[all-commits] [llvm/llvm-project] 935bac: [DAG] SimplifyDemandedBits - correctly adjust trun...

[Simon Pilgrim via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 935bacd3a7244f04b7f39818e3fc589529474d13
      https://github.com/llvm/llvm-project/commit/935bacd3a7244f04b7f39818e3fc589529474d13
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2021-01-21 (Thu, 21 Jan 2021)

  Changed paths:
    M llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp

  Log Message:
  -----------
  [DAG] SimplifyDemandedBits - correctly adjust truncated shift amount type

As noticed on D56387, for vectors we must always correctly adjust the shift amount type during truncation (not just after legalization). We were getting away with it as we currently only accepted scalars via the dyn_cast<ConstantSDNode>.


  Commit: 0ca81b90d19d395c4891b7507cec0f063dd26d22
      https://github.com/llvm/llvm-project/commit/0ca81b90d19d395c4891b7507cec0f063dd26d22
  Author: Simon Pilgrim <llvm-dev at redking.me.uk>
  Date:   2021-01-21 (Thu, 21 Jan 2021)

  Changed paths:
    M llvm/test/CodeGen/X86/uint_to_fp-3.ll

  Log Message:
  -----------
  [X86][SSE] Add uitofp(trunc(and(lshr(x,c)))) vector test

Reduced from regression reported by @hans on D56387


Compare: https://github.com/llvm/llvm-project/compare/ed2853d2c82d...0ca81b90d19d