[all-commits] [llvm/llvm-project] 6e90f1: [mlir][spirv] Drop support for SPV_NV_cooperative_...

[Jakub Kuderski via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 6e90f13cc9bc9dbc5c2c248d95c6e18a5fb021b4
      https://github.com/llvm/llvm-project/commit/6e90f13cc9bc9dbc5c2c248d95c6e18a5fb021b4
  Author: Jakub Kuderski <jakub at nod-labs.com>
  Date:   2024-01-08 (Mon, 08 Jan 2024)

  Changed paths:
    M mlir/include/mlir/Conversion/GPUToSPIRV/GPUToSPIRV.h
    M mlir/include/mlir/Conversion/Passes.td
    M mlir/include/mlir/Dialect/SPIRV/IR/SPIRVBase.td
    M mlir/include/mlir/Dialect/SPIRV/IR/SPIRVCooperativeMatrixOps.td
    M mlir/include/mlir/Dialect/SPIRV/IR/SPIRVTypes.h
    M mlir/lib/Conversion/GPUToSPIRV/GPUToSPIRVPass.cpp
    M mlir/lib/Conversion/GPUToSPIRV/WmmaOpsToSPIRV.cpp
    M mlir/lib/Dialect/SPIRV/IR/CastOps.cpp
    M mlir/lib/Dialect/SPIRV/IR/CooperativeMatrixOps.cpp
    M mlir/lib/Dialect/SPIRV/IR/SPIRVDialect.cpp
    M mlir/lib/Dialect/SPIRV/IR/SPIRVOps.cpp
    M mlir/lib/Dialect/SPIRV/IR/SPIRVTypes.cpp
    M mlir/lib/Target/SPIRV/Deserialization/DeserializeOps.cpp
    M mlir/lib/Target/SPIRV/Deserialization/Deserializer.cpp
    M mlir/lib/Target/SPIRV/Serialization/Serializer.cpp
    M mlir/test/Conversion/GPUToSPIRV/wmma-ops-to-spirv-khr-coop-matrix.mlir
    R mlir/test/Conversion/GPUToSPIRV/wmma-ops-to-spirv-nv-coop-matrix.mlir
    M mlir/test/Dialect/SPIRV/IR/cast-ops.mlir
    M mlir/test/Dialect/SPIRV/IR/composite-ops.mlir
    M mlir/test/Dialect/SPIRV/IR/khr-cooperative-matrix-ops.mlir
    M mlir/test/Dialect/SPIRV/IR/matrix-ops.mlir
    R mlir/test/Dialect/SPIRV/IR/nv-cooperative-matrix-ops.mlir
    M mlir/test/Dialect/SPIRV/IR/structure-ops.mlir
    M mlir/test/Dialect/SPIRV/IR/types.mlir
    M mlir/test/Target/SPIRV/matrix.mlir
    R mlir/test/Target/SPIRV/nv-cooperative-matrix-ops.mlir

  Log Message:
  -----------
  [mlir][spirv] Drop support for SPV_NV_cooperative_matrix (#76782)

This extension has been superseded by SPV_KHR_cooperative_matrix which
is supported across major vendors GPU like Nvidia, AMD, and Intel.

Given that the KHR version has been supported for nearly half a year,
drop the NV-specific extension to reduce the maintenance burden and code
duplication.