-
Notifications
You must be signed in to change notification settings - Fork 315
OpenJul 30, 2026
Due by September 2, 2026
•Last updated 48% complete
List view
0 of 48 selected 0 issues of 48 selected
cuda.core: make buffer cleanup context-safe
bugSomething isn't workingSomething isn't workingcuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleP0High priority - Must do!High priority - Must do!Status: Draft (not ready).Support CUDA core dump
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should dotriageNeeds the team's attentionNeeds the team's attentionStatus: Open.#1344 In NVIDIA/cuda-python;[cuda.core] Add support for host_launch (host callback nodes / host function launches)
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should dotriageNeeds the team's attentionNeeds the team's attentionStatus: Open.#2058 In NVIDIA/cuda-python;Add
cuda-ccclas a required dependency tocuda-pythoncuda.bindingsEverything related to the cuda.bindings moduleEverything related to the cuda.bindings moduleP1Medium priority - Should doMedium priority - Should dopackagingAnything related to wheels or Conda packagesAnything related to wheels or Conda packagesRFCPlans and announcementsPlans and announcementsStatus: Open.#691 In NVIDIA/cuda-python;Allow user-provided annotation for kernels generated from
ProgramandObjectCodecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Open.#1281 In NVIDIA/cuda-python;WIP: Support passing SMV as a kernel argument of type
cuda::std::mdspanblockedThis task is currently blocked by other tasksThis task is currently blocked by other taskscuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Draft (not ready).WIP: Add
cuda.core.utils.make_aligned_dtypecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Draft (not ready).Add
cuda.core.utils.get_aligned_dtypecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Open.#734 In NVIDIA/cuda-python;Provide a fast path for constructing a
StridedMemoryViewfrom acupy.ndarraycuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleenhancementAny code-related improvementsAny code-related improvementsP1Medium priority - Should doMedium priority - Should doStatus: Open.#748 In NVIDIA/cuda-python;[FEA]: private constructor prevents derived classes of cythonized Stream, Event
bugSomething isn't workingSomething isn't workingcuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduletriageNeeds the team's attentionNeeds the team's attentionStatus: Open.#750 In NVIDIA/cuda-python;[BUG]: Important CI failures are not team-visible
CI/CDCI/CD infrastructureCI/CD infrastructurefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Open.#2204 In NVIDIA/cuda-python;Support batched memory movement
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP1Medium priority - Should doMedium priority - Should doStatus: Open.#1333 In NVIDIA/cuda-python;Revisit user object lifetime testing
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleP1Medium priority - Should doMedium priority - Should dotestImprovements or additions to testsImprovements or additions to testsStatus: Open.#2044 In NVIDIA/cuda-python;Inconsistent lifetime management across
from_handle/from_*APIscuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleenhancementAny code-related improvementsAny code-related improvementsP1Medium priority - Should doMedium priority - Should doStatus: Open.#1989 In NVIDIA/cuda-python;CUDA graph phase N - graph updates
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP0High priority - Must do!High priority - Must do!Status: Open.#1330 In NVIDIA/cuda-python;demo: use pytest
subtestsin cuda.core system device loops to ensure full test coveragecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleP1Medium priority - Should doMedium priority - Should dotestImprovements or additions to testsImprovements or additions to testsStatus: Draft (not ready).Python latency testing & benchmarking
cuda.bindingsEverything related to the cuda.bindings moduleEverything related to the cuda.bindings modulecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleP1Medium priority - Should doMedium priority - Should doStatus: Open.#1580 In NVIDIA/cuda-python;EPIC: Port Rapids projects from pynvml to cuda.core.system
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleEPICSoul of a releaseSoul of a releaseP1Medium priority - Should doMedium priority - Should doStatus: Open.Measure and track
cuda.coreimport timecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleP1Medium priority - Should doMedium priority - Should doStatus: Open.#2045 In NVIDIA/cuda-python;cuda.core graph: host-callback capture path can leave a committed, unowned node
bugSomething isn't workingSomething isn't workingcuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleP1Medium priority - Should doMedium priority - Should doStatus: Open.#2317 In NVIDIA/cuda-python;Support passing structs as kernel
launch()argumentscuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core moduleenhancementAny code-related improvementsAny code-related improvementsP1Medium priority - Should doMedium priority - Should doStatus: Open.#2343 In NVIDIA/cuda-python;CI: Brain-dump the design doc
CI/CDCI/CD infrastructureCI/CD infrastructuredocumentationImprovements or additions to documentationImprovements or additions to documentationP0High priority - Must do!High priority - Must do!Status: Open.#2346 In NVIDIA/cuda-python;cuda.core: add append-only attachment ownership for executable graphs
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP0High priority - Must do!High priority - Must do!Status: Open.#2353 In NVIDIA/cuda-python;cuda.core: add atomic update methods for executable graph nodes
cuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulefeatureNew feature or requestNew feature or requestP0High priority - Must do!High priority - Must do!Status: Open.#2354 In NVIDIA/cuda-python;Test using pytest-run-parallel and related fixups in the tests
CI/CDCI/CD infrastructureCI/CD infrastructurecuda.bindingsEverything related to the cuda.bindings moduleEverything related to the cuda.bindings modulecuda.coreEverything related to the cuda.core moduleEverything related to the cuda.core modulecuda.pathfinderEverything related to the cuda.pathfinder moduleEverything related to the cuda.pathfinder moduleP1Medium priority - Should doMedium priority - Should doStatus: Open (in progress).