20:20:58frgo::notify drmeister There is a paper talking about emitting vtables from clang: https://arxiv.org/pdf/2003.04228.pdf
20:20:58Colleenfrgo: Got it. I'll let drmeister know as soon as possible.
20:23:23frgodrmeister: I assume you are aware of -fforce-emit-vtables ...
20:31:39drmeisterfrgo: The paper doesn't appear to talk about exporting vtable symbols.
20:31:39Colleendrmeister: frgo said 10 minutes, 41 seconds ago: There is a paper talking about emitting vtables from clang: https://arxiv.org/pdf/2003.04228.pdf
20:31:53drmeisterI'm trying the -fforce-emit-vtables option - I was not aware of that.
20:32:29frgoFrom reading that paper and then checking the command line options of clang and searching for vtable
20:35:51drmeisterI'm trying that option to see if it does anything wrt exporting vtable symbols.
20:36:20drmeisterThis paper is directly relevant to my problem - it suggests new C++ syntax: http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2018/p1263r0.pdf
20:46:30drmeisterNo - sadly, the -fforce-emit-vtables option does not change the linkage of the vtable symbols.
20:47:13drmeisterI've built a small test case and I'm wrestling with waf at the moment to relink the executable with an option that would export the symbols I need.