Chameleon > DevTalk
ATI/AMD GraphicsEnabler support (legacy, modules & missing devices)!
Azimutz:
--- Quote from: pauldmps on September 02, 2011, 06:45:18 AM ---^^ He is online everyday & not answering posts 4-5days old?
I understand that Devs do a hard work but when you've started a thread you must follow it.
--- End quote ---
pauldmps, you need to lower those anxiety levels :)
Concerning Chameleon GE, you were already answered on the forge issue (156) and your specific card was added.
About the Muxless thing, i have no idea about how OS X handles it. I only have access to nVidia's Optimus,
but i still didn't had time to try OS X in it. Same thing with Lion and i can't run it on the desktop so, can't help with it.
But, you can always try uninstalling Intel's drivers to see if you can get the ATI ones working; if you are successful with ATI,
then check if Muxless is viable and if not, the choice is yours on the graphics to keep.
For now it's all i have...
--- Quote from: dontomaso on August 26, 2011, 07:58:45 PM ---I have 10.7 Lion and also the AMD Radeon HD 6470M graphics in my Dell Vostro 3350 notebook, but this notebook has also the integrated Sandy Bridge Intel HD 3000 Onboard Graphics.
...
--- End quote ---
dontomaso, same as above, except for anxiety ;D
yep, Intel seems to work ootb with MacBookPro8,2/8,3 at least; 8,1 uses only Intel (Mactracker info) so i don't think it will do.
About how Chameleon works, from a couple of posts ago:
--- Quote ---... all GE does is write/pass some values to ioreg device tree,
that the drivers expect to find, similar to what can be done with DSDT editing.
--- End quote ---
if you want to know what's "written" check the code, ati.c; basically it's down to write the device path (devprop_add_device)
and the stuff that's not commented out on "dev_prop_t ati_devprop_list", using devprop_add_list function and pass it
much like using "device-properties" key. The rest is done by kernel and kexts.
You guessed right about the funny FB names :P yep, the FB's changed place on 10.6.8 and above, like you figured;
now we are limited to the ones present on the used Controller.
Throw me your subsys id/ven so i can add an entry for your card.
Is device id 6760 correct?
Zane:
--- Quote from: Azimutz on September 17, 2011, 01:55:25 PM ---
--- Quote from: Zane on September 16, 2011, 09:02:41 PM ---Done. Used Hoolock. HDMI does not work with Hoolock. One thing to note is that i get a scrambled screen for 1 second before it goes back to proper almost everytime OS X boots. Uploaded the bdmesg and ioreg for both Hoolock and Langur.
--- End quote ---
Zane, those ioreg's are useless :P use
--- Code: ---ioreg -lfw0
--- End code ---
bdmesg looks ok; just confirm me something...
If you boot without e.g. AtiConfig=Hoolock do you get Nomascus framebuffer?
And the subsys id/ven is correct?
Still, would like to take a look at ioregs...
--- End quote ---
You should have said so earlier! :p I'll get it to you in a bit.
Yes if i boot without Hoolock i get Nomascus and that doesnt work for me.
What do you mean the subsys id and vendor id is correct?
Azimutz:
--- Quote from: Zane on September 18, 2011, 06:34:44 AM ---You should have said so earlier! :p I'll get it to you in a bit.
--- End quote ---
yeah, yeah... more work less talk :P kidding... no hurries...
I mean, if it shows the card's correct subsys id/ven (1522103C), e.g. on bdmesg, ioreg, etc...?
dontomaso:
Hello Azimutz,
Thanks a lot for the info.
--- Quote ---Throw me your subsys id/ven so i can add an entry for your card.
Is device id 6760 correct?
--- End quote ---
I attached two jpg files, thanks.
Zane:
--- Quote from: Azimutz on September 18, 2011, 06:59:39 PM ---yeah, yeah... more work less talk :P kidding... no hurries...
I mean, if it shows the card's correct subsys id/ven (1522103C), e.g. on bdmesg, ioreg, etc...?
--- End quote ---
I believe the subssys id and vendor ID are correct. Didn't i attach a report in one of the previous posts?
Seems like i didn't, I will. Here is the ioreg -lfw0 for hoolock. I'll get the Langur one later.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version