https://gcc.gnu.org/bugzilla/show_bug.cgi?id=119244
--- Comment #2 from Joseph S. Myers <jsm28 at gcc dot gnu.org> --- If ABI support for _Float128 is added on an architecture that doesn't currently have it, we can also add such support on that architecture in glibc - but that will increase the minimum GCC version for building glibc on that architecture (and require a new feature for Versions files in glibc to declare lots of *f128 and *f64x functions as being introduced at the max of their existing version, and the version where an architecture gained _Float128 support).
