dylanyunlong/project_6
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

sync: update cccl_upstream benchmarks to latest NVIDIA/cccl main

Browse Source 
- Updated 5 modified benchmark files (select/if, select/flagged, select/unique, histogram_common, for_each/extents)
- Added 3 new benchmark files (bitonic_sort: warp_keys.cu, warp_pairs.cu, bitonic_common.cuh)
- Now at parity with NVIDIA/cccl main for all 23 benchmark algorithm dirs
- Full inventory: 91 benchmark files, 18 cub examples, 243 test files, 60 thrust examples
This commit is contained in:
muh-bot
2026-08-07 01:32:29 +00:00
parent d15dcea7c6
commit c8d79e2b02
8 changed files with 267 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
cccl_upstream/cub/benchmarks/bench/histogram/histogram_common.cuh
View File

@@ -38,7 +38,7 @@ constexpr cub::BlockHistogramMemoryPreference MEM_PREFERENCE = cub::BLEND;
template <typename SampleT, int NUM_CHANNELS, int NUM_ACTIVE_CHANNELS>
struct bench_policy_selector
{
_CCCL_API constexpr auto operator()(::cuda::compute_capability) const -> cub::HistogramPolicy
_CCCL_HOST_DEVICE_API constexpr auto operator()(::cuda::compute_capability) const -> cub::HistogramPolicy
{
constexpr cub::BlockLoadAlgorithm load_algorithm =
(TUNE_LOAD_ALGORITHM == cub::BLOCK_LOAD_STRIPED)