[Bug target/103750] [i386] GCC schedules KMOV instructions that destroys performance in loop

[cvs-commit at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103750

--- Comment #16 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by hongtao Liu <liuho...@gcc.gnu.org>:

https://gcc.gnu.org/g:1a7ce8570997eb1596c803443d20687b43fa2e47

commit r12-6103-g1a7ce8570997eb1596c803443d20687b43fa2e47
Author: liuhongt <hongtao....@intel.com>
Date:   Wed Dec 22 16:48:54 2021 +0800

    Combine vpcmpuw + zero_extend to vpcmpuw.

    vcmp{ps,ph,pd} and vpcmp{,u}{b,w,d,q} implicitly clear the upper bits
    of dest.

    gcc/ChangeLog:

            PR target/103750
            * config/i386/sse.md
            (*<avx512>_cmp<V48H_AVX512VL:mode>3_zero_extend<SWI248x:mode>):
            New pre_reload define_insn_and_split.
            (*<avx512>_cmp<VI12_AVX512VL:mode>3_zero_extend<SWI248x:mode>):
            Ditto.
            (*<avx512>_ucmp<VI12_AVX512VL:mode>3_zero_extend<SWI248x:mode>):
            Ditto.
            (*<avx512>_ucmp<VI48_AVX512VL:mode>3_zero_extend<SWI248x:mode>):
            Ditto.
            (*<avx512>_cmp<V48H_AVX512VL:mode>3_zero_extend<SWI248x:mode>_2):
            Ditto.
            (*<avx512>_cmp<VI12_AVX512VL:mode>3_zero_extend<SWI248x:mode>_2):
            Ditto.
            (*<avx512>_ucmp<VI12_AVX512VL:mode>3_zero_extend<SWI248x:mode>_2):
            Ditto.
            (*<avx512>_ucmp<VI48_AVX512VL:mode>3_zero_extend<SWI248x:mode>_2):
            Ditto.

    gcc/testsuite/ChangeLog:

            * gcc.target/i386/avx512bw-pr103750-1.c: New test.
            * gcc.target/i386/avx512bw-pr103750-2.c: New test.
            * gcc.target/i386/avx512f-pr103750-1.c: New test.
            * gcc.target/i386/avx512f-pr103750-2.c: New test.
            * gcc.target/i386/avx512fp16-pr103750-1.c: New test.
            * gcc.target/i386/avx512fp16-pr103750-2.c: New test.