gnome-shell crashed with SIGSEGV in _clutter_input_device_reset_scroll_info()

[Anthony]

[ Impact ]

Using Ubuntu 16.04 and 20.04 LTS Beta release. Was using gnome-shell and using Chrome web browser when computer became unresponsive to any commands or mouse clicks. About 3-4 minutes later a crash report popped up on my screen. I clicked report and now I am reporting this as a new bug as I have never had this happen before on previous versions of Ubuntu. I was attempting to copy and save an image from Google Chrome when this happened.

Similar crash still happening, as per https://errors.ubuntu.com/problem/fda12af21f79f7f1e23ab3cc830da47eca410baa

----

[ Test case ]

Run xinput and figure device IDs for your mouse.
Run this evil script with 9 changed (be sure to change both) to your id:

  for ((;;)) do xinput disable 9; echo Disabled; xinput enable 9; echo Enabled; done

While the script runs, open some context menu in some program. It may require a few clicks due to evil script enabling/disabling mouse.
This will sometimes crash GTK programs due to a mirror bug in GTK.

GNOME Shell should not crash

[ Regression potential ]

Devices might be not be properly recognized when plugged and unplugged

----

Description: Ubuntu Xenial Xerus (development branch)
Release: 16.04

lostprophet@DigitalDamage:~$ apt-cache policy google-chrome-stable
google-chrome-stable:
  Installed: 49.0.2623.75-1
  Candidate: 49.0.2623.75-1
  Version table:
 *** 49.0.2623.75-1 100
        100 /var/lib/dpkg/status

ProblemType: CrashDistroRelease: Ubuntu 16.04
Package: gnome-shell 3.18.4-0ubuntu2
ProcVersionSignature: Ubuntu 4.4.0-14.30-generic 4.4.5
Uname: Linux 4.4.0-14-generic x86_64
ApportVersion: 2.20-0ubuntu3
Architecture: amd64
CurrentDesktop: GNOME
Date: Thu Mar 17 22:20:56 2016
DisplayManager: lightdm
ExecutablePath: /usr/bin/gnome-shell
InstallationDate: Installed on 2016-02-24 (22 days ago)
InstallationMedia: Ubuntu 15.10 "Wily Werewolf" - Release amd64 (20151021)
ProcCmdline: /usr/bin/gnome-shell
ProcEnviron:
 LANGUAGE=en_US
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=<set>
 LANG=en_US.UTF-8
 SHELL=/bin/bash
SegvAnalysis:
 Segfault happened at: 0x7f634ba78320: mov 0xe0(%rdi),%rax
 PC (0x7f634ba78320) ok
 source "0xe0(%rdi)" (0x000000e0) not located in a known VMA region (needed readable region)!
 destination "%rax" ok
SegvReason: reading NULL VMASignal: 11SourcePackage: gnome-shell
StacktraceTop:
 ?? () from /usr/lib/x86_64-linux-gnu/libclutter-1.0.so.0
 ?? () from /usr/lib/x86_64-linux-gnu/libclutter-1.0.so.0
 ?? () from /usr/lib/x86_64-linux-gnu/libclutter-1.0.so.0
 clutter_x11_handle_event () from /usr/lib/x86_64-linux-gnu/libclutter-1.0.so.0
 ?? () from /usr/lib/libmutter.so.0
Title: gnome-shell crashed with SIGSEGV in clutter_x11_handle_event()
UpgradeStatus: Upgraded to xenial on 2016-03-03 (14 days ago)
UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo

[Apport retracing service (apport)]

StacktraceTop:
 _clutter_input_device_reset_scroll_info () from /tmp/apport_sandbox_MOoLFt/usr/lib/x86_64-linux-gnu/libclutter-1.0.so.0
 clutter_device_manager_xi2_translate_event () from /tmp/apport_sandbox_MOoLFt/usr/lib/x86_64-linux-gnu/libclutter-1.0.so.0
 clutter_backend_real_translate_event () from /tmp/apport_sandbox_MOoLFt/usr/lib/x86_64-linux-gnu/libclutter-1.0.so.0
 clutter_x11_handle_event () from /tmp/apport_sandbox_MOoLFt/usr/lib/x86_64-linux-gnu/libclutter-1.0.so.0
 x_event_source_dispatch () from /tmp/apport_sandbox_MOoLFt/usr/lib/libmutter.so.0

[Launchpad Janitor (janitor)]

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

[Marco Trevisan (Treviño) (3v1n0)]

Even though the trace is quite different, I'm confident that this is due to the same root cause of lp:1846403 and https://gitlab.gnome.org/GNOME/mutter/-/issues/1345

[Brian Murray (brian-murray)]

Hello Anthony, or anyone else affected,

Accepted mutter into focal-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/mutter/3.36.6-1ubuntu0.20.04.1 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 on 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, what testing has been performed on the package and change the tag from verification-needed-focal to verification-done-focal. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed-focal. In either case, without details of your testing we will not be able to proceed.

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

N.B. The updated package will be released to -updates after the bug(s) fixed by this package have been verified and the package has been in -proposed for a minimum of 7 days.

[Brian Murray (brian-murray)]

Hello Anthony, or anyone else affected,

Accepted mutter into focal-proposed. The package will build now and be available at https://launchpad.net/ubuntu/+source/mutter/3.36.6-1ubuntu0.20.04.2 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 on 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, what testing has been performed on the package and change the tag from verification-needed-focal to verification-done-focal. If it does not fix the bug for you, please add a comment stating that, and change the tag to verification-failed-focal. In either case, without details of your testing we will not be able to proceed.

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

N.B. The updated package will be released to -updates after the bug(s) fixed by this package have been verified and the package has been in -proposed for a minimum of 7 days.

[Daniel van Vugt (vanvugt)]

I am unable to reproduce/verify the crash here. It's also unclear if the attach upstream link is really related to this bug.

[Cristiano Fraga G. Nunes (cfgnunes)]

I can't reproduce the crash by using the steps mentioned in "[Test case]".

After following these steps, the process Xorg consume about 100% of CPU.

[Cristiano Fraga G. Nunes (cfgnunes)]

I've also tried to reproduce adding a delay (sleep 0.1) between the commands, but without success.

I've tried:
$ xinput (to get the <ID> number of my mouse)
$ while true; do xinput disable <ID>; sleep 0.1; xinput enable <ID>; sleep 0.1; done

[Cristiano Fraga G. Nunes (cfgnunes)]

This validation is blocking the commit for the bug 1892440 (that has exhaustively already validated).

[Daniel van Vugt (vanvugt)]

There's no sense stalling the entire focal update for just this fix. We have tried to verify it and the results are inconclusive.

Given that https://errors.ubuntu.com/ is showing the crash reliably up to and including gnome-shell 3.36.4, we can use that as the final verification. If the crash continues to be reported after that then we can just reopen this bug.

[Marco Trevisan (Treviño) (3v1n0)]

Daniel,

Detailed ways to reproduce the crash are explained upstream at https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/823#note_866367

The crash was happening easily by using the above script, if it doesn't anymore than it's a verification.
I also verified this in my setup, but preferred to have some other reports.

[Launchpad Janitor (janitor)]

This bug was fixed in the package mutter - 3.36.6-1ubuntu0.20.04.2

---------------
mutter (3.36.6-1ubuntu0.20.04.2) focal; urgency=medium

  * xrandr-scaling: Never try to set invalid screen sizes (again!)
    Looks like this change (introduced to fix LP:1889090) got lost while
    rebasing previous fixes, gbp has been a bad guy this time! (LP: #1896818)

mutter (3.36.6-1ubuntu0.20.04.1) focal; urgency=medium

  * Merge with debian, containing new stable release (LP: #1896332)
  * debian/patches:
    - Refresh patches and drop applied ones
    - Don't set the font-dpi on clutter-backend-x11 (LP: #1892440)
    - Cherry-pick upstream night-switch fixes for 3.36 branch
    - Cherry-pick upstream fixes for handling X11 devices removal
      (LP: #1558886)
  * xrandr-scaling: Don't update UI scale factor on xrandr-manager creation
    (LP: #1892440)
  * debian/control: Revert pipewire version bumping
  * Remaining changes with debian:
    - debian/control:
      + Update VCS flags to point to ubuntu salsa branch
    - debian/gbp.conf: update branch to point to ubuntu/master
    - debian/patches/x11-Add-support-for-fractional-scaling-using-Randr.patch:
      + X11: Add support for fractional scaling using Randr

mutter (3.36.6-1) unstable; urgency=medium

  * Team upload
  * New upstream release
  * Update symbols file
  * d/control.in, d/rules: Enable Pipewire on Debian (but not Ubuntu).
    Now that we have Pipewire 0.3, we can enable screencasting and remote
    desktop support again.

mutter (3.36.5-1) unstable; urgency=medium

  * Team upload
  * New upstream release
    - Screencast fixes and improvements
    - Fix glitches when subsurfaces extend outside the toplevel
    - Improve background display in overview workspace switcher
    - Fix wine copy & paste
    - Plug memory leaks
  * Drop patches that were applied upstream

mutter (3.36.4-1) unstable; urgency=medium

  * New upstream stable release (LP: #1887998)
    - Fix crash on area screenshots with fractional scaling
    - Do not paint textures of fully obscured windows
    - Turn off CRTCs as well when enabling DPMS
    - Improve selection support
    - Use a more appropriate combine function on opaque areas
    - Fix remote desktop being broken without screencast session
    - Fix popovers disappearing on wayland and HiDPI
    - Fixed crashes (LP: #1870867, LP: #1857947)
    - Plugged memory leaks
  * d/p/screen-cast-Let-the-reason-for-recording-determine-what-t.patch,
    d/p/screen-cast-src-Add-flag-to-maybe_record.patch,
    d/p/screen-cast-src-Fix-signedness-of-timestamp-field.patch,
    d/p/screen-cast-src-Make-record-functions-return-an-error-whe.patch,
    d/p/screen-cast-src-Make-the-two-record-vfuncs-more-similarly.patch,
    d/p/screen-cast-src-Record-follow-up-frame-after-timeout.patch,
    d/p/screen-cast-src-Remove-follow-up-timeout-source-on-disabl.patch,
    d/p/screen-cast-src-Use-G_USEC_PER_SEC-instead-of-1000000.patch,
    d/p/screen-cast-window-stream-src-Fix-indentation.patch:
    - Import more fixes for screencasting and remote desktop

 -- Marco Trevisan (Treviño) <email address hidden> Wed, 23 Sep 2020 22:03:53 +0200

[Brian Murray (brian-murray)]

The verification of the Stable Release Update for mutter has completed successfully and the package is now being 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 regressions.

[Daniel van Vugt (vanvugt)]

I'm still not sure what to make of this bug. The fix linked upstream:

  https://gitlab.gnome.org/GNOME/mutter/-/issues/1345

is supposedly in mutter 3.35.2 but there are plenty of these crashes after that:

  https://errors.ubuntu.com/problem/fda12af21f79f7f1e23ab3cc830da47eca410baa

So it seems like at least one of those should not be linked here. That said, if we don't see any more crashes in 3.36.6 and 3.38.1 then there's nothing left to do and resolving the confusion is not required.

[Daniel van Vugt (vanvugt)]

Still happening in 3.36.9 (bug 1965252 as well as thousands of reports in https://errors.ubuntu.com/problem/fda12af21f79f7f1e23ab3cc830da47eca410baa)

[Daniel van Vugt (vanvugt)]

There are no crashes after 3.38 though, so I'm still hopeful this has been fixed in Ubuntu 21.10 onward.