OSDN Git Service

NVPTX: Replace uses of cuda.syncthreads with nvvm.barrier0
authorJustin Bogner <mail@justinbogner.com>
Wed, 6 Jul 2016 20:02:45 +0000 (20:02 +0000)
committerJustin Bogner <mail@justinbogner.com>
Wed, 6 Jul 2016 20:02:45 +0000 (20:02 +0000)
commit45d8d9a866ac6f971981071a82693e9249d696d8
tree676849ffef644fb68e70e11f84ffba5961e83519
parentbf52e2c5a304439d988fcaf190a0e65491e00457
NVPTX: Replace uses of cuda.syncthreads with nvvm.barrier0

Everywhere where cuda.syncthreads or __syncthreads is used, use the
properly namespaced nvvm.barrier0 instead.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274664 91177308-0d34-0410-b5e6-96231b3b80d8
docs/LangRef.rst
docs/NVPTXUsage.rst
include/llvm/IR/IntrinsicsNVVM.td
lib/Target/NVPTX/NVPTXInstrInfo.cpp
lib/Target/NVPTX/NVPTXIntrinsics.td
test/CodeGen/NVPTX/MachineSink-convergent.ll
test/CodeGen/NVPTX/TailDuplication-convergent.ll
test/CodeGen/NVPTX/access-non-generic.ll
test/CodeGen/NVPTX/noduplicate-syncthreads.ll
test/Feature/intrinsic-noduplicate.ll
test/Transforms/FunctionAttrs/convergent.ll