https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116058
--- Comment #13 from Jeffrey A. Law <law at gcc dot gnu.org> --- I've done light testing. Essentially it gets us past the ICE. I'm still seeing a comparison failure, but that will be a distinct BZ once I get get my brain around what's going wrong. So ideally Andrew P would submit it upstream, but you could as well. Richard S has effectively ACK'd in this BZ, so it just needs a suitable ChangeLog and test on one of the primary platforms before going onto the trunk.
