[valgrind] [Bug 489338] arm64: Instruction fcvtas should round 322.5 to 323, but result is 322.
https://bugs.kde.org/show_bug.cgi?id=489338 --- Comment #1 from Bernhard Übelacker --- Created attachment 171102 --> https://bugs.kde.org/attachment.cgi?id=171102&action=edit 0001-arm64-Fix-fcvtas-instruction.patch I tried to look into it and I think the issue is the `fcvtas` gets stored in the intermediate representation with Irrm_NEAREST. When this gets translated back to the native instruction i guess it results in a `fcvtns`. Attached patch tries to preserve this difference by using Irrm_NEAREST_TIE_AWAY_0, at least the reproducer shows with it no longer a difference. -- You are receiving this mail because: You are watching all bug changes.
[valgrind] [Bug 489338] New: arm64: Instruction fcvtas should round 322.5 to 323, but result is 322.
https://bugs.kde.org/show_bug.cgi?id=489338 Bug ID: 489338 Summary: arm64: Instruction fcvtas should round 322.5 to 323, but result is 322. Classification: Developer tools Product: valgrind Version: 3.20.0 Platform: Debian unstable OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: vex Assignee: jsew...@acm.org Reporter: bernha...@mailbox.org Target Milestone: --- Created attachment 171101 --> https://bugs.kde.org/attachment.cgi?id=171101&action=edit Minimal reproducer: fp-valgrind-test.c Hello, I was investigating a debian bug, where the continuous integration failed because it receives a different value when running with valgrind. [1071656] I could track it down to a fcvtas instruction. => 0x4948 : fcvtas w5, d8 (With $d8=322.5) With plain GDB: (gdb) print $w5 $2 = 323 With valgrind+GDB: (gdb) print $w5 $2 = 322 Attached is a minimal reproducer. [1071656] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1071656 -- You are receiving this mail because: You are watching all bug changes.
[valgrind] [Bug 485134] New: vgcore files unusable in gdb / does not load debug symbols
https://bugs.kde.org/show_bug.cgi?id=485134 Bug ID: 485134 Summary: vgcore files unusable in gdb / does not load debug symbols Classification: Developer tools Product: valgrind Version: 3.16.1 Platform: Debian stable OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: jsew...@acm.org Reporter: bernha...@mailbox.org Target Milestone: --- Created attachment 168219 --> https://bugs.kde.org/attachment.cgi?id=168219&action=edit add_auxiliary_vector_to_vgcore_files.patch It seems gcc defaults now quite some time to create pie executables. And gdb seems to rely for core files of such executables to contain an auxv section. This was found in this Debian bug: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=993018 There are more details about reproducing it. I investigated back then and came up with an patch which makes valgrind write the auxv section into the core file. Unfortuantely forgot about it and now trying to upstream this. Would such a patch be acceptable? If yes, how should I proceed? Patch is attached and it seems still to apply to current valgrind 3.20.0 in Debian Trixie/testing. -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 473602] Konsole crashing when coming back from sleep (upon switching computers on KVM)
https://bugs.kde.org/show_bug.cgi?id=473602
Bernhard Übelacker changed:
What|Removed |Added
Product|konsole |kwin
Assignee|konsole-de...@kde.org |kwin-bugs-n...@kde.org
Version|23.04.3 |5.27.5
Component|general |general
--- Comment #3 from Bernhard Übelacker ---
Hello, I hope it is ok to change this bug to kwin.
Bottom line is:
kwin_x11 creates from this geometry {xp = 4, yp = 28, w = -8, h = -32} an event
with dimensions 65528 x 65504.
Would following modification be acceptable?
> --- kwin-5.27.5.orig/src/x11window.cpp
> +++ kwin-5.27.5/src/x11window.cpp
> @@ -3786,8 +3786,8 @@ void X11Window::sendSyntheticConfigureNo
> -u.event.width = Xcb::toXNative(m_clientGeometry.width());
> -u.event.height = Xcb::toXNative(m_clientGeometry.height());
> +u.event.width = Xcb::toXNative(std::max(0.0, m_clientGeometry.width()));
> +u.event.height = Xcb::toXNative(std::max(0.0,
> m_clientGeometry.height()));
--
You are receiving this mail because:
You are watching all bug changes.
[kde] [Bug 461723] konsole (or other applications) crash at disabling second screen
https://bugs.kde.org/show_bug.cgi?id=461723 --- Comment #13 from Bernhard Übelacker --- Hello, unfortunately this Qt upstream modifiction seems not to avoid this issue. Further debugging leads to kwin_x11 as causing this, details added in https://bugs.kde.org/show_bug.cgi?id=473602 -- You are receiving this mail because: You are watching all bug changes.
[konsole] [Bug 463475] Konsole crashes when monitor goes to sleep
https://bugs.kde.org/show_bug.cgi?id=463475 --- Comment #29 from Bernhard Übelacker --- Added some details about the possible source of this problematic huge dimensions to https://bugs.kde.org/show_bug.cgi?id=473602. -- You are receiving this mail because: You are watching all bug changes.
[konsole] [Bug 473602] Konsole crashing when coming back from sleep (upon switching computers on KVM)
https://bugs.kde.org/show_bug.cgi?id=473602
Bernhard Übelacker changed:
What|Removed |Added
CC||bernha...@mailbox.org
--- Comment #1 from Bernhard Übelacker ---
Hello,
I receive this fault every now and then.
I just found fullscreen processes crashing, therefore as a workaround it helps
to not make windows fullscreen,
instead leave it a little smaller.
I received these crashes either when enabling or disabling one of my two
screens, or when screens wakeup after energy saving.
My setup are two screens of different brand, therefore maybe different reaction
times. Different size but same resolution.
The right screen is the main one, the left currently disabled.
I report this issue some time ago in #461723 and brought this to the attention
of Qt developers
in https://bugreports.qt.io/browse/QTBUG-109226 .
They fixed some flaw, but this issue unfortunately remained active.
I found now again some time for a deeper look and found following before the
crash:
(with modified Qt library to output details of XCB_CONFIGURE_NOTIFY)
> export QT_LOGGING_RULES=qt.qpa.*=true
> gwenview picture.png
> ...
> qt.qpa.events: Event | XCB_CONFIGURE_NOTIFY(22) | sequence: 961 |
> 4-28-65528-65504
> qt.qpa.xcb: [ QWidgetWindow(0x556124696730, name="MainWindow#1Window") ]
> creating shared memory 17169384448 bytes for QSize(65528, 65504) depth 24
> bits 32
> qt.qpa.xcb: xcb_shm_create_segment() can't be called for size 17169384448,
> maximumallowed size is 4294967295
> ...
> 10 -- signal=11
> ...
> KCrash: Attempting to start /usr/lib/x86_64-linux-gnu/libexec/drkonqi
This huge resolution of 65528 width and 65504 height I could follow back to an
event XCB_CONFIGURE_NOTIFY
read from the X11 socket /tmp/.X11-unix/X0 in _xcb_in_read.
I suspected this might originate from kwin_x11 - and there function
X11Window::sendSyntheticConfigureNotify could fit.
And below I found m_clientGeometry got negative values for width and heigth,
which looks like it gets interpreted as a 16-bit unsigned int. (-32 == 0xffe0
== 65504).
Should therefore this issue be reassigned to kwin?
> Thread 1 hit Breakpoint 3, KWin::X11Window::sendSyntheticConfigureNotify
> (this=this@entry=0x557d9d033db0) at ./src/x11window.cpp:3791
> 3791u.event.border_width = 0;
> (rr) bt
> #0 KWin::X11Window::sendSyntheticConfigureNotify()
> (this=this@entry=0x557d9d033db0) at ./src/x11window.cpp:3791
> #1 0x7f9ffab7b961 in KWin::X11Window::updateServerGeometry()
> (this=this@entry=0x557d9d033db0) at ./src/x11window.cpp:4270
> #2 0x7f9ffab7fdf9 in KWin::X11Window::moveResizeInternal(QRectF const&,
> KWin::Window::MoveResizeMode) (this=0x557d9d033db0, rect=,
> mode=KWin::Window::MoveResizeMode::MoveResize) at ./src/x11window.cpp:4215
> #3 0x7f9ffaa8626c in KWin::PlacementTracker::restore(QString const&)
> (this=this@entry=0x7f9fd4011660, key=...) at ./src/placementtracker.cpp:110
> #4 0x7f9ffab6165a in KWin::Workspace::desktopResized()
> (this=this@entry=0x557d95264b30) at ./src/workspace.cpp:2251
> ...
> (rr) print this->m_clientGeometry
> $5 = {xp = 4, yp = 28, w = -8, h = -32}
> (rr) print u.event.width
> $6 = 65528
> (rr) print u.event.height
> $7 = 65504
> (rr) list
> 3784u.event.response_type = XCB_CONFIGURE_NOTIFY;
> 3785u.event.event = window();
> 3786u.event.window = window();
> 3787u.event.x = Xcb::toXNative(m_clientGeometry.x());
> 3788u.event.y = Xcb::toXNative(m_clientGeometry.y());
> 3789u.event.width = Xcb::toXNative(m_clientGeometry.width());
> 3790u.event.height = Xcb::toXNative(m_clientGeometry.height());
> 3791u.event.border_width = 0;
> 3792u.event.above_sibling = XCB_WINDOW_NONE;
> 3793u.event.override_redirect = 0;
> 3794xcb_send_event(kwinApp()->x11Connection(), true, c.event,
> XCB_EVENT_MASK_STRUCTURE_NOTIFY, reinterpret_cast(&u));
I suspect #464967 and #463475 are open duplicates to this bug.
Another small note: also non-Qt applications like gitk are sometimes affected
and crash with an X error.
--
You are receiving this mail because:
You are watching all bug changes.
[konsole] [Bug 463475] Konsole crashes when monitor goes to sleep
https://bugs.kde.org/show_bug.cgi?id=463475
--- Comment #28 from Bernhard Übelacker ---
Some short additions:
- I am getting this with X server.
The resize event way up in the stack looks like
the new width should change from 1920 to 65528 and the height from 996 to
65504.
With 4 bytes per pixel this would explain the later `dstBytesPerLine=262112`.
Having a buffer with this dimensions in memory would need ~16 GB.
#8 copy_unswapped (..., dstBytesPerLine=262112, ...) at
./src/plugins/platforms/xcb/qxcbbackingstore.cpp:547
...
#18 0x7fa7f3bbd421 in QWidgetWindow::handleResizeEvent () at
kernel/qwidgetwindow.cpp:841
(gdb) print event->s
$4 = {wd = 65528, ht = 65504}
(gdb) print event->olds
$5 = {wd = 1920, ht = 996}
--
You are receiving this mail because:
You are watching all bug changes.
[konsole] [Bug 463475] Konsole crashes when monitor goes to sleep
https://bugs.kde.org/show_bug.cgi?id=463475 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #27 from Bernhard Übelacker --- The same crash I reported already in: https://bugs.kde.org/show_bug.cgi?id=461723 Which led to this Qt bug report: https://bugreports.qt.io/browse/QTBUG-109226 Unfortunately this change in Qt did not reach the 5.15 branch? https://code.qt.io/cgit/qt/qtbase.git/commit/src/plugins/platforms/xcb/qxcbwindow.cpp?id=6a3627b6c5aa5109a80024f3d7b0f938504f7ffe But it made it at least into the Debian package: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1025997 https://sources.debian.org/src/qtbase-opensource-src/5.15.10%2Bdfsg-1/debian/patches/qxcbwindow_set_geometry.diff/ But the bad thing is, I still receive this crash with the Debian package (yesterday with kate). Therefore this patch is not sufficient. Other notes: - This happens for me just for just for maximized windows. As a workaround having them just near full screen seems avoiding the crash. - My last crash was with two monitors connected, but one was switched off. (Monitors of different brand, different size, same resolution.) - Initially I reported it when doing explicit changes to dual monitor setup, but the last crash was when monitor just went to standby and woke up again. -- You are receiving this mail because: You are watching all bug changes.
[apper] [Bug 463576] "Check For update" cause the crash for kded5 under Debian testing branch (apper segfault in Transaction::role() )
https://bugs.kde.org/show_bug.cgi?id=463576 --- Comment #4 from Bernhard Übelacker --- Bug #462706 states this to be no issue in kded framework and closed RESOLVED UPSTREAM. Therefore this issue was opened: https://github.com/PackageKit/PackageKit-Qt/issues/42 There it was stated that "TransactionWatcher.cpp comes from apper". And the events described in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1026062#35 might boil down to this question: Is it allowed for packagekit-qt to call deleteLater while still in the constructor (as far as I see it fails because it has role Transaction::RoleUnknown), or has TransactionWatcher to cope with this and e.g. not store such a "failed" Transaction in m_transactions? -- You are receiving this mail because: You are watching all bug changes.
[apper] [Bug 463576] "Check For update" cause the crash for kded5 under Debian testing branch (apper segfault in Transaction::role() )
https://bugs.kde.org/show_bug.cgi?id=463576 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #3 from Bernhard Übelacker --- > #4 0x7f555e391f90 in () at > /lib/x86_64-linux-gnu/libc.so.6 > #5 0x7f5508377ba4 in PackageKit::Transaction::role() const () at > ./src/transaction.cpp:297 > #6 0x7f550880daae in > TransactionWatcher::watchTransaction(QDBusObjectPath const&, bool) () at > ./apperd/TransactionWatcher.cpp:104 > #7 0x7f550880db99 in > TransactionWatcher::transactionListChanged(QStringList const&) () at > ./apperd/TransactionWatcher.cpp:85 > ... That backtrace looks similar to bug #462706 or bug #463626. And at Debian side: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1026062 -- You are receiving this mail because: You are watching all bug changes.
[apper] [Bug 463626] KDED: Job Failed
https://bugs.kde.org/show_bug.cgi?id=463626 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #1 from Bernhard Übelacker --- This might be a duplicate of bug #462706 ? -- You are receiving this mail because: You are watching all bug changes.
[apper] [Bug 462706] Since updating to 5.100 kded segfaults in Packagekit (likely during update check)
https://bugs.kde.org/show_bug.cgi?id=462706 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #3 from Bernhard Übelacker --- I tried to bring this to the attention of packagekit-qt developers in this bug report: https://github.com/PackageKit/PackageKit-Qt/issues/42 -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 463052] Krita crashes on double free when closing application
https://bugs.kde.org/show_bug.cgi?id=463052 --- Comment #1 from Bernhard Übelacker --- Overnight a local Debian package built of 5.1.3 finished with my mentioned change, and this one does really no longer crashes. For some reason also this night, the Debian Maintainer uploaded a new version 5.1.4, which states to fix this bug, too. This is really the case in my test environment. I failed to add in my previous information the output to stdout, especially important because of the "Duplicated profile" messages in 5.1.3, which are no longer visible with this new 5.1.4. 5.1.3: $ krita $ krita Lcms2 error: 13 LUT is not suitable to be saved as LutBToA Lcms2 error: 7 Couldn't write type 'mBA ' Lcms2 error: 13 LUT is not suitable to be saved as LutBToA Lcms2 error: 7 Couldn't write type 'mBA ' krita.lib.pigment: Duplicated profile "Artifex PS CMYK Profile" "/usr/share/color/icc/ghostscript/gray_to_k.icc" "" Lcms2 error: 13 LUT is not suitable to be saved as LutBToA Lcms2 error: 7 Couldn't write type 'mBA ' Lcms2 error: 13 LUT is not suitable to be saved as LutBToA Lcms2 error: 7 Couldn't write type 'mBA ' Lcms2 error: 13 LUT is not suitable to be saved as LutBToA Lcms2 error: 7 Couldn't write type 'mBA ' Lcms2 error: 13 LUT is not suitable to be saved as LutBToA Lcms2 error: 7 Couldn't write type 'mBA ' krita.lib.pigment: Duplicated profile "Artifex PS CMYK Profile" "/usr/share/color/icc/ghostscript/ps_cmyk.icc" "/usr/share/color/icc/ghostscript/gray_to_k.icc" Lcms2 error: 13 LUT is not suitable to be saved as LutBToA Lcms2 error: 7 Couldn't write type 'mBA ' Lcms2 error: 13 LUT is not suitable to be saved as LutBToA Lcms2 error: 7 Couldn't write type 'mBA ' krita.lib.pigment: Duplicated profile "Artifex Software sRGB ICC Profile" "/usr/share/color/icc/ghostscript/srgb.icc" "/usr/share/color/icc/ghostscript/default_rgb.icc" krita.lib.pigment: Replacing color space factory "LABA" "L*a*b*/Alpha (16-bit integer/channel)" with "LABA" "L*a*b* (16-bit integer/channel, unmanaged)" krita.lib.pigment: Replacing color space factory "RGBA" "RGB/Alpha (8-bit integer/channel)" with "RGBA" "RGB (8-bit integer/channel, unmanaged)" krita.lib.pigment: Replacing color space factory "RGBA16" "RGB/Alpha (16-bit integer/channel)" with "RGBA16" "RGB (16-bit integer/channel, unmanaged)" QObject::startTimer: Timers cannot have negative intervals /usr/lib/x86_64-linux-gnu/krita-python-libs/krita added to PYTHONPATH Lcms2 error: 13 LUT is not suitable to be saved as LutBToA Lcms2 error: 7 Couldn't write type 'mBA ' 19 -- exe=/usr/bin/krita 14 -- appname=krita 17 -- apppath=/usr/bin 10 -- signal=11 10 -- pid=42710 12 -- startupid=0 KCrash: crashing... crashRecursionCounter = 2 ... $ 5.1.4: $ krita QObject::startTimer: Timers cannot have negative intervals /usr/lib/x86_64-linux-gnu/krita-python-libs/krita added to PYTHONPATH $ -- You are receiving this mail because: You are watching all bug changes.
[krita] [Bug 463052] New: Krita crashes on double free when closing application
https://bugs.kde.org/show_bug.cgi?id=463052
Bug ID: 463052
Summary: Krita crashes on double free when closing application
Classification: Applications
Product: krita
Version: 5.1.3
Platform: Debian testing
OS: Linux
Status: REPORTED
Severity: normal
Priority: NOR
Component: * Unknown
Assignee: krita-bugs-n...@kde.org
Reporter: bernha...@mailbox.org
Target Milestone: ---
I tried investigating on Debian Bug https://bugs.debian.org/1025104.
And I could reproduce it in a i386 and amd64 VM.
The crash happens when main is already left and the exit handlers get executed:
Thread 1 received signal SIGSEGV, Segmentation fault.
0x7fd65c0bef73 in KoColorProfileStorage::Private::~Private
(this=0x557d2d687090, __in_chrg=) at
./libs/pigment/KoColorProfileStorage.cpp:42
42 delete p;
1: x/i $pc
=> 0x7fd65c0bef73 <_ZN21KoColorProfileStorage7PrivateD2Ev+787>: call
*0x8(%rax)
(rr) bt
#0 0x7fd65c0bef73 in KoColorProfileStorage::Private::~Private()
(this=0x557d2d687090, __in_chrg=) at
./libs/pigment/KoColorProfileStorage.cpp:42
#1 0x7fd65c0bcf71 in
QScopedPointerDeleter::cleanup(KoColorProfileStorage::Private*)
(pointer=0x557d2d687090) at
/usr/include/x86_64-linux-gnu/qt5/QtCore/qscopedpointer.h:60
#2
QScopedPointerDeleter::cleanup(KoColorProfileStorage::Private*)
(pointer=0x557d2d687090) at
/usr/include/x86_64-linux-gnu/qt5/QtCore/qscopedpointer.h:52
#3 QScopedPointer >::~QScopedPointer()
(this=, __in_chrg=) at
/usr/include/x86_64-linux-gnu/qt5/QtCore/qscopedpointer.h:107
#4 KoColorProfileStorage::~KoColorProfileStorage() (this=,
__in_chrg=) at ./libs/pigment/KoColorProfileStorage.cpp:56
#5 0x7fd65c0b531d in KoColorSpaceRegistry::Private::~Private()
(this=0x557d2d563d40, __in_chrg=) at
./libs/pigment/KoColorSpaceRegistry.cpp:42
#6 KoColorSpaceRegistry::~KoColorSpaceRegistry() (this=0x7fd65c3841f0
<(anonymous namespace)::Q_QGS_s_instance::innerFunction()::holder>,
__in_chrg=) at ./libs/pigment/KoColorSpaceRegistry.cpp:232
#7 0x7fd65c0b5519 in Holder::~Holder() (this=,
__in_chrg=) at ./libs/pigment/KoColorSpaceRegistry.cpp:39
#8 0x7fd65e2fa435 in __run_exit_handlers (status=0, listp=0x7fd65e48e820
<__exit_funcs>, run_list_atexit=run_list_atexit@entry=true,
run_dtors=run_dtors@entry=true) at ./stdlib/exit.c:113
#9 0x7fd65e2fa5aa in __GI_exit (status=) at
./stdlib/exit.c:143
#10 0x7fd65e2e3191 in __libc_start_call_main
(main=main@entry=0x557d2c4544c0 , argc=argc@entry=1,
argv=argv@entry=0x7ffdc9b620f8) at ../sysdeps/nptl/libc_start_call_main.h:74
#11 0x7fd65e2e3245 in __libc_start_main_impl (main=0x557d2c4544c0
, argc=1, argv=0x7ffdc9b620f8, init=,
fini=, rtld_fini=, stack_end=0x7ffdc9b620e8) at
../csu/libc-start.c:381
#12 0x557d2c458271 in _start ()
(rr) list
30 ~Private()
31 {
32 Q_FOREACH (KoColorProfile *p, profileMap) {
33 profileUniqueIdMap.remove(p->uniqueId());
34 delete p;
35 }
36 profileMap.clear();
37 Q_FOREACH (KoColorProfile *p, profileUniqueIdMap) {
38 delete p;
39 }
40 profileUniqueIdMap.clear();
41 Q_FOREACH(KoColorProfile *p, duplicates) {
42 delete p;
43 }
44 duplicates.clear();
45 }
Valgrind brought this to the point as double free:
==6079== Invalid read of size 4
==6079==at 0x7AC33A8: KoColorProfileStorage::Private::~Private()
(KoColorProfileStorage.cpp:42)
==6079==by 0x7AC12D2: cleanup (qscopedpointer.h:60)
==6079==by 0x7AC12D2: cleanup (qscopedpointer.h:52)
==6079==by 0x7AC12D2: ~QScopedPointer (qscopedpointer.h:107)
==6079==by 0x7AC12D2: KoColorProfileStorage::~KoColorProfileStorage()
(KoColorProfileStorage.cpp:56)
==6079==by 0x7AB8CE6: ~Private (KoColorSpaceRegistry.cpp:42)
==6079==by 0x7AB8CE6: KoColorSpaceRegistry::~KoColorSpaceRegistry()
(KoColorSpaceRegistry.cpp:232)
==6079==by 0x7AB8F07: (anonymous
namespace)::Q_QGS_s_instance::innerFunction()::Holder::~Holder()
(KoColorSpaceRegistry.cpp:39)
==6079==by 0x70D7B5A: __run_exit_handlers (exit.c:89)
==6079==by 0x70D7CB0: exit (exit.c:143)
==6079==by 0x70BF2A0: (below main) (libc_start_call_main.h:74)
==6079== Address 0xdfb8a18 is 0 bytes inside a block of size 8 free'd
==6079==at 0x48436C7: operator delete(void*, unsigned int)
(vg_replace_malloc.c:929)
==6079==by 0x7C6A74B: KoDummyColorProfile::~KoDummyColorProfile()
(KoDummyColorProfile.cpp:16)
==6079==by 0x7AC314B: KoColorProfileStorage::Private::~Private()
(KoColorProfileStorage.cpp:34)
==6079==by 0x7AC12D2: cleanup (qscopedpointer.h:60)
==6079==by 0x7AC12D2: cleanup (qscopedpointer.h:52)
==6079==by 0x7AC12D2: ~QScopedPointer (qscopedpointer.h:107)
==6079==by 0x7AC12D2: KoColorProfileStorage::~KoColorProfileStorage
[yakuake] [Bug 462487] Crash after terminal show
https://bugs.kde.org/show_bug.cgi?id=462487 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #1 from Bernhard Übelacker --- Backtrace looks similar to bug #461723. -- You are receiving this mail because: You are watching all bug changes.
[konsole] [Bug 462022] Konsole (first time) suddenly crashes when aligning to the right of my second monitor
https://bugs.kde.org/show_bug.cgi?id=462022 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #1 from Bernhard Übelacker --- Backtrace looks similar to bug #461723. -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 451110] Apps crash when I minimize windows to latte
https://bugs.kde.org/show_bug.cgi?id=451110 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #1 from Bernhard Übelacker --- Backtrace looks similar to bug #461723. -- You are receiving this mail because: You are watching all bug changes.
[kde] [Bug 461723] konsole (or other applications) crash at disabling second screen
https://bugs.kde.org/show_bug.cgi?id=461723 --- Comment #3 from Bernhard Übelacker --- Hello, upstream bug QTBUG-109226 got closed now with a commit: qtbase/dev: https://code.qt.io/cgit/qt/qtbase.git/commit/?id=6a3627b6c5aa5109a80024f3d7b0f938504f7ffe qtbase/6.4: https://code.qt.io/cgit/qt/qtbase.git/commit/?h=6.4&id=003d30fac2a75ee5f942917dbd4901536a742cbc Unfortunately it looks like the qt-5.15 cerry-pick is not publicly visible ... -- You are receiving this mail because: You are watching all bug changes.
[kde] [Bug 461723] konsole (or other applications) crash at disabling second screen
https://bugs.kde.org/show_bug.cgi?id=461723 --- Comment #2 from Bernhard Übelacker --- Hello Nate, thanks for looking into this report. I opened now https://bugreports.qt.io/browse/QTBUG-109226 -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 460594] Plasma freeze happening since 5.26.0 update
https://bugs.kde.org/show_bug.cgi?id=460594 --- Comment #25 from Bernhard Übelacker --- (In reply to Bernhard Übelacker from comment #24) > Thread 1 in each process seem to be middle in some processing, plasmashell > waiting for some mutex, kwin_x11 for some xcb request. Tried to look a little further in plasmashell process. It looks like thread 1 is inside some QSGThreadedRenderLoop and is waiting to be woken by some other thread: Thread 1 (Thread 0x7fd89f10aa00 (LWP 1857)): ... #4 ___pthread_cond_wait (cond=0x7fd87006c068, mutex=0x7fd87006c040) at ./nptl/pthread_cond_wait.c:618 #5 0x7fd8a3cd199b in QWaitConditionPrivate::wait(QDeadlineTimer) (deadline=..., this=0x7fd87006c040) at thread/qwaitcondition_unix.cpp:146 #6 QWaitCondition::wait(QMutex*, QDeadlineTimer) (this=, mutex=0x7fd87008b428, deadline=...) at thread/qwaitcondition_unix.cpp:225 #7 0x7fd8a5a0b968 in QSGThreadedRenderLoop::polishAndSync(QSGThreadedRenderLoop::Window*, bool) (this=this@entry=0x7fd898007220, w=w@entry=0x7fd8980319a0, inExpose=inExpose@entry=true) at scenegraph/qsgthreadedrenderloop.cpp:1603 #8 0x7fd8a5a0cc23 in QSGThreadedRenderLoop::handleExposure(QQuickWindow*) (this=0x7fd898007220, window=) at scenegraph/qsgthreadedrenderloop.cpp:1374 ... https://sources.debian.org/src/qtdeclarative-opensource-src/5.15.6%2Bdfsg-2/src/quick/scenegraph/qsgthreadedrenderloop.cpp/#L1603 1603 w->thread->waitCondition.wait(&w->thread->mutex); https://sources.debian.org/src/qtdeclarative-opensource-src/5.15.6%2Bdfsg-2/src/quick/scenegraph/qsgthreadedrenderloop.cpp/#L1374 1374 polishAndSync(w, true); There are a few other threads executing below `QSGRenderThread::run()` and also waiting ... : ... #6 QWaitCondition::wait(QMutex*, QDeadlineTimer) (this=this@entry=0x7fd870082768, mutex=mutex@entry=0x7fd870082760, deadline=...) at thread/qwaitcondition_unix.cpp:225 #7 0x7fd8a5a07c95 in QSGRenderThreadEventQueue::takeEvent(bool) (wait=true, this=0x7fd870082758) at scenegraph/qsgthreadedrenderloop.cpp:257 #8 QSGRenderThread::processEventsAndWaitForMore() (this=this@entry=0x7fd8700826c0) at scenegraph/qsgthreadedrenderloop.cpp:935 #9 0x7fd8a5a0a0c1 in QSGRenderThread::run() (this=0x7fd8700826c0) at scenegraph/qsgthreadedrenderloop.cpp:1052 ... But there is one thread that seems waiting in plasmashell process also for some xcb reply: (gdb) bt #0 0x7fd8a3b1afef in __GI___poll (fds=fds@entry=0x7fd72e7fa7a8, nfds=nfds@entry=1, timeout=timeout@entry=-1) at ../sysdeps/unix/sysv/linux/poll.c:29 #1 0x7fd8a61a4d12 in poll (__timeout=-1, __nfds=1, __fds=0x7fd72e7fa7a8) at /usr/include/x86_64-linux-gnu/bits/poll2.h:47 #2 _xcb_conn_wait (c=c@entry=0x56038ce45910, cond=cond@entry=0x7fd7180e71c8, vector=vector@entry=0x0, count=count@entry=0x0) at ../../src/xcb_conn.c:508 #3 0x7fd8a61a716a in xcb_wait_for_special_event (c=0x56038ce45910, se=0x7fd7180e71a0) at ../../src/xcb_in.c:806 #4 0x7fd89ddf1465 in dri3_wait_for_event_locked (draw=draw@entry=0x7fd7180f5798, full_sequence=full_sequence@entry=0x0) at ../src/loader/loader_dri3_helper.c:584 #5 0x7fd89ddf22a4 in dri3_find_back (draw=draw@entry=0x7fd7180f5798, prefer_a_different=) at ../src/loader/loader_dri3_helper.c:726 #6 0x7fd89ddf3da1 in dri3_get_buffer (format=format@entry=4099, buffer_type=buffer_type@entry=loader_dri3_buffer_back, draw=draw@entry=0x7fd7180f5798, driDrawable=) at ../src/loader/loader_dri3_helper.c:2030 #7 0x7fd89ddf40d3 in loader_dri3_get_buffers (driDrawable=, format=4099, stamp=0x7fd718088720, loaderPrivate=0x7fd7180f5798, buffer_mask=, buffers=0x7fd72e7fab00) at ../src/loader/loader_dri3_helper.c:2255 #8 0x7fd8966ae13f in dri_image_drawable_get_buffers (drawable=drawable@entry=0x7fd718088720, images=images@entry=0x7fd72e7fab00, statts=statts@entry=0x7fd71816ce78, statts_count=statts_count@entry=2) at ../src/gallium/frontends/dri/dri2.c:289 #9 0x7fd8966ae27f in dri2_allocate_textures (ctx=0x56038d312f10, drawable=0x7fd718088720, statts=0x7fd71816ce78, statts_count=2) at ../src/gallium/frontends/dri/dri2.c:458 #10 0x7fd8966b1265 in dri_st_framebuffer_validate (stctx=, stfbi=, statts=0x7fd71816ce78, count=2, out=0x7fd72e7face0) at ../src/gallium/frontends/dri/dri_drawable.c:82 #11 0x7fd896784d62 in st_framebuffer_validate (stfb=stfb@entry=0x7fd71816ca20, st=st@entry=0x56039097e4f0) at ../src/mesa/state_tracker/st_manager.c:238 #12 0x7fd896785405 in st_manager_validate_framebuffers (st=0x56039097e4f0) at ../src/mesa/state_tracker/st_manager.c:1333 #13 0x7fd89675f6ed in st_validate_state (st=st@entry=0x56039097e4f0, pipeline=pipeline@entry=ST_PIPELINE_CLEAR) at ../src/mesa/state_tracker/st_atom.c:231 #14 0x7fd896933314 in st_Clear (ctx=0x5603916d9e70, mask=50) at ../src/mesa/state_tracker/st_cb_clear.c:432 #15 0x7fd8a59af885 in QSGBatchRenderer::Renderer::renderBatches() (this=this@entry=0x7fd7180c23a0) at scenegraph/coreapi/qsgbatchrenderer.cpp:4022 #16
[kwin] [Bug 460594] Plasma freeze happening since 5.26.0 update
https://bugs.kde.org/show_bug.cgi?id=460594 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #24 from Bernhard Übelacker --- Created attachment 153835 --> https://bugs.kde.org/attachment.cgi?id=153835&action=edit plasmashell_and_kwin_x11_thread-apply-bt.txt I notice also such a hang. I left today at 12:05 the system. When I came back at 13:45 the time was stuck at 12:15, and three notifications from kdeconnect where still visible. To me it looks like this hang is likely to happen if kdeconnect is producing multiple notifications fast in a row. I assume this is the case when I changed WLAN accesspoints in this time and my phone is connecting to the new acesspoint. I gathered core dumps of kwin_x11 and plasmashell. Attached is a `thread apply all bt`, but there are many threads in each and am not sure which one is supposed to do something. Most of them are waiting for something, but I cannot say for what. If wanted, I can add more debug symbols. Thread 1 in each process seem to be middle in some processing, plasmashell waiting for some mutex, kwin_x11 for some xcb request. -- You are receiving this mail because: You are watching all bug changes.
[kde] [Bug 461723] New: konsole (or other applications) crash at disabling second screen
https://bugs.kde.org/show_bug.cgi?id=461723 Bug ID: 461723 Summary: konsole (or other applications) crash at disabling second screen Classification: I don't know Product: kde Version: unspecified Platform: Debian testing OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: unassigned-b...@kde.org Reporter: bernha...@mailbox.org Target Milestone: --- Hello, I received a crash of konsole when I disabled a second screen via systemsettings. This screen is left of my main screen and has a lower resolution. The crash looks like caused by a calculation in copy_unswapped, which does some pointer arithmetic, but unfortunately the offset gets negative, and therefore unmapped memory is tried to be accessed. Otherwise it looks like yy might be related to a pixel resolution, but my screens are nowhere near a height of 8256 pixel. I received this crash two weeks ago also in konsole and dolphin. This bug might be a duplicate of Bug 461563 and/or Bug 451110. I collected the cores of the three crashes, so I can lookup somthing if needed. STEPS TO REPRODUCE Unfortunately I did not yet try to reproduce it this time. Last time I could not get it crash when I tried to reproduce it. SOFTWARE/OS VERSIONS Operating System: Debian GNU/Linux KDE Plasma Version: 5.26.0 KDE Frameworks Version: 5.98.0 Qt Version: 5.15.6 Kernel Version: 6.0.0-2-amd64 (64-bit) Graphics Platform: X11 Processors: 16 × AMD Ryzen 7 1700 Eight-Core Processor Memory: 31.1 GiB of RAM Graphics Processor: AMD Radeon RX 460 Graphics ADDITIONAL INFORMATION (gdb) bt #0 0x7f009bcfe32f in __GI___poll (fds=0x7ffc26bb9058, nfds=1, timeout=1000) at ../sysdeps/unix/sysv/linux/poll.c:29 #1 0x7f009d975160 in ?? () from /lib/x86_64-linux-gnu/libKF5Crash.so.5 #2 0x7f009d975b67 in KCrash::defaultCrashHandler(int) () from /lib/x86_64-linux-gnu/libKF5Crash.so.5 #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 #21 0x7f009cb11e1c in QWindowSystemInterface::sendWindowSystemEvents (flags=flags@entry=...) at kernel/qwindowsysteminterface.cpp:1169 #22 0x7f00962fc0fa in xcbSourceDispatch (source=) at ./src/plugins/platforms/xcb/qxcbeventdispatcher.cpp:105 #23 0x7f009a9da799 in g_main_context_dispatch () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #24 0x7f009a9daa28 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #25 0x7f009a9daabc in g_main_context_iteration () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #26 0x7f009c7094b6 in QEventDispatcherG
[konsole] [Bug 454122] neofetch and right mouse button menu at the same time will cause a crash
https://bugs.kde.org/show_bug.cgi?id=454122
--- Comment #7 from Bernhard Übelacker ---
A short addition: I could reproduce it inside a rr-debugger recording.
It looks to me like updateForegroundProcessInfo (which sets
_foregroundProcessInfo)
was called the last time even before the neofetch process got started.
Therefore _foregroundProcessInfo contains that old "state" while
isForegroundProcessActive is returning the current "state".
Another place which calls isForegroundProcessActive
is calling updateForegroundProcessInfo immediately after it.
Therefore maybe this change might be considered?
-if (isForegroundProcessActive() && _foregroundProcessInfo->isValid())
{
+if (isForegroundProcessActive() && updateForegroundProcessInfo() &&
_foregroundProcessInfo && _foregroundProcessInfo->isValid()) {
--
You are receiving this mail because:
You are watching all bug changes.
[konsole] [Bug 454122] neofetch and right mouse button menu at the same time will cause a crash
https://bugs.kde.org/show_bug.cgi?id=454122
Bernhard Übelacker changed:
What|Removed |Added
CC||bernha...@mailbox.org
--- Comment #6 from Bernhard Übelacker ---
I received this crash too with current Debian testing.
Below the top of the backtrace without and the full backtrace with debug
symbols installed.
While in my first crash I think I had also a ssh connection open, I could
reproduce the issue
- with installing neofetch
- starting a new konsole
- starting inside neofetch
- right click while neofetch was still running.
Interesting looks the call to `Konsole::ProcessInfo::isValid` with this=0x0.
I assume these bugs are duplicates to this one: 458157 460764
And bug 391447 seems also kind of related, while it was closed with a commit
changing this:
-if (isForegroundProcessActive()) {
+if (isForegroundProcessActive() && _foregroundProcessInfo->isValid())
{
Maybe extend this check by following:
-if (isForegroundProcessActive() && _foregroundProcessInfo->isValid())
{
+if (isForegroundProcessActive() && _foregroundProcessInfo &&
_foregroundProcessInfo->isValid()) {
Kind regards,
Bernhard
Application: Konsole (konsole), signal: Segmentation fault
[KCrash Handler]
#4 0x7f85356d0ad0 in Konsole::ProcessInfo::isValid() const () from
/lib/x86_64-linux-gnu/libkonsoleprivate.so.1
#5 0x7f85357609f4 in Konsole::Session::getUrl() () from
/lib/x86_64-linux-gnu/libkonsoleprivate.so.1
#6 0x7f8535768b71 in Konsole::SessionController::url() const () from
/lib/x86_64-linux-gnu/libkonsoleprivate.so.1
#7 0x7f8535776bdc in
Konsole::SessionController::showDisplayContextMenu(QPoint const&) () from
/lib/x86_64-linux-gnu/libkonsoleprivate.so.1
...
konsole-dbgsym 4:22.08.1-1
konsole-kpart-dbgsym 4:22.08.1-1
libqt5core5a-dbgsym 5.15.6+dfsg-2
libqt5gui5-dbgsym 5.15.6+dfsg-2
libqt5widgets5-dbgsym 5.15.6+dfsg-2
libglib2.0-0-dbgsym 2.74.1-1
Application: Konsole (konsole), signal: Segmentation fault
[KCrash Handler]
#4 Konsole::ProcessInfo::isValid (this=0x0) at
/usr/include/x86_64-linux-gnu/qt5/QtCore/qflags.h:151
#5 0x7f85357609f4 in Konsole::Session::getUrl (this=0x5646bfb35520) at
./src/session/Session.cpp:1232
#6 0x7f8535768b71 in Konsole::SessionController::url (this=) at ./src/session/SessionController.cpp:379
#7 0x7f8535776bdc in Konsole::SessionController::showDisplayContextMenu
(this=0x5646bf563fb0, position=...) at ./src/session/SessionController.cpp:1891
#8 0x7f85340e8caf in QtPrivate::QSlotObjectBase::call (a=0x7ffe7043e5c0,
r=0x5646bf563fb0, this=0x5646bfbdbce0) at
../../include/QtCore/../../src/corelib/kernel/qobjectdefs_impl.h:398
#9 doActivate (sender=0x5646bf95cea0, signal_index=12,
argv=0x7ffe7043e5c0) at kernel/qobject.cpp:3919
#10 0x7f85340e1f4f in QMetaObject::activate
(sender=sender@entry=0x5646bf95cea0, m=m@entry=0x7f8535803be0
,
local_signal_index=local_signal_index@entry=5, argv=argv@entry=0x7ffe7043e5c0)
at kernel/qobject.cpp:3979
#11 0x7f85356ba575 in Konsole::TerminalDisplay::configureRequest
(this=this@entry=0x5646bf95cea0, _t1=...) at
./obj-x86_64-linux-gnu/src/konsoleprivate_autogen/DRAQINE3W2/moc_TerminalDisplay.cpp:398
#12 0x7f853573fc8f in Konsole::TerminalDisplay::mousePressEvent
(this=0x5646bf95cea0, ev=0x7ffe7043eb10) at
./src/terminalDisplay/TerminalDisplay.cpp:1168
#13 0x7f8534da4ce8 in QWidget::event (this=0x5646bf95cea0,
event=0x7ffe7043eb10) at kernel/qwidget.cpp:9034
#14 0x7f8534d62f5e in QApplicationPrivate::notify_helper
(this=this@entry=0x5646bf4211d0, receiver=receiver@entry=0x5646bf95cea0,
e=e@entry=0x7ffe7043eb10) at kernel/qapplication.cpp:3637
#15 0x7f8534d6b502 in QApplication::notify (this=,
receiver=0x5646bf95cea0, e=) at kernel/qapplication.cpp:3081
#16 0x7f85340b1718 in QCoreApplication::notifyInternal2
(receiver=0x5646bf95cea0, event=0x7ffe7043eb10) at
kernel/qcoreapplication.cpp:1064
#17 0x7f8534d6960e in QApplicationPrivate::sendMouseEvent
(receiver=receiver@entry=0x5646bf95cea0, event=event@entry=0x7ffe7043eb10,
alienWidget=, nativeWidget=0x5646bf74f790,
buttonDown=buttonDown@entry=0x7f85352a69f0 ,
lastMouseReceiver=..., spontaneous=true, onlyDispatchEnterLeave=false) at
kernel/qapplication.cpp:2619
#18 0x7f8534dbdaf8 in QWidgetWindow::handleMouseEvent (this=0x5646bf779c80,
event=0x7ffe7043edc0) at kernel/qwidgetwindow.cpp:683
#19 0x7f8534dc0e80 in QWidgetWindow::event (this=0x5646bf779c80,
event=0x7ffe7043edc0) at kernel/qwidgetwindow.cpp:300
#20 0x7f8534d62f5e in QApplicationPrivate::notify_helper (this=, receiver=0x5646bf779c80, e=0x7ffe7043edc0) at
kernel/qapplication.cpp:3637
#21 0x7f85340b1718 in QCoreApplication::notifyInternal2
(receiver=0x5646bf779c80, event=0x7ffe7043edc0) at
kernel/qcoreapplication.cpp:1064
#22 0x7f853453d6bd in
[valgrind] [Bug 454925] New: Stracktraces contain addresses that are off by one, except frame 0
https://bugs.kde.org/show_bug.cgi?id=454925 Bug ID: 454925 Summary: Stracktraces contain addresses that are off by one, except frame 0 Product: valgrind Version: unspecified Platform: Debian stable OS: Linux Status: REPORTED Severity: minor Priority: NOR Component: general Assignee: jsew...@acm.org Reporter: bernha...@mailbox.org Target Milestone: --- Created attachment 149505 --> https://bugs.kde.org/attachment.cgi?id=149505&action=edit Attempt-to-fix-EIP-pointer-in-stacktrace.patch The addresses in stacktraces shown by valgrind are off by one, except for frame 0. This seems to be explained in this commit: https://sourceware.org/git/?p=valgrind.git;a=patch;f=coregrind/m_stacktrace.c;h=efaaa0b74a695a68ef5e758102d3fe3920e121fd So valgrind is decrementing the address by one to get into the call instruction. But when showing the stacktrace this decremention gets not corrected later. Below example shows the same process, one with the valgrind stacktrace and one with the gdb stacktrace. A disassemble shows valgrind's address in the last byte of the call instruction, while gdb shows the next byte, the return address of the call instruction. $ valgrind --vgdb=yes --vgdb-error=0 rc ==4357== Invalid free() / delete / delete[] / realloc() ==4357==at 0x4840EB7: free (vg_replace_malloc.c:872) ==4357==by 0x1134BF: ??? (in /usr/bin/rc.byron) ==4357==by 0x118B7C: ??? (in /usr/bin/rc.byron) ==4357==by 0x118473: ??? (in /usr/bin/rc.byron) ==4357==by 0x486F6E5: rl_completion_matches (in /usr/lib/i386-linux-gnu/libreadline.so.8.1) ==4357==by 0x118760: ??? (in /usr/bin/rc.byron) ... $ gdb $(which rc) (gdb) target remote | /usr/bin/vgdb --pid=4357 (gdb) cont (gdb) bt #0 0x04840eb7 in _vgr10050ZU_VgSoSynsomalloc_free (p=0x4b2e758) at m_replacemalloc/vg_replace_malloc.c:872 #1 0x001134c0 in ?? () #2 0x00118b7d in ?? () #3 0x00118474 in ?? () #4 0x0486f6e6 in rl_completion_matches () from /lib/i386-linux-gnu/libreadline.so.8 #5 0x00118761 in ?? () ... Dump of assembler code from 0x4840eb5 to 0x4840ebd: 0x04840eb5 <_vgr10050ZU_VgSoSynsomalloc_free+101>: 87 db xchg %ebx,%ebx 0x04840eb7 <_vgr10050ZU_VgSoSynsomalloc_free+103>: 89 55 e0mov %edx,-0x20(%ebp) 0x04840eba <_vgr10050ZU_VgSoSynsomalloc_free+106>: 8b 45 e0mov -0x20(%ebp),%eax End of assembler dump. ... (gdb) disassemble /r 0x001134ba,0x001134c6 Dump of assembler code from 0x1134ba to 0x1134c6: 0x001134ba: 50 push %eax 0x001134bb: e8 c0 6b ff ff call 0x10a080 0x001134c0: 83 c4 10add$0x10,%esp 0x001134c3: 83 c4 08add$0x8,%esp End of assembler dump. Attached patch is an attempt to correct just the output of the stacktrace in `VG_(describe_IP)` while leaving the internal used address unchanged. SOFTWARE/OS VERSIONS Linux: valgrind 1:3.16.1-1 amd64 (Bullseye) valgrind 1:3.18.1-1 i386 (Bookworm) -- You are receiving this mail because: You are watching all bug changes.
[ksysguard] [Bug 350140] KSysGuard crashed on close
https://bugs.kde.org/show_bug.cgi?id=350140 --- Comment #58 from Bernhard Übelacker --- (I am no longer sure about my relation ot the systemload in my previous post, as I was not able to reproduce the crash with neither version running without any debugger.) I could reproduce the crash in 3 out of 3 attempts running with "rr record -n" without D29616 applied. Then I applied D29616 on top of the Debian package ksysguard 5.20.5-2. With this 3 out of 3 attemts did not show a crash anymore. So I assume the patch successfully avoids the crash. -- You are receiving this mail because: You are watching all bug changes.
[ksysguard] [Bug 350140] KSysGuard crashed on close
https://bugs.kde.org/show_bug.cgi?id=350140
Bernhard Übelacker changed:
What|Removed |Added
CC||bernha...@mailbox.org
--- Comment #56 from Bernhard Übelacker ---
I received some similar crashes when closing ksysguard.
Therefore I tried to do some more investigation using rr-debugger.
It was quite good reproducible by opening ksysguard, then watching
a partial build of e.g. wine (many short lived processes).
There I could track it down to following free,
which destroys the QLabel in sbCpuStat:
(rr) reverse-cont
Continuing.
Thread 1 hit Breakpoint 6, __GI___libc_free (mem=0x562c360be2e0) at
malloc.c:3092
3092in malloc.c
9: Toplevel = (class TopLevel *) 0x562c35c4ecf0
10: Toplevel->sbCpuStat = (class QLabel *) 0x562c360be2e0
11: Toplevel->sbCpuStat->d_ptr = {d = 0x562c36043d30}
12: x/i $pc
=> 0x7f2c687dd720 <__GI___libc_free>: sub$0x18,%rsp
(rr) bt
#0 __GI___libc_free (mem=0x562c360be2e0) at malloc.c:3092
#1 0x7f2c66eacbde in QObjectPrivate::deleteChildren()
(this=this@entry=0x562c360403c0) at kernel/qobject.cpp:2104
#2 0x7f2c67b04736 in QWidget::~QWidget() (this=0x562c36040380,
__in_chrg=) at kernel/qwidget.cpp:1522
#3 0x7f2c67c72319 in QStatusBar::~QStatusBar() (this=0x562c36040380,
__in_chrg=) at widgets/qstatusbar.cpp:252
#4 0x7f2c66eacbde in QObjectPrivate::deleteChildren()
(this=this@entry=0x562c35c6d420) at kernel/qobject.cpp:2104
#5 0x7f2c67b04736 in QWidget::~QWidget() (this=0x562c35c4ecf0,
__in_chrg=) at kernel/qwidget.cpp:1522
#6 0x7f2c68945aa7 in TopLevel::~TopLevel() (this=0x562c35c4ecf0,
__in_chrg=, __vtt_parm=) at
/usr/include/ksysguard/ksgrd/SensorClient.h:44
#7 TopLevel::~TopLevel() (this=0x562c35c4ecf0, __in_chrg=,
__vtt_parm=) at
./obj-x86_64-linux-gnu/gui/kdeinit_ksysguard_autogen/EWIEGA46WW/../../../../gui/ksysguard.h:41
#8 0x7f2c66eaedaf in QObject::event(QEvent*) (this=0x562c35c4ecf0,
e=0x562c36389a10) at kernel/qobject.cpp:1301
#9 0x7f2c6834f659 in KXmlGuiWindow::event(QEvent*) () at
/usr/lib/x86_64-linux-gnu/libKF5XmlGui.so.5
#10 0x7f2c67ac915f in QApplicationPrivate::notify_helper(QObject*, QEvent*)
(this=, receiver=0x562c35c4ecf0, e=0x562c36389a10) at
kernel/qapplication.cpp:3632
#11 0x7f2c66e82fca in QCoreApplication::notifyInternal2(QObject*, QEvent*)
(receiver=0x562c35c4ecf0, event=0x562c36389a10) at
kernel/qcoreapplication.cpp:1063
#12 0x7f2c66e85a01 in QCoreApplicationPrivate::sendPostedEvents(QObject*,
int, QThreadData*) (receiver=0x0, event_type=0, data=0x562c35acb8f0) at
kernel/qcoreapplication.cpp:1817
#13 0x7f2c66edae93 in postEventSourceDispatch(GSource*, GSourceFunc,
gpointer) (s=0x562c35c59fc0) at kernel/qeventdispatcher_glib.cpp:277
#14 0x7f2c5da0ee6b in g_main_dispatch (context=0x7f2c4c005000) at
../../../glib/gmain.c:3325
#15 g_main_context_dispatch (context=0x7f2c4c005000) at
../../../glib/gmain.c:4043
#16 0x7f2c5da0f118 in g_main_context_iterate
(context=context@entry=0x7f2c4c005000, block=block@entry=1,
dispatch=dispatch@entry=1, self=) at ../../../glib/gmain.c:4119
#17 0x7f2c5da0f1cf in g_main_context_iteration (context=0x7f2c4c005000,
may_block=may_block@entry=1) at ../../../glib/gmain.c:4184
#18 0x7f2c66eda51f in
QEventDispatcherGlib::processEvents(QFlags)
(this=0x562c35c6b380, flags=...) at kernel/qeventdispatcher_glib.cpp:423
#19 0x7f2c66e8198b in
QEventLoop::exec(QFlags)
(this=this@entry=0x7ffcffd2dea0, flags=..., flags@entry=...) at
../../include/QtCore/../../src/corelib/global/qflags.h:69
#20 0x7f2c66e89c00 in QCoreApplication::exec() () at
../../include/QtCore/../../src/corelib/global/qflags.h:121
#21 0x7f2c6898206f in kdemain(int, char**) (argc=,
argv=) at ./gui/ksysguard.cpp:627
#22 0x7f2c68779d0a in __libc_start_main (main=0x562c354bc050 ,
argc=1, argv=0x7ffcffd2e0c8, init=, fini=,
rtld_fini=, stack_end=0x7ffcffd2e0b8) at ../csu/libc-start.c:308
#23 0x562c354bc08a in _start ()
(rr) when
Current event: 531099
Then a little later the same also destroyed TopLevel object is referenced,
and an assignment to the sbCpuStat QLabel is done.
(rr) cont
Continuing.
Thread 1 received signal SIGSEGV, Segmentation fault.
0x7f2c66d247a9 in operator== (s1=, s2="CPU: 15 %\302\234\061\065 %") at
text/qstring.cpp:3437
3437if (s1.d->size != s2.d->size)
9: Toplevel = (TopLevel *) 0x562c35c4ecf0
10: Toplevel->sbCpuStat = (QLabel *) 0x562c360be2e0
11: Toplevel->sbCpuStat->d_ptr = {d = 0x562c3609fb90}
12: x/i $pc
=> 0x7f2c66d247a9 <_ZeqRK7QStringS1_+9>:movslq 0x4(%rax),%rsi
(rr) bt
#0 0x7f2c66d247a9 in operator==(QString const&, QString const&) (s1=, s2="CPU:
15 %\302\234\061\065 %") at text/qstring.cpp:3437
#1 0x7f2c67bfd3e5 in QLabel::setText(QString const&) (this=0x562c360be2e0,
text="CP
[kio-extras] [Bug 415693] MTP randomly crashes kdeinit5
https://bugs.kde.org/show_bug.cgi?id=415693 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #1 from Bernhard Übelacker --- This downstream bug report seems related: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=920772#20 -- You are receiving this mail because: You are watching all bug changes.
[plasmashell] [Bug 411106] New: Remove disconnected network devices from Network Monitor.
https://bugs.kde.org/show_bug.cgi?id=411106 Bug ID: 411106 Summary: Remove disconnected network devices from Network Monitor. Product: plasmashell Version: 5.14.5 Platform: Debian stable OS: Linux Status: REPORTED Severity: wishlist Priority: NOR Component: general Assignee: k...@davidedmundson.co.uk Reporter: bernha...@mailbox.org CC: plasma-b...@kde.org Target Milestone: 1.0 USB network devices or VirtualBox bridges are just temporarily up and active in the system. The Network Monitor Widget automatically adds such interfaces when such devices get connected. Unfortunately it does not remove them when disconnected. This patch tries to implement also the removal and works in current Debian testing/buster. But it touches the common file Applet.qml, so it may have unwanted side effects to other widgets in its current state. Betriebssystem: Debian GNU/Linux KDE-Plasma-Version: 5.14.5 Qt-Version: 5.11.3 KDE-Frameworks-Version: 5.54.0 Kernel-Version: 4.19.0-4-amd64 Art des Betriebssystems: 64-bit Prozessoren: 16 × AMD Ryzen 7 1700 Eight-Core Processor Speicher: 15,5 GiB Arbeitsspeicher The bug tracker mentioned to open a proposed change in phabricator, so I did here: https://phabricator.kde.org/D21035 -- You are receiving this mail because: You are watching all bug changes.
[marble] [Bug 378885] Marble crashes sometime when using the vector map theme
https://bugs.kde.org/show_bug.cgi?id=378885 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #12 from Bernhard Übelacker --- Hello, hit this crash at 2018-04-13 in current Debian Buster too, with: marble 4:17.08.3-3.1 libqt5gui5 5.11.3+dfsg1-1 Could not reproduce it later again. Was moving the mouse, maybe scrolling, while marble was loading Vektor-OSM data over a slow mobile connection. Thread 1 (Thread 0x7fc1489ed940 (LWP 5283)): [KCrash Handler] #6 0x7fc15317327b in QPixmap::size() const () at /usr/lib/x86_64-linux-gnu/libQt5Gui.so.5 #7 0x7fc154698748 in () at /usr/lib/x86_64-linux-gnu/libmarblewidget-qt5.so.28 #8 0x7fc15469a303 in () at /usr/lib/x86_64-linux-gnu/libmarblewidget-qt5.so.28 #9 0x7fc1545e96ca in Marble::MarbleMap::hasFeatureAt(QPoint const&) const () at /usr/lib/x86_64-linux-gnu/libmarblewidget-qt5.so.28 #10 0x7fc1546339dc in Marble::MarbleDefaultInputHandler::adjustCursorShape(QPoint const&, QPoint const&) () at /usr/lib/x86_64-linux-gnu/libmarblewidget-qt5.so.28 #11 0x7fc154634fba in Marble::MarbleDefaultInputHandler::handleMouseEvent(QMouseEvent*) () at /usr/lib/x86_64-linux-gnu/libmarblewidget-qt5.so.28 ... -- You are receiving this mail because: You are watching all bug changes.
[ksplash] [Bug 405446] ksplashqml hits its hard timeout of 30 seconds because setStage(6) is needed twice
https://bugs.kde.org/show_bug.cgi?id=405446 Bernhard Übelacker changed: What|Removed |Added Attachment #118782|0 |1 is obsolete|| --- Comment #8 from Bernhard Übelacker --- Comment on attachment 118782 --> https://bugs.kde.org/attachment.cgi?id=118782 ksplashqml: Quit on first call to setStage with stage == 6. Moving the assignment is not correct as there are really 7 stages. -- You are receiving this mail because: You are watching all bug changes.
[ksplash] [Bug 405446] ksplashqml hits its hard timeout of 30 seconds because setStage(6) is needed twice
https://bugs.kde.org/show_bug.cgi?id=405446 --- Comment #7 from Bernhard Übelacker --- Hello Kai Uwe Broulik, with the patch from #405444 I now receive that missing "kcminit" stage. Maybe that stage "kcminit" arrives even before the "kinit" and got lost because of the not yet up dbus interface. So this issue might be just closed. Still adding that stage "kcminit" to the comment in splashapp.cpp might be an improvement. Kind regards, Bernhard -- You are receiving this mail because: You are watching all bug changes.
[ksplash] [Bug 405446] ksplashqml hits its hard timeout of 30 seconds because setStage(6) is needed twice
https://bugs.kde.org/show_bug.cgi?id=405446 Bernhard Übelacker changed: What|Removed |Added Attachment #118783|0 |1 is obsolete|| --- Comment #6 from Bernhard Übelacker --- Comment on attachment 118783 --> https://bugs.kde.org/attachment.cgi?id=118783 ksplashqml: Add some logging to get details when stages were reached. Better version added to #405444. -- You are receiving this mail because: You are watching all bug changes.
[ksplash] [Bug 405444] ksplashqml hits its hard timeout of 30 seconds because of failing qdbus call kinit
https://bugs.kde.org/show_bug.cgi?id=405444 --- Comment #4 from Bernhard Übelacker --- Created attachment 118829 --> https://bugs.kde.org/attachment.cgi?id=118829&action=edit ksplashqml: Add some logging to get details when stages were reached. -- You are receiving this mail because: You are watching all bug changes.
[ksplash] [Bug 405444] ksplashqml hits its hard timeout of 30 seconds because of failing qdbus call kinit
https://bugs.kde.org/show_bug.cgi?id=405444
--- Comment #3 from Bernhard Übelacker ---
Hello Kai Uwe Broulik,
> > Can you give the patch https://phabricator.kde.org/D19753 a try?
> I will rebuild a package with just that change and report back.
I build the package with D19753 applied and the first some boots did not
show these "Cannot find 'org.kde.KSplash.setStage' ..." messages anymore.
Additional I had modified my logging patch and now I also get the
"kcminit" stage even before the "kinit".
Might the "kcminit" also just got lost because the
dbus interface was not yet up?
Kind regards,
Bernhard
2019-03-15 22:34:12.728 SplashApp::setStage("initial")
2019-03-15 22:34:12.728 SplashApp::setStage("initial"): m_stages.count==1
2019-03-15 22:34:12.728 SplashApp::setStage(1), m_stage==0
2019-03-15 22:34:12.803 SplashApp::setStage("kcminit")
2019-03-15 22:34:12.803 SplashApp::setStage("kcminit"): m_stages.count==2
2019-03-15 22:34:12.803 SplashApp::setStage(2), m_stage==1
2019-03-15 22:34:12.811 SplashApp::setStage("kinit")
2019-03-15 22:34:12.811 SplashApp::setStage("kinit"): m_stages.count==3
2019-03-15 22:34:12.811 SplashApp::setStage(3), m_stage==2
2019-03-15 22:34:12.891 SplashApp::setStage("ksmserver")
2019-03-15 22:34:12.891 SplashApp::setStage("ksmserver"): m_stages.count==4
2019-03-15 22:34:12.891 SplashApp::setStage(4), m_stage==3
2019-03-15 22:34:13.119 SplashApp::setStage("ready")
2019-03-15 22:34:13.119 SplashApp::setStage("ready"): m_stages.count==5
2019-03-15 22:34:13.119 SplashApp::setStage(5), m_stage==4
2019-03-15 22:34:13.440 SplashApp::setStage("wm")
2019-03-15 22:34:13.440 SplashApp::setStage("wm"): m_stages.count==6
2019-03-15 22:34:13.440 SplashApp::setStage(6), m_stage==5
2019-03-15 22:34:15.841 SplashApp::setStage("desktop")
2019-03-15 22:34:15.841 SplashApp::setStage("desktop"): m_stages.count==7
2019-03-15 22:34:15.841 SplashApp::setStage(7), m_stage==6
2019-03-15 22:34:15.841 SplashApp::setStage() before exit
--
You are receiving this mail because:
You are watching all bug changes.
[ksplash] [Bug 405444] ksplashqml hits its hard timeout of 30 seconds because of failing qdbus call kinit
https://bugs.kde.org/show_bug.cgi?id=405444
--- Comment #2 from Bernhard Übelacker ---
Hello Kai Uwe Broulik,
thanks for your time.
> Do you use a fancy splash screen theme? Are you running Wayland?
No fancy splash that I am aware of - visually equal to a current standard
debian buster installation - just the black screen with logo in the middle
and the "circling circle". ("ksplashqml Breeze --pid")
Also not running Wayland - I am still on Xorg with AMD graphics drivers.
> Can you give the patch https://phabricator.kde.org/D19753 a try?
I will rebuild a package with just that change and report back.
Kind regards,
Bernhard
--
You are receiving this mail because:
You are watching all bug changes.
[ksplash] [Bug 405446] ksplashqml hits its hard timeout of 30 seconds because setStage(6) is needed twice
https://bugs.kde.org/show_bug.cgi?id=405446 --- Comment #5 from Bernhard Übelacker --- Hello Kai Uwe Broulik, thats strange because I could get with my patches applied just the 6 stages mentioned in the comment in [1]. Also the comment of the lastest git version just mentions these 6. So the comment might be incomplete and I have to investigate why I don't get the "kcminit" stage? At a first sight I don't see a failure of kcminit_startup - I will have a look and report back. [1] https://sources.debian.org/src/plasma-workspace/4:5.14.5.1-1/ksplash/ksplashqml/splashapp.cpp/#L40 [2] https://cgit.kde.org/plasma-workspace.git/tree/ksplash/ksplashqml/splashapp.cpp#n40 -- You are receiving this mail because: You are watching all bug changes.
[ksplash] [Bug 405446] ksplashqml hits its hard timeout of 30 seconds because setStage(6) is needed twice
https://bugs.kde.org/show_bug.cgi?id=405446
--- Comment #3 from Bernhard Übelacker ---
(In reply to Kai Uwe Broulik from comment #2)
> Is this still an issue or just a symptom of Bug 405444?
I think both are two distinct issues:
- in 405444 we are just collecting 5 different elements in m_stages and
therefore the most we reach is a call setStage(5)
and never enter the exit path.
- in 405446 we reach setStage(6), but need any other call setStage(int)
following to enter the exit path.
But there might be a reason for that ordering in
setStage(int) that I may have overlooked.
E.g. two dbus calls setStage("desktop") are expected to be received?
Therefore I am less sure about 405446.
--
You are receiving this mail because:
You are watching all bug changes.
[ksplash] [Bug 405446] ksplashqml hits its hard timeout of 30 seconds because setStage(6) is needed twice
https://bugs.kde.org/show_bug.cgi?id=405446 --- Comment #1 from Bernhard Übelacker --- Created attachment 118783 --> https://bugs.kde.org/attachment.cgi?id=118783&action=edit ksplashqml: Add some logging to get details when stages were reached. -- You are receiving this mail because: You are watching all bug changes.
[ksplash] [Bug 405446] New: ksplashqml hits its hard timeout of 30 seconds because setStage(6) is needed twice
https://bugs.kde.org/show_bug.cgi?id=405446 Bug ID: 405446 Summary: ksplashqml hits its hard timeout of 30 seconds because setStage(6) is needed twice Product: ksplash Version: 5.14.5 Platform: Debian testing OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: unassigned-b...@kde.org Reporter: bernha...@mailbox.org Target Milestone: --- Created attachment 118782 --> https://bugs.kde.org/attachment.cgi?id=118782&action=edit ksplashqml: Quit on first call to setStage with stage == 6. As far as I see the dbus interface setStage is called with 6 different strings. These get stored in m_stages and an setStage overload is called with the number of elements in m_stages. When setStage is called with parameter stage==6 another it gets not into the closing path immediately because another setStage call is needed for some reason. I could not find a reason for the need of this second setStage call and therefore want to ask if the assignment has really to be done before the check "m_stage == 6"? With this I get the ksplashqml appearance of less than 5 seconds instead of 30. The second attached patch was used to get some logging from ksplashqml. Operating System: Debian GNU/Linux KDE Plasma Version: 5.14.5 Qt Version: 5.11.3 KDE Frameworks Version: 5.54.0 Kernel Version: 4.19.0-2-amd64 OS Type: 64-bit Processors: 16 × AMD Ryzen 7 1700 Eight-Core Processor Memory: 15.5 GiB of RAM -- You are receiving this mail because: You are watching all bug changes.
[ksplash] [Bug 405444] New: ksplashqml hits its hard timeout of 30 seconds because of failing qdbus call kinit
https://bugs.kde.org/show_bug.cgi?id=405444 Bug ID: 405444 Summary: ksplashqml hits its hard timeout of 30 seconds because of failing qdbus call kinit Product: ksplash Version: 5.14.5 Platform: Debian testing OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: unassigned-b...@kde.org Reporter: bernha...@mailbox.org Target Milestone: --- Created attachment 118780 --> https://bugs.kde.org/attachment.cgi?id=118780&action=edit ksplashqml: Retry in start script if dbus interface is not yet up. When /usr/bin/startkde starts it first executes ksplashqml around line 134. Later around line 297 is a call "qdbus ... kinit" to increase the stage in ksplashqml. Unfortunately it looks like on fast enough systems the qdbus call is faster reached as ksplashqml has opened the dbus interface. This results in the message in .xsession-errors: Cannot find 'org.kde.KSplash.setStage' in object /KSplash at org.kde.KSplash Therefore ksplashqml seems to never reach setStage(6) to finish its execution and therefore runs into the hard limit of 30 seconds. Attached patch tries to retry the qdus call when it returns not success. Unfortunately the overhead of the loop might be already enough to have the dbus interface already up, because I did never see the logging when doing a retry. Operating System: Debian GNU/Linux KDE Plasma Version: 5.14.5 Qt Version: 5.11.3 KDE Frameworks Version: 5.54.0 Kernel Version: 4.19.0-2-amd64 OS Type: 64-bit Processors: 16 × AMD Ryzen 7 1700 Eight-Core Processor Memory: 15.5 GiB of RAM -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 391214] kwin crash
https://bugs.kde.org/show_bug.cgi?id=391214 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #3 from Bernhard Übelacker --- This bug looks like a duplicate of bug #390892. At least the frames 6 to 19 are equal. -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 392614] kwin_wayland crash when theme applied
https://bugs.kde.org/show_bug.cgi?id=392614 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #4 from Bernhard Übelacker --- Might be a duplicate of bug #395346. -- You are receiving this mail because: You are watching all bug changes.
[plasmashell] [Bug 396258] wayland: using window decorations installed from store crashes plasma when trying to close any window
https://bugs.kde.org/show_bug.cgi?id=396258 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #7 from Bernhard Übelacker --- Might this be a duplicate of #395346? -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 396096] /usr/bin/kwin_wayland asserts: Scene::windowClosed: assert(m_windows.contains(c));
https://bugs.kde.org/show_bug.cgi?id=396096 --- Comment #2 from Bernhard Übelacker --- A small addition: This bug seems more to be the duplicate of bug #395346, where a workaround was implemented that led to another crash in bug #395732, where the previous workaround got reverted. The first bug mentions also [QTBUG-68997]. Therefore if I read the bugs correct there is no workaround anymore in place and currently the only way to avoid this bug is to wait for Qt 5.12. Does this qualify #395346 to get reopened? [QTBUG-68997] https://bugreports.qt.io/browse/QTBUG-68997 -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 396096] New: /usr/bin/kwin_wayland asserts: Scene::windowClosed: assert(m_windows.contains(c));
https://bugs.kde.org/show_bug.cgi?id=396096 Bug ID: 396096 Summary: /usr/bin/kwin_wayland asserts: Scene::windowClosed: assert(m_windows.contains(c)); Product: kwin Version: 5.12.5 Platform: Debian testing OS: Linux Status: UNCONFIRMED Severity: normal Priority: NOR Component: wayland-generic Assignee: kwin-bugs-n...@kde.org Reporter: bernha...@mailbox.org Target Milestone: --- Just tried a plasma wayland session and found following when closing a window. Might be related to using non default theme (changed it last year because of a Qt bug). What looks suspicious is that KWin::ShellClient::destroyClient is twice in the callstack for the same object. If there is anything else I can provide, the core file is still there and all needed debug packages installed. (Not sure about the component selected for this bug report.) Core was generated by `/usr/bin/kwin_wayland --xwayland --libinput --exit-with-session=/usr/lib/x86_64'. Program terminated with signal SIGABRT, Aborted. (gdb) bt #0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51 #1 0x7fc92210830f in __GI_abort () at abort.c:100 #2 0x7fc9220ff9da in __assert_fail_base (fmt=0x7fc909213129 , assertion=assertion@entry=0x7fc925110272 "m_windows.contains(c)", file=file@entry=0x7fc9251102e0 "/build/kwin-posrbs/kwin-5.12.5/scene.cpp", line=line@entry=414, function=function@entry=0x7fc925110480 "void KWin::Scene::windowClosed(KWin::Toplevel*, KWin::Deleted*)") at assert.c:92 #3 0x7fc9220ffa52 in __GI___assert_fail (assertion=assertion@entry=0x7fc925110272 "m_windows.contains(c)", file=file@entry=0x7fc9251102e0 "/build/kwin-posrbs/kwin-5.12.5/scene.cpp", line=line@entry=414, function=function@entry=0x7fc925110480 "void KWin::Scene::windowClosed(KWin::Toplevel*, KWin::Deleted*)") at assert.c:101 #4 0x7fc92501c81f in KWin::Scene::windowClosed (this=0x5638a2713d60, c=, deleted=0x5638a2ecbd40) at ./scene.cpp:414 #5 0x7fc9250edc43 in KWin::Scene::qt_static_metacall (_o=, _c=, _id=, _a=) at ./obj-x86_64-linux-gnu/kwin_autogen/EWIEGA46WW/moc_scene.cpp:98 #6 0x7fc922cfb195 in QMetaObject::activate (sender=sender@entry=0x5638a2d19c90, signalOffset=, local_signal_index=local_signal_index@entry=5, argv=argv@entry=0x7fff3fa8d320) at kernel/qobject.cpp:3767 #7 0x7fc922cfb867 in QMetaObject::activate (sender=sender@entry=0x5638a2d19c90, m=m@entry=0x7fc92539fc80 , local_signal_index=local_signal_index@entry=5, argv=argv@entry=0x7fff3fa8d320) at kernel/qobject.cpp:3629 #8 0x7fc9250e68de in KWin::Toplevel::windowClosed (this=this@entry=0x5638a2d19c90, _t1=, _t1@entry=0x5638a2d19c90, _t2=, _t2@entry=0x5638a2ecbd40) at ./obj-x86_64-linux-gnu/kwin_autogen/EWIEGA46WW/moc_toplevel.cpp:718 #9 0x7fc92508b9ac in KWin::ShellClient::destroyClient (this=0x5638a2d19c90) at ./shell_client.cpp:362 #10 0x7fc922cfb2af in QtPrivate::QSlotObjectBase::call (a=0x7fff3fa8d490, r=0x5638a2d19c90, this=0x5638a2d20430) at ../../include/QtCore/../../src/corelib/kernel/qobjectdefs_impl.h:378 #11 QMetaObject::activate (sender=sender@entry=0x5638a29c21d0, signalOffset=, local_signal_index=local_signal_index@entry=0, argv=argv@entry=0x7fff3fa8d490) at kernel/qobject.cpp:3750 #12 0x7fc922cfb867 in QMetaObject::activate (sender=sender@entry=0x5638a29c21d0, m=m@entry=0x7fc923124a60 , local_signal_index=local_signal_index@entry=0, argv=argv@entry=0x7fff3fa8d490) at kernel/qobject.cpp:3629 #13 0x7fc922cfb90f in QObject::destroyed (this=this@entry=0x5638a29c21d0, _t1=, _t1@entry=0x5638a29c21d0) at .moc/moc_qobject.cpp:214 #14 0x7fc922d020ca in QObject::~QObject (this=, __in_chrg=) at kernel/qobject.cpp:902 #15 0x7fc924127397 in KWayland::Server::XdgTopLevelV6Interface::~XdgTopLevelV6Interface (this=0x5638a29c21d0, __in_chrg=) at ./src/server/xdgshell_v6_interface_p.h:92 #16 KWayland::Server::XdgTopLevelV6Interface::~XdgTopLevelV6Interface (this=0x5638a29c21d0, __in_chrg=) at ./src/server/xdgshell_v6_interface_p.h:92 #17 0x7fc922cfbd20 in QObject::event (this=0x5638a29c21d0, e=) at kernel/qobject.cpp:1238 #18 0x7fc92328e6cc in QApplicationPrivate::notify_helper (this=this@entry=0x5638a260ae80, receiver=receiver@entry=0x5638a29c21d0, e=e@entry=0x7fc8f8006220) at kernel/qapplication.cpp:3732 #19 0x7fc923295e84 in QApplication::notify (this=0x7fff3fa8e800, receiver=0x5638a29c21d0, e=0x7fc8f8006220) at kernel/qapplication.cpp:3491 #20 0x7fc922ccb938 in QCoreApplication::notifyInternal2 (receiver=0x5638a29c21d0, event=event@entry=0x7fc8f8006220) at kernel/qcoreapplication.cpp:1050 #21 0x7fc922cce4cd in QCoreApplication::sendEvent (event=0x7fc8f8006220, receiver=) at ../../include/QtCore/../../src/corelib/kernel/qcoreapplication.h:234 #22 QCoreApplicationPrivate::sendPostedEvents (receiver=0x0, event_type=52, data=0x5638a2613330) at kernel/qcoreapplication.cpp:
[kdeplasma-addons] [Bug 358126] Nonexistent network interface displayed on graph
https://bugs.kde.org/show_bug.cgi?id=358126 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #3 from Bernhard Übelacker --- I think I hit the same or similar issue. I placed the network monitor widget on my panel ages ago. Using the configuration dialog somehow I ended with a section in ~/.config/plasma-org.kde.plasma.desktop-appletsrc like this: [Containments][2][Applets][18] immutability=1 plugin=org.kde.plasma.systemmonitor.net [Containments][2][Applets][18][Configuration][ConfigDialog] DialogHeight=540 DialogWidth=720 [Containments][2][Applets][18][Configuration][General] sources=network%2Finterfaces%2Feth0%2Ftransmitter%2Fdata,network%2Finterfaces%2Fenp37s0%2Ftransmitter%2Fdata,network%2Finterfaces%2Fenp6s0%2Ftransmitter%2Fdata Now if the interface is removed e.g. by switching the mainboard started again the configuration persists and is shown in the applet. Unfortunately the configuration dialog just shows existing interfaces and therefore nonexisting cannot be removed. So it would be nice to add in the config dialog also the interfaces currently shown in the applet. A workaround for me was to remove the applet and add it again that it got put on a new index. But the section "[Containments][2][Applets][18][Configuration][General]" stays in the file. Another workaround is probably just to logout and remove the section "[Configuration][General]" from ~/.config/plasma-org.kde.plasma.desktop-appletsrc Files containing the applet might be these: applets/systemmonitor/net/contents/ui/net.qml applets/systemmonitor/net/contents/ui/netConfig.qml So this might be reassigned to product "plasmashell", component "System Monitor". Seen on Debian Stretch, plasma-workspace 5.8.6-2.1. -- You are receiving this mail because: You are watching all bug changes.
[plasmashell] [Bug 370195] Bug#839716: plasma-widgets: For easier bug reports widgets should have an "About" Section containing the package they belong to.
https://bugs.kde.org/show_bug.cgi?id=370195 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #1 from Bernhard Übelacker --- It would already be very helpful if such an "About" section would show the shared object or qml file name. With that it would already be easier to identify the package that provides the file. If there are some unique translations used, one can try following page to get a hint to which package that belongs: https://l10n.kde.org/dictionary/search-translations.php -- You are receiving this mail because: You are watching all bug changes.
[systemsettings] [Bug 373628] systemsettings5 crashes when closing or going back
https://bugs.kde.org/show_bug.cgi?id=373628 --- Comment #99 from Bernhard Übelacker --- Just a note for completeness: Upstream bug [QTBUG-57714] got now closed and a fix got commited [1]. It is included since Qt release v5.9.0-beta2 and in the v5.8 branch waiting for the next v5.8.x release. And in duplicate [QTBUG-58562] is a commit mentioned [2] as the patch causing this issue which on the other side is said to be the fix for [361236]. [QTBUG-57714] https://bugreports.qt.io/browse/QTBUG-57714 [QTBUG-58562] https://bugreports.qt.io/browse/QTBUG-58562 [361236] https://bugs.kde.org/show_bug.cgi?id=361236#c113 [1] http://code.qt.io/cgit/qt/qtdeclarative.git/commit/?id=d438be92dd7068fef94ce98e1ec039fe0ef4f3b3 [2] http://code.qt.io/cgit/qt/qtdeclarative.git/commit/?id=4d375f3f2b95bfcb322402df3525db6cc7a723ce -- You are receiving this mail because: You are watching all bug changes.
[plasmashell] [Bug 359664] xembedsniproxy random crashes (w/core dumps)
https://bugs.kde.org/show_bug.cgi?id=359664 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #3 from Bernhard Übelacker --- I received a similar crash in Debian Stretch with plasma-workspace and plasma-workspace-dbgsym in version 4:5.8.6-2. At the time it crashed I assume just kradio4 tray icon was running additionally to the default ones. (Saw just later the crash in dmesg.) # coredumpctl gdb Core was generated by `/usr/bin/xembedsniproxy'. Program terminated with signal SIGSEGV, Segmentation fault. #0 SNIProxy::getImageNonComposite (this=this@entry=0x5636e8f4b460) at ./xembed-sni-proxy/sniproxy.cpp:273 273 QImage naiveConversion = QImage(image->data, image->width, image->height, QImage::Format_ARGB32); [Current thread is 1 (Thread 0x7f15c8ff25c0 (LWP 1559))] (gdb) bt #0 SNIProxy::getImageNonComposite (this=this@entry=0x5636e8f4b460) at ./xembed-sni-proxy/sniproxy.cpp:273 #1 0x5636e7509507 in SNIProxy::update (this=0x5636e8f4b460) at ./xembed-sni-proxy/sniproxy.cpp:214 #2 0x5636e7506583 in FdoSelectionManager::nativeEventFilter (this=0x7fff86289810, eventType=..., message=, result=) at ./xembed-sni-proxy/fdoselectionmanager.cpp:154 #3 0x7f15c7b0ab0f in QAbstractEventDispatcher::filterNativeEvent(QByteArray const&, void*, long*) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 #4 0x7f15beaa2334 in QXcbConnection::handleXcbEvent(xcb_generic_event_t*) () from /usr/lib/x86_64-linux-gnu/libQt5XcbQpa.so.5 #5 0x7f15beaa3015 in QXcbConnection::processXcbEvents() () from /usr/lib/x86_64-linux-gnu/libQt5XcbQpa.so.5 #6 0x7f15c7b3a499 in QObject::event(QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 #7 0x7f15c7b0d87a in QCoreApplication::notify(QObject*, QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 #8 0x7f15c7b0d9e0 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 #9 0x7f15c7b1016d in QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 #10 0x7f15c7b61c43 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 #11 0x7f15c3b757f7 in g_main_context_dispatch () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #12 0x7f15c3b75a60 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #13 0x7f15c3b75b0c in g_main_context_iteration () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #14 0x7f15c7b6204f in QEventDispatcherGlib::processEvents(QFlags) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 #15 0x7f15c7b0b9ca in QEventLoop::exec(QFlags) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 #16 0x7f15c7b1413c in QCoreApplication::exec() () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5 #17 0x5636e7503972 in main (argc=, argv=) at ./xembed-sni-proxy/main.cpp:68 (gdb) list SNIProxy::getImageNonComposite ... 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); (gdb) print/x image $1 = (gdb) disassemble SNIProxy::getImageNonComposite ... 0x5636e750922a <+106>: mov%rax,%r12 0x5636e750922d <+109>: callq 0x5636e7502bf0 => 0x5636e7509232 <+114>: movzwl 0x2(%rax),%ecx 0x5636e7509236 <+118>: movzwl (%rax),%edx (gdb) print/x $ecx $3 = 0xc6f61ffa (gdb) print/x $rax $4 = 0x0 As far as I found the callq instruction is supposed to write the return value into the rax register. So I assume function xcb_image_get returned a NULL pointer. Shouldn't there be a "if (image)" before the QImage construction? Are some more informations needed from the core dump? Kind regards, Bernhard -- You are receiving this mail because: You are watching all bug changes.
[systemsettings] [Bug 373628] systemsettings5 crashes when closing or going back
https://bugs.kde.org/show_bug.cgi?id=373628 --- Comment #67 from Bernhard Übelacker --- Just as a information. Attached to the upstream bug QTBUG-57714 is a patch against qtdeclarative-opensource-src-5.7.1 to attempt to fix the problem there. In my short tests it worked as expected. -- You are receiving this mail because: You are watching all bug changes.
[ksmserver] [Bug 369391] ksmserver captures and discards standard output/error - debug information lost
https://bugs.kde.org/show_bug.cgi?id=369391 Bernhard Übelacker changed: What|Removed |Added CC||bernha...@mailbox.org --- Comment #1 from Bernhard Übelacker --- Created attachment 103756 --> https://bugs.kde.org/attachment.cgi?id=103756&action=edit ksmserver: Make sure stdout and stderr gets forwarded to .xsession-errors. In my Debian Stretch system I noticed that no logging from e.g. powerdevil arrives in .xsession-errors. Just before opening a new bug I found that Jonathan Marten already described just the exact same problem. Please find attached a patch against Debian version of plasma-workspace_5.8.4-1, enabling the forwarding to ksmservers stdout/stderr. -- You are receiving this mail because: You are watching all bug changes.
[konsole] [Bug 369050] Konsole crashs randomly.
https://bugs.kde.org/show_bug.cgi?id=369050
--- Comment #19 from Bernhard Übelacker ---
Created attachment 103561
--> https://bugs.kde.org/attachment.cgi?id=103561&action=edit
gdb session with some pretty-printers showing the event details.
Hello,
It took some time to get the pretty-printers in place.
Please see attached file for more informations on the
QDBusCallDeliveryEvent object.
Who is sending the message I could not find out.
Right now no "org.a11y.atspi.Registry" is visible
in qdbusviewer, just a "org.a11y.Bus" ?
At least in [1] Qt5 seems to listen to this
"EventListenerDeregistered" message.
The message member seems to contain a member arguments [2],
a QList, containing 2 QVariant of type = 10.
type = 10 == QVariant::String == QMetaType::QString ?
The error message:
Internal error: got invalid meta type 11 (QStringList)
when trying to convert to meta type 10 (QString)
So it looks like the arguments got converted in [3] somehow to
a QStringList instead a QString, and do therefore not
match the "eventListenerDeregistered" signature anymore.
Kind regards,
Bernhard
[1]
qtbase-opensource-src-5.7.1+dfsg/src/platformsupport/linuxaccessibility/atspiadaptor.cpp:140
success = success &&
m_dbus->connection().connect(QLatin1String("org.a11y.atspi.Registry"),
QLatin1String("/org/a11y/atspi/registry"),
QLatin1String("org.a11y.atspi.Registry"),
QLatin1String("EventListenerDeregistered"), this,
SLOT(eventListenerDeregistered(QString,QString)));
[2]
(gdb) print *(e->message.d_ptr)
$8 = {
arguments = QList = {
[0] = {
d = {
data = {
..
ptr = 0x7fb1100117f0,
shared = 0x7fb1100117f0
},
type = 10,
is_shared = 0,
is_null = 0
}
},
[1] = {
d = {
data = {
..
ptr = 0x7fb1100172c0,
shared = 0x7fb1100172c0
},
type = 10,
is_shared = 0,
is_null = 0
}
}
},
...
interface = "org.a11y.atspi.Registry",
name = "EventListenerDeregistered",
...
[3] qtbase-opensource-src-5.7.1+dfsg/src/dbus/qdbusintegrator.cpp:935
const QVariant &arg = msg.arguments().at(i - 1);
if (arg.userType() == id)
...
else if (arg.userType() == qMetaTypeId()) {
...
} else {
qFatal("Internal error: got invalid meta type %d (%s) "
"when trying to convert to meta type %d (%s)",
arg.userType(), QMetaType::typeName(arg.userType()),
id, QMetaType::typeName(id));
}
--
You are receiving this mail because:
You are watching all bug changes.
[konsole] [Bug 369050] Konsole crashs randomly.
https://bugs.kde.org/show_bug.cgi?id=369050
Bernhard Übelacker changed:
What|Removed |Added
CC||bernha...@mailbox.org
--- Comment #18 from Bernhard Übelacker ---
I think I got the same issue here.
It happened when I switched a some video playing in VLC (started via dolphin).
When switching back I noticed my konsoles did disappear.
(In reply to Martin Sandsmark from comment #17)
> Can someone see if they can find the actual error message ... or in
> .xsession-errors?
Internal error: got invalid meta type 11 (QStringList) when trying to convert
to meta type 10 (QString)
KCrash: crashing... crashRecursionCounter = 2
KCrash: Application Name = kdeinit5 path = /usr/bin pid = 12699
KCrash: Arguments: /usr/bin/kdeinit5 --workdir /home/bernhard/data
I still have the process existing and trying to inspect
with gdb with debian debug symbols installed.
Unfortunately nearly all that I want to see is ...
...
(gdb) up
#12 0x7fb12f591b2c in QApplicationPrivate::notify_helper (this=, receiver=0xed8120, e=0x7fb110016490) at kernel/qapplication.cpp:3799
3799bool consumed = receiver->event(e);
(gdb) print *e
$57 = (QDBusCallDeliveryEvent) { = { = {_vptr.QEvent =
0x7fb127d49eb0 , d = 0x0, t = 43, posted
= 0, spont = 0, m_accept = 1, reserved = 0}, slotObj_ = 0x0, sender_ =
0x7fb110012500, signalId_ = -1, nargs_ = 0, types_ = 0x0, args_ = 0x0,
semaphore_ = 0x0, callFunction_ = 0x0, method_offset_ = 0, method_relative_ =
6}, connection = {d = 0x7fb110012500}, message = {d_ptr = 0x7fb1100139d0},
metaTypes = {d = 0xad1490}, flags = 0}
(gdb) print *(e->message.d_ptr)
$60 = {arguments = {> = {}, {p =
{d = 0x7fb110017290}, d = 0x7fb110017290}}, service = {d = 0x7fb110017260},
path = {d = 0x7fb110017200}, interface = {d = 0x7fb1100171b0}, name = {d =
0x7fb1100170f0}, message = {d = 0x7fb130da6e80 },
signature = {d = 0x7fb1100173a0}, msg = 0x7fb110012d60, reply = 0x0, localReply
= 0x0, ref = {> = {> = {_q_value =
{> = {_M_i = 1}, }}, },
}, type = QDBusMessage::SignalMessage, delayedReply = 0,
localMessage = 0, parametersValidated = 0, autoStartService = 1}
(gdb) print *(e->metaTypes.d)
$63 = { = {ref = {atomic = {_q_value = {> =
{_M_i = 2}, }}}, size = 3, alloc = 10, capacityReserved = 0,
offset = 24}, }
(gdb) printq5string e->message.d_ptr.name
EventListenerDeregistered
(gdb) printq5string e->message.d_ptr.service
:1.7
(gdb) printq5string e->message.d_ptr.path
/org/a11y/atspi/registry
(gdb) printq5string e->message.d_ptr.interface
org.a11y.atspi.Registry
(gdb) printq5string e->message.d_ptr.signature
ss
Is there anything else someone is interested in while the process still exists?
--
You are receiving this mail because:
You are watching all bug changes.
[kscreenlocker] [Bug 373628] systemsettings5 crashes when closing or going back from lock screen configuration.
https://bugs.kde.org/show_bug.cgi?id=373628 --- Comment #3 from Bernhard Übelacker --- Thank you for the feedback. I created https://bugreports.qt.io/browse/QTBUG-57714. -- You are receiving this mail because: You are watching all bug changes.
[kscreenlocker] [Bug 373628] systemsettings5 crashes when closing or going back from lock screen configuration.
https://bugs.kde.org/show_bug.cgi?id=373628 --- Comment #1 from Bernhard Übelacker --- Created attachment 102875 --> https://bugs.kde.org/attachment.cgi?id=102875&action=edit Minimal testcase This test creates a main window with a single QQuickWidget. Via setContextProperty the pointer to the MainWindow instance is stored. If at Qml side a property of this instance is accessed we get this crash in the destructor. (An exmple gdb run on my machine is included.) -- You are receiving this mail because: You are watching all bug changes.
[kscreenlocker] [Bug 373628] New: systemsettings5 crashes when closing or going back from lock screen configuration.
https://bugs.kde.org/show_bug.cgi?id=373628 Bug ID: 373628 Summary: systemsettings5 crashes when closing or going back from lock screen configuration. Product: kscreenlocker Version: unspecified Platform: Debian testing OS: Linux Status: UNCONFIRMED Severity: normal Priority: NOR Component: kcm Assignee: plasma-b...@kde.org Reporter: bernha...@mailbox.org CC: bhus...@gmail.com, mgraess...@kde.org Target Milestone: --- Created attachment 102773 --> https://bugs.kde.org/attachment.cgi?id=102773&action=edit Debugging session. On my Debian testing desktop I noticed a crash when browsing in systemsettings to the lock screen configuration and trying to go back or closing the application. Please see attached file at the end for backtrace of the crash. After "some time" looking into it I believe that the crash occours because of the following events: - ScreenLockerKcmForm constructor calls QQmlContext::setContextProperty (configDialog). That way a pointer to ScreenLockerKcmForm get stored as some Qml/QV4 objects (QObjectWrapper). - ScreenLockerKcmForm destructor is called. - The inherited QWidget destructor is called. - The QObjectPrivate member declarativeData is set to 0 (qwidget.cpp:1679) - QObjectPrivate::deleteChildren is called. - The QObjectPrivate member currentChildBeingDeleted is set to the next one to be deleted. (qobject.cpp:1968) Unfortunately is declarativeData and currentChildBeingDeleted sharing the same memory - therefore declarativeData is now not 0 anymore. (qobject_p.h:236) - One of the childs tries to clean up the Qml/QV4 magic. - Now in QObjectWrapper::destroyObject a QQmlData pointer is retrieved. That way the QObject pointer currentChildBeingDeleted is now retrieved via declarativeData as a QQmlData pointer. Can someone at least reproduce the crash? Is there something I should include in another debugging session? I think this also is somehow related to the crash filed in bug #361236. At least it ends in the same QV4::Value::as member (qv4value_p.h:351). Kind regards, Bernhard -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 361236] Aurorae crashes in QQmlBinding::write on creation - Qt 5.6
https://bugs.kde.org/show_bug.cgi?id=361236 --- Comment #66 from Bernhard Übelacker --- After the latest updates in Debian Testing I am not able to reproduce the crash anymore by switching from Breeze to Plastik decorations. The updates contained mostly Qt from 5.6.1 to 5.7.1~20161021. And some packages libkf5* and kwin* got just rebuilt at the same version. -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 361236] Aurorae crashes in QQmlBinding::write on creation - Qt 5.6
https://bugs.kde.org/show_bug.cgi?id=361236
--- Comment #59 from Bernhard Übelacker ---
Tried to find out what in method QV4::Value::as should happen.
I _think_ the loop is entered with more or less the _val member
casted to const QV4::VTable *.
At least when I try this manually I end up with the same invalid pointer.
So it looks like the parents parent pointer got truncated somehow:
0x7fbc4810
0x7fbc48105d72
QV4::Value::as
const VTable *vt = m()->vtable();
while (vt) {
if (vt == T::staticVTable())
return static_cast(this);
vt = vt->parent;
}
return 0;
# gdb -q --pid 9545
...
(gdb) set height 0
(gdb) set width 0
(gdb) directory /path/to/qtdeclarative-opensource-src-5.6.1/src/qml
...
(gdb) source /usr/share/doc/kdesdk-scripts/examples/kde-devel-gdb
(gdb) up
...
(gdb)
#6 QV4::Value::as (this=0x7fbc2d3bd320) at
../../include/QtQml/5.6.1/QtQml/private/../../../../../src/qml/jsruntime/qv4value_p.h:351
warning: Source file is more recent than executable.
351 vt = vt->parent;
...
(gdb) print vt
$1 = (const QV4::VTable *) 0x7fbc4810
(gdb) print *vt
Cannot access memory at address 0x7fbc4810
(gdb) up
#7 QQmlBinding::write (this=this@entry=0x55a99a0, core=..., result=...,
isUndefined=, flags=...) at qml/qqmlbinding.cpp:257
257 if (const QV4::QQmlValueTypeWrapper *vtw = result.as()) {
(gdb) print/x result._val
$2 = 0x7fbc48105c30
(gdb) print/x *(const QV4::VTable *)result._val
$3 = {parent = 0x7fbc48105d72, isExecutionContext = 0x0, isString = 0x0,
isObject = 0x0, isFunctionObject = 0x0, isErrorObject = 0x0, isArrayData = 0x0,
unused = 0x0, type = 0x0, className = 0x0, destroy = 0x0, markObjects = 0x0,
isEqualTo = 0x0}
(gdb) print/x ((const QV4::VTable *)result._val)->parent
$4 = 0x7fbc48105d72
(gdb) print/x *(((const QV4::VTable *)result._val)->parent)
$5 = {parent = 0x7fbc4810, isExecutionContext = 0x0, isString = 0x0, isObject =
0x0, isFunctionObject = 0x0, isErrorObject = 0x0, isArrayData = 0x0, unused =
0x0, type = 0x0, className = 0x0, destroy = 0x0, markObjects = 0x0, isEqualTo =
0x0}
(gdb) print/x (((const QV4::VTable *)result._val)->parent)->parent
$6 = 0x7fbc4810
(gdb) print/x *const QV4::VTable *)result._val)->parent)->parent)
Cannot access memory at address 0x7fbc4810
(gdb) print &QV4::QQmlValueTypeWrapper::static_vtbl
$9 = (const QV4::ObjectVTable *) 0x7fbc5e97f100
--
You are receiving this mail because:
You are watching all bug changes.
[kwin] [Bug 361236] Aurorae crashes in QQmlBinding::write on creation - Qt 5.6
https://bugs.kde.org/show_bug.cgi?id=361236
Bernhard Übelacker changed:
What|Removed |Added
CC||bernha...@mailbox.org
--- Comment #58 from Bernhard Übelacker ---
I received the same crash in current Debian Stretch.
(qtbase5 5.6.1, kwin-x11, 5.8.2, was a dist-upgrade from Jessie)
For me it is reproducible on every switch from Breeze to Plastik
window decorations.
apt-get install kwin-x11-dbgsym qtbase5-dbg kwin-common-dbgsym
qtdeclarative5-dbg qtdeclarative5-dev
(gdb) up
#7 QQmlBinding::write (this=this@entry=0x55a99a0, core=..., result=...,
isUndefined=, flags=...) at qml/qqmlbinding.cpp:257
257 if (const QV4::QQmlValueTypeWrapper *vtw = result.as()) {
(gdb) print core.propType
$7 = 39
/* seems to be /* QMetaType::QObjectStar */
(gdb) down
#6 QV4::Value::as (this=0x7fbc2d3bd320) at
../../include/QtQml/5.6.1/QtQml/private/../../../../../src/qml/jsruntime/qv4value_p.h:351
351 vt = vt->parent;
(gdb) print vt
$11 = (const QV4::VTable *) 0x7fbc4810
(gdb) print *vt
Cannot access memory at address 0x7fbc4810
Is there anything I can supply (other than a small test case)?
--
You are receiving this mail because:
You are watching all bug changes.
