[GM965/GL960] Pressing Suspend key turns off display instead of suspending

[Janne Hyötylä]

Binary package hint: gnome-power-manager

Laptop: Dell Latitude D630.

When pressing the suspend key (Fn-F1), only the display gets turned off, instead of suspending the laptop.

In the settings, I have the following:
When the suspend button is pressed: Suspend.

The suspend key is recognized correctly. Here the xev output:

KeyPress event, serial 34, synthetic NO, window 0x1400001,
    root 0xff, subw 0x0, time 21205448, (100,89), root:(1302,142),
    state 0x0, keycode 213 (keysym 0x1008ffa7, XF86Suspend), same_screen YES,
    XLookupString gives 0 bytes:
    XmbLookupString gives 0 bytes:
    XFilterEvent returns: False

KeyRelease event, serial 34, synthetic NO, window 0x1400001,
    root 0xff, subw 0x0, time 21205448, (100,89), root:(1302,142),
    state 0x0, keycode 213 (keysym 0x1008ffa7, XF86Suspend), same_screen YES,
    XLookupString gives 0 bytes:
    XFilterEvent returns: False

Then I looked in gconf, XF86Suspend is only found in this key:
/apps/gnome_settings_daemon/keybindings/sleep
so this should be correct.

Output of
$ gnome-power-manager --verbose | tee gpm.debug.log.txt
attached.

You can see keycode 213 (XF86Suspend) being "grabbed" during initialisation (first few rows).

I pressed the button at time 18:24:00, nothing was generated in the log as you can see.

Compared to that, I pressed the power button at 18:25:00, selected suspend from the menu, and then resumed at 18:25:20.

So looks to me that the suspend key is intercepted before GPM gets it (?).

ProblemType: Bug
Architecture: i386
Date: Sat Aug 8 22:05:24 2009
DistroRelease: Ubuntu 9.10
Package: gnome-power-manager 2.27.5-0ubuntu1
ProcEnviron:
 PATH=(custom, user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-5.24-generic
SourcePackage: gnome-power-manager
Uname: Linux 2.6.31-5-generic i686

[Scott Howard (showard314)]

Thank you for taking the time to report this bug and helping to make Ubuntu better. This particular bug has already been reported and is a duplicate of bug 407491, so it is being marked as such. Please look at the other bug report to see if there is any missing information that you can provide, or to see if there is a workaround for the bug. Additionally, any further discussion regarding the bug should occur in the other report. Please continue to report any other bugs you may find.

[Janne Hyötylä (janne-hyotyla)]

Hi Scott,

This is not a duplicate of Bug 407491. I tried the fixed version from a PPA mentioned there and this problem still persists. Moreover, I am using i386, and as I understood from the other bug, it affects only amd64.

I updated the bug description, because I found out that Fn-F1 now turns off the display instead of suspending on both AC and battery.

When I initially installed karmic (between alpha3 and 4), the suspend button was definitely working as intended for a short time. I don't know what could have triggered this bug except maybe it happened after I started gnome-power-preferences for the first time after installation.

[Janne Hyötylä (janne-hyotyla)]

Further: Now when I suspend and resume, the screen is not locked anymore (i.e. I don't have to type in a password). This was definitely not the case directly after installation. Screen locking e.g. by manually activating screen saver works normally. I have a feeling these issues could be related. Or should I open a new bug for this one?

[Scott Howard (showard314)]

Thanks for testing it and confirming that your bug isn't a duplicate. To debug hotkeys, see and follow the inatructions at:
https://wiki.ubuntu.com/Hotkeys/Troubleshooting
to make sure that ubuntu recognizes (Fn-F1) as suspend and not "blank screen"

Also, see
https://wiki.ubuntu.com/DebuggingGNOMEPowerManager
specifically the section:
Getting info from GPM
to get verbose output as to what GPM is doing.

We have to find if the problem is with the linux kernel misidentifying the suspend key as blank screen, or if the problem is in GPM sending a blank screen signal instead of a suspend signal. Could you debug it and report here with the output of the tests?

As for the other problem, I think you should make another report. I remember seeing someone describe something similar (screen isn't locked on resume), but I haven't found the report yet, so it would be good to have yours and we can combine them later when working on them.

[Janne Hyötylä (janne-hyotyla)]

The suspend key is recognized correctly. Here the xev output:

KeyPress event, serial 34, synthetic NO, window 0x1400001,
    root 0xff, subw 0x0, time 21205448, (100,89), root:(1302,142),
    state 0x0, keycode 213 (keysym 0x1008ffa7, XF86Suspend), same_screen YES,
    XLookupString gives 0 bytes:
    XmbLookupString gives 0 bytes:
    XFilterEvent returns: False

KeyRelease event, serial 34, synthetic NO, window 0x1400001,
    root 0xff, subw 0x0, time 21205448, (100,89), root:(1302,142),
    state 0x0, keycode 213 (keysym 0x1008ffa7, XF86Suspend), same_screen YES,
    XLookupString gives 0 bytes:
    XFilterEvent returns: False

Then I looked in gconf, XF86Suspend is only found in this key:
/apps/gnome_settings_daemon/keybindings/sleep
so this should be correct.

I was unable to debug GPM. When running
$ gnome-power-manager --verbose --no-daemon
(after killing running instances)
I get no output to the console whatsoever..?

When no instance gnome-power-manager is running, Fn-F1 still blanks the screen. Other buttons like the power button do not work anymore.

[Scott Howard (showard314)]

Thanks for the info, it does look like ubuntu recognizes the suspend
key being pressed.
Could you try:
killall gnome-power-manager
gnome-power-manager --verbose

(I could get the --verbose to work that way)
There should be a lot of messages in the terminal at first and more
once you hit the suspend key.

That's interesting that with GPM killed, the screen is blanked - it
may be something in the kernel (or xserver) after all. Let's try the
gnome-power-manager --verbose first and see if that works.

2009/8/12 Janne Hyötylä <email address hidden>:
> The suspend key is recognized correctly. Here the xev output:
>
> KeyPress event, serial 34, synthetic NO, window 0x1400001,
>    root 0xff, subw 0x0, time 21205448, (100,89), root:(1302,142),
>    state 0x0, keycode 213 (keysym 0x1008ffa7, XF86Suspend), same_screen YES,
>    XLookupString gives 0 bytes:
>    XmbLookupString gives 0 bytes:
>    XFilterEvent returns: False
>
> KeyRelease event, serial 34, synthetic NO, window 0x1400001,
>    root 0xff, subw 0x0, time 21205448, (100,89), root:(1302,142),
>    state 0x0, keycode 213 (keysym 0x1008ffa7, XF86Suspend), same_screen YES,
>    XLookupString gives 0 bytes:
>    XFilterEvent returns: False
>
>
> Then I looked in gconf, XF86Suspend is only found in this key:
> /apps/gnome_settings_daemon/keybindings/sleep
> so this should be correct.
>
>
> I was unable to debug GPM. When running
> $ gnome-power-manager --verbose --no-daemon
> (after killing running instances)
> I get no output to the console whatsoever..?
>
> When no instance gnome-power-manager is running, Fn-F1 still blanks the
> screen. Other buttons like the power button do not work anymore.
>
> ** Changed in: gnome-power-manager (Ubuntu)
>       Status: Incomplete => New
>
> --
> Pressing Suspend key turns off display instead of suspending
> https://bugs.launchpad.net/bugs/410829
> You received this bug notification because you are a direct subscriber
> of the bug.
>
> Status in “gnome-power-manager” package in Ubuntu: New
>
> Bug description:
> Binary package hint: gnome-power-manager
>
> Laptop: Dell Latitude D630.
>
> When pressing the suspend key (Fn-F1), only the display gets turned off, instead of suspending the laptop.
>
> In the settings, I have the following:
> When the suspend button is pressed: Suspend.
>
> ProblemType: Bug
> Architecture: i386
> Date: Sat Aug  8 22:05:24 2009
> DistroRelease: Ubuntu 9.10
> Package: gnome-power-manager 2.27.5-0ubuntu1
> ProcEnviron:
>  PATH=(custom, user)
>  LANG=en_US.UTF-8
>  SHELL=/bin/bash
> ProcVersionSignature: Ubuntu 2.6.31-5.24-generic
> SourcePackage: gnome-power-manager
> Uname: Linux 2.6.31-5-generic i686
>

[Janne Hyötylä (janne-hyotyla)]

Output of
$ gnome-power-manager --verbose | tee gpm.debug.log.txt
attached.

You can see keycode 213 (XF86Suspend) being "grabbed" during initialisation (first few rows).

I pressed the button at time 18:24:00, nothing was generated in the log as you can see.

Compared to that, I pressed the power button at 18:25:00, selected suspend from the menu, and then resumed at 18:25:20.

So looks to me that the suspend key is intercepted before GPM gets it (?).

[Scott Howard (showard314)]

Thanks for debugging this. I'm also adding xorg to this bugreport since it appears something in X is intercepting the keyevent and blanking the screen. Could you do the following:

Paste in output of: lspci -nn | grep VGA
Attach /var/log/Xorg.0.log

To triagers/developers:
1) xev is showing the key press event for XF86Suspend.
2) with no GPM running, the screen is turned off on keypress
3) on key press with GPM running, GPM does not tell screen is turned off, but it turns off anyways

[Janne Hyötylä (janne-hyotyla)]

$ lspci -nn | grep VGA
00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller [8086:2a02] (rev 0c)

[Scott Howard (showard314)]

Thanks for the information. I'm marking as confirmed since the needed logs are here.

[Geir Ove Myhr (gomyhr)]

As far as I understand the package xserver-xorg-video-intel only has the driver for the video card, and key strokes are handled by some other part of xorg. Assigning back to xorg.

[eljoest (eljoest)]

I'm seeing similar behavior on a Dell Latitude D830 with an external keyboard from Logitech.
The previous installation of Jaunty suspended the machine when the "Suspend" button on the external keyboard was pressed, Karmic simply blanks the screen. The key is mapped to XF86Sleep as intended.

$ lspci -nn | grep VGA
00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller [8086:2a02] (rev 0c)

[Bryce Harrington (bryce)]

[This is an automatic notification.]

Hi Janne,

This bug was reported against an earlier version of Ubuntu, can you
test if it still occurs on Lucid?

Please note we also provide technical support for older versions of
Ubuntu, but not in the bug tracker. Instead, to raise the issue through
normal support channels, please see:

    http://www.ubuntu.com/support

If you are the original reporter and can still reproduce the issue on
Lucid, please run the following command to refresh the report:

  apport-collect 410829

If you are not the original reporter, please file a new bug report, so
we can work with you as the original reporter instead (you can reference
bug 410829 in your report if you think it may be related):

  ubuntu-bug xorg

If by chance you can no longer reproduce the issue on Lucid or if you
feel it is no longer relevant, please mark the bug report 'Fix Released'
or 'Invalid' as appropriate, at the following URL:

  https://bugs.launchpad.net/ubuntu/+bug/410829

[Janne Hyötylä (janne-hyotyla)]

This has been fixed in Lucid, at least partially.
Now the suspend key hibernates the laptop, regardless of the setting in power manager. I will open a new bug for that.

[Owen Williams (ywwg)]

I started seeing this bug in Ubuntu Natty, and I don't know if the cause was the same but I'm adding my experience here for others that find it via googling. In my case, it turned out that since the upgrade, my "suspend" button was now being interpreted as a "hibernate" button instead. There was no entry in gconf for a hibernate command, so it didn't know what to do.

The error in gnome-power-manager was:

- action: /apps/gnome-power-manager/buttons/hibernate set to blank (The hibernate button has been pressed.)

Adding a key called /apps/gnome-power-manager/buttons/hibernate and setting that to the value of "suspend" fixed the problem