[PATCH] D83723: [OpenMP] Generalize CGOpenMPRuntimeNVPTX as CGOpenMPRuntimeGPU

[Saiyedul Islam via Phabricator via cfe-commits]

saiislam created this revision.
saiislam added reviewers: ABataev, jdoerfert, JonChesterfield, grokos.
Herald added subscribers: cfe-commits, aaron.ballman, sstefan1, jfb, guansong, yaxunl, mgorny, jvesely, jholewinski.
Herald added a project: clang.

Refactors CGOpenMPRuntimeNVPTX as CGOpenMPRuntimeGPU to make it a
generalization for OpenMP GPU Codegen. Target specific specialized
methods for NVPTX and AMDGCN are defined in their corresponding
classes CGOpenMPRuntimeNVPTX and CGOpenMPRuntimeAMDGCN, respectively.

This paves the way for a clean and easy to maintain AMDGCN implementation
for OpenMP GPU Codegen. Imeplementaion of getGPUWarpSize() method (earlier
getNVPTXWarpSize()) has been provided as a proof of concept.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D83723

Files:
  clang/lib/CodeGen/CGOpenMPRuntimeAMDGCN.cpp
  clang/lib/CodeGen/CGOpenMPRuntimeAMDGCN.h
  clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
  clang/lib/CodeGen/CGOpenMPRuntimeGPU.h
  clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
  clang/lib/CodeGen/CGOpenMPRuntimeNVPTX.h
  clang/lib/CodeGen/CMakeLists.txt
  clang/lib/CodeGen/CodeGenModule.cpp

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D83723.277576.patch
Type: text/x-patch
Size: 299212 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20200713/31d8a5c1/attachment-0001.bin>