5:57:34drmeisterpaulapatience[m]: I got a response from Lang Hames (the developer of the new ORC JIT in llvm) "I think that was bad internal logic in LLJIT in LLVM 9.0. It should be fixed in LLVM 11 (it may be fixed in LLVM 10, but I don't recall)"
10:32:27selwyndoes clasp use a specific endianness everywhere? if so, what is it?
15:53:24Bikeso although it seems like backtraces are going to be a mess for the forseeable future, we ought to come up with a stable interface for accessing the information.
15:55:15Bikeso that we don't have to push a new commit to slime every week when some llvm change renders our previous approach futile, that sort of thing
16:19:18drmeisterWhat is the issue with backtraces at the moment?
16:19:54Bikenothing specific, we've just changed everything around a lot
16:20:37drmeisterLately yes. Does this need a good EXT API?
16:21:34drmeisterWe have everything we need now to create a public facing API - I believe.
16:22:19Bikeyes, we should have a public facing API in ext: or possibly its own package.