[ROCm] xdist: select GPU from runner-allocated ROCR slice#793
Open
mminutoli wants to merge 1 commit into
Open
Conversation
CI runners isolate each job to a subset of the host's physical GPUs by injecting ROCR_VISIBLE_DEVICES (e.g. "0,3,4,5"); these are absolute physical indices and may be non-contiguous. Re-setting ROCR_VISIBLE_DEVICES does not compose across processes, so the previous `xdist_worker_number % num_rocm_devices` selection could land a worker on a GPU owned by another job running on the same host. Pick the device from the inherited ROCR_VISIBLE_DEVICES slice instead, and fall back to the old behaviour when no slice is injected (local dev).
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
CI runners isolate each job to a subset of the host's physical GPUs by injecting ROCR_VISIBLE_DEVICES (e.g. "0,3,4,5"); these are absolute physical indices and may be non-contiguous. Re-setting ROCR_VISIBLE_DEVICES does not compose across processes, so the previous
xdist_worker_number % num_rocm_devicesselection could land a worker on a GPU owned by another job running on the same host. Pick the device from the inherited ROCR_VISIBLE_DEVICES slice instead, and fall back to the old behaviour when no slice is injected (local dev).