Skip to content

[Docs] update FAQ with logprobs MQ limits and deprecation#5368

Merged
Jiang-Jia-Jun merged 3 commits intoPaddlePaddle:developfrom
sunlei1024:doc/faq_logprobs
Dec 4, 2025
Merged

[Docs] update FAQ with logprobs MQ limits and deprecation#5368
Jiang-Jia-Jun merged 3 commits intoPaddlePaddle:developfrom
sunlei1024:doc/faq_logprobs

Conversation

@sunlei1024
Copy link
Collaborator

Motivation

Improve the FAQ documentation by adding details about message size limits when logprobs is enabled and clarifying that the current System V message-queue–based communication mechanism will be deprecated in future versions.

Modifications

  • Updated FAQ section describing inference stalls caused by System V Message Queue limits.
  • Added detailed message-size calculation examples for MTP and non-MTP modes.
  • Added guidance on adjusting kernel.msgmax and kernel.msgmnb.
  • Added a deprecation notice indicating that this communication method will be removed in future releases.

Usage or Command

N/A — documentation-only update.

Accuracy Tests

N/A — no changes to model behavior or inference logic.

Checklist

  • Add at least a tag in the PR title. Suggested: [Docs]
  • Format updated documentation.
  • No unit tests needed (documentation-only).
  • No accuracy results needed.
  • Not a cherry-pick PR.

@Jiang-Jia-Jun Jiang-Jia-Jun merged commit 3697110 into PaddlePaddle:develop Dec 4, 2025
6 of 8 checks passed
@codecov-commenter
Copy link

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (develop@5cd17fd). Learn more about missing BASE report.

Additional details and impacted files
@@            Coverage Diff             @@
##             develop    #5368   +/-   ##
==========================================
  Coverage           ?   59.71%           
==========================================
  Files              ?      325           
  Lines              ?    40286           
  Branches           ?     6100           
==========================================
  Hits               ?    24057           
  Misses             ?    14337           
  Partials           ?     1892           
Flag Coverage Δ
GPU 59.71% <ø> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@paddle-bot
Copy link

paddle-bot bot commented Dec 4, 2025

Thanks for your contribution!

liyonghua0910 pushed a commit to liyonghua0910/FastDeploy that referenced this pull request Dec 5, 2025
…le#5368)

* [doc] update FAQ with logprobs MQ limits and deprecation

* [doc] update FAQ with logprobs MQ limits and deprecation

* update faq
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants