[Bug target/77881] [5/6/7 Regression] Non-optimal signed comparison on x86_64 since r146817

jakub at gcc dot gnu.org gcc-bugzilla@gcc.gnu.org
Thu Oct 6 12:32:00 GMT 2016


https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77881

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Well, generally if we want to store a < 0 or a >= 0 into a bool/int variable
etc., then the right shift is desirable, because we avoid branching or
sets/setns, but not when the result is used in a conditional jump.


More information about the Gcc-bugs mailing list