Experimental Physics and
Industrial Control System

Michael Davidsaver via Core-talk <core-talk at aps.anl.gov> · Fri, 24 Apr 2020 14:36:55 -0700

FYI. glibc at least seems not to follow the required ordering when destructing
thread_local objects on the main thread.

I read the language at

https://en.cppreference.com/w/cpp/utility/program/exit

as saying that thread_local destruction and atexit callbacks should be mixed
based on the reverse order in of constructors and atexit() calls.
However, glibc strictly destroys all thread_local before running
any atexit callbacks.

https://github.com/bminor/glibc/blob/5f72f9800b250410cad3abfeeb09469ef12b2438/stdlib/exit.c#L38

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System