sm_6x GPUs add atomic operations that explicitly specify set of threads that observe operation as atomic.
This patch implements these instructions along with matching intrinsics.
Details
Details
Diff Detail
Diff Detail
- Repository
- rL LLVM
Paths
| Differential D24943
[NVPTX] Added intrinsics for atom.gen.{sys|cta}.* instructions. ClosedPublic Authored by tra on Sep 26 2016, 4:11 PM.
Details Summary sm_6x GPUs add atomic operations that explicitly specify set of threads that observe operation as atomic.
Diff Detail
Event Timelinetra updated this object. tra added a child revision: D24944: [CUDA] Added __nvvm_atom_{sys|cta}_* builtins for sm_60 GPUs..Sep 26 2016, 4:45 PM jlebar edited edge metadata. Comment ActionsYay, this is great.
This revision is now accepted and ready to land.Sep 27 2016, 2:07 PM Closed by commit rL282607: [NVPTX] Added intrinsics for atom.gen.{sys|cta}.* instructions. (authored by tra). · Explain WhySep 28 2016, 10:34 AM This revision was automatically updated to reflect the committed changes.
Revision Contents
Diff 72857 llvm/trunk/include/llvm/IR/IntrinsicsNVVM.td
llvm/trunk/lib/Target/NVPTX/NVPTX.td
llvm/trunk/lib/Target/NVPTX/NVPTXISelLowering.cpp
llvm/trunk/lib/Target/NVPTX/NVPTXInstrInfo.td
llvm/trunk/lib/Target/NVPTX/NVPTXIntrinsics.td
llvm/trunk/lib/Target/NVPTX/NVPTXSubtarget.h
llvm/trunk/lib/Target/NVPTX/NVPTXSubtarget.cpp
llvm/trunk/lib/Target/NVPTX/NVPTXTargetTransformInfo.cpp
llvm/trunk/test/CodeGen/NVPTX/atomics-with-scope.ll
|