https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56997
--- Comment #17 from Bernd Edlinger <bernd.edlinger at hotmail dot de> --- (In reply to Andrew Pinski from comment #16) > This testcase fails on aarch64 when SLOW_UNALIGNED_ACCESS is true. hmm, yes. there are targets that define SLOW_UNALIGNED_ACCESS=1, but they also define STRICT_ALIGNMENT=1 at the same time. probably this combination is not tested at all. does it pass if you use -mstrict-align ?
