microsoft/mscclpp
1
0
Fork 0
mirror of https://github.com/microsoft/mscclpp.git synced 2026-05-25 15:24:43 +00:00
Code Issues Packages Projects Releases Wiki Activity
669 Commits 110 Branches 16 Tags
2127a3ba298b9947498fcebfa96a673fca287d99
Commit Graph

386 Commits

Author SHA1 Message Date
Changho Hwang
2127a3ba29 Improve CMake options (#376) 
* Let all CMake option names start with `MSCCLPP_`
* Explain the `MSCCLPP_BUILD_PYTHON_BINDINGS` option in readme

---------

Co-authored-by: Binyang Li <binyli@microsoft.com>
 2024-11-22 01:54:11 +00:00
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
Ziyue Yang
3e51e9b359 Fix missing packet parameter for executor (#385) 2024-11-19 08:36:37 +08:00
Binyang Li
1baea89fa0 Fix light load bug (#379) 
Fix lightLoadExecutionPlan issue.
An execution context many have multi device execution plans. These plans
share the channel connections which are constructed before.
A deviceExecutionPlanKey is introduced to identify these plans. We can
get the current device execution plan key via:
`contexts.currentDevicePlan`
 2024-11-13 07:58:43 +00:00
Caio Rocha
d5d608abdc Fixing Bug Const Offset in Execution Plan (#380) 
The offset was not differentiating between the buffer types, causing the
offset to be incorrect when the buffer type was not `SCRATCH`.

---------

Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2024-11-11 20:02:02 -08:00
Changho Hwang
85fdde7a73 Lazily create the context stream (#381) 
Create the context stream only when needed.
 2024-11-11 10:39:32 +08:00
Caio Rocha
c6e06cfad7 Executor AllGather In-Place Support (#365) 2024-10-21 05:45:56 -07:00
Changho Hwang
0c150e5166 Fix copyright messages (#367) 2024-10-17 21:25:46 -07:00
Caio Rocha
08a0cec2eb Fixing RegisterMemory Allocation for ProxyChannels (#353) 
Co-authored-by: Binyang Li <binyli@microsoft.com>
Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2024-09-24 23:01:41 -07:00
Ziyue Yang
5c4e105814 Fix NPKit exit event offset (#356) 2024-09-19 13:35:44 +08:00
Binyang Li
b30bb260e3 Tune threads per block for mscclpp executor (#345) 2024-09-18 17:21:47 -07:00
Binyang Li
7bedb25054 Add proxy channel related operations (#351) 
Add Flush, PutWithSignal, PutWithFlushAndSignal operation
 2024-09-15 13:24:57 -07:00
Binyang Li
26a87535f9 Fix bug for construct sempaphore (#341) 
Current semaphore construction requires two-way communication, e.g., to
construct a semaphore signaling from rank 0 to rank 1, both rank 0 and
rank 1 need to send a message to each other. This PR fixes an executor
bug that fails to conduct two-way communication for constructing such
one-way semaphores, and instead hangs during the semaphore construction.
In the future, we may need to change the implementation to construct
semaphore via one-way communication.

---------

Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2024-09-04 19:42:03 +08:00
Changho Hwang
72b99a4229 Fix for ROCm 6.0 (#347) 2024-09-01 20:22:33 -07:00
Caio Rocha
4eca6f1e95 Support executors to send packets over ProxyChannel (#344) 
Co-authored-by: Binyang Li <binyli@microsoft.com>
 2024-08-30 22:10:33 +00:00
Caio Rocha
1af62ea43d ProxyChannel Support in Executor (#342) 
Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2024-08-27 10:09:44 -07:00
Changho Hwang
1e82dd444f Make ibverbs optional at compile time (#340) 
Co-authored-by: Caio Rocha <caiorocha@microsoft.com>
 2024-08-21 12:47:05 -07:00
Caio Rocha
ead4efc315 Dynamically load libibverbs (#337) 2024-08-13 23:48:39 -07:00
Changho Hwang
8c6fb429e9 bfloat16 support (#336) 
* Add bfloat16 support for executor and NCCL interface
* Changed `gpu_data_types.hpp` into an internal header file
 2024-08-12 15:41:58 -07:00
caiomcbr
67eb9b04cc NCCL API Executor Integration (#331) 
Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2024-07-25 15:05:02 -07:00
Roshan Dathathri
f131fae3ec Add support for different vector sizes in multimem instructions (#332) 2024-07-25 10:14:02 -07:00
Ziyue Yang
b5a48f836c Separate NPKit CPU timestamp access from different blocks for AMD platform (#321) 
Reference: https://github.com/ROCm/rccl/pull/1229
 2024-07-02 19:36:48 +08:00
Ziyue Yang
f29095b3b1 Fix NPKit support for AMD (#312) 2024-06-14 16:22:14 +08:00
Ziyue Yang
76328fe623 Add NPKit GPU event support (#310) 2024-06-13 13:59:50 +08:00
Binyang Li
80aefe55bc Cumulative Updates (#309) 
Bug fix: Unable to execute communication primitives with the same
execution plan but varying message sizes.
Add reduce_packets OP
 2024-06-12 19:17:57 +08:00
Changho Hwang
1f62dfd7cd Add C++ executor test (#304) 
- Add C++ executor test
- Fix executor bugs for packet operation
- Enhance executor_test.py

---------

Co-authored-by: Binyang Li <binyli@microsoft.com>
 2024-05-29 10:54:36 +00:00
Binyang Li
3a18068cd4 Fix security issue (#305) 
Change sprintf to snprintf to avoid potential security issue
 2024-05-25 23:12:57 -07:00
Binyang Li
6226556ce2 Optimized the execution kernel (#294) 2024-05-03 11:54:50 -07:00
Binyang Li
5628362715 Resolve multi-nodes test failure issue (#295) 
Fix bug, resolve multi-nodes test failure issue.
 2024-04-26 13:06:57 +08:00
Changho Hwang
d4ede480f4 Ethernet support (#284) 
Co-authored-by: Binyang Li <binyli@microsoft.com>
Co-authored-by: Caio Rocha <caiorocha@microsoft.com>
 2024-04-25 11:06:43 -07:00
Changho Hwang
89896ff94f Include GPU data types only for kernel code (#292) 2024-04-24 20:55:02 -07:00
Changho Hwang
6c1fa5307c Refactoring NVLS interfaces (#293) 
Move NVLS details from the core to a separate interface
 2024-04-24 10:05:41 -07:00
Changho Hwang
9934c982a8 Seperate headers for GPU data types (#291) 
Prevent unnecessarily including data type headers in everywhere.
 2024-04-19 05:52:43 +00:00
Roshan Dathathri
41e0964d93 Allow binding allocated memory to NVLS multicast pointer (#290) 
And change NVLS multimem instructions to static functions
 2024-04-18 17:11:31 -07:00
Binyang Li
64d837f9ab Add executor to execute schedule-plan file (#283) 
Add executor to execute the JSON schedule file generated by msccl-tools

---------

Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2024-04-18 19:10:41 +00:00
Changho Hwang
9406123711 Fix a typo name (#286) 2024-04-17 23:45:46 +00:00
Changho Hwang
5ba6ce00c7 Fix bootstrapping mechanism (#278) 
Co-authored-by: Binyang Li <binyli@microsoft.com>
Co-authored-by: Pashupati Kumar <74680231+pash-msft@users.noreply.github.com>
 2024-03-27 10:24:24 +08:00
Changho Hwang
d34e097b40 Fix wrong offset calculation (#257) 2024-02-06 08:55:43 +08:00
Saeed Maleki
91d592dcc0 NVLS support. (#250) 
Co-authored-by: Saeed Maleki <saemal@microsoft.com>
Co-authored-by: Binyang Li <binyli@microsoft.com>
Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2024-02-04 20:46:10 -08:00
Binyang Li
163cba08c8 Update interface to let user change fifo size (#243) 
Related with this issue:
https://github.com/microsoft/mscclpp/issues/242. The user may use more
threads than the number specified in `fifo_size` to interact with the
FIFO. In this case, there will be unexpected behavior.
Update the interface to let user change fifo size on their demands.
 2024-01-09 22:14:36 -08:00
Binyang Li
e7d3e2d44b Fix crash in static variable deconstructor (#238) 
According to https://en.cppreference.com/w/cpp/utility/program/exit, `The last destructor for thread-local objects is [sequenced-before](https://en.cppreference.com/w/cpp/language/eval_order) the first destructor for a static object.`
Change the code to avoid this case.

---------

Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2023-12-25 14:01:28 +00:00
Changho Hwang
5fa5bd2706 Check nvidia_peermem during runtime (#234) 2023-12-25 12:02:10 +08:00
Changho Hwang
544ff0c21d ROCm support (#213) 
Co-authored-by: Binyang Li <binyli@microsoft.com>
 2023-11-24 16:41:56 +08:00
Changho Hwang
e710701728 Warning ahead of CQ being full (#202) 2023-11-15 08:03:29 +00:00
Changho Hwang
7686e15fbd Allow infinite waiting (#200) 2023-10-23 12:28:05 +08:00
Saeed Maleki
85e8017535 Atomic for semaphores instead of fences (#188) 
Co-authored-by: Pratyush Patel <pratyushpatel.1995@gmail.com>
Co-authored-by: Esha Choukse <eschouks@microsoft.com>
Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2023-10-13 18:57:08 +08:00
Saeed Maleki
c4785c9591 Improve debugging messages (#195) 
Debugging information to understand what connections are being made.

---------

Co-authored-by: Changho Hwang <changhohwang@microsoft.com>
 2023-10-13 16:55:52 +08:00
Saeed Maleki
148681b4bc Fix a pytest bug (#196) 2023-10-13 16:39:43 +08:00
Changho Hwang
8c0f9e84d0 v0.3.0 (#171) 2023-10-11 22:35:54 +08:00
Changho Hwang
6c0ee72916 Construct ProxyChannel with shared pointers (#184) 2023-09-18 05:46:23 +00:00