Comment # 17 on bug 1141883 from Fabian Vogt

(In reply to Fabian Vogt from comment #15)
> (In reply to Martin Li��ka from comment #14)
> > Fabian: Any guess what can be broken based on the back-trace?
> 
> I made a patch: https://phabricator.kde.org/D22719
> 
> I don't know why it only breaks with LTO though.

While the patch is correct (it's possible to have argc == 0), I think the root
cause for pinentry-qt is something else though. As during QApplication
construction, argc = 1 and stored inside QCoreApplicationPrivate, it shouldn't
suddenly change to 0 during teardown. I'll have a closer look tomorrow.