This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: __cxa_atexit question
> No, I was thinking of putting actual definitions of __cxa_atexit and
> __cxa_finalize (and atexit) in libstdc++.
I don't think atexit is appropriate. However, the other two seem like
candidates. Without the atexit hooks seems kinda pointless tho.
From:
http://www.codesourcery.com/cxx-abi/abi.html#obj-ctor
"It is expected that implementations supporting both C and C++ will integrate this capability into the libc atexit implementation so that C-only DSOs will nevertheless interact with C++ programs in a C++-standard-conforming manner. No user interface to __cxa_atexit is supported, so the user is not able to register an atexit function with a parameter or a home DSO."
At this point, there are a lot of conditional libsupc++ symbols
defined, although most of the extensions to this point are for ARM. If
you persist in this approach for the __cxa_atexit issue, I'd like to
take the opportunity to clean up the ARM bits, and make the ARM exports
target-dependent, which I've been threatening to do for some time.
-benjamin