[X86] Add masked predicate execution variants for instructions in X86FixupInstTuning
authorNoah Goldstein <goldstein.w.n@gmail.com>
Wed, 8 Mar 2023 16:57:46 +0000 (10:57 -0600)
committerNoah Goldstein <goldstein.w.n@gmail.com>
Wed, 8 Mar 2023 17:57:41 +0000 (11:57 -0600)
commit8ac8c579e214e682ec72afa7e50a3198164182d7
treeb082bd7746123419cd793cff58e84f2c60ba4b48
parent6b29a6f27d40cfc4225fe14d0f0a298fde611cad
[X86] Add masked predicate execution variants for instructions in X86FixupInstTuning

Masked variants of UNPCKLPD, UNPCKHPD, and PERMILPS were missing and
be transformed with the same logic as their non-masked counterparts.

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D144763
llvm/lib/Target/X86/X86FixupInstTuning.cpp
llvm/test/CodeGen/X86/avx512-intrinsics-fast-isel.ll
llvm/test/CodeGen/X86/avx512-intrinsics-upgrade.ll
llvm/test/CodeGen/X86/avx512-shuffles/in_lane_permute.ll
llvm/test/CodeGen/X86/avx512vl-intrinsics-upgrade.ll
llvm/test/CodeGen/X86/tuning-shuffle-permilps-avx512.ll
llvm/test/CodeGen/X86/tuning-shuffle-unpckpd-avx512.ll
llvm/test/CodeGen/X86/vector-shuffle-combining-avx512f.ll