Binyang Li 28a57b0610 NVLS support for msccl++ executor (#375) ...

- Support mote datatype for multicast operation
- Add new OP MULTI_LOAD_REDUCE_STORE to support NVLS
- Modify allocSharedPhysicalCuda, which return std::shared_ptr<T>
instead of std::shared_ptr<PhysicalCudaMemory>
- Add Python support for allocSharedPhysicalCuda

Test passed for `allreduce_nvls.json`

2024-11-20 06:43:28 +00:00

..

bootstrap

Fix copyright messages (#367)

2024-10-17 21:25:46 -07:00

executor

NVLS support for msccl++ executor (#375)

2024-11-20 06:43:28 +00:00

include

NVLS support for msccl++ executor (#375)

2024-11-20 06:43:28 +00:00

npkit

Separate NPKit CPU timestamp access from different blocks for AMD platform (#321)

2024-07-02 19:36:48 +08:00

.gitignore

[python] switch to setup.py to build package

2023-04-12 12:29:17 -07:00

c_style_remnants.cc

Add license comments (#106)

2023-06-25 12:40:12 +08:00

CMakeLists.txt

Add executor to execute schedule-plan file (#283)

2024-04-18 19:10:41 +00:00

communicator.cc

Refactoring NVLS interfaces (#293)

2024-04-24 10:05:41 -07:00

connection.cc

Fixing RegisterMemory Allocation for ProxyChannels (#353)

2024-09-24 23:01:41 -07:00

context.cc

Lazily create the context stream (#381)

2024-11-11 10:39:32 +08:00

core.cc

Ethernet support (#284)

2024-04-25 11:06:43 -07:00

cuda_utils.cc

ROCm support (#213)

2023-11-24 16:41:56 +08:00

debug.cc

Fix copyright messages (#367)

2024-10-17 21:25:46 -07:00

endpoint.cc

Fix copyright messages (#367)

2024-10-17 21:25:46 -07:00

errors.cc

Add executor to execute schedule-plan file (#283)

2024-04-18 19:10:41 +00:00

fifo.cc

Fix bootstrapping mechanism (#278)

2024-03-27 10:24:24 +08:00

ib.cc

Make ibverbs optional at compile time (#340)

2024-08-21 12:47:05 -07:00

numa.cc

ROCm support (#213)

2023-11-24 16:41:56 +08:00

nvls.cc

NVLS support for msccl++ executor (#375)

2024-11-20 06:43:28 +00:00

proxy_channel.cc

Update interface to let user change fifo size (#243)

2024-01-09 22:14:36 -08:00

proxy.cc

Update interface to let user change fifo size (#243)

2024-01-09 22:14:36 -08:00

registered_memory.cc

NVLS support for msccl++ executor (#375)

2024-11-20 06:43:28 +00:00

semaphore.cc

ROCm support (#213)

2023-11-24 16:41:56 +08:00

sm_channel.cc

Pytest (#162)

2023-09-01 21:22:11 +08:00

utils_internal.cc

Fix security issue (#305)

2024-05-25 23:12:57 -07:00

utils.cc

NVLS support for msccl++ executor (#375)

2024-11-20 06:43:28 +00:00