[all-commits] [llvm/llvm-project] ffbffa: [PowerPC] Improve codegen for int-to-fp conversion...

[Conanap via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: ffbffaf6b6b0fc06abb7b43ec8de8bc61d941bc7
      https://github.com/llvm/llvm-project/commit/ffbffaf6b6b0fc06abb7b43ec8de8bc61d941bc7
  Author: Albion Fung <albion.fung at ibm.com>
  Date:   2021-05-11 (Tue, 11 May 2021)

  Changed paths:
    M llvm/lib/Target/PowerPC/PPCInstrVSX.td
    M llvm/test/CodeGen/PowerPC/uint-to-fp-v4i32.ll
    A llvm/test/CodeGen/PowerPC/vec-extract-itofp.ll
    M llvm/test/CodeGen/PowerPC/vec_conv_i16_to_fp32_elts.ll
    M llvm/test/CodeGen/PowerPC/vec_conv_i8_to_fp32_elts.ll

  Log Message:
  -----------
  [PowerPC] Improve codegen for int-to-fp conversion of subword vector extract

When an integer is converted into floating point in subword vector extract,
it can be done in 2 instructions instead of the 3+ instructions it generates
right now. This patch removes the uncessary generation.

Differential: https://reviews.llvm.org/D100604