Bug#566547: linux-image-2.6.32-trunk-amd64: Segmentation fault during halt

2010-01-23 Thread Aaron Barany
Package: linux-2.6
Version: 2.6.32-5
Severity: normal

After updating to the latest kernel, the halt process seg faults on shutdown, 
and the computer never fully turns off. Reboot appears to work fine. If there's 
a log someplace where I can get the stack backtrace that prints out I can post 
that. I do remember that the function at the top of the backtrace was 
hrtimer_cpu_notify.


-- Package-specific info:
** Version:
Linux version 2.6.32-trunk-amd64 (Debian 2.6.32-5) (b...@decadent.org.uk) (gcc 
version 4.3.4 (Debian 4.3.4-6) ) #1 SMP Sun Jan 10 22:40:40 UTC 2010

** Command line:
BOOT_IMAGE=/boot/vmlinuz-2.6.32-trunk-amd64 
root=UUID=43acb8fe-ae48-4a5a-8f5f-959b3d09d4db ro vga=786 quiet quiet

** Tainted: P (1)
 * Proprietary module has been loaded.

** Kernel log:
[   10.193827] HDA Intel :02:00.1: setting latency timer to 64
[   11.115123] Adding 11719376k swap on /dev/sda5.  Priority:-1 extents:1 
across:11719376k 
[   11.541695] loop: module loaded
[   11.923431] EXT4-fs (sdc1): warning: maximal mount count reached, running 
e2fsck is recommended
[   11.947476] EXT4-fs (sdc1): mounted filesystem with ordered data mode
[   11.991099] fuse init (API version 7.13)
[   13.408872] ip_tables: (C) 2000-2006 Netfilter Core Team
[   13.434111] nf_conntrack version 0.5.0 (16384 buckets, 65536 max)
[   13.434238] CONFIG_NF_CT_ACCT is deprecated and will be removed soon. Please 
use
[   13.434239] nf_conntrack.acct=1 kernel parameter, acct=1 nf_conntrack module 
option or
[   13.434240] sysctl net.netfilter.nf_conntrack_acct=1 to enable it.
[   13.783745] r8169: eth0: link up
[   13.783749] r8169: eth0: link up
[   20.388546] Bluetooth: Core ver 2.15
[   20.388591] NET: Registered protocol family 31
[   20.388593] Bluetooth: HCI device and connection manager initialized
[   20.388595] Bluetooth: HCI socket layer initialized
[   20.418012] Bluetooth: L2CAP ver 2.14
[   20.418015] Bluetooth: L2CAP socket layer initialized
[   20.422641] Bluetooth: RFCOMM TTY layer initialized
[   20.422649] Bluetooth: RFCOMM socket layer initialized
[   20.422651] Bluetooth: RFCOMM ver 1.11
[   20.437367] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
[   20.437370] Bluetooth: BNEP filters: protocol multicast
[   20.455692] Bridge firewalling registered
[   20.513761] Bluetooth: SCO (Voice Link) ver 0.6
[   20.513763] Bluetooth: SCO socket layer initialized
[   21.435995] lp: driver loaded but no devices found
[   21.446242] ppdev: user-space parallel port driver
[   23.374467] Linux agpgart interface v0.103
[   23.378751] fglrx: module license 'Proprietary. (C) 2002 - ATI Technologies, 
Starnberg, GERMANY' taints kernel.
[   23.378756] Disabling lock debugging due to kernel taint
[   23.439878] [fglrx] Maximum main memory to use for locked dma buffers: 5781 
MBytes.
[   23.440003] [fglrx]   vendor: 1002 device: 9440 count: 1
[   23.440229] [fglrx] ioport: bar 4, base 0xc000, size: 0x100
[   23.440240] pci :02:00.0: PCI INT A - GSI 16 (level, low) - IRQ 16
[   23.440244] pci :02:00.0: setting latency timer to 64
[   23.440368] [fglrx] Kernel PAT support is enabled
[   23.440380] [fglrx] module loaded - fglrx 8.68.2 [Nov 24 2009] with 1 minors
[   24.198702]   alloc irq_desc for 32 on node -1
[   24.198705]   alloc kstat_irqs on node -1
[   24.198717] fglrx_pci :02:00.0: irq 32 for MSI/MSI-X
[   24.199461] [fglrx] Firegl kernel thread PID: 2772
[   24.380561] eth0: no IPv6 routers present
[   25.025086] [fglrx] Gart USWC size:1279 M.
[   25.025090] [fglrx] Gart cacheable size:508 M.
[   25.025096] [fglrx] Reserved FB block: Shared offset:0, size:100 
[   25.025099] [fglrx] Reserved FB block: Unshared offset:fbff000, size:401000 
[   25.025102] [fglrx] Reserved FB block: Unshared offset:3fffb000, size:5000 
[   31.786935] firewire_core: giving up on config rom for node id ffc0
[   35.009204] CPU0 attaching NULL sched-domain.
[   35.009208] CPU1 attaching NULL sched-domain.
[   35.009211] CPU2 attaching NULL sched-domain.
[   35.009213] CPU3 attaching NULL sched-domain.
[   35.009215] CPU4 attaching NULL sched-domain.
[   35.009217] CPU5 attaching NULL sched-domain.
[   35.009220] CPU6 attaching NULL sched-domain.
[   35.009222] CPU7 attaching NULL sched-domain.
[   35.078575] CPU0 attaching sched-domain:
[   35.078579]  domain 0: span 0,4 level SIBLING
[   35.078582]   groups: 0 (cpu_power = 589) 4 (cpu_power = 589)
[   35.078589]   domain 1: span 0-7 level MC
[   35.078591]groups: 0,4 (cpu_power = 1178) 1,5 (cpu_power = 1178) 2,6 
(cpu_power = 1178) 3,7 (cpu_power = 1178)
[   35.078602] CPU1 attaching sched-domain:
[   35.078604]  domain 0: span 1,5 level SIBLING
[   35.078607]   groups: 1 (cpu_power = 589) 5 (cpu_power = 589)
[   35.078612]   domain 1: span 0-7 level MC
[   35.078615]groups: 1,5 (cpu_power = 1178) 2,6 (cpu_power = 1178) 3,7 
(cpu_power = 1178) 0,4 (cpu_power = 1178)
[   35.078625] CPU2 attaching sched-domain:
[   35.078627]  domain 0: span 2,6 level SIBLING
[   35.078629]   groups: 2 

Bug#566547: linux-image-2.6.32-trunk-amd64: Segmentation fault during halt

2010-01-23 Thread Ben Hutchings
On Sat, 2010-01-23 at 12:54 -0800, Aaron Barany wrote:
 Package: linux-2.6
 Version: 2.6.32-5
 Severity: normal
 
 After updating to the latest kernel, the halt process seg faults on
 shutdown, and the computer never fully turns off. Reboot appears to
 work fine. If there's a log someplace where I can get the stack
 backtrace that prints out I can post that. I do remember that the
 function at the top of the backtrace was hrtimer_cpu_notify.

It will be too late to log anything to a file.  But if you can take a
photo of the screen, please send that.

 -- Package-specific info:
 ** Version:
 Linux version 2.6.32-trunk-amd64 (Debian 2.6.32-5) (b...@decadent.org.uk) 
 (gcc version 4.3.4 (Debian 4.3.4-6) ) #1 SMP Sun Jan 10 22:40:40 UTC 2010
 
 ** Command line:
 BOOT_IMAGE=/boot/vmlinuz-2.6.32-trunk-amd64 
 root=UUID=43acb8fe-ae48-4a5a-8f5f-959b3d09d4db ro vga=786 quiet quiet
 
 ** Tainted: P (1)
  * Proprietary module has been loaded.
[...]

Please check whether this is reproducible if you remove the fglrx module
before shutting down.

Ben.

-- 
Ben Hutchings
Any smoothly functioning technology is indistinguishable from a rigged demo.


signature.asc
Description: This is a digitally signed message part


Bug#566547: linux-image-2.6.32-trunk-amd64: Segmentation fault during halt

2010-01-23 Thread Aaron Barany
Here a the photo of the screen with the segfault. (
http://akb825.com/images/halt%20segfault.jpg)

I couldn't figure out how to unload the fglrx module, since rmmod always
threw an error saying that it is in use. However, if I boot in recovery mode
it doesn't load fglrx, but it still segfaults if I call halt after logging
in.

Thanks,
Aaron

On Sat, Jan 23, 2010 at 2:38 PM, Ben Hutchings b...@decadent.org.uk wrote:

 On Sat, 2010-01-23 at 12:54 -0800, Aaron Barany wrote:
  Package: linux-2.6
  Version: 2.6.32-5
  Severity: normal
 
  After updating to the latest kernel, the halt process seg faults on
  shutdown, and the computer never fully turns off. Reboot appears to
  work fine. If there's a log someplace where I can get the stack
  backtrace that prints out I can post that. I do remember that the
  function at the top of the backtrace was hrtimer_cpu_notify.

 It will be too late to log anything to a file.  But if you can take a
 photo of the screen, please send that.

  -- Package-specific info:
  ** Version:
  Linux version 2.6.32-trunk-amd64 (Debian 2.6.32-5) (b...@decadent.org.uk)
 (gcc version 4.3.4 (Debian 4.3.4-6) ) #1 SMP Sun Jan 10 22:40:40 UTC 2010
 
  ** Command line:
  BOOT_IMAGE=/boot/vmlinuz-2.6.32-trunk-amd64
 root=UUID=43acb8fe-ae48-4a5a-8f5f-959b3d09d4db ro vga=786 quiet quiet
 
  ** Tainted: P (1)
   * Proprietary module has been loaded.
 [...]

 Please check whether this is reproducible if you remove the fglrx module
 before shutting down.

 Ben.

 --
 Ben Hutchings
 Any smoothly functioning technology is indistinguishable from a rigged
 demo.



Bug#566547: linux-image-2.6.32-trunk-amd64: Segmentation fault during halt

2010-01-23 Thread Ben Hutchings
On Sat, 2010-01-23 at 19:13 -0800, Aaron Barany wrote:
 Here a the photo of the screen with the segfault.
 (http://akb825.com/images/halt%20segfault.jpg)
 
 I couldn't figure out how to unload the fglrx module, since rmmod
 always threw an error saying that it is in use. However, if I boot in
 recovery mode it doesn't load fglrx, but it still segfaults if I call
 halt after logging in.

OK, so it's not the fault of fglrx.

Please report this at http://bugzilla.kernel.org under product
'Process Management', component 'Other'.  Then let us know the bug
number so we can track it.

Ben.

-- 
Ben Hutchings
Any smoothly functioning technology is indistinguishable from a rigged demo.


signature.asc
Description: This is a digitally signed message part


Bug#566547: linux-image-2.6.32-trunk-amd64: Segmentation fault during halt

2010-01-23 Thread Aaron Barany
I have added the bug listing as you requested. It is bug #15117. (
http://bugzilla.kernel.org/show_bug.cgi?id=15117)

Thanks,
Aaron

On Sat, Jan 23, 2010 at 7:39 PM, Ben Hutchings b...@decadent.org.uk wrote:

 On Sat, 2010-01-23 at 19:13 -0800, Aaron Barany wrote:
  Here a the photo of the screen with the segfault.
  (http://akb825.com/images/halt%20segfault.jpg)
 
  I couldn't figure out how to unload the fglrx module, since rmmod
  always threw an error saying that it is in use. However, if I boot in
  recovery mode it doesn't load fglrx, but it still segfaults if I call
  halt after logging in.

 OK, so it's not the fault of fglrx.

 Please report this at http://bugzilla.kernel.org under product
 'Process Management', component 'Other'.  Then let us know the bug
 number so we can track it.

 Ben.

 --
 Ben Hutchings
 Any smoothly functioning technology is indistinguishable from a rigged
 demo.