Adrian, On Mon, Oct 3, 2016 at 6:36 PM, John Paul Adrian Glaubitz <glaub...@physik.fu-berlin.de> wrote: > On 10/01/2016 02:16 PM, Mathieu Malaterre wrote: >> 3. radeonfb should be able to kick out offb >> >> In case (2) above cannot be solved in time. We can also fix the color >> inversion another way. Technically some PowerMac uses a radeon based >> graphic card, so we could fix the radeonfb module issue so that it is >> properly loaded during d-i. >> >> So who can help me with: >> >> https://bugs.debian.org/826629 ? > > I will do a fresh install of Debian Stretch using the current d-i images > tomorrow on my > Mac Mini G4 to see what I can do. I have previously fixed some issues > regarding "radeonfb". > The primary problem back then was that "radeonfb" left the GPU in a weird > state from where > the "radeon" KMS module could not initialize it. > > Problem was that that CONFIG_FB_RADEON was configured with =y instead of =m > meaning that > I had to disable "radeonfb" on the kernel command line using > "video:radeonfb=off". This > is also mentioned on the Ubuntu Wiki [1].
I am glad you mention this ! In fact if you read carefully the red/blue color inversion, you'll discover this particular message: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=825840#59 I'd appreciate if you could also run the installer with offb:off. What is happening is that CONFIG_FB_RADEON=y make sure to use radeonfb (I was not able to find exactly why it skips offb), while CONFIG_FB_RADEON=m is doing something like this: 1. Try hardware framebuffer: offb 2. Try framebuffer: radeonfb However step (2) above fails as explained in #826629. Thus I fear CONFIG_FB_RADEON=m only fixes the *symptoms*. I would really appreciate if you could confirm that `radeonfb` properly works on your Mac Mini G4. -M
