[all-commits] [llvm/llvm-project] aaddca: [mlir][GPUToSPIRV] Modify the lowering of gpu.bloc...

[MaheshRavishankar via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: aaddca1efd1d5a66d014023b9649cb273a84a7ae
      https://github.com/llvm/llvm-project/commit/aaddca1efd1d5a66d014023b9649cb273a84a7ae
  Author: MaheshRavishankar <ravishankarm at google.com>
  Date:   2020-02-08 (Sat, 08 Feb 2020)

  Changed paths:
    M mlir/lib/Conversion/GPUToSPIRV/ConvertGPUToSPIRV.cpp
    M mlir/test/Conversion/GPUToSPIRV/builtins.mlir
    M mlir/test/Conversion/GPUToSPIRV/load-store.mlir

  Log Message:
  -----------
  [mlir][GPUToSPIRV] Modify the lowering of gpu.block_dim to be consistent with Vulkan SPEC

The existing lowering of gpu.block_dim added a global variable with
the WorkGroupSize decoration. This raises an error within
Vulkan/SPIR-V validation since Vulkan requires this to have a constant
initializer. This is not yet supported in SPIR-V dialect. Changing the
lowering to return the workgroup size as a constant value instead,
obtained from spv.entry_point_abi attribute gets around the issue for
now. The validation goes through since the workgroup size is specified
using spv.execution_mode operation.