public inbox for gentoo-user@lists.gentoo.org
 help / color / mirror / Atom feed
* [gentoo-user] xorg segfaults if I have an encrypted volume mounted
@ 2010-06-24 22:53 Remy Blank
  2010-06-25 12:52 ` [gentoo-user] " Remy Blank
  0 siblings, 1 reply; 4+ messages in thread
From: Remy Blank @ 2010-06-24 22:53 UTC (permalink / raw
  To: gentoo-user

[-- Attachment #1: Type: text/plain, Size: 2932 bytes --]

xorg started segfaulting here on startup, at the point where it should
detect input devices:

[  1198.330] (II) AIGLX: enabled GLX_MESA_copy_sub_buffer
[  1198.330] (II) AIGLX: enabled GLX_INTEL_swap_event
[  1198.330] (II) AIGLX: enabled GLX_SGI_make_current_read
[  1198.330] (II) AIGLX: GLX_EXT_texture_from_pixmap backed by buffer
objects
[  1198.330] (II) AIGLX: Loaded and initialized /usr/lib/dri/i965_dri.so
[  1198.330] (II) GLX: Initialized DRI2 GL provider for screen 0
[  1198.331] (II) intel(0): Setting screen physical size to 507 x 317
[  1198.437]
Backtrace:
[  1198.437] 0: /usr/bin/X (xorg_backtrace+0x38) [0x80ae1c8]
[  1198.437] 1: /lib/libudev.so.0 (0xb782b000+0x33d2) [0xb782e3d2]
[  1198.437] Segmentation fault at address 0x7974702f
[  1198.437]
Fatal server error:
[  1198.437] Caught signal 11 (Segmentation fault). Server aborting

For reference, here's the same location in a good start:

[  2832.766] (II) AIGLX: enabled GLX_MESA_copy_sub_buffer
[  2832.766] (II) AIGLX: enabled GLX_INTEL_swap_event
[  2832.766] (II) AIGLX: enabled GLX_SGI_make_current_read
[  2832.766] (II) AIGLX: GLX_EXT_texture_from_pixmap backed by buffer
objects
[  2832.766] (II) AIGLX: Loaded and initialized /usr/lib/dri/i965_dri.so
[  2832.766] (II) GLX: Initialized DRI2 GL provider for screen 0
[  2832.767] (II) intel(0): Setting screen physical size to 507 x 317
[  2832.906] (II) config/udev: Adding input device Video Bus
(/dev/input/event8)
[  2832.906] (**) Video Bus: Applying InputClass "evdev keyboard catchall"
[  2832.906] (**) Video Bus: Applying InputClass "Keyboard-all"
[  2832.906] (II) LoadModule: "evdev"
[  2832.906] (II) Loading /usr/lib/xorg/modules/input/evdev_drv.so
[  2832.906] (II) Module evdev: vendor="X.Org Foundation"
[  2832.906]    compiled for 1.8.1.901, module version = 2.3.2
[  2832.907]    Module class: X.Org XInput Driver
[  2832.907]    ABI class: X.Org XInput driver, version 9.0

So it's segfaulting in libudev. And where it's getting weird is that it
only segfaults if I have one particular encrypted container mapped. The
container is a file mapped to /dev/loop0, opened with "cryptsetup
luksOpen" as "/dev/mapper/crypt-morpheus.athome". If I "luksClose" it
(but keep /dev/loop0), xorg starts normally.

Strangely, there is another encrypted container "/dev/mapper/crypt-swap"
mapped to my swap partition, but that one doesn't seem to interfere.

This symptom has only started today, and I haven't updated anything
udev- or xorg-related. However, I reboot rarely, and I did reboot today,
so it may be due to an earlier update.

Some version info:

  xorg-server-1.8.1.901
  xf86-input-evdev-2.3.2
  udev-149
  gentoo-sources-2.6.31-r6

Has anyone seen anything similar? Any idea how I could either work
around the issue or debug it? I have tried strace but couldn't extract
any meaningful information.

-- Remy


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [gentoo-user] Re: xorg segfaults if I have an encrypted volume mounted
  2010-06-24 22:53 [gentoo-user] xorg segfaults if I have an encrypted volume mounted Remy Blank
@ 2010-06-25 12:52 ` Remy Blank
  2010-06-25 14:46   ` James L
  0 siblings, 1 reply; 4+ messages in thread
From: Remy Blank @ 2010-06-25 12:52 UTC (permalink / raw
  To: gentoo-user

[-- Attachment #1: Type: text/plain, Size: 1243 bytes --]

Remy Blank wrote:
> Some version info:
> 
>   xorg-server-1.8.1.901
>   xf86-input-evdev-2.3.2
>   udev-149
>   gentoo-sources-2.6.31-r6
> 
> Has anyone seen anything similar? Any idea how I could either work
> around the issue or debug it? I have tried strace but couldn't extract
> any meaningful information.

For the record, I have update to gentoo-sources-2.6.34-r1, and the issue
seems to be fixed. I still don't understand why this has suddenly
started happening, as I had been running 2.6.31-r6 for months.

Now on to my next issue: my laptop powers off by itself after a few
hours of doing nothing. I first thought that it was a hardware issue,
but it just switched off before my eyes a few minutes ago: the screen
went blank, and I could see the caps lock and scroll lock LEDs blink for
~30 seconds, before the laptop powered off. This seems to be a strong
hint at a kernel panic.

Unfortunately, as I was in X, I didn't get to see the panic, and it
wasn't logged either. Does anyone know how to get the kernel panic
message in such a case? I have enabled remote logging with syslog-ng,
but I suspect that this won't be enough. I have found kdump, but setting
it up seems to be quite tricky.

-- Remy


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [gentoo-user] Re: xorg segfaults if I have an encrypted volume  mounted
  2010-06-25 12:52 ` [gentoo-user] " Remy Blank
@ 2010-06-25 14:46   ` James L
  2010-06-25 16:09     ` Remy Blank
  0 siblings, 1 reply; 4+ messages in thread
From: James L @ 2010-06-25 14:46 UTC (permalink / raw
  To: gentoo-user

> Remy Blank wrote:
> my laptop powers off by itself after a few
> hours of doing nothing. I first thought that it was a hardware issue,
> but it just switched off before my eyes a few minutes ago: the screen
> went blank, and I could see the caps lock and scroll lock LEDs blink for
> ~30 seconds, before the laptop powered off. This seems to be a strong
> hint at a kernel panic.
>
> Unfortunately, as I was in X, I didn't get to see the panic, and it
> wasn't logged either. Does anyone know how to get the kernel panic
> message in such a case? I have enabled remote logging with syslog-ng,
> but I suspect that this won't be enough. I have found kdump, but setting
> it up seems to be quite tricky.
>
> -- Remy

Turn off all power management software.
My guess is it is going into one of the power saving modes that ends up
crashing it.




^ permalink raw reply	[flat|nested] 4+ messages in thread

* [gentoo-user] Re: xorg segfaults if I have an encrypted volume  mounted
  2010-06-25 14:46   ` James L
@ 2010-06-25 16:09     ` Remy Blank
  0 siblings, 0 replies; 4+ messages in thread
From: Remy Blank @ 2010-06-25 16:09 UTC (permalink / raw
  To: gentoo-user

[-- Attachment #1: Type: text/plain, Size: 628 bytes --]

James L wrote:
> Turn off all power management software.
> My guess is it is going into one of the power saving modes that ends up
> crashing it.

Thanks for the suggestion. The only "power management software" that I
have installed is the cpufreq ondemand governor, which shouldn't use the
power saving modes AFAIK.

I have logged the CPU temperature, remaining battery capacity, charger
state and loadavg every 10 seconds, and the CPU is mostly idle (0.00
loadavg, ~40 °C), the battery full, and the charger connected. Then
suddenly, after ~15 hours, blinking keyboard lights and finally power off.

-- Remy


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2010-06-25 16:11 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-06-24 22:53 [gentoo-user] xorg segfaults if I have an encrypted volume mounted Remy Blank
2010-06-25 12:52 ` [gentoo-user] " Remy Blank
2010-06-25 14:46   ` James L
2010-06-25 16:09     ` Remy Blank

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox