Feature/rccl install - #299
Open
ahskabir wants to merge 4 commits into
Open
Conversation
- Test suite that allows user to install rccl-tests and
RCCL library (optional) from rocm-systems repo
Signed-off-by: Ahsan Kabir <Ahsan.Kabir@amd.com>
- Sparse-clone RCCL and rccl-tests from rocm-systems. - Build rccl-tests with custom RCCL (CUSTOM_RCCL_LIB) and optional amdclang++. - Reuse prebuilt RCCL lib (under rccl_lib_install_dir) when available. Signed-off-by: Ahsan Kabir <Ahsan.Kabir@amd.com> Co-authored-by: sarachoi-amd <sarachoi@amd.com> Co-authored-by: Ahsan Kabir <Ahsan.Kabir@amd.com>
sarachoi-amd
force-pushed
the
feature/rccl_install
branch
from
August 13, 2026 21:59
9e51528 to
294d9e5
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Motivation
This feature will allow an user to install RCCL test and optionally build RCCL library as well. Both
rccllibrary andrccl-testswill be built and installed from the rocm-systems repository:Technical Details
The rocm-systems repository: https://github.com/rocm/rocm-systems. As for RCCL library is concerned, user has either of the two choices: Use library from within the rocm installation /opt/rocm/lib Or build it from rocm-systems
The option will be provided whether or not user wants to build RCCL library. Note that rccl-tests build requires installation of RCCL library plus installation of OpenMPI. The configuration file for this test suite will ask for: Path to OpenMPI installation
Test Plan
Installation is tested. Both RCCL library and rccl-tests were built. After built RCCL test was run and performance of all_reduce was validated.
Test Result
all_reduce perf result: The RCCL test completed successfully and achieved the expected bandwidth (Size: 16GB → BusBW: 360 GB/s.
Submission Checklist