Re: [radeonhd] [PATCH] Somehow /force/ backlight control
![](https://seccdn.libravatar.org/avatar/269a245a17cfd3594519d252d8682ca7.jpg?s=120&d=mm&r=g)
Rafa? Mi?ecki writes:
... unfortunately, yes. Lemme try and add the AtomBIOS variant and see if this will work for you. (In fact you could try if this would work for you by enabling AtomBIOS with the "UseAtomBIOS" option.
So I can add "UseAtomBIOS", remove all NOT_YET and this should work for me? Because you mentioned earlier it's not ready yet...?
You don't even have to remove the NOT_YET with thoe UseAtomBIOS option. I'm currently looking at why I added the NOT_YETs. It's working on RS690 (LVTMA). I still need to try it on DIG.
We've already got a PCI subvendor ID based quirks list. We could add this to the flags section there.
But these quirks are connectors touching, right? Connectors should be the same of every 1002:95c4, so that's easy. Unfortunately 1002:95c4 can be monted in machine with backlight control based on on-GPU-chip registers, as well as in machine with different backlight control. So we need more specific devices table.
No, it is per subvendor/subsys ID, thus exactly what we need. Cheers, Egbert. -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org
![](https://seccdn.libravatar.org/avatar/1964acfaf091476dc23e13fcb67f2e0c.jpg?s=120&d=mm&r=g)
W dniu 30 czerwca 2009 17:32 użytkownik Egbert Eich
> So I can add "UseAtomBIOS", remove all NOT_YET and this should work > for me? Because you mentioned earlier it's not ready yet...?
You don't even have to remove the NOT_YET with thoe UseAtomBIOS option. I'm currently looking at why I added the NOT_YETs. It's working on RS690 (LVTMA). I still need to try it on DIG.
OK, I just tested it and it works actually. I decided to check out atombios's check about backlight control. 1) In rhd_dig.c we call RhdAtomSetupBacklightControlProperty which returns bl level 2) in rhd_atomout.c: RhdAtomSetupBacklightControlProperty calls RHDAtomBIOSScratchBlLevel 3) in rhd_biosscratch.c: RHDAtomBIOSScratchBlLevel just check 0x18 / 0x172C register. So is this AtomBIOS based backlight control just using another register? No special AtomBIOS commands?
> > We've already got a PCI subvendor ID based quirks list. We could add this > > to the flags section there. > > But these quirks are connectors touching, right? Connectors should be > the same of every 1002:95c4, so that's easy. Unfortunately 1002:95c4 > can be monted in machine with backlight control based on on-GPU-chip > registers, as well as in machine with different backlight control. So > we need more specific devices table.
No, it is per subvendor/subsys ID, thus exactly what we need.
Didn't know that, thanks. -- Rafał Miłecki -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org
![](https://seccdn.libravatar.org/avatar/e6be5c2aafd856f8e01e8b0fbd66999e.jpg?s=120&d=mm&r=g)
2009/6/30 Rafał Miłecki
W dniu 30 czerwca 2009 17:32 użytkownik Egbert Eich
napisał: > So I can add "UseAtomBIOS", remove all NOT_YET and this should work > for me? Because you mentioned earlier it's not ready yet...?
You don't even have to remove the NOT_YET with thoe UseAtomBIOS option. I'm currently looking at why I added the NOT_YETs. It's working on RS690 (LVTMA). I still need to try it on DIG.
OK, I just tested it and it works actually.
I decided to check out atombios's check about backlight control.
1) In rhd_dig.c we call RhdAtomSetupBacklightControlProperty which returns bl level
2) in rhd_atomout.c: RhdAtomSetupBacklightControlProperty calls RHDAtomBIOSScratchBlLevel
3) in rhd_biosscratch.c: RHDAtomBIOSScratchBlLevel just check 0x18 / 0x172C register.
Those are the bios scratch regs.
So is this AtomBIOS based backlight control just using another register? No special AtomBIOS commands?
As I said previously, IIRC, it's controlled by LCD1OutputControl on pre-DCE3 chips, and DIG2TransmitterControl on DCE3 and newer chips. As I recall, you set the requested level in one of the bios scratch regs and then execute the table with the action set to the appropriate LCD_BLOFF/BLON/BL_BRIGHTNESS_CONTROL for the table. See atombios.h. Alex -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org
participants (3)
-
Alex Deucher
-
Egbert Eich
-
Rafał Miłecki