Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add GpuBuffer class #423

Open
wants to merge 22 commits into
base: main
Choose a base branch
from
Open

Add GpuBuffer class #423

wants to merge 22 commits into from

Conversation

chhwang
Copy link
Contributor

@chhwang chhwang commented Dec 19, 2024

  • Renamed and moved mem alloc functions into the mscclpp::detail:: namespace (now mscclpp::detail::gpuCalloc*<T>())
  • Deprecated constructor-calling mem alloc functions (mscclpp::makeShared*<T>() and mscclpp::makeUnique*<T>())
  • Added a new mscclpp::GpuBuffer<T>() class that should be used in general for allocating communication buffers
  • Added a new mscclpp.utils.GpuBuffer Python class that inherits cupy.ndarray and allocates using mscclpp::gpuMemAlloc
  • Renamed mscclpp::memcpyCuda*<T>() functions into mscclpp::gpuMemcpy*<T>() for name consistency
  • A few fixes in NVLS memory allocation
  • Tackled minor compiler warnings

@chhwang
Copy link
Contributor Author

chhwang commented Dec 19, 2024

/azp run

Copy link

Azure Pipelines could not run because the pipeline triggers exclude this branch/path.

@chhwang chhwang marked this pull request as draft December 19, 2024 23:49
Base automatically changed from chhwang/build to main December 20, 2024 00:51
@chhwang
Copy link
Contributor Author

chhwang commented Dec 20, 2024

/azp run

Copy link

Azure Pipelines could not run because the pipeline triggers exclude this branch/path.

@chhwang chhwang marked this pull request as ready for review December 20, 2024 00:53
@chhwang
Copy link
Contributor Author

chhwang commented Dec 20, 2024

/azp run

Copy link

Azure Pipelines successfully started running 3 pipeline(s).

@chhwang chhwang linked an issue Dec 20, 2024 that may be closed by this pull request
@chhwang
Copy link
Contributor Author

chhwang commented Dec 20, 2024

/azp run

Copy link

Azure Pipelines successfully started running 3 pipeline(s).

@chhwang
Copy link
Contributor Author

chhwang commented Jan 2, 2025

/azp run

Copy link

Azure Pipelines successfully started running 3 pipeline(s).

@chhwang
Copy link
Contributor Author

chhwang commented Jan 2, 2025

/azp run

Copy link

Azure Pipelines successfully started running 3 pipeline(s).

@chhwang
Copy link
Contributor Author

chhwang commented Jan 3, 2025

/azp run mscclpp-ut

Copy link

Azure Pipelines successfully started running 1 pipeline(s).

@chhwang
Copy link
Contributor Author

chhwang commented Jan 3, 2025

/azp run mscclpp-ut

Copy link

Azure Pipelines successfully started running 1 pipeline(s).

src/gpu_utils.cc Show resolved Hide resolved
test/mscclpp-test/allreduce_test.cu Show resolved Hide resolved
python/mscclpp_benchmark/mscclpp_op.py Show resolved Hide resolved
@chhwang
Copy link
Contributor Author

chhwang commented Jan 4, 2025

/azp run mscclpp-ut

Copy link

Azure Pipelines successfully started running 1 pipeline(s).

@chhwang
Copy link
Contributor Author

chhwang commented Jan 5, 2025

/azp run mscclpp-ut

Copy link

Azure Pipelines successfully started running 1 pipeline(s).

@chhwang
Copy link
Contributor Author

chhwang commented Jan 5, 2025

/azp run mscclpp-test

Copy link

Azure Pipelines successfully started running 1 pipeline(s).

@chhwang chhwang changed the title Add gpuMemAlloc function Add GpuBuffer class Jan 6, 2025
@chhwang chhwang requested a review from Binyang2014 January 6, 2025 01:26
@chhwang
Copy link
Contributor Author

chhwang commented Jan 6, 2025

/azp run mscclpp-ut

Copy link

Azure Pipelines successfully started running 1 pipeline(s).

@chhwang
Copy link
Contributor Author

chhwang commented Jan 6, 2025

/azp run mscclpp-ut

Copy link

Azure Pipelines successfully started running 1 pipeline(s).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[Bug] Proxy channel over CudaIPC on AMD GPUs
2 participants