[HD 3400] SIGWINCH causes xserver to briefly hang

Bug #367694 reported by aarons on 2009-04-27
12
This bug affects 2 people
Affects Status Importance Assigned to Milestone
fglrx-installer (Ubuntu)
Medium
Unassigned

Bug Description

Binary package hint: xorg

I have traced this bug back to one of the xorg packages. Some change between 8.10 and 9.04 causes SIGWINCH to hang the system for about 2 seconds. I use compiz, but I observed the problem with the fallback window manager as well. Opening windows, resizing them, restoring them, and opening the second tab in gnome-terminal all cause this 2 second hang where the cursor still moves, but everything else stops updating (such as gkrellm).

I am running the ati provided fglrx drivers on x86_64. This bug is not tied to a specific version of that driver.

Here are the packages currently held:

x11-xserver-utils 7.3+5
xorg 1:7.4~5ubuntu3
xserver-common 2:1.5.2-2ubuntu3
xserver-xorg 1:7.4~5ubuntu3
xserver-xorg-core 2:1.5.2-2ubuntu3
xserver-xorg-dev 2:1.5.2-2ubuntu3
xserver-xorg-input-all 1:7.4~5ubuntu3
xserver-xorg-input-evdev 1:2.0.99+git20080912-0
xserver-xorg-input-kbd 1:1.3.1-1ubuntu2
xserver-xorg-input-mouse 1:1.3.0-1build1
xserver-xorg-input-synaptics 0.15.2-0ubuntu7
xserver-xorg-video-all 1:7.4~5ubuntu3
xserver-xorg-video-ati 1:6.9.0+git20081003.f9

There are other xorg-video packages held, but they aren't used.

As I said, I have the ati fglrx driver installed via the ati provided script ati-driver-installer-9-4-x86.x86_64.run, but I should note that before I installed that my display was all garbled after my update to 9.04.

display: :0.0 screen: 0
OpenGL vendor string: ATI Technologies Inc.
OpenGL renderer string: ATI Mobility Radeon HD 3400 Series
OpenGL version string: 2.1.8591

 [fglrx] module loaded - fglrx 8.60.3 [Apr 1 2009] with 1 minors

[lspci]
00:00.0 Host bridge [0600]: Intel Corporation Mobile 4 Series Chipset Memory Controller Hub [8086:2a40] (rev 07)
     Subsystem: Dell Device [1028:029f]
01:00.0 VGA compatible controller [0300]: ATI Technologies Inc Mobility Radeon HD 3400 Series [1002:95c4]
     Subsystem: Dell Device [1028:029f]

Martin Olsson (mnemo) wrote :

Thank you for taking the time to report this bug. Even though you've provided a lot of information already, can you please run this command:
apport-collect 367694

This will automatically attach more detailed hw/sw info to the bug in the form of xorg.log, dmesg output etc

Changed in xorg (Ubuntu):
status: New → Incomplete

Architecture: amd64
DistroRelease: Ubuntu 9.04
NonfreeKernelModules: fglrx
Package: xorg 1:7.4~5ubuntu3
PackageArchitecture: amd64
ProcEnviron:
 SHELL=/bin/bash
 PATH=(custom, user)
 LANG=en_US.UTF-8
ProcVersion: Linux version 2.6.28-11-generic (buildd@crested) (gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4) ) #42-Ubuntu SMP Fri Apr 17 01:58:03 UTC 2009
Uname: Linux 2.6.28-11-generic x86_64
UserGroups: adm admin audio cdrom dialout lpadmin mpd plugdev sambashare

aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :

apport-collect crashed. I think I probably screwed up my system when I downgraded a bunch of packages back to 8.10 while trying to determine which one was actually causing this bug. I've since restored everything but the xorg packages, but there are probably some packages missing on my system now. Also though, because I'm running that older xorg, I'm not sure all the apport-collect info will actually be helpful, since I'm not experiencing the bug right now. I can upgrade those packages again and rerun apport-collect, but this is really a huge annoyance and if this bug is going to be low in the queue I'm not going to be able to keep things broken while waiting for the fix.

anyway, here is what apport-collect said:

Collecting apport information for source package xorg...
hook /usr/share/apport/package-hooks//source_xorg.py crashed:
Traceback (most recent call last):
  File "/usr/lib/python2.6/dist-packages/apport/report.py", line 570, in add_hooks_info
    symb['add_info'](self)
  File "/usr/share/apport/package-hooks//source_xorg.py", line 104, in add_info
    report['fglrx-loaded'] = matches
  File "/usr/lib/python2.6/dist-packages/problem_report.py", line 479, in __setitem__
    assert k.replace('.', '').isalnum()
AssertionError
Uploading additional information to Launchpad bug...
   short text data...
   attachment: Dependencies.txt...
   attachment: LsHal.txt...
   attachment: LsMod.txt...
   attachment: LsPci.txt...
   attachment: XorgConf.txt...
   attachment: XorgLog.txt...
Traceback (most recent call last):
  File "/usr/bin/apport-collect", line 173, in <module>
    upload(report, bug)
  File "/usr/bin/apport-collect", line 97, in upload
    filename=part.get_filename(), is_patch=False)
  File "/usr/lib/python2.6/dist-packages/launchpadlib/resource.py", line 337, in __call__
    args[key] = simplejson.dumps(value, cls=DatetimeJSONEncoder)
  File "/var/lib/python-support/python2.6/simplejson/__init__.py", line 243, in dumps
    **kw).encode(obj)
  File "/var/lib/python-support/python2.6/simplejson/encoder.py", line 360, in encode
    return encode_basestring_ascii(o)
UnicodeDecodeError: 'utf8' codec can't decode byte 0x81 in position 12396: unexpected code byte

If you can tell what I'm missing, I'll fix that too. I don't want to report that as its own bug since I suspect the problem is unique to my system.

Martin Olsson (mnemo) wrote :

Maybe you can manually attach xorg.log plus output of "xrandr --verbose" and "glxinfo".

In general I would not expect that quick turn around on this bug, we're getting bucket loads of bugs right now due to the april release and also we don't have source code for fglrx. If you can narrow down the bug report further that would be great, and then report it upstream to either fglrx devs or maybe some X.org package at bugs.freedesktop.org

Architecture: amd64
DistroRelease: Ubuntu 9.04
NonfreeKernelModules: fglrx
Package: xorg 1:7.4~5ubuntu18
PackageArchitecture: amd64
ProcEnviron:
 SHELL=/bin/bash
 PATH=(custom, user)
 LANG=en_US.UTF-8
ProcVersion: Linux version 2.6.28-11-generic (buildd@crested) (gcc version 4.3.3 (Ubuntu 4.3.3-5ubuntu4) ) #42-Ubuntu SMP Fri Apr 17 01:58:03 UTC 2009
Uname: Linux 2.6.28-11-generic x86_64
UserGroups: adm admin audio cdrom dialout lpadmin mpd plugdev sambashare

aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :
aarons (aarons-ubuntu) wrote :
Martin Olsson (mnemo) on 2009-04-27
affects: xorg (Ubuntu) → fglrx-installer (Ubuntu)
Changed in fglrx-installer (Ubuntu):
status: Incomplete → Confirmed
Bryce Harrington (bryce) on 2009-04-28
description: updated
Bryce Harrington (bryce) on 2009-06-26
summary: - SIGWINCH causes xserver to briefly hang
+ [HD 3400] SIGWINCH causes xserver to briefly hang
Bryce Harrington (bryce) wrote :

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

Changed in fglrx-installer (Ubuntu):
status: Confirmed → New
status: New → Incomplete
aarons (aarons-ubuntu) wrote :

I tested the new driver with the standard xorg packages from jaunty, and then with the upgraded packages from the ppa, but the problem still exists in both. If all this is is a repackaging of the 9.6 driver from ati, then I've actually already tested that. I upgrade the driver using their script when I see a new one, because I'm always hoping this issue (or one of the other less annoying issues) will be resolved by the new version. So far only one annoying issue ever has been.

Bryce Harrington (bryce) wrote :

Thanks for testing, yes it's the same driver, just repackaged for Ubuntu.

Changed in fglrx-installer (Ubuntu):
importance: Undecided → Medium
status: Incomplete → Confirmed
Bryce Harrington (bryce) on 2009-08-13
tags: added: jaunty

I'm still experiencing this with fglrx 8.66.10 (Sep 3 2009) / Catalyst 9.10 from the Ubuntu repos on 9.10 x86-64 with Radeon HD 3400 hardware (in a Thinkpad T400), and I'd say this would be (currently) the biggest drawback of the fglrx driver.

I've updated the upstream report also, at http://ati.cchtml.com/show_bug.cgi?id=1546 - which has the URL of this LP bug report too.

If you want to raise the priority of this bug with ATI, it may be worthwhile selecting 'Vote for this bug' (may need to be logged in) at:
http://ati.cchtml.com/show_bug.cgi?id=1546

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.