Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Dantrell's GNOME Without Systemd Project (Part 1)
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3, ... 27, 28, 29  Next  
This topic is locked: you cannot edit posts or make replies.    Gentoo Forums Forum Index Unsupported Software
View previous topic :: View next topic  
Author Message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Wed Dec 23, 2015 9:22 pm    Post subject: Reply with quote

Shibotto wrote:
dantrell wrote:
Are you experiencing wallpaper corruption when resuming from suspend?

So it seems
http://i.imgur.com/li3UPrZ.jpg
This or similar patterns is what I get when I resume from suspend. Using Nvidia 340.96.

Thanks for confirming the issue. I'll be giving priority to correcting this although it may take a while since I'm traveling (holidays and all).
_________________
Dantrell B.
Back to top
View user's profile Send private message
Anarcho
Advocate
Advocate


Joined: 06 Jun 2004
Posts: 2970
Location: Germany

PostPosted: Mon Jan 04, 2016 1:00 pm    Post subject: Reply with quote

Hi Dantrell,

first, thanks much for the systemd-free gnome!

I just made a clean fresh install with Gnome-3.18 from your overlay. Everything seems to work but the laptop does not enter sleep mode when I close the lid. Do you have any hint where to look? ACPID logs the close event.

Best regards,

Matthias
_________________
...it's only Rock'n'Roll, but I like it!
Back to top
View user's profile Send private message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Tue Jan 05, 2016 6:49 pm    Post subject: Reply with quote

Anarcho wrote:
Hi Dantrell,

first, thanks much for the systemd-free gnome!

I just made a clean fresh install with Gnome-3.18 from your overlay. Everything seems to work but the laptop does not enter sleep mode when I close the lid. Do you have any hint where to look? ACPID logs the close event.

Best regards,

Matthias

Thanks for confirming that closing a laptop's lid does not trigger suspend/hibernate. I have been trying to get feedback on that specific functionality since GNOME 3.12 as I don't have a device to test it out on.

Now that I know it's broken I have a fix in mind. It might take a while to deploy though as I'm still traveling. If you cant wait, I believe it's GNOME Settings Daemon that needs to be patched.
_________________
Dantrell B.
Back to top
View user's profile Send private message
Anarcho
Advocate
Advocate


Joined: 06 Jun 2004
Posts: 2970
Location: Germany

PostPosted: Tue Jan 05, 2016 8:22 pm    Post subject: Reply with quote

Confirming the bug was the easy part ;) I'm glad that it wasn't my fault!

Thanks to your hint I did take a look at the code in gnome-settings-daemon and from what I found out is that the real suspend command is not actually called (strange is, I couldn't find the call in the unmodified source, too). I modified your patch with the following patch:

Code:
--- gnome-settings-daemon-3.16.2-restore-deprecated-code.patch   2016-01-05 21:16:38.613012284 +0000
+++ gnome-settings-daemon-3.16.2-restore-deprecated-code.patch_patched   2016-01-05 21:07:22.430041002 +0000
@@ -604,6 +604,17 @@
  do_lid_closed_action (GsdPowerManager *manager)
  {
          /* play a sound, using sounds from the naming spec */
+@@ -1307,6 +1307,10 @@
+                         /* We put the screensaver on * as we're not suspending,
+                          * but the lid is closed */
+                         lock_screensaver (manager);
++                } else {
++                        g_debug ("Starting suspend");
++         // do the real suspend now
++                   do_power_action_type (manager, GSD_POWER_ACTION_SUSPEND);
+                 }
+         }
+ }
 @@ -1908,6 +1932,24 @@
                  handle_wake_up_screen (GSD_POWER_MANAGER (user_data));
  }


This adds the call to suspend and works for me. Feel free if you want to integrate.

Best regards,

Matthias

P.S. It works for me but I didn't dig too deep to check if it's really correct.
_________________
...it's only Rock'n'Roll, but I like it!
Back to top
View user's profile Send private message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Sat Jan 16, 2016 8:42 am    Post subject: Reply with quote

I'm surprisingly still traveling but I somehow managed to get the first round of updates in for the year. This includes fixing "suspend on laptop lid close" functionality.

Next on the agenda is tackling wallpaper corruption when resuming from suspend for GNOME 3.16+. I'd like to say this will be done soon but the reality is that it will take a while longer.

Thanks again to everyone who contributed.
_________________
Dantrell B.
Back to top
View user's profile Send private message
simonvanderveldt
Apprentice
Apprentice


Joined: 26 Jan 2016
Posts: 151

PostPosted: Sat Apr 02, 2016 11:05 am    Post subject: Reply with quote

Tried to update my system today and ran into this, anybody have a suggestion what could be causing it? I did notice an update to gentoo-overlay-dantrell-gnome from 2 days ago, but don't immediately see the reason why this is clashing.
http://pastebin.com/deFxvMf2
Back to top
View user's profile Send private message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Sun Apr 03, 2016 3:13 am    Post subject: Reply with quote

simonvanderveldt wrote:
Tried to update my system today and ran into this, anybody have a suggestion what could be causing it? I did notice an update to gentoo-overlay-dantrell-gnome from 2 days ago, but don't immediately see the reason why this is clashing.
http://pastebin.com/deFxvMf2

Turns out this was due to a missing USE flag so sync up and you should be good to go.

In other news... GNOME 3.20 is underway.
_________________
Dantrell B.
Back to top
View user's profile Send private message
simonvanderveldt
Apprentice
Apprentice


Joined: 26 Jan 2016
Posts: 151

PostPosted: Sun Apr 03, 2016 12:37 pm    Post subject: Reply with quote

dantrell wrote:
simonvanderveldt wrote:
Tried to update my system today and ran into this, anybody have a suggestion what could be causing it? I did notice an update to gentoo-overlay-dantrell-gnome from 2 days ago, but don't immediately see the reason why this is clashing.
http://pastebin.com/deFxvMf2

Turns out this was due to a missing USE flag so sync up and you should be good to go.

Thanks for the quick fix, just updated and everything's fine again :)

Quote:
In other news... GNOME 3.20 is underway.

Awesome! Noticed that in that table 3.18 is "only" silver, but it's been running 100% stable for me the past +-3 months. Is there anything in particular that's not working? Can I help with trying something/logs to try to fix it?
Anyway, thanks for the work you're doing on this, I wouldn't be able to run my machines the way I want it without your overlays :)
Back to top
View user's profile Send private message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Mon Apr 04, 2016 10:23 am    Post subject: Reply with quote

simonvanderveldt wrote:
Awesome! Noticed that in that table 3.18 is "only" silver, but it's been running 100% stable for me the past +-3 months. Is there anything in particular that's not working? Can I help with trying something/logs to try to fix it?

Loosely speaking, 3 aspects are graded: accessibility, usability and stability. Each qualifying aspect gives a grade increase.

GNOME 3.16 and 3.18 are graded at silver because they have gone backwards in accessibility and usability compared to GNOME 3.14 due to certain changes in GTK+, GNOME Shell and Nautilus. GNOME (upstream) would have to revert those changes for me to increase the grading.

I will document this in the overview in case others have wondered as well.

simonvanderveldt wrote:
Anyway, thanks for the work you're doing on this, I wouldn't be able to run my machines the way I want it without your overlays :)

Thanks for the support and speaking of which, GNOME 3.20 is now available for testing. Enjoy. :P
_________________
Dantrell B.
Back to top
View user's profile Send private message
saboya
Guru
Guru


Joined: 28 Nov 2006
Posts: 552
Location: Brazil

PostPosted: Wed Apr 06, 2016 2:41 pm    Post subject: Reply with quote

First of all, thanks for your work, I would not be using Gnome if it wasn't for your overlays.

Now, 2 this:

1 - How do I turn off suspend on close lid? There's an option in the tweak tool for that but it doesn't seem to work. Tested on 3.16 and 3.18. If you could give me some points on where to look I could try to fix it myself and submit a patch.
2 - Is there anything in particular you are looking to test in 3.20? Maybe I could give something back and contribute with some testing and even some patches.

Regards
Back to top
View user's profile Send private message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Thu Apr 07, 2016 1:28 pm    Post subject: Reply with quote

saboya wrote:
First of all, thanks for your work, I would not be using Gnome if it wasn't for your overlays.

Now, 2 this:

1 - How do I turn off suspend on close lid? There's an option in the tweak tool for that but it doesn't seem to work. Tested on 3.16 and 3.18. If you could give me some points on where to look I could try to fix it myself and submit a patch.
2 - Is there anything in particular you are looking to test in 3.20? Maybe I could give something back and contribute with some testing and even some patches.

Regards

Of course, just when the "suspend on laptop lid close" functionality is finally fixed, someone wants it gone. :lol:

Looks like something trivial was missed, perhaps in GNOME Settings Daemon again. I'll check and let you know.

For now, you can try setting the following in /etc/UPower/UPower.conf:

Code:
IgnoreLid=true

As for GNOME 3.20, at this stage I'm just making sure it all builds and runs (which it does, from the reports I have gotten). Just remember it won't be considered "stable" until 3.20.2.
_________________
Dantrell B.
Back to top
View user's profile Send private message
saboya
Guru
Guru


Joined: 28 Nov 2006
Posts: 552
Location: Brazil

PostPosted: Fri Apr 08, 2016 2:18 am    Post subject: Reply with quote

dantrell wrote:
Of course, just when the "suspend on laptop lid close" functionality is finally fixed, someone wants it gone. :lol:

Looks like something trivial was missed, perhaps in GNOME Settings Daemon again. I'll check and let you know.

For now, you can try setting the following in /etc/UPower/UPower.conf:

Code:
IgnoreLid=true


That did the trick, thanks.
Back to top
View user's profile Send private message
simonvanderveldt
Apprentice
Apprentice


Joined: 26 Jan 2016
Posts: 151

PostPosted: Thu Apr 14, 2016 5:41 pm    Post subject: Reply with quote

dantrell wrote:
Of course, just when the "suspend on laptop lid close" functionality is finally fixed, someone wants it gone. :lol:

I'm very happy that suspend on lid close works, I'm going to be using the gnome without SystemD setup I have running on my desktop on my laptop soon so having a working suspend on lid close is definitely appreciated! :)

Quote:
As for GNOME 3.20, at this stage I'm just making sure it all builds and runs (which it does, from the reports I have gotten). Just remember it won't be considered "stable" until 3.20.2.

I'll probably start testing/running it in a month, I'll report back if/when I do.

Small question, is there a specific reason gnome-control-center needs networkmanager with the modemmanager USE flag?
https://github.com/dantrell/gentoo-overlay-dantrell-gnome-3-18/blob/master/gnome-base/gnome-control-center/gnome-control-center-3.18.2.ebuild#L68


Last edited by simonvanderveldt on Fri Apr 15, 2016 7:48 am; edited 1 time in total
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Fri Apr 15, 2016 7:46 am    Post subject: Reply with quote

Good work, guys; keep it up.

Don't ever give up, Dantrell.. ;-)
Back to top
View user's profile Send private message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Fri Apr 15, 2016 8:09 pm    Post subject: Reply with quote

simonvanderveldt wrote:
dantrell wrote:
As for GNOME 3.20, at this stage I'm just making sure it all builds and runs (which it does, from the reports I have gotten). Just remember it won't be considered "stable" until 3.20.2.

I'll probably start testing/running it in a month, I'll report back if/when I do.

Small question, is there a specific reason gnome-control-center needs networkmanager with the modemmanager USE flag?
https://github.com/dantrell/gentoo-overlay-dantrell-gnome-3-18/blob/master/gnome-base/gnome-control-center/gnome-control-center-3.18.2.ebuild#L68

Yes, it is because it's a required dependency of GNOME Control Center.

steveL wrote:
Good work, guys; keep it up.

Don't ever give up, Dantrell.. ;-)

Thanks for the continued suport and on that note, GNOME 3.20.1 is now available for testing. :)
_________________
Dantrell B.
Back to top
View user's profile Send private message
saboya
Guru
Guru


Joined: 28 Nov 2006
Posts: 552
Location: Brazil

PostPosted: Fri Apr 29, 2016 5:23 am    Post subject: Reply with quote

Just switched to 3.20 today, everything seems fine.

For those using touchpads like me, Gnome 3.20 no longer supports synaptics, so you have to switch to libinput, which is still USE-masekd in xorg-drivers. You have to unmask the USE flag and set it you make.conf. Touchpad sensitivy feels awkard right (seems more responsive tho) now but tap to click works just fine.

*edit* - Actually, I had 1 problem: I believe gnome-extra/sushi is missing a slot-dependency: media-libs/clutter-gst:3.0. I got a compile error for the package and had to manually emerge clutter-gst in order to fix it.
Back to top
View user's profile Send private message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Fri Apr 29, 2016 1:08 pm    Post subject: Reply with quote

saboya wrote:
Just switched to 3.20 today, everything seems fine.

For those using touchpads like me, Gnome 3.20 no longer supports synaptics, so you have to switch to libinput, which is still USE-masekd in xorg-drivers. You have to unmask the USE flag and set it you make.conf. Touchpad sensitivy feels awkard right (seems more responsive tho) now but tap to click works just fine.

*edit* - Actually, I had 1 problem: I believe gnome-extra/sushi is missing a slot-dependency: media-libs/clutter-gst:3.0. I got a compile error for the package and had to manually emerge clutter-gst in order to fix it.

Thanks for the feedback, the dependencies for sushi have been updated accordingly. You can now remove clutter-gst from your world file if you didn't oneshot it.
_________________
Dantrell B.
Back to top
View user's profile Send private message
Shibotto
Apprentice
Apprentice


Joined: 19 Jun 2015
Posts: 155
Location: CET/CEST

PostPosted: Thu May 26, 2016 9:33 am    Post subject: Reply with quote

Hi, sorry to bother you again, but here's something that's been bugging me since GNOME 3.18 (using 3.20 now). As soon as GDM starts it triggers screen powersave the same as when a certain timeout has been reached. Since my dialectic is worse than a monkey, I made a short video:

https://www.youtube.com/watch?v=TDhxUiARJ1w

This only occurs when GDM starts (service xdm start or poweron, since it is in default runlevel), not when I logout from a GNOME session. Any idea what I should look for?

SOLVED: this no longer occurs as of GNOME 3.24 with elogind, as expected from dropping ConsoleKit.


Last edited by Shibotto on Mon May 29, 2017 8:20 am; edited 1 time in total
Back to top
View user's profile Send private message
simonvanderveldt
Apprentice
Apprentice


Joined: 26 Jan 2016
Posts: 151

PostPosted: Thu May 26, 2016 9:42 am    Post subject: Reply with quote

AFAIK that's caused by a "feature" in the nvidia binary drivers. Do you use those?
Back to top
View user's profile Send private message
Shibotto
Apprentice
Apprentice


Joined: 19 Jun 2015
Posts: 155
Location: CET/CEST

PostPosted: Thu May 26, 2016 10:12 am    Post subject: Reply with quote

simonvanderveldt wrote:
AFAIK that's caused by a "feature" in the nvidia binary drivers. Do you use those?


Yes I do, 340.96 (310M). Do you have more information about this issue or some kind of workaround? I tried with systemd based distros (Fedora and Arch) and I cannot reproduce this.
Back to top
View user's profile Send private message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Thu May 26, 2016 5:30 pm    Post subject: Reply with quote

Shibotto wrote:
As soon as GDM starts it triggers screen powersave the same as when a certain timeout has been reached.

This is actually a quirk that happens with GDM and ConsoleKit (systemd, as you found, apparently isn't affected). It has been present as early as GNOME 3.12 and may, or may not be, NVIDIA-driver specific.

I looked into this a while ago but did not find an obvious cause. I will check again, but this issue has a lower priority than porting the code to resolve wallpaper corruption when resuming from suspend.

The source is most likely GDM itself and/or GNOME Shell and is probably trivial to fix, just a bit tricky to find.
_________________
Dantrell B.
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Fri May 27, 2016 12:33 am    Post subject: Reply with quote

Shibotto wrote:
As soon as GDM starts it triggers screen powersave the same as when a certain timeout has been reached.

dantrell wrote:
This is actually a quirk that happens with GDM and ConsoleKit (systemd, as you found, apparently isn't affected). It has been present as early as GNOME 3.12 and may, or may not be, NVIDIA-driver specific.

Ah the "quirk"; what a handy way for an upstream to avoid responsibility.. ;)
Quote:
I looked into this a while ago but did not find an obvious cause. I will check again, but this issue has a lower priority than porting the code to resolve wallpaper corruption when resuming from suspend.

Hmm for users on nvidia-based machines, it will have a much higher priority, I suspect; getting it to work at all across all those boxen, could well be seen as more important than suspend/resume for laptops or phones.
"Use nouveau" is not an answer, of course. It's the kind of thing that smells of systemdbust, or Microshite.

What do you recall of it; what is the '"feature" in the nvidia binary drivers' you mentioned?
Quote:
The source is most likely GDM itself and/or GNOME Shell and is probably trivial to fix, just a bit tricky to find.

Sure. It's not your (personal, individual) responsibility.
Back to top
View user's profile Send private message
simonvanderveldt
Apprentice
Apprentice


Joined: 26 Jan 2016
Posts: 151

PostPosted: Fri May 27, 2016 7:51 am    Post subject: Reply with quote

dantrell wrote:
I looked into this a while ago but did not find an obvious cause. I will check again, but this issue has a lower priority than porting the code to resolve wallpaper corruption when resuming from suspend.

There is a bug on this somewhere, but I can't find it anymore.

From what I remember it's a timing issue that only manifests itself with the nvidia binary driver where the driver expects a certain sort of signal within a pretty short amount of time and if that doesn't happen it puts the output/monitor into sleep mode. (could be that I'm wrong, this is written down from a vague recollection of things).

steveL wrote:
Hmm for users on nvidia-based machines, it will have a much higher priority, I suspect; getting it to work at all across all those boxen, could well be seen as more important than suspend/resume for laptops or phones.
"Use nouveau" is not an answer, of course. It's the kind of thing that smells of systemdbust, or Microshite.

It's actually not really a big issue because the monitor just turns black, once you move your mouse or type something it turns on again and it works as expected.
So while it obviously is behaviour that shouldn't happen the actual impact is not that big.
Back to top
View user's profile Send private message
dantrell
l33t
l33t


Joined: 01 Jun 2007
Posts: 915
Location: Earth

PostPosted: Fri May 27, 2016 9:43 pm    Post subject: Reply with quote

simonvanderveldt wrote:
There is a bug on this somewhere, but I can't find it anymore.

After some investigation, I found Unix & Linux Stack Exchange question #272005 (Change GDM display timeout) which led me to Gnome bug #710904 (Gnome screen turns off display after lock).

This was enough to help me pin down the relevant code in the relevant packages. A courtesy glance showed nothing that is NVIDIA specific nor revealed why systemd isn't affected, however, I discovered the presence of code that results in the display aggressively blanking out when not in use (as well as utilizing "barebones" transition times) in the interest of power saving.

After some experimental changes, the new behaviour I accomplished is that on initial startup (i.e. the greeter) and when locking, the display will dim instead of blanking out completely. Recovery is instant compared to now where you have to wait for the display to turn back on. What I would add on to improve this, is using the value from GNOME Control Panel -> Power -> Blank Screen to determine the idle delay before turning off the display.

The dimming effect may actually be NVIDIA-driver specific so more testing is needed (from what I see, it should just stay on), but, if it's not, is dimming sufficient, or is there a preference for the screen to remain on for the duration of the idle period?

Is there a better design?
_________________
Dantrell B.
Back to top
View user's profile Send private message
simonvanderveldt
Apprentice
Apprentice


Joined: 26 Jan 2016
Posts: 151

PostPosted: Fri May 27, 2016 10:27 pm    Post subject: Reply with quote

dantrell wrote:
After some investigation, I found Unix & Linux Stack Exchange question #272005 (Change GDM display timeout) which led me to Gnome bug #710904 (Gnome screen turns off display after lock).

That bug :( 2.5 years old, 80+ responses and no fix, sometimes one has to wonder what's going on in the GNOME camp.

Anyway,
Quote:
This was enough to help me pin down the relevant code in the relevant packages. A courtesy glance showed nothing that is NVIDIA specific nor revealed why systemd isn't affected, however, I discovered the presence of code that results in the display aggressively blanking out when not in use (as well as utilizing "barebones" transition times) in the interest of power saving.

After some experimental changes, the new behaviour I accomplished is that on initial startup (i.e. the greeter) and when locking, the display will dim instead of blanking out completely. Recovery is instant compared to now where you have to wait for the display to turn back on. What I would add on to improve this, is using the value from GNOME Control Panel -> Power -> Blank Screen to determine the idle delay before turning off the display.

The dimming effect may actually be NVIDIA-driver specific so more testing is needed (from what I see, it should just stay on), but, if it's not, is dimming sufficient, or is there a preference for the screen to remain on for the duration of the idle period?

Is there a better design?

Sounds great! Maybe the dimming isn't necessary/could be disabled? Does it just show the screen but with less brightness?
Using the Blank screen time as the delay to blank the screen sounds like a nice improvement.
Back to top
View user's profile Send private message
Display posts from previous:   
This topic is locked: you cannot edit posts or make replies.    Gentoo Forums Forum Index Unsupported Software All times are GMT
Goto page Previous  1, 2, 3, ... 27, 28, 29  Next
Page 2 of 29

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum