thinkpad_acpi complains that hotkey-setup is doing the wrong thing

Bug #256887 reported by Matt Zimmerman
16
This bug affects 2 people
Affects Status Importance Assigned to Milestone
hotkey-setup (Ubuntu)
Fix Released
Undecided
Unassigned

Bug Description

Binary package hint: hotkey-setup

I noticed this today (on Intrepid):

[13681.078899] thinkpad_acpi: setting the hotkey mask to 0x00ffffff is likely not the best way to go about it
[13681.078899] thinkpad_acpi: please consider using the driver defaults, and refer to up-to-date thinkpad-acpi documentation

/etc/init.d/hotkey-setup says:

do_thinkpad () {
    . /usr/share/hotkey-setup/ibm.hk
    modprobe thinkpad-acpi

    VIDEO=`xorg_driver`
    case $VIDEO in
        intel|ati|radeon)
            echo 0xffffff > $THINKPAD_PROC_HOTKEY
        ;;
    esac

Revision history for this message
Reinhard Tartler (siretart) wrote :

I can confirm this issue on an Thinkpad X60s

Changed in hotkey-setup:
status: New → Confirmed
Revision history for this message
Reinhard Tartler (siretart) wrote :

15:44:48 < shenki> mjg59: do you know what this message is about:
15:44:52 < shenki> "thinkpad_acpi: setting the hotkey mask to 0x00ffffff is likely not the best way to go about it"
15:45:17 < mjg59> Yes, with recent thinkpad_acpis userspace shouldn't be setting hotkey_mask
15:48:38 < shenki> mjg59: where could that be getting set?
15:50:07 < mjg59> /etc/modprobe.d somewhere?

Revision history for this message
Joel Stanley (shenki) wrote :

23:14 < shenki> mjg59: do you know what this message is about:
23:14 < shenki> "thinkpad_acpi: setting the hotkey mask to 0x00ffffff is likely
                not the best way to go about it"
23:15 < mjg59> Yes, with recent thinkpad_acpis userspace shouldn't be setting
               hotkey_mask

Seeing this on a Thinkpad X300.

Revision history for this message
Matt Zimmerman (mdz) wrote : Re: [Bug 256887] Re: thinkpad_acpi complains that hotkey-setup is doing the wrong thing

On Thu, Aug 28, 2008 at 02:01:10PM -0000, Reinhard Tartler wrote:
> 15:44:48 < shenki> mjg59: do you know what this message is about:
> 15:44:52 < shenki> "thinkpad_acpi: setting the hotkey mask to 0x00ffffff is likely not the best way to go about it"
> 15:45:17 < mjg59> Yes, with recent thinkpad_acpis userspace shouldn't be setting hotkey_mask
> 15:48:38 < shenki> mjg59: where could that be getting set?
> 15:50:07 < mjg59> /etc/modprobe.d somewhere?

It's set in /etc/init.d/hotkey-setup

--
 - mdz

Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package hotkey-setup - 0.1-23ubuntu4

---------------
hotkey-setup (0.1-23ubuntu4) intrepid; urgency=low

  * init.d: No need to mess with the Thinkpad hotkey mask anymore.
    Current version of thinkpad_acpi suggests using the default.
    (LP: #256887)

 -- Timo Aaltonen <email address hidden> Mon, 01 Sep 2008 10:29:57 +0300

Changed in hotkey-setup:
status: Confirmed → Fix Released
Revision history for this message
Matt Zimmerman (mdz) wrote :

I've revised this further in Intrepid. There were some problems with the original fix, and it turns out that the driver defaults are actually not good enough for us.

We now take a more conservative approach, and use the driver defaults except for enabling the few things we seem to need.

hotkey-setup (0.1-23ubuntu6) intrepid; urgency=low

  * Revert 0.1-23ubuntu4 and use a different method to set the hotkey mask:

    + Add hotkey mask definitions to ibm.hk
    + Use the sysfs interface rather than the old procfs interface
    + Unconditionally try to enable the bits for volume and brightness
    + Restore the test for the thinkpad key, and start the daemon if
      setting the corresponding bit doesn't work
    + For everything else, just accept the driver defaults

  * Thanks to Matthew Garrett for background on the old code and the
    thinkpad_acpi driver

  * This should address LP: #222796 without reopening LP: #256887

 -- Matt Zimmerman <email address hidden> Mon, 13 Oct 2008 11:21:30 +0100

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

Other bug subscribers

Remote bug watches

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