[Core] Change multiproc method default to spawn - #8576

👋 Hi! Thank you for contributing to the vLLM project.
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. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can do one of these:

• Add ready label to the PR
• Enable auto-merge.

🚀

for reference since i was just looking around ... some other places I've seen advise against fork breaking things (multithreaded things in particular)

• https://docs.habana.ai/en/latest/PyTorch/Getting_Started_with_PyTorch_and_Gaudi/Getting_Started_with_PyTorch.html?highlight=multiprocessing#to[…]ers -- "If training scripts use multiprocessing with multiple workers for PyTorch dataloader, change the start method to spawn or forkserver ..."
• Use 'spawn' multiprocessing method instead of fork instructlab/instructlab#957 -- "The fork method is not safe in multi-threaded environments. It is also incompatible with some hardware accelerators."
• https://pytorch.org/docs/stable/multiprocessing.html#sharing-cuda-tensors -- "Sharing CUDA tensors between processes is supported only in Python 3, using a spawn or forkserver start methods."
• https://pytorch.org/docs/stable/notes/multiprocessing.html -- "The CUDA runtime does not support the fork start method; either the spawn or forkserver start method are required to use CUDA in subprocesses."

In some private chat, it was pointed out that this will break user code using the LLM class if their code is not protected by if name == “main” so this isn’t a good idea for now at least