D26764: Another revert to unbreak crash fix

[Konstantin Kharlamov]

kkharlamov added a comment.


  In D26764#600607 <https://phabricator.kde.org/D26764#600607>, @anthonyfieroni wrote:
  
  > Actually remove all atexit usage, it does not needed at all
  >  https://en.cppreference.com/w/cpp/utility/program/exit
  >
  >   The destructors of objects with thread local storage duration that are associated with the current thread, the destructors of objects with static storage duration, and the functions registered with std::atexit are executed concurrently, while maintaining the following guarantees:
  
  
  Right, but as I mentioned doing that causes another crash. Stacktrace:
  
    #0  0x0000000000000018 in ?? ()
    #1  0x00007ffff78cd1d6 in ?? () from /usr/lib/libKF5GlobalAccel.so.5
    #2  0x00007ffff5cb8792 in qt_call_post_routines() () from /usr/lib/libQt5Core.so.5
    #3  0x00007ffff67f5b37 in QApplication::~QApplication() () from /usr/lib/libQt5Widgets.so.5
    #4  0x00007ffff553d6a7 in __run_exit_handlers () from /usr/lib/libc.so.6
    #5  0x00007ffff553d85e in exit () from /usr/lib/libc.so.6
    #6  0x00007ffff552615a in __libc_start_main () from /usr/lib/libc.so.6
    #7  0x000055555555506e in _start ()
  
  Btw, if somebody wants to make it work,  I pushed the code here <https://invent.kde.org/konstantink/konsole/commit/1e2e0a1cb03aec1be71ad473d4d54fd6c5a46696>. If you make it work, you can take ownership over the changes too, I don't mind.

REPOSITORY
  R319 Konsole

REVISION DETAIL
  https://phabricator.kde.org/D26764

To: sandsmark, #konsole, hindenburg, kkharlamov
Cc: anthonyfieroni, konsole-devel, hindenburg, #konsole, hiangel, cent, EspadaV8, fbampaloukas, thsurrel, ngraham, maximilianocuria
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/konsole-devel/attachments/20200125/a73baf71/attachment-0001.html>