Are you using a slideshow wallpaper?

Yes, I am.

Created attachment 123232 [details]
New crash information added by DrKonqi

plasmashell (5.17.0) using Qt 5.13.1

- What I was doing when the application crashed:
The pc was open and I was not in front of it.
Going back to my pc I've found the bug tracker open.

- Unusual behavior I noticed:
Since the last update I've noticed that a folder view plasmoid continues to extend over the wallpaper even though I continue to try to reduce its height in order to avoid.

-- Backtrace (Reduced):
#6  0x00007fb3b3b0d771 in  () at /usr/lib/libQt5Core.so.5
#7  0x00007fb3b3b0f2df in  () at /usr/lib/libQt5Core.so.5
#8  0x00007fb3b3b1170a in  () at /usr/lib/libQt5Core.so.5
#9  0x00007fb3b3b11d4c in  () at /usr/lib/libQt5Core.so.5
#10 0x00007fb3b3b11f93 in QSortFilterProxyModel::invalidate() () at /usr/lib/libQt5Core.so.5

Created attachment 123233 [details]
Overlay betweer folder view and wallpaper image

This didn't happened with the previous plasma version because I was able to resize the folder view (so maybe the plasma crash did not happened with the previous version only because there wasn't any overlay)

Can you please install debug symbols for plasma-workspace and qt core so we get a better understanding of what's going on

(In reply to Kai Uwe Broulik from comment #5)
> Can you please install debug symbols for plasma-workspace and qt core so we
> get a better understanding of what's going on

I am trying.
As far as I've understood to do that in archlinux I've to compile from sources the packages with specific options and install them and then wait fr the next time the issue happens again.
The problem is that I've found the plasma-workspace package, but not the qt core (I've tried searching for qt-core).
Do you now the name of the package?

Are you using random order?
I think I could reproduce it (also with Qt 5.13.1)
#0  0x00007ffff1f49dc0 in QSortFilterProxyModelPrivate::build_source_to_proxy_mapping(QVector<int> const&, QVector<int>&) const (proxy_to_source=..., source_to_proxy=..., this=0x5555560e8dc0) at itemmodels/qsortfilterproxymodel.cpp:1184
#1  0x00007ffff1f4ba5f in QSortFilterProxyModelPrivate::create_mapping(QModelIndex const&) const (this=this@entry=0x5555560e8dc0, source_parent=...) at itemmodels/qsortfilterproxymodel.cpp:522
#2  0x00007ffff1f4cf6a in QSortFilterProxyModelPrivate::update_source_sort_column() (this=this@entry=0x5555560e8dc0) at itemmodels/qsortfilterproxymodel.cpp:637
#3  0x00007ffff1f4d351 in QSortFilterProxyModelPrivate::_q_clearMapping() (this=this@entry=0x5555560e8dc0) at itemmodels/qsortfilterproxymodel.cpp:486
#4  0x00007ffff1f4d4d9 in QSortFilterProxyModel::invalidate() (this=0x5555560e8da0) at itemmodels/qsortfilterproxymodel.cpp:2918
#5  0x00007fffc10658c8 in Image::nextSlide() (this=0x5555560e5de0) at /home/david/src/kde/kde/workspace/plasma-workspace/wallpapers/image/image.cpp:775

Yes, I am using random order.
I am still trying to recompile the packages with symbols, but being the first time that I attempt to do that at the moment I'm still facing the problem that the locally compiled package files differ from those of the repository one even before adding the debug option. Probably because I have not compiled in a chroot.

For reference the easiest way to reproduce this is to have a slideshow in random order with as few images as possible and a short timeout.
The crash happens sometimes when invalidate() is called to generate a new random order after all images have been shown once.

(In reply to David Redondo from comment #9)
> For reference the easiest way to reproduce this is to have a slideshow in
> random order with as few images as possible and a short timeout.
> The crash happens sometimes when invalidate() is called to generate a new
> random order after all images have been shown once.

Should I install the debug packages or the info I could obtain doing that is not needed anymore?

*** Bug 413068 has been marked as a duplicate of this bug. ***

*** Bug 413082 has been marked as a duplicate of this bug. ***

*** Bug 412462 has been marked as a duplicate of this bug. ***

*** Bug 413016 has been marked as a duplicate of this bug. ***

@Xwang: That is not neccessary anymore, we found the reason. Thanks a lot!

*** Bug 413099 has been marked as a duplicate of this bug. ***

We are working on a fix for this: https://phabricator.kde.org/D24723

*** Bug 413139 has been marked as a duplicate of this bug. ***

*** Bug 413140 has been marked as a duplicate of this bug. ***

*** Bug 413166 has been marked as a duplicate of this bug. ***

*** Bug 413159 has been marked as a duplicate of this bug. ***

*** Bug 413141 has been marked as a duplicate of this bug. ***

*** Bug 413059 has been marked as a duplicate of this bug. ***

Created attachment 123348 [details]
New crash information added by DrKonqi

plasmashell (5.17.0) using Qt 5.13.1

PC was running during the night, came back in the morning and found full black screen.
kstart plasmashell via yakuake gave me a running desktop back

crash is also listed in systemd log

Process 2352 (plasmashell) of user 1000 dumped core.
                                                
                                                Stack trace of thread 2352:
                                                #0  0x00007f7d2384cd90 n/a (libQt5Core.so.5)
                                                #1  0x00007f7d2385295a n/a (libQt5Core.so.5)
                                                #2  0x00007f7c7d02048a n/a (libplasma_wallpaper_imageplugin.so)
                                                #3  0x00007f7d238a9638 _ZN11QMetaObject8activateEP7QObjectiiPPv (libQt5Core.so.5)
                                                #4  0x00007f7d238a9eea _ZN7QObject5eventEP6QEvent (libQt5Core.so.5)
                                                #5  0x00007f7d24435c62 _ZN19QApplicationPrivate13notify_helperEP7QObjectP6QEvent (libQt5Widgets.so.5)
                                                #6  0x00007f7d2443f1e0 _ZN12QApplication6notifyEP7QObjectP6QEvent (libQt5Widgets.so.5)
                                                #7  0x00007f7d2387e572 _ZN16QCoreApplication15notifyInternal2EP7QObjectP6QEvent (libQt5Core.so.5)
                                                #8  0x00007f7d23881208 _ZN23QCoreApplicationPrivate16sendPostedEventsEP7QObjectiP11QThreadData (libQt5Core.so.5)
                                                #9  0x00007f7d238d5343 n/a (libQt5Core.so.5)
                                                #10 0x00007f7d21e9f6b3 g_main_context_dispatch (libglib-2.0.so.0)
                                                #11 0x00007f7d21ea1460 n/a (libglib-2.0.so.0)
                                                #12 0x00007f7d21ea149f g_main_context_iteration (libglib-2.0.so.0)
                                                #13 0x00007f7d238d4981 _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt5Core.so.5)
                                                #14 0x00007f7d2387d0db _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt5Core.so.5)
                                                #15 0x00007f7d23884d42 _ZN16QCoreApplication4execEv (libQt5Core.so.5)
                                                #16 0x000055b34c6dd206 n/a (plasmashell)
                                                #17 0x00007f7d23264e0b __libc_start_main (libc.so.6)
                                                #18 0x000055b34c6dd63a _start (plasmashell)

-- Backtrace (Reduced):
#6  0x00007f0451138d90 in  () at /usr/lib64/libQt5Core.so.5
#7  0x00007f045113e95a in  () at /usr/lib64/libQt5Core.so.5
#8  0x00007f03f9c1348a in  () at /usr/lib64/qt5/qml/org/kde/plasma/wallpapers/image/libplasma_wallpaper_imageplugin.so
[...]
#10 0x00007f0451195eea in QObject::event(QEvent*) () at /usr/lib64/libQt5Core.so.5
#11 0x00007f0451d21c62 in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /usr/lib64/libQt5Widgets.so.5

*** Bug 413219 has been marked as a duplicate of this bug. ***

I think Bug 350148 is also related to this bug. It causes the .xsession-errors to grow to over 20MB just after login.

*** Bug 413252 has been marked as a duplicate of this bug. ***

Git commit a1cf305ffb21b8ae8bbaf4d6ce03bbaa94cff405 by David Redondo.
Committed on 21/10/2019 at 08:51.
Pushed by davidre into branch 'Plasma/5.17'.

Fix slideshow crashing in invalidate()

Summary:
QSortFilterProxyModel uses std::stable_sort internally which requires that the
comparison function generates a strict weak ordering. Returning true or false
randomly didn't fullfil this requirement causing a crash in some calls to invalidate.
To keep the random order consistent a vector of row indices is used which records
the current random order.
FIXED-IN: 5.17.1

Test Plan:
To reproduce the bug use a slideshow in random order with few pictures and a small
time intervall.

Reviewers: #plasma, broulik

Reviewed By: #plasma, broulik

Subscribers: davidedmundson, broulik, plasma-devel

Tags: #plasma

Differential Revision: https://phabricator.kde.org/D24723

M  +1    -1    wallpapers/image/image.cpp
M  +55   -4    wallpapers/image/slidefiltermodel.cpp
M  +6    -1    wallpapers/image/slidefiltermodel.h

https://commits.kde.org/plasma-workspace/a1cf305ffb21b8ae8bbaf4d6ce03bbaa94cff405

*** Bug 413225 has been marked as a duplicate of this bug. ***

*** Bug 413155 has been marked as a duplicate of this bug. ***

*** Bug 413273 has been marked as a duplicate of this bug. ***

*** Bug 413193 has been marked as a duplicate of this bug. ***