View Single Post
Old 07-14-12, 05:04 AM   #28
windchine
Registered User
 
Join Date: Jun 2012
Posts: 3
Default Re: Xorg server crashes in nvidia 304.22

Confirmed same symptoms as 295 with beta 304.22 using Fedora 17 and graphics card:

01:00.0 VGA compatible controller: nVidia Corporation G92M [Quadro FX 3600M] (rev a2)

During failure machine can still be accessed via LAN with ssh and "top" shows Xorg spinning one CPU core at near 100% utilisation:

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1267 root 20 0 63052 43m 11m R 99.7 1.4 13:38.92 Xorg

Tail end of Xorg.0.log shows:

[ 68.163] (II) NVIDIA(GPU-0): "1920x1200_60" : 1920 x 1200 @ 60.0 Hz (from: EDID)
[ 68.163] (II) NVIDIA(GPU-0): --- End of ModePool for Seiko/Epson (DFP-0): ---
[ 68.163] (II) NVIDIA(GPU-0):
[ 131.753] (WW) NVIDIA(0): WAIT (2, 6, 0x8000, 0x0000c990, 0x0000d23c)
[ 138.753] (WW) NVIDIA(0): WAIT (1, 6, 0x8000, 0x0000c990, 0x0000d23c)
[ 140.535] [mi] EQ overflowing. Additional events will be discarded until existing events are processed.
[ 140.535]
[ 140.535] Backtrace:
[ 140.536] 0: /usr/bin/Xorg (xorg_backtrace+0x4a) [0x80ac24a]
[ 140.536] 1: /usr/bin/Xorg (mieqEnqueue+0x22b) [0x81aa77b]
[ 140.536] 2: /usr/bin/Xorg (0x8047000+0x4577c) [0x808c77c]
[ 140.536] 3: /usr/bin/Xorg (xf86PostMotionEventM+0xfa) [0x80d99ca]
[ 140.536] 4: /usr/bin/Xorg (xf86PostMotionEvent+0xaa) [0x80d9c1a]
[ 140.536] 5: /usr/lib/xorg/modules/input/synaptics_drv.so (0xb47c7000+0x470e) [0xb47cb70e]
[ 140.536] 6: /usr/lib/xorg/modules/input/synaptics_drv.so (0xb47c7000+0x6bf1) [0xb47cdbf1]
[ 140.536] 7: /usr/bin/Xorg (0x8047000+0x81fa2) [0x80c8fa2]
[ 140.536] 8: /usr/bin/Xorg (0x8047000+0xa8185) [0x80ef185]
[ 140.536] 9: (vdso) (__kernel_sigreturn+0x0) [0xb7719400]
[ 140.536] 10: (vdso) (__kernel_vsyscall+0x10) [0xb7719424]
[ 140.536] 11: /lib/libc.so.6 (__gettimeofday+0x16) [0x43ad1986]
[ 140.536] 12: /usr/lib/xorg/modules/drivers/nvidia_drv.so (0xb4e4b000+0xe014c) [0xb4f2b14c]
[ 140.536]
[ 140.536] [mi] These backtraces from mieqEnqueue may point to a culprit higher up the stack.
[ 140.536] [mi] mieq is *NOT* the cause. It is a victim.
[ 141.760] (WW) NVIDIA(0): WAIT (2, 6, 0x8000, 0x0000c990, 0x0000e348)
[ 148.760] (WW) NVIDIA(0): WAIT (1, 6, 0x8000, 0x0000c990, 0x0000e348)
[ 148.760] [mi] Increasing EQ size to 512 to prevent dropped events.
[ 148.760] [mi] EQ processing has resumed after 27 dropped events.
[ 148.760] [mi] This may be caused my a misbehaving driver monopolizing the server's resources.

Attempt to remotly close X and return to VGA text console using "telinit 3" results in partially functional remote ssh sessions and the following VGA text console diagnostics after X has partially closed:

Message from syslogd@laptop at Jul 14 09:32:25 ...
kernel:[ 1564.038952] BUG: soft lockup - CPU#1 stuck for 22s! [gmain:1966]

Message from syslogd@laptop at Jul 14 09:32:25 ...
kernel:[ 1564.038980] Process gmain (pid: 1966, ti=efb72000 task=f02d7110 task.ti=efb72000)

Message from syslogd@laptop at Jul 14 09:32:25 ...
kernel:[ 1564.038980] Stack:

Message from syslogd@laptop at Jul 14 09:32:25 ...
kernel:[ 1564.038980] Call Trace:

Message from syslogd@laptop at Jul 14 09:32:25 ...
kernel:[ 1564.038980] Code: 0d 01 00 83 c4 10 ba 2f 00 00 00 eb 11 89 83 4c 22 00 00 50 e8 f0 f7 00 00 83 c4 04 89 c2 89 d0 83 c5 08 5b c3 b8 2f 00 00 00 c3 <57> 56 53 8b 7c 24 10 8b 5c 24 14 89 fe 57 ff 57 6c 83 c4 04 83
windchine is offline   Reply With Quote