Live CD boots to a black screen on Blueberry G3 iMac

[Ian McCall]

Trying Breezy Badger Preview 5.10 on a Blueberry G3 iMac, http://lowendmac.com/imacs/blue.shtml. I
note three problems:

1. Just before the Ubuntu logo appears with the scrolling text beneath, a message appears saying
"Bummer: can't run /sbin/debian-installer"

2. With the Ubuntu logo on-screen a message flashes up "Initialising general console font - failed"

3. The screen turns to black - no X server or GNOME environment appears. This occurs even with the X
server told to restrict itself to 800x600, a resolution I know is supported on this machine.

[Dennis Kaarsemaker (dennis)]

Thank you for your bug report. The preview release has been obsoleted with
colony 5 and several daily cd's. Please retry with colony 5

[Matt Zimmerman (mdz)]

Please also try burning the image at a slower speed; many problems like this are
caused by CD-R media faults

[Ian McCall (ian-eruvia)]

(In reply to comment #2)
> Please also try burning the image at a slower speed; many problems like this are
> caused by CD-R media faults

I can confirm it's not the media, as the same Live CD works perfectly on my 867Mhz G4 Powerbook. I'll download the
newer image and give that a try tomorrow - will let people know how I get on.

[Matt Zimmerman (mdz)]

(In reply to comment #3)
> (In reply to comment #2)
> > Please also try burning the image at a slower speed; many problems like this are
> > caused by CD-R media faults
>
> I can confirm it's not the media, as the same Live CD works perfectly on my
867Mhz G4 Powerbook. I'll download the
> newer image and give that a try tomorrow - will let people know how I get on.

That is promising, but it isn't a litmus test. Many discs can be read in one
drive and not in another.

[twongkee (twongkee)]

also happens on iMac Snow DV with colony 5.

5.04 works fine on that system. burned the disk very slow 8x so no cd problems...

[Ian McCall (ian-eruvia)]

OK - retried it on the Blueberry. Same result with Colony 5, except that I now no longer have any failures regarding console
fonts or similar (ie. points 1 and 2 of the original report no longer occur).

I tried again using the options 'live video=ofonly'. This time I saw more errors, a large number were about missing symbols when
trying to start the X server, and it died with a signal 7. However, the log also warned that seeing missing symol reports might
not be the true cause of the error.

It still didn't boot with these options - it died, with the last on-screen message being 'Starting anac(h)ronistic scheduler'. I
don't know if that's actually related however, it's possible the X and GDM problems in the background caused it.

Shame I can't get the log off the machine - is there any way of starting a second console? I tried Apple-F2 but got nothing.

Cheers,
Ian

[Daniel Stone (daniels)]

mdz, can you please give these people directions for getting the information I
need to resolve this bug?

[Matt Zimmerman (mdz)]

The live CD spawns shells on consoles 1-6. On my powerpc system (I use a
standard US layout USB keyboard), alt+f1 through alt+f6 work just fine.

You can also try booting with "live casper-udeb/runlevel=S" to get to
single-user mode, and launch the X server by hand.

[twongkee (twongkee)]

got it working now; copied the screen freq from my xorg.conf from 5.04

[twongkee (twongkee)]

Created an attachment (id=4481)
X -probeonly on my iMac DV (snow G3, 500 Mhz ATI card)

[twongkee (twongkee)]

(From update of attachment 4481)
next is breezy

[twongkee (twongkee)]

Created an attachment (id=4482)
X -probeonly from iMac G3 DV; breezy (after I added the monitor settings
manually)

[twongkee (twongkee)]

Still present in release 5-10 live
workaround: go to console and vi /etc/X11/xorg.conf
replace HorizSync and VertRefresh in monitor section as follows:

Section "Monitor"
        Identifier "Generic Monitor"
        Option "DPMS"
        HorizSync 60-60
        VertRefresh 43-117
EndSection

restart X

[Daniel Stone (daniels)]

could you please also attach the output of sudo ddcprobe?

[twongkee (twongkee)]

(In reply to comment #14)
> could you please also attach the output of sudo ddcprobe?

ubuntu@ubuntu:~$ sudo ddcprobe
oem: ATY Rage128
memory: 8192kb
edid: 1 1
id: 059d
eisa: APP059d
serial: 01010101
manufacture: 0 1998
input: analog signal.
screensize: 27 20
gamma: 2.500000
dpms: non-RGB, no active off, no suspend, no standby
timing: 1024x768@75 Hz (VESA)
dtiming: 640x480@116
dtiming: 800x600@94
monitorrange: 60-60, 75-117
monitorname: iMac
(In reply to comment #14)
> could you please also attach the output of sudo ddcprobe?

[twongkee (twongkee)]

Still happens on Dapper, February release.

[twongkee (twongkee)]

ubuntu@ubuntu:~$ sudo ddcprobe
oem: ATY Rage128
memory: 8192kb
edid:
edid: 1 1
id: 059d
eisa: APP059d
serial: 01010101
manufacture: 0 1998
input: analog signal.
screensize: 27 20
gamma: 2.500000
dpms: non-RGB, no active off, no suspend, no standby
timing: 1024x768@75 Hz (VESA)
dtiming: 640x480@116
dtiming: 800x600@94
monitorrange: 60-60, 75-117
monitorname: iMac
ubuntu@ubuntu:~$

again sudo ddcprobe

[Matt Zimmerman (mdz)]

Please attach /var/log/Xorg.0.log and /etc/X11/xorg.conf

[twongkee (twongkee)]

/var/log/Xorg.0.log:
I) R128(0): Not using default mode "1024x768" (hsync out of range)
(II) R128(0): Not using default mode "512x384" (hsync out of range)
(II) R128(0): Not using default mode "1024x768" (hsync out of range)
(II) R128(0): Not using default mode "512x384" (hsync out of range)
(II) R128(0): Not using default mode "1024x768" (hsync out of range)
(II) R128(0): Not using default mode "512x384" (hsync out of range)
(II) R128(0): Not using default mode "1152x864" (hsync out of range)
(II) R128(0): Not using default mode "576x432" (hsync out of range)
(II) R128(0): Not using default mode "1280x960" (hsync out of range)
(II) R128(0): Not using default mode "640x480" (hsync out of range)
(II) R128(0): Not using default mode "1280x960" (hsync out of range)
(II) R128(0): Not using default mode "640x480" (hsync out of range)
(II) R128(0): Not using default mode "1280x1024" (hsync out of range)
(II) R128(0): Not using default mode "640x512" (hsync out of range)
(II) R128(0): Not using default mode "1280x1024" (hsync out of range)
(II) R128(0): Not using default mode "640x512" (hsync out of range)
(II) R128(0): Not using default mode "1280x1024" (hsync out of range)
(II) R128(0): Not using default mode "640x512" (hsync out of range)
(II) R128(0): Not using default mode "1600x1200" (hsync out of range)
(II) R128(0): Not using default mode "800x600" (hsync out of range)
(II) R128(0): Not using default mode "1600x1200" (hsync out of range)
(II) R128(0): Not using default mode "800x600" (hsync out of range)
(II) R128(0): Not using default mode "1600x1200" (hsync out of range)
(II) R128(0): Not using default mode "800x600" (hsync out of range)
(II) R128(0): Not using default mode "1600x1200" (hsync out of range)
(II) R128(0): Not using default mode "800x600" (hsync out of range)
(II) R128(0): Not using default mode "1600x1200" (hsync out of range)
(II) R128(0): Not using default mode "800x600" (hsync out of range)
(II) R128(0): Not using default mode "1792x1344" (insufficient memory for mode)
(II) R128(0): Not using default mode "896x672" (hsync out of range)
(II) R128(0): Not using default mode "1792x1344" (insufficient memory for mode)
(II) R128(0): Not using default mode "896x672" (hsync out of range)
(II) R128(0): Not using default mode "1856x1392" (insufficient memory for mode)
(II) R128(0): Not using default mode "928x696" (hsync out of range)
(II) R128(0): Not using default mode "1856x1392" (insufficient memory for mode)
(II) R128(0): Not using default mode "928x696" (hsync out of range)
(II) R128(0): Not using default mode "1920x1440" (insufficient memory for mode)
(II) R128(0): Not using default mode "960x720" (hsync out of range)
(II) R128(0): Not using default mode "1920x1440" (insufficient memory for mode)
(II) R128(0): Not using default mode "960x720" (hsync out of range)
(II) R128(0): Not using default mode "832x624" (vrefresh out of range)
(II) R128(0): Not using default mode "416x312" (vrefresh out of range)
(II) R128(0): Not using default mode "1152x864" (hsync out of range)
(II) R128(0): Not using default mode "576x432" (hsync out of range)
(II) R128(0): Not using default mode "1400x1050" (hsync out of range)
(II) R128(0): Not ...

[twongkee (twongkee)]

default (unedited) xorg.conf, that does not work, is the one generated by the live CD:

# /etc/X11/xorg.conf (xorg X Window System server configuration file)
#
# This file was generated by dexconf, the Debian X Configuration tool, using
# values from the debconf database.
#
# Edit this file with caution, and see the /etc/X11/xorg.conf manual page.
# (Type "man /etc/X11/xorg.conf" at the shell prompt.)
#
# This file is automatically updated on xserver-xorg package upgrades *only*
# if it has not been modified since the last upgrade of the xserver-xorg
# package.
#
# If you have edited this file but would like it to be automatically updated
# again, run the following command:
# sudo dpkg-reconfigure -phigh xserver-xorg

Section "Files"
        FontPath "/usr/share/X11/fonts/misc"
        FontPath "/usr/share/X11/fonts/cyrillic"
        FontPath "/usr/share/X11/fonts/100dpi/:unscaled"
        FontPath "/usr/share/X11/fonts/75dpi/:unscaled"
        FontPath "/usr/share/X11/fonts/Type1"
        FontPath "/usr/share/X11/fonts/100dpi"
        FontPath "/usr/share/X11/fonts/75dpi"
        # path to defoma fonts
        FontPath "/var/lib/defoma/x-ttcidfont-conf.d/dirs/TrueType"
EndSection

Section "Module"
        Load "i2c"
        Load "bitmap"
        Load "ddc"
        Load "dri"
        Load "extmod"
        Load "freetype"
        Load "glx"
        Load "int10"
        Load "type1"
        Load "vbe"
EndSection

Section "InputDevice"
        Identifier "Generic Keyboard"
        Driver "kbd"
        Option "CoreKeyboard"
        Option "XkbRules" "xorg"
        Option "XkbModel" "pc104"
        Option "XkbLayout" "us"
EndSection

Section "InputDevice"
        Identifier "Configured Mouse"
        Driver "mouse"
        Option "CorePointer"
        Option "Device" "/dev/input/mice"
        Option "Protocol" "ExplorerPS/2"
        Option "Emulate3Buttons" "true"
EndSection

Section "Device"
        Identifier "ATI Technologies, Inc. Rage 128 PR/PRO AGP 4x TMDS"
        Driver "ati"
        Option "UseFBDev" "true"
EndSection

Section "Monitor"
        Identifier "Generic Monitor"
        Option "DPMS"
        HorizSync 28-51
        VertRefresh 43-60
EndSection

Section "Screen"
        Identifier "Default Screen"
        Device "ATI Technologies, Inc. Rage 128 PR/PRO AGP 4x TMDS"
        Monitor "Generic Monitor"
        DefaultDepth 24
        SubSection "Display"
                Depth 1
                Modes "1024x768" "800x600" "640x480"
        EndSubSection
        SubSection "Display"
                Depth 4
                Modes "1024x768" "800x600" "640x480"
        EndSubSection
        SubSection "Display"
                Depth 8
                Modes "1024x768" "800x600" "640x480"
        EndSubSection
        SubSection "Display"
                Depth ...

[twongkee (twongkee)]

didn't notice the attachment link on the new interface...
sorry.

[hackerb9 (benjamin-a-wong)]

Bug still exists on the latest powerpc live CD, Dapper Flight 5, dated March 10, 2006.

The xorg.conf file is not getting the monitor timings from ddcprobe. Editing it by hand to reflect the output of ddcprobe works.

What creates the xorg.conf file?

[Henrik Nilsen Omma (henrik)]

We seem to have useful debugging infor for dapper -> setting to confirmed. The failure of ddcprobe to get the freq. is of course a problem.

[Fabio Massimo Di Nitto (fabbione)]

I fixed this bug between beta1 and beta2 of dapper. The problem was that we were not writing sync ranges on powerpc due to a wrong check. We do now.

Fabio

[Fabio Massimo Di Nitto (fabbione)]

Re-opening.

[RHTopics (tyrht01)]

This problem occurs with the ShipIt CD, "Version 6.06 LTS for your Mac".

[ReverendJW (reverendjw)]

I have a iMac DV (G3) and ran into this problem using Kubuntu 6.06 desktop image. I am happy to report that the freq workaround mentioned above (https://launchpad.net/distros/ubuntu/+source/xorg/+bug/22976/comments/13) successfully corrected the problem for me.

I have attached my xorg.conf file which is unaltered from initial load except for the freq settings.

[Jamie Kahn Genet (jamiekg)]

On my Rev c tray-loading iMac 233MHz G3 with 512MB RAM:

Using the Ubuntu 5.10 LiveCD I encoutered the problem decribed here: <http://ubuntuforums.org/showthread.php?t=75604&highlight=imac+blank+screen> and used the author's fix to get it working. Sadly the same bug - so far as I can tell - still exists in my new 6.06 Live/Install CD. The fix that worked in 5.10 no longer works for me :-(

[Jamie Kahn Genet (jamiekg)]

ok, tried on all my old CRT iMacs. I do the following:

Quote:
Originally Posted by twongkee
drop to console (ctrl - option F1)
** this means once you boot up to a blank screen hit the 'ctrl' key, the 'option' key and the F1 key all at the same time. you'll be in a terminal window. **

** if you need to log in use the name ubuntu to log in. **

and edit the X config file as follows:

** type this in exactly as you see it (vi is a terminal editor)**
sudo vi /etc/X11/xorg.conf

change the frequencies in monitor section as follows:
** find this part in the file you are now in (xorg.conf) and make these changes **

Section "Monitor"
Identifier "Generic Monitor"
Option "DPMS"
HorizSync 60-60
VertRefresh 43-117
EndSection

** after the changes then hit the 'esc' or 'escape' key, then type ':wq'
** no 'quotes' -- but yes the ':' is part of the command
** that will write the changes to the file and quit from vi **

** restart X by typing: sudo /etc/init.d/gdm restart
** or maybe just hit 'ctrl' 'alt' 'delete'
that should be it. now ubuntu should start right up.

sadly under 6.06 it doesn't restart. Instead I get, upon saving the changes to /etc/X11/xorg.conf, and then typing sudo /etc/init.d/gdm restart the following error: "Starting GNOME Display Manager... [fail]"

[Joakim Larsson (joakim-bildrulle)]

As far as I can tell from my recent experiences the black screen means that the ATI r128 drivers uses the VGA output instead of the builtin CRT. I speculate that this is a PC inheritance. The VGA output is well hidden under a hatch at the back, on my 400 MHz grape iMac DV at least, so I attached a VGA screen there. The X output nicelly appeared on the VGA screen only.The framebuffer driver used before X is up outputs console on both outputs. Unfortunality a bug:

https://launchpad.net/distros/ubuntu/+source/xserver-xorg-video-ati/+bug/63072

inhibits the UseFBDev option wich has to be turned to "false" in the xorg.conf and the Monitor options doesn't seem to have any effect.

[Dustin Kirkland  (kirkland)]

I was having the problems described above installing Edubuntu Edgy on some of my wife's school's older G3 iMacs (400-500MHz, 256MB RAM). Note that with 128MB of RAM, these machines were basically useless (256MB and they're usable).

I managed to work around the problem described above as follows:
- Boot the Edubuntu 6.10 Edgy LiveCD
- At the boot prompt, specify single user mode with the "single" kernel option
  boot: live single
- Once the machine comes up to a shell, Change the following in /etc/X11/xorg.conf (patch attached)
  HorizSync 60-60
  VertRefresh 75-117
  and
  DefaultDepth 15
- startx

After that, I was able to run the live distro, and successfully perform an installation. Note that at this point, you're running X as root, which is generally not preferred).

Following the installation, the xorg.conf changes were preserved, which was nice. However, the additional kernel boot parameter "single" was also preserved. So after the installation completed, the machine rebooted again into single-user mode where I had to remove the "single" kernel boot option and reinstall yaboot via ybin.
# sed -i "s/single//" /etc/yaboot.conf
# ybin --verbose
# reboot

After that, my installation worked fine.

----
To the Ubuntu developers... The proper monitor ranges should be detectable from ddcprobe at boot time on these machines, so I'd think you ought to be able to populate the HorizSync and VertRefresh properly in the next Fiesty ISOs. As for the DefaultDepth... This might ultimately be a problem with X for not failing back to a lower, working color depth. But for the sake of Fiesty ISOs booting on G3 iMacs (plenty still hanging around our school systems at least), any chance you could detect them, and special case their DefaultDepth down to 15bit?

Thanks,
:-Dustin

[Timothy Smith (tas50)]

Can someone test this under a Feisty build so we can see if this bug still exists?

[Ian McCall (ian-eruvia)]

I can test under Feisty over the next weekend - will post the results here as soon as done.

[Ian McCall (ian-eruvia)]

Well, I finally got a chance to try but the PPC beta disc I tried would no longer boot the G3 at all - I got a Kernel Loading Failure on the Blueberry G3 iMac I tried this with.

Can anyone point me to a working PPC Feisty ISO?

[James Powell (powellj)]

I wish to raise awareness that setting the VertRefresh to 75-117 on a G3 iMac makes it possible for X to drive the CRT at refresh rates that are not supported by the hardware. It appears to run well but a month or so later your CRT is killed by this. I let the magic smoke out of one iMac and I had almost killed a second one when I realized that the refresh rate is probably to blame.

I do not have a good answer as to the right information to put into xorg.conf, but I am %95 certain that 75-117 is not it.

[James Powell (powellj)]

I have to retract my previous post. I did manage to destroy one and a half CRT iMacs but it was not at all Ubuntu-related. The good folks at FreeGeek confirmed for me that they never see these problems.

What explains my explosions I think is the accumulation of dust on a coil related to the CRT. The dust causes the coil to conduct where it should not, and it heats up and eventually shorts out.

I saved the second iMac by opening it up and thoroughly dusting it out with an air duster. It now runs without any overheating or display glitches.

As a humble karmic offering for my over-reaction allow me to offer these ModeLine entries which when added to /etc/X11/xorg.conf give Feisty Fawn the ability to change monitor resolutions using System/Settings/Screen Resolution on a slot-loading iMac:

---<begin>---
#(**) ATI(0): *Mode "1024x768": 78.5 MHz, 59.8 kHz, 74.8 Hz
Modeline "1024x768" 78.50 1024 1033 1129 1312 768 769 772 800 +hsync +vsync
# (**) ATI(0): *Mode "800x600": 62.4 MHz, 60.0 kHz, 94.9 Hz
Modeline "800x600" 62.40 800 825 905 1040 600 601 604 632 +hsync +vsync
# (**) ATI(0): *Mode "640x480": 49.9 MHz, 60.0 kHz, 116.7 Hz
Modeline "640x480" 49.90 640 653 717 832 480 481 484 514 +hsync +vsync
# (**) ATI(0): Default mode "512x384": 39.4 MHz, 60.1 kHz, 75.1 Hz (D)
Modeline "512x384" 39.40 512 520 568 656 384 384 386 400 doublescan +hsync +vsync
---<end>---

Add those to Section "Monitor" and you are good to go.

cheers,
  James

[Timo Aaltonen (tjaalton)]

Lowering the priority, since ppc is not an officially supported architecture.

[Timo Jyrinki (timo-jyrinki)]

It would be nice if you could try Ubuntu 8.04 LTS development version's live CD and see if this problem is finally fixed. For example the current one is at: http://cdimage.ubuntu.com/ports/daily-live/20080310/hardy-desktop-powerpc.iso

The link will be gone in a few days, after that the latest one can be found in subdirectories of http://cdimage.ubuntu.com/ports/daily-live/

Ubuntu 8.04 has a new 6.8.0 version of xserver-xorg-video-ati, which has a very large number of bug fixes etc.

[Ian McCall (ian-eruvia)]

Tested using your 20080310 download link - same problem.

The bootup process looks like:

1. The brown desktop background flashes up for a second or so
2. The screen then blanks and I get the text boot options
3. After a while, screen turns white for a few seconds then goes black again
4. A text cursor can be seen flashing in the top-left of the screen

That's it - nothing else happens. Tested on the Blueberry G3 iMac that I used when originally filing this bug.

[Jamie Kahn Genet (jamiekg)]

On 10/03/2008, at 10:47 PM, Ian McCall wrote:

> Tested using your 20080310 download link - same problem.
>
> The bootup process looks like:
>
> 1. The brown desktop background flashes up for a second or so
> 2. The screen then blanks and I get the text boot options
> 3. After a while, screen turns white for a few seconds then goes
> black again
> 4. A text cursor can be seen flashing in the top-left of the screen
>
> That's it - nothing else happens. Tested on the Blueberry G3 iMac
> that I
> used when originally filing this bug.
>
> --
> Live CD boots to a black screen on Blueberry G3 iMac
> https://bugs.launchpad.net/bugs/22976
> You received this bug notification because you are a direct subscriber
> of the bug.

Similar thing is still happening here on a lime Rev c G3 iMac.

Regards,
  Jamie Kahn Genet
--
"The surest sign that intelligent life exists elsewhere in the
universe is that it has never tried to contact us." - Calvin and
Hobbes (Bill Watterson)

[Niels Kristian Bech Jensen (nkbjensen)]

The bug is still here on my Blueberry G3 iMac (slotloading) with the hardy 20080416 live cd. I found two problems:

1. The BusID was set to PCI:6:3:0 in xorg.conf - it should have been PCI:0:16:0.

2. Even with the correct BusID in xorg.conf the sync freqs were not probed correctly.

Here are the changes I had to make to xorg.conf to get X started:

--- xorg.conf.orig 2008-04-17 05:48:45.820016000 +0000
+++ xorg.conf 2008-04-17 06:53:33.362385848 +0000
@@ -31,12 +31,14 @@

 Section "Device"
        Identifier "Configured Video Device"
- BusID "PCI:6:3:0"
+ BusID "PCI:0:16:0"
        Option "UseFBDev" "true"
 EndSection

 Section "Monitor"
        Identifier "Configured Monitor"
+ HorizSync 60-60
+ VertRefresh 75-117
 EndSection

 Section "Screen"

I have attached the original Xorg.0.log file from the live cd before changing xorg.conf.

[Bryce Harrington (bryce)]

Is this still an issue with Intrepid?

[Dustin Kirkland  (kirkland)]

On Wed, Jul 30, 2008 at 11:04 AM, Bryce Harrington
<email address hidden> wrote:
> Is this still an issue with Intrepid?

Sorry, I no longer have access to the hardware :-(

:-Dustin

[Ian McCall (ian-eruvia)]

On 30 Jul 2008, at 17:04, Bryce Harrington wrote:

> Is this still an issue with Intrepid?
>
> ** Changed in: xserver-xorg-video-ati (Ubuntu)
> Status: Confirmed => Incomplete
>
> --
> Live CD boots to a black screen on Blueberry G3 iMac
> https://bugs.launchpad.net/bugs/22976
> You received this bug notification because you are a direct subscriber
> of the bug.

Will try in next day or so and update.

Ian

[Jamie Kahn Genet (jamiekg)]

I can confirm that it is.

On 31/07/2008, at 4:22 AM, Dustin Kirkland wrote:

> On Wed, Jul 30, 2008 at 11:04 AM, Bryce Harrington
> <email address hidden> wrote:
>> Is this still an issue with Intrepid?
>
> Sorry, I no longer have access to the hardware :-(
>
> :-Dustin
>
> --
> Live CD boots to a black screen on Blueberry G3 iMac
> https://bugs.launchpad.net/bugs/22976
> You received this bug notification because you are a direct subscriber
> of the bug.

--
"On two occasions, I have been asked [by members of Parliament],
'Pray, Mr. Babbage, if you put into the machine wrong figures, will
the right answers come out?' I am not able to rightly apprehend the
kind of confusion of ideas that could provoke such a question."
- Charles Babbage (1791-1871)

[Niels Kristian Bech Jensen (nkbjensen)]

It is still not fixed in intrepid.

[Bryce Harrington (bryce)]

Forwarding this bug from a Ubuntu user (other users report similar behavior on various iMac flavors):
https://bugs.edge.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/22976

"I found two problems:

1. The BusID was set to PCI:6:3:0 in xorg.conf - it should have been PCI:0:16:0.

2. Even with the correct BusID in xorg.conf the sync freqs were not probed correctly.

Here are the changes I had to make to xorg.conf to get X started:

--- xorg.conf.orig 2008-04-17 05:48:45.820016000 +0000
+++ xorg.conf 2008-04-17 06:53:33.362385848 +0000
@@ -31,12 +31,14 @@

 Section "Device"
        Identifier "Configured Video Device"
- BusID "PCI:6:3:0"
+ BusID "PCI:0:16:0"
        Option "UseFBDev" "true"
 EndSection

 Section "Monitor"
        Identifier "Configured Monitor"
+ HorizSync 60-60
+ VertRefresh 75-117
 EndSection

 Section "Screen"

Here is the original Xorg.0.log file from the live cd before changing xorg.conf:
http://launchpadlibrarian.net/13534788/Xorg.0.log

[agd5f (agd5f)]

This is an r128 chip not a radeon. Secondly, these look to be issues with the livecd and xorg.conf rather than driver issues.

[Bryce Harrington (bryce)]

I've gone ahead and forwarded the bug upstream here. Please subscribe to that bug report so if upstream has further questions or needs you to test something, you can work with them directly:

https://bugs.freedesktop.org/show_bug.cgi?id=17344

Also, it would be helpful if you could update to the latest Intrepid (Alpha-4 + updates), re-test, and confirm the issue still occurs. We uploaded Debian's latest git snapshot of the -ati driver yesterday, and upstream would probably appreciate it if you could verify against that.

[Bryce Harrington (bryce)]

Upstream says it's a configuration issue, not a driver issue. So you'll need to run an alpha-4 or newer livecd (or do a fresh installation) and re-confirm the problem. Also please attach your /var/log/Xorg.0.log and /etc/X11/xorg.conf after reproducing the problem.

[Bryce Harrington (bryce)]

Also, this was mis-filed against -ati, not -r128 as it should have been.

[randyworkman (randyworkman)]

I can confirm that Live CD boots to a black screen in both i-mac/400mhz and e-mac/1ghz from 5.10 to 8.04. 5.04 boots normally. The e-mac has ATI graphics. None of the workarounds posted on related bug topics have yielded any results.

[Niels Kristian Bech Jensen (nkbjensen)]

We cannot test the original bug with Intrepid since there are no PowerPC live CDs.

[Niels Kristian Bech Jensen (nkbjensen)]

Still not fixed in jaunty 20081226. Now it boots to a black screen frosen solid (I cannot change to a VT by ctrl-alt-F1).

[Niels Kristian Bech Jensen (nkbjensen)]

Installing from jaunty 20090112 alternate left me with an empty (zero length) /etc/X11/xorg.conf file. I copied the /etc/X11/xorg.conf.failsafe file to /etc/X11/xorg.conf and rebooted. This gave me a usable screen with a grainy look and some strange color artifacts.

I have attached the /etc/X11/xorg.conf.failsafe and corresponding /var/log/Xorg.0.log files.

[Tormod Volden (tormodvolden)]

The Jaunty PowerPC live CD has recently been updated. Can you please try it? http://cdimage.ubuntu.com/ports/daily-live/current/

General instructions for debugging "black screen" issues from a live CD:

Please try a Jaunty live CD: http://cdimage.ubuntu.com/releases/jaunty/

At the CD boot menu, add "text" to the boot options:
https://help.ubuntu.com/community/BootParameters

The CD will now boot but not start X. At the text prompt, enter this command:
 xinit -e 'sh -c "xrandr --verbose > xrandr.log" '
This will try to start X and log some information about the active screen resolution to the file xrandr.log before it stops X again. Copy the corresponding X log to the current directory:
 cp /var/log/Xorg.0.log .

Now edit xorg.conf to make X use the "vesa" (failsafe) driver:
 nano /etc/X11/xorg.conf
and change the Device section to look like this:

Section "Device"
       Identifier "Configured Video Device"
       Driver "vesa"
EndSection

Then start X which should give you a desktop session, although with low resolution:
 startx

>From here, attach the xrandr.log and Xorg.0.log (which are in the Home
Folder) to this bug report. If for some reason you can not log into
launchpad from the live CD (network problems etc), you can also copy the
files to a USB drive now and attach them from another computer or
session later.

See also https://wiki.ubuntu.com/X/Troubleshooting/BlankScreen

[Niels Kristian Bech Jensen (nkbjensen)]

There is an improvement with the 20090321 daily-live cdrom. It boots into a lowres black X screen with the message:

Ubuntu is running in low-graphics mode.
Your screen, graphics card, and input device settings
could not be detected correctly. You will need to
configure these yourself.

I will investigate this further.

[Tormod Volden (tormodvolden)]

Sounds like the r128 driver fails to start, and the server restarts with the "vesa" driver. Please attach the files as described in my previous comment.

[Niels Kristian Bech Jensen (nkbjensen)]

The line

xinit -e 'sh -c "xrandr --verbose > xrandr.log" '

does not write a xrandr.log file. I have attached the /var/log/Xorg.0.log file. The test was done with the 20090321 daily live cd with boot parameters "live-nosplash text" ("live-nosplash" due to bug #216666).

[Niels Kristian Bech Jensen (nkbjensen)]

Here is an image of the screen output from xinit -e 'sh -c "xrandr --verbose > xrandr.log" '.

[Tormod Volden (tormodvolden)]

Niels, can you please report this (No devices detected) upstream at bugs.freedesktop.org and attach your log there? Attach also the output from "lspci -vvnn".

[Niels Kristian Bech Jensen (nkbjensen)]

It works with the jaunty 20090331 daily live cdrom with a somewhat grainy look though. It uses the fbdev driver as shown in the attached /var/log/Xorg.0.log file.

[Tormod Volden (tormodvolden)]

Thanks. I don't know why the fbdev driver is now picked. It is supposed to work with the r128 driver, so I will leave this bug open.

[Niels Kristian Bech Jensen (nkbjensen)]

Until the problem with the r128 driver is fixed I suggest you put this line in the Screen section of /etc/X11/xorg.conf to avoid the color artifacts caused by the default 8 bit framebuffer:

DefaultFbBpp 32

[Thomas Ribbrock (emgaron+ubuntu)]

I wonder whether in Intrepid this is related to the bug report I filed a few days ago (I could not find this one then): https://bugs.launchpad.net/bugs/398204 . As far as I remember, I have already tried playing with "DefaultFbBpp" to no avail, but I'll re-check, just to be sure.

[Thomas Ribbrock (emgaron+ubuntu)]

Ok, I've re-installed 9.04PPC on my iMac and tried this again. "DefaultFbBpp" does *not* work. If I add this statement to the screen section, X fails with this error:

(II) FBDEV(0): using default device
(II) Running in FRAMEBUFFER Mode
(--) FBDEV(0): Depth 24, (**) framebuffer bpp 32
(==) FBDEV(0): RGB weight 888
(==) FBDEV(0): Default visual is TrueColor
(==) FBDEV(0): Using gamma correction (1.0, 1.0, 1.0)
(II) FBDEV(0): hardware: OFfb ATY,Rage12 (video memory: 600kB)
(II) FBDEV(0): checking modes against framebuffer device...
(II) FBDEV(0): mode "1024x768" test failed
(II) FBDEV(0): mode "800x600" test failed
(II) FBDEV(0): mode "640x480" test failed
(II) FBDEV(0): checking modes against monitor...
(--) FBDEV(0): Virtual size is 800x600 (pitch 800)
(**) FBDEV(0): Built-in mode "current": 100.0 MHz, 119.0 kHz, 186.0 Hz
(II) FBDEV(0): Modeline "current"x0.0 100.00 800 816 824 840 600 616 624 640 -hsy
nc -vsync -csync (119.0 kHz)
(==) FBDEV(0): DPI set to (96, 96)
(II) Loading sub module "fb"
(II) LoadModule: "fb"
(II) Loading /usr/lib/xorg/modules//libfb.so
(II) Module fb: vendor="X.Org Foundation"
        compiled for 1.6.0, module version = 1.0.0
        ABI class: X.Org ANSI C Emulation, version 0.4
(**) FBDEV(0): using shadow framebuffer
(II) Loading sub module "shadow"
(II) LoadModule: "shadow"
(II) Loading /usr/lib/xorg/modules//libshadow.so
(II) Module shadow: vendor="X.Org Foundation"
        compiled for 1.6.0, module version = 1.1.0
        ABI class: X.Org ANSI C Emulation, version 0.4
(==) Depth 24 pixmap format is 32 bpp
(EE) FBDEV(0): FBIOPUT_VSCREENINFO succeeded but modified mode
(EE) FBDEV(0): mode initialization failed

[Niels Kristian Bech Jensen (nkbjensen)]

The problem is still present in lucid daily-live 20100408.

[Niels Kristian Bech Jensen (nkbjensen)]

Here is the /var/log/Xorg.0.log file from booting with the lucid 20100411 desktop cdrom. Let me know if you need other information to fix this problem.

[Niels Kristian Bech Jensen (nkbjensen)]

It seems the r128 driver crashes at startup on Lucid release. The attached xorg.conf can be used as a workaround.

[Tormod Volden (tormodvolden)]

Is anything else than UseFBDev needed?

[Niels Kristian Bech Jensen (nkbjensen)]

The HorizSync and VertRefresh lines are needed to get the correct resolution (1024x768@75Hz).

[Tormod Volden (tormodvolden)]

The need for HorizSync and VertRefresh lines is probably due to failing DDC probing of the monitor, and would be a separate bug. This is unfortunately seen for several drivers/cards/monitor combinations.

[Niels Kristian Bech Jensen (nkbjensen)]

The real bug is the r128 server crashing. Could the system be made to default to the fbdev server when that happens?

I will try to get some debug information on the crash.

[Niels Kristian Bech Jensen (nkbjensen)]

Here are the /var/log/Xorg.0.log file from the Lucid release desktop cdrom and the minimal xorg.conf file to work around the bug. A curious (but probably unrelated) observation is that using

Driver "fbdev"

in xorg.conf gives a grainy looking screen (seems like 8-bit depth) while

Option "UseFBDev" "true"

gives the correct colors.

[Bryce Harrington (bryce)]

This bug report seems to have devolved into an amalgam of different issues, and the upstreamed bug report is closed as invalid due to being not a driver problem but rather xorg.conf problems.

Re-test against natty and file new bug reports for each separate problem.