gahow/agentic-kvc
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add vLLM v0.18.1 source tree with KV transfer abort fix

Browse Source 
third_party/vllm/ now tracked in git for direct patch management.
Based on vLLM v0.18.1 release with one patch applied:

  vllm/v1/core/sched/scheduler.py:
    Replace fatal assert with graceful skip when KV transfer callback
    arrives for an already-aborted request during PD disaggregated serving.

Future vLLM modifications should be made directly in third_party/vllm/
and committed normally. The patches/ directory is kept as documentation
of what changed from upstream.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Gahow Wang
2026-05-22 00:30:38 +08:00
parent b6591950bc
commit 445e491123
4285 changed files with 1111303 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
third_party/vllm/.github/workflows/add_label_automerge.yml vendored Normal file
View File

@@ -0,0 +1,23 @@
name: Add label on auto-merge enabled
permissions:
pull-requests: write
on:
pull_request_target:
types:
- auto_merge_enabled
jobs:
add-label-on-auto-merge:
runs-on: ubuntu-latest
steps:
- name: Add label
uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
with:
script: |
github.rest.issues.addLabels({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: context.issue.number,
labels: ['ready']
})
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

32
third_party/vllm/.github/workflows/cleanup_pr_body.yml vendored Normal file
View File

@@ -0,0 +1,32 @@
name: Cleanup PR Body
on:
pull_request_target:
types: [opened, reopened, edited]
permissions:
pull-requests: write
jobs:
update-description:
runs-on: ubuntu-latest
steps:
- name: Checkout repository
uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # v6.0.1
- name: Set up Python
uses: actions/setup-python@83679a892e2d95755f2dac6acb0bfd1e9ac5d548 # v6.1.0
with:
python-version: '3.12'
cache: 'pip'
- name: Install Python dependencies
run: |
python3 -m pip install --upgrade pip
python3 -m pip install regex
- name: Update PR description
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: bash .github/scripts/cleanup_pr_body.sh "${{ github.event.number }}"

386
third_party/vllm/.github/workflows/issue_autolabel.yml vendored Normal file
View File

@@ -0,0 +1,386 @@
name: Label issues based on keywords
on:
issues:
types: [opened, edited, reopened]
permissions:
issues: write # needed so the workflow can add labels
contents: read
concurrency:
group: issue-labeler-${{ github.event.issue.number }}
cancel-in-progress: true
jobs:
add-labels:
runs-on: ubuntu-latest
steps:
- name: Label issues based on keywords
id: label-step
uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
with:
script: |
// Configuration: Add new labels and keywords here
const labelConfig = {
rocm: {
// Keyword search - matches whole words only (with word boundaries)
keywords: [
{
term: "composable kernel",
searchIn: "both"
},
{
term: "rccl",
searchIn: "body" // only search in body
},
{
term: "migraphx",
searchIn: "title" // only search in title
},
{
term: "hipgraph",
searchIn: "both"
},
{
term: "ROCm System Management Interface",
searchIn: "body"
},
],
// Substring search - matches anywhere in text (partial matches)
substrings: [
{
term: "VLLM_ROCM_",
searchIn: "both"
},
{
term: "aiter",
searchIn: "title"
},
{
term: "rocm",
searchIn: "title"
},
{
term: "amd",
searchIn: "title"
},
{
term: "hip-",
searchIn: "both"
},
{
term: "gfx",
searchIn: "both"
},
{
term: "cdna",
searchIn: "both"
},
{
term: "rdna",
searchIn: "both"
},
{
term: "torch_hip",
searchIn: "body" // only in body
},
{
term: "_hip",
searchIn: "both"
},
{
term: "hip_",
searchIn: "both"
},
// ROCm tools and libraries
{
term: "hipify",
searchIn: "both"
},
],
// Regex patterns - for complex pattern matching
regexPatterns: [
{
pattern: "\\bmi\\d{3}[a-z]*\\b",
description: "AMD GPU names (mi + 3 digits + optional letters)",
flags: "gi",
searchIn: "both" // "title", "body", or "both"
}
],
},
cpu: {
// Keyword search - matches whole words only (with word boundaries)
keywords: [
{
term: "CPU Backend",
searchIn: "title"
},
{
term: "x86",
searchIn: "title"
},
{
term: "ARM",
searchIn: "title"
},
{
term: "Apple Silicon",
searchIn: "title"
},
{
term: "IBM Z",
searchIn: "title"
},
],
},
// Add more label configurations here as needed
// example: {
// keywords: [...],
// substrings: [...],
// regexPatterns: [...]
// },
};
// Helper function to create regex based on search type
function createSearchRegex(term, type) {
// Escape special regex characters in the term
const escapedTerm = term.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
switch (type) {
case 'keyword':
// Word boundary search - matches whole words only
return new RegExp(`\\b${escapedTerm}\\b`, "gi");
case 'substring':
// Substring search - matches anywhere in the text
return new RegExp(escapedTerm, "gi");
default:
throw new Error(`Unknown search type: ${type}`);
}
}
// Helper function to find matching terms in text with line information
function findMatchingTermsWithLines(text, searchTerms = [], searchType = 'keyword', searchLocation = '') {
const matches = [];
const lines = text.split('\n');
for (const termConfig of searchTerms) {
let regex;
let term, searchIn, pattern, description, flags;
// Handle different input formats (string or object)
if (typeof termConfig === 'string') {
term = termConfig;
searchIn = 'both'; // default
} else {
term = termConfig.term;
searchIn = termConfig.searchIn || 'both';
pattern = termConfig.pattern;
description = termConfig.description;
flags = termConfig.flags;
}
// Skip if this term shouldn't be searched in the current location
if (searchIn !== 'both' && searchIn !== searchLocation) {
continue;
}
// Create appropriate regex
if (searchType === 'regex') {
regex = new RegExp(pattern, flags || "gi");
} else {
regex = createSearchRegex(term, searchType);
}
const termMatches = [];
// Check each line for matches
lines.forEach((line, lineIndex) => {
const lineMatches = line.match(regex);
if (lineMatches) {
lineMatches.forEach(match => {
termMatches.push({
match: match,
lineNumber: lineIndex + 1,
lineContent: line.trim(),
searchType: searchType,
searchLocation: searchLocation,
originalTerm: term || pattern,
description: description,
// Show context around the match in the line
context: line.length > 100 ?
line.substring(Math.max(0, line.toLowerCase().indexOf(match.toLowerCase()) - 30),
line.toLowerCase().indexOf(match.toLowerCase()) + match.length + 30) + '...'
: line.trim()
});
});
}
});
if (termMatches.length > 0) {
matches.push({
term: term || (description || pattern),
searchType: searchType,
searchLocation: searchLocation,
searchIn: searchIn,
pattern: pattern,
matches: termMatches,
count: termMatches.length
});
}
}
return matches;
}
// Helper function to check if label should be added
async function processLabel(labelName, config) {
const body = context.payload.issue.body || "";
const title = context.payload.issue.title || "";
core.notice(`Processing label: ${labelName}`);
core.notice(`Issue Title: "${title}"`);
core.notice(`Issue Body length: ${body.length} characters`);
let shouldAddLabel = false;
let allMatches = [];
let reason = '';
const keywords = config.keywords || [];
const substrings = config.substrings || [];
const regexPatterns = config.regexPatterns || [];
core.notice(`Searching with ${keywords.length} keywords, ${substrings.length} substrings, and ${regexPatterns.length} regex patterns`);
// Search in title
if (title.trim()) {
core.notice(`Searching in title: "${title}"`);
const titleKeywordMatches = findMatchingTermsWithLines(title, keywords, 'keyword', 'title');
const titleSubstringMatches = findMatchingTermsWithLines(title, substrings, 'substring', 'title');
const titleRegexMatches = findMatchingTermsWithLines(title, regexPatterns, 'regex', 'title');
allMatches.push(...titleKeywordMatches, ...titleSubstringMatches, ...titleRegexMatches);
}
// Search in body
if (body.trim()) {
core.notice(`Searching in body (${body.length} characters)`);
const bodyKeywordMatches = findMatchingTermsWithLines(body, keywords, 'keyword', 'body');
const bodySubstringMatches = findMatchingTermsWithLines(body, substrings, 'substring', 'body');
const bodyRegexMatches = findMatchingTermsWithLines(body, regexPatterns, 'regex', 'body');
allMatches.push(...bodyKeywordMatches, ...bodySubstringMatches, ...bodyRegexMatches);
}
if (allMatches.length > 0) {
core.notice(`Found ${allMatches.length} matching term(s):`);
for (const termMatch of allMatches) {
const locationText = termMatch.searchLocation === 'title' ? 'title' : 'body';
const searchInText = termMatch.searchIn === 'both' ? 'both' : termMatch.searchIn;
if (termMatch.searchType === 'regex') {
core.notice(` 📍 Regex: "${termMatch.term}" (pattern: ${termMatch.pattern}) found ${termMatch.count} time(s) in ${locationText} (configured to search in: ${searchInText}):`);
} else {
core.notice(` 📍 Term: "${termMatch.term}" (${termMatch.searchType} search) found ${termMatch.count} time(s) in ${locationText} (configured to search in: ${searchInText}):`);
}
// Show details for each match
termMatch.matches.forEach((match, index) => {
core.notice(` ${index + 1}. Line ${match.lineNumber} in ${match.searchLocation}: "${match.match}" [${match.searchType}]`);
if (match.description) {
core.notice(` Description: ${match.description}`);
}
core.notice(` Context: ${match.context}`);
if (match.lineContent !== match.context) {
core.notice(` Full line: ${match.lineContent}`);
}
});
}
shouldAddLabel = true;
const totalMatches = allMatches.reduce((sum, t) => sum + t.count, 0);
const titleMatches = allMatches.filter(t => t.searchLocation === 'title').reduce((sum, t) => sum + t.count, 0);
const bodyMatches = allMatches.filter(t => t.searchLocation === 'body').reduce((sum, t) => sum + t.count, 0);
const keywordMatches = allMatches.filter(t => t.searchType === 'keyword').reduce((sum, t) => sum + t.count, 0);
const substringMatches = allMatches.filter(t => t.searchType === 'substring').reduce((sum, t) => sum + t.count, 0);
const regexMatches = allMatches.filter(t => t.searchType === 'regex').reduce((sum, t) => sum + t.count, 0);
reason = `Found ${totalMatches} total matches (${titleMatches} in title, ${bodyMatches} in body) - ${keywordMatches} keyword matches, ${substringMatches} substring matches, ${regexMatches} regex matches`;
}
core.notice(`Final decision: ${shouldAddLabel ? 'ADD LABEL' : 'DO NOT ADD LABEL'}`);
core.notice(`Reason: ${reason || 'No matching terms found'}`);
if (shouldAddLabel) {
const existingLabels = context.payload.issue.labels.map(l => l.name);
if (!existingLabels.includes(labelName)) {
await github.rest.issues.addLabels({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: context.issue.number,
labels: [labelName],
});
core.notice(`Label "${labelName}" added. ${reason}`);
return true;
}
core.notice(`Label "${labelName}" already present.`);
return false;
}
core.notice(`No matching terms found for label "${labelName}".`);
return false;
}
// Process all configured labels
const labelsAddedResults = await Promise.all(
Object.entries(labelConfig).map(([labelName, config]) =>
processLabel(labelName, config).then(added => ({ labelName, added }))
)
);
const numLabelsAdded = labelsAddedResults.filter(r => r.added).length;
core.notice(`Processing complete. ${numLabelsAdded} label(s) added.`);
// Return which labels were added for the next step
const addedLabels = labelsAddedResults.filter(r => r.added).map(r => r.labelName);
core.setOutput('labels_added', JSON.stringify(addedLabels));
return addedLabels;
- name: CC users for labeled issues
if: steps.label-step.outputs.labels_added != '[]'
uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
with:
script: |
// Configuration: Map labels to GitHub users to CC
// You can add multiple users per label, and multiple label configurations
const ccConfig = {
rocm: {
users: ['hongxiayang', 'tjtanaa', 'vllmellm'], // Add more users as needed: ['user1', 'user2', 'user3']
message: 'CC {users} for ROCm-related issue' // {users} will be replaced with @mentions
},
// Add more label -> user mappings here
// Example:
// cuda: {
// users: ['user1', 'user2'],
// message: 'CC {users} for CUDA-related issue'
// },
// performance: {
// users: ['perfexpert'],
// message: 'CC {users} for performance issue'
// },
};
const labelsAdded = JSON.parse('${{ steps.label-step.outputs.labels_added }}');
core.notice(`Labels added: ${labelsAdded.join(', ')}`);
// Get existing comments to check for already mentioned users
const comments = await github.rest.issues.listComments({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: context.issue.number,
});
const issueBody = context.payload.issue.body || '';
const allExistingText = issueBody + '\n' + comments.data.map(c => c.body).join('\n');
// Process each label that was added
for (const label of labelsAdded) {
if (ccConfig[label]) {
const config = ccConfig[label];
const usersToMention = [];
// Check which users haven't been mentioned yet
for (const user of config.users) {
const mentionPattern = new RegExp(`@${user}\\b`, 'i');
if (!mentionPattern.test(allExistingText)) {
usersToMention.push(user);
} else {
core.notice(`@${user} already mentioned for label "${label}", skipping`);
}
}
// Post comment if there are users to mention
if (usersToMention.length > 0) {
const mentions = usersToMention.map(u => `@${u}`).join(' ');
const message = config.message.replace('{users}', mentions);
await github.rest.issues.createComment({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: context.issue.number,
body: message
});
core.notice(`CC comment added for label "${label}": ${mentions}`);
} else {
core.notice(`All users for label "${label}" already mentioned, skipping comment`);
}
}
}

84
third_party/vllm/.github/workflows/macos-smoke-test.yml vendored Normal file
View File

@@ -0,0 +1,84 @@
name: macOS Apple Silicon Smoke Test
on:
push:
branches:
- main
workflow_dispatch: # Manual trigger
permissions:
contents: read
jobs:
macos-m1-smoke-test:
runs-on: macos-latest
timeout-minutes: 30
steps:
- uses: actions/checkout@v6.0.1
- uses: astral-sh/setup-uv@v7
with:
enable-cache: true
cache-dependency-glob: |
requirements/**/*.txt
pyproject.toml
python-version: '3.12'
- name: Create virtual environment
run: |
uv venv
echo "$GITHUB_WORKSPACE/.venv/bin" >> "$GITHUB_PATH"
- name: Install dependencies and build vLLM
run: |
uv pip install -r requirements/cpu-build.txt --index-strategy unsafe-best-match
uv pip install -r requirements/cpu.txt --index-strategy unsafe-best-match
uv pip install -e . --no-build-isolation
env:
CMAKE_BUILD_PARALLEL_LEVEL: 4
- name: Verify installation
run: |
python -c "import vllm; print(f'vLLM version: {vllm.__version__}')"
- name: Smoke test vllm serve
run: |
# Start server in background
vllm serve Qwen/Qwen3-0.6B \
--max-model-len=2K \
--load-format=dummy \
--hf-overrides '{"num_hidden_layers": 2}' \
--enforce-eager \
--port 8000 &
SERVER_PID=$!
# Wait for server to start
for i in {1..30}; do
if curl -s http://localhost:8000/health > /dev/null; then
echo "Server started successfully"
break
fi
if [ "$i" -eq 30 ]; then
echo "Server failed to start"
kill "$SERVER_PID"
exit 1
fi
sleep 2
done
# Test health endpoint
curl -f http://localhost:8000/health
# Test completion
curl -f http://localhost:8000/v1/completions \
-H "Content-Type: application/json" \
-d '{
"model": "Qwen/Qwen3-0.6B",
"prompt": "Hello",
"max_tokens": 5
}'
# Cleanup
kill "$SERVER_PID"

17
third_party/vllm/.github/workflows/matchers/actionlint.json vendored Normal file
View File

@@ -0,0 +1,17 @@
{
"problemMatcher": [
{
"owner": "actionlint",
"pattern": [
{
"regexp": "^(?:\\x1b\\[\\d+m)?(.+?)(?:\\x1b\\[\\d+m)*:(?:\\x1b\\[\\d+m)*(\\d+)(?:\\x1b\\[\\d+m)*:(?:\\x1b\\[\\d+m)*(\\d+)(?:\\x1b\\[\\d+m)*: (?:\\x1b\\[\\d+m)*(.+?)(?:\\x1b\\[\\d+m)* \\[(.+?)\\]$",
"file": 1,
"line": 2,
"column": 3,
"message": 4,
"code": 5
}
]
}
]
}

17
third_party/vllm/.github/workflows/matchers/markdownlint.json vendored Normal file
View File

@@ -0,0 +1,17 @@
{
"problemMatcher": [
{
"owner": "markdownlint",
"pattern": [
{
"regexp": "^([^:]*):(\\d+):?(\\d+)?\\s([\\w-\\/]*)\\s(.*)$",
"file": 1,
"line": 2,
"column": 3,
"code": 4,
"message": 5
}
]
}
]
}

16
third_party/vllm/.github/workflows/matchers/mypy.json vendored Normal file
View File

@@ -0,0 +1,16 @@
{
"problemMatcher": [
{
"owner": "mypy",
"pattern": [
{
"regexp": "^(.+):(\\d+):\\s(error|warning):\\s(.+)$",
"file": 1,
"line": 2,
"severity": 3,
"message": 4
}
]
}
]
}

28
third_party/vllm/.github/workflows/pre-commit.yml vendored Normal file
View File

@@ -0,0 +1,28 @@
name: pre-commit
on:
pull_request:
push:
branches: [main]
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
cancel-in-progress: ${{ github.event_name == 'pull_request' }}
permissions:
contents: read
jobs:
pre-commit:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # v6.0.1
- uses: actions/setup-python@83679a892e2d95755f2dac6acb0bfd1e9ac5d548 # v6.1.0
with:
python-version: "3.12"
- run: echo "::add-matcher::.github/workflows/matchers/actionlint.json"
- run: echo "::add-matcher::.github/workflows/matchers/markdownlint.json"
- run: echo "::add-matcher::.github/workflows/matchers/mypy.json"
- uses: pre-commit/action@2c7b3805fd2a0fd8c1884dcaebf91fc102a13ecd # v3.0.1
with:
extra_args: --all-files --hook-stage manual

54
third_party/vllm/.github/workflows/reminder_comment.yml vendored Normal file
View File

@@ -0,0 +1,54 @@
name: PR Reminder Comment Bot
permissions:
pull-requests: write
on:
pull_request_target:
types: [opened]
jobs:
pr_reminder:
runs-on: ubuntu-latest
steps:
- name: Remind to run full CI on PR
uses: actions/github-script@ed597411d8f924073f98dfc5c65a23a2325f34cd # v8.0.0
with:
script: |
try {
// Get the PR author
const prAuthor = context.payload.pull_request.user.login;
// Check if this is the author's first PR in this repository
// Use GitHub's search API to find all PRs by this author
const { data: searchResults } = await github.rest.search.issuesAndPullRequests({
q: `repo:${context.repo.owner}/${context.repo.repo} type:pr author:${prAuthor}`,
per_page: 100
});
const authorPRCount = searchResults.total_count;
console.log(`Found ${authorPRCount} PRs by ${prAuthor}`);
// Only post comment if this is the first PR (only one PR by this author)
if (authorPRCount === 1) {
console.log(`Posting welcome comment for first-time contributor: ${prAuthor}`);
await github.rest.issues.createComment({
owner: context.repo.owner,
repo: context.repo.repo,
issue_number: context.issue.number,
body: '👋 Hi! Thank you for contributing to the vLLM project.\n\n' +
'💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.\n\n' +
'Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run `fastcheck` CI which starts running only a small and essential subset of CI tests to quickly catch errors. \n\n' +
'You ask your reviewers to trigger select CI tests on top of `fastcheck` CI. \n\n' +
'Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.\n\n' +
'To run CI, PR reviewers can either: Add `ready` label to the PR or enable auto-merge.\n\n' +
'If you have any questions, please reach out to us on Slack at https://slack.vllm.ai.\n\n' +
'🚀'
});
} else {
console.log(`Skipping comment for ${prAuthor} - not their first PR (${authorPRCount} PRs found)`);
}
} catch (error) {
console.error('Error checking PR history or posting comment:', error);
// Don't fail the workflow, just log the error
}
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

22
third_party/vllm/.github/workflows/scripts/build.sh vendored Normal file
View File

@@ -0,0 +1,22 @@
#!/bin/bash
set -eux
python_executable=python$1
cuda_home=/usr/local/cuda-$2
# Update paths
PATH=${cuda_home}/bin:$PATH
LD_LIBRARY_PATH=${cuda_home}/lib64:$LD_LIBRARY_PATH
# Install requirements
$python_executable -m pip install -r requirements/build.txt -r requirements/cuda.txt
# Limit the number of parallel jobs to avoid OOM
export MAX_JOBS=1
# Make sure release wheels are built for the following architectures
export TORCH_CUDA_ARCH_LIST="7.0 7.5 8.0 8.6 8.9 9.0+PTX"
bash tools/check_repo.sh
# Build
$python_executable setup.py bdist_wheel --dist-dir=dist

20
third_party/vllm/.github/workflows/scripts/create_release.js vendored Normal file
View File

@@ -0,0 +1,20 @@
// Uses GitHub's API to create the release and wait for result.
// We use a JS script since github CLI doesn't provide a way to wait for the release's creation and returns immediately.
module.exports = async (github, context, core) => {
try {
const response = await github.rest.repos.createRelease({
draft: false,
generate_release_notes: true,
name: process.env.RELEASE_TAG,
owner: context.repo.owner,
prerelease: true,
repo: context.repo.repo,
tag_name: process.env.RELEASE_TAG,
});
core.setOutput('upload_url', response.data.upload_url);
} catch (error) {
core.setFailed(error.message);
}
}

23
third_party/vllm/.github/workflows/scripts/cuda-install.sh vendored Normal file
View File

@@ -0,0 +1,23 @@
#!/bin/bash
# Replace '.' with '-' ex: 11.8 -> 11-8
cuda_version=$(echo "$1" | tr "." "-")
# Removes '-' and '.' ex: ubuntu-20.04 -> ubuntu2004
OS=$(echo "$2" | tr -d ".\-")
# Installs CUDA
wget -nv "https://developer.download.nvidia.com/compute/cuda/repos/${OS}/x86_64/cuda-keyring_1.1-1_all.deb"
sudo dpkg -i cuda-keyring_1.1-1_all.deb
rm cuda-keyring_1.1-1_all.deb
sudo apt -qq update
sudo apt -y install "cuda-${cuda_version}" "cuda-nvcc-${cuda_version}" "cuda-libraries-dev-${cuda_version}"
sudo apt clean
# Test nvcc
PATH=/usr/local/cuda-$1/bin:${PATH}
nvcc --version
# Log gcc, g++, c++ versions
gcc --version
g++ --version
c++ --version

56
third_party/vllm/.github/workflows/scripts/env.sh vendored Normal file
View File

@@ -0,0 +1,56 @@
#!/bin/bash
# This file installs common linux environment tools
export LANG C.UTF-8
# python_version=$1
sudo apt-get update && \
sudo apt-get install -y --no-install-recommends \
software-properties-common \
sudo apt-get install -y --no-install-recommends \
build-essential \
apt-utils \
ca-certificates \
wget \
git \
vim \
libssl-dev \
curl \
unzip \
unrar \
cmake \
net-tools \
sudo \
autotools-dev \
rsync \
jq \
openssh-server \
tmux \
screen \
htop \
pdsh \
openssh-client \
lshw \
dmidecode \
util-linux \
automake \
autoconf \
libtool \
net-tools \
pciutils \
libpci-dev \
libaio-dev \
libcap2 \
libtinfo5 \
fakeroot \
devscripts \
debhelper \
nfs-common
# Remove github bloat files to free up disk space
sudo rm -rf "/usr/local/share/boost"
sudo rm -rf "$AGENT_TOOLSDIRECTORY"
sudo rm -rf "/usr/share/dotnet"

15
third_party/vllm/.github/workflows/scripts/pytorch-install.sh vendored Normal file
View File

@@ -0,0 +1,15 @@
#!/bin/bash
python_executable=python$1
pytorch_version=$2
cuda_version=$3
# Install torch
$python_executable -m pip install numpy pyyaml scipy ipython mkl mkl-include ninja cython typing pandas typing-extensions dataclasses setuptools && conda clean -ya
$python_executable -m pip install torch=="${pytorch_version}+cu${cuda_version//./}" --extra-index-url "https://download.pytorch.org/whl/cu${cuda_version//./}"
# Print version information
$python_executable --version
$python_executable -c "import torch; print('PyTorch:', torch.__version__)"
$python_executable -c "import torch; print('CUDA:', torch.version.cuda)"
$python_executable -c "from torch.utils import cpp_extension; print (cpp_extension.CUDA_HOME)"

54
third_party/vllm/.github/workflows/stale.yml vendored Normal file
View File

@@ -0,0 +1,54 @@
name: 'Close inactive issues and PRs'
on:
schedule:
# Daily at 1:30 AM UTC
- cron: '30 1 * * *'
jobs:
close-issues-and-pull-requests:
# Prevents triggering on forks or other repos
if: github.repository == 'vllm-project/vllm'
permissions:
issues: write
pull-requests: write
actions: write
runs-on: ubuntu-latest
steps:
- uses: actions/stale@997185467fa4f803885201cee163a9f38240193d # v10.1.1
with:
# Increasing this value ensures that changes to this workflow
# propagate to all issues and PRs in days rather than months
operations-per-run: 1000
exempt-draft-pr: true
exempt-issue-labels: 'keep-open'
exempt-pr-labels: 'keep-open'
labels-to-add-when-unstale: 'unstale'
labels-to-remove-when-stale: 'unstale'
days-before-issue-stale: 90
days-before-issue-close: 30
stale-issue-label: 'stale'
stale-issue-message: >
This issue has been automatically marked as stale because it has not
had any activity within 90 days. It will be automatically closed if no
further activity occurs within 30 days. Leave a comment if
you feel this issue should remain open. Thank you!
close-issue-message: >
This issue has been automatically closed due to inactivity. Please
feel free to reopen if you feel it is still relevant. Thank you!
days-before-pr-stale: 90
days-before-pr-close: 30
stale-pr-label: 'stale'
stale-pr-message: >
This pull request has been automatically marked as stale because it
has not had any activity within 90 days. It will be automatically
closed if no further activity occurs within 30 days. Leave a comment
if you feel this pull request should remain open. Thank you!
close-pr-message: >
This pull request has been automatically closed due to inactivity.
Please feel free to reopen if you intend to continue working on it.
Thank you!