Skip to content

Pull requests: NVIDIA/nccl

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
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

GIN: multi-thread GIN proxy with per-thread endpoint assignment
#2279 opened Jul 9, 2026 by mozarhua Contributor Loading…
Add profiler proxy poll backoff
#2277 opened Jul 8, 2026 by pawelchcki Loading…
efa_gda: Add support for EFA GDA in GIN
#2273 opened Jul 7, 2026 by bhasunit Contributor Loading…
Add GIN GPU/NIC peer affinity mapping
#2269 opened Jul 5, 2026 by hzt123123 Loading…
Avoid RMA proxy when cuMem host is unavailable
#2253 opened Jun 25, 2026 by jm99k56 Loading…
fix: propagate error from ncclCudaContextTrack
#2249 opened Jun 24, 2026 by EylonKrause Loading…
net_ib: free GIN barrier buffer on failure
#2227 opened Jun 9, 2026 by wanglei875 Contributor Loading…
Remove misleading NCCL kernel protocol suffix
#2224 opened Jun 8, 2026 by sanrise Loading…
Add NCCL task protocol trace metadata
#2223 opened Jun 8, 2026 by sanrise Loading…
ProTip! no:milestone will show everything without a milestone.