Skip to content

Conversation

@Zhuul
Copy link
Owner

@Zhuul Zhuul commented Sep 2, 2025

This PR was automatically created because workflow files were updated while syncing with upstream.
Please review and merge.

Zhuul and others added 30 commits April 23, 2025 05:35
Add GitHub Actions workflow to sync with upstream repository.

* Create a new workflow file `.github/workflows/sync_with_upstream.yml`.
* Trigger the workflow on a daily schedule and on push events to the main branch.
* Add steps to fetch changes from the upstream repository.
* Add steps to merge upstream changes with the fork.
* Create a new branch if merge conflicts arise.
* Send notifications if manual intervention is required for conflict resolution.

---

For more details, open the [Copilot Workspace session](https://copilot-workspace.githubnext.com/Zhuul/vllm?shareId=XXXX-XXXX-XXXX-XXXX).
Add sync worker to detect changes and merge with fork
* **.github/workflows/sync_with_upstream.yml**
  - Add error handling for merge conflicts
  - Add logging for debugging and monitoring

* **.buildkite/scripts/run-multi-node-test.sh**
  - Add retry mechanism for failed Docker container starts
  - Add logging for debugging and monitoring
This reverts commit 8458f5e.
- Created TROUBLESHOOTING-WSL-GPU.md for comprehensive GPU troubleshooting steps in WSL2 with Podman.
- Added check-venv.sh to verify Python virtual environment setup within the container.
- Introduced check-wsl-gpu.sh for diagnosing WSL2 + GPU configuration issues.
- Implemented manage-container.sh for managing the vLLM development container lifecycle.
- Developed run-vllm-dev-fedora.ps1 and run-vllm-dev-fedora.sh for launching the vLLM development container with GPU support.
- Added setup-wsl-gpu.sh for installing NVIDIA Container Toolkit in WSL2.
Zhuul and others added 28 commits August 14, 2025 02:44
…ccache+compat LD paths; add LOCAL_MIRROR option; SHM/tmpfs; prune extras to minimal; add .dockerignore; Podman-first run scripts updated
…nd usability; add ccache support, local mirror option, and streamline dev setup process
…cript for better build logging and error handling
- Added new Podman-based scripts for running and managing vLLM containers.
- Deprecated old run-vllm-dev.ps1 and run-vllm-dev.sh scripts, redirecting to new Podman scripts.
- Implemented a comprehensive test script for vLLM container functionality.
- Created a patches directory with an apply_patches.sh script for managing patches.
- Added README files for better documentation across extras, patches, podman, secrets, storage, and testing directories.
- Introduced GPU status checking and diagnostics in the new Podman scripts.
- Established a secrets directory for local-only secret management.
- Developed storage helpers for managing external volumes for models and caches.
- Created a testing harness with a matrix for models/environments and scripts for running tests and comparing results.
…-cac904ae723a

Fix pre-commit linting errors in extras/ directory
…rt and dependency management

- Upgrade CUDA version to 13.0.0 and adjust base image flavor.
- Enhance build environment configuration with new arguments for better flexibility.
- Remove deprecated scripts and streamline setup process.
- Ensure compatibility with Python 3.12 and update PyTorch installation commands.
@Zhuul Zhuul force-pushed the main branch 2 times, most recently from 56fd993 to f51bf9a Compare September 27, 2025 01:43
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