[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-11 Thread Thiago Macieira
https://bugs.kde.org/show_bug.cgi?id=510785

Thiago Macieira  changed:

   What|Removed |Added

Version|QT 6.11.1,6.12.0|Qt 6.12.0,6.11.1,6.10.3
  Fixed/Implemented||
 In||

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-11 Thread TraceyC
https://bugs.kde.org/show_bug.cgi?id=510785

TraceyC  changed:

   What|Removed |Added

Version||QT 6.11.1,6.12.0
  Fixed/Implemented||
 In||

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-10 Thread Christoph Cullmann
https://bugs.kde.org/show_bug.cgi?id=510785

--- Comment #17 from Christoph Cullmann  ---
(In reply to Thiago Macieira from comment #15)
> Submitted a workaround in Qt.

Thanks!

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-10 Thread TraceyC
https://bugs.kde.org/show_bug.cgi?id=510785

--- Comment #16 from TraceyC  ---
Thanks for working through this, it's going to avoid a lot of users submitting
these error lines in unrelated bug reports.

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-09 Thread Thiago Macieira
https://bugs.kde.org/show_bug.cgi?id=510785

--- Comment #15 from Thiago Macieira  ---
Submitted a workaround in Qt.

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-09 Thread Christoph Cullmann
https://bugs.kde.org/show_bug.cgi?id=510785

--- Comment #14 from Christoph Cullmann  ---
KAboutData::processCommandLine  

shows the issue for

❯ kwrite --author 
qt.multimedia.symbolsresolver: Couldn't load pipewire-0.3 library
qt.multimedia.symbolsresolver: Couldn't resolve pipewire-0.3 symbols
kwrite was written by:
Christoph Cullmann 
Dominik Haumann 
Sven Brauch 
Kåre Särs 
Waqar Ahmed 
Anders Lund 
Joseph Wenninger 
Hamish Rodda 
Alexander Neundorf 
Waldo Bastian 
Charles Samuels 
Matt Newell 
Michael Bartl 
Michael McCallum 
Jochen Wilhemly 
Michael Koch 
Christian Gebauer 
Simon Hausmann 
Glen Parker 
Scott Manson 
John Firebaugh 
Pablo Martín 
Gerald Senarclens de Grancy 
Please use https://bugs.kde.org to report bugs.
QThreadStorage: entry 1 destroyed before end of thread 0x55744d9dfc20
QThreadStorage: entry 0 destroyed before end of thread 0x55744d9dfc20

If we use the non-cleanup _exit instead, we avoid that, naturally, that seems
not like a good fix.

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-09 Thread Christoph Cullmann
https://bugs.kde.org/show_bug.cgi?id=510785

--- Comment #13 from Christoph Cullmann  ---
(In reply to Christoph Cullmann from comment #12)
> (In reply to Thiago Macieira from comment #11)
> > From QCommandLineParser::showVersion()
> 
> The code there does some better cleanup, should we try to do the same in our
> framework?

Ok, I see, that crashes more or less in QCommandLineParser::showVersion() like
we crash in KAboutData::processCommandLine for the --help stuff.

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-09 Thread Christoph Cullmann
https://bugs.kde.org/show_bug.cgi?id=510785

--- Comment #12 from Christoph Cullmann  ---
(In reply to Thiago Macieira from comment #11)
> From QCommandLineParser::showVersion()

The code there does some better cleanup, should we try to do the same in our
framework?

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-09 Thread Thiago Macieira
https://bugs.kde.org/show_bug.cgi?id=510785

--- Comment #11 from Thiago Macieira  ---
>From QCommandLineParser::showVersion()

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-09 Thread Christoph Cullmann
https://bugs.kde.org/show_bug.cgi?id=510785

--- Comment #10 from Christoph Cullmann  ---
(In reply to Thiago Macieira from comment #9)
> Most likely, this is the result of a Q_GLOBAL_STATIC(QThreadStorage, y)
> that was already destroyed. It's difficult to get this right because
> Q_GLOBAL_STATIC will destroy early, before the thread exit. But it's also
> the right thing to do.
> 
> The other thing is this warning is suppressed when
> QCoreApplication::instance() is null. So why is it not null during
> application exit?

I think the issue is that stuff like

'kate --version'

triggers plain exit.

(same for other variants that exit early)

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-09 Thread TraceyC
https://bugs.kde.org/show_bug.cgi?id=510785

TraceyC  changed:

   What|Removed |Added

   See Also||https://bugs.kde.org/show_b
   ||ug.cgi?id=502874

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-09 Thread Thiago Macieira
https://bugs.kde.org/show_bug.cgi?id=510785

Thiago Macieira  changed:

   What|Removed |Added

 CC||[email protected]

--- Comment #9 from Thiago Macieira  ---
Most likely, this is the result of a Q_GLOBAL_STATIC(QThreadStorage, y) that
was already destroyed. It's difficult to get this right because Q_GLOBAL_STATIC
will destroy early, before the thread exit. But it's also the right thing to
do.

The other thing is this warning is suppressed when QCoreApplication::instance()
is null. So why is it not null during application exit?

-- 
You are receiving this mail because:
You are watching all bug changes.

[plasmashell] [Bug 510785] QThreadStorage entry n destroyed before end of thread errors on command line after Qt applications exit normally

2026-03-09 Thread TraceyC
https://bugs.kde.org/show_bug.cgi?id=510785

TraceyC  changed:

   What|Removed |Added

  Component|general |general
  Version First|25.08.2 |6.4.5
Reported In||
Product|konsole |plasmashell
   Target Milestone|--- |1.0
 CC||[email protected],
   ||[email protected]
   Assignee|[email protected]   |[email protected]
URL||https://qt-project.atlassia
   ||n.net/browse/QTBUG-144862
Summary|QThreadStorage errors after |QThreadStorage entry n
   |running "konsole --help"|destroyed before end of
   ||thread  errors on command
   ||line after Qt applications
   ||exit normally

--- Comment #8 from TraceyC  ---
The same lines are shown with just konsole --help and other programs, so it's
not specifically a konsole bug. It looks like a Qt bug.
I've opened a bug with the Qt folks for this. 
https://qt-project.atlassian.net/browse/QTBUG-144862

Also see
https://discuss.kde.org/t/qthreadstorage-spamming/34576

These lines are emitted by plasma-discover as well, with just the --help flag.
Also with gwenview --help, dolphin --help, digikam --help and probably others.
This probably affects any Qt app.
My system has Qt 6.10.2

ex:
% plasma-discover --help   
Usage: plasma-discover [options] urls
An application explorer
(snip)

Arguments:
  urlsSupports appstream: url scheme
QThreadStorage: entry 2 destroyed before end of thread 0x5576d11c9c40
QThreadStorage: entry 1 destroyed before end of thread 0x5576d11c9c40

A quick Bugzilla search shows 25 open bugs with the same string, appearing in
comments as early as April 2025

https://bugs.kde.org/buglist.cgi?bug_status=UNCONFIRMED&bug_status=CONFIRMED&bug_status=ASSIGNED&bug_status=REOPENED&f0=OP&f1=OP&f2=product&f3=component&f4=alias&f5=short_desc&f7=content&f8=CP&f9=CP&j1=OR&list_id=3458084&longdesc=destroyed%20before%20end%20of%20thread&longdesc_type=substring&o2=substring&o3=substring&o4=substring&o5=substring&o6=substring&o7=matches&query_format=advanced&v2=QThreadStorage&v3=QThreadStorage&v4=QThreadStorage&v5=QThreadStorage&v6=QThreadStorage&v7=%22QThreadStorage%22

-- 
You are receiving this mail because:
You are watching all bug changes.