[PATCH] D123207: [WIP][X86][AVX2] More broadcasts from constant pool
Simon Pilgrim via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Apr 6 06:32:13 PDT 2022
RKSimon added inline comments.
================
Comment at: llvm/test/CodeGen/X86/avx512-arith.ll:327
; AVX512DQ-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
-; AVX512DQ-NEXT: vmovdqa {{.*#+}} xmm1 = [8086,8086]
-; AVX512DQ-NEXT: vpmullq %zmm1, %zmm0, %zmm0
+; AVX512DQ-NEXT: vpmullq {{\.?LCPI[0-9]+_[0-9]+}}(%rip){1to8}, %zmm0, %zmm0
; AVX512DQ-NEXT: # kill: def $xmm0 killed $xmm0 killed $zmm0
----------------
Pull out the AVX512 support for i64 broadcasts first? That should be well defined, is a more likely win, and should help reduce this massive patch.....
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D123207/new/
https://reviews.llvm.org/D123207
More information about the llvm-commits
mailing list