https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77468
PeteVine <tulipawn at gmail dot com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|DUPLICATE |--- --- Comment #22 from PeteVine <tulipawn at gmail dot com> --- @ktkachov I've just retested with your patches from PR68664 and it looks like this could be a different issue, concerning Cortex-A53 codegen exclusively. Using -mcpu=thunderx demonstrates the problem (as well as gcc7 superiority probably): http://openbenchmarking.org/result/1702146-RI-1609039HA18