https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82016
Michael Meissner <meissner at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #7 from Michael Meissner <meissner at gcc dot gnu.org> --- Yes, change 251814 should have fixed the issue. Change 251814 was where I changed the underlying code to enable the _Float128 support by default on 64-bit Linux PowerPC systems if they were configured with an assembler that understands VSX instructions. I'm not sure why the wide int changes seemed to trigger the test.