docs: Physical GPU Passthrough for KubeVirt VMs solution (ACP-53455)

[Muyan0828]

Summary

Adds a KB Solution article: Physical GPU Passthrough for KubeVirt Virtual Machines on ACP.

It documents enabling physical GPU passthrough for KubeVirt VMs on ACP via the NVIDIA GPU Operator cluster plugin (vm-passthrough sandbox mode), replacing the legacy manual procedure from the user manual.

Sections:

• Overview & constraints (IOMMU/VT-d, vfio-pci, one GPU per VM)
• Prerequisites (ACP + KubeVirt + GPU node, IOMMU, uninstall host NVIDIA driver)
• Install the GPU Operator via Marketplace > Cluster Plugins
• Verify (ClusterPolicy ready, sandbox operands running)
• Configure KubeVirt permittedHostDevices
• Create a VM with a passthrough GPU
• Uninstall / cleanup

Notes

• Delivered as a kind: Solution KB article (English only; id auto-generated by build).
• Installation uses the cluster-plugin method; the plugin package bundles chart + images (incl. arm64), so no manual image list.

Jira: ACP-53455

🤖 Generated with Claude Code

Summary by CodeRabbit

• Documentation
  • Added a comprehensive guide for enabling physical GPU passthrough for KubeVirt Virtual Machines on Alauda Container Platform using NVIDIA GPU Operator, including prerequisites, IOMMU/host driver setup, GPU Operator installation and verification, KubeVirt configuration, and VM creation steps (plus removal/uninstall instructions).

[coderabbitai]

Warning

Review limit reached

@Muyan0828, we couldn't start this review because you've reached your PR review rate limit.

More reviews will be available in 41 minutes. Learn how PR review limits work.

Your organization has used up its prepaid credits, and credit purchases are no longer available. Enable the review add-on in the billing tab to keep reviews running — you're only billed for reviews past your plan's rate limits ($0.25/file).

⌛ How to resolve this issue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based credits.

🚦 How do rate limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please see our Fair Usage Limits Policy for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 29e63a74-9f75-494b-be88-4768c5b56778

📥 Commits

Reviewing files that changed from the base of the PR and between 13603c1 and fbf1275.

📒 Files selected for processing (1)

• docs/en/solutions/Physical_GPU_Passthrough_for_KubeVirt_Virtual_Machines_on_ACP.md

Walkthrough

A new documentation page describes how to configure Physical GPU passthrough for KubeVirt virtual machines on Alauda Container Platform using the NVIDIA GPU Operator in vm-passthrough sandbox mode.

Changes

Physical GPU Passthrough Documentation

Layer / File(s)	Summary
Overview and constraints docs/en/solutions/Physical_GPU_Passthrough_for_KubeVirt_Virtual_Machines_on_ACP.md	Introduces the page structure and documents the passthrough requirements and limits, including IOMMU/VT-d, single-GPU assignment, and the host NVIDIA driver constraint.
Prerequisites and operator setup docs/en/solutions/Physical_GPU_Passthrough_for_KubeVirt_Virtual_Machines_on_ACP.md	Shows how to enable IOMMU, remove the host NVIDIA driver, install the NVIDIA GPU Operator plugin in vm-passthrough mode, and verify passthrough-capable allocatable GPU resources.
KubeVirt passthrough configuration docs/en/solutions/Physical_GPU_Passthrough_for_KubeVirt_Virtual_Machines_on_ACP.md	Documents enabling disableMDevConfiguration, identifying pciDeviceSelector and resourceName, and patching kubevirt-hyperconverged to register passthrough GPUs in permittedHostDevices.
VM selection and cleanup docs/en/solutions/Physical_GPU_Passthrough_for_KubeVirt_Virtual_Machines_on_ACP.md	Explains setting the Physical GPU (Alpha) VM parameter, removing the GPU passthrough entry with a JSON patch, and uninstalling the GPU Operator plugin.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~5 minutes

Suggested reviewers

• fanzy618

Poem

🐇 Hop hop, a GPU tale is spun,
IOMMU wakes and the driver is undone.
A VM picks its shiny card with care,
Then back to the burrow, clean andేర్?

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly matches the new documentation page about physical GPU passthrough for KubeVirt VMs on ACP.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch acp-53455-gpu-passthrough-kb

---

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In
`@docs/en/solutions/Physical_GPU_Passthrough_for_KubeVirt_Virtual_Machines_on_ACP.md`:
- Line 40: Add language specifiers to four fenced code blocks in the Markdown
file to improve syntax highlighting. The GRUB configuration block containing
GRUB_CMDLINE_LINUX needs a bash specifier, the grub2-mkconfig command block
needs a bash specifier, the dmesg command output block needs a bash specifier,
and the JSON output example block containing nvidia.com/GK210GL_TESLA_K80 needs
a json specifier. For each of these code blocks, change the opening triple
backticks from ``` to ```bash or ```json as appropriate based on the content
type.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 239e88e0-68fd-42d0-8483-fe61caa44ea3

📥 Commits

Reviewing files that changed from the base of the PR and between e20db9e and 2efdc51.

📒 Files selected for processing (1)

• docs/en/solutions/Physical_GPU_Passthrough_for_KubeVirt_Virtual_Machines_on_ACP.md