https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93738
--- Comment #20 from Jeffrey A. Law <law at gcc dot gnu.org> --- I don't think so. We generally are concerned with incorrect code, ICEs and the like. A small missed optimization, even though it's a regression, doesn't seem to have the right risk/reward. So if your patch fixes this bug for power, I'd recommend closing as resolved without a backport.
