> > As we move towards enabling PKRU on the host, due to some customer > requests, I have to wonder if PKRU-disabled is the norm. > > In other words, is this a likely() or unlikely() optimization? I think it should be likely() as PKU was introduced very early in the Skylake-SP server cores many years ago. Today I think all recent client CPUs should have PKU on default if I am not mistaken. So yeah, adding a likely() probably should help prevent the compiler from evicting this code chunk to the end of function. Thanks. -Mingwei
