Skip to content

Support CUDA 13 scan comparators #310

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
jeongseok-meta wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Support CUDA 13 scan comparators #310

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
9 changes: 5 additions & 4 deletions nerfacc/cuda/csrc/scan_cub.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@
* Copyright (c) 2022 Ruilong Li, UC Berkeley.
*/

#include <thrust/functional.h>
#include <thrust/iterator/reverse_iterator.h>
#include "include/utils_cuda.cuh"
#include "include/utils.cub.cuh"
Expand All @@ -22,7 +23,7 @@ inline void exclusive_sum_by_key(
TORCH_CHECK(num_items <= std::numeric_limits<long>::max(),
"cub ExclusiveSumByKey does not support more than LONG_MAX elements");
CUB_WRAPPER(cub::DeviceScan::ExclusiveSumByKey, keys, input, output,
num_items, cub::Equality(), at::cuda::getCurrentCUDAStream());
num_items, thrust::equal_to<int64_t>(), at::cuda::getCurrentCUDAStream());
}

template <typename KeysInputIteratorT, typename ValuesInputIteratorT, typename ValuesOutputIteratorT>
Expand All @@ -32,7 +33,7 @@ inline void inclusive_sum_by_key(
TORCH_CHECK(num_items <= std::numeric_limits<long>::max(),
"cub InclusiveSumByKey does not support more than LONG_MAX elements");
CUB_WRAPPER(cub::DeviceScan::InclusiveSumByKey, keys, input, output,
num_items, cub::Equality(), at::cuda::getCurrentCUDAStream());
num_items, thrust::equal_to<int64_t>(), at::cuda::getCurrentCUDAStream());
}

template <typename KeysInputIteratorT, typename ValuesInputIteratorT, typename ValuesOutputIteratorT>
Expand All @@ -42,7 +43,7 @@ inline void exclusive_prod_by_key(
TORCH_CHECK(num_items <= std::numeric_limits<long>::max(),
"cub ExclusiveScanByKey does not support more than LONG_MAX elements");
CUB_WRAPPER(cub::DeviceScan::ExclusiveScanByKey, keys, input, output, Product(), 1.0f,
num_items, cub::Equality(), at::cuda::getCurrentCUDAStream());
num_items, thrust::equal_to<int64_t>(), at::cuda::getCurrentCUDAStream());
}

template <typename KeysInputIteratorT, typename ValuesInputIteratorT, typename ValuesOutputIteratorT>
Expand All @@ -52,7 +53,7 @@ inline void inclusive_prod_by_key(
TORCH_CHECK(num_items <= std::numeric_limits<long>::max(),
"cub InclusiveScanByKey does not support more than LONG_MAX elements");
CUB_WRAPPER(cub::DeviceScan::InclusiveScanByKey, keys, input, output, Product(),
num_items, cub::Equality(), at::cuda::getCurrentCUDAStream());
num_items, thrust::equal_to<int64_t>(), at::cuda::getCurrentCUDAStream());
}
#endif

Expand Down