make training and sweeps work on amd gpu hip/rocm#562
Open
sozforex wants to merge 2 commits into
Open
Conversation
Support AMD GPU sweep execution, keep early-stop diagnostics out of persisted metrics, and add the HIP/CUDA shim needed for ROCm builds.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Allows training and sweeps to run on amd gpu [hip/rocm],
cuda specific code is hipified during build and is placed in build/hip/src.
There has been a problem with sweep_obj being passed between processes when protein sweep is on gpu [apparently special synchronization handling is implemented for cuda, but not rocm in torch], so this PR contains a change related to that.
Tested only breakout sweep / eval on amd gpu.
Mostly done through codex gpt-5.5.
@jsuarez5341 please let me know if you care to merge amd gpu support in a shape similar to this draft PR