[FFe] graphics_driver script does not report proprietary driver version

On 12-10-02 09:47 AM, Sylvain Pineau wrote:
> Public bug reported:
>
> I have a system with a NVIDIA graphic chip running the prop drivers.
> This is the output of the checkbox graphics_driver script:
>
> $ ./graphics_driver
> ------------- VIDEO DRIVER INFORMATION -------------
> ERROR: No video driver loaded! Possibly in failsafe mode!
> ------------- HYBRID GRAPHICS CHECK ----------------
> Graphics Chipset: NVIDIA (10de:0a74)
> Graphics Chipset: NVIDIA (10de:0866)
> Loaded DDX Drivers: nvidia, dri2
> Hybrid Graphics: no
>
> But nvidia-settings reports the right version:
>
> $ nvidia-settings -v | egrep -o 'version [0-9\.]+'
> version 304.51
>
> The graphics_driver script should not report an error when prop drivers
> are in use, the warning about a failsafe mode is confusing in that case.

graphics_driver parses /var/log/Xorg.0.log to determine the driver that's in
use. Could you attach that file so we can use it to diagnose this? thanks!

>
> ** Affects: checkbox
> Importance: Undecided
> Status: New
>

This is the diff from the upstream changelog. Since Checkbox has no "upstream changelog" proper, we maintain changes in debian/changelog, so it's the diff of that file.

The stray unrelated line is there due to whitespace removal so it got picked up by diff.

Here's the build log, produced using sbuild with these parameters:

sbuild -A --arch=amd64 -d quantal checkbox_0.14.8.dsc

Here is the console output when installing:

DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=12.10
DISTRIB_CODENAME=quantal
DISTRIB_DESCRIPTION="Ubuntu quantal (development branch)"

Replacing a previous checkbox install:

ubuntu@ubuntu:~$ sudo dpkg -i checkbox_0.14.8_amd64.deb
(Reading database ... 152427 files and directories currently installed.)
Preparing to replace checkbox 0.14.6 (using checkbox_0.14.8_amd64.deb) ...
Unpacking replacement checkbox ...
Setting up checkbox (0.14.8) ...
Processing triggers for man-db ...
Processing triggers for hicolor-icon-theme ...

ubuntu@ubuntu:~$ sudo dpkg -i checkbox-qt_0.14.8_amd64.deb
(Reading database ... 152518 files and directories currently installed.)
Preparing to replace checkbox-qt 0.14.6 (using checkbox-qt_0.14.8_amd64.deb) ...
Unpacking replacement checkbox-qt ...
Setting up checkbox-qt (0.14.8) ...
Processing triggers for man-db ...
Processing triggers for desktop-file-utils ...
Processing triggers for bamfdaemon ...
Rebuilding /usr/share/applications/bamf.index...
Processing triggers for gnome-menus ...

Fresh Install:
ubuntu@ubuntu:~$ sudo dpkg -i checkbox_0.14.8_amd64.deb
Selecting previously unselected package checkbox.
(Reading database ... 152089 files and directories currently installed.)
Unpacking checkbox (from checkbox_0.14.8_amd64.deb) ...
Setting up checkbox (0.14.8) ...
Processing triggers for man-db ...
Processing triggers for hicolor-icon-theme ...

ubuntu@ubuntu:~$ sudo dpkg -i checkbox-qt_0.14.8_amd64.deb
Selecting previously unselected package checkbox-qt.
(Reading database ... 152501 files and directories currently installed.)
Unpacking checkbox-qt (from checkbox-qt_0.14.8_amd64.deb) ...
Setting up checkbox-qt (0.14.8) ...
Processing triggers for man-db ...
Processing triggers for desktop-file-utils ...
Processing triggers for bamfdaemon ...
Rebuilding /usr/share/applications/bamf.index...
Processing triggers for gnome-menus ...

Approved to get this uploaded today or tomorrow (that's before Final Freeze, which is October 9th 2100UTC).

This bug was fixed in the package checkbox - 0.14.8

---------------
checkbox (0.14.8) quantal; urgency=low

  * New upstream release (LP: #1061359)

  * Launchpad automated translation updates

  [Sylvain Pineau]
  * setup.py, qt/checkbox-qt.ui: Removed the old UI design file (LP: #1049912)
  * scripts/display_resource, jobs/resource.txt.in: Added a new display
    resource script to properly handle connector names returned by proprietary
    drivers (LP: #956139) (LP: #992727)
  * jobs/optical.txt.in, scripts/optical_write_test: Filter ANSI escape char
    outputed by wodim and removed some stray whitespace (LP: #1052803)
  * checkbox/parsers/udevadm.py: Improved wireless devices detection.
    The wireless category is now set if the subsystem is equal to ieee80211
    (LP: #855382)
  * debian/control, jobs/esata.txt.in, jobs/firewire.txt.in, jobs/usb.txt.in,
    jobs/mediacard.txt.in: Remove udisks package references in all of the jobs
    that use removable_storage scripts (LP: #1059620). Used wrap-and-sort in
    debian/control for clarity (Depends).
  * scripts/graphics_driver: Added NVIDIA driver detection (LP: #1060211)

  [Brendan Donegan]
  * jobs/cpu.txt.in - add environ field containing
    CHECKBOX_DATA to allow that environment variable to be used in the command
    (LP: #1049595)
  * jobs/wireless.txt.in - replace use of network_wireless_test in wireless_scanning
    with a simple Bash script using nmcli and delete network_wireless_test (LP: #900370)
  * [UIFe] qt/frontend/qtfront.ui, qt/frontend/qtfront.cpp - set focus to Continue
    button and make it the default so that it can be 'clicked' using Enter,
    as well as renaming it to 'continueButton' (LP: #1052506)
  * jobs/usb.txt.in - Fix dependencies of USB tests so that things work properly if
    the usb/storage-automated test fails (LP: #987876)
  * scripts/network_info - add exception handling to file reading so that
    sensible values are given if the required file cannot be read (LP: #823606)

  [Marc Tardif]
  * qt/frontend/qtfront.ui: Fixed warnings when building with
    qtcreator (LP: #1053126)
  * plugins/suites_prompt.py: Fixed tree view in selection window (LP: #1056432)
  * plugins/apport_prompt.py: Removed apport plugin that caused crashes
    when trying to send bug report (LP: #1047857)
  * jobs/optical.txt.in: Fixed missing category assignment in optical
    dvd write tests (LP: #1057762)

  [Jeff Marcom]
  * jobs/optical.txt.in: Changed optical read/write job commands to use
    a more reliable dev path (LP: #990560)

  [Daniel Manrique]
  * Added consecutive numbering to messages sent to the backend, so the
    frontend knows to discard out-of-sequence messages. (LP: #886118)
 -- Daniel Manrique <email address hidden> Wed, 03 Oct 2012 17:23:34 -0400