gnome-shell crashed with SIGSEGV in meta_window_get_monitor() from ffi_call_unix64() from ffi_call() from gjs_invoke_c_function() from function_call()

Bug #1724439 reported by hat121 on 2017-10-18
438
This bug affects 46 people
Affects Status Importance Assigned to Milestone
Mutter
Fix Released
Medium
gnome-shell (Ubuntu)
High
Marco Trevisan (Treviño)
Bionic
High
Marco Trevisan (Treviño)
mutter (Ubuntu)
High
Marco Trevisan (Treviño)
Bionic
High
Marco Trevisan (Treviño)

Bug Description

https://errors.ubuntu.com/problem/3e662975e4b7e6829b9d68d1c2b06f429e44c854

---

left virtual box to update win10 came back and got error message

ProblemType: Crash
DistroRelease: Ubuntu 17.10
Package: gnome-shell 3.26.1-0ubuntu3
ProcVersionSignature: Ubuntu 4.13.0-16.19-generic 4.13.4
Uname: Linux 4.13.0-16-generic x86_64
NonfreeKernelModules: nvidia_uvm nvidia_drm nvidia_modeset nvidia
ApportVersion: 2.20.7-0ubuntu3
Architecture: amd64
CrashCounter: 1
CurrentDesktop: ubuntu:GNOME
Date: Tue Oct 17 23:37:32 2017
DisplayManager: gdm3
ExecutablePath: /usr/bin/gnome-shell
InstallationDate: Installed on 2017-04-18 (182 days ago)
InstallationMedia: Ubuntu-GNOME 17.04 "Zesty Zapus" - Release amd64 (20170412)
ProcCmdline: /usr/bin/gnome-shell
ProcEnviron:
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=<set>
 LANG=en_US.UTF-8
 SHELL=/bin/bash
SegvAnalysis:
 Segfault happened at: 0x7feb2ed42d94 <meta_window_get_monitor+4>: mov 0x18(%rax),%eax
 PC (0x7feb2ed42d94) ok
 source "0x18(%rax)" (0x00000018) not located in a known VMA region (needed readable region)!
 destination "%eax" ok
SegvReason: reading NULL VMA
Signal: 11
SourcePackage: gnome-shell
StacktraceTop:
 meta_window_get_monitor () from /usr/lib/x86_64-linux-gnu/libmutter-1.so.0
 ffi_call_unix64 () from /usr/lib/x86_64-linux-gnu/libffi.so.6
 ffi_call () from /usr/lib/x86_64-linux-gnu/libffi.so.6
 ?? () from /usr/lib/libgjs.so.0
 ?? () from /usr/lib/libgjs.so.0
Title: gnome-shell crashed with SIGSEGV in meta_window_get_monitor()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo

hat121 (hat-121) wrote :

StacktraceTop:
 meta_window_get_monitor (window=0x5598fc58db60) at core/window.c:3739
 ffi_call_unix64 () at ../src/x86/unix64.S:76
 ffi_call (cif=cif@entry=0x5598fc916818, fn=<optimized out>, rvalue=<optimized out>, rvalue@entry=0x7ffe2184a728, avalue=avalue@entry=0x7ffe2184a620) at ../src/x86/ffi64.c:525
 gjs_invoke_c_function (context=context@entry=0x5598f9fa8c30, function=function@entry=0x5598fc916800, obj=..., obj@entry=..., args=..., js_rval=..., r_value=r_value@entry=0x0) at gi/function.cpp:1033
 function_call (context=0x5598f9fa8c30, js_argc=0, vp=0x7ffe2184aac8) at gi/function.cpp:1351

Changed in gnome-shell (Ubuntu):
importance: Undecided → Medium
tags: removed: need-amd64-retrace
information type: Private → Public

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

Changed in gnome-shell (Ubuntu):
status: New → Confirmed
Daniel van Vugt (vanvugt) wrote :
summary: - gnome-shell crashed with SIGSEGV in meta_window_get_monitor()
+ gnome-shell crashed with SIGSEGV in meta_window_get_monitor() from
+ ffi_call_unix64() from ffi_call() from gjs_invoke_c_function() from
+ function_call()
description: updated
Changed in gnome-shell:
importance: Unknown → Medium
status: Unknown → Confirmed
affects: gnome-shell → mutter
Changed in mutter:
importance: Medium → Unknown
status: Confirmed → Unknown
Changed in mutter:
importance: Unknown → Medium
status: Unknown → Confirmed
Brian Neltner (neltnerb) wrote :

This bug is coming up every single time I turn on my monitor. I play sound with the monitor off, and sound plays up until the moment I turn on the monitor, after which libmutter causes gnome-shell to crash. dmesg and crash log timing coincides exactly with monitor turn on time.

tags: added: bionic
tags: added: bugpattern-needed
Brian Neltner (neltnerb) wrote :

What is a bugpattern? Happy to try to create this, but my web search for the term is just showing the reports that have it rather than a definition.

Daniel van Vugt (vanvugt) wrote :
Daniel van Vugt (vanvugt) wrote :

Bumped to High.

This bug is the most common gnome-shell crash (not counting those caused by Xwayland) that is still affecting 18.04: https://errors.ubuntu.com/?package=gnome-shell&period=month

Changed in gnome-shell (Ubuntu):
importance: Medium → High
Daniel van Vugt (vanvugt) wrote :

It looks like the upstream fix(es) have landed ready for mutter 3.26.3. Although there are a couple so I'm not sure.

Changed in mutter:
status: Confirmed → In Progress
Jeremy Nickurak (nickurak) wrote :

Between gnome bugs 788834 and 788882, there are 6 patches, only 2 of which are committed. 2 are marked reviewed, 1 is marked accept-commit now, and 1 has no status.

Daniel van Vugt (vanvugt) wrote :

Zero reports of this in 3.27 so far. Fingers crossed it stays that way.

Well, my bug #1756036 just got (automatically-i-think) marked as a duplicate of this one, and I'm on 18.04 (gnome-shell 3.27.92-0ubuntu1) So that's one, Daniel. ;-) I'd been getting this on some mornings, but the bug report was getting rejected because of outdated packages until this morning. Interestingly it doesn't seem to happen again later in the day, so, as noted before, length-of-sleep may be a factor.

I reported a lot on these bugs in 17.10, until I eventually gave up and just used my (half-the-speed) mac for a while. Freshly-installed 18.04 and tried again, and I *am* having far less trouble. When it does go wrong, it "fails better", in that it seems to recover reasonably well and leave me with a working session, even if there are a couple of side-effects (eg: having both dash-to-dock and the dash in activities/applications) which clear on an orderly logout/login.



Rachel, no need to switch to Mac to avoid this problem. What I ended up
doing was this:

$ sudo add-apt-repository ppa:kubuntu-ppa/backports
$ sudo apt update && sudo apt upgrade
$ sudo apt install kubuntu-desktop

--
Daniel Wilches

On Thu, Mar 15, 2018 at 4:21 AM, Rachel Greenham <<email address hidden>
> wrote:

> Well, my bug #1756036 just got (automatically-i-think) marked as a
> duplicate of this one, and I'm on 18.04 (gnome-shell 3.27.92-0ubuntu1)
> So that's one, Daniel. ;-) I'd been getting this on some mornings, but
> the bug report was getting rejected because of outdated packages until
> this morning. Interestingly it doesn't seem to happen again later in the
> day, so, as noted before, length-of-sleep may be a factor.
>
> I reported a lot on these bugs in 17.10, until I eventually gave up and
> just used my (half-the-speed) mac for a while. Freshly-installed 18.04
> and tried again, and I *am* having far less trouble. When it does go
> wrong, it "fails better", in that it seems to recover reasonably well
> and leave me with a working session, even if there are a couple of side-
> effects (eg: having both dash-to-dock and the dash in
> activities/applications) which clear on an orderly logout/login.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1724439
>
> Title:
> gnome-shell crashed with SIGSEGV in meta_window_get_monitor() from
> ffi_call_unix64() from ffi_call() from gjs_invoke_c_function() from
> function_call()
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mutter/+bug/1724439/+subscriptions
>

not that we should let this (bug report) thread descend into KDE
advocacy but I did actually try that, last time around. I'm afraid I
couldn't get the hang of it, and then plasma went all weird on me, like
the plasma background not filling the screen kind of weird. But I did
try it. :-) Used to use KDE back in the day.

--
Rachel

On 15/03/18 13:39, dwilches wrote:
> ​
> ​
> Rachel, no need to switch to Mac to avoid this problem. What I ended up
> doing was this:
>
> $ sudo add-apt-repository ppa:kubuntu-ppa/backports
> $ sudo apt update && sudo apt upgrade
> $ sudo apt install kubuntu-desktop
>
>
> --
> Daniel Wilches
>
>
> On Thu, Mar 15, 2018 at 4:21 AM, Rachel Greenham <<email address hidden>
>> wrote:
>> Well, my bug #1756036 just got (automatically-i-think) marked as a
>> duplicate of this one, and I'm on 18.04 (gnome-shell 3.27.92-0ubuntu1)
>> So that's one, Daniel. ;-) I'd been getting this on some mornings, but
>> the bug report was getting rejected because of outdated packages until
>> this morning. Interestingly it doesn't seem to happen again later in the
>> day, so, as noted before, length-of-sleep may be a factor.
>>
>> I reported a lot on these bugs in 17.10, until I eventually gave up and
>> just used my (half-the-speed) mac for a while. Freshly-installed 18.04
>> and tried again, and I *am* having far less trouble. When it does go
>> wrong, it "fails better", in that it seems to recover reasonably well
>> and leave me with a working session, even if there are a couple of side-
>> effects (eg: having both dash-to-dock and the dash in
>> activities/applications) which clear on an orderly logout/login.
>>
>> --
>> You received this bug notification because you are subscribed to the bug
>> report.
>> https://bugs.launchpad.net/bugs/1724439
>>
>> Title:
>> gnome-shell crashed with SIGSEGV in meta_window_get_monitor() from
>> ffi_call_unix64() from ffi_call() from gjs_invoke_c_function() from
>> function_call()
>>
>> To manage notifications about this bug go to:
>> https://bugs.launchpad.net/mutter/+bug/1724439/+subscriptions
>>

Daniel van Vugt (vanvugt) wrote :

A (set of) fix(es) is in progress here:
https://bugzilla.gnome.org/show_bug.cgi?id=788834

tags: added: rls-bb-incoming
Will Cooke (willcooke) on 2018-04-03
Changed in gnome-shell (Ubuntu):
assignee: nobody → Daniel van Vugt (vanvugt)
tags: removed: rls-bb-incoming
Changed in gnome-shell (Ubuntu Bionic):
assignee: Daniel van Vugt (vanvugt) → Marco Trevisan (Treviño) (3v1n0)
Per-Inge (per-inge-hallin) wrote :

I have earlier reported a private bug #1761149 and bug #1761973
According to information it is a duplicate this bug. When I'm asked to report the bug, this bug number or my earlier reported bugs isn't among the suggested bugs.
I am used to turn of the monitor, when I'm not going to use the computer for some time. When I turn on the monitor I most of get this crash, but I have also got a problem where Ubuntu dock is replaced with the "standard Gnome dock" and I have to click Activities to see the favorite applications.
My monitor is a MSI Optix MAG27CQ which is set to 144 MHz refresh rate.

Grzegorz Daniluk (dgregd) wrote :

I just checked Ubuntu 18.04 beta 2. The bug is still present :(

Soluvious (soluvious) wrote :

Looks like they have a patch for this issue in commit https://gitlab.gnome.org/GNOME/mutter/commit/71b4ef59 however it doesn't look like it will roll until all of these items are fixed as well https://bugzilla.gnome.org/show_bug.cgi?id=788882 .

Daniel van Vugt (vanvugt) wrote :

Commit 71b4ef59 is actually commit 6dcce19932506233e0663450aa8d62a0037132c4 which is already released in mutter 3.27.91 and later. So I guess we're waiting on the other bits.

wdoekes (walter+ubuntu) wrote :

I just upgraded to Ubuntu 18.04 with libmutter 3.28.0-2 and ran into this; monitor poweroff/poweron yields a crash of gnome-shell. The good news is that the applications are not killed like they were in 17.10 (with wayland?). But waiting for the ubuntu-bug crash file to get generated isn't a good use of my time either.

For me, applying the following patch works:

  https://bug788834.bugzilla-attachments.gnome.org/attachment.cgi?id=361331

Patched binaries are here:

  https://downloads.osso.nl/libmutter-3.28.0-ubuntu-bionic/mutter_3.28.0-2osso1/

Cheers,
Walter Doekes
OSSO B.V.

Changed in mutter (Ubuntu Bionic):
status: New → In Progress
Changed in gnome-shell (Ubuntu Bionic):
status: Confirmed → In Progress
Changed in mutter (Ubuntu Bionic):
importance: Undecided → High
assignee: nobody → Marco Trevisan (Treviño) (3v1n0)
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package gnome-shell - 3.28.1-0ubuntu1

---------------
gnome-shell (3.28.1-0ubuntu1) bionic; urgency=medium

  [ Jeremy Bicha ]
  * New upstream release
  * Drop obsolete patches:
    - 27-nm-libexec-path.patch
    - fix-wayland-vbox-crash.patch

  [ Marco Trevisan (Treviño) ]
  Cherry-pick patches applied to upstream git master:
  * ui-Theme-lookup-should-respect-XDG_DATA_DIRS.patch:
    - Renamed from ubuntu_theme_lookup_xdg.patch and cherry-picked from
      upstream
  * workspaceThumbnail-only-update-_porthole-if-the-overview-.patch:
    workspaceThumbnail-rebuild-thumbnails-if-workareas-size-c.patch:
    workspaceThumbnail-initialize-porthole-based-on-workArea.patch:
    - Cherry-picks from upstream, avoid unneeded computations in activities
  * popupMenu-Fix-wrong-call-to-clutter_actor_add_child.patch:
    - Cherry-pick from upstream
  * volume-Add-back-sound-feedback-on-scroll.patch:
    - Fix regression causing missing feedback on volume slider scroll

  Add patches proposed upstream:
  * shell-ignore-invalid-window-monitor-index.patch:
    - Fix crash on accessing on invalid monitor windows (LP: #1724439)
  * StIcon-only-compute-shadow-pipeline-when-the-texture-is-p.patch:
    - Don't try to compute shadows on not allocated icons (LP: #1723378)
  * js-fix-invalid-access-errors.patch:
    - Fix javascript errors (LP: #1747566)
  * workspace-fix-repositioned-windows-in-activities.patch:
    - Ensure windows thumbnails coordinates are correct (LP: #1653153)
  * st-texture-cache-Cancel-sliced-image-loading-on-target-ac.patch:
    - Fix possible crash on cache loading
  * st-texture-cache-Don-t-add-NULL-textures-to-cache.patch:
    - Fix crash when deleting NULL textures from cache (LP: #1754445)

 -- Marco Trevisan (Treviño) <email address hidden> Tue, 17 Apr 2018 14:40:17 -0400

Changed in gnome-shell (Ubuntu Bionic):
status: In Progress → Fix Released
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package mutter - 3.28.1-1

---------------
mutter (3.28.1-1) unstable; urgency=medium

  [ Jeremy Bicha ]
  * New upstream release
    - Fix window button spacing when display is scaled (LP: #1725133)
  * Bump Standards-Version to 4.1.4

  [ Marco Trevisan (Treviño) ]
  * Add patches proposed upstream:
  * theme-frames-Use-surface-device-scale-instead-of-cairo_sc.patch:
    - theme, frames: Use surface device scale instead of cairo_scale
      (LP: #1764554)
  * theme-use-gtk_render_icon_suface-to-paint-button-icon.patch:
    - theme: use gtk_render_icon_suface to paint button icon
      (LP: #1764558)
  * theme-load-icons-as-Gtk-does-with-fallback-and-RTL-suppor.patch:
    - theme: load icons as Gtk does with fallback and RTL support
  * clutter-Smooth-out-master-clock-to-smooth-visuals.patch:
    - clutter: Smooth out master clock to smooth visuals
  * core-Return-1-if-meta_window_get_monitor-is-called-on-an-.patch:
    - core: Return -1 if meta_window_get_monitor is called on an
      unmanaged window (LP: #1724439)
  * backends-Move-MetaOutput-crtc-field-into-private-struct.patch:
    - backends: Move MetaOutput::crtc field into private struct
      (LP: #1703668)
  * backends-Add-logical-monitor-monitor-output-crtc-ref-chai.patch:
    - backends: Add logical monitor -> monitor -> output -> crtc ref
      chain (LP: #1703668)

 -- Jeremy Bicha <email address hidden> Mon, 16 Apr 2018 22:35:14 -0400

Changed in mutter (Ubuntu Bionic):
status: In Progress → Fix Released
Per-Inge (per-inge-hallin) wrote :
Download full text (4.2 KiB)

I checked and I don't get this bug any longer

2018-04-18 23:41 GMT+02:00 Launchpad Bug Tracker <<email address hidden>
>:

> This bug was fixed in the package mutter - 3.28.1-1
>
> ---------------
> mutter (3.28.1-1) unstable; urgency=medium
>
> [ Jeremy Bicha ]
> * New upstream release
> - Fix window button spacing when display is scaled (LP: #1725133)
> * Bump Standards-Version to 4.1.4
>
> [ Marco Trevisan (Treviño) ]
> * Add patches proposed upstream:
> * theme-frames-Use-surface-device-scale-instead-of-cairo_sc.patch:
> - theme, frames: Use surface device scale instead of cairo_scale
> (LP: #1764554)
> * theme-use-gtk_render_icon_suface-to-paint-button-icon.patch:
> - theme: use gtk_render_icon_suface to paint button icon
> (LP: #1764558)
> * theme-load-icons-as-Gtk-does-with-fallback-and-RTL-suppor.patch:
> - theme: load icons as Gtk does with fallback and RTL support
> * clutter-Smooth-out-master-clock-to-smooth-visuals.patch:
> - clutter: Smooth out master clock to smooth visuals
> * core-Return-1-if-meta_window_get_monitor-is-called-on-an-.patch:
> - core: Return -1 if meta_window_get_monitor is called on an
> unmanaged window (LP: #1724439)
> * backends-Move-MetaOutput-crtc-field-into-private-struct.patch:
> - backends: Move MetaOutput::crtc field into private struct
> (LP: #1703668)
> * backends-Add-logical-monitor-monitor-output-crtc-ref-chai.patch:
> - backends: Add logical monitor -> monitor -> output -> crtc ref
> chain (LP: #1703668)
>
> -- Jeremy Bicha <email address hidden> Mon, 16 Apr 2018 22:35:14 -0400
>
> ** Changed in: mutter (Ubuntu Bionic)
> Status: In Progress => Fix Released
>
> --
> You received this bug notification because you are subscribed to a
> duplicate bug report (1761149).
> https://bugs.launchpad.net/bugs/1724439
>
> Title:
> gnome-shell crashed with SIGSEGV in meta_window_get_monitor() from
> ffi_call_unix64() from ffi_call() from gjs_invoke_c_function() from
> function_call()
>
> Status in Mutter:
> In Progress
> Status in gnome-shell package in Ubuntu:
> Fix Released
> Status in mutter package in Ubuntu:
> Fix Released
> Status in gnome-shell source package in Bionic:
> Fix Released
> Status in mutter source package in Bionic:
> Fix Released
>
> Bug description:
> https://errors.ubuntu.com/problem/3e662975e4b7e6829b9d68d1c2b06f
> 429e44c854
>
> ---
>
> left virtual box to update win10 came back and got error message
>
> ProblemType: Crash
> DistroRelease: Ubuntu 17.10
> Package: gnome-shell 3.26.1-0ubuntu3
> ProcVersionSignature: Ubuntu 4.13.0-16.19-generic 4.13.4
> Uname: Linux 4.13.0-16-generic x86_64
> NonfreeKernelModules: nvidia_uvm nvidia_drm nvidia_modeset nvidia
> ApportVersion: 2.20.7-0ubuntu3
> Architecture: amd64
> CrashCounter: 1
> CurrentDesktop: ubuntu:GNOME
> Date: Tue Oct 17 23:37:32 2017
> DisplayManager: gdm3
> ExecutablePath: /usr/bin/gnome-shell
> InstallationDate: Installed on 2017-04-18 (182 days ago)
> InstallationMedia: Ubuntu-GNOME 17.04 "Zesty Zapus" - Release amd64
> (20170412)
> ProcCmdline: /usr/bin/gnome-shell
> ProcEnviro...

Read more...

Great to hear, were you getting this regularly?

Per-Inge (per-inge-hallin) wrote :

Yes every time. I changed method and used Power/Automatic suspend instead,
which actually is more convenient.

2018-04-24 23:12 GMT+02:00 Marco Trevisan (Treviño) <mail@3v1n0.net>:

> Great to hear, were you getting this regularly?
>
> --
> You received this bug notification because you are subscribed to a
> duplicate bug report (1760485).
> https://bugs.launchpad.net/bugs/1724439
>
> Title:
> gnome-shell crashed with SIGSEGV in meta_window_get_monitor() from
> ffi_call_unix64() from ffi_call() from gjs_invoke_c_function() from
> function_call()
>
> Status in Mutter:
> In Progress
> Status in gnome-shell package in Ubuntu:
> Fix Released
> Status in mutter package in Ubuntu:
> Fix Released
> Status in gnome-shell source package in Bionic:
> Fix Released
> Status in mutter source package in Bionic:
> Fix Released
>
> Bug description:
> https://errors.ubuntu.com/problem/3e662975e4b7e6829b9d68d1c2b06f
> 429e44c854
>
> ---
>
> left virtual box to update win10 came back and got error message
>
> ProblemType: Crash
> DistroRelease: Ubuntu 17.10
> Package: gnome-shell 3.26.1-0ubuntu3
> ProcVersionSignature: Ubuntu 4.13.0-16.19-generic 4.13.4
> Uname: Linux 4.13.0-16-generic x86_64
> NonfreeKernelModules: nvidia_uvm nvidia_drm nvidia_modeset nvidia
> ApportVersion: 2.20.7-0ubuntu3
> Architecture: amd64
> CrashCounter: 1
> CurrentDesktop: ubuntu:GNOME
> Date: Tue Oct 17 23:37:32 2017
> DisplayManager: gdm3
> ExecutablePath: /usr/bin/gnome-shell
> InstallationDate: Installed on 2017-04-18 (182 days ago)
> InstallationMedia: Ubuntu-GNOME 17.04 "Zesty Zapus" - Release amd64
> (20170412)
> ProcCmdline: /usr/bin/gnome-shell
> ProcEnviron:
> PATH=(custom, no user)
> XDG_RUNTIME_DIR=<set>
> LANG=en_US.UTF-8
> SHELL=/bin/bash
> SegvAnalysis:
> Segfault happened at: 0x7feb2ed42d94 <meta_window_get_monitor+4>:
> mov 0x18(%rax),%eax
> PC (0x7feb2ed42d94) ok
> source "0x18(%rax)" (0x00000018) not located in a known VMA region
> (needed readable region)!
> destination "%eax" ok
> SegvReason: reading NULL VMA
> Signal: 11
> SourcePackage: gnome-shell
> StacktraceTop:
> meta_window_get_monitor () from /usr/lib/x86_64-linux-gnu/
> libmutter-1.so.0
> ffi_call_unix64 () from /usr/lib/x86_64-linux-gnu/libffi.so.6
> ffi_call () from /usr/lib/x86_64-linux-gnu/libffi.so.6
> ?? () from /usr/lib/libgjs.so.0
> ?? () from /usr/lib/libgjs.so.0
> Title: gnome-shell crashed with SIGSEGV in meta_window_get_monitor()
> UpgradeStatus: No upgrade log present (probably fresh install)
> UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mutter/+bug/1724439/+subscriptions
>

--
Skickat från min nya mailadress

Brian Neltner (neltnerb) wrote :

After upgrading to Bionic (and no longer using Wayland), I also no longer get this bug.

Changed in mutter:
status: In Progress → Fix Released
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.