Bug 748358

Summary: [RV380] server stuck in infinite loop
Product: [Fedora] Fedora Reporter: Oron Peled <oron>
Component: xorg-x11-drv-atiAssignee: Jérôme Glisse <jglisse>
Status: CLOSED CANTFIX QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: 15CC: jglisse, mcepl, xgl-maint
Target Milestone: ---Keywords: Triaged
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard: [cat:crash]
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-03-28 09:37:10 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
Xorg.0.log with a backtrace of stuck Xserver
none
Oops after soft-reboot (reboot command), after the server was stuck none

Description Oron Peled 2011-10-24 09:08:14 UTC
Created attachment 529757 [details]
Xorg.0.log with a backtrace of stuck Xserver

Description of problem:
  The Xserver randomly stuck (every few hours/days):
   * The mouse pointer disappear
   * The keyboard does not respond (even pressing caps-lock does not
     affect the caps-lock-led)
  Otherwise the host is OK (can be accessed from the network)

  The attached Xorg.0.log contains a backtrace that hints the Xserver
  is stuck somewhere in the input subsystem, but the trace also shows
  libdrm_radeon, so it may be related.

Version-Release number of selected component (if applicable):
  xorg-x11-drv-ati-6.14.1-2.20110525gitfe5c42f51.fc15.i686

  Installed on a fully updated F15

How reproducible:
  Be (un-)lucky ?

Additional info:
  $ lspci -nn | grep VGA
    01:00.0 VGA compatible controller [0300]: ATI Technologies Inc RV370 5B60 [Radeon X300 (PCIE)] [1002:5b60] (rev ff)

  This is not a new regression, I had these random problems at least throughout F15
  (maybe earlier, not sure). Due to no keyboard response I initially thought the
  host was completely stuck. Only later I had a chance to access it from the
  network and find out that only the Xserver was stuck (with the log).

  Restarting the Xserver, "solve" the problem.

Comment 1 Matěj Cepl 2011-10-24 12:30:06 UTC
Backtrace:
[ 82941.633] 0: /usr/bin/Xorg (xorg_backtrace+0x3c) [0x80a408c]
[ 82941.633] 1: /usr/bin/Xorg (mieqEnqueue+0x1c1) [0x809e1c1]
[ 82941.633] 2: /usr/bin/Xorg (xf86PostMotionEventM+0xbf) [0x80c54ef]
[ 82941.633] 3: /usr/bin/Xorg (xf86PostMotionEventP+0x68) [0x80c5638]
[ 82941.633] 4: /usr/lib/xorg/modules/input/evdev_drv.so (0x183000+0x2def) [0x185def]
[ 82941.633] 5: /usr/lib/xorg/modules/input/evdev_drv.so (0x183000+0x3ee1) [0x186ee1]
[ 82941.633] 6: /usr/bin/Xorg (0x8047000+0x6b9b0) [0x80b29b0]
[ 82941.633] 7: /usr/bin/Xorg (0x8047000+0x123725) [0x816a725]
[ 82941.633] 8: (vdso) (__kernel_sigreturn+0x0) [0x643400]
[ 82941.633] 9: (vdso) (__kernel_vsyscall+0x2) [0x643416]
[ 82941.633] 10: /lib/libc.so.6 (ioctl+0x19) [0x4d58b299]
[ 82941.633] 11: /usr/lib/libdrm.so.2 (drmIoctl+0x34) [0x411df984]
[ 82941.633] 12: /usr/lib/libdrm.so.2 (drmCommandWriteRead+0x3d) [0x411e236d]
[ 82941.633] 13: /usr/lib/libdrm_radeon.so.1 (0xfc8000+0x15c2) [0xfc95c2]
[ 82941.633] 14: /usr/lib/libdrm_radeon.so.1 (0xfc8000+0x1834) [0xfc9834]
[ 82941.633] 15: /usr/lib/libdrm_radeon.so.1 (radeon_bo_map+0x1b) [0xfcae3b]
[ 82941.633] 16: /usr/lib/xorg/modules/drivers/radeon_drv.so (0x4c6000+0xa4e0c) [0x56ae0c]
[ 82941.633] 17: /usr/lib/xorg/modules/libexa.so (0x3d6000+0xb69f) [0x3e169f]
[ 82941.633] 18: /usr/bin/Xorg (0x8047000+0x163107) [0x81aa107]
[ 82941.633] 19: /usr/bin/Xorg (0x8047000+0xa5234) [0x80ec234]
[ 82941.633] 20: /usr/bin/Xorg (0x8047000+0x27425) [0x806e425]
[ 82941.633] 21: /usr/bin/Xorg (0x8047000+0x2a4e7) [0x80714e7]
[ 82941.633] 22: /usr/bin/Xorg (0x8047000+0x1d38c) [0x806438c]
[ 82941.633] 23: /lib/libc.so.6 (__libc_start_main+0xf3) [0x4d4d2413]
[ 82941.633] 24: /usr/bin/Xorg (0x8047000+0x1d6a1) [0x80646a1]

Comment 2 Matěj Cepl 2011-10-24 12:32:28 UTC
Pleae, provide exact output of the command

rpm -q xorg-x11-server-Xorg libdrm xorg-x11-drv-{evdev,ati}

thank you

Comment 3 Oron Peled 2011-10-30 12:21:40 UTC
Sorry about the delay, I have access to this (work) computer only on specific days.

# rpm -q xorg-x11-server-Xorg libdrm xorg-x11-drv-{evdev,ati}
xorg-x11-server-Xorg-1.10.4-1.fc15.i686
libdrm-2.4.26-1.fc15.i686
xorg-x11-drv-evdev-2.6.0-4.fc15.i686
xorg-x11-drv-ati-6.14.1-2.20110525gitfe5c42f51.fc15.i686

Comment 4 Oron Peled 2011-10-30 12:58:12 UTC
Created attachment 530827 [details]
Oops after soft-reboot (reboot command), after the server was stuck

After the Xserver was stuck:
 * Logged on remotely via ssh
 * systemctl stop/start prefdm.service several times
 * Graphics display only shows the last framebuffer state (frozen)
 * Textual (console) display fully functional
 * Run the reboot command
 * Gdm shows the greeter
 * Logged on to GNOME (for testing, normally I use KDE)
 * Got the attached kernel Oops.

I'll repeat now with hard-reset...

Comment 5 Oron Peled 2011-10-30 13:10:10 UTC
Another small information. During boot this computer displays (and always
did) a textual progress bar and not the graphical "water-drop" progress.

However, login to gnome-shell does *not* complain about lack of graphics support
and seems to be working with the usual graphical effects (until it stuck too).

Comment 6 Matěj Cepl 2011-10-31 06:58:49 UTC
Frame 1: /usr/bin/Xorg (mieqEnqueue+0x1c1) [0x809e1c1]
	/usr/src/debug/xorg-server-1.10.4/mi/mieq.c:179
	mieqEnqueue
Frame 2: /usr/bin/Xorg (xf86PostMotionEventM+0xbf) [0x80c54ef]
	/usr/src/debug/xorg-server-1.10.4/hw/xfree86/common/xf86Xinput.c:1057
	xf86PostMotionEventM
Frame 3: /usr/bin/Xorg (xf86PostMotionEventP+0x68) [0x80c5638]
	/usr/src/debug/xorg-server-1.10.4/hw/xfree86/common/xf86Xinput.c:1009
	xf86PostMotionEventP
Frame 6: /usr/bin/Xorg (0x8047000+0x6b9b0) [0x80b29b0]
	/usr/src/debug/xorg-server-1.10.4/hw/xfree86/common/xf86Events.c:300
	xf86SigioReadInput
Frame 7: /usr/bin/Xorg (0x8047000+0x123725) [0x816a725]
	/usr/src/debug/xorg-server-1.10.4/hw/xfree86/os-support/linux/../shared/sigio.c:106
	xf86SIGIO
Frame 13: /usr/lib/libdrm_radeon.so.1 (0xfc8000+0x15c2) [0xfc95c2]
	/usr/src/debug/libdrm-2.4.26/radeon/radeon_bo_gem.c:216
	bo_wait
Frame 14: /usr/lib/libdrm_radeon.so.1 (0xfc8000+0x1834) [0xfc9834]
	/usr/src/debug/libdrm-2.4.26/radeon/radeon_bo_gem.c:187
	bo_map
Frame 17: /usr/lib/xorg/modules/libexa.so (0x3d6000+0xb69f) [0x3e169f]
	/usr/src/debug/xorg-server-1.10.4/exa/exa_accel.c:1299
	exaGetImage
Frame 18: /usr/bin/Xorg (0x8047000+0x163107) [0x81aa107]
	/usr/src/debug/xorg-server-1.10.4/mi/misprite.c:436
	miSpriteGetImage
Frame 19: /usr/bin/Xorg (0x8047000+0xa5234) [0x80ec234]
	/usr/src/debug/xorg-server-1.10.4/composite/compinit.c:150
	compGetImage
Frame 20: /usr/bin/Xorg (0x8047000+0x27425) [0x806e425]
	/usr/src/debug/xorg-server-1.10.4/dix/dispatch.c:2164
	DoGetImage
Frame 21: /usr/bin/Xorg (0x8047000+0x2a4e7) [0x80714e7]
	/usr/src/debug/xorg-server-1.10.4/dix/dispatch.c:431
	Dispatch
Frame 22: /usr/bin/Xorg (0x8047000+0x1d38c) [0x806438c]
	/usr/src/debug/xorg-server-1.10.4/dix/main.c:287
	main

Comment 7 Oron Peled 2012-03-28 09:37:10 UTC
Few weeks ago, this computer was gone (Looks like MB problem, upon power-on
does not even reach the BIOS screen).

So I guess we can close this bug. Since no one else seem to have this,
it raise a suspicion that it was due to some hardware problem -- it
was a very old computer.