[all-commits] [llvm/llvm-project] 74bf0b: [mlir] Lower math dialect later in gpu-lower-to-nv...

[Guray Ozen via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 74bf0b1cd9dde53b2df08966bcf2f91d4909ccd6
      https://github.com/llvm/llvm-project/commit/74bf0b1cd9dde53b2df08966bcf2f91d4909ccd6
  Author: Guray Ozen <guray.ozen at gmail.com>
  Date:   2024-01-31 (Wed, 31 Jan 2024)

  Changed paths:
    M mlir/lib/Dialect/GPU/Pipelines/GPUToNVVMPipeline.cpp
    A mlir/test/Dialect/GPU/test-nvvm-pipeline.mlir

  Log Message:
  -----------
  [mlir] Lower math dialect later in gpu-lower-to-nvvm-pipeline (#78556)

This PR moves lowering of math dialect later in the pipeline. Because
math dialect is lowered correctly by `createConvertGpuOpsToNVVMOps` for
GPU target, and it needs to run it first.