cccl
cpp MediumNVIDIA/cccl
2,239 stars
367 forks
20 open issues
Active Mar 2026
Beginner-Friendly Issues 20
Issues tagged for new contributors
[FEA]: Memory resources supporting alignments greater than 256-bytes
#8157 · Mar 24, 2026
[FEA]: Add cuda::ptx wrappers for warp-level mma.sync instructions
#8146 · Mar 24, 2026
needs triage
[BUG]: Compile time for thrust::tabulate with large unary op functor
#8122 · Mar 20, 2026
[FEA]: Investigate performance regression with `cuda::std::partition`
#8112 · Mar 19, 2026
libcu++
[FEA]: Investigate performance regression with `cuda::std::transform_exclusive_scan`
#8111 · Mar 19, 2026
[FEA]: Investigate remaining performance issues with PSTL algorithms
#8110 · Mar 19, 2026
libcu++
[spike] Investigate enabling using CUB cooperative algorithms from C++ tile kernels
#8089 · Mar 18, 2026
[spike] Investigate implementing CCCL algorithms as tile kernels
#8087 · Mar 18, 2026
[THEME] CUDA C++ Tile
#8086 · Mar 18, 2026
`thrust::is_partitioned` is slower than it should be
#8085 · Mar 18, 2026
good first issue thrust
[BUG]: Cannot export CMake interface target which links to CCCL::CUB
#8069 · Mar 17, 2026
good first issue cub infra agent
[FEA]: Cast any_resource to underlying type
#7994 · Mar 11, 2026
[FEA]: Add batched binary search algorithm for sorted needles and haystack
#7964 · Mar 10, 2026
In cccl runtime documentation link to API reference section where applicable
#7963 · Mar 9, 2026
Use the new tuning API internally for `detail::segmented_sort::dispatch`
#7962 · Mar 9, 2026
Use the new tuning API internally for `detail::topk::dispatch`
#7961 · Mar 9, 2026