https://bugs.kde.org/show_bug.cgi?id=528089
--- Comment #6 from Russell Haley <[email protected]> --- Created attachment 198116 --> https://bugs.kde.org/attachment.cgi?id=198116&action=edit thread stacks at time 1, when right panel is gone I tried to attach gdb to the running /usr/bin/plasmashell pid, and it spent several minutes downloading, growing my ~/.cache/debuginfod_client to over 7 GiB (from maybe 1.5 GiB? IDK why it downloaded more than when inspecting with coredumpctl.). Then gdb itself crashed, with > /usr/include/c++/16/string_view:291: constexpr const > std::basic_string_view<_CharT, _Traits>::value_type& > std::basic_string_view<_CharT, _Traits>::back() const [with _CharT = char; > _Traits = std::char_traits<char>; const_reference = const char&]: Assertion > 'this->_M_len > 0' failed. It *does not* fail if I don't let it use debuginfod. I attached gdb, `continue`'d, set the display-off timeout to 1 minute, and waited for the screen to blank. The first time, the problem failed to reproduce because I was on a virtual desktop with only 6 windows open (firefox, gvim, 2x konsole, abrt-applet, drkonqi-coredump-gui). Apparently having lots of open windows open (some GPU resource?) is a necessary precondition. I tried again on a viritual desktop with 34 windows, and waited about 2 hours after the screen blanked to wake it up again, ^C gdb, and capture backtraces. The first is immediately after wakeup. In this state, my right monitor (the displayport one that falls off the bus) has no panel, and all of the windows from both monitors are shown in the left monitor's panel. -- You are receiving this mail because: You are watching all bug changes.
