xorg hangs on boot (feisty)

Bug #82875 reported by GiuseppeVerde
6
Affects Status Importance Assigned to Milestone
xserver-xorg-video-ati (Ubuntu)
Fix Released
Undecided
Unassigned

Bug Description

Binary package hint: xorg

When booting, xorg hangs. I get a black screen (not even a mouse cursor), and am unable to switch virtual consoles. When I use the SaK (Alt+SysRq+k), X dies, I'm asked if I want to start X on a new display (because the old one is in use). Upon selecting [No], X starts fine.

System: Inspiron 8600 with ATI Radeon Mobile 9600 (RV350). More data will be attached shortly.

Revision history for this message
GiuseppeVerde (launchpad-digitasaru) wrote :
Revision history for this message
GiuseppeVerde (launchpad-digitasaru) wrote :
Revision history for this message
GiuseppeVerde (launchpad-digitasaru) wrote :
Revision history for this message
GiuseppeVerde (launchpad-digitasaru) wrote :

kernel output:

[ 61.004000] Bluetooth: L2CAP ver 2.8
[ 61.004000] Bluetooth: L2CAP socket layer initialized
[ 61.060000] Bluetooth: HIDP (Human Interface Emulation) ver 1.1
[ 61.196000] Bluetooth: RFCOMM socket layer initialized
[ 61.196000] Bluetooth: RFCOMM TTY layer initialized
[ 61.196000] Bluetooth: RFCOMM ver 1.8
[ 62.196000] ip_tables: (C) 2000-2006 Netfilter Core Team
[ 62.496000] Netfilter messages via NETLINK v0.30.
[ 62.516000] nf_conntrack version 0.5.0 (8192 buckets, 65536 max)
[ 73.304000] SysRq : SAK
[ 73.308000] SAK: killed process 1 (init): fd#0 opened to the tty
[ 73.308000] SAK: killed process 2516 (logd): fd#0 opened to the tty
[ 62.496000] Netfilter messages via NETLINK v0.30.
[ 62.516000] nf_conntrack version 0.5.0 (8192 buckets, 65536 max)
[ 73.304000] SysRq : SAK
[ 73.308000] SAK: killed process 1 (init): fd#0 opened to the tty
[ 73.308000] SAK: killed process 2516 (logd): fd#0 opened to the tty
[ 73.308000] SAK: killed process 5192 (gdm): fd#6 opened to the tty
[ 73.764000] [drm] Initialized drm 1.1.0 20060810
[ 73.792000] ACPI: PCI Interrupt 0000:01:00.0[A] -> Link [LNKA] -> GSI 11 (lev
el, low) -> IRQ 11
[ 73.792000] [drm] Initialized radeon 1.25.0 20060524 on minor 0
[ 74.548000] agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0.
[ 74.548000] agpgart: Putting AGP V2 device at 0000:00:00.0 into 1x mode
[ 74.548000] agpgart: Putting AGP V2 device at 0000:01:00.0 into 1x mode
[ 74.852000] [drm] Setting GART location based on new memory map
[ 74.852000] [drm] Loading R300 Microcode
[ 74.852000] [drm] writeback test succeeded in 1 usecs
[ 78.860000] mtrr: no MTRR for d0000000,8000000 found
[ 83.344000] agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0.
[ 83.344000] agpgart: Putting AGP V2 device at 0000:00:00.0 into 1x mode
[ 83.344000] agpgart: Putting AGP V2 device at 0000:01:00.0 into 1x mode
[ 83.616000] [drm] Setting GART location based on new memory map
[ 83.616000] [drm] Loading R300 Microcode
[ 83.616000] [drm] writeback test succeeded in 1 usecs
[ 86.136000] input: Bluetooth HID Boot Protocol Device as /class/input/input7
[ 117.356000] ADDRCONF(NETDEV_CHANGE): ipw2100: link becomes ready
[ 126.164000] ieee80211_crypt: registered algorithm 'WEP'
[ 128.172000] ipw2100: no IPv6 routers present

Revision history for this message
Paddy de Búrca (paddy-deburca) wrote :

I can confirm a similar problem with ATI Radeon 9600.

If the graphic card driver is set to "ati" i get a blank screen - nothing works (that I know of) to regain control of the machine except the power button.

When I change the driver to "fbdev" it works.

I attach a tar file with
:. Xorg.0.log-ati
:. Xorg.0.log-fbdev
:. lspci-vvnn.txt
:. dmesg.log
:. uname-a.txt
:. xorg.conf

Paddy.

Revision history for this message
GiuseppeVerde (launchpad-digitasaru) wrote :

I no longer have a hang. paddy: is it working for you now too?

Revision history for this message
Paddy de Búrca (paddy-deburca) wrote :

Sorry to say - I still have the same problem.

I just changed fbdev with ati in the xorg.conf file and restarted X - black screen. I could not regain control of the machine and had to power cycle.

I am glad to see that your problem has been resolved.

I am performing a update/upgrade as I write but I don't see any X related updated. I will try again later all the same.

Paddy.

Revision history for this message
Paddy de Búrca (paddy-deburca) wrote :

Problem is solved!!

I understand that this is a feature that is added to the radeon driver by debian/ubuntu team(s). The code does not cater for ppc users and the driver tries to find/load an x86 bios even on a ppc machine.

The solution you may ask? Add the following to the ATI driver section
    Option "noint10" "yes"

Please find attached my xorg.conf file.

Any now even Beryl works like a dream.... and glxgears gives
6240 frames in 5.0 seconds = 1247.842 FPS
6435 frames in 5.0 seconds = 1286.823 FPS
6360 frames in 5.0 seconds = 1271.958 FPS
6508 frames in 5.0 seconds = 1301.523 FPS
6436 frames in 5.0 seconds = 1287.071 FPS

and I was getting over 1800 FPS too at one point :-)

Revision history for this message
Timo Aaltonen (tjaalton) wrote :

The original bug has been fixed in feisty, so I'll close this one.

Paddy, your bug is #93509.

Changed in xorg:
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.