Skip to content

Pull requests: NVIDIA/cccl

New pull request New
203 Open 4,047 Closed
203 Open 4,047 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[Backport 3.2.x] [cuda.compute] Add dependency on nvidia-nvvm #6909
#6949 opened Dec 11, 2025 by shwina • Draft
2 tasks
2
Avoid using _CCCL_UNREACHABLE() unless it's necessary
#6948 opened Dec 11, 2025 by davebayer • Draft
5
Remove [[nodiscard]] from barrier's .arrive(...) method backport branch/3.1.x backport branch/3.2.x
#6947 opened Dec 11, 2025 by davebayer Loading…
1
@davebayer
5
[Backport 3.0] Properly specialize cub functions for __nv_bfloat16 (#6931)
#6946 opened Dec 11, 2025 by miscco Loading…
2 tasks
[Backport 3.1] Properly specialize cub functions for __nv_bfloat16 (#6931)
#6945 opened Dec 11, 2025 by miscco Loading…
2 tasks
1
[Backport 3.0] Add _CCCL_DECLSPEC_EMPTY_BASES to mdspan features (#6444)
#6944 opened Dec 11, 2025 by miscco Loading…
1
[DRAFT] Optimize offset based DeviceSegmentedReduce for small and medium segment sizes
#6942 opened Dec 11, 2025 by srinivasyadav18 Loading…
4 tasks
1
1
[cuda.compute] Refactor code for creating void* wrappers to avoid ODR violations
#6941 opened Dec 10, 2025 by shwina Loading…
2 tasks
3
[Backport branch/3.2.x] Properly specialize cub functions for __nv_bfloat16
#6940 opened Dec 10, 2025 by github-actions bot Loading…
4
[Backport branch/3.2.x] [libcu++] Rename device_transform back to launch_transform
#6937 opened Dec 10, 2025 by github-actions bot Loading…
3
[Backport branch/3.2.x] [libcu++] Static assert that resource is copyable in buffer constructors
#6936 opened Dec 10, 2025 by github-actions bot Loading…
4
[Backport branch/3.2.x] Make sure we actually use overflow builtins
#6934 opened Dec 10, 2025 by github-actions bot Loading…
3
[Backport branch/3.2.x] Add missing nvrtc nv target archs
#6933 opened Dec 10, 2025 by github-actions bot Loading…
4
[BACKPORT 3.0] Update cuda/ptx instructions to support all new SM architectures in CTK 13 (#5600)
#6930 opened Dec 10, 2025 by miscco Loading…
7
Expose not guaranteed determinism to reduce in cuda.compute
#6926 opened Dec 9, 2025 by NaderAlAwar Loading…
1 of 2 tasks
8
Implement std::geometric_distribution
#6924 opened Dec 9, 2025 by RAMitchell Loading…
1
Enable <cuda/device> APIs to be used on device
#6922 opened Dec 9, 2025 by davebayer • Draft
3
Add internal cuda::__is_device_memory 3.2.0 Targeted for 3.2.0 release
#6918 opened Dec 8, 2025 by fbusato Loading…
@fbusato
10
Remove need for hardcoded LevelT for histogram in c.parallel and cuda.compute
#6915 opened Dec 8, 2025 by NaderAlAwar Loading…
2 tasks
1
11
Implement the new tuning API for DeviceTransform
#6914 opened Dec 8, 2025 by bernhardmgruber Loading…
4 of 7 tasks
1
24
inclusive_scan smoke test
#6913 opened Dec 8, 2025 by viralbhadeshiya Loading…
1 of 2 tasks
5
Smoke test for exclusive_scan
#6912 opened Dec 8, 2025 by viralbhadeshiya Loading…
1 of 2 tasks
5
Add NVTX annotations to cuda.compute user-facing APIs
#6906 opened Dec 8, 2025 by Copilot AI Loading…
1 of 2 tasks
1
@shwina
5
Allow if consteval in device code with nvcc 13.1 backport branch/3.2.x
#6902 opened Dec 7, 2025 by davebayer • Draft
5
[CUB] Reduce BlockAdjacentDifference shared memory usage by 50%
#6901 opened Dec 7, 2025 by Aminsed Loading…
1
1
ProTip! Add no:assignee to see everything that’s not assigned.