[all-commits] [llvm/llvm-project] 21d91a: [libomptarget][devicertl] Replace lanemask with ui...

Jon Chesterfield via All-commits all-commits at lists.llvm.org
Wed Aug 18 12:47:54 PDT 2021


  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 21d91a8ef319eec9c2c272e19beee726429524aa
      https://github.com/llvm/llvm-project/commit/21d91a8ef319eec9c2c272e19beee726429524aa
  Author: Jon Chesterfield <jonathanchesterfield at gmail.com>
  Date:   2021-08-18 (Wed, 18 Aug 2021)

  Changed paths:
    M clang/test/OpenMP/nvptx_parallel_codegen.cpp
    M llvm/include/llvm/Frontend/OpenMP/OMPIRBuilder.h
    M llvm/include/llvm/Frontend/OpenMP/OMPKinds.def
    M llvm/lib/Frontend/OpenMP/OMPIRBuilder.cpp
    M llvm/test/Transforms/OpenMP/add_attributes.ll
    M openmp/libomptarget/DeviceRTL/include/Interface.h
    M openmp/libomptarget/DeviceRTL/src/Synchronization.cpp
    M openmp/libomptarget/deviceRTLs/common/src/sync.cu
    M openmp/libomptarget/deviceRTLs/interface.h

  Log Message:
  -----------
  [libomptarget][devicertl] Replace lanemask with uint64 at interface

Use uint64_t for lanemask on all GPU architectures at the interface
with clang. Updates tests. The deviceRTL is always linked as IR so the zext
and trunc introduced for wave32 architectures will fold after inlining.

Simplification partly motivated by amdgpu gfx10 which will be wave32 and
is awkward to express in the current arch-dependant typedef interface.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D108317




More information about the All-commits mailing list