On Wed, 2009-06-24 at 15:26 +0200, Rafał Miłecki wrote:
2009/6/24 Matthias Hopf
: Do you know a version where it worked? Possibly to git-bisect this? git commits efa0b5475517 and 742a4b4e are likely candidates for breaking anything regarding properties.
That is probably still the best thing you can do. Maybe compiling with -DDEBUG before and producing a -logverbose 7 log file. But I'm unsure whether this will include the necessary information.
Starting from efa0b5475517 I don't see _Backlight anymore for my PANEL.
The (really) weird thing is this was working for you Vladimir earlier.
It works fine for me just after it was announced in this list. Even more, gnome still may change desplay brightness somehow (most probably direct with sysctl)
In my case there was _Backlight property with completly wrong value, and every try of changing it crashed driver. So you saw _Backlight property because of bug and setting this property... shouldn't work for you actually ;)
Not sure, what exactly property you talking about, but it was controlled by xbacklight.
Real code for backlight controll is available, it's just disabled for some reason. Grep rhd_lvtma.c for "backlight" - it's inside #ifdef NOT_YET. I enabled this property locally and it was working fine. But I didn't care much about checking that as I'll try to implement backlight in KMS (proper place for this).
Can someone explaing me why we have two functions LVDSSetBacklight? In: rhd_dig.c rhd_lvtma.c
-- Rafał Miłecki -- Vladimir B. Grebenschikov vova@fbsd.ru -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org