[all-commits] [llvm/llvm-project] cf964e: [VE] v512i1 mask arithmetic isel

[Simon Moll via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: cf964eb5bd666c870cc21963fc5bf017699c29d7
      https://github.com/llvm/llvm-project/commit/cf964eb5bd666c870cc21963fc5bf017699c29d7
  Author: Simon Moll <simon.moll at emea.nec.com>
  Date:   2022-02-21 (Mon, 21 Feb 2022)

  Changed paths:
    M llvm/lib/Target/VE/VECustomDAG.cpp
    M llvm/lib/Target/VE/VECustomDAG.h
    M llvm/lib/Target/VE/VEISelLowering.cpp
    M llvm/lib/Target/VE/VEISelLowering.h
    M llvm/lib/Target/VE/VEInstrInfo.td
    M llvm/lib/Target/VE/VEInstrPatternsVec.td
    A llvm/test/CodeGen/VE/Packed/mask_binary.ll

  Log Message:
  -----------
  [VE] v512i1 mask arithmetic isel

Packed vector and mask registers (v512) are composed of two v256
subregisters that occupy the even and odd element positions.  We add
packing support SDNodes (vec_unpack_lo|hi and vec_pack) and splitting of
v512i1 mask arithmetic ops with those.

Reviewed By: kaz7

Differential Revision: https://reviews.llvm.org/D120053