Jonathan Wakely

If libgstreamer-1.0.so.0 includes the linunwind symbols then it should use a linker script to avoid them being exported. Otherwise you get incompatible symbols with the same name.

> You could probably omit it and say: Yes, the same thought had occurred to me. I'll push an update to the branch.

> I do not know if this is a g++ behaviour exceptional to the standard C++ compilers. Why are you using g++ to compile this, it's C code? The meaning...

If you are using `gcc` not `g++` then the problem is likely related to https://gcc.gnu.org/gcc-5/porting_to.html#inline

> If you are using `gcc` not `g++` then the problem is likely related to https://gcc.gnu.org/gcc-5/porting_to.html#inline So then it's probably this ^^ That link explains how to make code written...

If you're linking a mix of C and C++ then using `g++` to link is generally correct. That's not the problem here, the problem is the change to what `inline`...

@sumariva is using `CFLAGS="-I$HOME/include"` If that overrides the default CFLAGS and replaces any optimization options like `-O2` then it means the code will be compiled with no optimization, and so...

The `-I` option is a preprocessor option, so use `CPPFLAGS` for that, not `CFLAGS`, and not `CPP_FLAGS`.

> I’m not certain about the copyright of this — ISO says that they only hold copyright over published versions, but I couldn’t find information about copyright of drafts. https://en.cppreference.com/w/cpp/language/memory_model...

Submitting it here is fine - makes it easier for me to fetch the change and merge it locally. I haven't looked at the patch but it sounds sane, I'll...