https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123267
--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> --- (In reply to Sergei Trofimovich from comment #3) > For me bisect landed on r16-6333-gac64ceb33bf05b "ifcvt: cond zero arith: > handle subreg for shift count". That is what I expected to be one of the ifcvt patches. Also I am reducing this. Should be done in an hour or 2.
