docs(core-concepts): clarify which cu/nvml symbols libvgpu.so intercepts

[ipsitapp8]

What type of PR is this?

/kind documentation

What this PR does / why we need it:

The GPU virtualization page says HAMi-Core hijacks NVIDIA library functions "starting with cu/nvml", and that it intercepts "all function calls starting with cu and nvml". That is broader than what libvgpu.so does.

dlsym is overridden in src/libvgpu.c, but what comes back is decided by a lookup, not by the prefix:

• nvml* goes to __dlsym_hook_section_nvml(), which is an explicit list of DLSYM_HOOK_FUNC(...) entries. A symbol that is not in the list returns NULL and falls through to the real dlsym.

• cu* is resolved against libvgpu.so itself first. If HAMi-Core defines no wrapper for it, it falls through the same way. The cuGetProcAddress path goes through __dlsym_hook_section(), which is the same kind of list.

• One family is skipped outright, in find_symbols_in_table() in src/cuda/hook.c:

  /* Skip CUDA graph functions: let them fall through to real driver */
  if (strncmp(symbol, "cuGraph", 7) == 0) {
      return NULL;
  }

This matters when the page is used to debug an isolation gap. "Every cu/nvml call is intercepted" and "the calls in the hook table are intercepted" point at different causes for an allocation that got past the limit.

The change touches the HAMi-Core row in the component table and the dlsym paragraph in the interception flow section, in English and Chinese. Text only, no links or page structure changed.

AI assistance: I used Claude Code to check the existing wording against the HAMi-core source.

Which issue(s) this PR fixes:

None. 4 changed lines, below the 100 line threshold that requires an issue first.

Checklist:

• npm run lint and npm run format:check pass
• npm run build succeeds for both en and zh
• Chinese translation updated if English docs changed (or noted why not)
• Commits are signed off (git commit -s)

Note on the first box: markdownlint is clean across the repo and Prettier is clean on both changed files. Running npm run format:check on a Windows checkout reports every file in the repo, including untouched ones on master, because core.autocrlf gives CRLF line endings locally. That is a local artifact, not something in this diff.

Summary by CodeRabbit

• Documentation
  • Clarified that only explicitly registered NVIDIA functions are intercepted by the virtualization library.
  • Documented that unregistered functions resolve normally to the underlying driver.
  • Updated both English and Chinese GPU virtualization documentation.

[netlify]

✅ Deploy Preview for project-hami ready!

Name	Link
🔨 Latest commit	b3db299
🔍 Latest deploy log	https://app.netlify.com/projects/project-hami/deploys/6a7ac59e0aec2400088fd02d
😎 Deploy Preview	https://deploy-preview-754--project-hami.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.
🤖 Make changes	Run an agent on this branch

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: d7e5ba57-5176-4111-9996-6c6987ecdebe

📥 Commits

Reviewing files that changed from the base of the PR and between fe70b57 and b3db299.

📒 Files selected for processing (2)

• docs/core-concepts/gpu-virtualization.md
• i18n/zh/docusaurus-plugin-content-docs/current/core-concepts/gpu-virtualization.md

---

📝 Walkthrough

Walkthrough

The English and Chinese GPU virtualization documentation now states that libvgpu.so intercepts only cu and nvml functions listed in its hook table. Unlisted functions resolve normally to the real driver.

Changes

GPU virtualization documentation

Layer / File(s)	Summary
Hook-table interception semantics docs/core-concepts/gpu-virtualization.md, i18n/zh/docusaurus-plugin-content-docs/current/core-concepts/gpu-virtualization.md	The documentation describes selective interception of listed NVIDIA functions and normal resolution of unlisted functions.

Estimated code review effort: 1 (Trivial) | ~3 minutes

Suggested reviewers: mesutoezdil

🚥 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 and concisely describes the documentation change to clarify which cu/nvml symbols libvgpu.so intercepts.
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.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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.}

[rootsongjc]

/lgtm

[hami-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: ipsitapp8, rootsongjc
Once this PR has been reviewed and has the lgtm label, please assign windsonsea for approval. For more information see the Kubernetes Code Review Process.

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment