| View previous topic :: View next topic |
| Author |
Message |
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
 Posted: Mon Jun 18, 2007 7:59 pm Post subject: Radeon 9200 on ppc no longer works with X11? |
 |
|
I have a PPC G4 that at one time had a working X setup with the "ATI Radeon Mac Edition 9200" in it. Lately, while keeping the software updated, I was using it as a server and didn't notice that X got broken along the way. I see some other have had problem and that there is a note in the FAQ about:
| Quote: | | Due to a mistake in creating the default options for GRP, the drivers for r128 and radeon were left turned off in the 2007.0 release. Please add the appropriate VIDEO_CARDS line to /etc/make.conf and recompile xorg-server and mesa to rebuild the driver. Sorry about the trouble! |
.As long as it's been possible, I had | Code: | | VIDEO_CARDS="radeon fbdev" |
in make.conf, so is the above suggestion to re-emerge media-libs/mesa?
The error I'm getting in Xorg.0.log is: | Code: | | (EE) No devices detected |
which would make me think the problem is something else, perhaps in the kernel, but I don't see anything obviously wrong there. I just recently changed to the 2007.0 profile, so perhaps there is something in that. Below are some more details, particularly if rebuilding mesa isn't the right solution to my problem.I These are just pieces of the puzzle. More complete details are in the files at http://www.pdc.kth.se/~mike/PPC_X11/. | Code: | zgrep '^CONFIG.*DRM' /proc/config.gz
CONFIG_DRM=m
CONFIG_DRM_TDFX=m
CONFIG_DRM_R128=m
CONFIG_DRM_RADEON=m
|
| Code: | # uname -a
Linux bunny 2.6.20-gentoo-r5 #1 Wed Apr 11 12:01:09 CEST 2007 ppc 7400, altivec supported PowerMac3,3 GNU/Linux
|
Installed:
| Code: | x11-drivers/xf86-video-ati 6.6.3
x11-base/xorg-server 1.2.0-r3
x11-base/xorg-x11 7.2 |
lspci | Code: |
0001:11:02.0 VGA compatible controller: ATI Technologies Inc RV280 [Radeon 9200] (rev 01) |
lsmod: | Code: | lsmod
Module Size Used by
ide_floppy 18176 0
uninorth_agp 9484 1
agpgart 30504 1 uninorth_agp
sungem 30436 0
sungem_phy 11552 1 sungem
|
I can modprobe "radeon" and "r128" without problem, but that does result in a different error message in Xorg.0.log.
As it is right now, I have only been able to get fbdev working with 8-bit depth. glxgears is so slow the gears are almost static.
Thanks for any tips! 
/iMike
PS: I re-emerged | Code: | x11-base/xorg-server-1.2.0-r3
media-libs/mesa-6.5.2-r1 |
but it made no difference. |
|
| Back to top |
|
 |
DirtyHairy
l33t
Joined: 03 Jul 2006
Posts: 608
Location: Würzburg, Deutschland
|
| Posted: Tue Jun 19, 2007 9:08 am Post subject: |
|
|
| Your problem cannot be connected to the gentoo 2007.0 issue if you didn't install the GRP packages, but just updated your profile (I'm on a G4 with a radeon 8500 and it works perfectly). Could you please post your xorg.conf? |
|
| Back to top |
|
|
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
| Posted: Tue Jun 19, 2007 9:41 am Post subject: |
|
|
The xorg.conf is in with the other details mentioned at http://www.pdc.kth.se/~mike/PPC_X11/.
Yes, I strongly suspect it's something else too. By the way, I only posted one example of xorg.conf I've tried, and the corresponding Xorg.0.log file. I did not post my currently working (though poorly) fbdev xorg.conf.
/iMike |
|
| Back to top |
|
|
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
| Posted: Tue Jun 19, 2007 10:06 am Post subject: |
|
|
Another thing to note is that my Radeon card is a PCI card, not an AGP card. There is nothing in the AGP slot of this machine. I don't know if that is what "No devices detected" is referring to or not. At the end of Xorg.0.log you also have "no screens found", but I am assuming this is a result of "no device is detected", meaning no graphics card, therefore no screens. Wrong assumption?
/iMike |
|
| Back to top |
|
|
JoseJX
Retired Dev
Joined: 28 Apr 2002
Posts: 2774
|
| Posted: Tue Jun 19, 2007 12:45 pm Post subject: |
|
|
Can I see the output of:
| Code: | | sudo lspci -s 0001:11:02.0 -n |
Perhaps the PCI ID of this card was accidentally dropped from the list of supported cards.
_________________
Gentoo PPC FAQ: http://www.gentoo.org/doc/en/gentoo-ppc-faq.xml |
|
| Back to top |
|
|
DirtyHairy
l33t
Joined: 03 Jul 2006
Posts: 608
Location: Würzburg, Deutschland
|
| Posted: Tue Jun 19, 2007 1:46 pm Post subject: |
|
|
| I second JoseJX. In addition, you could remove the busid line from xorg.conf, if you shifted the card around lately, it might have a different ID. |
|
| Back to top |
|
|
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
| Posted: Tue Jun 19, 2007 5:15 pm Post subject: |
|
|
bunny ~ # lspci -s 0001:11:02.0 -n
0001:11:02.0 0300: 1002:5961 (rev 01)
I'll try removing BusID "PCI:17:2:0" from xorg.conf too and see if that helps. I don't recall moving the card, but I could have sometime. I recall seeing once a command the translated the lspci numbers to the BusID numbers xorg uses, but I haven't stumbled upon that again.
Should it perhaps be helpful, lshw gives: | Code: | *-display
description: VGA compatible controller
product: RV280 [Radeon 9200]
vendor: ATI Technologies Inc
physical id: 2
bus info: pci@0001:11:02.0
version: 01
width: 32 bits
clock: 33MHz
capabilities: pm vga bus_master cap_list
configuration: driver=radeonfb latency=16 mingnt=8
|
and lspci -vvv gives: | Code: | 0001:11:02.0 VGA compatible controller: ATI Technologies Inc RV280 [Radeon 9200] (rev 01) (prog-if 00 [VGA])
Subsystem: ATI Technologies Inc RV280 [Radeon 9200]
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B-
Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
Latency: 16 (2000ns min), Cache Line Size: 32 bytes
Interrupt: pin A routed to IRQ 52
Region 0: Memory at 90000000 (32-bit, prefetchable) [size=256M]
Region 1: I/O ports at 1000 [size=256]
Region 2: Memory at 80090000 (32-bit, non-prefetchable) [size=64K]
Expansion ROM at 800a0000 [size=128K]
Capabilities: [50] Power Management version 2
Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 PME-Enable- DSel=0 DScale=0 PME-
|
Thanks for the help!
/iMike |
|
| Back to top |
|
|
JoseJX
Retired Dev
Joined: 28 Apr 2002
Posts: 2774
|
| Posted: Tue Jun 19, 2007 5:53 pm Post subject: |
|
|
| Code: | | #define PCI_CHIP_RV280_5961 0x5961 |
Hmm, nope. That's still in there.
The numbers lspci uses are in hex, X uses decimal, that's the only difference afaik. So, the 11 in your lspci output is the same as the 17 in your BusID line.
Try adding this to the device section of your X config:
| Code: | | Option "ForcePCIMode" "True" |
_________________
Gentoo PPC FAQ: http://www.gentoo.org/doc/en/gentoo-ppc-faq.xml |
|
| Back to top |
|
|
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
| Posted: Wed Jun 20, 2007 10:08 am Post subject: |
|
|
First I tried just removing the BusID line, but got the same result. Next I tried the ForcePCIMode option, but also no luck. I also commented out DRI, but nothing worked their either.
Although it worked previously, could it be something with the modules? I see from | Code: | bunny ~ # lsmod
Module Size Used by
sg 31584 0
ide_floppy 18176 0
uninorth_agp 9484 1
agpgart 30504 1 uninorth_agp
sungem 30436 0
sungem_phy 11552 1 sungem
bunny ~ # modprobe radeon
bunny ~ # lsmod
Module Size Used by
radeon 139080 0
drm 83832 1 radeon
sg 31584 0
ide_floppy 18176 0
uninorth_agp 9484 1
agpgart 30504 2 drm,uninorth_agp
sungem 30436 0
sungem_phy 11552 1 sungem
|
that when I load radeon, it loads drm, which appears dependent on agpgart, but I don't have a card in the AGP slot. A problem?
Looking at:
| Code: | bunny ~ # zgrep RADEON /proc/config.gz
CONFIG_DRM_RADEON=m
CONFIG_FB_RADEON=y
CONFIG_FB_RADEON_I2C=y
# CONFIG_FB_RADEON_DEBUG is not set
|
should I perhaps have everything as a module instead of built in, and perhaps turn on RADEON_DEBUG?
Just a few ideas. I may be searching in the dark here.
/iMike |
|
| Back to top |
|
|
DirtyHairy
l33t
Joined: 03 Jul 2006
Posts: 608
Location: Würzburg, Deutschland
|
| Posted: Wed Jun 20, 2007 11:30 am Post subject: |
|
|
Sorry, I am at a loss here. I doubt this problem is related to the drm modules; they are important only for 3D acceleration, 2D should work without them.
The only other idea I can produce is to disable all other framebuffer drivers apart from the radeon framebuffer, and perhaps (but this is really a desperate suggestion since I don't understand what it does) add "Option "useFBdev" "true"" to the device section (I'm under the impression that this makes the radeon driver aware of the framebuffer driver which also tries to access the card, but I'm happily corrected on that one) |
|
| Back to top |
|
|
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
| Posted: Thu Jun 21, 2007 10:30 am Post subject: |
|
|
I have now tried adding | Code: | | Option "useFBdev" "true" |
in xorg.conf, but still no luck. This is highly frustrating since I have a pair of older Macs (B&W G3s) running the same kernel, same X, but with older R128 cards and they work fine. One is running Xfce the other KDE. There, of course, the kernel is using:
| Code: | CONFIG_DRM=y
CONFIG_DRM_R128=y |
I guess the next stop is to re-compile the kernel on my G4 with RADEON_DEBUG and submit a bugzilla report. Or, are there other ideas? I still hold out hope it is just something stupid I am totally blind to just now.  |
|
| Back to top |
|
|
JoseJX
Retired Dev
Joined: 28 Apr 2002
Posts: 2774
|
| Posted: Thu Jun 21, 2007 12:28 pm Post subject: |
|
|
Can you try just running the command below? This will run the X server, but just return a list of PCI devices that it sees, it might be helpful in tracking this problem down.
_________________
Gentoo PPC FAQ: http://www.gentoo.org/doc/en/gentoo-ppc-faq.xml |
|
| Back to top |
|
|
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
| Posted: Thu Jun 21, 2007 7:44 pm Post subject: |
|
|
| Code: | bunny ~ # Xorg -scanpci
Probing for PCI devices (Bus:Device:Function)
(0:11:0) Apple Computer Inc. UniNorth AGP |
Given
| Code: | bunny ~ # lspci
0000:00:0b.0 Host bridge: Apple Computer Inc. UniNorth AGP
0001:10:0b.0 Host bridge: Apple Computer Inc. UniNorth PCI
0001:10:0d.0 PCI bridge: Digital Equipment Corporation DECchip 21154 (rev 05)
0001:11:02.0 VGA compatible controller: ATI Technologies Inc RV280 [Radeon 9200] (rev 01)
0001:11:07.0 Class ff00: Apple Computer Inc. KeyLargo Mac I/O (rev 03)
0001:11:08.0 USB Controller: Apple Computer Inc. KeyLargo USB
0001:11:09.0 USB Controller: Apple Computer Inc. KeyLargo USB
0001:11:0a.0 FireWire (IEEE 1394): Texas Instruments TSB12LV26 IEEE-1394 Controller (Link)
0002:21:0b.0 Host bridge: Apple Computer Inc. UniNorth Internal PCI
0002:21:0f.0 Ethernet controller: Apple Computer Inc. UniNorth GMAC (Sun GEM) (rev 01) |
does that mean it is only seeing the AGP slot? |
|
| Back to top |
|
|
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
| Posted: Thu Jun 21, 2007 9:09 pm Post subject: |
|
|
Some perhaps helpful excerpts from dmesg: | Code: |
...
Using PowerMac machine description
Total memory = 512MB; using 1024kB for hash table (at cff00000)
Linux version 2.6.20-gentoo-r5 (root@bunny) (gcc version 4.1.1 (Gentoo 4.1.1-r3)) #1 Wed Apr 11 12:01:09 CEST 2007
Found UniNorth memory controller & host bridge @ 0xf8000000 revision: 0x08
...
PowerMac motherboard: PowerMac G4 AGP Graphics
Entering add_active_range(0, 0, 131072) 0 entries of 256 used
Found UniNorth PCI host bridge at 0x00000000f0000000. Firmware bus number: 0->0
Found UniNorth PCI host bridge at 0x00000000f2000000. Firmware bus number: 0->1
Found UniNorth PCI host bridge at 0x00000000f4000000. Firmware bus number: 0->0
...
PCI: Enabling device 0001:11:02.0 (0006 -> 0007)
radeonfb: Found Open Firmware ROM Image
radeonfb: Retrieved PLL infos from Open Firmware
radeonfb: Reference=27.00 MHz (RefDiv=12) Memory=200.00 Mhz, System=250.00 MHz
radeonfb: PLL min 12000 max 35000
...
radeonfb: Monitor 1 type DFP found
radeonfb: EDID probed
radeonfb: Monitor 2 type no found
Display is GTF capable
Console: switching to colour frame buffer device 200x75
radeonfb (0001:11:02.0): ATI Radeon Ya
...
Linux agpgart interface v0.101 (c) Dave Jones
agpgart: Detected Apple UniNorth chipset
agpgart: configuring for size idx: 8
agpgart: AGP aperture is 32M @ 0x0 |
It appears the kernal, at least, finds the Radeon card and senses a GTF-capable display attached. |
|
| Back to top |
|
|
JoseJX
Retired Dev
Joined: 28 Apr 2002
Posts: 2774
|
|
| Back to top |
|
|
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
|
| Back to top |
|
|
iMike
Apprentice
Joined: 01 Apr 2005
Posts: 217
Location: Stockholm, Sweden
|
| Posted: Wed Jun 27, 2007 3:10 pm Post subject: |
|
|
No indication thus far that anyone is watching these bugs. Is this the "upstream" that you meant?
/iMike |
|
| Back to top |
|
|
JoseJX
Retired Dev
Joined: 28 Apr 2002
Posts: 2774
|
|
| Back to top |
|
|
|
Gentoo on PPC
