[Bug c++/92731] Data race on exception object thrown from std::future

2019-11-29 Thread rrrlasse at hotmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92731 --- Comment #4 from Lasse Reinhold --- Some simplifications, and also tried with atomics instead of mutexes, to no avail: #include #include #include #include std::function job; std::atomic has_job{ false }; int main() { std::thread t =

[Bug c++/92731] Data race on exception object thrown from std::future

2019-11-29 Thread rrrlasse at hotmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92731 --- Comment #2 from Lasse Reinhold --- As far as I remember I could only add a single file when I created the bug report, and the option to add more showed up after completion. Anyway, the c++ code isn't very idiomatic, but I can't see why it wo

[Bug c++/92731] New: Data race on exception object thrown from std::future

2019-11-29 Thread rrrlasse at hotmail dot com
Component: c++ Assignee: unassigned at gcc dot gnu.org Reporter: rrrlasse at hotmail dot com Target Milestone: --- Created attachment 47397 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=47397&action=edit TSan output and .ii file The source is compiled with: g++

[Bug c++/81152] False strict-aliasing warning

2017-06-21 Thread rrrlasse at hotmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81152 --- Comment #1 from Lasse Reinhold --- Created attachment 41598 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41598&action=edit Non-stack overflowing version

[Bug c++/81152] New: False strict-aliasing warning

2017-06-21 Thread rrrlasse at hotmail dot com
++ Assignee: unassigned at gcc dot gnu.org Reporter: rrrlasse at hotmail dot com Target Milestone: --- Version 8.0.0 20170620 (and also version 7.1) gives following false warning: me@ubuntu:~$ g++ -Wall -Wextra -O3 -std=c++14 -fstrict-aliasing -Wstrict-aliasing /mnt/hgfs/D/c.cpp /mnt/hgfs/D