should go to sleep with throttled screensaver

[Tormod Volden]

When I press the sleep button, I can see the screen go blank, then I see a screensaver hack starting for a second, then it goes blank again before the laptop goes into sleep.

When I wake up the laptop again, the screensaver hack comes on for a second, before the screen goes blank to show the password dialog.

It would be better if g-p-m told the screensaver to lock without starting any hack, like if using the --throttle option.

This is not only esthetics, but I have some GL hacks that doesn't like to be put to sleep on my hardware, and a throttle would work around that.

[Tormod Volden (tormodvolden)]

g-p-m "doing gnome-screensaver lock".

A tempting fix would be to call gpm_screensaver_enable_throttle (TRUE) before gpm_screensaver_lock, and similarly dethrottle after poke at wake up. This would not save any throttle state that the user himself had set, but this is not taken care of when for instance closing/opening the lid either.

[Richard Hughes (richard-hughes)]

This is better handled in CVS a few weeks ago, as the g-s API changed to use cookies (so one program could have multiple locks). I'm not sure the 2-14 branch should call gpm_screensaver_enable_throttle as this will do bad things when the user actually wants g-p-m to throttle.

[Tormod Volden (tormodvolden)]

If I understand correctly the 2-14 branch is already doing this (calling gpm_screensaver_enable_throttle to enable/disable throttle) at lid close/open, right?

[Ralf Nieuwenhuijsen (ralf-nieuwenhuijsen)]

Is this bug still valid?

[Tormod Volden (tormodvolden)]

Yes, this is still the same in Feisty.

[Tormod Volden (tormodvolden)]

Same behaviour in Gutsy.

[Caroline Ford (secretlondon)]

I'm guessing you still have the same problem in hardy.

Could you look at https://wiki.ubuntu.com/DebuggingGNOMEPowerManager and give us some information?

dbus-monitor --session "type='signal',interface='org.freedesktop.PowerManagement.Backlight'"

gconftool --recursive-list /apps/gnome-power-manager > gpm.gconf.values.txt

I'm not sure what else might be useful.

[Tormod Volden (tormodvolden)]

I am not sure why the backlight signaling is relevant here. What I see is a problem, is that the sleep function starts a screensaver before sending the computer off to sleep. I understand that it does this as a way to have the screen locked at resume, but in that case it should start a "throttled" screensaver which just blanks (and locks) the screen.

I don't think this has anything to do with my hardware or configuration (except having set the screensaver preference to something else than "Blank") so I think anyone can confirm this by noticing a screensaver starting just before the computer goes to sleep. Unless your hardware is so fast that you can't catch it.

[Tormod Volden (tormodvolden)]

I am setting this back to New since I don't see any relevant information is missing. I still see this in Hardy, the screensaver hack is started for a second before the computer goes to sleep. Unfortunately I can not try suspend in Intrepid yet, since my installation is on a USB drive, and this breaks resume.

[Scott Howard (showard314)]

Thanks for your comments. Does this still exist in Jaunty (or Karmic development release if you have it)? If so, we could mark this as confirmed since we have the necessary information to work on this.

[Tormod Volden (tormodvolden)]

Yes, this is the same in Karmic. Should be very simple for anyone to confirm: Just choose a screensaver in the preferences (for instance Swirl is good because is starts up fast and fills the screen. Suspend the machine and watch as the screensaver starts running for a second before the machine stops.

[Scott Howard (showard314)]

Thank you for your bug report. This bug is probably better handeld and discussed upstream, and has been reported to the developers of the software. You can track it and make comments at:
http://bugzilla.gnome.org/show_bug.cgi?id=586555

[Tormod Volden (tormodvolden)]

I sent a patch upstream, test packages are in my PPA.

[Tormod Volden (tormodvolden)]

http://git.gnome.org/cgit/gnome-power-manager/commit/?id=5d3b1cecc36038695eee13dae6d8eeb68e58d984

[Tormod Volden (tormodvolden)]

Note that the throttling has now been introduced in the karmic indicator-session, so the the debdiff here will make sure that the suspend/resume cycle will look the same whether it is invoked by the menu or the sleep button.

[Martin Pitt (pitti)]

Applied to trunk:

http://git.gnome.org/cgit/gnome-power-manager/commit/?id=5d3b1cecc36038695eee13dae6d8eeb68e58d984

I'm asking Richard whether this will get into 2.28.1 (which we'll upload next week anyway with the new gnome point release), or whether we should cherrypick this.

[Martin Pitt (pitti)]

Richard confirmed that this will land on the 2.28 branch, so we'll get it with next week's gnome updates.

[Martin Pitt (pitti)]

This was included in 2.8.1.