[libomptarget][nfc] Move three more files to common

[JonChesterfield]

Summary

[libomptarget][nfc] Move three more files to common

[JonChesterfield]

After this patch common/ actually contains more SLOC than nvptx/. Parts of libcall.cu and data_sharing.cu are architecture agnostic, reduction.cu probably needs a bespoke implementation for amdgcn.

[ABataev]

openmp/libomptarget/deviceRTLs/common/src/parallel.cu

1

GPU OpenMP parallel implementation?

[JonChesterfield]

• update comment

[JonChesterfield]

• update comment

[ABataev]

openmp/libomptarget/deviceRTLs/common/src/support.cu

1

Same here, GPU

openmp/libomptarget/deviceRTLs/common/src/sync.cu

1

GPU

[JonChesterfield]

Update comment, then fix the fallout from arc diff master --update using an out of date master. Apologies to everyone who was transiently subscribed to this.

[JonChesterfield]

• Update two more comments as requested

[ABataev]

LG

openmp/libomptarget/deviceRTLs/common/src/parallel.cu

-//===---- parallel.cu - NVPTX OpenMP parallel implementation ----- CUDA -*-===//
+//===---- parallel.cu - GPU OpenMP parallel implementation ------- CUDA -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 //
 // Parallel implemention in the GPU. Here is the pattern:

openmp/libomptarget/deviceRTLs/common/src/support.cu

-//===--------- support.cu - NVPTX OpenMP support functions ------- CUDA -*-===//
+//===--------- support.cu - GPU OpenMP support functions --------- CUDA -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 //
 // Wrapper implementation to some functions natively supported by the GPU.

openmp/libomptarget/deviceRTLs/common/src/sync.cu

-//===------------ sync.h - NVPTX OpenMP synchronizations --------- CUDA -*-===//
+//===------------ sync.cu - GPU OpenMP synchronizations ---------- CUDA -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 //
 // Include all synchronization.

openmp/libomptarget/deviceRTLs/nvptx/CMakeLists.txt

if(LIBOMPTARGET_DEP_CUDA_FOUND)
 
   set(cuda_src_files
       ${devicertl_common_directory}/src/cancel.cu
       ${devicertl_common_directory}/src/critical.cu
       src/data_sharing.cu
       src/libcall.cu
       ${devicertl_common_directory}/src/loop.cu
       ${devicertl_common_directory}/src/omptarget.cu
-      src/parallel.cu
+      ${devicertl_common_directory}/src/parallel.cu
       src/reduction.cu
-      src/support.cu
-      src/sync.cu
+      ${devicertl_common_directory}/src/support.cu
+      ${devicertl_common_directory}/src/sync.cu
       ${devicertl_common_directory}/src/task.cu
   )
 
   set(omp_data_objects src/omp_data.cu)
 
   # Get the compute capability the user requested or use SM_35 by default.
   # SM_35 is what clang uses by default.
   set(default_capabilities 35)