xrandr Xerrors with the nvidia binary drivers (optimus)

StacktraceTop:
 XSync (dpy=0x2560760, discard=0) at ../../src/Sync.c:44
 gdk_flush () at /build/buildd/gtk+3.0-3.8.4/./gdk/gdkdisplay.c:478
 gnome_rr_config_apply_with_time (config=config@entry=0x25c2e70, screen=<optimized out>, timestamp=timestamp@entry=0, error=error@entry=0x7fff02dca878) at gnome-rr-config.c:1312
 apply_configuration (manager=manager@entry=0x25c2df0, config=config@entry=0x25c2e70, timestamp=timestamp@entry=0, show_error=show_error@entry=1, save_configuration=save_configuration@entry=0) at gsd-xrandr-manager.c:501
 apply_default_boot_configuration (timestamp=0, mgr=0x25c2df0) at gsd-xrandr-manager.c:1901

Status changed to 'Confirmed' because the bug affects multiple users.

Is this hardware has Nvidia Optimus technology? Nvidia-prime was installed?

If "yes", I think that Bug #1227747 is an duplicate of this.

op 18-10-13 14:02, Sergio Benjamim schreef:
> Is this hardware has Nvidia Optimus technology? Nvidia-prime was
> installed?
>
> If "yes", I think that Bug #1227747 is an duplicate of this.
>
In my case it is both a yes.

op 18-10-13 14:02, Sergio Benjamim schreef:
> Is this hardware has Nvidia Optimus technology? Nvidia-prime was
> installed?
>
> If "yes", I think that Bug #1227747 is an duplicate of this.
>
Yes it is nvidia geforce GT 620M 1GB (with NVIDIA OPTIMUS) in a acer
v5-571g.
I'm thinking it has something to do with the combination of nvidia-prime
and psensor or indicator multiload. Since I disabled both (actually this
morning) it seems my system is a bit more stable, allthoug I still have
a lot of pauses....

My laptop has an optimus graphics card.

I don't have nvidia-prime installed. I condifured it manually since nvidia-prime doesn't configure it for gdm only for lightdm.

and the problem persists, so it's not specifically to do with nvidia-prime.

With nvidia-prime, nvidia-319 drivers, and a GT650M I faced the same problem. The desktop theme was broken, and the keyboard was in English layout, while I configured it for French.

I can reproduce this when using the nvidia driver and offloading rendering to it using RandR on an NVIDIA Optimus system. This is what nvidia-prime does and what NVIDIA suggests on their website.

I've assigned the bug report to both the nvidia driver and the modesettings driver since they are both used. The modesettings driver is the one which ultimately deals with RandR (and therefore the gnome-settings-daemon).

Apparently, if I use the "intel" driver instead of the "modesettings" driver, I can't reproduce the problem. This is not a bug in the nvidia driver, it's definitely an issue in the modesetting driver.

I take back what I said, I had a false positive. I probably know what's going on but I need to investigate this a little further.

Apparently the modesetting driver doesn't play well when trying to set the primary output (which does not seem to be supported by the driver).

In any case, we should really make sure to trap the XError and to ignore it when setting the primary output, instead of only ignoring the error when we look for the primary output. This solves the problem in libgnome-desktop.

The debdiff looks fine to me, thanks Alberto! Could you upstream the bug/patch and include the bug reference in the header? Then feel free to upload

I've released upstream modesetting 0.8.1 which contains the fix for the duplicate VGA-0 output. I've uploaded a package to debian, which should be synced to ubuntu today or tomorrow.

SRU request:

Please accept gnome-desktop3 (3.4.2-0ubuntu0.2) in precise-proposed and gnome-desktop3 (3.8.4-0ubuntu1.1) in saucy-proposed.

[Impact]
In Precise and Saucy, we support hybrid graphics, and this issue prevents users from enabling external screens (the ones connected to the Intel card), and to benefit from all the features that the gnome-settings-daemon provides (font settings, input settings, theming, etc.).

[Test Case]
After installing the update, users should be able to open the Displays configuration tool, and the gnome-settings-daemon will be running (as shown by the "ps aux | grep gnome-settings-daemon" command).

[Regression Potential]
None. We simply trap and ignore an error when setting the primary output, as the rest of the program already does when getting the primary output.

Duplicate output names that crash gsd fixed with xserver-xorg-video-modesetting 0.8.1-1.

Hello odror, or anyone else affected,

Accepted gnome-desktop3 into saucy-proposed. The package will build now and be available at http://launchpad.net/ubuntu/+source/gnome-desktop3/3.8.4-0ubuntu1.1 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, and change the tag from verification-needed to verification-done. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed. In either case, details of your testing will help us make a better decision.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance!

Hello odror, or anyone else affected,

Accepted gnome-desktop3 into precise-proposed. The package will build now and be available at http://launchpad.net/ubuntu/+source/gnome-desktop3/3.4.2-0ubuntu0.2 in a few hours, and then in the -proposed repository.

Please help us by testing this new package. See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Your feedback will aid us getting this update out to other Ubuntu users.

If this package fixes the bug for you, please add a comment to this bug, mentioning the version of the package you tested, and change the tag from verification-needed to verification-done. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed. In either case, details of your testing will help us make a better decision.

Further information regarding the verification process can be found at https://wiki.ubuntu.com/QATeam/PerformingSRUVerification . Thank you in advance!

Hello Brian

Using the proposed package fixes the XReply bug.

I now experience other bugs from other display related packages. I set up my display incorrectly at first so they just might be reports of that. I'll reboot and see if they remain.

Otherwise the theme is correct, the keyboard layout holds and key combinations work again. Thanks.

An update:

No other crashes after rebooting.

Booting with a secondary display attached causes it to fail miserably though. Attaching the display after the xsession is up and running works as it should. I'll play around with the xorg,conf and see if I can sort it out. I don't think it has anything to do with this package. (Probably just a mistake in xorg.conf from my part)

Thanks again.

the package doesn't affect the boot process. That would be an issue in the xserver (i.e. a separate problem).

I've installed the nvidia-319, nvidia-settings-319, nvidia-primus on a ubuntu 13.10 (unity) and had the "[xrandr]: gnome-settings-daemon crashed with signal 5 in _XReply() " when i logged in, so I ran "sudo apt-get build-dep nvidia-319" and reboot the pc, and the error was gone.

PS:
- i've only experienced error on login, so i don't now if it fixed all the problems or just the login one.
- some themes were broken like chromium and stills broken (i don't know if it's a gtk2 or gtk3 theme).

Installed nvidia-319, nvidia-settings-319, nvidia-prime on a ubuntu 13.10 64-bit (unity) (Sony Vaio SBS15127PXB), and it worked perfectly!

glxspheres runnind about 620fps!!! :D

GREAT WORK! Thanks a lot!

Problem persists with nvidia-331 with kernel 3.12

@Sayantan: what version of Ubuntu are you running and what version of libgnome-desktop did you install?

This bug was fixed in the package gnome-desktop3 - 3.4.2-0ubuntu0.2

---------------
gnome-desktop3 (3.4.2-0ubuntu0.2) precise-proposed; urgency=low

  * debian/patches/ignore_errors_with_primary_outputs.patch:
    - Ignore failures when setting the primary output (LP: #1224254).
 -- Alberto Milone <email address hidden> Tue, 03 Dec 2013 16:30:22 +0100

The verification of the Stable Release Update for gnome-desktop3 has completed successfully and the package has now been released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regresssions.

Status changed to 'Confirmed' because the bug affects multiple users.

This bug was fixed in the package gnome-desktop3 - 3.8.4-0ubuntu1.1

---------------
gnome-desktop3 (3.8.4-0ubuntu1.1) saucy-proposed; urgency=low

  * debian/patches/ignore_errors_with_primary_outputs.patch:
    - Ignore failures when setting the primary output (LP: #1224254).
 -- Alberto Milone <email address hidden> Tue, 03 Dec 2013 17:25:35 +0100

saucy has seen the end of its life and is no longer receiving any updates. Marking the saucy task for this ticket as "Won't Fix".

The Precise Pangolin has reached end of life, so this bug will not be fixed for that release