microsoft/mscclpp issues and pull requests

#377 - [Bug] flush() hang bug.

Issue - State: closed - Opened by TonyWu199 3 months ago

#376 - Improve CMake options

Pull Request - State: closed - Opened by chhwang 3 months ago

#375 - NVLS support for msccl++ executor

Pull Request - State: closed - Opened by Binyang2014 3 months ago

#374 - [docs] fix quickstart link

Pull Request - State: closed - Opened by jeffra 3 months ago

#373 - Update README.md

Pull Request - State: closed - Opened by adk9 3 months ago - 1 comment

#372 - Fix in-place all-gather input buffer in executor_test

Pull Request - State: closed - Opened by yzygitzh 4 months ago

#371 - Support Simple protocol plans using scratch buffer

Pull Request - State: open - Opened by yzygitzh 4 months ago - 2 comments

#370 - Update docker image for cuda12.4

Pull Request - State: closed - Opened by Binyang2014 4 months ago

#370 - Update docker image for cuda12.4

Pull Request - State: closed - Opened by Binyang2014 4 months ago

#369 - Fix algo repo name

Pull Request - State: closed - Opened by Binyang2014 4 months ago

#369 - Fix algo repo name

Pull Request - State: closed - Opened by Binyang2014 4 months ago

#368 - [Feature] Immediate data upon signal & wait

Issue - State: open - Opened by chhwang 4 months ago

#367 - Fix copyright messages

Pull Request - State: closed - Opened by chhwang 4 months ago

#367 - Fix copyright messages

Pull Request - State: closed - Opened by chhwang 4 months ago

#366 - signal/poll optimization

Pull Request - State: open - Opened by liangyuRain 4 months ago - 9 comments

#366 - signal/poll optimization

Pull Request - State: open - Opened by liangyuRain 4 months ago - 9 comments

#365 - Executor AllGather In-Place Support

Pull Request - State: closed - Opened by caiomcbr 4 months ago

#365 - Executor AllGather In-Place Support

Pull Request - State: closed - Opened by caiomcbr 4 months ago

#364 - Perf optimization & support clipping

Pull Request - State: closed - Opened by chhwang 4 months ago - 1 comment

#364 - Perf optimization & support clipping

Pull Request - State: closed - Opened by chhwang 4 months ago - 1 comment

#363 - Fix NCCL API bugs

Pull Request - State: closed - Opened by chhwang 4 months ago - 2 comments

#363 - Fix NCCL API bugs

Pull Request - State: closed - Opened by chhwang 4 months ago - 2 comments

#362 - [Perf] Failed to reproduce the performance result for Single-node AllReduce mentioned in README.md

Issue - State: closed - Opened by FC-Li 4 months ago - 5 comments

#361 - [Feature] Can NVIDIA and AMD communicate?

Issue - State: closed - Opened by liuyang6055 4 months ago - 1 comment

#360 - Allreduce performance optimization and correctness fix

Pull Request - State: open - Opened by nusislam 4 months ago - 1 comment

#359 - [Bug] Can't launch allreduce test

Issue - State: closed - Opened by chenhongyu2048 5 months ago - 1 comment

#358 - Update integration-test-rocm.yml for Azure Pipelines

Pull Request - State: closed - Opened by Binyang2014 5 months ago

#357 - Update ROCm CI

Pull Request - State: closed - Opened by chhwang 5 months ago - 1 comment

#356 - Fix NPKit exit event offset

Pull Request - State: closed - Opened by yzygitzh 5 months ago

#355 - Use IB transport flags only when an IB device exists

Pull Request - State: closed - Opened by chhwang 5 months ago

#354 - [TEST PR] Update ib.cc

Pull Request - State: closed - Opened by EricWangCN 5 months ago

#353 - Fixing RegisterMemory Allocation for ProxyChannels

Pull Request - State: closed - Opened by caiomcbr 5 months ago

#352 - Fixing RegisterMemory Allocation for ProxyChannels

Pull Request - State: closed - Opened by caiomcbr 5 months ago

#351 - Add proxy channel related operations

Pull Request - State: closed - Opened by Binyang2014 5 months ago

#349 - [Bug] libmscclpp_nccl fails linking using ROCm 6.0

Issue - State: closed - Opened by corey-derochie-amd 5 months ago - 2 comments

#348 - [Doc] mscclpp docs

Pull Request - State: closed - Opened by Binyang2014 5 months ago

#346 - Add CI for rocm

Pull Request - State: closed - Opened by Binyang2014 5 months ago

#345 - Tune threads per block for mscclpp executor

Pull Request - State: closed - Opened by Binyang2014 5 months ago

#326 - Multi-stream CUDA IPC

Pull Request - State: open - Opened by chhwang 7 months ago

#100 - Fix hanging bootstrap issues

Pull Request - State: closed - Opened by chhwang over 1 year ago - 3 comments

#100 - Fix hanging bootstrap issues

Pull Request - State: closed - Opened by chhwang over 1 year ago - 3 comments

#99 - [Feature] Support `POLL_PRINT_ON_STUCK`

Issue - State: closed - Opened by chhwang over 1 year ago

#99 - [Feature] Support `POLL_PRINT_ON_STUCK`

Issue - State: closed - Opened by chhwang over 1 year ago

#98 - Make bootstrap use persistent sockets

Pull Request - State: closed - Opened by olsaarik over 1 year ago - 3 comments

#98 - Make bootstrap use persistent sockets

Pull Request - State: closed - Opened by olsaarik over 1 year ago - 3 comments

#97 - Create Azure pipeline for multi-node tests

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#97 - Create Azure pipeline for multi-node tests

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#96 - Saemal/atomic signal

Pull Request - State: closed - Opened by chhwang over 1 year ago

#96 - Saemal/atomic signal

Pull Request - State: closed - Opened by chhwang over 1 year ago

#95 - Misc updates

Pull Request - State: closed - Opened by chhwang over 1 year ago

#95 - Misc updates

Pull Request - State: closed - Opened by chhwang over 1 year ago

#94 - Use MSCCL++ Docker image for CodeQL

Pull Request - State: closed - Opened by chhwang over 1 year ago

#94 - Use MSCCL++ Docker image for CodeQL

Pull Request - State: closed - Opened by chhwang over 1 year ago

#93 - Setup Azure pipeline to run mscclpp-test

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#93 - Setup Azure pipeline to run mscclpp-test

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#92 - [Bug] Hanging bootstrap communication

Issue - State: closed - Opened by chhwang over 1 year ago - 4 comments

#92 - [Bug] Hanging bootstrap communication

Issue - State: closed - Opened by chhwang over 1 year ago - 4 comments

#91 - A test for the core assumption of data + signal of MSCCL++

Issue - State: closed - Opened by saeedmaleki over 1 year ago - 2 comments

#91 - A test for the core assumption of data + signal of MSCCL++

Issue - State: closed - Opened by saeedmaleki over 1 year ago - 2 comments

#90 - [Performance] AllReduce performance debugging

Issue - State: closed - Opened by chhwang over 1 year ago - 5 comments

#90 - [Performance] AllReduce performance debugging

Issue - State: closed - Opened by chhwang over 1 year ago - 5 comments

#89 - MSCCL++ v0.3.0 Release Plan

Issue - State: closed - Opened by chhwang over 1 year ago - 1 comment

#89 - MSCCL++ v0.3.0 Release Plan

Issue - State: closed - Opened by chhwang over 1 year ago - 1 comment

#88 - Update docs

Pull Request - State: closed - Opened by chhwang over 1 year ago

#88 - Update docs

Pull Request - State: closed - Opened by chhwang over 1 year ago

#87 - [mscclpp-test] Add AllToAll tests

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#87 - [mscclpp-test] Add AllToAll tests

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#86 - Cleanup old files and functions

Pull Request - State: closed - Opened by chhwang over 1 year ago

#86 - Cleanup old files and functions

Pull Request - State: closed - Opened by chhwang over 1 year ago

#85 - Add packet copy (LL) for AllReduce

Pull Request - State: closed - Opened by chhwang over 1 year ago - 4 comments

#85 - Add packet copy (LL) for AllReduce

Pull Request - State: closed - Opened by chhwang over 1 year ago - 4 comments

#84 - Implement host offload algorithm for allgather

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#84 - Implement host offload algorithm for allgather

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#83 - [mscclpp-test] Add AllReduce and AllToAll tests

Pull Request - State: closed - Opened by chhwang over 1 year ago - 4 comments

#83 - [mscclpp-test] Add AllReduce and AllToAll tests

Pull Request - State: closed - Opened by chhwang over 1 year ago - 4 comments

#82 - Remove alloc.h and beef up cuda_utils.hpp

Pull Request - State: closed - Opened by olsaarik over 1 year ago

#82 - Remove alloc.h and beef up cuda_utils.hpp

Pull Request - State: closed - Opened by olsaarik over 1 year ago

#81 - Update unit tests

Pull Request - State: closed - Opened by chhwang over 1 year ago - 1 comment

#81 - Update unit tests

Pull Request - State: closed - Opened by chhwang over 1 year ago - 1 comment

#80 - Fix CodeQL workflow

Pull Request - State: closed - Opened by chhwang over 1 year ago

#80 - Fix CodeQL workflow

Pull Request - State: closed - Opened by chhwang over 1 year ago

#79 - Update CI

Pull Request - State: closed - Opened by chhwang over 1 year ago

#79 - Update CI

Pull Request - State: closed - Opened by chhwang over 1 year ago

#78 - Add allgather test to mscclpp-test

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago - 3 comments

#78 - Add allgather test to mscclpp-test

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago - 3 comments

#77 - Rewrite mscclpp-test with cpp style API

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#77 - Rewrite mscclpp-test with cpp style API

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#76 - Cuda smart pointers

Pull Request - State: closed - Opened by olsaarik over 1 year ago

#76 - Cuda smart pointers

Pull Request - State: closed - Opened by olsaarik over 1 year ago

#75 - Merge api-extension branch to main

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#75 - Merge api-extension branch to main

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#74 - offloading allgather to CPU entirely

Pull Request - State: closed - Opened by saeedmaleki over 1 year ago

#74 - offloading allgather to CPU entirely

Pull Request - State: closed - Opened by saeedmaleki over 1 year ago

#73 - Refine exception

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#73 - Refine exception

Pull Request - State: closed - Opened by Binyang2014 over 1 year ago

#72 - Add Doxygen-based document

Pull Request - State: closed - Opened by yzygitzh almost 2 years ago

#72 - Add Doxygen-based document

Pull Request - State: closed - Opened by yzygitzh almost 2 years ago

#71 - Resolve conflict and merge main branch to api-extension

Pull Request - State: closed - Opened by Binyang2014 almost 2 years ago

#71 - Resolve conflict and merge main branch to api-extension

Pull Request - State: closed - Opened by Binyang2014 almost 2 years ago

#70 - Update sm copy test

Pull Request - State: closed - Opened by Binyang2014 almost 2 years ago - 2 comments

GitHub / microsoft/mscclpp issues and pull requests