https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101891
--- Comment #8 from Qing Zhao <qing.zhao at oracle dot com> --- > On May 24, 2022, at 11:42 AM, arjan at linux dot intel.com > <gcc-bugzi...@gcc.gnu.org> wrote: > > --- Comment #7 from Arjan van de Ven <arjan at linux dot intel.com> --- > from a performance angle, the xor-only sequence is not so great at all; modern > CPUs are really good at eliminating mov's so that's why the code originally > was > added to do a combo of xor and mov.. Are you saying that the Xor-only sequence is slower than the previous XOR + MOV sequence? If so, can you explain a little bit more on why? And do you have any data for this claim? Thanks.