Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Hardware][CPU] Support chunked-prefill and prefix-caching on CPU #10355

Merged
merged 13 commits into from
Nov 20, 2024

Conversation

bigPYJ1151
Copy link
Contributor

This PR provides CPU chunked-prefill and prefix-caching support. For now only FP32 and BF16 are supported, ipex-2.6 will provide the FP16 support.

PR Checklist (Click to Expand)

Thank you for your contribution to vLLM! Before submitting the pull request, please ensure the PR meets the following criteria. This helps vLLM maintain the code quality and improve the efficiency of the review process.

PR Title and Classification

Only specific types of PRs will be reviewed. The PR title is prefixed appropriately to indicate the type of change. Please use one of the following:

  • [Bugfix] for bug fixes.
  • [CI/Build] for build or continuous integration improvements.
  • [Doc] for documentation fixes and improvements.
  • [Model] for adding a new model or improving an existing model. Model name should appear in the title.
  • [Frontend] For changes on the vLLM frontend (e.g., OpenAI API server, LLM class, etc.)
  • [Kernel] for changes affecting CUDA kernels or other compute kernels.
  • [Core] for changes in the core vLLM logic (e.g., LLMEngine, AsyncLLMEngine, Scheduler, etc.)
  • [Hardware][Vendor] for hardware-specific changes. Vendor name should appear in the prefix (e.g., [Hardware][AMD]).
  • [Misc] for PRs that do not fit the above categories. Please use this sparingly.

Note: If the PR spans more than one category, please include all relevant prefixes.

Code Quality

The PR need to meet the following code quality standards:

  • We adhere to Google Python style guide and Google C++ style guide.
  • Pass all linter checks. Please use format.sh to format your code.
  • The code need to be well-documented to ensure future contributors can easily understand the code.
  • Include sufficient tests to ensure the project to stay correct and robust. This includes both unit tests and integration tests.
  • Please add documentation to docs/source/ if the PR modifies the user-facing behaviors of vLLM. It helps vLLM user understand and utilize the new features or changes.

Adding or changing kernels

Each custom kernel needs a schema and one or more implementations to be registered with PyTorch.

  • Make sure custom ops are registered following PyTorch guidelines: Custom C++ and CUDA Operators and The Custom Operators Manual
  • Custom operations that return Tensors require meta-functions. Meta-functions should be implemented and registered in python so that dynamic dims can be handled automatically. See above documents for a description of meta-functions.
  • Use torch.libary.opcheck() to test the function registration and meta-function for any registered ops. See tests/kernels for examples.
  • When changing the C++ signature of an existing op, the schema must be updated to reflect the changes.
  • If a new custom type is needed, see the following document: Custom Class Support in PT2.

Notes for Large Changes

Please keep the changes as concise as possible. For major architectural changes (>500 LOC excluding kernel/data/config/test), we would expect a GitHub issue (RFC) discussing the technical design and justification. Otherwise, we will tag it with rfc-required and might not go through the PR.

What to Expect for the Reviews

The goal of the vLLM team is to be a transparent reviewing machine. We would like to make the review process transparent and efficient and make sure no contributor feel confused or frustrated. However, the vLLM team is small, so we need to prioritize some PRs over others. Here is what you can expect from the review process:

  • After the PR is submitted, the PR will be assigned to a reviewer. Every reviewer will pick up the PRs based on their expertise and availability.
  • After the PR is assigned, the reviewer will provide status update every 2-3 days. If the PR is not reviewed within 7 days, please feel free to ping the reviewer or the vLLM team.
  • After the review, the reviewer will put an action-required label on the PR if there are changes required. The contributor should address the comments and ping the reviewer to re-review the PR.
  • Please respond to all comments within a reasonable time frame. If a comment isn't clear or you disagree with a suggestion, feel free to ask for clarification or discuss the suggestion.

Thank You

Finally, thank you for taking the time to read these guidelines and for your interest in contributing to vLLM. Your contributions make vLLM a great tool for everyone!

Copy link

👋 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.

🚀

@mergify mergify bot added documentation Improvements or additions to documentation ci/build labels Nov 15, 2024
@bigPYJ1151 bigPYJ1151 force-pushed the upstream_chunked_prefill branch from e9ddc0e to d33a175 Compare November 15, 2024 07:14
# reset distributed env properly. Use a value > 1 just when you test.
@pytest.mark.parametrize("tensor_parallel_size", [1])
@pytest.mark.parametrize("attention_backend", ["TORCH_SDPA"])
@pytest.mark.cpu_only
Copy link
Member

Choose a reason for hiding this comment

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

Can you define this mark in pyproject.toml?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Let's use the cpu_model defined before.

.buildkite/run-cpu-test.sh Show resolved Hide resolved
@DarkLight1337
Copy link
Member

I'll leave the detailed review to @Isotr0py and @WoosukKwon .

vllm/worker/cpu_model_runner.py Outdated Show resolved Hide resolved
vllm/worker/cpu_model_runner.py Outdated Show resolved Hide resolved
vllm/worker/cpu_model_runner.py Outdated Show resolved Hide resolved
Copy link

mergify bot commented Nov 17, 2024

This pull request has merge conflicts that must be resolved before it can be
merged. Please rebase the PR, @bigPYJ1151.

https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork

@mergify mergify bot added the needs-rebase label Nov 17, 2024
@bigPYJ1151 bigPYJ1151 force-pushed the upstream_chunked_prefill branch from 96cafe5 to a7534ec Compare November 18, 2024 09:28
@mergify mergify bot removed the needs-rebase label Nov 18, 2024
@bigPYJ1151 bigPYJ1151 force-pushed the upstream_chunked_prefill branch from a7534ec to f43c276 Compare November 18, 2024 09:30
@Isotr0py Isotr0py added the ready ONLY add when PR is ready to merge/full CI is needed label Nov 18, 2024
@Isotr0py
Copy link
Collaborator

Seems that the kernel tests are failing, perhaps we need merging from main branch?

@bigPYJ1151 bigPYJ1151 force-pushed the upstream_chunked_prefill branch from f43c276 to 3fb5054 Compare November 18, 2024 12:04
@bigPYJ1151
Copy link
Contributor Author

Hmm, seems the kernel test are broken before, they were also failed in some previous commits.

@Isotr0py
Copy link
Collaborator

cc @youkaichao Seems that VLLM config is not initialized in kernel tests.

@youkaichao
Copy link
Member

cc @youkaichao Seems that VLLM config is not initialized in kernel tests.

should be fixed by #10431 . please merge main to fix the test error.

@bigPYJ1151 bigPYJ1151 force-pushed the upstream_chunked_prefill branch 4 times, most recently from e219bab to eb1c775 Compare November 19, 2024 02:51
@DarkLight1337
Copy link
Member

Should be fixed now.

Copy link
Collaborator

@Isotr0py Isotr0py left a comment

Choose a reason for hiding this comment

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

Looks good to me now! Thanks for implementing this!

@bigPYJ1151 bigPYJ1151 force-pushed the upstream_chunked_prefill branch from eb1c775 to 883c6f5 Compare November 20, 2024 01:49
Copy link
Member

@mgoin mgoin left a comment

Choose a reason for hiding this comment

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

LGTM just some nits

tests/basic_correctness/test_chunked_prefill.py Outdated Show resolved Hide resolved
vllm/attention/backends/torch_sdpa.py Outdated Show resolved Hide resolved
vllm/worker/cpu_model_runner.py Outdated Show resolved Hide resolved
@DarkLight1337
Copy link
Member

Hmm, CPU tests failed.

@bigPYJ1151
Copy link
Contributor Author

@DarkLight1337 ops, a conflict with #10441...

@DarkLight1337 DarkLight1337 enabled auto-merge (squash) November 20, 2024 05:20
auto-merge was automatically disabled November 20, 2024 05:56

Head branch was pushed to by a user without write access

@DarkLight1337 DarkLight1337 enabled auto-merge (squash) November 20, 2024 06:03
@bigPYJ1151
Copy link
Contributor Author

@DarkLight1337 The failed tests are not related to this PR. Can we ignore these cases and merge this PR manually?

@DarkLight1337
Copy link
Member

I don't have permission to force merge. Can you just merge from main?

Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
Signed-off-by: jiang1.li <[email protected]>
auto-merge was automatically disabled November 20, 2024 09:00

Head branch was pushed to by a user without write access

@bigPYJ1151 bigPYJ1151 force-pushed the upstream_chunked_prefill branch from 504c82b to fda58f7 Compare November 20, 2024 09:00
@bigPYJ1151
Copy link
Contributor Author

@DarkLight1337 No problem, let's try again and pray for all cases green😂

@DarkLight1337 DarkLight1337 enabled auto-merge (squash) November 20, 2024 09:10
@DarkLight1337 DarkLight1337 merged commit 63f1fde into vllm-project:main Nov 20, 2024
54 checks passed
@bigPYJ1151 bigPYJ1151 deleted the upstream_chunked_prefill branch November 20, 2024 11:00
tlrmchlsmth pushed a commit to neuralmagic/vllm that referenced this pull request Nov 23, 2024
sleepwalker2017 pushed a commit to sleepwalker2017/vllm that referenced this pull request Dec 13, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ci/build documentation Improvements or additions to documentation ready ONLY add when PR is ready to merge/full CI is needed x86 CPU
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants