[PATCH] D39060: AMDGPU: Lower buffer store and atomic intrinsics manually

[Marek Olšák via Phabricator via llvm-commits]

mareko created this revision.
Herald added subscribers: t-tye, tpr, dstuttard, yaxunl, wdng, kzhuravl.

Without this, SIMemoryLegalizer inserts s_waitcnt vmcnt(0) before every
buffer store and atomic instruction.


https://reviews.llvm.org/D39060

Files:
  lib/Target/AMDGPU/AMDGPUISelLowering.cpp
  lib/Target/AMDGPU/AMDGPUISelLowering.h
  lib/Target/AMDGPU/BUFInstructions.td
  lib/Target/AMDGPU/SIISelLowering.cpp
  lib/Target/AMDGPU/SIInstrInfo.td
  lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
  test/CodeGen/AMDGPU/llvm.amdgcn.buffer.atomic.ll
  test/CodeGen/AMDGPU/llvm.amdgcn.buffer.store.format.ll
  test/CodeGen/AMDGPU/llvm.amdgcn.buffer.store.ll
  test/CodeGen/AMDGPU/llvm.amdgcn.image.atomic.ll
  test/CodeGen/AMDGPU/llvm.amdgcn.image.ll
  test/CodeGen/AMDGPU/llvm.amdgcn.s.waitcnt.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D39060.119497.patch
Type: text/x-patch
Size: 30910 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171018/73e3afc2/attachment.bin>