in the near future.
* we might want to make audio on/off output specific attributes that can also be set/unset through xrandr. That's something i also had in mind, and there are other options in the HDMI info packets that are quite interesting, like colormetry (you can set the black/white reference value) or aspect ratio (some DLP Projectors can actually switch between 16:9,4:3 and 9:16,3:4 without loss of luminescence and my TV switches between 16:9 and 4:3 when i set this option by adding black borders).
I will keep working on this in the near future. And by the way is anybody else working on EEDID support? I know this really doesn't belong into radeonhd, but since x server 1.5.0 we have the raw EEDID data right in front of us, with no help from x to decode this. Bye, Christian. PS: Since my last convincement worked so good: Getting the 3D/xvideo support released would really make a difference between the radeonhd and radeon driver :) -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org
On Mon, Oct 27, 2008 at 10:47:29PM +0100, Christian König wrote:
Is the LVDS problem HDMI related, or does it make problems on the Sony TV anyway?
A problem with the LVDS (TMDSB) output and this TV. Sadly, i had to return the TV, because people were complaining about not being able to play the wii.
About the options: three things there...
* we might want to get rid of the audio option altogether. The other option takes care of this already. And we might want to load the basic infrastructure anyway. These option wasn't used for so much anyway, i just wanted to have a global switch to turn everything off in case of a problem. I think we can remove this if there are no regressions related to this patch.
Ok :)
* we might want to match the connector names with the connector names (without analog/digital), as reported by xrandr. That was my intention anyway, but after the discussion about xrandr connector names started i didn't want do stick my nose into code properly changing in the near future.
Just go ahead, and change when the naming actually changes.
* we might want to make audio on/off output specific attributes that can also be set/unset through xrandr. That's something i also had in mind,
Ok :)
and there are other options in the HDMI info packets that are quite interesting, like colormetry (you can set the black/white reference value) or aspect ratio (some DLP Projectors can actually switch between 16:9,4:3 and 9:16,3:4 without loss of luminescence and my TV switches between 16:9 and 4:3 when i set this option by adding black borders).
I will keep working on this in the near future. And by the way is anybody else working on EEDID support? I know this really doesn't belong into radeonhd, but since x server 1.5.0 we have the raw EEDID data right in front of us, with no help from x to decode this.
We can always go and talk to i2c directly and read things out immediately.
PS: Since my last convincement worked so good: Getting the 3D/xvideo support released would really make a difference between the radeonhd and radeon driver :)
Well, do you have the code sitting around ready and waiting like last time? :p Luc Verhaegen -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org
participants (3)
-
Christian König
-
libv@kemper.freedesktop.org
-
Luc Verhaegen