> The patch punishes reload of alternative pair of "d, Z" for > movsi_internal1. The reload occurs if 'Z' doesn't match and generates an > additional insn. So the memory reload should be punished. > > Bootstrapped and tested on powerpc64le-linux with no regressions. Is this > okay for trunk? Any recommendations? Thanks a lot. > > > ChangeLog > > 2021-09-29 Haochen Gui <guih...@linux.ibm.com> > > gcc/ > * gcc/config/rs6000/rs6000.md (movsi_internal1): disparages > slightly the alternative 'Z' of "lfiwzx" when reload is needed.
Capitalize "D" of disparages. Should this disparage "stfiwzx" also? Carl Love saw poor code generation at -O0 for a trivial example where a stack store moved the value to an FPR and used stfiwx. Thanks, David
