Skip to content

[None][chore] Update flashinfer-python from 0.6.11.post1 to 0.6.12rc1#14512

Open
yihwang-nv wants to merge 1 commit into
NVIDIA:mainfrom
yihwang-nv:yihwang-nv/update_flashinfer_0.6.12rc1
Open

[None][chore] Update flashinfer-python from 0.6.11.post1 to 0.6.12rc1#14512
yihwang-nv wants to merge 1 commit into
NVIDIA:mainfrom
yihwang-nv:yihwang-nv/update_flashinfer_0.6.12rc1

Conversation

@yihwang-nv
Copy link
Copy Markdown
Collaborator

@yihwang-nv yihwang-nv commented May 25, 2026

Summary

  • Bump flashinfer-python from 0.6.11.post1 to 0.6.12rc1
  • 0.6.12rc1 is not yet published to PyPI, so requirements.txt pins to the GitHub tag via git+https://github.com/flashinfer-ai/flashinfer.git@v0.6.12rc1
  • Updated version pins in requirements.txt, security_scanning/pyproject.toml, and ATTRIBUTIONS-Python.md
  • security_scanning/poetry.lock intentionally left untouched (RC has no PyPI hashes); maintainers regenerate it separately

Test plan

  • pip install -r requirements.txt installs successfully (requires building flashinfer from source)
  • pytest tests/unittest/_torch/flashinfer/ -v
  • pytest tests/unittest/_torch/attention/test_flashinfer_attention.py -v
  • CI pre-merge passes

Summary by CodeRabbit

  • Chores
    • Updated flashinfer-python dependency to v0.6.12rc1.

Review Change Stack

Bump flashinfer-python dependency to the v0.6.12rc1 release candidate.
The wheel is not yet published to PyPI, so requirements.txt pins to the
GitHub tag via a git+https URL. Updated version pins in requirements.txt,
security_scanning/pyproject.toml, and ATTRIBUTIONS-Python.md.

Signed-off-by: yihwang-nv <yihwang@nvidia.com>
@yihwang-nv yihwang-nv requested a review from a team as a code owner May 25, 2026 04:39
@yihwang-nv
Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast --add-multi-gpu-test

@coderabbitai
Copy link
Copy Markdown
Contributor

coderabbitai Bot commented May 25, 2026

📝 Walkthrough

Walkthrough

This PR updates the flashinfer-python dependency from version 0.6.11.post1 to 0.6.12rc1 across three files: production requirements switch to a git-sourced installation, the security scanning configuration updates the pinned version, and the attribution document reflects the new version.

Changes

Dependency Version Upgrade

Layer / File(s) Summary
Dependency version update
requirements.txt, security_scanning/pyproject.toml, ATTRIBUTIONS-Python.md
flashinfer-python is upgraded to 0.6.12rc1 in production requirements (switching from PyPI-pinned to git-sourced at tag v0.6.12rc1), security scanning configuration (exact version pin), and attribution documentation.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Possibly related PRs

  • NVIDIA/TensorRT-LLM#14076: Same dependency update pattern upgrading flashinfer-python across the same three files.
  • NVIDIA/TensorRT-LLM#13992: Related version bump of flashinfer-python (0.6.10 → 0.6.11) across requirements, pyproject.toml, and attribution files.
  • NVIDIA/TensorRT-LLM#13064: Coordinated flashinfer-python version pin update across the same set of configuration and documentation files.

Suggested reviewers

  • wenmingw
  • cascade812
  • juney-nvidia
🚥 Pre-merge checks | ✅ 5
✅ Passed checks (5 passed)
Check name Status Explanation
Title check ✅ Passed The title accurately summarizes the main change: updating flashinfer-python from 0.6.11.post1 to 0.6.12rc1, with proper format including ticket reference and type label.
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.
Description check ✅ Passed The PR description is comprehensive and addresses the template requirements effectively despite not using the exact template structure.

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

✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests

Comment @coderabbitai help to get the list of available commands and usage tips.

Copy link
Copy Markdown
Contributor

@coderabbitai coderabbitai Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 2

🤖 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 `@requirements.txt`:
- Line 57: Replace the mutable tag in the requirements entry for
flashinfer-python with the full commit SHA to ensure reproducible installs:
locate the line containing "flashinfer-python @
git+https://github.com/flashinfer-ai/flashinfer.git@v0.6.12rc1#egg=flashinfer-python"
and change the ref after the @ to the immutable commit
"529e592e9d8c5677b25c3cb38358bbd4def6c6dd" so it reads the same URL but with
"`@529e592e9d8c5677b25c3cb38358bbd4def6c6dd`" instead of "`@v0.6.12rc1`".

In `@security_scanning/pyproject.toml`:
- Line 57: The pyproject pin for the dependency "flashinfer-python
(==0.6.12rc1)" is out of sync with the lockfile which still records
0.6.11.post1; regenerate or update the lockfile so both declare the same
version: either change the pyproject constraint to match the lockfile or run the
dependency manager's lock/update command to produce a poetry.lock that contains
flashinfer-python==0.6.12rc1, then commit the updated lockfile to ensure
scanners read consistent metadata.
🪄 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: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: fb28b5ac-7611-4a92-b5d3-88c272aa7e47

📥 Commits

Reviewing files that changed from the base of the PR and between 90694fe and b248761.

📒 Files selected for processing (3)
  • ATTRIBUTIONS-Python.md
  • requirements.txt
  • security_scanning/pyproject.toml

Comment thread requirements.txt
patchelf
einops
flashinfer-python==0.6.11.post1
flashinfer-python @ git+https://github.com/flashinfer-ai/flashinfer.git@v0.6.12rc1#egg=flashinfer-python
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major | ⚡ Quick win

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
# Verify current tag target and whether requirements.txt uses an immutable SHA.
set -euo pipefail

echo "Resolving flashinfer tag:"
git ls-remote https://github.com/flashinfer-ai/flashinfer.git refs/tags/v0.6.12rc1

echo
echo "Current flashinfer requirement line:"
rg -n '^flashinfer-python\s+@\s+git\+https://github.com/flashinfer-ai/flashinfer.git@' requirements.txt

Repository: NVIDIA/TensorRT-LLM

Length of output: 297


Pin flashinfer-python to an immutable commit SHA

requirements.txt currently pins flashinfer-python to the mutable tag @v0.6.12rc1 (tag currently points to 529e592e9d8c5677b25c3cb38358bbd4def6c6dd). Pin the full commit SHA for reproducibility and supply-chain integrity.

Suggested update
-flashinfer-python @ git+https://github.com/flashinfer-ai/flashinfer.git@v0.6.12rc1#egg=flashinfer-python
+flashinfer-python @ git+https://github.com/flashinfer-ai/flashinfer.git@529e592e9d8c5677b25c3cb38358bbd4def6c6dd#egg=flashinfer-python
📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change
flashinfer-python @ git+https://github.com/flashinfer-ai/flashinfer.git@v0.6.12rc1#egg=flashinfer-python
flashinfer-python @ git+https://github.com/flashinfer-ai/flashinfer.git@529e592e9d8c5677b25c3cb38358bbd4def6c6dd#egg=flashinfer-python
🤖 Prompt for 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.

In `@requirements.txt` at line 57, Replace the mutable tag in the requirements
entry for flashinfer-python with the full commit SHA to ensure reproducible
installs: locate the line containing "flashinfer-python @
git+https://github.com/flashinfer-ai/flashinfer.git@v0.6.12rc1#egg=flashinfer-python"
and change the ref after the @ to the immutable commit
"529e592e9d8c5677b25c3cb38358bbd4def6c6dd" so it reads the same URL but with
"`@529e592e9d8c5677b25c3cb38358bbd4def6c6dd`" instead of "`@v0.6.12rc1`".

"patchelf (>=0.17.2.4,<0.18.0.0)",
"einops (>=0.8.2,<0.9.0)",
"flashinfer-python (==0.6.11.post1)",
"flashinfer-python (==0.6.12rc1)",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major | 🏗️ Heavy lift

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
# Verify pyproject/lock mismatch for flashinfer-python in security_scanning inputs.
set -euo pipefail

echo "pyproject dependency:"
rg -n 'flashinfer-python' security_scanning/pyproject.toml

echo
echo "poetry.lock package stanza:"
rg -n -A6 -B2 'name = "flashinfer-python"' security_scanning/poetry.lock

Repository: NVIDIA/TensorRT-LLM

Length of output: 402


Align flashinfer-python version between security_scanning/pyproject.toml and security_scanning/poetry.lock

security_scanning/pyproject.toml pins "flashinfer-python (==0.6.12rc1)", while security_scanning/poetry.lock still lists flashinfer-python as 0.6.11.post1. Update/regenerate security_scanning/poetry.lock (or ensure the security scanner uses the same dependency source) to prevent stale version metadata during scans.

🤖 Prompt for 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.

In `@security_scanning/pyproject.toml` at line 57, The pyproject pin for the
dependency "flashinfer-python (==0.6.12rc1)" is out of sync with the lockfile
which still records 0.6.11.post1; regenerate or update the lockfile so both
declare the same version: either change the pyproject constraint to match the
lockfile or run the dependency manager's lock/update command to produce a
poetry.lock that contains flashinfer-python==0.6.12rc1, then commit the updated
lockfile to ensure scanners read consistent metadata.

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

PR_Github #50142 [ run ] triggered by Bot. Commit: b248761 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

PR_Github #50142 [ run ] completed with state SUCCESS. Commit: b248761
/LLM/main/L0_MergeRequest_PR pipeline #39692 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

  • Please check the failed tests and fix your PR
  • If you cannot view the failures, ask the CI triggerer to share details
  • Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

@yihwang-nv
Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast --add-multi-gpu-test

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

PR_Github #50226 [ run ] triggered by Bot. Commit: b248761 Link to invocation

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.

2 participants