MSCCL++ v0.7.0

What's Changed

• Move pipeline to official org by @Binyang2014 in #406
• Disable CuMemMap check for ROCm by @Binyang2014 in #411
• NVLS support for NCCL API by @Binyang2014 in #410
• Supporting multi-node executors in NCCL API by @caiomcbr in #412
• Fix synchronization in allreduce8 kernel by @dsidler in #407
• Add ncclBcast / ncclBroadcast support by @SreevatsaAnantharamu in #419
• Update README by @Binyang2014 in #414
• Fix nccl-test failure issue by @Binyang2014 in #421
• Tackle build warnings by @chhwang in #422
• trigger ci for release branches by @Binyang2014 in #426
• Fix CI trigger issue by @Binyang2014 in #428
• Fix typos in the pipeline by @chhwang in #420
• Update version number by @Binyang2014 in #433
• Enhance the nccl error message handling by @seagater in #434
• [NPKIT] Adding the NPKIT support for kernel allreduce7 in mscclpp-nccl by @PedramAlizadeh in #399
• Fix azure pipeline by @Binyang2014 in #437
• Add GpuBuffer class by @chhwang in #423
• Fix CMake build messages by @chhwang in #443
• Flushing Proxy Channels at CPU side upon reaching the Inflight Request Limit by @caiomcbr in #415
• Fix Python binding of exceptions by @chhwang in #444
• Auto-update version numbers in CMakeLists.txt by @chhwang in #450
• Resolve cuMemMap error by @Binyang2014 in #451
• Manage runtime environments by @chhwang in #452
• Lazily create streams for CudaIpcConnection by @chhwang in #449
• Fix PR #449 by @chhwang in #453
• Merge mscclpp-lang to mscclpp project by @Binyang2014 in #442
• Renaming channels by @chhwang in #436
• Add multi-nodes example & update doc by @Binyang2014 in #455
• Adjusting BFS to seek circular dependencies in the msccl-tools DAG by @caiomcbr in #459
• remove unnecessary sync by @Binyang2014 in #461
• Support ReduceScatter in the NCCL interface by @caiomcbr in #460
• Updating MSCCLLang Examples by @caiomcbr in #462
• Disable channel cache by @seagater in #463
• Adjusting AllGather Collective in MSCCLLang by @caiomcbr in #466
• Adding Read Put Packet operation at Executor by @caiomcbr in #441
• NPKit Support to Read Put Packet Operation by @caiomcbr in #471
• Adjust NPKit IB Event by @caiomcbr in #472
• Fix minor typos and errors in documentation by @RyoYang in #474
• Improving Get Operation at MSCCLLang by @caiomcbr in #475
• Fix memory OOM issue by @Binyang2014 in #479
• Mark mscclpp-test as deprecated in the doc by @chhwang in #478
• Update allgather fallback algo by @Binyang2014 in #476
• Add min operation for allreduce by @Binyang2014 in #481
• NCCL API CI Test for ReduceScatter by @caiomcbr in #465
• Fix correctness issue when mscclppDisableChannelCache set to true by @Binyang2014 in #483
• nccl/rccl integration by @seagater in #469
• Fix reduceMin failaure issue by @Binyang2014 in #486
• Reduce Operation Support to the Executor by @caiomcbr in #484
• Add CI test for fallback allgather, allreduce, broadcastand reducescatter to NCCL operations by @seagater in #485
• Remove the requirement for CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_FABRIC_SUPPORTED for NVLS support by @Binyang2014 in #489
• Add CUDA 12.8 images by @chhwang in #488
• Add a devcontainer configuration by @chhwang in #490
• Fix CMake installation in Dockerfile for arm64 by @chhwang in #491
• Export mscclpp GpuBuffer to dlpack format by @Binyang2014 in #492
• Fix the virtual address mapping issue of cuMemMap in fallback code by @seagater in #501
• Improve signal/wait performance and fix barrier issue by @Binyang2014 in #499
• Fix performance issue introduced in PR: 499 by @Binyang2014 in #505
• Add flag to disable nvls by @Binyang2014 in #500
• Optimized allreduce fallback for ~10KB sizes by @chhwang in #506
• Automatic creation of Scratch Buffer at MSCCLLang by @caiomcbr in #510
• Use implicit ctors for default device ctors by @chhwang in #512
• apps/nccl: fix a bug in allreduce kernels for graph mode by @nusislam in #502
• Revised MemoryChannel interfaces by @chhwang in #508
• Fix #508 by @chhwang in #515
• Add NVLS based fallback algo by @Binyang2014 in #507
• Enhance Collective Check at MSCCLang by @caiomcbr in #511
• Support ibv_reg_dmabuf_mr for buffer allocated by cuMemMalloc by @seagater in #513
• Fix the issue of echo message for nccl fallback in CI test by @seagater in #520
• Asynchronous setup by @chhwang in #514
• Adding maxSpinCount to port channel flush by @Binyang2014 in #518
• Fix device assert by @chhwang in #522
• Fix #514 by @chhwang in #521
• Add a CMake option MSCCLPP_GPU_ARCHS by @chhwang in #525
• Update citations by @chhwang in #524
• Set Up a CI Pipeline for H100 by @Binyang2014 in #526
• Properly setting up the device in Ethernet Connection by @caiomcbr in #527
• Add device semaphore API by @Binyang2014 in #523
• Address NVCC warning #20012-D by @chhwang in #528
• Rename ChannelTrigger fields and check field values in debug builds by @chhwang in #529
• DLPack fixes by @chhwang in #537
• Improved documentation & minor interface revision by @chhwang in #541
• Use a stream pool for gpuCalloc*() by @chhwang in #509
• Multi-stream CUDA IPC by @chhwang in #326
• Fix #509 by @chhwang in #546
• Fix build processes by @chhwang in #545
• Do not use tail replica by default by @chhwang in #544
• DeviceSemaphore fix by @Binyang2014 in #553
• Fix some typos in docs by @Edenzzzz in #555
• New FIFO test by @chhwang in #558
• FIFO improvements by @chhwang in #557
• Fix #557 by @chhwang in #560
• Support connection between local endpoints by @chhwang in #561
• Fix multi-nodes CI pipeline by @Binyang2014 in #564
• Support any GPUs per node for NCCL_API by @Binyang2014 in #566
• Fix pytest failure by @Binyang2014 in #567
• Fix a FIFO correctness bug by @chhwang in #549
• New semaphore constructors by @chhwang in #559
• Revise NVLS interface by @chhwang in #458
• update readme & bump version by @Binyang2014 in #550

New Contribu...

MSCCL++ v0.6.0

Highlight

• Improved NCCL API integration in MSCCL++ for better performance and usability
• Enhanced execution plan-based executor in MSCCL++
• Fixed several bugs to improve stability and reliability

What's Changed

• Add support for different vector sizes in multimem instructions by @roshandathathri in #332
• NCCL API Executor Integration by @caiomcbr in #331
• Fix missing import in executor test by @yzygitzh in #334
• bfloat16 support by @chhwang in #336
• Dynamically load libibverbs by @caiomcbr in #337
• Auto-tune vector sizes for NVLS allreduce6 by @roshandathathri in #338
• Make ibverbs optional at compile time by @chhwang in #340
• ProxyChannel Support in Executor by @caiomcbr in #342
• Support executors to send packets over ProxyChannel by @caiomcbr in #344
• Fix for ROCm 6.0 by @chhwang in #347
• Fix bug for construct sempaphore by @Binyang2014 in #341
• Add proxy channel related operations by @Binyang2014 in #351
• Add CI for rocm by @Binyang2014 in #346
• Tune threads per block for mscclpp executor by @Binyang2014 in #345
• Fix NPKit exit event offset by @yzygitzh in #356
• Use IB transport flags only when an IB device exists by @chhwang in #355
• Update ROCm CI by @chhwang in #357
• Fixing RegisterMemory Allocation for ProxyChannels by @caiomcbr in #353
• Fix NCCL API bugs by @chhwang in #363
• Perf optimization & support clipping by @chhwang in #364
• Fix copyright messages by @chhwang in #367
• [Doc] mscclpp docs by @Binyang2014 in #348
• Executor AllGather In-Place Support by @caiomcbr in #365
• Fix algo repo name by @Binyang2014 in #369
• Update docker image for cuda12.4 by @Binyang2014 in #370
• Fix in-place all-gather input buffer in executor_test by @yzygitzh in #372
• [docs] fix quickstart link by @jeffra in #374
• Add kernel-based verification for executor_test by @yzygitzh in #378
• Lazily create the context stream by @chhwang in #381
• Fixing Bug Const Offset in Execution Plan by @caiomcbr in #380
• Fix light load bug by @Binyang2014 in #379
• Small Adjust in Test Data AllGather at Executor Test by @caiomcbr in #384
• Fix missing packet parameter for executor by @yzygitzh in #385
• NVLS support for msccl++ executor by @Binyang2014 in #375
• Fix typo by @Binyang2014 in #389
• Improve CMake options by @chhwang in #376
• Fixing Message Boundary AllReduce Fallback Code by @caiomcbr in #391
• Fix mscclpp_benchmark by @Binyang2014 in #392
• Add cross threadblock barrier by @Binyang2014 in #383
• AllGather Executor Support in NCCL Interface by @caiomcbr in #393
• Providing reduce-scatter test support by @caiomcbr in #390
• Select algo according to json config by @Binyang2014 in #396
• Add connection events for NPKit by @yzygitzh in #386
• Revised ProxyChannel interfaces by @chhwang in #400
• Setup pipeline for mscclpp over nccl by @Binyang2014 in #401
• Exception Max Number Operation per Tb by @caiomcbr in #405
• Reduce memory usage for scratch buffer by @Binyang2014 in #403
• [Cherry-pick] Move pipeline to official org (#406) by @Binyang2014 in #416
• [Cherry-pick] trigger ci for release branches (#426) by @Binyang2014 in #427
• [Cherry-pick] Disable CuMemMap check for ROCm (#411) by @Binyang2014 in #424
• [Cherry-pick] NVLS support for NCCL API (#410) by @Binyang2014 in #425
• [Cherry-pick] Fix nccl-test failure issue (#421) by @Binyang2014 in #429

New Contributors

• @jeffra made their first contribution in #374

Full Changelog: v0.5.2...v0.6.0

MSCCL++ v0.5.2

What's Changed

• Add C++ executor test by @chhwang in #304
• Cumulative Updates by @Binyang2014 in #309
• Add NPKit GPU event support by @yzygitzh in #310
• Fix NPKit support for AMD by @yzygitzh in #312
• Add "packet type" option for executor test by @Binyang2014 in #313
• Add support for multicast reduce insruction by @roshandathathri in #316
• Update quickstart.md by @angelica-moreira in #314
• Simplify/improve barrier in AllReduce6 by @roshandathathri in #317
• Support NCCL APIs by @caiomcbr in #319
• Update allreduce_bench.py by @angelica-moreira in #318
• Separate NPKit CPU timestamp access from different blocks for AMD platform by @yzygitzh in #321
• AllReduce Kernel for Small Messages by @caiomcbr in #322
• Resolve clang++ warnings by @chhwang in #325
• Support to write packets via uint2 by @Binyang2014 in #327
• Double buffering for NCCL APIs by @caiomcbr in #324
• v0.5.2 by @chhwang in #328

New Contributors

• @angelica-moreira made their first contribution in #314
• @caiomcbr made their first contribution in #319

Full Changelog: v0.5.1...v0.5.2

MSCCL++ v0.5.1

What's Changed

• Upgrade gtest by @chhwang in #300
• Rename executor.cpp to executor_py.cpp by @chhwang in #301
• Fix assert declaration & add a compile test by @chhwang in #303
• Fix security issue by @Binyang2014 in #305
• v0.5.1 by @chhwang in #308

Full Changelog: v0.5.0...v0.5.1

MSCCL++ v0.5.0

What's Changed

• Fix a typo name by @chhwang in #286
• Add executor to execute schedule-plan file by @Binyang2014 in #283
• Allow binding allocated memory to NVLS multicast pointer by @roshandathathri in #290
• Seperate headers for GPU data types by @chhwang in #291
• Refactoring NVLS interfaces by @chhwang in #293
• Include GPU data types only for kernel code by @chhwang in #292
• Ethernet support by @chhwang in #284
• Resolve multi-nodes test failure issue by @Binyang2014 in #295
• Move pipeline to Azure org by @Binyang2014 in #296
• Optimized the execution kernel by @Binyang2014 in #294
• Allow obtaining cuda stream handle from PyTorch stream when launching kernel by @aashaka in #297
• v0.5.0 by @chhwang in #298

New Contributors

• @roshandathathri made their first contribution in #290

Full Changelog: v0.4.3...v0.5.0

MSCCL++ v0.4.3

What's Changed

• Add optional prefix to installation paths by @chhwang in #235
• Fix #235 by @chhwang in #239
• Check nvidia_peermem during runtime by @chhwang in #234
• Do not check value of __HIP_PLATFORM_AMD__ by @chhwang in #240
• Fix crash in static variable deconstructor by @Binyang2014 in #238
• Update interface to let user change fifo size by @Binyang2014 in #243
• Mask each fields of the trigger by @chhwang in #244
• Minor improvement on device syncer by @chhwang in #231
• remove make pylib-copy command by @Binyang2014 in #249
• Increase MSCCLPP_BITS_REGMEM_HANDLE to 9 by @aashaka in #251
• Add putWithSignal() latency tests by @chhwang in #246
• NVLS support. by @saeedmaleki in #250
• Fix wrong offset calculation by @chhwang in #257
• Fix NVLS support by @chhwang in #258
• Allow MSCCL++ CommGroup to take PyTorch tensors in args by @aashaka in #255
• Fix multi-nodes test failure by @Binyang2014 in #262
• Allow semaphores and memory to be registered separately in ProxyService by @aashaka in #264
• Remove cuda-python from project by @Binyang2014 in #245
• Fix the comm.py for nvls by @saeedmaleki in #267
• New packet format & optimizations by @chhwang in #256
• Fix multi-node ci pipeline by @Binyang2014 in #272
• add launch_bounds for mscclpp_test by @Binyang2014 in #273
• Fix bootstrapping mechanism by @chhwang in #278
• v0.4.3 by @chhwang in #279

New Contributors

• @aashaka made their first contribution in #251

Full Changelog: v0.4.2...v0.4.3

MSCCL++ v0.4.2

What's Changed

• Include cstdint in packet_device.hpp by @chhwang in #233
• Fix & improve perf for ROCm by @chhwang in #232
• v0.4.2 by @chhwang in #236

Full Changelog: v0.4.1...v0.4.2

MSCCL++ v0.4.1

What's Changed

• Fix performance downgrade issue & update doc by @Binyang2014 in #229
• Add a documentation issue template by @chhwang in #230

Full Changelog: v0.4.0...v0.4.1

MSCCL++ v0.4.0

• Add Python benchmark
• Update documentation
• Add ROCm support
• Bug fixes

See details from #160.

MSCCL++ v0.3.0

• Updated interfaces
• Add Python bindings and interfaces
• Add Python unit tests
• Add more configurable parameters
• Add a new single-node AllReduce kernel
• Fix bugs

See details from #89.

Full Changelog: v0.2.0...v0.3.0