Search results

[BUG]: STF multi-GPU host_launch write intermittently triggers cudaErrorLaunchFailure with registered host memory

Status: Open.

#7490 In NVIDIA/cccl;

· 19970126ljl opened on Feb 4, 2026

Update cuda/atomic backend to be able to use new NVVM intrinsics when available

Status: Open.

#7480 In NVIDIA/cccl;

· jrhemstad opened on Feb 3, 2026

Refactor cccl.c unique_by_key to new tuning API

Status: Open.

#7479 In NVIDIA/cccl;

· jrhemstad opened on Feb 3, 2026

Refactor cccl.c three_way_partition to new tuning API

Status: Open.

#7478 In NVIDIA/cccl;

· jrhemstad opened on Feb 3, 2026

Refactor cccl.c segmented_sort to use new tuning API

Status: Open.

#7477 In NVIDIA/cccl;

· jrhemstad opened on Feb 3, 2026

Refactor cccl.c scan to use tuning API

Status: Open.

#7476 In NVIDIA/cccl;

· jrhemstad opened on Feb 3, 2026

Refactor cccl.c <code>histogram</code> to use new tuning API

Status: Open.

#7475 In NVIDIA/cccl;

· jrhemstad opened on Feb 3, 2026

[RFC] Decide the fate of CUB iterators

Status: Open.

#7468 In NVIDIA/cccl;

· bernhardmgruber opened on Feb 3, 2026

Finalize replacing thrust with cuda iterators

Status: Open.

#7467 In NVIDIA/cccl;

· bernhardmgruber opened on Feb 3, 2026

Remove all CUB dispatchers

Status: Open.

#7466 In NVIDIA/cccl;

· bernhardmgruber opened on Feb 3, 2026

Make the new CUB tuning API public

Status: Open.

#7465 In NVIDIA/cccl;

· bernhardmgruber opened on Feb 3, 2026

Use the new CUB tuning APIs internally

Status: Open.

#7464 In NVIDIA/cccl;

· bernhardmgruber opened on Feb 3, 2026