https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116058
--- Comment #7 from Jeffrey A. Law <law at gcc dot gnu.org> --- You might be barking up the wrong tree here. My gut tells me this isn't the core problem and a bootstrap with your patch just failed in the exact same place as before. My suspicion is your patch works around the problem in the reduced testcase, but that the real bug is elsewhere.
