[all-commits] [llvm/llvm-project] b43701: [OpenMPIRBuilder] Do not call host runtime for GPU...

[Dominik Adamski via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: b4370140b4467eddc42e7b8075959c692daaf3e8
      https://github.com/llvm/llvm-project/commit/b4370140b4467eddc42e7b8075959c692daaf3e8
  Author: Dominik Adamski <dominik.adamski at amd.com>
  Date:   2024-01-31 (Wed, 31 Jan 2024)

  Changed paths:
    M llvm/lib/Frontend/OpenMP/OMPIRBuilder.cpp
    A mlir/test/Target/LLVMIR/omptarget-teams-llvm.mlir

  Log Message:
  -----------
  [OpenMPIRBuilder] Do not call host runtime for GPU teams codegen (#79984)

Patch ensures that host runtime functions are not called for handling
OpenMP teams clause on the device.

GPU code for pragma `omp target teams distribute parallel do` will
require only one call to OpenMP loop-worksharing GPU runtime. Support
for it will be added later.

This patch does not include changes required for handling `omp target
teams` for the host side.