[PATCH] D24623: AMDGPU: Implement memory model

[Konstantin Zhuravlyov via Phabricator via llvm-commits]

kzhuravl updated this revision to Diff 107323.
kzhuravl added a comment.

Fix atomic fence expansion for workgroup/wavefront/singlethread scopes.


https://reviews.llvm.org/D24623

Files:
  lib/Target/AMDGPU/AMDGPU.h
  lib/Target/AMDGPU/AMDGPUMachineModuleInfo.cpp
  lib/Target/AMDGPU/AMDGPUMachineModuleInfo.h
  lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
  lib/Target/AMDGPU/BUFInstructions.td
  lib/Target/AMDGPU/CMakeLists.txt
  lib/Target/AMDGPU/FLATInstructions.td
  lib/Target/AMDGPU/SIDefines.h
  lib/Target/AMDGPU/SIInstrFormats.td
  lib/Target/AMDGPU/SIInstructions.td
  lib/Target/AMDGPU/SIMemoryLegalizer.cpp
  test/CodeGen/AMDGPU/fence-amdgiz.ll
  test/CodeGen/AMDGPU/flat_atomics.ll
  test/CodeGen/AMDGPU/flat_atomics_i64.ll
  test/CodeGen/AMDGPU/global_atomics.ll
  test/CodeGen/AMDGPU/global_atomics_i64.ll
  test/CodeGen/AMDGPU/memory-legalizer-atomic-cmpxchg.ll
  test/CodeGen/AMDGPU/memory-legalizer-atomic-fence.ll
  test/CodeGen/AMDGPU/memory-legalizer-atomic-load.ll
  test/CodeGen/AMDGPU/memory-legalizer-atomic-rmw.ll
  test/CodeGen/AMDGPU/memory-legalizer-atomic-store.ll
  test/CodeGen/AMDGPU/memory-legalizer-invalid-syncscope.ll
  test/CodeGen/AMDGPU/syncscopes.ll
  test/CodeGen/MIR/AMDGPU/memory-legalizer-atomic-insert-end.mir

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D24623.107323.patch
Type: text/x-patch
Size: 109811 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170719/0f193c18/attachment-0001.bin>