diff options
| author | Alexey Bataev <a.bataev@hotmail.com> | 2019-02-20 16:36:22 +0000 |
|---|---|---|
| committer | Alexey Bataev <a.bataev@hotmail.com> | 2019-02-20 16:36:22 +0000 |
| commit | 8061acd501f1cb6c00a886f4ee5cb9adc6cda39a (patch) | |
| tree | a4c87ec4d6dfe10c577d2c352be2dc1a41ce6bb5 /clang/lib/Frontend/CompilerInvocation.cpp | |
| parent | c4d07554e44867191492a102ea3639276ba1ece1 (diff) | |
| download | bcm5719-llvm-8061acd501f1cb6c00a886f4ee5cb9adc6cda39a.tar.gz bcm5719-llvm-8061acd501f1cb6c00a886f4ee5cb9adc6cda39a.zip | |
[OPENMP][NVPTX]Use faster teams reduction algorithm.
A faster way to reduce the values in teams reductions was found, the
codegen is updated to use this faster algorithm and new runtime functions.
llvm-svn: 354479
Diffstat (limited to 'clang/lib/Frontend/CompilerInvocation.cpp')
| -rw-r--r-- | clang/lib/Frontend/CompilerInvocation.cpp | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp index d4094ac5aea..1a33a00004d 100644 --- a/clang/lib/Frontend/CompilerInvocation.cpp +++ b/clang/lib/Frontend/CompilerInvocation.cpp @@ -2840,6 +2840,9 @@ static void ParseLangArgs(LangOptions &Opts, ArgList &Args, InputKind IK, Opts.OpenMPCUDABlocksPerSM = getLastArgIntValue(Args, options::OPT_fopenmp_cuda_blocks_per_sm_EQ, Opts.OpenMPCUDABlocksPerSM, Diags); + Opts.OpenMPCUDAReductionBufNum = getLastArgIntValue( + Args, options::OPT_fopenmp_cuda_teams_reduction_recs_num_EQ, + Opts.OpenMPCUDAReductionBufNum, Diags); } // Prevent auto-widening the representation of loop counters during an |
