[all-commits] [llvm/llvm-project] 3a4ad4: [DAGCombiner] Combine trunc (splat_vector x) -> sp...

[Luke Lau via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 3a4ad45a2c5e700b0245db85b0364f5a62fa4359
      https://github.com/llvm/llvm-project/commit/3a4ad45a2c5e700b0245db85b0364f5a62fa4359
  Author: Luke Lau <luke at igalia.com>
  Date:   2023-08-30 (Wed, 30 Aug 2023)

  Changed paths:
    M llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
    M llvm/test/CodeGen/RISCV/rvv/vnsrl-sdnode.ll
    M llvm/test/CodeGen/RISCV/rvv/vwsll-sdnode.ll
    M llvm/test/CodeGen/RISCV/rvv/vwsll-vp.ll

  Log Message:
  -----------
  [DAGCombiner] Combine trunc (splat_vector x) -> splat_vector (trunc x)

>From the discussion in https://reviews.llvm.org/D158853, moving the truncate
into the splat helps more splatted scalar operands get selected on RISC-V, and
also avoids the need for splat_vector_parts on RV32.

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D159147