cuda.core next

Issues

• cuda.core: support "3D copies with attributes" (follow-on to issue #2365)

  cuda.core
  Enhancement

  NVIDIA/cuda-python#2660 · Filter by author juenglin opened on Aug 18, 2026 · cuda.core next

• fix(core): reject use_libdevice on the NVRTC and PTX backends

  bugcuda.coreP1
  Unread
  NVIDIA/cuda-python#2573·LeSingh1 opened Aug 9, 2026 · Updated Oct 27, 2026
  6

  

• [Pitch] Nightly pytest-randomly flake finder + AI fixer

  P2
  NVIDIA/cuda-python#2088 · Filter by author rwgk opened on May 14, 2026 · cuda.core next
  2

  

• cuda.core: release synchronous allocations in stream order through a host callback instead of blocking

  cuda.coreenhancementP2
  NVIDIA/cuda-python#2989 · Filter by author Andy-Jost opened on Oct 1, 2026 · cuda.core next
  

• Add a snippet-based release note approach

  CI/CDcuda.bindingscuda.corecuda.pathfinder
  Unread
  NVIDIA/cuda-python#3034·mdboom opened Oct 6, 2026 · Updated Oct 28, 2026
  20

  

• [BUG]: NVVM backend silently ignores 12 ProgramOptions fields that NVRTC emits

  bugcuda.coreP2
  Bug

  NVIDIA/cuda-python#2954 · Filter by author Arthur031221 opened on Sep 26, 2026 · cuda.core next
  8

  

• cuda.core: should sticky CUDA errors derive from BaseException?

  cuda.coreRFC
  Enhancement

  NVIDIA/cuda-python#2761 · Filter by author Andy-Jost opened on Sep 3, 2026 · cuda.core next
  

• MNT: Switch to the "src" layout

  cuda.bindingscuda.corecuda.pathfindersupport
  Task

  NVIDIA/cuda-python#3048 · Filter by author leofang opened on Oct 7, 2026 · cuda.core next

• CI: Add nvmath-python integration tests to nightly CI

  CI/CDP1test
  NVIDIA/cuda-python#2340 · Filter by author leofang opened on Jul 11, 2026 · cuda.core next
  3

  

• cuda.core: transfer VMM IPC allocations one at a time

  cuda.coreenhancementP2
  NVIDIA/cuda-python#3110 · Filter by author Andy-Jost opened on Oct 14, 2026 · cuda.core next
  

• cuda.core: ending a GraphBuilder after an invalidated conditional-body capture crashes the process (CUDA driver bug)

  blockedbugcuda.coreP2
  NVIDIA/cuda-python#2918 · Filter by author Andy-Jost opened on Sep 18, 2026 · cuda.core next

• cuda.core: forward ofast_compile to the NVVM backend once libNVVM supports -Ofast-compile

  cuda.coreenhancementP2
  NVIDIA/cuda-python#3206 · Filter by author Andy-Jost opened on Oct 23, 2026 · cuda.core next
  1

  

• CI: Remove runner certificate trust store probe workflow

  CI/CDP1
  NVIDIA/cuda-python#3257 · Filter by author leofang opened on Oct 27, 2026 · cuda.core next
  1

  

• Address known thread-safety issues in cuda.core/look into call_once pattern

  cuda.coreenhancement
  Enhancement

  NVIDIA/cuda-python#1389 · Filter by author leofang opened on Dec 16, 2025 · cuda.core next
  3

  

• [DOC]: 7 dead external links in samples/*/README.md fail pre-commit run --all-files (lychee 404s)

  documentation
  Documentation

  NVIDIA/cuda-python#3075 · Filter by author lijinf2 opened on Oct 12, 2026 · cuda.core next
  

• StridedMemoryView.__dlpack__ raises BufferError for bfloat16 views

  bugcuda.coregood first issueP2
  NVIDIA/cuda-python#3337 · Filter by author 0z5a opened on Nov 3, 2026 · cuda.core next
  3

  

• Provide a fast path for constructing a StridedMemoryView from a cupy.ndarray

  blockedcuda.coreP1performance
  NVIDIA/cuda-python#748 · Filter by author leofang opened on Jul 8, 2025 · cuda.core next
  3

  

• Add cuda.core.utils.get_aligned_dtype

  cuda.corefeatureP1
  NVIDIA/cuda-python#734 · Filter by author leofang opened on Jun 27, 2025 · cuda.core next
  4

  

• Allow user-provided annotation for kernels generated from Program andObjectCode

  cuda.corefeatureP1
  NVIDIA/cuda-python#1281 · Filter by author leofang opened on Nov 20, 2025 · cuda.core next
  2

  

• WIP: Support passing SMV as a kernel argument of type cuda::std::mdspan

  blockedcuda.corefeatureP1
  Unread
  NVIDIA/cuda-python#1387·leofang opened Dec 16, 2025 · Updated Nov 5, 2026·Draft
  3

  

• WIP: Add cuda.core.utils.make_aligned_dtype

  cuda.corefeatureP1
  Unread
  NVIDIA/cuda-python#1636·leofang opened Feb 18, 2026 · Updated Nov 5, 2026·Draft
  6

  

• [DOC]: instructions to compile PTX compatible with system version of cuda-toolkit

  cuda.coredocumentationgood first issueP2
  Documentation

  NVIDIA/cuda-python#3379 · Filter by author murdock-grewar opened on Nov 5, 2026 · cuda.core next
  

• cuda.core tests: run kernel tests on PinnedMemoryResource buffers

  cuda.coreP1test
  NVIDIA/cuda-python#3414 · Filter by author Andy-Jost opened on Nov 9, 2026 · cuda.core next

• cuda.core: PinnedMemoryResource grants access to the current device only

  cuda.coreenhancementP1
  NVIDIA/cuda-python#3415 · Filter by author Andy-Jost opened on Nov 9, 2026 · cuda.core next

• [doc-only] Explain cuda.core support through concrete examples

  cuda.coredocumentation
  Unread
  NVIDIA/cuda-python#3390·rwgk opened Nov 5, 2026 · Updated Nov 10, 2026·Draft
  4