Skip to content

Pull requests: microsoft/mscclpp

New pull request New
24 Open 630 Closed
24 Open 630 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

support rocm7.2
#819 opened Jun 17, 2026 by Binyang2014 Contributor • Draft
1
MoE Commnucator design doc
#818 opened Jun 16, 2026 by Binyang2014 Contributor Loading…
Add support for allgather nvls algorithms
#817 opened Jun 8, 2026 by Empyreus Contributor Loading…
25
Extending DSL Executor CI tests for multi node envinronment
#812 opened May 22, 2026 by caiomcbr Contributor Loading…
2
GB200 support: SendRecv DSL collective and per-channel executor connections
#810 opened May 20, 2026 by Binyang2014 Contributor Loading…
9
Change DefaultMaxCqPollNum to 256 & minor IB CI fixes
#808 opened May 20, 2026 by chhwang Contributor Loading…
2
Comprehensive GDRCopy version check
#805 opened May 15, 2026 by chhwang Contributor Loading…
6
Multinode NVL/NVLS supports
#798 opened May 6, 2026 by Binyang2014 Contributor Loading…
12
Migrate MoE dispatch / combine primitives into MSCCL++
#796 opened May 4, 2026 by seagater Contributor Loading…
26
Add FP8 conversion unit tests
#789 opened Apr 17, 2026 by Binyang2014 Contributor Loading…
Add atomicAdd method for PortChannel
#784 opened Apr 13, 2026 by chhwang Contributor • Draft
Support TBG inside Pipeline
#783 opened Apr 11, 2026 by caiomcbr Contributor Loading…
9
TorchComms Integration for MSCCL++
#771 opened Apr 5, 2026 by michael-beebe Loading…
20
Unique QP per channel and env-controlled GID index for executor
#762 opened Mar 9, 2026 by Binyang2014 Contributor • Draft
10
Align the data element counts for Allreduce nvls kernel
#697 opened Dec 1, 2025 by seagater Contributor Loading…
9
Alltoallv with dynamic execution plan template
#629 opened Sep 9, 2025 by seagater Contributor Loading…
Track Intra Rank Data Dependency and Automatically Sync Thread in the same Rank
#627 opened Sep 8, 2025 by caiomcbr Contributor • Draft
Fifo test with data transfer between multiple GPUs
#596 opened Aug 5, 2025 by seagater Contributor Loading…
Try to extend the connection
#578 opened Jul 23, 2025 by Binyang2014 Contributor • Draft
Fifo test with multiple gpus
#573 opened Jul 22, 2025 by seagater Contributor Loading…
33
Add indirect connection support
#571 opened Jul 19, 2025 by hfhongzy Loading…
4
Exchange recvbuff at the beginning of the CUDA kernel call.
#464 opened Feb 19, 2025 by SreevatsaAnantharamu Contributor Loading…
1
A two-stage copy design with scratch buffer
#445 opened Jan 8, 2025 by SreevatsaAnantharamu Contributor Loading…
1
Support Simple protocol plans using scratch buffer
#371 opened Oct 22, 2024 by yzygitzh Contributor Loading…
6
ProTip! Add no:assignee to see everything that’s not assigned.