Issues

• [cudax][cuco] Add cuco Bloom filter byte-budget construction

  cucocudax
  NVIDIA/cccl#12202 · Filter by author sleeepyjack opened on Oct 22, 2026

• [cudax][cuco] Align cuco Bloom filter WordBytes policy API

  cucocudax
  NVIDIA/cccl#12201 · Filter by author sleeepyjack opened on Oct 22, 2026

• [cudax][cuco] Add cuco Bloom filter persisting-L2 access

  cucocudax
  NVIDIA/cccl#12200 · Filter by author sleeepyjack opened on Oct 22, 2026

• [cudax][cuco] Port cuco Bloom filter union and intersection

  cucocudax
  NVIDIA/cccl#12199 · Filter by author sleeepyjack opened on Oct 22, 2026

• [cuda.compute]: cached algorithm wrappers and operator cache keys keep the caller's device arrays alive, leading to OOM in long sweeps

  NVIDIA/cccl#12187 · Filter by author NaderAlAwar opened on Oct 22, 2026
  

• Should environments with duplicate properties, e.g. two tune(...), be rejected? The query uses the first one and ignores the rest

  NVIDIA/cccl#12181 · Filter by author bernhardmgruber opened on Oct 22, 2026
  

• [BUG]: Missing block barriers in libcu++ group_memcpy_async, arrive_tx and pipeline_arrive_on tests

  needs triage
  NVIDIA/cccl#12176 · Filter by author SUKUNA-AI opened on Oct 22, 2026
  1

• [BUG]: CUDASTF tests and examples read fence() timing events without waiting; threads_multiple_graphs unfreezes too early

  needs triage
  NVIDIA/cccl#12175 · Filter by author SUKUNA-AI opened on Oct 22, 2026
  1

• Add BLOCK_LOAD_STRIPED_VECTORIZE to cub::BlockLoad

  NVIDIA/cccl#12170 · Filter by author bernhardmgruber opened on Oct 22, 2026

• [CUB] Use device-word vectors in StoreDirectBlockedVectorized like BlockLoad

  NVIDIA/cccl#12162 · Filter by author bernhardmgruber opened on Oct 22, 2026
  

• Close performance gap between lookup of freshly built and deserialized algorithms

  NVIDIA/cccl#12161 · Filter by author oleksandr-pavlyk opened on Oct 22, 2026

• [FEA]: Accept Min/Max operators in keys-only sort in cub::detail::WarpBitonicSort

  cub
  NVIDIA/cccl#12160 · Filter by author lennartvoelz opened on Oct 22, 2026
  7

• [FEA]: Add CUDA Fast Math configuration in CI

  infra
  NVIDIA/cccl#12151 · Filter by author fbusato opened on Oct 21, 2026

• [BUG]: ReverseIterator + n moves the underlying iterator forward instead of backward

  needs triage
  NVIDIA/cccl#12141 · Filter by author v0ropaev opened on Oct 21, 2026
  1

• [BUG]: Thrust vector insert() writes out of bounds when size() + n overflows

  needs triage
  NVIDIA/cccl#12136 · Filter by author SUKUNA-AI opened on Oct 21, 2026
  1

• [BUG]: shared_memory_barrier status inspection traps on GeForce sm_120: cudaFabricOpErrorStatusCount returns cudaErrorNotSupported

  needs triage
  NVIDIA/cccl#12130 · Filter by author SUKUNA-AI opened on Oct 21, 2026
  1

  

• [BUG]: cuda.compute caches treat lambda x: x * 1 and lambda x: x * 1.0 as the same operator

  needs triage
  NVIDIA/cccl#12128 · Filter by author v0ropaev opened on Oct 21, 2026
  1

• [libcu++] cuda::std::atomic_ref does not compile with host-only MSVC and NVHPC

  NVIDIA/cccl#12120 · Filter by author SamMausberg opened on Oct 21, 2026
  1

  

• [BUG]: libcu++ NVRTC header parsing cost and PCH footprint

  needs triage
  Bug

  NVIDIA/cccl#12057 · Filter by author lamarrr opened on Oct 17, 2026
  2

  

• [FEA]: Make __resizable_buffer public in some capacity

  NVIDIA/cccl#12051 · Filter by author KyleFromNVIDIA opened on Oct 16, 2026

• [FEA]: a typed pipeline for cuda.coop on Numba-CUDA-MLIR

  cuda.coopneeds triage
  NVIDIA/cccl#12022 · Filter by author diptorupd opened on Oct 15, 2026
  1

  

• [BUG]: Using ProxyArray with cuda.compute.make_merge_sort fails to compile

  Bug

  NVIDIA/cccl#12021 · Filter by author oleksandr-pavlyk opened on Oct 15, 2026
  2

• [CUB] Track warp reduction optimizations

  NVIDIA/cccl#11999 · Filter by author fbusato opened on Oct 13, 2026
  2

• Investigate implementing some of c2h helpers' machine for cuda::buffer support to the CUDA buffer or memory resources

  NVIDIA/cccl#11990 · Filter by author oleksandr-pavlyk opened on Oct 12, 2026

• [BUG]: Incorrect test asserts in libcudacxx/test/libcudacxx/std - they constant evaluate on Clang

  needs triage
  Bug

  NVIDIA/cccl#11989 · Filter by author seanbaxter opened on Oct 12, 2026
  2