X server quits "BUG: scheduling while atomic:"

[lada]

Xserver falls and if fglrx driver is used, it writes "BUG: scheduling while atomic:" and some 4 digit number and then a small hexadecimal 0x0000002 number. If not using 3d accelerated driver (fglrx), Xserver respawns to gdm login screen.

Bug is repeatable on 2.6.27-11 distribution kernel as well as on 2.6.24.4 and 2.6.27.10 vanilla kernels.

When I launch cadsoft's eagle and click on menu, it happens all the time.

I suppose it is in Xserver, not in kernel or fglrx driver itself.

It started to behave this way after upgrade to 8.10 from 8.04 LTS

Eagle is version 5.3.0

I could downgrade some package and try to find out if it persists.

ProblemType: Bug
Architecture: i386
DistroRelease: Ubuntu 8.10
ExecutablePath: /usr/bin/yelp
Package: yelp 2.24.0-0ubuntu2
ProcEnviron:
 PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
 LANG=cs_CZ.UTF-8
 SHELL=/bin/bash
SourcePackage: yelp
Uname: Linux 2.6.27-11-generic i686

[lada (lada)]

Now it happened again, randomly. It keeps writing (in text mode)
[seconds.microseconds] BUG: scheduling while atomic: Xorg/5278/0x000000002

where 5278 is PID of Xorg process. Last view in process manages has shown do_select next to Xorg process, then kaboom - it fell. But Amarok played a song till its end.

When I ssh to this machine from another, I see this:

top - 06:41:43 up 5:41, 2 users, load average: 3.86, 2.82, 1.52
Tasks: 120 total, 5 running, 113 sleeping, 0 stopped, 2 zombie
Cpu(s): 20.0%us, 76.3%sy, 3.7%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 904032k total, 895776k used, 8256k free, 24936k buffers
Swap: 2245280k total, 661376k used, 1583904k free, 294284k cached

  PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
 5278 root 20 0 0 0 0 R 34.3 0.0 4:09.18 Xorg
 4606 root 20 0 1828 432 412 R 30.0 0.0 0:48.23 dd
 4608 klog 20 0 3660 472 420 S 23.7 0.1 0:37.50 klogd
 4554 syslog 20 0 1888 660 540 S 8.0 0.1 0:12.43 syslogd
11902 root 30 10 120m 115m 4172 R 3.3 13.0 1:28.33 update-apt-xapi
 5701 lada 20 0 28040 3560 2840 S 0.3 0.4 11:28.86 pulseaudio
    1 root 20 0 2948 548 492 S 0.0 0.1 0:04.68 init
    2 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 kthreadd
    3 root 15 -5 0 0 0 S 0.0 0.0 0:01.46 ksoftirqd/0
    4 root RT -5 0 0 0 S 0.0 0.0 0:00.00 watchdog/0
    5 root 15 -5 0 0 0 S 0.0 0.0 0:00.06 events/0
    6 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 khelper
   45 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 kintegrityd/0
   47 root 15 -5 0 0 0 S 0.0 0.0 0:00.80 kblockd/0
   49 root 15 -5 0 0 0 S 0.0 0.0 0:00.01 kacpid
   50 root 15 -5 0 0 0 S 0.0 0.0 0:00.01 kacpi_notify
  130 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 cqueue

Load seemed high before it fell, so maybe it was on a brink of collapse.

[lada (lada)]

And BTW, it's a Thinkpad T42p with 2Gigs of RAM, 2.1GHz, nothing special, WiFi turned off, maybe interrupts?

lada@T42p:~$ cat /proc/interrupts
           CPU0
  0: 1537574 XT-PIC-XT timer
  1: 727 XT-PIC-XT i8042
  2: 0 XT-PIC-XT cascade
  4: 4 XT-PIC-XT
  5: 1861805 XT-PIC-XT Intel 82801DB-ICH4
  7: 2 XT-PIC-XT parport0
  8: 94 XT-PIC-XT rtc0
  9: 832912 XT-PIC-XT acpi
 11: 42137 XT-PIC-XT eth0
 12: 260151 XT-PIC-XT i8042
 14: 130687 XT-PIC-XT ata_piix
 15: 120 XT-PIC-XT ata_piix
NMI: 0 Non-maskable interrupts
LOC: 784398 Local timer interrupts
SPU: 0 Spurious interrupts
ERR: 0
MIS: 0

lspci -v shows (only FireGL shown):
01:00.0 VGA compatible controller: ATI Technologies Inc M10 NT [FireGL Mobility T2] (rev 80)
        Subsystem: IBM Device 054f
        Flags: bus master, fast Back2Back, 66MHz, medium devsel, latency 255, IRQ 11
        Memory at e0000000 (32-bit, prefetchable) [size=128M]
        I/O ports at 3000 [size=256]
        Memory at c0100000 (32-bit, non-prefetchable) [size=64K]
        [virtual] Expansion ROM at c0120000 [disabled] [size=128K]
        Capabilities: <access denied>
        Kernel driver in use: fglrx_pci
        Kernel modules: radeonfb, fglrx

[lada (lada)]

top - 07:02:10 up 5 min, 2 users, load average: 1.26, 0.93, 0.41
Tasks: 75 total, 3 running, 72 sleeping, 0 stopped, 0 zombie
Cpu(s): 19.3%us, 80.7%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 904032k total, 420304k used, 483728k free, 19604k buffers
Swap: 2245280k total, 0k used, 2245280k free, 279452k cached

  PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
 6452 root 20 0 194m 23m 7200 R 99.7 2.6 1:41.82 Xorg

Now it fell and no writing on a terminal, just running all time, memory usage not increasing

[Bryce Harrington (bryce)]

I've posted a new version of the -fglrx driver to our xorg-edgers PPA,
would you mind testing it either on Jaunty or Karmic and see if it
resolves this bug?

Get fglrx 8.620 here:

  https://edge.launchpad.net/~xorg-edgers/+archive/ppa

[Bryce Harrington (bryce)]

We're closing this bug since it is has been some time with no response from the original reporter. However, if the issue still exists please feel free to reopen with the requested information. Also, if you could, please test against the latest development version of Ubuntu, since this confirms the bug is one we may be able to pass upstream for help.