https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69611
--- Comment #5 from joseph at codesourcery dot com <joseph at codesourcery dot com> --- I think it's OK for stage 4 - the t-hardfp point is that you'd get a smaller, faster libgcc on FreeBSD that way, by not compiling soft-fp at all for non-float128 hard float.
