Bug 350143

Summary:	Plasmashell unable to start with latest Qt 5.5 branch
Product:	[Plasma] plasmashell	Reporter:	Hrvoje Senjan <hrvoje.senjan>
Component:	general	Assignee:	David Edmundson <kde>
Status:	RESOLVED UPSTREAM
Severity:	crash	CC:	bhush94, plasma-bugs
Priority:	NOR
Version:	master
Target Milestone:	1.0
Platform:	Other
OS:	Linux
Latest Commit:		Version Fixed In:
Sentry Crash Report:

Description Hrvoje Senjan 2015-07-12 08:40:30 UTC

I'm guessing it's due to eeee9f1466ffaf17f81d635c2e43dca76ce5b021 in qtdeclarative

Reproducible: Always




backtrace:

Thread 1 (Thread 0x7f222f1d5880 (LWP 1916)):
[KCrash Handler]
#6  0x0000000000000000 in  ()
#7  0x00007f222ccea745 in QSGOpaqueTextureMaterial::setTexture(QSGTexture*) (this=this@entry=0x1c88208, texture=texture@entry=0x1c89150) at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/scenegraph/util/qsgtexturematerial.cpp:195
#8  0x00007f222cce9b7b in QSGSimpleTextureNode::setTexture(QSGTexture*) (this=this@entry=0x1c880d0, texture=0x1c89150) at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/scenegraph/util/qsgsimpletexturenode.cpp:229
#9  0x00007f222ce1f878 in QQuickCanvasItem::updatePaintNode(QSGNode*, QQuickItem::UpdatePaintNodeData*) (this=<optimized out>, oldNode=<optimized out>) at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/items/context2d/qquickcanvasitem.cpp:765
#10 0x00007f222cd3c430 in QQuickWindowPrivate::updateDirtyNode(QQuickItem*) (this=this@entry=0xe8beb0, item=item@entry=0x149f360) at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/items/qquickwindow.cpp:2984
#11 0x00007f222cd3cc78 in QQuickWindowPrivate::updateDirtyNodes() (this=this@entry=0xe8beb0) at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/items/qquickwindow.cpp:2722
#12 0x00007f222cd3cdc0 in QQuickWindowPrivate::syncSceneGraph() (this=this@entry=0xe8beb0) at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/items/qquickwindow.cpp:341
#13 0x00007f222cd0da54 in QSGGuiThreadRenderLoop::renderWindow(QQuickWindow*) (this=this@entry=0x9480d0, window=0xb06b90) at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/scenegraph/qsgrenderloop.cpp:372
#14 0x00007f222cd0ebc1 in QSGGuiThreadRenderLoop::event(QEvent*) (this=0x9480d0, e=<optimized out>) at /usr/src/debug/qtdeclarative-opensource-src-5.5.1/src/quick/scenegraph/qsgrenderloop.cpp:474
#15 0x00007f222a40d8fc in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /usr/lib64/libQt5Widgets.so.5
#16 0x00007f222a412a06 in QApplication::notify(QObject*, QEvent*) () at /usr/lib64/libQt5Widgets.so.5
#17 0x00007f2229098b33 in QCoreApplication::notifyInternal(QObject*, QEvent*) () at /usr/lib64/libQt5Core.so.5
#18 0x00007f22290eb1fd in QTimerInfoList::activateTimers() () at /usr/lib64/libQt5Core.so.5
#19 0x00007f22290eb711 in  () at /usr/lib64/libQt5Core.so.5
#20 0x00007f2224ee0cc7 in g_main_context_dispatch () at /usr/lib64/libglib-2.0.so.0
#21 0x00007f2224ee0ef8 in  () at /usr/lib64/libglib-2.0.so.0
#22 0x00007f2224ee0f9c in g_main_context_iteration () at /usr/lib64/libglib-2.0.so.0
#23 0x00007f22290ec34f in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib64/libQt5Core.so.5
#24 0x00007f222909647a in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () at /usr/lib64/libQt5Core.so.5
#25 0x00007f222909e13d in QCoreApplication::exec() () at /usr/lib64/libQt5Core.so.5
#26 0x0000000000432fc0 in main ()

Comment 1 David Edmundson 2015-07-12 09:41:41 UTC

>QQuickCanvasItem

The only applet I can think of that uses this are the calendar full view and the system load viewer.
Do you have one of these?

Could you also file a bug on Qt's bug tracker.

Comment 2 Hrvoje Senjan 2015-07-12 09:50:26 UTC

(In reply to David Edmundson from comment #1)
> >QQuickCanvasItem
> 
> The only applet I can think of that uses this are the calendar full view and
> the system load viewer.
> Do you have one of these?
Yep, i have the load viewer in panel.

> Could you also file a bug on Qt's bug tracker.
Sure. Trying first qtdeclarative 5.5 branch with mentioned commit reverted just to confirm there's the regression.

Comment 3 David Edmundson 2015-08-30 00:56:32 UTC

Could you put in a link to the upstream bug please.