problems triggered when I open the laptop lid

[Nicolò Chieffo]

Binary package hint: gnome-settings-daemon

The problems I have are minor, and are not present if gnome-settings-daemon is
not running.

They come every time I open the laptop lid (of course after having closed it,
no matter for how long)
1) the first bulk of keypresses lag some seconds (no matter when I hit the
keys, it can also be 1 minute after I opened the lid) then the keyboard goes
back to normal work
2) some panel icons (apparently all applets) suddenly disappear and reappear
(like a flahs). This does not always happen

ProblemType: Bug
Architecture: amd64
Date: Sat Feb 20 11:03:42 2010
DistroRelease: Ubuntu 10.04
Package: gnome-settings-daemon 2.29.90-0ubuntu1
ProcVersionSignature: Ubuntu 2.6.32-13.19-generic
SourcePackage: gnome-settings-daemon
Uname: Linux 2.6.32-13-generic x86_64

[Kai Jauch (kaijauch)]

This is the output of "gnome-settings-daemon --no-daemon --debug", started with an open lid. I closed the lid once and opened it again.

[Kai Jauch (kaijauch)]

I suspect that the following is happening:

- lid close causes a (fake) fn-f8 keypress (display toggle on the Dell Latitude E6400, gsd calls it fn-f7), gsd reconfigures outputs (all outputs off)
- lid open causes another (fake) fn-f8 keypress, gsd reconfigures outputs (LVDS on)

So, for one lid close and open, gsd should receive two (fake) fn-f8 keypresses and reconfigure the outputs each time. Unfortunately any keypress after a fn-f8 keypress will cause a duplicate fn-f8 keypress to appear (bug #535348), causing this:

- lid close -> fake fn-f8 keypress, gsd reconfigures outputs
- lid open -> fake fn-f8 keypress, which is causing a duplicate fn-f8 keypress to appear. gsd reconfigures outputs twice
- if you hit any key at any time after you've opened the lid, this causes another duplicate fn-f8 keypress to appear. gsd reacts and again reconfigures outputs, causing the visual lag and flashing of icons.

This is visible in the log I attached before, gsd claims to handle "fn-f7"-events four times.

[Kai Jauch (kaijauch)]

Fixing bug #535348 should provide correct behaviour. gsd should reconfigure the outputs after the lid was closed (all outputs are disconnected on lid-close) and again after it was opened (LVDS1 reconnected).

[Nicolò Chieffo (yelo3)]

Just for a curiosity, if I press fn-f8 in the terminal or a Gtk widget
with text, the letter "p" is printed

[Chris Coulson (chrisccoulson)]

On Wed, 2010-03-17 at 09:29 +0000, Nicolò Chieffo wrote:
> Just for a curiosity, if I press fn-f8 in the terminal or a Gtk widget
> with text, the letter "p" is printed
>
If you share similar hardware to me, then that is bug 539477

[Nicolò Chieffo (yelo3)]

Thanks

[Nicolò Chieffo (yelo3)]

Actually when closing and opening the lid, it generates 4 keypress and
release numbered 227.

[Nicolò Chieffo (yelo3)]

The problems I have are minor, and are highlighted by gnome-settings-daemon

Every time I open the laptop lid:
1) the first bulk of keypresses lag some seconds (no matter when I hit the
keys, it can also be 1 minute after I opened the lid) then the keyboard goes
back to normal work
2) some panel icons (apparently all applets) suddenly disappear and reappear
(like a flash). This does not always happen

Without gnome-settings-daemon running, everything is normal.

After having filed a bug to gnome-settings-daemon, developers discovered that some xrandr events are delayed until a key is pressed, and asked to file a bug to Xorg.

[Nicolò Chieffo (yelo3)]

As asked by Chase Douglas (in the upstream bug) I widened the target to Xorg, because he discovered that an xrandr event is delayed until a key is pressed.

[Nicolò Chieffo (yelo3)]

I hope that someone reads this message.

I watched the file /var/log/messages and I found this output

AFTER CLOSING THE LAPTOP LID:
un 6 17:20:14 Latitude-E6400 kernel: [ 5874.520456] Skipping EDID probe due to cached edid
Jun 6 17:20:15 Latitude-E6400 kernel: [ 5875.407947] dell-wmi: Received unknown WMI event (0x11)

AFTER OPENING THE LAPTOP LID:
Jun 6 17:20:27 Latitude-E6400 kernel: [ 5887.682786] Skipping EDID probe due to cached edid
Jun 6 17:20:27 Latitude-E6400 kernel: [ 5887.919909] dell-wmi: Received unknown WMI event (0x11)

AFTER USING THE KEYBOARD:
Jun 6 17:20:40 Latitude-E6400 kernel: [ 5900.510227] Skipping EDID probe due to cached edid
Jun 6 17:20:40 Latitude-E6400 kernel: [ 5901.330366] Skipping EDID probe due to cached edid

Can you please tell me if you think that this problem is in xrandr, or in the kernel? Thanks

[Nicolò Chieffo (yelo3)]

I've just tried from a fedora 13 live session. This problem is not
present! Which patch has fedora that ubuntu doesn't have?

[Sebastien Bacher (seb128)]

http://cvs.fedoraproject.org/viewvc/F-13/gnome-settings-daemon/ seems to indicator they have no g-s-d change, could be due to some difference in drivers versions, xorg, etc

[Nicolò Chieffo (yelo3)]

Well, as reported in the upstream bug, they tend to think that the
cause is in xorg: after hitting any keyboard key another randr event
is fired, so displays get reconfigured and I see the lag + flicker.

[Jeremy Sequoia (jeremyhu)]

I don't see how this could be happening in xorg-server ... my guess is that it is kernel related.

Keith, can you comment?

Nicolo, are you still experiencing with newer kernels? Can you attach system information:

lspci -v
dmesg
/var/log/Xorg.*.log
etc

[Nicolò Chieffo (yelo3)]

Hello! The problem is that now when I close the lid my laptop suspends, and there's no more an option to disable this behavior.
Any help? (I'm running the developement version of ubuntu)

[Nicolò Chieffo (yelo3)]

Hi back, I managed to remove the automatic sleep changing the dconf keys to "blank" instead of "suspend".
The bug is still present!

I'll attach the requested files

[Nicolò Chieffo (yelo3)]

Created attachment 47265
dmesg

[Nicolò Chieffo (yelo3)]

Created attachment 47266
lspci

[Nicolò Chieffo (yelo3)]

Created attachment 47267
Xorg.0.log

[Chris Wilson (ickle)]

This just sounds like silly userspace...

However, one question I can easily google: http://askubuntu.com/questions/15520/how-can-i-tell-ubuntu-to-do-nothing-when-i-close-my-laptop-lid

And I'd be concerned about userspace crashing like that upon resume...

[Sebastien Bacher (seb128)]

the issue should be fixed in the current version

[Chris Wilson (ickle)]

No evidence presented here for the delayed xrandr events.