feat(gpu): add sriov-report mode to control PF/VF advertisement - #3962
Merged
Conversation
Allow advertising only the PF while leaving host VFs enabled, so pods can keep a stable renderD node across recreates. Also skip node-register PCI ID failures without dropping the GPU from kubelet.
|
The latest updates on your projects. Learn more about Vercel for GitHub. 1 Skipped Deployment
|
eball
approved these changes
Aug 17, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Background
Allow advertising only the PF while leaving host VFs enabled, so pods can keep a stable renderD node across recreates. Also skip node-register PCI ID failures without dropping the GPU from kubelet.
Target Version for Merge
v1.12.7
Related Issues
None
PRs Involving Sub-Systems
feat(gpu): add sriov-report mode to control PF/VF advertisement intel-device-plugins-for-kubernetes#3
Other information:
Note
Medium Risk
Changes kubelet GPU device advertisement on Intel GPU nodes during upgrade; misconfiguration could affect GPU scheduling until the DaemonSet rolls out successfully.
Overview
Rolls out an updated Intel GPU device plugin for v1.12.7 clusters so SR-IOV-capable nodes advertise only the physical function (PF) to the scheduler while host VFs can stay enabled—aimed at keeping a stable renderD node across pod recreates. The bundled manifest now passes
-sriov-report=pfand bumpsbeclab/intel-device-pluginfrom 1.1.1 to 1.1.2 (behavior aligned with the forked intel-device-plugins change).Upgrade path adds an
ApplyIntelGPUPluginlocal task (kubectl apply ofwizard/config/gpu/intel/gpu-plugin.yaml, retried like other apply tasks) and wires it into a new daily upgrader1.12.7-20260817viaupgradeIntelGPUPlugin()ahead of the standard prepare steps.Reviewed by Cursor Bugbot for commit de576aa. Bugbot is set up for automated code reviews on this repo. Configure here.