https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100903
--- Comment #19 from Jason Merrill <jason at gcc dot gnu.org> --- (In reply to Philipp Fent from comment #18) > (In reply to Jonathan Wakely from comment #17) > > Do you mean comparing durations to each other, or when comparing the result > > of such a comparison to 0? The former would be surprising > > I also see this without any comparison to 0, but only in template functions. For information, this regression is due to Patrick's r16-2519-gba5a6787374dea (and followups) rewriting the < into <=> < 0 in a template so it looks like that's what the user wrote. I'm fixing it so it won't warn either way.
