https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58909
--- Comment #7 from Brooks Moses <brooks at gcc dot gnu.org> --- Right, understood, but Roland seemed pretty insistent in the discussion on bug 57740 that this was a libstdc++ bug, not a glibc bug. Your contention is that this is invalid because he's wrong about that, I take it? And that rather than libstdc++ documenting workarounds, this should be fixed by the "libpthread.a is a single object file" fix? Unfortunately, it seems fairly clear that filing this as a glibc bug would have Roland closing it as invalid there, and this will simply not get fixed. :(
