System lockup with catalyst 8.01

[jasonxh]

Watch out for authatieventsd.sh & atieventsd!

In order to help track down the problem, I installed this new driver again. I did all the tricks bridgman suggested in xorg.conf, still no go. I noticed a subtle difference as compared to 8.40.4, i.e. the last OLD codebase driver which I'm still stick to for all these months. It's this authatieventsd.sh script. Using the OLD driver, I never see it in running processes, so I assume it's only run at the X startup, and quits when it finishes its job. But in the NEW driver, this process keeps running

sh -c /etc/ati/authatieventsd.sh grant :0 (null)

So I did some experiments after switching to console

1. Issue a gdm stop, which takes CONSIDERABLY longer than with the OLD driver. However, X is still running and I can not kill it.
2. Manually kill the authatieventsd.sh process, and X is automatically released. (Note that authatieventsd.sh will be respawned by atieventsd immediately, which is OK)
3. Issue a gdm start, which seems normal.

And whoops, the gdm (and X) was successfully restarted. While I don't know if authatieventsd.sh is supposed to be running all the time in the new driver, I'm almost sure that it has something to do with the freezing problem. Hope this gives clues.

My xorg.conf and Xorg.0.log should give you every detail of my configuration and hardware.

xorg.conf:

# /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 xorg.conf(5) manual page.
# (Type "man 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 "ServerLayout"
	Identifier     "Default Layout"
	Screen         "Default Screen" 0 0
	InputDevice    "Generic Keyboard"
	InputDevice    "Configured Mouse"
	InputDevice    "Logitech VX Revolution"
	InputDevice    "stylus" "SendCoreEvents"
	InputDevice    "cursor" "SendCoreEvents"
	InputDevice    "eraser" "SendCoreEvents"
	InputDevice    "Synaptics Touchpad"
EndSection

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

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

Section "ServerFlags"
	Option	    "AIGLX" "off"
EndSection

Section "InputDevice"
	Identifier  "Generic Keyboard"
	Driver      "kbd"
	Option	    "CoreKeyboard"
	Option	    "XkbRules" "xorg"
	Option	    "XkbModel" "pc105"
	Option	    "XkbLayout" "us"
EndSection

Section "InputDevice"
	Identifier  "Configured Mouse"
	Driver      "mouse"
	Option	    "CorePointer"
	Option	    "Device" "/dev/input/mice"
	Option	    "Protocol" "ImPS/2"
	Option	    "Emulate3Buttons" "true"
	Option	    "EmulateWheel" "true"
	Option	    "EmulateWheelButton" "2"
	Option	    "XAxisMapping" "6 7"
	Option	    "YAxisMapping" "4 5"
EndSection

Section "InputDevice"
	Identifier  "Logitech VX Revolution"
	Driver      "evdev"
	Option	    "SendCoreEvents" "true"
	Option	    "vendor" "0x046d"
	Option	    "product" "0xc521"
	Option	    "version" "0x0111"
	Option	    "evBits" "+0-2 -3"
	Option	    "HWHEELRelativeAxisButtons" "7 6"
EndSection

Section "InputDevice"
	Identifier  "Synaptics Touchpad"
	Driver      "synaptics"
	Option	    "SendCoreEvents" "true"
	Option	    "Device" "/dev/psaux"
	Option	    "Protocol" "auto-dev"
	Option	    "HorizScrollDelta" "0"
	Option	    "SHMConfig" "true"
EndSection

Section "InputDevice"
	Identifier  "stylus"
	Driver      "wacom"
	Option	    "Device" "/dev/input/wacom"
	Option	    "Type" "stylus"
	Option	    "ForceDevice" "ISDV4"# Tablet PC ONLY
EndSection

Section "InputDevice"
	Identifier  "eraser"
	Driver      "wacom"
	Option	    "Device" "/dev/input/wacom"
	Option	    "Type" "eraser"
	Option	    "ForceDevice" "ISDV4"# Tablet PC ONLY
EndSection

Section "InputDevice"
	Identifier  "cursor"
	Driver      "wacom"
	Option	    "Device" "/dev/input/wacom"
	Option	    "Type" "cursor"
	Option	    "ForceDevice" "ISDV4"# Tablet PC ONLY
EndSection

Section "Monitor"
	Identifier   "Builtin Monitor"
	HorizSync    30.0 - 67.0
	VertRefresh  50.0 - 75.0
	Option	    "DPMS"
EndSection

Section "Device"
	Identifier  "ATI Mobility Radeon X1300 Internal"
	Driver      "fglrx"
	Option	    "VideoOverlay" "off"
	Option	    "OpenGLOverlay" "off"
	Option	    "TexturedVideo" "off"
	BusID       "PCI:1:0:0"
EndSection

Section "Screen"
	Identifier "Default Screen"
	Device     "ATI Mobility Radeon X1300 Internal"
	Monitor    "Builtin Monitor"
	DefaultDepth     24
	SubSection "Display"
		Depth     1
		Modes    "1400x1050" "1280x1024" "1024x768" "800x600" "640x480"
	EndSubSection
	SubSection "Display"
		Depth     4
		Modes    "1400x1050" "1280x1024" "1024x768" "800x600" "640x480"
	EndSubSection
	SubSection "Display"
		Depth     8
		Modes    "1400x1050" "1280x1024" "1024x768" "800x600" "640x480"
	EndSubSection
	SubSection "Display"
		Depth     15
		Modes    "1400x1050" "1280x1024" "1024x768" "800x600" "640x480"
	EndSubSection
	SubSection "Display"
		Depth     16
		Modes    "1400x1050" "1280x1024" "1024x768" "800x600" "640x480"
	EndSubSection
	SubSection "Display"
		Depth     24
		Modes    "1400x1050" "1280x1024" "1024x768" "800x600" "640x480"
	EndSubSection
EndSection

Section "DRI"
	Mode         0666
EndSection

Section "Extensions"
	Option	    "Composite" "Disable"
EndSection

P.S. As ivanovic mentioned, the freezing problem does not occur EVERY time; it's often after some time of use, like 10 minutes or so. I've never been able to figure out what exactly makes the difference. If you're repeating my steps above and not experiencing noticable lag of gdm stop, then you're in the lucky cycle, which means you can then directly issue a gdm start. In other words, the problem didn't occur in your test. However if it does occur and you skip step 2, then you will fail to start X with the message like "an X server has already started on display :0.0".

[jasonxh]

Watch out for authatieventsd.sh & atieventsd! (cont'd)

Xorg.0.log: (unable to post full log due to length limit)

X Window System Version 7.2.0
Release Date: 22 January 2007
X Protocol Version 11, Revision 0, Release 7.2
Build Operating System: Linux Ubuntu
Current Operating System: Linux jason-laptop 2.6.23.13-custom #3 SMP Sat Jan 12 17:18:22 EST 2008 i686
Build Date: 18 January 2008
	Before reporting problems, check http://wiki.x.org
	to make sure that you have the latest version.
Module Loader present

(II) LoadModule: "fglrx"
(II) Loading /usr/lib/xorg/modules/drivers//fglrx_drv.so
(II) Module fglrx: vendor="FireGL - ATI Technologies Inc."
(II) fglrx(0): pEnt->device->identifier=0x81e4650
(II) fglrx(0): === [atiddxPreInit] === begin
(II) fglrx(0): PCI bus 1 card 0 func 0
(**) fglrx(0): Depth 24, (--) framebuffer bpp 32
(II) fglrx(0): Pixel depth = 24 bits stored in 4 bytes (32 bpp pixmaps)
(==) fglrx(0): Default visual is TrueColor
(**) fglrx(0): Option "OpenGLOverlay" "off"
(**) fglrx(0): Option "VideoOverlay" "off"
(**) fglrx(0): Option "TexturedVideo" "off"
(**) fglrx(0): Option "DPMS"
(II) fglrx(0): PCS database file /etc/ati/amdpcsdb not found
(II) fglrx(0):   Creating PCS database from initial defaults instead
(==) fglrx(0): RGB weight 888
(II) fglrx(0): Using 8 bits per RGB (8 bit DAC)
(==) fglrx(0): Gamma Correction for I is 0x06419064
(==) fglrx(0): Gamma Correction for II is 0x06419064
(==) fglrx(0): Buffer Tiling is ON
(--) fglrx(0): Chipset: "ATI Mobility Radeon X1300" (Chipset = 0x7149)
(--) fglrx(0): (PciSubVendor = 0x17aa, PciSubDevice = 0x2005)
(--) fglrx(0): board vendor info: third party graphics adapter - NOT original ATI
(--) fglrx(0): Linear framebuffer (phys) at 0xd8000000
(--) fglrx(0): MMIO registers at 0xee100000
(==) fglrx(0): ROM-BIOS at 0x000c0000
(II) fglrx(0): Primary V_BIOS segment is: 0xc000
(II) fglrx(0): VESA BIOS detected
(II) fglrx(0): VESA VBE Version 3.0
(II) fglrx(0): VESA VBE Total Mem: 16384 kB
(II) fglrx(0): VESA VBE OEM: ATI ATOMBIOS
(II) fglrx(0): VESA VBE OEM Software Rev: 9.12
(II) fglrx(0): VESA VBE OEM Vendor: (C) 1988-2005, ATI Technologies Inc. 
(II) fglrx(0): VESA VBE OEM Product: M52CSP
(II) fglrx(0): VESA VBE OEM Product Rev: 01.00
(II) fglrx(0): ATI Video BIOS revision 9 or later detected
(II) Loading sub module "fglrxdrm"
(II) LoadModule: "fglrxdrm"
(II) Loading /usr/lib/xorg/modules/linux//libfglrxdrm.so
(II) Module fglrxdrm: vendor="FireGL - ATI Technologies Inc."
(II) fglrx(0): Using adapter: PCI:1:0:0.
(--) fglrx(0): VideoRAM: 65536 kByte, Type: DDR1
(II) fglrx(0): PCIE card detected
(WW) fglrx(0): board is an unknown third party board, chipset is supported
(II) fglrx(0): Connected Display1: LCD on internal LVDS [lvds]
(II) fglrx(0): Display1 EDID data ---------------------------
(II) fglrx(0): Manufacturer: LEN  Model: 4022  Serial#: 0
(II) fglrx(0): Year: 2006  Week: 29
(II) fglrx(0): EDID Version: 1.3
(II) fglrx(0): Digital Display Input
(II) fglrx(0): Max H-Image Size [cm]: horiz.: 29  vert.: 21
(II) fglrx(0): Gamma: 2.20
(II) fglrx(0): DPMS capabilities: StandBy Suspend Off; RGB/Color Display
(II) fglrx(0): First detailed timing is preferred mode
(II) fglrx(0): redX: 0.610 redY: 0.330   greenX: 0.300 greenY: 0.530
(II) fglrx(0): blueX: 0.150 blueY: 0.130   whiteX: 0.313 whiteY: 0.329
(II) fglrx(0): Supported VESA Video Modes:
(II) fglrx(0): 640x480@60Hz
(II) fglrx(0): 800x600@60Hz
(II) fglrx(0): 1024x768@60Hz
(II) fglrx(0): Manufacturer's mask: 0
(II) fglrx(0): Supported Future Video Modes:
(II) fglrx(0): #0: hsize: 1280  vsize 1024  refresh: 60  vid: 32897
(II) fglrx(0): Supported additional Video Mode:
(II) fglrx(0): clock: 108.0 MHz   Image Size:  287 x 215 mm
(II) fglrx(0): h_active: 1400  h_sync: 1448  h_sync_end 1560 h_blank_end 1688 h_border: 0
(II) fglrx(0): v_active: 1050  v_sync: 1051  v_sync_end 1054 v_blanking: 1066 v_border: 0
(II) fglrx(0): Supported additional Video Mode:
(II) fglrx(0): clock: 90.0 MHz   Image Size:  287 x 215 mm
(II) fglrx(0): h_active: 1400  h_sync: 1448  h_sync_end 1560 h_blank_end 1688 h_border: 0
(II) fglrx(0): v_active: 1050  v_sync: 1051  v_sync_end 1054 v_blanking: 1066 v_border: 0
(II) fglrx(0):  LTD141EN9B
(II) fglrx(0): EDID (in hex):
(II) fglrx(0): 	00ffffffffffff0030ae224000000000
(II) fglrx(0): 	1d100103801d1578ea6f959c544c8726
(II) fglrx(0): 	21505421080081800101010101010101
(II) fglrx(0): 	010101010101302a7820511a10403070
(II) fglrx(0): 	13001fd71000001825237820511a1040
(II) fglrx(0): 	307013001fd7100000180000000f0090
(II) fglrx(0): 	43329043280f010030649055000000fe
(II) fglrx(0): 	004c5444313431454e39420a20200045
(II) fglrx(0): End of Display1 EDID data --------------------
(WW) fglrx(0): Only one display is connnected,so single mode is enabled
(II) fglrx(0): Primary Controller - LCD on internal LVDS
(II) fglrx(0): Internal Desktop Setting: 0x00000001
(II) fglrx(0): POWERplay version 3.  5 power states available:
(II) fglrx(0):   1. 392/324MHz @ 60Hz [enable load balancing]
(II) fglrx(0):   2. 128/135MHz @ 60Hz [low voltage, enable sleep]
(II) fglrx(0):   3. 209/135MHz @ 60Hz [low voltage, enable sleep]
(II) fglrx(0):   4. 324/135MHz @ 60Hz [enable sleep]
(II) fglrx(0):   5. 338/324MHz @ 60Hz [enable sleep, thermal diode mode]
(==) fglrx(0): Qbs is not supported in this release. Disabled.
(==) fglrx(0): FAST_SWAP disabled
(==) fglrx(0):  PseudoColor visuals disabled
(==) fglrx(0): Using gamma correction (1.0, 1.0, 1.0)
(==) fglrx(0): Center Mode is disabled 
(==) fglrx(0): TMDS coherent mode is enabled 
(II) fglrx(0): Total of 15 modes found for primary display.
(--) fglrx(0): Virtual size is 1400x1050 (pitch 0)
(--) fglrx(0): Display dimensions: (290, 210) mm
(--) fglrx(0): DPI set to (122, 126)
(--) fglrx(0): Virtual size is 1400x1050 (pitch 1408)
(==) fglrx(0): NoAccel = NO
(==) fglrx(0): NoDRI = NO
(II) Loading sub module "fglrxdrm"
(II) LoadModule: "fglrxdrm"
(II) Reloading /usr/lib/xorg/modules/linux//libfglrxdrm.so
(II) fglrx(0): Depth moves disabled by default
(==) fglrx(0): Capabilities: 0x00000000
(==) fglrx(0): CapabilitiesEx: 0x00000000
(==) fglrx(0): cpuFlags: 0x8000001d
(==) fglrx(0): OpenGL ClientDriverName: "fglrx_dri.so"
(**) fglrx(0): ATI GART size: 256 MB
(II) fglrx(0): [pcie] 258048 kB allocated
(II) fglrx(0): [drm] DRM buffer queue setup: nbufs = 100 bufsize = 65536
(==) fglrx(0): UseFastTLS=0
(==) fglrx(0): BlockSignalsOnLock=1
(II) fglrx(0): driver needs X.org 7.1.x.y with x.y >= 0.0
(II) fglrx(0): detected X.org 7.1.0.0
(II) fglrx(0): doing DRIScreenInit
(II) fglrx(0): [drm] DRM interface version 1.0
(II) fglrx(0): [drm] created "fglrx" driver at busid "PCI:1:0:0"
(II) fglrx(0): [drm] added 8192 byte SAREA at 0x2000
(II) fglrx(0): [drm] mapped SAREA 0x2000 to 0xb7c53000
(II) fglrx(0): [drm] framebuffer handle = 0x3000
(II) fglrx(0): [drm] added 1 reserved context for kernel
(II) fglrx(0): DRIScreenInit done
(II) fglrx(0): Kernel Module Version Information:
(II) fglrx(0):     Name: fglrx
(II) fglrx(0):     Version: 8.45.4
(II) fglrx(0):     Date: Jan 16 2008
(II) fglrx(0):     Desc: ATI FireGL DRM kernel module
(II) fglrx(0): Kernel Module version matches driver.
(II) fglrx(0): Kernel Module Build Time Information:
(II) fglrx(0):     Build-Kernel UTS_RELEASE:        2.6.23.13-custom
(II) fglrx(0):     Build-Kernel MODVERSIONS:        no
(II) fglrx(0):     Build-Kernel __SMP__:            no
(II) fglrx(0):     Build-Kernel PAGE_SIZE:          0x1000
(II) fglrx(0): [drm] register handle = 0x00004000
(II) fglrx(0): Interrupt handler installed at IRQ 16.
(II) fglrx(0): Exposed events to the /proc interface
(II) fglrx(0): DRI initialization successfull!
(II) fglrx(0): FBADPhys: 0xc0000000 FBMappedSize: 0x01007000
(II) fglrx(0): FBMM initialized for area (0,0)-(1408,2984)
(II) fglrx(0): FBMM auto alloc for area (0,0)-(1408,1050) (front color buffer - assumption)
(II) fglrx(0): Largest offscreen area available: 1408 x 1934
(==) fglrx(0): Backing store disabled
(**) fglrx(0): DPMS enabled
(II) fglrx(0): Using XFree86 Acceleration Architecture (XAA)
(II) fglrx(0): Acceleration enabled
(II) fglrx(0): X context handle = 0x1
(II) fglrx(0): [DRI] installation complete
(II) fglrx(0): Direct rendering enabled
(==) fglrx(0): Silken mouse enabled
(==) fglrx(0): Using hardware cursor

[oyvind]

ivanovic is right: atieventsd is the culprit

First of all, if you have useful input, please consider contributing to the bug report I've filed for this issue at: http://ati.cchtml.com/show_bug.cgi?id=992. Just don't expect to get any feedback .

Something about atieventsd and authatieventsd.sh was mentioned in the official 8.01-thread in connection with the freeze-on-logout problem. Ivanovic, in this thread, also points at atieventsd, so I decided to do some testing.

If I have atieventsd running, the X session will often freeze on logout. However, Linux itself doesn't crash, it's just the Xorg process that goes wild (which in many cases results in a feeling of total crash, but hey, this isn't Windows ). It seems that some kind of event is dispatched when the session ends, so that atieventsd spawns the following script:

/bin/sh -c /etc/ati/authatieventsd.sh grant :0 (null)

I'm not 100% sure those are exactly the args that are passed, but something like it (don't have it in front of me, logged in from a different machine). It seems this is causing some nasty race in combination with xauth+Xorg, which causes Xorg to crash/freeze. The spawned script does not exit, and Xorg consumes 100% CPU. The ~/.Xauthority file is locked.

I was able to get the box back up on its feet from an SSH-session by:
- Killing atieventsd and the script it spawned
- Shut down gdm
- Killing Xorg with kill -9
- Removing fglrx kernel module, then reload it.
- Fire up gdm, and boom, Xorg comes back up with the GDM greeter.

Now, for the fix that worked for me:
Nuke atieventsd. Prevent it from starting up, kill it, etc. On Ubuntu, I just removed all startup symlinks from the /etc/rcX.d/ directories. You can also execute "/usr/sbin/update-rc.d -f atieventsd remove" (as root). Then reboot. If you want the startup links restored, just reinstall the fglrx-package ..

I've done 25 login/logout cycles in a row without atieventsd running, and Xorg has never frozen.

What's the point with this extra daemon, anyways ? Seems like kind-of a hack. If it's supposed to help with hotplug keys on laptops it has certainly never worked for me (Thinkpad Z61m). Those buttons have never worked. I always use aticonfig+xrandr. Also, I don't think the packaged version of fglrx that comes with Ubuntu Gutsy (8.37.6) enables this daemon. It's gone from my startup sequence, at least, until I hear official news from AMD that this buggy behaviour has been fixed.

Thanks to Ivanovic for the hint.

[DirtyHairy]

Hmmm, if atieventsd is the culprit, then I would wager that the driver is in fact locking up on switching powerstates which is performed by atieventsd (which reproducibly locks up my machine, see my post in the main thread). You might want to check this assertion by manually trying to switch powerstates via aticonfig...

As this happens on every switch for me only when Textured2D is enabled, this also would explain why meddling with the textured options seems to reduce the chance of locking up for some people.

If this is indeed correct, then disabling atieventsd (and possibly any related acpid scripts) would be a temporary fix, although not a very satisfactory one --- powersaving is no gimmick but vital for laptop users.

[oyvind]

Hmmm, if atieventsd is the culprit, then I would wager that the driver is in fact locking up on switching powerstates which is performed by atieventsd (which reproducibly locks up my machine, see my post in the main thread). You might want to check this assertion by manually trying to switch powerstates via aticonfig...

Ah, I forgot that that this was part of the daemon's purpose. In fact, I've disabled automatic VGA powerstate switching (configured in /etc/default/fglrx on Ubuntu). I remember I had problems with that (suspend/resume related IIRC). Got a large, fresh battery, instead .

Just for reference, I have an ATI X1400 mobile, and I can switch between all three powerstates without any problems, using aticonfig. But I do have the freeze-on-logout-problem when atieventsd is running. I cannot switch powerstates when driving two screens, but that's not an error according to aticonfig output.

But why would the daemon switch powerstate on logout, in the first place ?

As this happens on every switch for me only when Textured2D is enabled, this also would explain why meddling with the textured options seems to reduce the chance of locking up for some people.

Perhaps, yes, I fiddled with video options at first, but don't remember the reason I thought that helped. Also, Textured2D is not an official option, but a "secret" beta/alpha thing, right ? Wouldn't expect that to work very well, when there are so many other problems. I tried it, but didn't like the horrible tearing and flickering.

If this is indeed correct, then disabling atieventsd (and possibly any related acpid scripts) would be a temporary fix, although not a very satisfactory one --- powersaving is no gimmick but vital for laptop users.

Yeah, batteries suck. But I've never noticed a huge gain in battery life with VGA powerstate at its lowest level. Maybe because my graphics card isn't very powerful and a bit old, don't know.

[bridgman]

1) The content of /etc/ati/amdpcsdb does override whatever is defined in /etc/X11/xorg.conf. IMO this is a bug...

Good point. I just realized that earlier this morning. Forcing the options off in xorg.conf won't do anything if they are forced on in amdpcsdb. I believe the settings can be changed via aticonfig rather than having to edit amdpcsdb but haven't played with that myself.

Having said that, it's not clear if these options actually make a difference with the lockup, so I wouldn't bother playing with them for now.

[ivanovic]

Good point. I just realized that earlier this morning. Forcing the options off in xorg.conf won't do anything if they are forced on in amdpcsdb. I believe the settings can be changed via aticonfig rather than having to edit amdpcsdb but haven't played with that myself.

The problem is that *most* of those options are not available in amdccle. There only a *very* minor subset of the stuff that gets set is available. And since the xorg.conf is basically setup by an admin, what the admin says should be valued, that is those values should override whatever is set in amdpcsdb.

Having said that, it's not clear if these options actually make a difference with the lockup, so I wouldn't bother playing with them for now.

Right, looks like the overlay options do not really are the reason for the lockup.

It is really great to get some feedback directly from the people at AMD/ATI, a whole lot better than just "Hmm, can you hear me? Is there anybody around?". Thanks for answering posts here in the forums, bridgman.

[ig-88]

I turning off the overlay options in /etc/ati/amdpcsdb doesn't solve the problems for me.

[yoshi314]

i didn't restart my system after installing 8.1 so i guess the daemon didn't work at all. x locks up with a cursor on screen. mouse doesn't work, system is totally locked up.

i hate to test it, doing a hard reset each time. x300se card here.

at least it's nice to know that some people of ati/amd show up on forums and actually care. all this together makes me look into ati linux future with hope.

[linrunner]

Hi!

My conclusion after 2 days of googling, testing and fiddling with almost every option:
the problem is somehow related to powerstate switching.

My system:
Thinkpad T60 (Model 2007-QPG)
X1400
Ubuntu 7.10 (Kernel 2.6.22-14.47)
Catalyst 8.1 installed according to
http://wiki.cchtml.com/index.php/Ubu...allation_Guide

My test configuration to reproduce the problem:
/etc/X11/xorg.conf

Section "Device"
        Identifier      "Generic Video Card"
        Driver          "fglrx"
        BusID           "PCI:1:0:0"
EndSection

/etc/default/acpi-support (abbrev.)

SAVE_VBE_STATE=false
POST_VIDEO=false
USE_DPMS=true

/etc/default/acpi-support (abbrev.)

FGLRX_ACPI_SWITCH_POWERSTATES=true

Because i'm not using 3d most of the time i'm reducing power consumption with:
/etc/X11/gdm/PostLogin/Default

/usr/bin/aticonfig --set-powerstate=1

Reproducing the problem:
As long as I'm not using any 3d features in my X session I can logout/shutdown without hang.
As soon as I try to activate Compiz (which by the way doesn't work with the driver ... ) or start fgl_glxgears, the following logout freezes the machine.

My current Workaround:
/etc/default/acpi-support

FGLRX_ACPI_SWITCH_POWERSTATES=[COLOR="Red"]false[/COLOR]

Setting

Option "NoDRI"

in xorg.conf works too.

Hope it helps!

linrunner