from:"Bernhard Übelacker"


Am 24.01.23 um 00:30 schrieb piorunz:

On 23/01/2023 22:46, Bernhard Übelacker wrote:

https://bugs.debian.org/1028197

On Mon, 23 Jan 2023 21:48:23 + piorunz  wrote:

On 23/01/2023 13:35, Bernhard Übelacker wrote:
  journalctl --user --since="2023-01-08 12:10" --until="2023-01-08 
12:12"

-- No entries --


Then please try something like this:

   journalctl --user --since="2023-01-08 11:08" --until="2023-01-08 
11:12"


That worked! Attached.




Jan 08 11:11:06 ryzen systemd-coredump[1987371]: Process 1341349 (kded5) of 
user 1000 dumped core.
...
 #3  0x7faafb991f90 n/a 
(libc.so.6 + 0x3bf90)
 #4  0x7faa7b2e8ba4 
_ZNK10PackageKit11Transaction4roleEv (libpackagekitqt5.so.1 + 0x1aba4)
 #5  0x7faa7b3a4aae n/a 
(kded_apperd.so + 0xeaae)
 #6  0x7faa7b3a4b99 n/a 
(kded_apperd.so + 0xeb99)
 #7  0x7faafb6e8fcf n/a 
(libQt5Core.so.5 + 0x2e8fcf)
 #8  0x7faa7b2dc095 
_ZN10PackageKit6Daemon22transactionListChangedERK11QStringList 
(libpackagekitqt5.so.1 + 0xe095)
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]: kscreen.xrandr: 
XRandR::setConfig
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]: kscreen.xrandr: Requested 
screen size is QSize(7680, 2160)
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]: kscreen.xrandr: Needed 
CRTCs:  2
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]: kscreen.xrandr: Actions 
to perform:
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]:  Primary Output: 
false
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]: kscreen.xrandr: 
Change Screen Size: false
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]: kscreen.xrandr: 
Disable outputs: false
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]: kscreen.xrandr: 
Change outputs: false
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]: kscreen.xrandr: 
Enable outputs: false
Jan 08 11:11:07 ryzen kscreen_backend_launcher[5795]: kscreen.xrandr: 
XRandR::setConfig done!
Jan 08 11:11:07 ryzen plasmashell[5575]: QFont::setPointSizeF: Point size <= 0 
(0.00), must be greater than 0
Jan 08 11:11:07 ryzen plasmashell[5575]: 25 -- exe=/usr/bin/plasmashell
Jan 08 11:11:07 ryzen plasmashell[5575]: KCrash: crashing... 
crashRecursionCounter = 2


Here was also just before such a kded5 crash right before.
Either this or the startup of the new kded5 might trigger this,
as it looks like setting up some screen parameters (xrandr).
And while doing this plasmashell wrote
this "QFont::setPointSizeF" message.

Those two upstream bugs appear, but both got closed,
because upstream could not reproduce them:
  https://bugs.kde.org/show_bug.cgi?id=453712
  https://bugs.kde.org/show_bug.cgi?id=462098

I don't know which package management you use,
but until there is a fix for Debian #1026062,
you could experiment with another package manager
like synaptic and uninstall apper for now, as a workaround.

Kind regards,
Bernhard

Bug#1028197: plasmashell: Program terminated with signal SIGSEGV


https://bugs.debian.org/1028197

On Mon, 23 Jan 2023 21:48:23 + piorunz  wrote:

On 23/01/2023 13:35, Bernhard Übelacker wrote:
  journalctl --user --since="2023-01-08 12:10" --until="2023-01-08 12:12"
-- No entries --


Then please try something like this:

  journalctl --user --since="2023-01-08 11:08" --until="2023-01-08 11:12"

Bug#1028208: plasma-discover: Program terminated with signal SIGSEGV


Am 23.01.23 um 23:07 schrieb piorunz:

On 23/01/2023 21:59, Bernhard Übelacker wrote:


Hello Piotr,
if it is not clear it would be possible to take
the bug number from the email addresses and open
the bug page e.g. https://bugs.debian.org/1028208.
This was just about the plasma-discover crash part.

However, you can also iterate through past logs by
showing the different boots like this:

$ journalctl --user --list-boots
IDX BOOT ID  FIRST ENTRY LAST 
ENTRY

...
-14  Thu 2022-12-29 16:49:33 CET Mon 
2023-01-09 12:05:55 CET

...


And then show the logging for your user at that boot and
scroll to the wanted time frame:

$ journalctl --user -b -14


Kind regards,
Bernhard



Thank you Bernhard,

I've extracted logs from that time, please see attached.




Hello Piotr,
ok, then I got distracted by this from coredumpctl, showing "GMT",
while it might be really localtime ... at least it is the same as in journalctl 
...
 Timestamp: Sun 2023-01-08 13:36:08 GMT (1h 29min ago)


Your log shows four times a crashing kded5 before, then the plasma-discover:
  Jan 08 13:35:03 ryzen systemd-coredump[2835427]: Process 1997855 (kded5) of 
user 1000 dumped core.
  Jan 08 13:35:04 ryzen systemd-coredump[2835810]: Process 2835418 (kded5) of 
user 1000 dumped core.
  Jan 08 13:35:06 ryzen systemd-coredump[2836051]: Process 2835806 (kded5) of 
user 1000 dumped core.
  Jan 08 13:35:08 ryzen systemd-coredump[2836163]: Process 2836048 (kded5) of 
user 1000 dumped core.
  Jan 08 13:36:09 ryzen systemd-coredump[2842084]: Process 2832872 
(plasma-discover) of user 1000 dumped core.


And before the plasma-discover PID 2832872 shows:
  Jan 08 13:35:00 ryzen plasmashell[2832872]: adding empty sources model 
QStandardItemModel(0x55cb965ccf70)
  Jan 08 13:35:00 ryzen plasmashell[2832872]: 
file:///usr/lib/x86_64-linux-gnu/qt5/qml/QtQuick/Controls.2/org.kde.desktop/private/TextFieldContextMenu.qml:159:5:
 QML MenuItem: Binding loop detected for property "implicitWidth"
  Jan 08 13:35:00 ryzen plasmashell[2832872]: 
file:///usr/lib/x86_64-linux-gnu/qt5/qml/org/kde/kirigami.2/private/PrivateActionToolButton.qml:76:5:
 QML Binding: Binding loop detected for property "value"
  Jan 08 13:35:00 ryzen plasmashell[2832872]: packagekitqt.transaction: Unknown Transaction 
property: "Sender" QVariant(QString, ":1.577")
  Jan 08 13:35:01 ryzen plasmashell[2832872]: packagekitqt.transaction: Unknown Transaction 
property: "Sender" QVariant(QString, ":1.577")
  Jan 08 13:35:01 ryzen plasmashell[2832872]: packagekitqt.transaction: Unknown Transaction 
property: "Sender" QVariant(QString, ":1.577")
  Jan 08 13:35:03 ryzen plasmashell[2832872]: took really long to fetch 
PackageKitBackend(0x55cb964b2e50)
  Jan 08 13:35:03 ryzen plasmashell[2832872]: 29 -- exe=/usr/bin/plasma-discover
  Jan 08 13:35:03 ryzen plasmashell[2832872]: KCrash: crashing... 
crashRecursionCounter = 2


So I assume this plasma-discover crash is a result
of the crash of kded5 with PID 1997855 in the same second before:
 #2  
_ZN6KCrash19defaultCrashHandlerEi (libKF5Crash.so.5 + 0x586d)
 #3  n/a (libc.so.6 + 0x3bf90)
 #4  
_ZNK10PackageKit11Transaction4roleEv (libpackagekitqt5.so.1 + 0x1aba4)
 #5  n/a (kded_apperd.so + 
0xeaae)
 #6  n/a (kded_apperd.so + 
0xeb99)
 #7  n/a (libQt5Core.so.5 + 
0x2e8fcf)
 #8  
_ZN10PackageKit6Daemon22transactionListChangedERK11QStringList 
(libpackagekitqt5.so.1 + 0xe095)

Which looks like that from Debian bug #1026062.


The 60 seconds between the crashes might be the timeout
of KCrash to give the user the chance to click the icon
in the systray, after that time it "forwards" the crash
to get collected by systemd-coredump.


So in the end it looks like plasma-discover is not prepared
for kded5 crashing, not ideal, but if kded5 gets fixed this
plasma-discover crash might disappear too.

Kind regards,
Bernhard

Bug#1028208: plasma-discover: Program terminated with signal SIGSEGV

Am 23.01.23 um 22:48 schrieb piorunz:

On 23/01/2023 13:35, Bernhard Übelacker wrote:

(gdb) bt
#0 0x7fbffc2a9ccc in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#1 0x7fbffc25aef2 in raise () from /lib/x86_64-linux-gnu/libc.so.6
#2 0x7fbffe79dbe1 in KCrash::defaultCrashHandler(int) () from
/lib/x86_64-linux-gnu/libKF5Crash.so.5

#3
#4 0x7fbffc2a9ccc in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#5 0x7fbffc25aef2 in raise () from /lib/x86_64-linux-gnu/libc.so.6
#6
#7 0x7fbffc2a9ccc in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#8 0x7fbffc25aef2 in raise () from /lib/x86_64-linux-gnu/libc.so.6
#9 0x7fbffad3c3b0 in g_closure_invoke () from
/lib/x86_64-linux-gnu/libgobject-2.0.so.0
#10 0x7fbffad4f1a5 in ?? () from
/lib/x86_64-linux-gnu/libgobject-2.0.so.0

Hello Piotr,
this raise might indicate that libgobject wanted
to stop the process for some reason.
This reason it might have written to stdout/stderr.

Therefore probably you can still retrieve this logging.
From the "Sun 2023-01-08 13:36:08 GMT" I assume it happend at 14:36
your local time? Then following might possibly show related output.
If there is some could you provide the related parts?

journalctl --user --since="2023-01-08 14:34" --until="2023-01-08 14:38"

Unfortunately there might also be some programs writing to
~/.xsession-errors,

there the lines are not prefixed with the date and time ...
So it might be difficult to find them there.

Was this just a "one-time" crash, or do you get that regularly?

Kind regards,
Bernhard

Hello Bernhard,

Thanks for your reply. I am not sure what bug you want me to reply to so
I reply to both.

Plasma, or kwin, or entire desktop crashes all the time. There most
likely are multiple bugs. This is one of them. Sometimes desktop crashes
and there is no SIGSEV. Usually I launch Konsole from shortcut, restart
kwin and plasmashell with --replace option, but this not always helps.

I have modern Radeon GPU, it probably adds fuel to the fire.
Can't wait for more updates regarding kwin/plasmashell/multi monitor
handling/sleeping and waking up the monitor. There is a lot of bugs in
these areas, when Radeon GPU is involved at least. I will be happy to
debug as much as I can, you can tell me what else to do.

$ journalctl --user --since="2023-01-08 14:34" --until="2023-01-08 14:38"
-- No entries --

journalctl --user --since="2023-01-08 12:10" --until="2023-01-08 12:12"
-- No entries --

.xsession-errors:

Xsession: X session started for pioruns at Sat 7 Jan 08:04:38 GMT 2023
dbus-update-activation-environment: setting
DBUS_SESSION_BUS_ADDRESS=unix:path=/run/user/1000/bus

dbus-update-activation-environment: setting DISPLAY=:0
dbus-update-activation-environment: setting
XAUTHORITY=/home/pioruns/.Xauthority

localuser:pioruns being added to access control list
dbus-update-activation-environment: setting GTK_MODULES=gail:atk-bridge
dbus-update-activation-environment: setting QT_ACCESSIBILITY=1
file:///home/pioruns/.local/share/plasma/look-and-feel/com.github.vinceliuice.Qogir-light/contents/splash/Splash.qml:89:
ReferenceError: bottomRect is not defined
file:///home/pioruns/.local/share/plasma/look-and-feel/com.github.vinceliuice.Qogir-light/contents/splash/Splash.qml:88:
ReferenceError: bottomRect is not defined
file:///home/pioruns/.local/share/plasma/look-and-feel/com.github.vinceliuice.Qogir-light/contents/splash/Splash.qml:89:
ReferenceError: bottomRect is not defined
file:///home/pioruns/.local/share/plasma/look-and-feel/com.github.vinceliuice.Qogir-light/contents/splash/Splash.qml:89:
ReferenceError: bottomRect is not defined
file:///home/pioruns/.local/share/plasma/look-and-feel/com.github.vinceliuice.Qogir-light/contents/splash/Splash.qml:88:
ReferenceError: bottomRect is not defined
file:///home/pioruns/.local/share/plasma/look-and-feel/com.github.vinceliuice.Qogir-light/contents/splash/Splash.qml:89:
ReferenceError: bottomRect is not defined
Xsession: X session started for pioruns at Mon 9 Jan 10:52:01 GMT 2023

This seems to be a session between 7th and 9th January.

Hello Piotr,
if it is not clear it would be possible to take
the bug number from the email addresses and open
the bug page e.g. https://bugs.debian.org/1028208.
This was just about the plasma-discover crash part.

However, you can also iterate through past logs by
showing the different boots like this:

$ journalctl --user --list-boots
IDX BOOT ID FIRST ENTRY LAST ENTRY
...
-14 Thu 2022-12-29 16:49:33 CET Mon 2023-01-09
12:05:55 CET
...

And then show the logging for your user at that boot and
scroll to the wanted time frame:

$ journalctl --user -b -14

Kind regards,
Bernhard

Bug#1028197: plasmashell: Program terminated with signal SIGSEGV


(gdb) bt
...
#4  
#5  0x7f54b82563fb in QQuickItem::~QQuickItem() () from 
/lib/x86_64-linux-gnu/libQt5Quick.so.5
#6  0x7f54b83d0045 in ?? () from /lib/x86_64-linux-gnu/libQt5Quick.so.5
#7  0x7f54b66dd53f in QObject::event(QEvent*) () from 
/lib/x86_64-linux-gnu/libQt5Core.so.5
#8  0x7f54b7362f5e in QApplicationPrivate::notify_helper(QObject*, QEvent*) 
() from /lib/x86_64-linux-gnu/libQt5Widgets.so.5
#9  0x7f54b66b17c8 in QCoreApplication::notifyInternal2(QObject*, QEvent*) 
() from /lib/x86_64-linux-gnu/libQt5Core.so.5
#10 0x7f54b66b4761 in QCoreApplicationPrivate::sendPostedEvents(QObject*, 
int, QThreadData*) () from /lib/x86_64-linux-gnu/libQt5Core.so.5
#11 0x7f54b670a1d3 in ?? () from /lib/x86_64-linux-gnu/libQt5Core.so.5
#12 0x7f54b46657a9 in g_main_context_dispatch () from 
/lib/x86_64-linux-gnu/libglib-2.0.so.0
#13 0x7f54b4665a38 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#14 0x7f54b4665acc in g_main_context_iteration () from 
/lib/x86_64-linux-gnu/libglib-2.0.so.0
#15 0x7f54b67098b6 in 
QEventDispatcherGlib::processEvents(QFlags) () 
from /lib/x86_64-linux-gnu/libQt5Core.so.5
#16 0x7f54b66b024b in 
QEventLoop::exec(QFlags) () from 
/lib/x86_64-linux-gnu/libQt5Core.so.5
#17 0x7f54b66b83b6 in QCoreApplication::exec() () from 
/lib/x86_64-linux-gnu/libQt5Core.so.5
#18 0x55937ecd0c6c in ?? ()
#19 0x7f54b624618a in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#20 0x7f54b6246245 in __libc_start_main () from 
/lib/x86_64-linux-gnu/libc.so.6


Hello Piotr,
I fear that backtrace is not that distinctive.
Upstream I found [1] a few bugs, but they also do
not give helpful pointers.

So here it might also be helpful to see what might be contained
in journalctl or in .xsession-errors:

  journalctl --user --since="2023-01-08 12:10" --until="2023-01-08 12:12"

Can you please give a rough estimate how often it does crash?

Kind regards,
Bernhard

[1]
  https://bugs.kde.org/show_bug.cgi?id=461464
  https://bugs.kde.org/show_bug.cgi?id=460477

Bug#1028208: plasma-discover: Program terminated with signal SIGSEGV


(gdb) bt
#0  0x7fbffc2a9ccc in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#1  0x7fbffc25aef2 in raise () from /lib/x86_64-linux-gnu/libc.so.6
#2  0x7fbffe79dbe1 in KCrash::defaultCrashHandler(int) () from 
/lib/x86_64-linux-gnu/libKF5Crash.so.5
#3  
#4  0x7fbffc2a9ccc in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#5  0x7fbffc25aef2 in raise () from /lib/x86_64-linux-gnu/libc.so.6
#6  
#7  0x7fbffc2a9ccc in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#8  0x7fbffc25aef2 in raise () from /lib/x86_64-linux-gnu/libc.so.6
#9  0x7fbffad3c3b0 in g_closure_invoke () from 
/lib/x86_64-linux-gnu/libgobject-2.0.so.0
#10 0x7fbffad4f1a5 in ?? () from /lib/x86_64-linux-gnu/libgobject-2.0.so.0



Hello Piotr,
this raise might indicate that libgobject wanted
to stop the process for some reason.
This reason it might have written to stdout/stderr.

Therefore probably you can still retrieve this logging.
From the "Sun 2023-01-08 13:36:08 GMT" I assume it happend at 14:36
your local time? Then following might possibly show related output.
If there is some could you provide the related parts?

journalctl --user --since="2023-01-08 14:34" --until="2023-01-08 14:38"

Unfortunately there might also be some programs writing to ~/.xsession-errors,
there the lines are not prefixed with the date and time ...
So it might be difficult to find them there.

Was this just a "one-time" crash, or do you get that regularly?

Kind regards,
Bernhard

Bug#1028083: plasma-discover: Discover tray app crashes kded5, causing other system tray icons to vanish


Am 22.01.23 um 14:19 schrieb Gregor Riepl:
FYI: My kded core dumps also contained something about a closed DBus 
connection in a different thread's stack trace. Perhaps this is related?


I am not able to exclude it to be related, maybe it is the start of the 
row of events. When debugging through it I got the impression that this 
Transaction object with RoleUnknown gets "marked for deletion" kind of 
unconditionally. But I am not much experienced with this source code.


Kind regards,
Bernhard

Bug#1028083: plasma-discover: Discover tray app crashes kded5, causing other system tray icons to vanish

2023-01-21 Thread Bernhard Übelacker




On Fri, 06 Jan 2023 16:50:02 +0100 Gregor Riepl  wrote:
...

#14 0x7f9c8c26f6ed in QObject::connect<...> (...) at 
/usr/include/x86_64-linux-gnu/qt5/QtCore/qobject.h:268
#15 TransactionJob::TransactionJob (...) at ./apperd/TransactionJob.cpp:47
#16 0x7f9c8c271648 in TransactionWatcher::transactionChanged (...) at 
./apperd/TransactionWatcher.cpp:211
#17 0x7f9c8c271ade in TransactionWatcher::watchTransaction (...) at 
./apperd/TransactionWatcher.cpp:106
#18 0x7f9c8c271b99 in TransactionWatcher::transactionListChanged (...) at 
./apperd/TransactionWatcher.cpp:85

...


Hello,
this seems similar to the backtrace in bug #1026062.
At least the "transactionListChanged" and the two lines above.

Kind regards,
Bernhard


https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1026062

Bug#1027326: kde-plasma-desktop: Process 1465 (kded5) of user 1000 dumped core.

2023-01-20 Thread Bernhard Übelacker


On Fri, 30 Dec 2022 09:46:49 -0300 craudio  wrote:


#8  0x7f2ba43e5b99 n/a (kded_apperd.so + 0xeb99)
#9  0x7f2becae8fcf n/a (libQt5Core.so.5 + 0x2e8fcf)
#10 0x7f2ba4323095 
_ZN10PackageKit6Daemon22transactionListChangedERK11QStringList 
(libpackagekitqt5.so.1 + 0xe095)



Hello,
this seems similar to the backtrace in bug #1026062.
At least the "transactionListChanged" and the two lines above.

Kind regards,
Bernhard


https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1026062

Bug#1026062: kded5: kded crashes with signal 11

2023-01-04 Thread Bernhard Übelacker


Short addition:
I tried to bring this to the attention of packagekit-qt
developers in this bug report:
  https://github.com/PackageKit/PackageKit-Qt/issues/42

Kind regards,
Bernhard

Bug#1026062: kded5: kded crashes with signal 11

2023-01-03 Thread Bernhard Übelacker


Dear Maintainer,
I was able to reproduce this issue inside a minimal
amd64 qemu VM running Bookworm/testing.

By editing the kded service unit [1] I could get valgrind have a look
at this issue and it produced a matching use-after-free [2].

By further editing the service unit I was also able to record
such a crash with rr-debugger, that allows debugging in
reverse execution direction.


So I guess these are the relevant events:

- A object of PackageKit::Transaction gets created [3] (0x562777928e00)

- Inside its constructor also a QDeferredDeleteEvent
  gets created [4] (0x562777885690)

- The PackageKit::Transaction object gets stored in the
  m_transactions container [5]

- Qt reaches its event loop and processes the QDeferredDeleteEvent
  and deletes the PackageKit::Transaction [6]

- In TransactionWatcher::watchTransaction the member m_transactions
  still holds a reference to the already deleted object and gets
  found because a matching "tid" gets processed.
  "Unfortunately" the memory of the PackageKit::Transaction was
  already modified therefore the segmentation fault follows. [7]


For comparision the matching systemd-coredump
report with mangled symbols in [8].

There are some reports in bugs.kde.org [9].
But in 462706 it is mentioned that this is an issue
in packagekit-qt, unfortunately could not yet
find a report there.


Kind regards,
Bernhard






[1]
/usr/lib/systemd/user/plasma-kded.service
-ExecStart=/usr/bin/kded5
+ExecStart=/usr/bin/valgrind /usr/bin/kded5
or
+ExecStart=/usr/bin/rr record /usr/bin/kded5

systemctl --user daemon-reload
systemctl --user start plasma-kded

Then trigger in Discover a package list update.




[2]
==12351== Invalid read of size 8
==12351==at 0x27458BA0: PackageKit::Transaction::role() const 
(transaction.cpp:297)
==12351==by 0x273B6AAD: TransactionWatcher::watchTransaction(QDBusObjectPath 
const&, bool) (TransactionWatcher.cpp:104)
==12351==by 0x273B6B98: TransactionWatcher::transactionListChanged(QStringList 
const&) (TransactionWatcher.cpp:85)
==12351==by 0x5B59FCE: ??? (in 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by 0x2744C094: PackageKit::Daemon::transactionListChanged(QStringList 
const&) (moc_daemon.cpp:419)
==12351==by 0x5B59FFB: ??? (in 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by 0x27464B37: TransactionListChanged (daemonproxy.moc:331)
==12351==by 0x27464B37: 
OrgFreedesktopPackageKitInterface::qt_static_metacall(QObject*, 
QMetaObject::Call, int, void**) (daemonproxy.moc:178)
==12351==by 0x27465D72: 
OrgFreedesktopPackageKitInterface::qt_metacall(QMetaObject::Call, int, void**) 
(daemonproxy.moc:288)
==12351==by 0x580361A: ??? (in 
/usr/lib/x86_64-linux-gnu/libQt5DBus.so.5.15.7)
==12351==by 0x5B4E76F: QObject::event(QEvent*) (in 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by 0x49CAF5D: QApplicationPrivate::notify_helper(QObject*, 
QEvent*) (in /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5.15.7)
==12351==by 0x5B227C7: QCoreApplication::notifyInternal2(QObject*, QEvent*) 
(in /usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==  Address 0x1a6a2a10 is 16 bytes inside a block of size 24 free'd
==12351==at 0x484371B: operator delete(void*) (vg_replace_malloc.c:923)
==12351==by 0x5B4E53E: QObject::event(QEvent*) (in 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by 0x49CAF5D: QApplicationPrivate::notify_helper(QObject*, 
QEvent*) (in /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5.15.7)
==12351==by 0x5B227C7: QCoreApplication::notifyInternal2(QObject*, QEvent*) 
(in /usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by 0x5B25760: QCoreApplicationPrivate::sendPostedEvents(QObject*, 
int, QThreadData*) (in /usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by 0x5B7B1D2: ??? (in 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by 0x6E1B7A8: g_main_context_dispatch (in 
/usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.7400.4)
==12351==by 0x6E1BA37: ??? (in 
/usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.7400.4)
==12351==by 0x6E1BACB: g_main_context_iteration (in 
/usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.7400.4)
==12351==by 0x5B7A8B5: 
QEventDispatcherGlib::processEvents(QFlags) (in 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by 0x5B2124A: 
QEventLoop::exec(QFlags) (in 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by 0x5B293B5: QCoreApplication::exec() (in 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==  Block was alloc'd at
==12351==at 0x4840F2F: operator new(unsigned long) (vg_replace_malloc.c:422)
==12351==by 0x273B693B: TransactionWatcher::watchTransaction(QDBusObjectPath 
const&, bool) (TransactionWatcher.cpp:95)
==12351==by 0x273B6B98: TransactionWatcher::transactionListChanged(QStringList 
const&) (TransactionWatcher.cpp:85)
==12351==by 0x5B59FCE: ??? (in 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.15.7)
==12351==by

Bug#1026426: plasma-workspace: No default wallpaper with black desktop, and ksplashqml hides desktop until hard time limit of 30 seconds is reached

2023-01-03 Thread Bernhard Übelacker


Dear Maintainer,
I retested this inside a minimal Bookworm/testing VM now because
the last plasma package updates seem to have settled a bit.
Unfortunately the black background
and the 30 seconds splash are still visible.


At least in this bug report at least the black
default wallpaper gets described:

   https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1026070#17



After an affected new user configures a wallpaper the
previously reported message from main.qml is gone,
but the 30 second splash is also still there.


To make the 30 second splash issue better visible I created
this upstream merge request:

   https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/2450/diffs

This one applied to the Debian package showed that
the last step "desktop" is missing.

See this for all needed steps:
   
https://invent.kde.org/plasma/plasma-workspace/-/blob/master/ksplash/ksplashqml/splashapp.cpp#L36


Kind regards,
Bernhard

Bug#1025997: libqt5core5a: Qt applications sometimes crashing at display configuration changes or power savings

2022-12-24 Thread Bernhard Übelacker


On Tue, 13 Dec 2022 01:28:39 +0100 =?UTF-8?Q?Bernhard_=c3=9cbelacker?= 
 wrote:


Dear Maintainer,
I experienced since about a month ago sometimes crashes in applications
running in fullscreen, when doing display configuration changes
or lately when waking up the screen from power saving.
(See below for an example backtrace.)

I opened these upstream bug reports, where the
Qt bug received a patch hopefully fixing this issue:

   https://bugs.kde.org/show_bug.cgi?id=461723
   https://bugreports.qt.io/browse/QTBUG-109226
   
https://code.qt.io/cgit/qt/qtbase.git/commit/?id=6a3627b6c5aa5109a80024f3d7b0f938504f7ffe
   (Unfortunately it looks like the qt-5.15 cerry-pick is not publicly visible.)

I create this bug also to ask how chances are that this
commit reaches Qt before or during the freeze period?

Kind regards,
Bernhard


I forgot to mention, this might also be related to a dual monitor setup.
With "power savings" is meant just the monitors turning off.

Short after opening this bug the Qt "Gerrit Bot" added the information
the fixed upstream version seems to be 5.15.12.

And reading up release plans [1] I think 5.15.12 is way out of reach,
therefore the next option might be picking the patch into the
packaged by Debian version 5.15.7, depending if there arrive other
reports about this issue.

[1] https://bugs.debian.org/1025715

Kind regards,
Bernhard

Bug#1026426: plasma-workspace: No default wallpaper with black desktop, and ksplashqml hides desktop until hard time limit of 30 seconds is reached

2022-12-19 Thread Bernhard Übelacker


Package: plasma-workspace
Version: 4:5.26.4.1-1
Severity: normal
X-Debbugs-Cc: bernha...@mailbox.org

Dear Maintainer,

I noticed since a week or so a change in behaviour of a plasma desktop.
If i install in a minimal VM the package task-kde-desktop after logging
into the desktop no wallpaper is shown - the desktop stays black.


Also the splash screen hides the desktop longer than usual.
This I think is related to following message:

Dez 19 23:56:38 debian plasmashell[1367]: 
file:///usr/share/plasma/wallpapers/org.kde.image/contents/ui/main.qml:151: 
TypeError: Cannot call method 'createObject' of undefined (exception occurred 
during delayed function evaluation)

Therefore the last stage in ksplashqml is never reached which would
release the sight to the desktop.

Upstream master had several changes, [1] at least avoids above message,
but still experienced the 30 seconds ksplashqml time.


I thought I would create this report because of the nearing freeze,
and I am not sure if there would be another upstream 5.26 release and it
gets packaged in time, and if such a release would contain fixes
from upstream master.

Kind regards,
Bernhard

[1] 
https://invent.kde.org/plasma/plasma-workspace/-/commit/125d39423fb9a24027ef165404fabc00e78cc961





-- System Information:
Debian Release: bookworm/sid
  APT prefers testing-debug
  APT policy: (500, 'testing-debug'), (500, 'testing')
Architecture: amd64 (x86_64)

Kernel: Linux 6.0.0-6-amd64 (SMP w/16 CPU threads; PREEMPT)
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages plasma-workspace depends on:
ii  dbus-user-session [default-dbus-session-bus]1.14.4-1
ii  drkonqi 5.26.4-1
ii  frameworkintegration5.100.0-1
ii  gdb 12.1-4
ii  init-system-helpers 1.65.2
ii  iso-codes   4.12.0-1
ii  kactivitymanagerd   5.26.4-1
ii  kded5   5.100.0-1
ii  kinit   5.100.0-1
ii  kio 5.100.0-2
ii  kpackagetool5   5.100.0-1
ii  kwin-common 4:5.26.4-1
ii  libappstreamqt2 0.15.5-1
ii  libc6   2.36-6
ii  libcolorcorrect54:5.26.4.1-1
ii  libcrypt1   1:4.4.33-1
ii  libfontconfig1  2.13.1-4.5
ii  libfreetype62.12.1+dfsg-3
ii  libgcc-s1   12.2.0-10
ii  libgps283.22-4.1+b1
ii  libice6 2:1.0.10-1
ii  libicu7272.1-3
ii  libkf5activities5   5.100.0-1
ii  libkf5activitiesstats1  5.100.0-1
ii  libkf5archive5  5.100.0-2
ii  libkf5authcore5 5.100.0-1
ii  libkf5baloo55.100.0-1
ii  libkf5bookmarks55.100.0-1
ii  libkf5calendarevents5   5.100.0-1
ii  libkf5completion5   5.100.0-1
ii  libkf5config-bin5.100.1-1
ii  libkf5configcore5   5.100.1-1
ii  libkf5configgui55.100.1-1
ii  libkf5configwidgets55.100.0-1
ii  libkf5coreaddons5   5.100.0-1
ii  libkf5crash55.100.0-1
ii  libkf5dbusaddons5   5.100.0-1
ii  libkf5declarative5  5.100.0-1
ii  libkf5globalaccel-bin   5.100.0-1
ii  libkf5globalaccel5  5.100.0-1
ii  libkf5guiaddons55.100.0-3
ii  libkf5holidays5 1:5.100.0-1
ii  libkf5i18n5 5.100.0-1
ii  libkf5iconthemes5   5.100.0-1
ii  libkf5idletime5 5.100.0-1
ii  libkf5jobwidgets5   5.100.0-1
ii  libkf5kcmutils5 5.100.0-1
ii  libkf5kexiv2-15.0.0 22.12.0-2
ii  libkf5kiocore5  5.100.0-2
ii  libkf5kiofilewidgets5

Bug#1025104: krita: segfault on exit

2022-12-14 Thread Bernhard Übelacker


Control: forwarded -1 https://bugs.kde.org/show_bug.cgi?id=463052


Dear Maintainer,
I tried to have a look if I can find some more information,
and opened an issue upstream here:

  https://bugs.kde.org/show_bug.cgi?id=463052

It looks like a double free because a pointer is hold in two lists.
And it can be reproduced quite easily in i386 and amd64 by really
just open and close Krita.

Kind regards,
Bernhard

Bug#1019401: QSGRenderThread[569483]: segfault after suspend/resume

2022-12-12 Thread Bernhard Übelacker


Hello,
that will still not be enough for further investigation,
but the given byte sequence in conjunction the package from
snapshot.d.o including the dbgsym package points to this location:

0x...cfb in QThreadStorageData::get() const at thread/qthreadstorage.cpp:122


@piorunz: a first step might be to install the package systemd-coredump.
That way a whole lot of information for a crash should appear in the journal.
This link might contain more hints: https://wiki.debian.org/HowToGetABacktrace


Kind regards,
Bernhard


(gdb) list qthreadstorage.cpp:114,125
114 void **QThreadStorageData::get() const
...
121 QVector  = data->tls;
122 if (tls.size() <= id)
123 tls.resize(id + 1);

Bug#1025997: libqt5core5a: Qt applications sometimes crashing at display configuration changes or power savings

2022-12-12 Thread Bernhard Übelacker


Package: libqt5core5a
Version: 5.15.6+dfsg-2
Severity: wishlist
X-Debbugs-Cc: bernha...@mailbox.org


Dear Maintainer,
I experienced since about a month ago sometimes crashes in applications
running in fullscreen, when doing display configuration changes
or lately when waking up the screen from power saving.
(See below for an example backtrace.)

I opened these upstream bug reports, where the
Qt bug received a patch hopefully fixing this issue:

  https://bugs.kde.org/show_bug.cgi?id=461723
  https://bugreports.qt.io/browse/QTBUG-109226
  
https://code.qt.io/cgit/qt/qtbase.git/commit/?id=6a3627b6c5aa5109a80024f3d7b0f938504f7ffe
  (Unfortunately it looks like the qt-5.15 cerry-pick is not publicly visible.)

I create this bug also to ask how chances are that this
commit reaches Qt before or during the freeze period?

Kind regards,
Bernhard


(gdb) bt
#3  
#4  __memcpy_avx_unaligned () at 
../sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S:228
#5  0x7f00962f142a in memmove (__len=262112, __src=0x7efb8bbe8810, 
__dest=) at 
/usr/include/x86_64-linux-gnu/bits/string_fortified.h:36
#6  copy_unswapped (rect=..., img=..., dstBytesPerLine=262112, 
dst=) at ./src/plugins/platforms/xcb/qxcbbackingstore.cpp:547
#7  native_sub_image (swap=false, rect=..., src=..., 
dstStride=262112, buffer=0x5557a9af9130) at 
./src/plugins/platforms/xcb/qxcbbackingstore.cpp:590
#8  QXcbBackingStoreImage::flushPixmap (this=0x5557a9af90b0, region=..., 
fullRegion=) at 
./src/plugins/platforms/xcb/qxcbbackingstore.cpp:669
#9  0x7f00962f1a29 in QXcbBackingStoreImage::flushPixmap (fullRegion=false, 
region=..., this=0x5557a9af90b0) at 
./src/plugins/platforms/xcb/qxcbbackingstore.cpp:627
#10 QXcbBackingStoreImage::put (this=0x5557a9af90b0, dst=85983245, region=..., 
offset=...) at ./src/plugins/platforms/xcb/qxcbbackingstore.cpp:741
#11 0x7f00962f2369 in QXcbBackingStore::flush 
(this=this@entry=0x5557a9a3b4d0, window=window@entry=0x5557a98d7c10, 
region=..., offset=...) at ./src/plugins/platforms/xcb/qxcbwindow.h:128
#12 0x7f009cd017b2 in QBackingStore::flush (this=this@entry=0x5557a9a4f510, 
region=..., window=0x5557a98d7c10, offset=...) at painting/qbackingstore.cpp:252
#13 0x7f009d37059f in QWidgetRepaintManager::flush 
(this=this@entry=0x5557a9d8fad0, widget=0x5557a98df320, region=..., 
widgetTextures=) at kernel/qwidgetrepaintmanager.cpp:1184
#14 0x7f009d372129 in QWidgetRepaintManager::flush (this=0x5557a9d8fad0) at 
kernel/qwidgetrepaintmanager.cpp:1082
#15 0x7f009d374270 in QWidgetRepaintManager::paintAndFlush 
(this=0x5557a9d8fad0) at kernel/qwidgetrepaintmanager.cpp:1014
#16 0x7f009d3bd341 in QWidgetWindow::handleResizeEvent 
(this=0x5557a98d7c10, event=0x7ffc26bba560) at kernel/qwidgetwindow.cpp:841
#17 0x7f009d3c10db in QWidgetWindow::event (this=0x5557a98d7c10, 
event=0x7ffc26bba560) at kernel/qwidgetwindow.cpp:322
#18 0x7f009d362f5e in QApplicationPrivate::notify_helper (this=, receiver=0x5557a98d7c10, e=0x7ffc26bba560) at kernel/qapplication.cpp:3637
#19 0x7f009c6b1718 in QCoreApplication::notifyInternal2(QObject*, QEvent*) 
() from /lib/x86_64-linux-gnu/libQt5Core.so.5
#20 0x7f009cb39bac in QGuiApplicationPrivate::processGeometryChangeEvent 
(e=) at kernel/qguiapplication.cpp:2610
...


-- System Information:
Debian Release: bookworm/sid
  APT prefers testing-debug
  APT policy: (500, 'testing-debug'), (500, 'testing')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Bug#1022106: plasma-workspace: plasma panel sometimes hangs

2022-11-13 Thread Bernhard Übelacker






plasma panel sometimes hangs, becomes irresponsive. this is preceded by
some notification appearing in the notification area, that becomes stuck
in the corner of the windows and cannot be dismissed. at the same time
all the panel becomes irresponsive to mouse clicks.
my only workaround is killing plasmashell.



Hello Fulvio,
have you possibly kdeconnect installed and those notification, as far as
they are still readable, are related to a connected Android phone?

Then this upstream issue might be related:
  https://bugs.kde.org/show_bug.cgi?id=442645


I ask because I have kdeconnect installed and see all some
days such a hang you are describing.
(But I don't get a call like in above upstream bug, it is just notifications.)


I am able to "free" my session by this sequence from SSH or virtual terminal 
session:
  killall plasmashell
  killall kwin_x11
  export DISPLAY=:0
  plasmashell --replace &
  kwin_x11 --replace &

(But unfortunately that environment is not exactly the
same so a restart of the session might be still a good idea,
or repeat from a still running konsole window with a good environment ...)


I generated a core file via gdb last time for both processes,
but when I inspected it now I did not see an obvious interesting thread...


Kind regards,
Bernhard

Bug#1019446: kde-config-cron: Binary missing

2022-09-24 Thread Bernhard Übelacker


On Fri, 09 Sep 2022 11:25:50 + Christian Buhtz  wrote:


installed "kde-config-cron" without having a full KDE desktop environment
enabled. I am using XFCE here but would like to have cron gui front-end.

After installation I can not find a binary for "kde-config-cron".



It seems to me that this isn't a stand alone package but something like an
addon or plugin.



Or is there a way for me to start that application?



Hello Christian,
I guess the kcm_cron.so is really just like a module
loaded by systemsettings5.

So maybe you can start it with the supplied desktop file,
or by "systemsettings5 kcm_cron".


# cat /usr/share/kservices5/kcm_cron.desktop
[Desktop Entry]
Exec=systemsettings5 kcm_cron


Kind regards,
Bernhard

Bug#996726: libkdecorations2-5v5 freeze KDE on login, general protection fault

2021-10-20 Thread Bernhard Übelacker


Hello Bob,

On Wed, 20 Oct 2021 15:33:47 +0800 Bob Wong  wrote:

Hello, I am unable to downgrade the kdecoration, if I try to downgrade
kdecoration, the apt will remove the whole kde desktop. So the only choice
for me is to upgrade. When will the new version of the kdecoration be
available in bookworm? As for now I can only use other desktop environment.
It's really inconvenient.


Unfortunately Patrick's link got line wrapped.

His link should read:
  https://snapshot.debian.org/package/kdecoration/4%3A5.21.5-2/

And the amd64 package would be:
  
https://snapshot.debian.org/archive/debian/20210817T204926Z/pool/main/k/kdecoration/libkdecorations2-5v5_5.21.5-2_amd64.deb

If you download this and install it by following command,
it should just downgrade this single package:
  dpkg -i libkdecorations2-5v5_5.21.5-2_amd64.deb

Kind regards,
Bernhard

Bug#996860: kded5: crash during init

2021-10-19 Thread Bernhard Übelacker


Control: merge -1 996859


Hello Achim,
I hope it is ok to try to merge those two bugs #996859 and #996860,
BTS takes some time to confirm creating a bug ...

Have you updated lately libkdecorations2-5v5 to 4:5.23.0-2 ?
Then you probably hit this bug #996726,
where a workaround is described by downgrading
to the package version before.

If you see a different issue, maybe you
could install systemd-coredump and provide the
output of "journalctl -e --no-pager" related to the crash.

Kind regards,
Bernhard


https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=996726

Bug#996761: kwin-x11: KWin crashes and doesn't start

2021-10-18 Thread Bernhard Übelacker


Hello Francisco, hello Albert,
this might be the same as in bug #996726
and not directly related to the nvidia driver.

I hit a crash in kwin_x11 with an AMD graphics card
and could workaround by installing
libkdecorations2-5v5_5.21.5-2_amd64.deb like
mentioned in #996726.

Kind regards,
Bernhard

https://bugs.debian.org/996726

Bug#996744: kde-config-gtk-style: Version mismatch segfaults kded5 repeatedly

2021-10-18 Thread Bernhard Übelacker


Hello Patrick,
I guess this is the same as described in bug #996726.

And I was able to workaround by installing
libkdecorations2-5v5_5.21.5-2_amd64.deb.

Kind regards,
Bernhard


Core was generated by `/usr/bin/kwin_x11 --replace'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x75d1f510 in KDecoration2::DecorationSettings::font 
(this=this@entry=0x556f0880) at /usr/include/c++/11/bits/unique_ptr.h:173
173 /usr/include/c++/11/bits/unique_ptr.h: Datei oder Verzeichnis nicht 
gefunden.
[Current thread is 1 (Thread 0x7047b9c0 (LWP 7026))]
(gdb) set width 0
(gdb) set pagination off
(gdb) bt
#0  0x75d1f510 in KDecoration2::DecorationSettings::font 
(this=this@entry=0x556f0880) at /usr/include/c++/11/bits/unique_ptr.h:173
#1  0x75d1f737 in operator() (__closure=) at 
./src/decorationsettings.cpp:19
#2  KDecoration2::DecorationSettings::DecorationSettings (this=0x556f0880, 
bridge=0x5573e230, parent=) at 
./src/decorationsettings.cpp:34
#3  0x77d68add in 
QSharedPointer::create
 () at /usr/include/c++/10/new:175
#4  KWin::Decoration::DecorationBridge::init (this=0x5573e230) at 
./decorations/decorationbridge.cpp:121
#5  KWin::Decoration::DecorationBridge::init (this=0x5573e230) at 
./decorations/decorationbridge.cpp:110
#6  0x77e6d5a7 in KWin::Workspace::Workspace (this=0x5574b450) at 
./workspace.cpp:187
#7  0x77dc6990 in KWin::Application::createWorkspace 
(this=this@entry=0x7fffe3a0) at ./main.cpp:272
#8  0xd41e in KWin::ApplicationX11::continueStartupWithScreens 
(this=0x7fffe3a0) at ./main_x11.cpp:272
...

Bug#987886: krita: Right-clicking on the canvas crashes krita

2021-05-18 Thread Bernhard Übelacker


Hello Torbjørn Birch Moltu,
I tried to reproduce this issue inside a virtual machine.
But there the menu opens without the issue.

Does this happen to you if you startup without the wacom input attached?

Next thing you could try is to startup krita this way:
  export MALLOC_CHECK_=3
  krita --new-image RGBA,U8,1600,1600

Does the KCrash window open or do you have the
sad smiley at the bottom right near the clock?

Kind regards,
Bernhard

Bug#985994: kwin-x11: crashes randomly on ALT-TAB for switching between windows

2021-04-10 Thread Bernhard Übelacker


Hello Russel,
thank you for the detailed backtraces.
Unfortunately they point into Qt's javascript module,
finally trying to copy memory to a null pointer.

I have not found a hint to similar crashes in
upstream bug tracker.

Could you please provide which theme you are using,
or if there are maybe some third party extensions in use?

Maybe you are aware of some not widely used settings
that you changed, which might be related?

Kind regards,
Bernhard

Bug#985994: kwin-x11: crashes randomly on ALT-TAB for switching between windows

2021-04-09 Thread Bernhard Übelacker


Hello Russel,
thanks for the fast answer, unfortunately the
backtrace is not yet enough expressive.

Maybe you could also install the following debug symbol packages?

libqt5qml5-dbgsym libqt5core5a-dbgsym libqt5widgets5-dbgsym libqt5gui5-dbgsym 
kwin-common-dbgsym kwin-x11-dbgsym

These are located in a separate debug symbol repository,
which has to be enabled first and is described here:
  https://wiki.debian.org/HowToGetABacktrace#Installing_the_debugging_symbols

Kind regards,
Bernhard




[KCrash Handler]
#4  0x7f42259bee08 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Qml.so.5
#5  0x7f4225acd3cc in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Qml.so.5
#6  0x001e0009 in ?? ()
#7  0x55623023d9d0 in ?? ()
#8  0x7f421a2d in ?? ()
#9  0x556230c214a0 in ?? ()

Bug#985197: kwin-wayland: Drag and drop a file in Dolphin makes KDE Plasma Wayland crash

2021-04-09 Thread Bernhard Übelacker


Hello Silvério,



Reading symbols from /usr/bin/kwin_wayland...
(No debugging symbols found in /usr/bin/kwin_wayland)
BFD: warning: /tmp/user/1000/coredump-ySV5m6 is truncated: expected core file
size >= 2365169664, found: 2147483648




it looks like for some reason the kwin_wayland exceeds
the address space limit of systemd-coredump, therefore
the results are not good.


If you want to give it another try, then please uncomment
and raise this limit from e.g. 2GB to 3GB like this:

/etc/systemd/coredump.conf
...
ProcessSizeMax=3G
ExternalSizeMax=3G
...

systemctl daemon-reload


Then the size warning hopefully goes away on the next
attempt, and a better backtrace might be printed.

Kind regards,
Bernhard

Bug#985994: kwin-x11: crashes randomly on ALT-TAB for switching between windows

2021-04-09 Thread Bernhard Übelacker


Hello Russell,
could you still see this issue?


Stack trace of thread 120123:

#0  0x7f49c3823ce1 n/a (/lib/x86_64-linux-gnu/libc-2.31.so 
(deleted) + 0x3bce1)



Were there more lines following in the "Stack trace of thread 120123"?

And as libc-2.31 is shown as deleted, I assume this
happened while libc package got updated while the kwin-x11
process was still using the previous version.

Kind regards,
Bernhard

Bug#985197: kwin-wayland: Drag and drop a file in Dolphin makes KDE Plasma Wayland crash

2021-03-31 Thread Bernhard Übelacker


Hello Silvério,



Am 31.03.21 um 20:42 schrieb Silvério Santos:

Hallo Bernhard,

Here is it:

Mär 31 20:11:25 systemname kernel: kwin_wayland[1834]: segfault at e8 ip 
7fe70ba08268 sp 7ffc9bbf64d0 error 4 in 
libKWaylandServer.so.5.20.5[7fe70b9b+b6000]
Mär 31 20:11:25 systemname kernel: Code: 00 00 48 8b 47 10 48 8b 40 48 c3 90 66 0f 1f 
44 00 00 41 57 41 56 41 55 41 54 41 89 f4 55 48 89 fd 53 48 83 ec 18 4c 8b 7f 10 
<49> 8b 87 c8 00 00 00 8b 48 0c 8b 58 08 89 4c 24 0c 89 5c 24 08 39
Mär 31 20:11:25 systemname systemd[1]: Created slice 
system-systemd\x2dcoredump.slice.
Mär 31 20:11:25 systemname systemd[1]: Started Process Core Dump (PID 3494/UID 
0).



Mär 31 20:12:04 systemname systemd-coredump[3495]: Process 1834 (kwin_wayland) 
of user 1000 dumped core.
   
   Stack trace of thread 1834:

   #0  0x7fe70ba08268 n/a 
(n/a + 0x0)
Mär 31 20:12:05 systemname systemd[1]:systemd-coredump@0-3494-0.service: 
Succeeded.




thanks for the quick response, but unfortunately this
is less informative as I expected it to be.

If you have the package gdb installed could you please
try following commands and forward the output:

coredumpctl gdb 1834
set pagination off
bt

(With 1834 or the last pid shown after kwin_wayland
from the latest crash.)

Kind regards,
Bernhard

Bug#985197: kwin-wayland: Drag and drop a file in Dolphin makes KDE Plasma Wayland crash

2021-03-29 Thread Bernhard Übelacker


Hello Santos,
I tried to reproduce this issue but did not see it inside a minimal VM.
Therefore tried to some more information from the Code bytes printed in
you journal/dmesg.

This should point to this instruction:
   0x7f6e16bd5268 
<_ZN14KWaylandServer16SurfaceInterface13frameRenderedEj+24>:  49 8b 87 c8 
00 00 00mov0xc8(%r15),%rax
A breakpoint set to this instruction points to this source line:
   0x...268 in QListData::isEmpty() const at 
/usr/include/x86_64-linux-gnu/qt5/QtCore/qlist.h:116

The unmangled function name should read like this:
   KWaylandServer::SurfaceInterface::frameRendered(unsigned int)


Unfortunately this seems not exact enough to take further actions.
Therefore I would suggest installing, if possible, the package
systemd-coredump. That way in the journal a simple backtrace should
be printed which might give an idea which functions are involved.
(As there may crash multiple processes as the wayland server dies,
make sure to get the lines after this line:
"Process ... (kwin_wayland) of user ... dumped core"


On the other side a search in upstream bugtracker returned [1].
I could immediately reproduce that crash by dragging an icon
from the kickoff menu onto a opened konsole window, which crashed
in the exact same instruction, pointed to by the code bytes from
the bug submission. Therefore I guess chances are not too bad
this bug is related.

Kind regards,
Bernhard

[1] https://bugs.kde.org/show_bug.cgi?id=428399

# single-use Bullseye/testing amd64 qemu VM 2021-03-29

echo "set enable-bracketed-paste off" >> /etc/inputrc; bash

apt update

# to speedup testing
mv /etc/manpath.config /etc/manpath.config.renamed
apt install libeatmydata1
export LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libeatmydata.so

apt dist-upgrade
apt install systemd-coredump mc gdb rr valgrind sddm plasma-workspace-wayland 
dolphin konsole lilypond-doc-html \
libkwaylandserver5-dbgsym
apt build-dep libkwaylandserver5




mkdir /home/benutzer/source/libkwaylandserver5/orig -p
cd/home/benutzer/source/libkwaylandserver5/orig
apt source libkwaylandserver5
cd






cp -a /usr/share/doc/lilypond/html/Documentation/58/lily-ab77784d.ly .

# could not reproduce by drag and drop ...




https://wiki.debian.org/InterpretingKernelOutputAtProcessCrash

# By bug submitter:
kernel: kwin_wayland[12033]: segfault at e8 ip 7fbbe21d5268 sp 
7ffe4b5000b0 error 4 in libKWaylandServer.so.5.20.5[7fbbe217d000+b6000]
kernel: Code: 00 00 48 8b 47 10 48 8b 40 48 c3 90 66 0f 1f 44 00 00 41 57 41 56 
41 55 41 54 41 89 f4 55 48 89 fd 53 48 83 ec 18 4c 8b 7f 10 <49> 8b 87 c8 00 00 
00 8b 48 0c 8b 58 08 89 4c 24 0c 89 5c 24 08 39

error 4 == 0b100
0: no page found
0: read access
1: user-mode access

echo -n "find /b ..., ..., 0x" && \
echo "00 00 48 8b 47 10 48 8b 40 48 c3 90 66 0f 1f 44 00 00 41 57 41 56 41 55 
41 54 41 89 f4 55 48 89 fd 53 48 83 ec 18 4c 8b 7f 10 <49> 8b 87 c8 00 00 00 8b 
48 0c 8b 58 08 89 4c 24 0c 89 5c 24 08 39" \
 | sed 's/[<>]//g' | sed 's/ /, 0x/g'

find /b ..., ..., 0x00, 0x00, 0x48, 0x8b, 0x47, 0x10, 0x48, 0x8b, 0x40, 0x48, 
0xc3, 0x90, 0x66, 0x0f, 0x1f, 0x44, 0x00, 0x00, 0x41, 0x57, 0x41, 0x56, 0x41, 
0x55, 0x41, 0x54, 0x41, 0x89, 0xf4, 0x55, 0x48, 0x89, 0xfd, 0x53, 0x48, 0x83, 
0xec, 0x18, 0x4c, 0x8b, 0x7f, 0x10, 0x49, 0x8b, 0x87, 0xc8, 0x00, 0x00, 0x00, 
0x8b, 0x48, 0x0c, 0x8b, 0x58, 0x08, 0x89, 0x4c, 0x24, 0x0c, 0x89, 0x5c, 0x24, 
0x08, 0x39

 
 
 
 
 
# dpkg -S libKWaylandServer.so.5.20.5
libkwaylandserver5:amd64: /usr/lib/x86_64-linux-gnu/libKWaylandServer.so.5.20.5

$ gdb -q --pid $(pidof kwin_wayland)
...
(gdb) set width 0
(gdb) set pagination off
(gdb) info share
FromTo  Syms Read   Shared Object Library
0x7f6e16b80ff0  0x7f6e16c32a66  Yes 
/lib/x86_64-linux-gnu/libKWaylandServer.so.5
...
(gdb) find /b 0x7f6e16b80ff0, 0x7f6e16c32a66, 0x00, 0x00, 0x48, 0x8b, 
0x47, 0x10, 0x48, 0x8b, 0x40, 0x48, 0xc3, 0x90, 0x66, 0x0f, 0x1f, 0x44, 0x00, 
0x00, 0x41, 0x57, 0x41, 0x56, 0x41, 0x55, 0x41, 0x54, 0x41, 0x89, 0xf4, 0x55, 
0x48, 0x89, 0xfd, 0x53, 0x48, 0x83, 0xec, 0x18, 0x4c, 0x8b, 0x7f, 0x10, 0x49, 
0x8b, 0x87, 0xc8, 0x00, 0x00, 0x00, 0x8b, 0x48, 0x0c, 0x8b, 0x58, 0x08, 0x89, 
0x4c, 0x24, 0x0c, 0x89, 0x5c, 0x24, 0x08, 0x39
0x7f6e16bd523e
1 pattern found.
(gdb) b * (0x7f6e16bd523e + 42)
Breakpoint 1 at 0x7f6e16bd5268: file 
/usr/include/x86_64-linux-gnu/qt5/QtCore/qlist.h, line 116.
(gdb) info b
Num Type   Disp Enb AddressWhat
1   breakpoint keep y   0x7f6e16bd5268 in QListData::isEmpty() 
const at /usr/include/x86_64-linux-gnu/qt5/QtCore/qlist.h:116
(gdb) disassemble /r 0x7f6e16bd523e, 0x7f6e16bd523e + 62
Dump of assembler code from 0x7f6e16bd523e to 0x7f6e16bd527c:
   0x7f6e16bd523e:  00 00   add%al,(%rax)
   0x7f6e16bd5240 <_ZNK14KWaylandServer16SurfaceInterface10compositorEv+0>: 
48 8b 47 10 mov0x10(%rdi),%rax
   0x7f6e16bd5244

Bug#979343: sddm: general protection fault in libQt5Qml.so.5.15.2

2021-03-24 Thread Bernhard Übelacker


Hello everyone,



I added it, and now I got one:
Tue 2021-03-23 20:20:40 CET2000   109   115  11 present 
/usr/bin/sddm-greeter

If I extract it, I get:
 Executable: /usr/bin/sddm-greeter

...

 #9  0x7fe7b41f5def __clone (libc.so.6 + 0xfddef)


With this "coredumpctl gdb 2000", and when you have gdb installed,
you should get a prompt "(gdb) ".
There a command "bt" should get a better backtrace than the automatic one.




You can get the core file, if you like, at
https://www.helgefjell.de/data/sddm.core


I tried to have a look at this one in the hope I have the same
package versions installed as you, and have received a backtrace
showing we are inside the __run_exit_handlers.
This might explain why you get no issue with it except the logging,
because this process has already done its main work
and is about to end itself.

What I further see is some object destruction going on
with mentioning QV4 - which I believe is tightly related to
Qts javascript engine.

And finally it is in a method QMetaType::destruct, which is
unfortunately about to call a function pointer m_destructor
that consists of some string data.

Getting a traps instead of a segfault might be because of
the function pointer using more than the lower 48 bits, to
which address space is currently limited?
At least a short test with the value 0x0070006d006f0063
leads to such a traps message, using 0x006d006f0063
shows a "segfault at" message in dmesg.

But having this string at this position might just be coincidence,
a few debugging details might be found in attached file.

Kind regards,
Bernhard


Core was generated by `/usr/bin/sddm-greeter --socket /tmp/sddm-:0-aSeIQL 
--theme /usr/share/sddm/them'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  QMetaType::destruct (data=0x563464af9d00, this=0x5634649ea3b8) at 
/usr/include/x86_64-linux-gnu/qt5/QtCore/qmetatype.h:2375
2375m_destructor(data);
[Current thread is 1 (Thread 0x7fe7b49fb840 (LWP 2000))]
(gdb) bt
#0  QMetaType::destruct (data=0x563464af9d00, this=0x5634649ea3b8) at 
/usr/include/x86_64-linux-gnu/qt5/QtCore/qmetatype.h:2375
#1  QV4::Heap::QQmlValueTypeWrapper::destroy (this=0x7fe79833d460) at 
qml/qqmlvaluetypewrapper.cpp:100
#2  0x7fe7b52fa55f in QV4::Chunk::sweep (this=0x7fe79833, 
engine=0x56346475ffe0) at memory/qv4mm.cpp:349
#3  0x7fe7b52fa7f3 in operator() (c=, __closure=) at memory/qv4mm.cpp:630
#4  std::__partition<__gnu_cxx::__normal_iterator 
>, QV4::BlockAllocator::sweep():: > (__pred=..., __last=0x2, 
__first=0x7fe79833) at /usr/include/c++/10/bits/stl_algo.h:1515
#5  std::partition<__gnu_cxx::__normal_iterator 
>, QV4::BlockAllocator::sweep():: > (__pred=..., __last=..., 
__first=...) at /usr/include/c++/10/bits/stl_algo.h:4673
#6  QV4::BlockAllocator::sweep (this=this@entry=0x56346442fa60) at 
memory/qv4mm.cpp:631
#7  0x7fe7b52fb415 in QV4::MemoryManager::sweep 
(this=this@entry=0x56346442fa50, lastSweep=lastSweep@entry=false, 
classCountPtr=classCountPtr@entry=0x0) at memory/qv4mm.cpp:994
#8  0x7fe7b52fbf2d in QV4::MemoryManager::runGC (this=0x56346442fa50) at 
memory/qv4mm.cpp:1054
#9  0x7fe7b52fddb5 in QV4::MemoryManager::allocate (size=32, 
allocator=0x56346442fa60, this=0x56346442fa50) at 
../../include/QtQml/5.15.2/QtQml/private/../../../../../src/qml/memory/qv4mm_p.h:307
#10 QV4::MemoryManager::allocString (this=this@entry=0x56346442fa50, 
unmanagedSize=) at memory/qv4mm.cpp:791
#11 0x7fe7b536418e in QV4::MemoryManager::allocWithStringData (arg1=..., unmanagedSize=, this=0x56346442fa50) at 
../../include/QtQml/5.15.2/QtQml/private/../../../../../src/qml/memory/qv4mm_p.h:217
#12 QV4::ExecutionEngine::newString (this=this@entry=0x56346475ffe0, s=...) at 
jsruntime/qv4engine.cpp:894
#13 0x7fe7b539f688 in QV4::ErrorPrototype::method_toString (b=, thisObject=0x7fe7986b9508) at jsruntime/qv4errorobject.cpp:352
#14 0x7fe7b541706f in QV4::FunctionObject::call (argc=0, argv=0x0, 
thisObject=0x7fe7986b9508, this=0x7fe7986b9530) at 
jsruntime/qv4functionobject_p.h:172
#15 QV4::RuntimeHelpers::ordinaryToPrimitive 
(engine=engine@entry=0x56346475ffe0, object=object@entry=0x7fe7986b9508, 
typeHint=typeHint@entry=0x7fe7986b9310) at jsruntime/qv4runtime.cpp:517
#16 0x7fe7b5417394 in QV4::RuntimeHelpers::objectDefaultValue 
(object=0x7fe7986b9508, object@entry=0x7fe7986b9518, typeHint=typeHint@entry=2) 
at jsruntime/qv4runtime.cpp:495
#17 0x7fe7b541bd75 in QV4::RuntimeHelpers::toPrimitive 
(typeHint=QV4::STRING_HINT, value=...) at jsruntime/qv4runtime_p.h:123
#18 QV4::Value::toQStringNoThrow (this=this@entry=0x7fe7986b9508) at 
jsruntime/qv4value.cpp:150
#19 0x7fe7b536d5de in QV4::ExecutionEngine::catchExceptionAsQmlError 
(this=this@entry=0x56346475ffe0) at 
../../include/QtQml/5.15.2/QtQml/private/../../../../../src/qml/jsruntime/qv4scopedvalue_p.h:234
#20 0x7fe7b5518412 in QQmlDelayedError::catchJavaScriptException 
(engine=0x56346475ffe0, this=0x5634647b3860)

Bug#979343: sddm: general protection fault in libQt5Qml.so.5.15.2

2021-03-20 Thread Bernhard Übelacker


Am 20.03.21 um 21:44 schrieb Helge Kreutzmann:

If I should install a -dbg version or something else please
inform me.


Thank you for the additional information.
There might really be something more. If you have not, is it possible
to install the package "systemd-coredump".

If then a crash happens again and it gets recorded it should
be mentioned in journalctl and this command should list them:
coredumpctl list
And a core should be stored. (But maybe just for the current boot)

There should be already a slight backtrace in journalctl
that might be helpful, but with 'coredumpctl gdb' and at the gdb
prompt with 'bt' it might reveal some more information.
Even better when sddm-dbgsym is installed, and if sufficiently RAM is 
available, libqt5qml5-dbgsym. (and some more not yet known in between ...)


There might be more pointer in:
https://wiki.debian.org/HowToGetABacktrace

Kind regards,
Bernhard

Bug#979343: sddm: general protection fault in libQt5Qml.so.5.15.2

2021-03-17 Thread Bernhard Übelacker


Hello Helge,
do you still see these messages in your logging?
If yes could you maybe add the surrounding logging
of one such fault? E.g. by something like this:
    journalctl | grep "traps:" -C20

Kind regards,
Bernhard

Bug#979343: sddm: general protection fault in libQt5Qml.so.5.15.2

2021-03-17 Thread Bernhard Übelacker


Hello Helge,
do you still see these messages in your logging?
If yes could you maybe add the surrounding logging
of one such fault? E.g. by something like this:
   journalctl | grep "traps:" -C20

Kind regards,
Bernhard

Bug#985194: konsole: blue bar on the left side when text is scrolling

2021-03-14 Thread Bernhard Übelacker


Hello Sandro, hello Norbert,
this sounds like the feature "Highlight lines coming into view" [1].

There is an option that looks like it could be disabled in:
Profiles - Scrollbars - "Highlight the lines coming into view"

Kind regards,
Bernhard

https://invent.kde.org/utilities/konsole/-/commit/dabac1f47ead4626f0ff83affb851309f10a6e58

Bug#983031: konqueror segfaults on starting

2021-03-06 Thread Bernhard Übelacker


Hello Joe,
this information might not be sufficient for the maintainer.

Could you try following to collect some more informations?

- install packages:
gdb psmisc konqueror-dbgsym

- stop all running instances of konqueror:
killall konqueror

- then start konqueror inside a debugger:
gdb -q -ex 'set pagination off' -ex 'run' -ex 'bt' -ex 'detach' -ex 'quit' 
--args konqueror


When the crash happens, a backtrace should be printed.
Please forward this output to this bug.

More debugging hints can be found here [1].

Kind regards,
Bernhard

[1] https://wiki.debian.org/HowToGetABacktrace

Bug#966174: okular: none

2020-10-17 Thread Bernhard Übelacker

Dear Maintainer,
I could reproduce it with the state of Debian testing at 2020-07-24.

Upstream bugs seem to be these:
  https://bugs.kde.org/show_bug.cgi?id=407338
  https://gitlab.freedesktop.org/poppler/poppler/-/issues/766

The issue seems to be resolved since poppler 0.77 and above.

Kind regards,
Bernhard


(gdb) bt
#0  0x7f298027ad75 in SECMOD_ReferenceModule (module=0x0) at pk11util.c:874
#1  0x7f298027b2fc in SECMOD_AddModule (newModule=0x5568b72ec210) at 
pk11util.c:568
#2  SECMOD_AddModule (newModule=0x5568b72ec210) at pk11util.c:546
#3  0x7f298027b3a0 in SECMOD_AddNewModuleEx (moduleName=, 
dllPath=0x7f29805e2e8f "libnssckbi.so", defaultMechanismFlags=0, 
cipherEnableFlags=0, modparms=0x0, nssparms=) at pk11util.c:722
#4  0x7f2980597529 in SignatureHandler::SignatureHandler 
(this=0x7ffe3141cfb0, p7=0x5568b7595f70 
"0\202)\206\006\t*\206H\206\367\r\001\a\002\240\202)w0\202)s\002\001\001\061\v0\t\006\005+\016\003\002\032\005",
 p7_length=13815) at ./poppler/SignatureHandler.cc:136
#5  0x7f298048cf2c in FormFieldSignature::validateSignature 
(forceRevalidation=, validationTime=4294967295, 
doVerifyCert=true, this=0x5568b7580420) at ./poppler/Form.cc:1722
#6  FormFieldSignature::validateSignature (this=0x5568b7580420, 
doVerifyCert=, forceRevalidation=, 
validationTime=4294967295) at ./poppler/Form.cc:1689
#7  0x7f29806a8384 in Poppler::FormFieldSignature::validate 
(this=this@entry=0x5568b7580320, opt=opt@entry=1, validationTime=...) at 
./qt5/src/poppler-form.cc:681
#8  0x7f29806a88e0 in Poppler::FormFieldSignature::validate 
(this=0x5568b7580320, 
opt=opt@entry=Poppler::FormFieldSignature::ValidateVerifyCertificate) at 
./qt5/src/poppler-form.cc:674
#9  0x7f298070180b in PopplerFormFieldSignature::PopplerFormFieldSignature 
(this=0x5568b75807d0, field=std::unique_ptr 
= {...}) at /usr/include/c++/9/bits/unique_ptr.h:360
...


# Bullseye/testing amd64 qemu VM 2020-07-24


approx:
debian-11-bullseye-snapshot.debian.org  
https://snapshot.debian.org/archive/debian/20200724T00Z/
debian-11-bullseye-debug-snapshot.debian.org
https://snapshot.debian.org/archive/debian-debug/20200724T00Z/


sources.list:
deb [check-valid-until=no] 
http://192.168.178.25:/debian-11-bullseye-snapshot.debian.org/ bullseye main
deb-src [check-valid-until=no] 
http://192.168.178.25:/debian-11-bullseye-snapshot.debian.org/ bullseye main
deb [check-valid-until=no] 
http://192.168.178.25:/debian-11-bullseye-debug-snapshot.debian.org/ 
bullseye-debug main



apt update
apt dist-upgrade


apt install systemd-coredump mc htop psmisc net-tools strace lightdm 
xserver-xorg openbox xterm gdb okular okular-dbgsym libkf5parts5-dbgsym 
libokular5core9-dbgsym libpoppler-qt5-1-dbgsym libpoppler82-dbgsym 
libnss3-dbgsym


wget https://www.boe.es/boe/dias/2018/04/26/pdfs/BOE-A-2018-5704.pdf


export DISPLAY=:0
okular BOE-A-2018-5704.pdf



benutzer@debian:~$ okular BOE-A-2018-5704.pdf
Icon theme "breeze" not found.
Icon theme "breeze" not found.
Speicherzugriffsfehler (Speicherabzug geschrieben)


root@debian:~# journalctl -e
Okt 17 11:31:01 debian kernel: okular[776]: segfault at 38 ip 7f298027ad75 
sp 7ffe3141ced0 error 4 in libnss3.so[7f2980233000+f2000]
Okt 17 11:31:01 debian kernel: Code: 84 fb ff 48 85 c0 74 0f 48 c7 00 00 00 00 
00 48 c7 40 08 00 00 00 00 48 83 c4 08 c3 66 0f 1f 84 00 00 00 00 00 41 54 49 
89 fc <48> 8b 7f 38 e8 12 9b fb ff 41 83 44 24 40 01 49 8b 7c 24 38 e8 b2
Okt 17 11:31:01 debian systemd[1]: Created slice 
system-systemd\x2dcoredump.slice.
Okt 17 11:31:01 debian systemd[1]: Started Process Core Dump (PID 811/UID 0).
Okt 17 11:31:02 debian systemd-coredump[812]: Process 776 (okular) of user 1000 
dumped core.
  
  Stack trace of thread 776:
  #0  0x7f298027ad75 
SECMOD_ReferenceModule (libnss3.so + 0x60d75)
  #1  0x7f298027b2fc n/a 
(libnss3.so + 0x612fc)
  #2  0x7f298027b3a0 
SECMOD_AddNewModuleEx (libnss3.so + 0x613a0)
  #3  0x7f2980597529 
_ZN16SignatureHandlerC2EPhi (libpoppler.so.82 + 0x228529)
  #4  0x7f298048cf2c 
_ZN18FormFieldSignature17validateSignatureEbbl (libpoppler.so.82 + 0x11df2c)
  #5  0x7f29806a8384 
_ZNK7Poppler18FormFieldSignature8validateEiRK9QDateTime (libpoppler-qt5.so.1 + 
0x48384)
  #6  0x7f29806a88e0 
_ZNK7Poppler18FormFieldSignature8validateENS0_15ValidateOptionsE 
(libpoppler-qt5.so.1 + 0x488e0)
  #7  0x7f298070180b n/a 
(okularGenerator_poppler.so + 0x2580b)
  #8  0x7f29806f5f94

Bug#969456: kmix 20.08.0 crashes on exit

2020-09-03 Thread Bernhard Übelacker

Package: kmix
Version: 4:20.08.0-1
Severity: normal
Forwarded: https://bugs.kde.org/show_bug.cgi?id=425469
Tags: upstream fixed-upstream


Dear Maintainer,
I found several kmix cores lately and investigated a bit.

It crashes when attempting to do some cleanup on process exit.
There unfortunately the QtPaMainLoop structure got deleted,
but later still accessed.

Upstream has already a patch in git.

Kind regards,
Bernhard



==3727== Invalid read of size 8
==3727==at 0x7FBA456: pa_srbchannel_free (srbchannel.c:364)
==3727==by 0x7FB7032: check_srbpending.part.0 (pstream.c:724)
==3727==by 0x7FB9319: pa_pstream_unlink (pstream.c:1190)
==3727==by 0x7FB9319: pa_pstream_unlink (pstream.c:1181)
==3727==by 0x6E1FCB4: context_unlink (context.c:223)
==3727==by 0x6E1FCB4: context_unlink (context.c:201)
==3727==by 0x6E1FE01: context_free (context.c:244)
==3727==by 0x4A60F87: Mixer_PULSE::~Mixer_PULSE() (mixer_pulse.cpp:1073)
==3727==by 0x4A60FA8: Mixer_PULSE::~Mixer_PULSE() (mixer_pulse.cpp:1080)
==3727==by 0x4A3E0C4: Mixer::~Mixer() (mixer.cpp:115)
==3727==by 0x4A3E0D8: Mixer::~Mixer() (mixer.cpp:116)
==3727==by 0x4A37C36: MixerToolBox::deinitMixer() (mixertoolbox.cpp:356)
==3727==by 0x12D3F7: KMixWindow::~KMixWindow() (kmix.cpp:139)
==3727==by 0x12D4D8: KMixWindow::~KMixWindow() (kmix.cpp:151)
==3727==  Address 0x10704168 is 88 bytes inside a block of size 112 free'd
==3727==at 0x4839EAB: operator delete(void*) (vg_replace_malloc.c:584)
==3727==by 0x4A60F40: operator() (unique_ptr.h:85)
==3727==by 0x4A60F40: ~unique_ptr (unique_ptr.h:361)
==3727==by 0x4A60F40: Mixer_PULSE::~Mixer_PULSE() (mixer_pulse.cpp:1056)
==3727==by 0x4A60FA8: Mixer_PULSE::~Mixer_PULSE() (mixer_pulse.cpp:1080)
==3727==by 0x4A3E0C4: Mixer::~Mixer() (mixer.cpp:115)
==3727==by 0x4A3E0D8: Mixer::~Mixer() (mixer.cpp:116)
==3727==by 0x4A37C36: MixerToolBox::deinitMixer() (mixertoolbox.cpp:356)
==3727==by 0x12D3F7: KMixWindow::~KMixWindow() (kmix.cpp:139)
==3727==by 0x12D4D8: KMixWindow::~KMixWindow() (kmix.cpp:151)
==3727==by 0x136533: KMixApp::~KMixApp() (KMixApp.cpp:58)
==3727==by 0x12A527: main (main.cpp:73)
==3727==  Block was alloc'd at
==3727==at 0x4838DEF: operator new(unsigned long) (vg_replace_malloc.c:342)
==3727==by 0x4A5E538: Mixer_PULSE::connectToDaemon() (mixer_pulse.cpp:961)
==3727==by 0x4A61387: Mixer_PULSE::Mixer_PULSE(Mixer*, int) 
(mixer_pulse.cpp:1037)
==3727==by 0x4A61745: PULSE_getMixer(Mixer*, int) (mixer_pulse.cpp:947)
==3727==by 0x4A3E9F7: Mixer::Mixer(QString const&, int) (mixer.cpp:102)
==3727==by 0x4A381FE: 
MixerToolBox::initMixerInternal(MixerToolBox::MultiDriverMode, QStringList 
const&, bool) (mixertoolbox.cpp:142)
==3727==by 0x4A39602: initMixer (mixertoolbox.cpp:273)
==3727==by 0x4A39602: MixerToolBox::initMixer(bool, QStringList const&, 
bool) (mixertoolbox.cpp:284)
==3727==by 0x133724: KMixWindow::KMixWindow(bool, bool) (kmix.cpp:91)
==3727==by 0x1363AD: KMixApp::createWindowOnce(bool, bool) [clone .part.0] 
(KMixApp.cpp:69)
==3727==by 0x136B41: createWindowOnce (KMixApp.cpp:125)
==3727==by 0x136B41: KMixApp::restoreSessionIfApplicable(bool, bool) 
(KMixApp.cpp:125)
==3727==by 0x136CDE: KMixApp::newInstance(QStringList const&, QString 
const&) (KMixApp.cpp:166)
==3727==by 0x12A4F1: main (main.cpp:84)



-- System Information:
Debian Release: bullseye/sid
  APT prefers testing-debug
  APT policy: (500, 'testing-debug'), (500, 'proposed-updates-debug'), (500, 
'testing')
Architecture: amd64 (x86_64)

Kernel: Linux 5.7.0-3-amd64 (SMP w/4 CPU threads)
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages kmix depends on:
ii  libasound2 1.2.3.2-1
ii  libc6  2.31-3
ii  libcanberra0   0.30-7
ii  libkf5configcore5  5.70.0-1
ii  libkf5configgui5   5.70.0-1
ii  libkf5configwidgets5   5.70.0-2
ii  libkf5coreaddons5  5.70.0-2
ii  libkf5dbusaddons5  5.70.0-1
ii  libkf5globalaccel-bin  5.70.0-1
ii  libkf5globalaccel5 5.70.0-1
ii  libkf5i18n55.70.0-1
ii  libkf5notifications5   5.70.0-1
ii  libkf5plasma5  5.70.1-1
ii  libkf5solid5   5.70.0-1
ii  libkf5widgetsaddons5   5.70.0-1
ii  libkf5windowsystem55.70.0-1
ii  libkf5xmlgui5  5.70.0-1+b1
ii  libpulse0  13.0-5
ii  libqt5core5a   5.14.2+dfsg-6
ii  libqt5dbus55.14.2+dfsg-6
ii  libqt5gui5 5.14.2+dfsg-6
ii  libqt5widgets5 5.14.2+dfsg-6
ii  libqt5xml5 5.14.2+dfsg-6
ii  libstdc++6 10.2.0-5

kmix recommends no packages.

kmix suggests no packages.

-- no debconf information

Bug#949710: /usr/bin/plasmashell: segfaults sporadically but only when using app dropdown menu

2020-02-07 Thread Bernhard Übelacker

Hello Sarah,
I am sorry, but the given information might not be enough
for the maintainers to solve the crash.

Usually when a segfault happens in 'dmesg' output should
appear two lines about this event. Maybe you could forward
these to the report too.

If it is possible to install systemd-coredump a backtrace
would be automatically added to the journal, which could
be viewed by 'journalctl --no-pager'.

'coredumpctl gdb' could also show these information and if
the package gdb is also installed a '(gdb)' prompt should
appear, where the command 'thread apply all bt full'
should provide detailed information where the crash happened.

This information would be much better if the package
plasma-workspace-dbgsym and maybe some for shared libraries
get installed from the repository described here [1].
(If output gets too much pages, maybe attach as file.)


Kind regards
Bernhard

[1] https://wiki.debian.org/HowToGetABacktrace#Installing_the_debugging_symbols

Bug#950537: lxterminal: wrong default PATH for superuser

2020-02-03 Thread Bernhard Übelacker

Hello Jan,
please leave the bugs email in the recipients or cc list to
have all information attached to the bug e.g. by using "reply all".

Nice to hear it works.
You can close a bug by just changing the email recipient from
'950...@bugs.debian.org' to '950537-d...@bugs.debian.org'.

Kind regards,
Bernhard



Am 03.02.20 um 15:55 schrieb Schweik Josef:
> Well, you are quite correct, "su -" is working fine!
> ...
> 
> Thanks. Jan
> 
> Please, how can I close this bug report?
> 
>

Bug#950537: lxterminal: wrong default PATH for superuser

2020-02-03 Thread Bernhard Übelacker

Hello Jan,
your issue might originate not from the terminal but from su itself.

This might be an interesting read:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=904988#20

But in short, is instead of 'su' using 'su -' working
for you like expected?

Kind regards,
Bernhard

Bug#924507: ksplashqml: hits its hard timeout of 30 seconds

2020-01-29 Thread Bernhard Übelacker

Control: found -1 plasma-workspace/4:5.14.5.1-5


Dear Maintainer,
I found today another device affected by this issue.
It got switched a few days ago from stable to current testing.

It shows the same 30 second splash screen.
A package build with the given patch did show the
splash just 14 seconds.

plasma-workspace version 5.15 or later contain this patch
and should much less likely show this issue.

Kind regards,
Bernhard

Bug#920772: kdeinit5 crashing at MTP transfer to android device.

2020-01-09 Thread Bernhard Übelacker

Dear Maintainer,
I got today the same crash as the submitter.

It happened short after disconnecting one android device,
connecting another and opening/retrying the MTP connection.

This upstream bug looks related:
  https://bugs.kde.org/show_bug.cgi?id=415693
(Seems to be also from buster due to address offsets.)


This line seems to call member getDevice even when there is no object:

  102 LIBMTP_mtpdevice_t *device = 
deviceCache->get(pathItems.at(0))->getDevice();

  https://sources.debian.org/src/kio-extras/4:18.08.3-1/mtp/kio_mtp.cpp/#L102


This line got removed upstream in this commit:
  
https://cgit.kde.org/kio-extras.git/commit/mtp/kio_mtp.cpp?id=aaa1edbb74c4fb01affbde7b79bb45d3a9b61f83

Which points to this task and among others this bug:
  https://phabricator.kde.org/T9390
  https://bugs.kde.org/show_bug.cgi?id=396527


Because the offending line and function removed,
current testing 4:19.08.1-1 might be have this bug fixed.

One upstream mentions that Nautilus was working find,
this might be a workaround, as I guess MTP with KDE
will stay kind of fragile in Buster.


Kind regards,
Bernhard


Without debug symbols:
[KCrash Handler]
#6  0x7f9b372b417a in ?? () from 
/usr/lib/x86_64-linux-gnu/qt5/plugins/kf5/kio/mtp.so
#7  0x7f9b372b9457 in ?? () from 
/usr/lib/x86_64-linux-gnu/qt5/plugins/kf5/kio/mtp.so
#8  0x7f9b372bd65b in ?? () from 
/usr/lib/x86_64-linux-gnu/qt5/plugins/kf5/kio/mtp.so
#9  0x7f9b3272f39f in KIO::SlaveBase::dispatch(int, QByteArray const&) () 
from /usr/lib/x86_64-linux-gnu/libKF5KIOCore.so.5
#10 0x7f9b3272f876 in KIO::SlaveBase::dispatchLoop() () from 
/usr/lib/x86_64-linux-gnu/libKF5KIOCore.so.5
#11 0x7f9b372be7fd in kdemain () from 
/usr/lib/x86_64-linux-gnu/qt5/plugins/kf5/kio/mtp.so
#12 0x556b97c3fe1c in ?? ()
#13 0x556b97c40eea in ?? ()
#14 0x556b97c418fb in ?? ()
#15 0x556b97c3c645 in ?? ()
#16 0x7f9b3646b09b in __libc_start_main (main=0x556b97c3bc70, argc=5, 
argv=0x7ffc69537f98, init=, fini=, 
rtld_fini=, stack_end=0x7ffc69537f88) at ../csu/libc-start.c:308
#17 0x556b97c3d2ca in ?? ()
[Inferior 1 (process 2264) detached]

With debug symbols:
Thread 1 (Thread 0x7f9b32b26780 (LWP 2264)):
[KCrash Handler]
#6  CachedDevice::getDevice (this=0x0) at ./mtp/devicecache.cpp:64
#7  0x7f9b372b9457 in MTPSlave::getPath (this=0x7ffc695377b0, path=...) at 
/usr/include/x86_64-linux-gnu/qt5/QtCore/qlist.h:115
#8  0x7f9b372bd65b in MTPSlave::fileSystemFreeSpace (this=0x7ffc695377b0, 
url=...) at ./mtp/kio_mtp.cpp:946
#9  0x7f9b3272f39f in KIO::SlaveBase::dispatch(int, QByteArray const&) () 
from /usr/lib/x86_64-linux-gnu/libKF5KIOCore.so.5
#10 0x7f9b3272f876 in KIO::SlaveBase::dispatchLoop() () from 
/usr/lib/x86_64-linux-gnu/libKF5KIOCore.so.5
#11 0x7f9b372be7fd in kdemain (argc=, argv=) 
at ./mtp/kio_mtp.cpp:56
#12 0x556b97c3fe1c in launch (argc=4, _name=0x556b993ab398 
"/usr/lib/x86_64-linux-gnu/qt5/plugins/kf5/kio/mtp.so", args=, 
cwd=, envc=0, envs=, reset_env=false, tty=0x0, 
avoid_loops=false, startup_id_str=0x556b97c43187 "0") at 
./src/kdeinit/kinit.cpp:706
#13 0x556b97c40eea in handle_launcher_request (sock=8, who=) 
at ./src/kdeinit/kinit.cpp:1146
#14 0x556b97c418fb in handle_requests (waitForPid=0) at 
./src/kdeinit/kinit.cpp:1339
#15 0x556b97c3c645 in main (argc=5, argv=) at 
./src/kdeinit/kinit.cpp:1785
[Inferior 1 (process 2264) detached]


-- System Information:
Debian Release: 10.2
  APT prefers stable-updates
  APT policy: (500, 'stable-updates'), (500, 'stable-debug'), (500, 
'proposed-updates-debug'), (500, 'proposed-updates'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.19.0-7-amd64 (SMP w/16 CPU cores)
Kernel taint flags: TAINT_WARN, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8), LANGUAGE= 
(charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages kio-extras depends on:
ii  kio5.54.1-1
ii  kio-extras-data4:18.08.3-1
ii  libc6  2.28-10
ii  libgcc11:8.3.0-6
ii  libkf5activities5  5.54.0-1
ii  libkf5archive5 5.54.0-1
ii  libkf5bookmarks5   5.54.0-1
ii  libkf5codecs5  5.54.0-1
ii  libkf5configcore5  5.54.0-1+deb10u1
ii  libkf5configgui5   5.54.0-1+deb10u1
ii  libkf5configwidgets5   5.54.0-1
ii  libkf5coreaddons5  5.54.0-1
ii  libkf5dbusaddons5  5.54.0-1
ii  libkf5dnssd5   5.54.0-1
ii  libkf5guiaddons5   5.54.0-1
ii  libkf5i18n55.54.0-1
ii  libkf5iconthemes5  5.54.0-1
ii  libkf5khtml5   5.54.0-1
ii  libkf5kiocore5 5.54.1-1
ii  libkf5kiofilewidgets5  5.54.1-1
ii  libkf5kiowidgets5  5.54.1-1
ii  libkf5parts5   5.54.0-1
ii  libkf5pty5 5.54.0-1
ii  libkf5service-bin  5.54.0-1
ii  libkf5service5 5.54.0-1
ii  libkf5solid5   5.54.0-1
ii

Bug#946420: kpat is spamming with error messages

2019-12-31 Thread Bernhard Übelacker

Control: notfound -1 4:18.04.1-1
Control: found -1 4:19.08.1-1
Control: tags -1 unreproducible moreinfo



Hello Hans,
I tried to reproduce inside a minimal plasma testing VM.
But I could not see a noticeable amount of messages in 
.xsession_errors while playing kpat.

If this is still visible on your system then
some more information could help.
- Which game have you played inside kpat?
- Some example lines of the messages in .xsession_errors

Kind regards,
Bernhard

Bug#942257: phonon-backend-gstreamer: Cannot upgrade phonon-backend-gstreamer-common to 4:4.9.1-2, this package requires 4:4.9.1-1

2019-12-12 Thread Bernhard Übelacker

Control: tags -1 - upstream


Hello Felix,
maybe the information from following bug is
relevant in this case too.

https://bugs.debian.org/942860

Kind regards,
Bernhard

Bug#942860: 942860: phonon: no functional current backend

2019-12-12 Thread Bernhard Übelacker

Hello Karl,
I am not involved in the packaging of phonon, but
still a few questions.

Have you any Qt4 applications installed?

Because src:phonon in version 4:4.10.3-2 was the last
version that built the Qt4 parts.

Since src:phonon 4:4.10.3-3 just the Qt5 packages
get build anymore.

And therefore I guess phonon and phonon-backend
got replaced by phonon4qt5 and phonon4qt5-backend
some time ago.

Maybe there should exist a transitional package
for upgrading phonon to phonon4qt5?

Kind regards,
Bernhard

Bug#946073: libkf5kiocore5: Several KDE applications crash when creating multiple nested new folders in the save as dialog.

2019-12-03 Thread Bernhard Übelacker

Package: libkf5kiocore5
Version: 5.54.1-1
Severity: normal
Tags: patch upstream


Dear Maintainer,
in the last year I hit a few crashes with kate,
without knowing how to reproduce the crash.

Today I found this upstream reports [1] and several
duplicates. With that information it was easy to reproduce
the crash by just creating several nested folders in the
save as dialog of okteta.

It got fixed upstream in [2] and is included in
upstream version 5.63.0.

As this may affect all users of the KDE save as dialog,
this might be considered for inclusion to stable?

Kind regards,
Bernhard


[1] https://bugs.kde.org/show_bug.cgi?id=401916
https://bugs.kde.org/show_bug.cgi?id=411310

[2] 
https://cgit.kde.org/kio.git/commit/src/core/kcoredirlister.cpp?id=f539b17ceec2a8c7a453228b7db62a7919b849be



-- System Information:
Debian Release: 10.2
  APT prefers stable-updates
  APT policy: (500, 'stable-updates'), (500, 'stable-debug'), (500, 
'proposed-updates'), (500, 'stable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.19.0-6-amd64 (SMP w/16 CPU cores)
Kernel taint flags: TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8), LANGUAGE= 
(charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages libkf5kiocore5 depends on:
ii  libacl12.2.53-4
ii  libc6  2.28-10
ii  libkf5auth55.54.0-2
ii  libkf5configcore5  5.54.0-1+deb10u1
ii  libkf5coreaddons5  5.54.0-1
ii  libkf5crash5   5.54.0-1
ii  libkf5dbusaddons5  5.54.0-1
ii  libkf5i18n55.54.0-1
ii  libkf5service-bin  5.54.0-1
ii  libkf5service5 5.54.0-1
ii  libqt5core5a   5.11.3+dfsg1-1+deb10u1
ii  libqt5dbus55.11.3+dfsg1-1+deb10u1
ii  libqt5network5 5.11.3+dfsg1-1+deb10u1
ii  libqt5xml5 5.11.3+dfsg1-1+deb10u1
ii  libstdc++6 8.3.0-6

libkf5kiocore5 recommends no packages.

libkf5kiocore5 suggests no packages.

-- no debconf information

# Buster/stable amd64

Thread 1 (Thread 0x7f25debb0c80 (LWP 2619)):
[KCrash Handler]
#6  0x7f25e4e16426 in KFileItem::~KFileItem() () from 
/usr/lib/x86_64-linux-gnu/libKF5KIOCore.so.5
#7  0x7f25e4eea06a in ?? () from 
/usr/lib/x86_64-linux-gnu/libKF5KIOCore.so.5
#8  0x7f25e4ee19cd in ?? () from 
/usr/lib/x86_64-linux-gnu/libKF5KIOCore.so.5
#9  0x7f25e381e9a3 in QMetaObject::activate(QObject*, int, int, void**) () 
from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#10 0x7f25e382a3f7 in QTimer::timeout(QTimer::QPrivateSignal) () from 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#11 0x7f25e381f09b in QObject::event(QEvent*) () from 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#12 0x7f25e42e14c1 in QApplicationPrivate::notify_helper(QObject*, QEvent*) 
() from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#13 0x7f25e42e8970 in QApplication::notify(QObject*, QEvent*) () from 
/usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#14 0x7f25e37f54f9 in QCoreApplication::notifyInternal2(QObject*, QEvent*) 
() from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#15 0x7f25e3845ba8 in QTimerInfoList::activateTimers() () from 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#16 0x7f25e3846404 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#17 0x7f25e1547f2e in g_main_context_dispatch () from 
/usr/lib/x86_64-linux-gnu/libglib-2.0.so.0
#18 0x7f25e15481c8 in ?? () from /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0
#19 0x7f25e154825c in g_main_context_iteration () from 
/usr/lib/x86_64-linux-gnu/libglib-2.0.so.0
#20 0x7f25e3846797 in 
QEventDispatcherGlib::processEvents(QFlags) () 
from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#21 0x7f25de7ec401 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5XcbQpa.so.5
#22 0x7f25e37f41cb in 
QEventLoop::exec(QFlags) () from 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#23 0x7f25e37fc1a2 in QCoreApplication::exec() () from 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#24 0x559a5cc4aa25 in ?? ()
#25 0x7f25e326809b in __libc_start_main (main=0x559a5cc48ac0, argc=3, 
argv=0x7fffb1e60268, init=, fini=, 
rtld_fini=, stack_end=0x7fffb1e60258) at ../csu/libc-start.c:308
#26 0x559a5cc4c0ba in _start ()
[Inferior 1 (process 2619) detached]


gdb -q --pid 2619

generate-core-file /coredumps/core.kate.20191114.2619


##


apt install dpkg-dev devscripts xserver-xorg plasma-desktop gdb kate 
kate-dbgsym libkf5kiocore5-dbgsym libqt5core5a-dbgsym libqt5widgets5-dbgsym 
libglib2.0-0-dbgsym libqt5gui5-dbgsym

gdb -q /usr/bin/kate --core core.kate.20191114.2619

set width 0
set pagination off
bt

(gdb) bt
#0  0x7f25e330a720 in __GI___nanosleep (requested_time=0x7fffb1e5eee0, 
remaining=0x7fffb1e5eee0) at ../sysdeps/unix/sysv/linux/nanosleep.c:28
#1  0x7f25e330a62a in __sleep (seconds=0) at ../sysdeps/posix/sleep.c:55
#2  0x7f25e482deea in ?? () from /usr/lib/x86_64-linux-gnu/libKF5Crash.so.5
#3  0x7f25e482e68e in ?? () from

Bug#891233: Bug #891233: kamoso: segmentation fault in kamoso in Debian 9 stable. Buster not affected

2019-09-01 Thread Bernhard Übelacker

Dear Maintainer,
I guess the actual segmentation fault is fixed since kamoso 3.2.4-1.
Instead it should print this message:

The webcam controller was unable to find or load wrappercamerabinsrc plugin;
please make sure all required gstreamer plugins are installed.

The last question would be, is this sufficient to close the bug,
or does kamoso need a dependency or recommends to
gstreamer1.0-plugins-bad?

Kind regards,
Bernhard

Bug#932808: baloo-kf5: Upon start baloo crashes with Segmentation fault following upgrade to buster

Control: tags -1 - moreinfo



Dear Maintainer,
I think the top frames with debug symbols would look like this:

(gdb) bt
#0  memcpy () in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#1  0x7feaa926d05d in Baloo::PostingCodec::decode () at 
./src/codecs/postingcodec.cpp:42
#2  0x7feaa925a679 in Baloo::PostingDB::get () at 
./src/engine/postingdb.cpp:100
#3  0x7feaa926a8ca in Baloo::WriteTransaction::commit () at 
./src/engine/writetransaction.cpp:275
#4  0x7feaa9260ea2 in Baloo::Transaction::commit () at 
./src/engine/transaction.cpp:269
...

This seems to point to upstream bug report [1].

The return value of function e.g. mdb_get might
not be checked enough.

An attempt to solve this seems to be commit [2].

Kind regards,
Bernhard


[1] https://bugs.kde.org/show_bug.cgi?id=367480
[2] 
https://cgit.kde.org/baloo.git/commit/src/engine/postingdb.cpp?id=eb68430ae5f4fc4c4a2e5b1798683ff6a639ad3d

Bug#932808: baloo-kf5: Upon start baloo crashes with Segmentation fault following upgrade to buster

Hello Leoš Pohl,
thanks for the additional information.

Please use the reply-all in your email program, to have
the ...@bugs.debian.org recipient in every answer, so
the debian bug tracker receives the messages too.

Just for future crashes (or if the Mainainter will requests it),
if you install the debug symbol packages, like described in [1],
the backtrace you can receive from DrKonqi gets way better,
and would e.g. already show the line information.

In your case these packages would be helpful:
baloo-kf5-dbgsym libkf5balooengine5-dbgsym

Kind regards,
Bernhard

[1] https://wiki.debian.org/HowToGetABacktrace#Installing_the_debugging_symbols

Bug#932808: Fwd: Bug#932808: baloo-kf5: Upon start baloo crashes with Segmentation fault following upgrade to buster



 Weitergeleitete Nachricht 
Datum:  Wed, 7 Aug 2019 15:44:44 -0400
Von:Leoš Pohl



Hello Bernhard,

since I could not find a way to forward the report directly
to this bug, please find screenshots here
https://snag.gy/mpTyWe.jpg and here https://snag.gy/lUKpRc.jpg
and the copied report from the second tab is below.


Application: Baloo File Indexing Daemon (baloo_file), signal: Segmentation fault
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[Current thread is 1 (Thread 0x7fad6cbec400 (LWP 8615))]

Thread 3 (Thread 0x7fad6bb57700 (LWP 8617)):
[KCrash Handler]
#6  0x7fad70df671b in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#7  0x7fad7178405d in ?? () from 
/usr/lib/x86_64-linux-gnu/libKF5BalooEngine.so.5
#8  0x7fad71771679 in Baloo::PostingDB::get(QByteArray const&) () from 
/usr/lib/x86_64-linux-gnu/libKF5BalooEngine.so.5
#9  0x7fad717818ca in Baloo::WriteTransaction::commit() () from 
/usr/lib/x86_64-linux-gnu/libKF5BalooEngine.so.5
#10 0x7fad71777ea2 in Baloo::Transaction::commit() () from 
/usr/lib/x86_64-linux-gnu/libKF5BalooEngine.so.5
#11 0x55f4dcf68f0d in ?? ()
#12 0x7fad7121d021 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#13 0x7fad71224aa7 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#14 0x7fad6fdadfa3 in start_thread () from 
/lib/x86_64-linux-gnu/libpthread.so.0
#15 0x7fad70d934cf in clone () from /lib/x86_64-linux-gnu/libc.so.6

Thread 2 (Thread 0x7fad6c68a700 (LWP 8616)):
#0  0x7fad70d88819 in poll () from /lib/x86_64-linux-gnu/libc.so.6
#1  0x7fad6f4e3136 in ?? () from /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0
#2  0x7fad6f4e325c in g_main_context_iteration () from 
/usr/lib/x86_64-linux-gnu/libglib-2.0.so.0
#3  0x7fad7141d87b in 
QEventDispatcherGlib::processEvents(QFlags) () 
from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#4  0x7fad713cb27b in 
QEventLoop::exec(QFlags) () from 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#5  0x7fad7121aec6 in QThread::exec() () from 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#6  0x7fad717a3545 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5DBus.so.5
#7  0x7fad71224aa7 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#8  0x7fad6fdadfa3 in start_thread () from 
/lib/x86_64-linux-gnu/libpthread.so.0
#9  0x7fad70d934cf in clone () from /lib/x86_64-linux-gnu/libc.so.6

Thread 1 (Thread 0x7fad6cbec400 (LWP 8615)):
#0  0x7fad70d88819 in poll () from /lib/x86_64-linux-gnu/libc.so.6
#1  0x7fad6f4e3136 in ?? () from /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0
#2  0x7fad6f4e325c in g_main_context_iteration () from 
/usr/lib/x86_64-linux-gnu/libglib-2.0.so.0
#3  0x7fad7141d87b in 
QEventDispatcherGlib::processEvents(QFlags) () 
from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#4  0x7fad713cb27b in 
QEventLoop::exec(QFlags) () from 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#5  0x7fad713d3262 in QCoreApplication::exec() () from 
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#6  0x55f4dcf63987 in ?? ()
#7  0x7fad70cbe09b in __libc_start_main () from 
/lib/x86_64-linux-gnu/libc.so.6
#8  0x55f4dcf63b2a in ?? ()
[Inferior 1 (process 8615) detached]

Bug#932808: baloo-kf5: Upon start baloo crashes with Segmentation fault following upgrade to buster

Control: tags -1 + moreinfo


Hello Leos Pohl,
you write when you execute "balooctl enable" you receive
some lines starting with "KCrash: ...".

When this happened, is there a small sad smiley [1] in the
system tray (normally at the bottom, left of the clock)?
If yes, you should receive by clicking on that a text describing
where the crash happened, which you could forward to this bug.

Kind regards,
Bernhard

[1] 
https://sources.debian.org/data/main/b/breeze-icons/4:5.54.0-1/icons/actions/16/tools-report-bug.svg

Bug#594506: konsole crashed (signal 11)

Hello David,
that was not the link I sent in my mail,
I sent a plain link to bugs.debian.org.
Maybe you are viewing my mail via google webmail client,
and that is adding something unwanted?

Kind regards,
Bernhard

Bug#931844: libqt5gui5: Endless loop in konsole showing midnight commander with activated orca screenreader.

Control: forwarded -1 https://bugreports.qt.io/browse/QTBUG-77017


Dear Maintainer,
I created an upstream bug report and forward this bug to it.

Kind regards,
Bernhard

Bug#594506: konsole crashed (signal 11)

Hello David,
I could find an issue with konsole freezing
and reported it in this bug:
https://bugs.debian.org/931844

Maybe you want to have a look if you find your issue there.

Kind regards,
Bernhard

Bug#931844: libqt5gui5: Endless loop in konsole showing midnight commander with activated orca screenreader.

Package: libqt5gui5
Version: 5.11.3+dfsg1-1
Severity: wishlist
Tags: upstream
Affects: konsole orca


Dear Maintainer,
I noted some recent additions to bug #594506 which I guess
describe a different problem. I did some investigations and
tripped on the issue with following steps:

- Installed in a qemu VM a minimal Buster/amd64 system,
  plus plasma-desktop, konsole, mc and orca.

- In systemsettings activate orca and changed orca to read
  the text below the mouse.

- Restarted

- Open konsole

- Inside that console, start "mc"

- Move the mouse over the konsole window

- Now konsole is frozen, endless looping
  in QAccessibleTextInterface::textAtOffset.

As far as I see the problem might be that boundary.position()
returns -1, therefore the condition to leave the second loop
in QAccessibleTextInterface::textAtOffset is never reached.

I could create a minimal reproducer that demonstrates the issue.
(reproduzer.tar.gz)

A workaround could be the attached patch that just leaves the
function in this situation with some logging.
(Leave-QAccessibleTextInterface__textAtOffset.patch)

Kind regards,
Bernhard




Thread 1 (Thread 0x7f2babb65840 (LWP 1312)):
#0  0x7f2bb0f08a34 in QTextBoundaryFinder::toNextBoundary 
(this=this@entry=0x7fff257d6d60) at tools/qtextboundaryfinder.cpp:366
#1  0x7f2bb14443ac in QAccessibleTextInterface::textAtOffset 
(this=, offset=0, boundaryType=QAccessible::WordBoundary, 
startOffset=0x7fff257d6e08, endOffset=0x7fff257d6e0c) at 
accessible/qaccessible.cpp:2321
#2  0x7f2bab7f28df in AtSpiAdaptor::textInterface (this=, 
interface=, function=..., message=..., connection=...) at 
atspiadaptor.cpp:1908
#3  0x7f2bab7f321d in AtSpiAdaptor::handleMessage 
(this=this@entry=0x5619664085d0, message=..., connection=...) at 
atspiadaptor.cpp:1284
#4  0x7f2bb072bdf9 in QDBusConnectionPrivate::activateObject 
(this=0x7f2b9c014520, node=..., msg=..., pathStartPos=27) at 
qdbusconnection_p.h:345
#5  0x7f2bb072e8ce in QDBusActivateObjectEvent::placeMetaCall 
(this=0x7f2b9c01ed00) at qdbusintegrator.cpp:1597
#6  0x7f2bb1056182 in QObject::event (this=0x5619664085d0, e=) at kernel/qobject.cpp:1251
#7  0x7f2bb1a474b1 in QApplicationPrivate::notify_helper 
(this=this@entry=0x561965f3e270, receiver=receiver@entry=0x5619664085d0, 
e=e@entry=0x7f2b9c01ed00) at kernel/qapplication.cpp:3726
#8  0x7f2bb1a4e950 in QApplication::notify (this=0x561965f3e030, 
receiver=0x5619664085d0, e=0x7f2b9c01ed00) at kernel/qapplication.cpp:3485
#9  0x7f2bb102c5a9 in QCoreApplication::notifyInternal2 
(receiver=0x5619664085d0, event=0x7f2b9c01ed00) at 
../../include/QtCore/5.11.3/QtCore/private/../../../../../src/corelib/thread/qthread_p.h:307
#10 0x7f2bb102f59b in QCoreApplication::sendEvent (event=0x7f2b9c01ed00, 
receiver=) at 
../../include/QtCore/../../src/corelib/kernel/qcoreapplication.h:234
#11 QCoreApplicationPrivate::sendPostedEvents (receiver=0x0, event_type=0, 
data=0x561965f3d600) at kernel/qcoreapplication.cpp:1744
#12 0x7f2bb107e233 in postEventSourceDispatch (s=0x561965f95a00) at 
kernel/qeventdispatcher_glib.cpp:276
#13 0x7f2baf2dff2e in g_main_dispatch (context=0x7f2ba4004ff0) at 
../../../glib/gmain.c:3182
#14 g_main_context_dispatch (context=context@entry=0x7f2ba4004ff0) at 
../../../glib/gmain.c:3847
#15 0x7f2baf2e01c8 in g_main_context_iterate 
(context=context@entry=0x7f2ba4004ff0, block=block@entry=1, 
dispatch=dispatch@entry=1, self=) at ../../../glib/gmain.c:3920
#16 0x7f2baf2e025c in g_main_context_iteration (context=0x7f2ba4004ff0, 
may_block=may_block@entry=1) at ../../../glib/gmain.c:3981
#17 0x7f2bb107d863 in QEventDispatcherGlib::processEvents 
(this=0x561965f8d0c0, flags=...) at kernel/qeventdispatcher_glib.cpp:422
#18 0x7f2bab7c93e1 in QPAEventDispatcherGlib::processEvents 
(this=0x561965f8d0c0, flags=...) at qeventdispatcher_glib.cpp:69
#19 0x7f2bb102b27b in QEventLoop::exec (this=this@entry=0x7fff257d7580, 
flags=..., flags@entry=...) at 
../../include/QtCore/../../src/corelib/global/qflags.h:140
#20 0x7f2bb1033262 in QCoreApplication::exec () at 
../../include/QtCore/../../src/corelib/global/qflags.h:120
#21 0x7f2bb146cbdc in QGuiApplication::exec () at 
kernel/qguiapplication.cpp:1762
#22 0x7f2bb1a47425 in QApplication::exec () at kernel/qapplication.cpp:2900
#23 0x7f2bb2a94827 in kdemain (argc=, argv=) 
at ./src/main.cpp:190
#24 0x7f2bb28cf09b in __libc_start_main (main=0x5619659be730 , 
argc=1, argv=0x7fff257d7838, init=, fini=, 
rtld_fini=, stack_end=0x7fff257d7828) at ../csu/libc-start.c:308
#25 0x5619659be76a in _start ()





-- System Information:
Debian Release: 10.0
  APT prefers stable-debug
  APT policy: (500, 'stable-debug'), (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 4.19.0-5-amd64 (SMP w/12 CPU cores)
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8), 
LANGUAGE=de_DE.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via

Bug#931646: Kmail crashes when I click on mail

2019-07-08 Thread Bernhard Übelacker

Hello Francis Laniel,
I am just looking at some crashes in some random packages,
and found this nouveau issue familiar.

I guess __pthread_cond_wait_common should be able to
handle a NULL for abstime - e.g. should "Block without a timeout".
https://sources.debian.org/src/glibc/2.28-10/nptl/pthread_cond_wait.c/#L499

There are also some many more bugs reported that all end with these
nouveau_pushbuf_kick/nouveau_pushbuf_data calls triggering
that "Assertion `kref' failed.", that might be related:
https://bugs.debian.org/927954
https://bugs.debian.org/929130
https://bugs.freedesktop.org/show_bug.cgi?id=91632
https://bugs.freedesktop.org/show_bug.cgi?id=98039
https://bugzilla.redhat.com/show_bug.cgi?id=1350275
https://bugs.archlinux.org/task/59057
https://bugreports.qt.io/browse/QTBUG-41242

The last one mentions as a workaround setting an
environment variable LIBGL_ALWAYS_SOFTWARE=1 or
QT_XCB_FORCE_SOFTWARE_OPENGL=1, maybe on of these still helps?

Kind regards,
Bernhard

Bug#594506: konsole crashed (signal 11)

2019-07-08 Thread Bernhard Übelacker

Hello David,
I am not involved in packaging, just tried to collect some more
information to this old bug.

I guess your issue migth not be that one discussed in this bug.
At least you might be able to provide the output of 'dmesg',
if in your case konsole really crashes.
Or after installing the package systemd-coredump and another
attempt to open konsole, you should receive more details
at the bottom of 'journalctl --no-pager', if this really is a crash.

If there is no usable information it is most likely a different issue,
and you might look to the bottom of your ~/.xsession-errors after
retrying to start konsole, and you may open a new bug then.

Kind regards,
Bernhard

Bug#928710: drkonqi: segfault after using reload button to get backtrace for dolphin

2019-05-09 Thread Bernhard Übelacker

Control: tags 928710 + upstream
Control: forwarded 928710 https://bugs.kde.org/show_bug.cgi?id=381644


Dear Maintainer,
above the upstream bug of this issue.

Kind regards,
Bernhard

Bug#918449: kate freeze when right-click the open/save file dialog

2019-05-05 Thread Bernhard Übelacker

Hello Joe Ma,

On Sun, 20 Jan 2019 19:59:27 +0800 Joe Ma  wrote:
> Hello Bernhard,
> For your question, I do execute your gdb command in the console by a shell
> script when kate freeze. What information do you want me to add? I will
> reply asap.

sorry for the late reply, I did not notice your answer as
the last mail got just sent to the bug tracking system.
Please use reply-all for answers.

Do you still see this problem?

Unfortunately I still can see no difference to a kate
running on my system that is not frozen.

You have KDE Plasma as desktop running or something else?
Have you some network filesystems mounted?

Kind regards,
Bernhard

Bug#886692: kate: crashing when opening a file from the Konsole CLI, when $(PWD) is too long

2019-05-03 Thread Bernhard Übelacker

Control: tags 886692 + unreproducible moreinfo



Hello Alberto,
I just tried to get some more information from this crash.

I assume this is not caused by the length of the current directory.
Instead I think this is caused by this file:

/DEFAULT/home/marmo/Work/NEGF/SPRKKR_LMAT_practice/.kateproject

If that does not exist the code path from your
backtrace is not entered.

Maybe you could forward this file to this bug?

Kind regards,
Bernhard

# Stretch amd64 qemu VM 2019-05-03


apt update
apt dist-upgrade


apt install systemd-coredump xserver-xorg dbus-x11 sddm plasma-desktop gdb mc 
git kate kate-dbgsym
apt build-dep kate


reboot




mkdir /home/benutzer/source/kate/orig -p
cd/home/benutzer/source/kate/orig
apt source kate
cd


mkdir .gdb
cd .gdb
git clone https://github.com/Lekensteyn/qt5printers.git
cat << EOF > qt5-gdbinit
python
import sys, os.path
sys.path.insert(0, os.path.expanduser('~/.gdb'))
import qt5printers
qt5printers.register_printers(gdb.current_objfile())
end
EOF
cd ..





mkdir /DEFAULT/home/marmo/Work/NEGF/SPRKKR_LMAT_practice -p
cd/DEFAULT/home/marmo/Work/NEGF/SPRKKR_LMAT_practice
echo test > negf.f
chown benutzer:benutzer /DEFAULT/home/marmo -R
cp -a negf.f /tmp




cat << EOF > .kateproject
{
   "name": "MyProject",
   "files": [ {
 "directory": "doc",
 "filters": ["*.tex", "Makefile"],
 "recursive": 1
   } ]
}
EOF







export DISPLAY=:0
cd/DEFAULT/home/marmo/Work/NEGF/SPRKKR_LMAT_practice
kate negf.f






gdb -q --args kate negf.f

set width 0
set pagination off
display/i $pc
b KatePluginManager::enablePluginGUI
run
disa 1
print 
_ZN17KatePluginManager15enablePluginGUIEP14KatePluginInfoP14KateMainWindowP11KConfigBase
b * $1 + 630
cont



set width 0
set pagination off
directory /home/benutzer/source/kate/orig/kate-16.08.3
display/i $pc
b KateProjectPluginView::KateProjectPluginView
y
run



set width 0
set pagination off
directory /home/benutzer/source/kate/orig/kate-16.08.3
source /home/benutzer/.gdb/qt5-gdbinit
display/i $pc
b KateProjectPlugin::projectForDir
y
b KateProjectPlugin::createProjectForFileName
y
run





set width 0
set pagination off
directory /home/benutzer/source/kate/orig/kate-16.08.3
source /home/benutzer/.gdb/qt5-gdbinit
display/i $pc
b KateProjectPluginView::viewForProject
y
run
disa 1
print KateProjectPluginView::viewForProject
b * $1 + 200
cont



benutzer@debian:/DEFAULT/home/marmo/Work/NEGF/SPRKKR_LMAT_practice$ gdb -q 
--args kate negf.f
Reading symbols from kate...Reading symbols from 
/usr/lib/debug/.build-id/f6/ac5a4ec48449ec7744c62c72f9f330bdae0094.debug...done.
done.
(gdb) set width 0
(gdb) set pagination off
(gdb) directory /home/benutzer/source/kate/orig/kate-16.08.3
Source directories searched: 
/home/benutzer/source/kate/orig/kate-16.08.3:$cdir:$cwd
(gdb) source /home/benutzer/.gdb/qt5-gdbinit
(gdb) display/i $pc
1: x/i $pc

(gdb) b KateProjectPluginView::viewForProject
Function "KateProjectPluginView::viewForProject" not defined.
Make breakpoint pending on future shared library load? (y or [n]) y
Breakpoint 1 (KateProjectPluginView::viewForProject) pending.
(gdb) run
Starting program: /usr/bin/kate negf.f
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7fffe291f700 (LWP 7958)]
[New Thread 0x7fffe19cb700 (LWP 7959)]
[New Thread 0x7fffdbfff700 (LWP 7961)]
[New Thread 0x7fffdb7fe700 (LWP 7962)]
[New Thread 0x7fffdaffd700 (LWP 7964)]
[New Thread 0x7fffda7fc700 (LWP 7965)]
[New Thread 0x7fffd9ffb700 (LWP 7966)]
[New Thread 0x7fffd97fa700 (LWP 7967)]
[New Thread 0x7fffd8ff9700 (LWP 7968)]
[New Thread 0x7fffb700 (LWP 7969)]
[New Thread 0x7fffbf7fe700 (LWP 7970)]
[New Thread 0x7fffbeffd700 (LWP 7971)]
[New Thread 0x7fffbe7fc700 (LWP 7972)]
[New Thread 0x7fffbdffb700 (LWP 7973)]
[New Thread 0x7fffbd7fa700 (LWP 7974)]
[New Thread 0x7fffbcff9700 (LWP 7975)]
[New Thread 0x7fff9bfff700 (LWP 7976)]
[New Thread 0x7fff9b7fe700 (LWP 7977)]

Thread 1 "kate" hit Breakpoint 1, KateProjectPluginView::viewForProject 
(this=this@entry=0x55acb330, project=0x55995430) at 
./addons/project/kateprojectpluginview.cpp:134
134 {
1: x/i $pc
=> 0x7fffe0d27560 :
push   %r15
(gdb) disa 1
(gdb) print KateProjectPluginView::viewForProject
$1 = {QPair (KateProjectPluginView * 
const, KateProject *)} 0x7fffe0d27560 

(gdb) b * $1 + 200
Breakpoint 2 at 0x7fffe0d27628: file 
./addons/project/kateprojectpluginview.cpp, line 190.
(gdb) cont
Continuing.

Thread 1 "kate" hit Breakpoint 2, 0x7fffe0d27628 in 
KateProjectPluginView::viewForProject (this=this@entry=0x55acb330, 
project=0x55995430) at ./addons/project/kateprojectpluginview.cpp:190
190 m_stackedProjectInfoViews->addWidget(infoView);
1: x/i $pc
=> 0x7fffe0d27628 :
callq  0x7fffe0d1ece0 <_ZN14QStackedWidget9addWidgetEP7QWidget@plt>
(gdb) bt
#0  0x7fffe0d27628 in KateProjectPluginView::viewForProject

Bug#927954: konqueror: Exit when opening http https' ftps' links

2019-04-28 Thread Bernhard Übelacker

Control: reassign 927954 libdrm-nouveau2 2.4.97-1
Control: affects 927954 + konqueror



Hello Osama Nasr,

> I'm so sorry for that,
> I just reinstalled KDE and every thing works fine,
> I'm sorry if that a waste of your time.

Glad to hear that it is working.
But as this might be a threading issue it might not
show up every time.


> Could you please explain me if this is a problem in my OS

I have no idea. I powered on a system with an older nvidia card
and when using the Nouveau driver glxinfo prints information
for that card, nothing about vmware.


The backtrace you supplied in message #17 points to following
source lines in package libdrm-nouveau2.

konqueror: ../nouveau/pushbuf.c:723: nouveau_pushbuf_data: Assertion `kref' 
failed.
Received signal 6
...
#7 0x7f73f7a180f2 __assert_fail
#8 0x7f73ee7a059f nouveau_pushbuf_data in nouveau_pushbuf_data () at 
../nouveau/pushbuf.c:723
#9 0x7f73ee7a0503 nouveau_pushbuf_data in nouveau_pushbuf_data () at 
../nouveau/pushbuf.c:715 at /lib/x86_64-linux-gnu/libdrm_nouveau.so.2
#10 0x7f73ee7a062fin pushbuf_submit ()   at 
../nouveau/pushbuf.c:326 at /lib/x86_64-linux-gnu/libdrm_nouveau.so.2
#11 0x7f73ee7a0a7fin pushbuf_flush ()at 
../nouveau/pushbuf.c:401 at /lib/x86_64-linux-gnu/libdrm_nouveau.so.2
#12 0x7f73ee7a1670 nouveau_pushbuf_kickin nouveau_pushbuf_kick () at 
../nouveau/pushbuf.c:775 at /lib/x86_64-linux-gnu/libdrm_nouveau.so.2
#13 0x7f73ed89dbc6 


So I am going to reassign this bug to package libdrm-nouveau2,
maybe they want to know about it.


Kind regards,
Bernhard

Bug#927954: konqueror: Exit when opening http https' ftps' links

2019-04-27 Thread Bernhard Übelacker

Hello Osama Nasr,

> oh, this system is running in a VM, that
> might be important information. 
> 
> But I'm runing it as my main OS, I mean I'm not runnig on VM.
> 
> BTW, when using Budgie, the problem disappear...

What output gives this command:
   lspci -nn | grep VGA

Kind regards,
Bernhard

Bug#927954: konqueror: Exit when opening http https' ftps' links

2019-04-27 Thread Bernhard Übelacker

Hello Osama Nasr,


> glxinfo -B
> name of display: :0
> display: :0  screen: 0
> direct rendering: Yes
> Extended renderer info (GLX_MESA_query_renderer):
> Vendor: VMware, Inc. (0x)
> Device: llvmpipe (LLVM 7.0, 256 bits) (0x)

oh, this system is running in a VM, that
might be important information.


> I cant install those packages.
> when hit apt install gdb konqueror-dbgsym libgl1-mesa-dri-dbgsym
> unable to locate packages , even after update.

The dbgsym packages are not in the regular repositories,
the mentioned link in my previous mail describes
what repositories contain them.


Kind regards,
Bernhard

Bug#920139: sddm: GTK and GNOME: Applications won't launch due error of glib2

2019-04-27 Thread Bernhard Übelacker

Hello Adrian,
so unfortunately it has still no clear error.

In your .xsession-errors is tcosmonitor mentioned.
As this package got 2016 removed [1] from debian - do you
have still installed tcosmonitor or tcosmonitor-common?

If yes, maybe that might cause some troubles?
If no, which file tries to load "tcosmonitor.shared"?

Do you have other packages not contained in testing and
any other sources.list, that may cause trouble?
LANG=C apt-show-versions | grep No

Back in 2010, was this system installed from a
regular debian installer?

Kind regards,
Bernhard

[1] https://tracker.debian.org/pkg/tcosmonitor

Bug#920139: sddm: GTK and GNOME: Applications won't launch due error of glib2

2019-04-26 Thread Bernhard Übelacker

Hello Adrian,


> Using a console login and startx, everything works fine though.
> 
> Logging in into amiwm works fine though!
> 

> I also can't use a session created through xdm or a login manager
> because a lot of GNOME and GTK complain or even seqfault and won't
> start after creating a session through xdm.
> 
> I suspect this is a general bug with login manager created X sessions.
> 
> I also have to note, this is is a installation from the year 2010,
> always kept updated to newest Debian/testing. 
> 
> Maybe something broke meanwhile.

So I have installed in a minimal Buster amd64 VM these packages:
xdm gnome
And configured xdm as default.
With that I could login just fine, so this seems not an issue
in a fresh installation.


In this configuration it looks like gnome is running with xorg.
Therefore you might also have a look into $HOME/.xsession-errors
and /var/log/Xorg.0.log or /var/log/Xorg.1.log.

Do you get dropped back to xdm immediately or do you stay at e.g. a
black screen?

Is there anything in the syslog above the "Started User Manager for..."
message in the same second and a few before?


> Does this read any package maintainer?

Currently this bug is assigned to package sddm, so just
sddm maintainer might have this on the screen.
Also your last email describes kind of a different issue,
with not even sddm involved.

Because you say this can be seen with sddm and xdm I guess
the problem might not be in these packages, therefore reassigning
to gnome-session might attract the right people?

Kind regards,
Bernhard

Bug#927954: konqueror: Exit when opening http https' ftps' links

2019-04-26 Thread Bernhard Übelacker

Hello Osama Nasr,


> Thanks a lot for your help, but actually i don't need that app, I just
> was reporting this bug.

I am also not involved in packaging konqueror, just tried to collect
some more information, for the maintainers to work with.


> Do you need me to install proprietary drivers and test it ?

I think this would gain not much for this bug, it would just be
a workaround, but as far as other applications are working for you
this might not be neccessary.


What might be interesting would be a backtrace with debug symbols
installed like described in [1].

Following steps should produce such a file that might be useful
for the maintainer and could be forwarded to this bug:

- Install packages: gdb konqueror-dbgsym libgl1-mesa-dri-dbgsym

- Start konqueror as usual

- Attach gdb to it with this command line:
script -a gdb_konqueror_$(date +%Y-%m-%d_%H-%M-%S).log -c "gdb -q -batch 
-ex 'set width 0' -ex 'set pagination off' -ex 'cont' -ex 'display/i $pc' -ex 
'bt' -ex 'info share' -ex 'info reg' -ex 'bt full' --pid $(pidof konqueror)"

- Then open the ususal webpage to trigger the crash.


Kind regards,
Bernhard


[1] https://wiki.debian.org/HowToGetABacktrace#Installing_the_debugging_symbols

Bug#927954: konqueror: Exit when opening http https' ftps' links

Hello Osama Nasr,


> By the way, I've installed GIMP, although there was a bug
> (libmypaint-common #906144).
> I don't know if that have a relation or not.

On a short look I guess this is not related.



> konqueror: ../nouveau/pushbuf.c:723: nouveau_pushbuf_data: Assertion 
> `kref' failed.
> Received signal 6
> #0 0x7f73c396cbde 
> #1 0x7f73c396ccf0 
> #2 0x7f73c396d327 
> #3 0x7f73f7a1f940 
> #4 0x7f73f7a1f8bb gsignal
> #5 0x7f73f7a0a535 abort
> #6 0x7f73f7a0a40f 
> #7 0x7f73f7a180f2 __assert_fail
> #8 0x7f73ee7a059f nouveau_pushbuf_data
> #9 0x7f73ee7a0503 nouveau_pushbuf_data
> #10 0x7f73ee7a062f 
> #11 0x7f73ee7a0a7f 
> #12 0x7f73ee7a1670 nouveau_pushbuf_kick
> #13 0x7f73ed89dbc6 
> #14 0x7f73ed9c5faf 
> #15 0x7f73ed5a0b37 
> #16 0x7f73ee77b243 
> #17 0x7f73effa8146 
> #18 0x7f73f60b6c4b QOpenGLContext::swapBuffers()
> ...

I guess this is the most interesting part and points to
an error in the nouveau driver.

This bug might be related:

https://bugs.freedesktop.org/show_bug.cgi?id=91632

It may help if you also add the output of this command, if installed:

glxinfo -B


Kind regards,
Bernhard

Bug#927954: konqueror: Exit when opening http https' ftps' links

Hello Osama Nasr,

please use the "reply all" to answer - that way the information
is automatically stored also in the bugs web site:

https://bugs.debian.org/927954

Kind regards,
Bernhard




Am 25.04.19 um 18:19 schrieb Osama Nasr:
> By the way, I've installed GIMP, although there was a bug
> (libmypaint-common #906144).
> I don't know if that have a relation or not.
> 
> On Thu, 25 Apr 2019, 6:01 pm Osama Nasr,  > wrote:
> 
> osama@debian:~$ konqueror
> Illegal icon group:  7
>  HTML :  "\n PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\"\n   
> \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\
> ">\n\n xmlns=\"http://www.w3.org/1999/xhtml\
> ">\n\n\t HREF=\"file:/usr/share/konqueror/about/\">\n\n   name=\"generator\" content=\n  \"HTML Tidy for Linux/x86 (vers 1st
> August 2004), see www.w3.org \" />\n\n   type=\"text/css\">\n  /* \"file:///usr/share/kf5/infopage/kde_infopage.css\"; /*
> kde_infopage.css */\n /* maybe @import \"kde_infopage_rtl.css\";
> */\n    @import \"konq.css\";\n  /*]]>*/\n  \n\n  Be
> free.\n\n\n\n  \n     id=\"headerL\">\n    \n\n     id=\"title\">\n  Konqueror \n    \n\n   
> \n  Be free. \n    \n 
> \n\n  \n  \n     id=\"barT\"> id=\"barTC\">\n    \n   id=\"barR\">\n     class=\"bar_text\">\n  Konqueror is a web browser, file
> manager and universal document viewer.\n  \n  \n     class=\"selected\">Starting
> Points\n     href=\"about:konqueror/intro\">Introduction\n     href=\"about:konqueror/tips\">Tips\n   
> Specifications\n  \n    \n  \n   
> \n     id=\"barBR\">\n  \n\n 
> \n   id=\"boxTL\"> id=\"boxTC\">\n    \n   id=\"boxR\">\n    \n\t align=\"center\">\n\t  \n\t  \n\t\t href=\"file:///home/osama\"> src=\"file:///usr/share/icons/Adwaita/48x48/places/user-home.png\"
> height=\"48\" width=\"48\" />\n\t\t\n\t   valign=\"bottom\" style=\"padding-bottom: 4px; padding-left:
> 6px;\">\n\t\t  Home
> FolderYour personal
> files\n\t\t\n     
> nowrap>\n\t\t\n\t\t  
>  
>  src=\"file:///usr/share/icons/Adwaita/48x48/status/user-trash-full.png\"
> height=\"48\" width=\"48\" />\n\t\t\n\t   valign=\"bottom\" style=\"padding-bottom: 4px; padding-left:
> 6px;\">\n\t\t    Trash id=\"subtext\">Browse and restore the trash\n\t\t 
> \n  \n   height=\".25em\">\n\t  \n\t\t\n\t\t   href=\"remote:/\"> src=\"file:///usr/share/icons/Adwaita/48x48/places/folder-remote.png\"
> height=\"48\" width=\"48\" />\n\t\t\n\t   valign=\"bottom\" style=\"padding-bottom: 4px; padding-left:
> 6px;\">\n\t\t  Network Folders id=\"subtext\">Shared files and folders\n\t\t 
> \n    \n\t\t   
> \n\t\t   width=\"48\" />\n    \n\t   valign=\"bottom\" style=\"padding-bottom: 4px; padding-left:
> 6px;\">\n\t\t   Bookmarks id=\"subtext\">Quick access to your
> bookmarks\n\t\t  \n\t\t\n   colspan=\"5\" height=\".25em\">\n\t\t   
> \n\t\t     href=\"about:konqueror/intro\"> src=\"file:///usr/share/icons/Adwaita/16x16/actions/go-next.png\"
> width=\"16\" height=\"16\">Next: An Introduction to
> Konqueror\n\t\t\t\n\n    \n 
> \n    \n     id=\"boxBL\"> id=\"boxBC\">\n  \n\n   id=\"footerL\"> id=\"footerR\">\n\n\n\n"
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine: Should no longer be called
> QPainter::begin: Paint device returned engine == 0, type: 1
> QPainter::translate: Painter not active
> QPainter::setClipRect: Painter not active
> QWidget::paintEngine: Should no longer be called
> QWidget::paintEngine:

Bug#927940: [Windows Subsystem for Linux] Applications cannot find libQt5Core.so.5

Control: retitle 927940 [Windows Subsystem for Linux] Applications cannot find 
libQt5Core.so.5


Hello Ryo,

> I encountered this problem with my WSL environment.Not quite the usual kernel 
> ... ;-)

A google search leads to this information [1]
and this bug [2].

There a workaround is provided by stripping
the section .note.ABI-tag from the file.

Kind regards,
Bernhard

[1] 
https://superuser.com/questions/1347723/arch-on-wsl-libqt5core-so-5-not-found-despite-being-installed
[2] https://github.com/Microsoft/WSL/issues/3023

Bug#927954: konqueror: Exit when opening http https' ftps' links

Hello Osama Nasr,
I tried to reproduce this issue in a minimal Unstable VM
with plasma-desktop and konqueror installed.
But could not reproduce it.

Could you try to start it from a konsole and forward
the output you get there after you hit this bug?

And have you configured something non-default in
  systemsettings - applications - Web Browser

Does it happen if you create a new user and login there too?

Kind regards,
Bernhard

Bug#927940: libqt5core5a: Applications rely on libQt5Core.so.5 cannot find libQt5Core.so.5

Hello Ryo IGARASHI,
I just trying to triage this bug, and have in a minimal
Buster amd64 VM installed just paraview installed and
could not reproduce the issue.

Could you please check the output in your system for any
differences to the outputs below:

dpkg -l | grep -E "paraview|libqt5core5a"
md5sum /usr/lib/x86_64-linux-gnu/libQt5Core.so.5* 
/usr/lib/paraview/paraview 
env | grep -E "^LD"

If they are equal, maybe the output of following call
could give a hint?

LD_DEBUG=libs ldd /usr/lib/paraview/paraview
LD_DEBUG=files ldd /usr/lib/paraview/paraview

Kind regards,
Bernahrd


PS.: I received following in my VM:

root@debian:~# dpkg -l | grep -E "paraview|libqt5core5a"
ii  libqt5core5a:amd64   5.11.3+dfsg1-1   amd64 
   Qt 5 core module
ii  paraview 5.4.1+dfsg4-3.1+b2   amd64 
   Parallel Visualization Application
ii  paraview-doc 5.4.1+dfsg4-3.1  all   
   Parallel Visualization Application. Comprehensive documentation
ii  paraview-python  5.4.1+dfsg4-3.1+b2   amd64 
   Parallel Visualization Application. python-support

root@debian:~# md5sum /usr/lib/x86_64-linux-gnu/libQt5Core.so.5* 
/usr/lib/paraview/paraview   
4acffc4ba6ef46508773294c59849734  /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
4acffc4ba6ef46508773294c59849734  
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.11
4acffc4ba6ef46508773294c59849734  
/usr/lib/x86_64-linux-gnu/libQt5Core.so.5.11.3
bd7619e0e0a906543e62ec737aa7ba10  /usr/lib/paraview/paraview

root@debian:~# env | grep -E "^LD"
root@debian:~#

Bug#925025: kate: Crash on file open dialog -> change directory

2019-04-22 Thread Bernhard Übelacker

Hello Dave Kitchen,
I am not involved in packaging kate, just trying to collect
some more information.

You wrote you open via nautilus - so you are running a
Gnome/Wayland desktop session?

I tried to reproduce the issue in a Gnome/Wayland session
but could not find a problem. In kate the native Gnome
file open dialog opens and let me select a file and open it.

>From my point of view kate should not be able to cause this,
even if it wants to.

Therefore I guess the maintainer will need more information on this.

When you open kate from a terminal could you collect its
output e.g.
   script -c "kate" -a kate_$(date +%Y-%m-%d_%H-%M-%S).log

Are there relevant entries in the system log e.g.
   journalctl --no-pager

Does this happen also with another or new user at this system?

Kind regards,
Bernhard

Bug#920139: sddm: GTK and GNOME: Applications won't launch due error of glib2

2019-04-08 Thread Bernhard Übelacker

Hello Adrian,

> I don't have any of those old GNOME applications installed, you mentioned.

Then these files should not be there I guess like e.g.:
/usr/share/glib-2.0/schemas/org.gnome.EasyTAG.gschema.xml

On a system where e.g. easytag is installed a 'dpkg -S' returns this:
$ dpkg -S /usr/share/glib-2.0/schemas/org.gnome.EasyTAG.gschema.xml
easytag: /usr/share/glib-2.0/schemas/org.gnome.EasyTAG.gschema.xml


However, if you suspect something of the environment may be wrong,
you could try this:

- login through sddm
env | sort > /home/user/env-enlightment.txt

- login through another login manager
env | sort > /home/user/env-other.txt

- compare:
diff -Nurp /home/user/env-other.txt /home/user/env-enlightment.txt

- then you may search if GSETTINGS_SCHEMA_DIR in the differences shown.

- or set the different variables one by one in a terminal by
export VAR=value
(call from there one of the crashing programs)
  and check if that changes behaviour.

Kind regards,
Bernhard

Bug#924507: ksplashqml: hits its hard timeout of 30 seconds

2019-03-22 Thread Bernhard Übelacker

Dear Maintainer,
so both issues have the same root and upstream commited this,
to open the dbus interface as soon as possible.

https://cgit.kde.org/plasma-workspace.git/commit/?id=9a4bb220c162f031b4d45ed04f9e156b6d0b852e

Kind regards,
Bernhard

Bug#924402: spyder3 Segmentation fault just after console comes up

2019-03-21 Thread Bernhard Übelacker

Control: tags 924402 + upstream


Hello Dimitry,
thanks for your quick response.

> Can I ask you to submit your patch upstream to [1], either as a pull
> request or as a simple issue?

I just created this pull request:

https://github.com/annulen/webkit/pull/803

Kind regards,
Bernhard

Bug#909366: libqt5webkit5: segfault in JSC::Structure::globalObject()

2019-03-21 Thread Bernhard Übelacker

Hello Everyone,
I think the same issue manifested in bug #924402.

To me it looks like the problem started when -fpermissive
got added to get the i386 build working again.

Without that a compiler error "invalid conversion" got hit,
because a derived class missed the attribute fastcall.

In bug #924402 a patch is attached that may solve this issue.
Maybe both bugs should be merged.

Kind regards,
Bernhard


https://bugs.debian.org/924402

Bug#924402: spyder3 Segmentation fault just after console comes up

2019-03-21 Thread Bernhard Übelacker

Control: tags 924402 + patch


Hello Everyone,
I continued to look into the "invalid conversion" and think
the problem is member JSImageConstructor::construct is
missing the __attribute__((fastcall)) aka JSC_HOST_CALL.

Attached patch to qtwebkit-opensource-src-5.212.0~alpha2
removes the -fpermissive and adds one attribute JSC_HOST_CALL.

With that I could build a package in a i386 qemu VM and
with that installed spyder3 did not crash anymore at startup.

Kind regards,
Bernhard
>From dc7e0a11c23ac81773b58dae59f4482af6a4c603 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Bernhard=20=C3=9Cbelacker?= 
Date: Thu, 21 Mar 2019 21:13:08 +0100
Subject: [PATCH] Bug#924402: Add missing attribute

---
 Source/WebCore/bindings/js/JSImageConstructor.cpp | 2 +-
 debian/rules  | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/Source/WebCore/bindings/js/JSImageConstructor.cpp b/Source/WebCore/bindings/js/JSImageConstructor.cpp
index 9236906..de553f5 100644
--- a/Source/WebCore/bindings/js/JSImageConstructor.cpp
+++ b/Source/WebCore/bindings/js/JSImageConstructor.cpp
@@ -44,7 +44,7 @@ template<> JSValue JSImageConstructor::prototypeForStructure(VM& vm, const JSDOM
 return JSHTMLElement::getConstructor(vm, );
 }
 
-template<> EncodedJSValue JSImageConstructor::construct(ExecState* state)
+template<> EncodedJSValue JSC_HOST_CALL JSImageConstructor::construct(ExecState* state)
 {
 JSImageConstructor* jsConstructor = jsCast(state->callee());
 Document* document = jsConstructor->document();
diff --git a/debian/rules b/debian/rules
index 365b37a..f428006 100755
--- a/debian/rules
+++ b/debian/rules
@@ -44,7 +44,6 @@ ifneq (,$(filter $(DEB_HOST_ARCH_CPU),ppc64el s390x))
 	EXTRA_CMAKE_ARGUMENTS += -DUSE_SYSTEM_MALLOC=ON
 endif
 
-DEB_CXXFLAGS_MAINT_APPEND += -fpermissive
 
 %:
 	dh $@ --buildsystem=cmake+ninja --with pkgkde_symbolshelper
-- 
2.20.1

Bug#923720: plasma-discover: Plasma-discover segfaults on Stretch

2019-03-18 Thread Bernhard Übelacker

reassign 923720 libappstreamqt2 0.10.6-2

Bug#923720: plasma-discover: Plasma-discover segfaults on Stretch

2019-03-18 Thread Bernhard Übelacker

Control: reassign 923720 libappstreamqt2/0.10.6-2
Control: affects 923720 plasma-discover
Control: fixed 923720 libappstreamqt2/0.11.3-1
Control: fixed 923720 plasma-discover/5.10.5-1
Control: tags 923720 + upstream fixed-upstream patch



Hello Everyone,
tried to get some more information from the backtrace.

I could not reproduce it but I think in this case method
AppStream::Pool::load got called with strerror being a null
pointer and for some reason the pool could not be loaded,
therefore line 77 was reached, trying to dereference strerror.

(gdb) list AppStream::Pool::load(QString*)   
71
72  bool Pool::load(QString* strerror)
73  {
74  g_autoptr(GError) error = nullptr;
75  bool ret = as_pool_load (d->m_pool, NULL, );
76  if (!ret && error) {
77  *strerror = QString::fromUtf8(error->message);  <<<
78  }
79  return ret;
80  }
81

This led to upstream fix in package appstream, available since 0.11.3: [1] [2]
Another fix was done in discover before, available since v5.10.5: [3] [4]

Therefore I assume this just affects Stretch.

Kind regards,
Bernhard


[1] https://github.com/ximion/appstream/pull/126
[2] 
https://github.com/ximion/appstream/commit/32f1445fd3f348598edd5e24e29ad3644c299639
[3] https://bugs.kde.org/show_bug.cgi?id=382916
[4] 
https://cgit.kde.org/discover.git/commit/?id=3a718124d45d60c49bb586e14d348f233178b34b

# Stretch amd64 qemu VM

apt update
apt dist-upgrade

apt install devscripts dpkg-dev systemd-coredump gdb xserver-xorg sddm 
plasma-desktop muon libappstreamqt2-dbgsym plasma-discover-dbgsym 
libglib2.0-0-dbg


systemctl start sddm



mkdir /tmp/source/appstream/orig -p
cd/tmp/source/appstream/orig
apt source appstream
cd



###

export DISPLAY=:0
# plasma-discover
gdb -q --args plasma-discover


set width 0
set pagination off
directory /tmp/source/appstream/orig/appstream-0.10.6
display/i $pc
break AppStream::Pool::load
y
run
disa 1.1
disa 1.3
disa 1.4
cont
bt






benutzer@debian:~$ gdb -q --args plasma-discover
Reading symbols from plasma-discover...Reading symbols from 
/usr/lib/debug/.build-id/8e/af6f71ec2d372a44c646c9eb0311f4bb45dd50.debug...done.
done.
(gdb) set width 0
(gdb) set pagination off
(gdb) directory /tmp/source/appstream/orig/appstream-0.10.6
Source directories searched: 
/tmp/source/appstream/orig/appstream-0.10.6:$cdir:$cwd
(gdb) display/i $pc
1: x/i $pc

(gdb) break AppStream::Pool::load
Function "AppStream::Pool::load" not defined.
Make breakpoint pending on future shared library load? (y or [n]) y
Breakpoint 1 (AppStream::Pool::load) pending.
(gdb) run
Starting program: /usr/bin/plasma-discover 
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7fffe66c2700 (LWP 2475)]
[New Thread 0x7fffe5a39700 (LWP 2476)]
[New Thread 0x7fffe5238700 (LWP 2478)]
[New Thread 0x7fffd278d700 (LWP 2479)]
[New Thread 0x7fffd1f8c700 (LWP 2480)]
[New Thread 0x7fffd178b700 (LWP 2481)]
[New Thread 0x7fffd0f8a700 (LWP 2482)]
[New Thread 0x7fffd0789700 (LWP 2483)]
[New Thread 0x7fffcff88700 (LWP 2484)]
[New Thread 0x7fffcf787700 (LWP 2485)]
[New Thread 0x7fffcef86700 (LWP 2486)]
[New Thread 0x7fffce01e700 (LWP 2487)]
[New Thread 0x7fffcd81c700 (LWP 2488)]
file:///usr/lib/x86_64-linux-gnu/qt5/qml/org/kde/kirigami/GlobalDrawer.qml:213:9:
 QML Flickable: Binding loop detected for property "contentWidth"
invalid kns backend! "" because: "Couldn't find knsrc file: comic.knsrc"

Thread 1 "plasma-discover" hit Breakpoint 1, 0x7fffc77e3cd0 in 
AppStream::Pool::load()@plt () from 
/usr/lib/x86_64-linux-gnu/qt5/plugins/discover/packagekit-backend.so
1: x/i $pc
=> 0x7fffc77e3cd0 <_ZN9AppStream4Pool4loadEv@plt>:  jmpq   *0x2305ca(%rip)  
  # 0x7fffc7a142a0
(gdb) disa 1.1
(gdb) disa 1.3
(gdb) disa 1.4
(gdb) cont
Continuing.

Thread 1 "plasma-discover" hit Breakpoint 1, AppStream::Pool::load 
(this=this@entry=0x55dcd3c8, strerror=strerror@entry=0x0) at 
./qt/pool.cpp:73
73  {
1: x/i $pc
=> 0x7fffc738d020 :push   %r12
(gdb) next
75  bool ret = as_pool_load (d->m_pool, NULL, );
1: x/i $pc
=> 0x7fffc738d038 : mov0x10(%rdi),%rax
(gdb) 
74  g_autoptr(GError) error = nullptr;
1: x/i $pc
=> 0x7fffc738d03c : movq   $0x0,0x10(%rsp)
(gdb) 
75  bool ret = as_pool_load (d->m_pool, NULL, );
1: x/i $pc
=> 0x7fffc738d045 : test   %rax,%rax
(gdb) 
76  if (!ret && error) {
1: x/i $pc
=> 0x7fffc738d065 : jne0x7fffc738d0b8 


(gdb) bt
#0  0x7fffc738d065 in AppStream::Pool::load(QString*) 
(this=this@entry=0x55dcd3c8, strerror=strerror@entry=0x0) at 
./qt/pool.cpp:76
#1  0x7fffc738d147 in AppStream::Pool::load() 
(this=this@entry=0x55dcd3c8) at ./qt/pool.cpp:69
#2  0x7fffc77ea7f3 in PackageKitBackend::PackageKitBackend(QObject*) 
(this=0x55dcd3b0, parent=) at

Bug#920139: sddm: GTK and GNOME: Applications won't launch due error of glib2

2019-03-16 Thread Bernhard Übelacker

Control: tags 920139 + unreproducible moreinfo


Hello Adrian Immanuel,
I am sorry for the late reply.

First a question: Do you still observe this fault?


I took a look today inside the md5sums you supplied
and tried to reproduce this setup inside a VM.

Unfortunately I found following old packages that are not
packaged in Debian anymore or maybe not installed in the
latest version:

  nautilus-pastebin: last built 2012, found just at snapshot.debian.org
  geoclue: last in jessie
  easytag: not in the lastest version?
  command-runner-applet: last in jessie
  gnome-search-tool: last in stretch
  sflphone-gnome: last in stretch

Can you please have a look at these packages, if you
really use them?
If not you might uninstall them and test if you still
can observe the not launching applications?

If you use synaptic, there should be a folder with outdated packages.


However, I tried to put your gsettings.compiled into that VM
and could login into GNOME and GNOME Classic sessions without
hitting that fault (traps: gnome-session-b...).

Kind regards,
Bernhard


# Stretch amd64 qemu VM 2019-03-16


apt update
apt dist-upgrade


#


# approx and sources.list to Buster snapshot from 20190210

debian-10-buster-snapshot.debian.org  
https://snapshot.debian.org/archive/debian/20190210T00Z/
debian-10-buster-debug-snapshot.debian.org  
https://snapshot.debian.org/archive/debian-debug/20190210T00Z/

deb [check-valid-until=no] 
http://192.168.178.25:/debian-10-buster-snapshot.debian.org/ buster main


#


apt update
apt dist-upgrade
reboot


apt install dpkg-dev devscripts systemd-coredump mc htop gdb strace 
xserver-xorg sddm
apt install gnome gnucash anjuta-extras tali telepathy-logger gthumb 
gnome-builder gnome-panel gjots2 gobby cinnamon-desktop-data 
ukui-settings-daemon seahorse-nautilus nemo gnome-photos dconf-editor easytag 
meld sound-juicer gnome-system-log virt-manager light-locker 
telepathy-mission-control-5 oregano gbonds-data gnomint liferea teg zapping 
accerciser blueman pulseaudio-module-gsettings almanah caribou-antler bijiben 
gnome-boxes gucharmap deja-dup gnome-dictionary empathy eog-plugins 
epiphany-browser font-manager geary gedit-latex-plugin 
gedit-source-code-browser-plugin ghex gimagereader gitg glabels gnome-nettool 
gnome-subtitles gnote gnumeric goobox gpaste gtranslator gwaei latexila 
gnome-multi-writer gnome-packagekit planner polari gnome-shell-mailnag 
gnome-shell-extension-weather gnome-shell-timer gnome-system-tools telegnome 
gtk-3-examples gtkhash libgtk-3-dev atril caja nemiver soundconverter gnomekiss 
ukwm mousepad



#



md5sum /usr/share/glib-2.0/schemas/* | sort -k 2



#


wget 
"https://bugs.debian.org/cgi-bin/bugreport.cgi?att=2;bug=920139;filename=usr-share-glib-2.0-schemas-md5sums.txt;msg=41;
 -O usr-share-glib-2.0-schemas-md5sums.txt
cat usr-share-glib-2.0-schemas-md5sums.txt | sort -k 2


#





nautilus-pastebin: last built 2012, found just at snapshot.debian.org
geoclue:  last in jessie
easytag not in the lastest version?
command-runner-applet:  last in jessie
gnome-search-tool: last in stretch
sflphone-gnome: last in stretch









wget 
"https://bugs.debian.org/cgi-bin/bugreport.cgi?att=1;bug=920139;filename=gschemas.compiled;msg=41;
 -O /usr/share/glib-2.0/schemas/gschemas.compiled
reboot

Bug#924507: ksplashqml: hits its hard timeout of 30 seconds

2019-03-13 Thread Bernhard Übelacker

Package: plasma-workspace
Version: 4:5.14.5.1-1
Severity: normal
Tags: patch upstream



Dear Maintainer,

I found that the splash screen takes longer than some month ago and did
some investigations that led to following two upstream bugs I reported.

Bug 405444 - ksplashqml hits its hard timeout of 30 seconds because of failing 
qdbus call kinit
Bug 405446 - ksplashqml hits its hard timeout of 30 seconds because setStage(6) 
is needed twice

Because of the near buster release they might be evaluated independent of
upstream for its use in buster to avoid bad user impressions.

The first one probably just manifests on faster systems when the
dbus interface of ksplashqml is not yet up.

The second one might need two setStage(6) calls to arrive when we
maybe just receive one.

Both changes reduced the time ksplashqml is shown at my
system from 30 to 5 seconds.

Kind regards,
Bernhard


https://bugs.kde.org/show_bug.cgi?id=405444
https://bugs.kde.org/show_bug.cgi?id=405446

Bug#910731: kaddressbook: crashes at startup

2019-03-05 Thread Bernhard Übelacker

Hello Sandro,

Am 05.03.19 um 00:43 schrieb Sandro Knauß:
> Btw. please use the bugnumber-qu...@bugs.debian.org rarely, because than 
> maintainers like me don't gets any mails, that's why I missed your 
> conversation.

Sorry for using -quiet - Pino Toscano made me already
aware to avoid -quiet.

One other thing that may be related is that I found using
BUGNUMBER-submitter@ very convenient, but I have the
impression that then the submitter gets by default
the BUGNUMBER-quiet@ as answer to address.

Kind regards,
Bernhard

Bug#920139: sddm: GTK and GNOME: Applications won't launch due error of glib2

2019-02-09 Thread Bernhard Übelacker

Hello Adrian,
maybe one of the installed source files lead
to a bad gschemas.compiled file?

Could you create a file with the md5sums of all
the xml files that seem to get combined into that
gschemas.compiled file.

 md5sum /usr/share/glib-2.0/schemas/* > 
~/usr-share-glib-2.0-schemas-md5sums.txt

Also you may attach the gschemas.compiled too?

Kind regards,
Bernhard

Bug#920139: sddm: GTK and GNOME: Applications won't launch due error of glib2

2019-02-04 Thread Bernhard Übelacker

Control: tags 920139 + moreinfo


Hello Adrian,

Am 03.02.19 um 09:24 schrieb Adrian Immanuel Kiess:
> The bug is, like I see it, that the applications cannot find the
> gsettings schema directory.

>From my point of view it might be more the file gschemas.compiled
inside that directory. Does that exist on your system?

> When setting export GSETTINGS_SCHEMA_DIR="/usr/share/glib-2.0/schemas/" 
> in my .xinitrc I can launch GTK and GNOME applications when the
> xsession ist started with startx.
> 
> Therefor setting GSETTINGS_SCHEMA_DIR in /etc/environment maybe fixes 
> the issue, which I have not tried yet.

My previous test was inside a minimal buster amd64 VM where I installed
just "systemd-coredump xserver-xorg sddm gnome-session" and there I can
login in sddm to a "GNOME on Xorg" session without showing that problem.
I searched that VM and could find no file setting that environment.

I tried renaming that file gschemas.compiled and setting the environment
like you did - but I still got the trap.

Therefore you might also install a coredump collector
like systemd-coredump.
That way after such an unsuccessful logon attempt you can list with:

coredumpctl list

And produce an exact backtrace in which function that error is
thrown by this command:

coredumpctl gdb [PID]
bt

Best would be if debug symbol packages
gnome-session-bin-dbgsym libglib2.0-0-dbgsym
are installed like described in [1].

> Shall I resubmit the bug against gsettings-desktop-schemas package?

If you mean with resumit to create a new bug, that should not be needed
as this bug can be reassigned to another package too.
Which version of package gsettings-desktop-schemas have you installed?

dpkg -l | grep gsettings-desktop-schemas

Kind regards,
Bernhard

[1] https://wiki.debian.org/HowToGetABacktrace#Installing_the_debugging_symbols

Bug#908130: gcompris-qt: Gcompris is crashing after making first move in "End of chess game"

2019-01-31 Thread Bernhard Übelacker

Hello Rokas,
I just tried to collect some more information and
could reproduce the crash with Qt 5.11.1+dfsg-8.

With version 5.11.2+dfsg-3 and current 5.11.3+dfsg-2
I was not able anymore to reproduce.

Can you confirm that the bug is not visible
anymore on your installation?

Kind regards,
Bernhard


(With buster/testing as of date 20181001)
(gdb) bt
#0  0x7f341f02c0a6 in QV4::Value::isObject (this=0x7f3415dca9c8) at 
../../include/QtQml/5.11.1/QtQml/private/../../../../../src/qml/memory/qv4heap_p.h:101
#1  QV4::Value::objectValue (this=0x7f3415dca9c8) at jsruntime/qv4value_p.h:368
#2  QV4::Value::as (this=0x7f3415dca9c8) at 
jsruntime/qv4managed_p.h:254
#3  QV4::Scoped::Scoped (v=..., scope=..., 
this=) at jsruntime/qv4scopedvalue_p.h:223
#4  QV4::Runtime::method_loadProperty (engine=0x5608459f1fd0, object=..., 
nameIndex=) at jsruntime/qv4runtime.cpp:726
#5  0x7f33e9f841b8 in ?? ()
#6  0x7f3415dca8f0 in ?? ()
#7  0x7f3415dca908 in ?? ()
#8  0x5608459f1fd0 in ?? ()
#9  0x in ?? ()

# Stretch amd64 qemu VM 2019-01-31


approx:
debian-10-buster-snapshot.debian.org  
https://snapshot.debian.org/archive/debian/20180901T00Z/
debian-10-buster-debug-snapshot.debian.org  
https://snapshot.debian.org/archive/debian-debug/20180901T00Z/

# snapshot
deb [check-valid-until=no] 
http://192.168.178.25:/debian-10-buster-snapshot.debian.org/ buster main
#deb-src [check-valid-until=no] 
http://192.168.178.25:/debian-10-buster-snapshot.debian.org/ buster main

#deb [check-valid-until=no] 
http://192.168.178.25:/debian-10-buster-debug-snapshot.debian.org/ 
buster-debug main




apt update
apt dist-upgrade

apt install systemd-coredump xserver-xorg lightdm openbox gcompris-qt
apt install mc htop gdb strace dpkg-dev devscripts
apt install gcompris-qt-dbgsym libqt5qml5-dbgsym libqt5core5a-dbgsym 
libglib2.0-0-dbgsym


reboot




[  418.415478] gcompris-qt[15563]: segfault at c ip 7f579cd530a6 sp 
7ffcb3f17630 error 4 in libQt5Qml.so.5.11.1[7f579cbfd000+298000]

root@debian:~# coredumpctl list
TIMEPID   UID   GID SIG COREFILE  EXE
Thu 2019-01-31 17:48:50 CET   15563  1000  1000  11 present   
/usr/games/gcompris-qt

root@debian:~# coredumpctl gdb 15563
   PID: 15563 (gcompris-qt)
   UID: 1000 (benutzer)
   GID: 1000 (benutzer)
Signal: 11 (SEGV)
 Timestamp: Thu 2019-01-31 17:48:49 CET (36s ago)
  Command Line: gcompris-qt
Executable: /usr/games/gcompris-qt
 Control Group: /user.slice/user-1000.slice/session-4.scope
  Unit: session-4.scope
 Slice: user-1000.slice
   Session: 4
 Owner UID: 1000 (benutzer)
   Boot ID: 47f6f3a26a5049c9a0846c2cb1d7d083
Machine ID: 9e5901179cfe4b73bc18669e6a6e0ab9
  Hostname: debian
   Storage: 
/var/lib/systemd/coredump/core.gcompris-qt.1000.47f6f3a26a5049c9a0846c2cb1d7d083.15563.154895332900.lz4
   Message: Process 15563 (gcompris-qt) of user 1000 dumped core.


Stack trace of thread 15563:
#0  0x7f579cd530a6 
_ZN3QV47Runtime19method_loadPropertyEPNS_15ExecutionEngineERKNS_5ValueEi 
(libQt5Qml.so.5)
#1  0x7f575aa601b8 n/a (n/a)
#2  0x7f579cd44fd1 n/a (libQt5Qml.so.5)
#3  0x7f579cd4f9bb 
_ZN3QV47Runtime15method_callNameEPNS_15ExecutionEngineEiPNS_5ValueEi 
(libQt5Qml.so.5)
#4  0x7f575aa5c0a0 n/a (n/a)
#5  0x7f579cd44fd1 n/a (libQt5Qml.so.5)
#6  0x7f579cd4f161 
_ZN3QV47Runtime16method_callValueEPNS_15ExecutionEngineERKNS_5ValueEPS3_i 
(libQt5Qml.so.5)
#7  0x7f575aa5c6a9 n/a (n/a)
#8  0x7f579cd44fd1 n/a (libQt5Qml.so.5)
#9  0x7f579cd4f161 
_ZN3QV47Runtime16method_callValueEPNS_15ExecutionEngineERKNS_5ValueEPS3_i 
(libQt5Qml.so.5)
#10 0x7f575aa5c6a9 n/a (n/a)
#11 0x7f579cd44fd1 n/a (libQt5Qml.so.5)
#12 0x7f579cd53665 
_ZN3QV47Runtime19method_callPropertyEPNS_15ExecutionEngineEPNS_5ValueEiS4_i 
(libQt5Qml.so.5)
#13 0x7f579cd46bb6 n/a (libQt5Qml.so.5)
#14 0x7f579cd4f9bb 
_ZN3QV47Runtime15method_callNameEPNS_15ExecutionEngineEiPNS_5ValueEi 
(libQt5Qml.so.5)
#15 0x7f579cd46a2c n/a (libQt5Qml.so.5)
#16 0x7f579cd53665 
_ZN3QV47Runtime19method_callPropertyEPNS_15ExecutionEngineEPNS_5ValueEiS4_i 
(libQt5Qml.so.5)
#17 0x7f579cd46bb6 n/a (libQt5Qml.so.5)
#18 0x7f579cd4f9bb 
_ZN3QV47Runtime15method_callNameEPNS_15ExecutionEngineEiPNS_5ValueEi 
(libQt5Qml.so.5)
#19 0x7f579cd46a2c n/a (libQt5Qml.so.5)
#20 0x7f579cd53665 
_ZN3QV47Runtime19method_callPropertyEPNS_15ExecutionEngineEPNS_5ValueEiS4_i 
(libQt5Qml.so.5)
#21 0x7f579cd46bb6 n/a (libQt5Qml.so.5)

Bug#920139: sddm: GTK and GNOME: Applications won't launch due error of glib2

2019-01-31 Thread Bernhard Übelacker

Hello Adrian Immanuel Kiess,
I tried to search the net about this message:
"GLib-GIO-ERROR: No GSettings schemas are installed on the system"

And found results pointing to .../share/glib-2.0/schemas/.
Therefore renamed following file:
/usr/share/glib-2.0/schemas/gschemas.compiled

And received on the next logon attempt a similar trap
message as you did.

Therefore I assume this is based on this file missing or corrupt,
and unrelated to the sddm package, but not sure to which package
to reassign.

Are all packages in a configured state? Does following do any changes:
apt install -f

I could trigger a recreation when the file was missing by e.g.:
apt install --reinstall yelp

Were there any faults when installing/upgrading
some gnome packages lately?
Could you inspect this file on your system
or let it be recreated?

Kind regards,
Bernhard



root@debian:~# journalctl | grep traps -B4
Jan 31 16:30:09 debian gnome-session[1233]: gnome-session-binary[1233]: 
GLib-GIO-ERROR: No GSettings schemas are installed on the system
Jan 31 16:30:09 debian gnome-session[1233]: aborting...
Jan 31 16:30:09 debian gnome-session-binary[1233]: GLib-GIO-ERROR: No GSettings 
schemas are installed on the system
   aborting...
Jan 31 16:30:09 debian kernel: traps: gnome-session-b[1233] trap int3 
ip:7ff639d19be5 sp:7fff4eda4490 error:0 in 
libglib-2.0.so.0.5800.2[7ff639ce1000+7e000]



(gdb) bt
#0  0x7ff639d19be5 in _g_log_abort (breakpoint=1) at 
../../../glib/gmessages.c:554
#1  0x7ff639d1aefd in g_logv (log_domain=0x7ff639f6f4e7 "GLib-GIO", 
log_level=G_LOG_LEVEL_ERROR, format=, 
args=args@entry=0x7fff4eda4560) at ../../../glib/gmessages.c:1371
#2  0x7ff639d1b0cf in g_log (log_domain=log_domain@entry=0x7ff639f6f4e7 
"GLib-GIO", log_level=log_level@entry=G_LOG_LEVEL_ERROR, 
format=format@entry=0x7ff639f8e0d8 "No GSettings schemas are installed on the 
system") at ../../../glib/gmessages.c:1413
#3  0x7ff639f1f171 in g_settings_set_property (object=, 
prop_id=, value=0x7fff4eda46c0, pspec=) at 
../../../gio/gsettings.c:585
#4  0x7ff639dfb8c9 in object_set_property (nqueue=0x561977752880, 
value=, pspec=0x561977747e50, object=0x561977743210) at 
../../../gobject/gobject.c:1447
#5  g_object_new_internal (class=class@entry=0x561977769520, 
params=params@entry=0x7fff4eda47f0, n_params=n_params@entry=1) at 
../../../gobject/gobject.c:1839
#6  0x7ff639dfd3a4 in g_object_new_valist (object_type=, 
first_property_name=first_property_name@entry=0x7ff639f8dbf3 "schema-id", 
var_args=var_args@entry=0x7fff4eda4940) at ../../../gobject/gobject.c:2128
#7  0x7ff639dfd6d9 in g_object_new (object_type=, 
first_property_name=first_property_name@entry=0x7ff639f8dbf3 "schema-id") at 
../../../gobject/gobject.c:1648
#8  0x7ff639f1f1c5 in g_settings_new 
(schema_id=schema_id@entry=0x561975bb51c5 "org.gnome.SessionManager") at 
../../../gio/gsettings.c:965
#9  0x561975ba7fec in gsm_manager_init (manager=0x56197776a100) at 
../gnome-session/gsm-manager.c:3231
#10 0x7ff639e190b7 in g_type_create_instance (type=) at 
../../../gobject/gtype.c:1864
#11 0x7ff639dfafe8 in g_object_constructor (type=, 
n_construct_properties=3, construct_params=0x561977767c50) at 
../../../gobject/gobject.c:2154
#12 0x561975ba8458 in gsm_manager_constructor (type=94667378430464, 
n_construct_properties=3, construct_properties=0x561977767c50) at 
../gnome-session/gsm-manager.c:2158
#13 0x7ff639dfb34b in g_object_new_with_custom_constructor (n_params=2, 
params=0x7fff4eda4da0, class=0x561977769270) at ../../../gobject/gobject.c:1723
#14 g_object_new_internal (class=class@entry=0x561977769270, 
params=params@entry=0x7fff4eda4da0, n_params=n_params@entry=2) at 
../../../gobject/gobject.c:1803
#15 0x7ff639dfd3a4 in g_object_new_valist (object_type=, 
first_property_name=first_property_name@entry=0x561975bb7d25 "client-store", 
var_args=var_args@entry=0x7fff4eda4ef0) at ../../../gobject/gobject.c:2128
#16 0x7ff639dfd6d9 in g_object_new (object_type=, 
first_property_name=first_property_name@entry=0x561975bb7d25 "client-store") at 
../../../gobject/gobject.c:1648
#17 0x561975ba9393 in gsm_manager_new (client_store=0x561977760a80, 
failsafe=0) at ../gnome-session/gsm-manager.c:3283
#18 0x561975bb2bf7 in create_manager () at ../gnome-session/main.c:140
#19 on_bus_acquired (connection=, name=, 
data=) at ../gnome-session/main.c:164
#20 0x7ff639f3aff4 in connection_get_cb (source_object=, 
res=0x7ff628004500, user_data=0x561977768570) at 
../../../gio/gdbusnameowning.c:487
#21 0x7ff639edc5b9 in g_task_return_now (task=0x7ff628004500) at 
../../../gio/gtask.c:1148
#22 0x7ff639edd036 in g_task_return (task=0x7ff628004500, type=) at ../../../gio/gtask.c:1206
#23 0x7ff639f2d412 in bus_get_async_initable_cb 
(source_object=0x56197774e260, res=0x5619777671f0, 
user_data=user_data@entry=0x7ff628004500) at ../../../gio/gdbusconnection.c:7314
#24

Bug#919966: akregator: symbol lookup error: /usr/lib/x86_64-linux-gnu/libKF5NewStuff.so.5: undefined symbol: _ZN7KNSCore6Engine15signalErrorCodeERKNS_9ErrorCodeERK7QStringRK8QVariant

2019-01-21 Thread Bernhard Übelacker

Hello Francois Marier,
this looks like a duplicate of bug:
  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=919765

Kind regards,
Bernhard

Bug#896670: kde: Plasma menu text query kills plasma and leaves kwin zombie

Hello André Isidoro Fernandes Esteves,

one more possibly useful thing to consider:

If you could install a core dump collector like package
systemd-coredump, then the system would store a file of
each crash for later inspections.
Also it would already print a backtrace in the system journal.

When then a crash happens these command might be used:
coredumpctl list
coredumpctl gdb 
backtrace

Kind regards,
Bernhard

Bug#904094: plasma-workspace: plasmashell crash when search in the menu bar writing more than 2 characters

Control: tags -1 + moreinfo

Hello Daniel Sutil,
just got to this bug report.
Do you still get this crash?

If it got solved for you, you could close this bug by sending
an email to 904094-d...@bugs.debian.org

If you still get it, you could follow these steps:
- install package gdb
- open a konsole
gdb -q -ex 'set width 0' -ex 'set pagination off' -ex 'cont' -ex 'thread 
apply all backtrace full' -ex 'detach' -ex 'quit' -ex 'info share' --pid 
$(pidof plasmashell) 2>&1 | tee -a ~/904094_gdb_plasmashell_$(date 
+%Y-%m-%d_%H-%M-%S).txt
- trigger the crash
- that command should create in your home directory a file
  containing backtraces of the crash of your plasmashell process.
- forward that file to this bug.

Kind regards,
Bernhard

Bug#857264: plasma-workspace: xembedsniproxy often segfaults

Control: forward -1 https://bugs.kde.org/show_bug.cgi?id=359664
Control: tags -1 upstream fixed-upstream
Control: fixed -1 plasma-workspace/4:5.12.0-2


Dear Maintainer, hello Michael Meier,
from the segfault line I would suspect that this
crash happened in sniproxy.cpp, line 273.

There image may contain NULL that gets unconditionally
dereferenced.

Upstream would have fixed it in [1], contained first in 5.12.
Also applied in branch 5.8 [2], contained first in 5.8.9.
Upstream bug report would be [3].

This bug may still show up in current stretch version.

Kind regards,
Bernhard


[1] 
https://cgit.kde.org/plasma-workspace.git/commit/?id=e2b7c395ecdb660b7bec960f3c938fba175ca4f8
[2] 
https://cgit.kde.org/plasma-workspace.git/commit/?h=Plasma/5.8=12e3568042fb365aad3eccf2fefa58bbeb065210
[3] https://bugs.kde.org/show_bug.cgi?id=359664

---

   0x5556022d <+109>:   callq  0x9bf0 
   0x55560256 <+150>:   mov%rax,%rbx

271
272 // Don't hook up cleanup yet, we may use a different QImage after 
all
273 QImage naiveConversion = QImage(image->data, image->width, 
image->height, QImage::Format_ARGB32);
   0x55560232 <+114>:   movzwl 0x2(%rax),%ecx   

   0x55560236 <+118>:   movzwl (%rax),%edx

---

259 QImage SNIProxy::getImageNonComposite() const
260 {
...
270 xcb_image_t *image = xcb_image_get(c, m_windowId, 0, 0, 
geom->width, geom->height, 0x, XCB_IMAGE_FORMAT_Z_PIXMAP);
271
272 // Don't hook up cleanup yet, we may use a different QImage after 
all
273 QImage naiveConversion = QImage(image->data, image->width, 
image->height, QImage::Format_ARGB32);

# Jessie amd64 qemu VM

apt update
apt dist-upgrade


approx:
debian-9-stretch-snapshot.debian.org
https://snapshot.debian.org/archive/debian/20170310T00Z/
debian-9-stretch-debug-snapshot.debian.org  
https://snapshot.debian.org/archive/debian-debug/20170310T00Z/


sources.list:
# snapshot
deb [check-valid-until=no] 
http://192.168.178.25:/debian-9-stretch-snapshot.debian.org/ stretch main
deb-src [check-valid-until=no] 
http://192.168.178.25:/debian-9-stretch-snapshot.debian.org/ stretch main
deb [check-valid-until=no] 
http://192.168.178.25:/debian-9-stretch-debug-snapshot.debian.org/ 
stretch-debug main

apt-get -o Acquire::Check-Valid-Until=false -o Acquire::Languages=none update
apt dist-upgrade
reboot


apt install xserver-xorg sddm kde-plasma-desktop binutils gdb elfutils 
plasma-workspace-dbgsym
apt install dpkg-dev devscripts




mkdir source/plasma-workspace/orig -p
cdsource/plasma-workspace/orig
apt source plasma-workspace
cd


benutzer@debian:~$ dpkg -l | grep plasma-workspace
ii  plasma-workspace  4:5.8.4-1 
 amd64Plasma Workspace for KF5
ii  plasma-workspace-dbgsym   4:5.8.4-1 
 amd64Debug symbols for plasma-workspace


##


From https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=857264:
[   49.752174] xembedsniproxy[5458]: segfault at 2 ip 561b56ae2232 sp 
7b32c330 error 4 in xembedsniproxy[561b56ad6000+14000]



https://www.enodev.fr/posts/decode-segfault-errors-in-dmesg.html
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/arch/x86/mm/fault.c?h=linux-4.9.y#n31


--


/*
 * Page fault error code bits:
 *
 *   bit 0 ==0: no page found   1: protection fault
 *   bit 1 ==0: read access 1: write access
 *   bit 2 ==0: kernel-mode access  1: user-mode access
 *   bit 3 ==   1: use of reserved bit detected
 *   bit 4 ==   1: fault was an instruction fetch
 *   bit 5 ==   1: protection keys block access
 */
enum x86_pf_error_code {

PF_PROT =   1 << 0,
PF_WRITE=   1 << 1,
PF_USER =   1 << 2,
PF_RSVD =   1 << 3,
PF_INSTR=   1 << 4,
PF_PK   =   1 << 5,
};


"error 4" == 0b100

bit 0 == 0: no page found
bit 1 == 0: read access
bit 2 == 1: user-mode access


--



# Create a 'unstripped' binary

benutzer@debian:~$ gdb -q -ex quit --args /usr/bin/xembedsniproxy
Reading symbols from /usr/bin/xembedsniproxy...Reading symbols from 
/usr/lib/debug/.build-id/78/036ce04f32f09dd53a566ea3e172453fd64037.debug...done.
done.


cp -a /usr/bin/xembedsniproxy /tmp/xembedsniproxy
cp -a /usr/lib/debug/.build-id/78/036ce04f32f09dd53a566ea3e172453fd64037.debug 
/tmp/036ce04f32f09dd53a566ea3e172453fd64037.debug
eu-unstrip /tmp/xembedsniproxy /tmp/036ce04f32f09dd53a566ea3e172453fd64037.debug


benutzer@debian:~$ objdump --disassemble 
/tmp/036ce04f32f09dd53a566ea3e172453fd64037.debug |

Bug#894748: plasma-workspace: Plasmashell crashes when notification shown