[all-commits] [llvm/llvm-project] abe71b: [OpenMP][NFC] Delete dead code

[Johannes Doerfert via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: abe71b77f9ac2fc689e899fa6aa2486120c53912
      https://github.com/llvm/llvm-project/commit/abe71b77f9ac2fc689e899fa6aa2486120c53912
  Author: Johannes Doerfert <johannes at jdoerfert.de>
  Date:   2023-11-06 (Mon, 06 Nov 2023)

  Changed paths:
    M clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp

  Log Message:
  -----------
  [OpenMP][NFC] Delete dead code


  Commit: 921bd299134fbe17c676b2486af269e18281def4
      https://github.com/llvm/llvm-project/commit/921bd299134fbe17c676b2486af269e18281def4
  Author: Johannes Doerfert <johannes at jdoerfert.de>
  Date:   2023-11-06 (Mon, 06 Nov 2023)

  Changed paths:
    M clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
    M clang/test/OpenMP/nvptx_teams_reduction_codegen.cpp
    M clang/test/OpenMP/reduction_implicit_map.cpp
    M clang/test/OpenMP/target_teams_generic_loop_codegen.cpp

  Log Message:
  -----------
  [OpenMP] Remove alignment for global <-> local reduction functions

The alignment did likely not help much but increases the memory
requirement. Note that half of the affected accesses are all performed
by a single thread in each block. The reads are by consecutive threads
in a single block.


  Commit: 3de645efe30b83ba1b6d7e500486c4f441a17a61
      https://github.com/llvm/llvm-project/commit/3de645efe30b83ba1b6d7e500486c4f441a17a61
  Author: Johannes Doerfert <johannes at jdoerfert.de>
  Date:   2023-11-06 (Mon, 06 Nov 2023)

  Changed paths:
    M clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
    M clang/test/OpenMP/nvptx_target_parallel_reduction_codegen.cpp
    M clang/test/OpenMP/nvptx_target_parallel_reduction_codegen_tbaa_PR46146.cpp
    M clang/test/OpenMP/nvptx_teams_reduction_codegen.cpp
    M clang/test/OpenMP/reduction_implicit_map.cpp
    M clang/test/OpenMP/target_teams_generic_loop_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 llvm/test/Transforms/OpenMP/always_inline_device.ll
    M llvm/test/Transforms/OpenMP/custom_state_machines.ll
    M llvm/test/Transforms/OpenMP/custom_state_machines_pre_lto.ll
    M llvm/test/Transforms/OpenMP/custom_state_machines_remarks.ll
    M llvm/test/Transforms/OpenMP/deduplication_target.ll
    M llvm/test/Transforms/OpenMP/get_hardware_num_threads_in_block_fold.ll
    M llvm/test/Transforms/OpenMP/global_constructor.ll
    M llvm/test/Transforms/OpenMP/globalization_remarks.ll
    M llvm/test/Transforms/OpenMP/gpu_state_machine_function_ptr_replacement.ll
    M llvm/test/Transforms/OpenMP/is_spmd_exec_mode_fold.ll
    M llvm/test/Transforms/OpenMP/nested_parallelism.ll
    M llvm/test/Transforms/OpenMP/parallel_level_fold.ll
    M llvm/test/Transforms/OpenMP/remove_globalization.ll
    M llvm/test/Transforms/OpenMP/replace_globalization.ll
    M llvm/test/Transforms/OpenMP/single_threaded_execution.ll
    M llvm/test/Transforms/OpenMP/spmdization.ll
    M llvm/test/Transforms/OpenMP/spmdization_assumes.ll
    M llvm/test/Transforms/OpenMP/spmdization_constant_prop.ll
    M llvm/test/Transforms/OpenMP/spmdization_guarding.ll
    M llvm/test/Transforms/OpenMP/spmdization_guarding_two_reaching_kernels.ll
    M llvm/test/Transforms/OpenMP/spmdization_indirect.ll
    M llvm/test/Transforms/OpenMP/spmdization_kernel_env_dep.ll
    M llvm/test/Transforms/OpenMP/spmdization_no_guarding_two_reaching_kernels.ll
    M llvm/test/Transforms/OpenMP/spmdization_remarks.ll
    M llvm/test/Transforms/OpenMP/value-simplify-openmp-opt.ll
    M llvm/test/Transforms/PhaseOrdering/openmp-opt-module.ll
    M mlir/test/Target/LLVMIR/omptarget-region-device-llvm.mlir
    M openmp/libomptarget/DeviceRTL/include/Interface.h
    M openmp/libomptarget/DeviceRTL/src/Reduction.cpp
    M openmp/libomptarget/include/Environment.h
    M openmp/libomptarget/plugins-nextgen/common/PluginInterface/PluginInterface.cpp

  Log Message:
  -----------
  [OpenMP][NFC] Split the reduction buffer size into two components

Before we tracked the size of the teams reduction buffer in order to
allocate it at runtime per kernel launch. This patch splits the number
into two parts, the size of the reduction data (=all reduction
variables) and the (maximal) length of the buffer. This will allow us to
allocate less if we need less, e.g., if we have less teams than the
maximal length. It also allows us to move code from clangs codegen into
the runtime as we now know how large the reduction data is.


Compare: https://github.com/llvm/llvm-project/compare/e28c393bd1c1...3de645efe30b