On Tue, 5 Jan 2021, Jakub Jelinek wrote:
> Hi!
>
> As requested in the PR, the one's complement abs can be done more
> efficiently without cmov or branching.
>
> Had to change the ifcvt-onecmpl-abs-1.c testcase, we no longer optimize
> it in ifcvt, on x86_64 with -m32 we generate in the end the
Hi!
As requested in the PR, the one's complement abs can be done more
efficiently without cmov or branching.
Had to change the ifcvt-onecmpl-abs-1.c testcase, we no longer optimize
it in ifcvt, on x86_64 with -m32 we generate in the end the exact same
code, but with -m64:
movl%edi,