Skip to content
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.

Added launch bounds to the reduce kernels #16397

Merged
merged 4 commits into from
Oct 31, 2019
Merged

Added launch bounds to the reduce kernels #16397

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
991302d
Added launch bounds to the reduce_kernel_M1
ptrendx Oct 8, 2019
72b6aeb
Trigger CI
ptrendx Oct 8, 2019
e6e4ef4
Reretrigger the CI
ptrendx Oct 9, 2019
0c1931e
Merge branch 'upstream' into pr_issue_16338
ptrendx Oct 19, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions src/operator/tensor/broadcast_reduce-inl.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -262,6 +262,7 @@ __global__ void reduce_lines_kernel(const int N, const int M, const bool addto,
}

template<typename Reducer, int ndim, typename AType, typename DType, typename OType, typename OP>
__launch_bounds__(kMaxThreadsPerBlock)
__global__ void reduce_kernel_M1(const int N, const bool addto,
const DType* __restrict big, OType *small, const Shape<ndim> bshape,
const Shape<ndim> sshape) {
Expand All @@ -277,6 +278,7 @@ __global__ void reduce_kernel_M1(const int N, const bool addto,
}

template<typename Reducer, int ndim, typename DType, typename OP1, typename OP2>
__launch_bounds__(kMaxThreadsPerBlock)
__global__ void reduce_kernel_M1(const int N, const bool addto,
const DType* __restrict big,
const DType* __restrict lhs,
Expand Down