[all-commits] [llvm/llvm-project] 56cd3b: [X86] Directly emit VBROADCAST_LOAD from constant ...

[topperc via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 56cd3bc209e00b419911e88672bb5c6488b4c5ab
      https://github.com/llvm/llvm-project/commit/56cd3bc209e00b419911e88672bb5c6488b4c5ab
  Author: Craig Topper <craig.topper at intel.com>
  Date:   2020-03-03 (Tue, 03 Mar 2020)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp

  Log Message:
  -----------
  [X86] Directly emit VBROADCAST_LOAD from constant pool in lowerBuildVectorAsBroadcast

Also add a DAG combine to combine different sized broadcasts from
constant pool to avoid a regression.

Differential Revision: https://reviews.llvm.org/D75412


  Commit: 3c4e6355936fa88023cc400e1d9251cc350353b9
      https://github.com/llvm/llvm-project/commit/3c4e6355936fa88023cc400e1d9251cc350353b9
  Author: Craig Topper <craig.topper at intel.com>
  Date:   2020-03-03 (Tue, 03 Mar 2020)

  Changed paths:
    M llvm/lib/Target/X86/X86ISelLowering.cpp
    M llvm/test/CodeGen/X86/avg.ll
    M llvm/test/CodeGen/X86/bitcast-int-to-vector-bool-sext.ll
    M llvm/test/CodeGen/X86/bitcast-int-to-vector-bool-zext.ll
    M llvm/test/CodeGen/X86/bitcast-int-to-vector-bool.ll
    M llvm/test/CodeGen/X86/broadcast-elm-cross-splat-vec.ll
    M llvm/test/CodeGen/X86/horizontal-reduce-umax.ll
    M llvm/test/CodeGen/X86/horizontal-reduce-umin.ll
    M llvm/test/CodeGen/X86/i64-to-float.ll
    M llvm/test/CodeGen/X86/pr30284.ll
    M llvm/test/CodeGen/X86/vec_uaddo.ll
    M llvm/test/CodeGen/X86/vec_umulo.ll
    M llvm/test/CodeGen/X86/vec_usubo.ll
    M llvm/test/CodeGen/X86/vector-blend.ll
    M llvm/test/CodeGen/X86/vector-fshl-256.ll
    M llvm/test/CodeGen/X86/vector-fshr-256.ll
    M llvm/test/CodeGen/X86/vector-sext.ll
    M llvm/test/CodeGen/X86/vector-shift-ashr-256.ll
    M llvm/test/CodeGen/X86/vector-shuffle-256-v16.ll
    M llvm/test/CodeGen/X86/vector-shuffle-256-v32.ll

  Log Message:
  -----------
  [X86] Always emit an integer vbroadcast_load from lowerBuildVectorAsBroadcast regardless of AVX vs AVX2

If we go with D75412, we no longer depend on the scalar type directly. So we don't need to avoid using i64. We already have AVX1 fallback patterns with i32 and i64 scalar types so we don't need to avoid using integer types on AVX1.

Differential Revision: https://reviews.llvm.org/D75413


Compare: https://github.com/llvm/llvm-project/compare/22dd23583a61...3c4e6355936f