Repository navigation
fix(plugin): serve whole GPUs when no HAMi-tracked pod is pending - #161
Merged
Merged
Conversation
Allocate returned "no binding pod found" for a Pod that carries none of HAMi's bind-time annotations (placed by another scheduler or alongside a second device plugin), so the whole-GPU fallback from the kubelet device ids was never reached and the Pod failed with UnexpectedAdmissionError. Treat ErrNoPendingPod as that case: use an empty Pod, serve the kubelet ids as whole GPUs, and skip the node-lock release and dmem cap that need a real Pod. Other lookup errors still fail Allocate. Signed-off-by: mesutoezdil <mesudozdil@gmail.com>
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: moezdil The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars. ⚙️ Run configuration
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
/kind bug
Allocate returned "no binding pod found" for a Pod that carries none of HAMi's bind-time annotations (placed by another scheduler or next to a second device plugin), so the whole-GPU fallback from #18 was never reached and the Pod failed with UnexpectedAdmissionError. This treats that case as a whole-GPU allocation from the kubelet device ids and skips the node-lock release and dmem cap that need a real Pod; other pending-pod lookup errors still fail Allocate.
Refs #18
Reproduced and verified on an MI325X node: a Pod with
hami.io/webhook: ignoreandschedulerName: default-schedulerfailed admission before and now runs with HSA_CU_MASK, ROCR_VISIBLE_DEVICES and HIP_VISIBLE_DEVICES set.AI assistance was used for this PR.