Skip to content

[FEAT]Support reset prefix cache by specified device #15003

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.

Sign up for GitHub

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

Jump to bottom
Merged
merged 3 commits into from
Mar 19, 2025
Merged

[FEAT]Support reset prefix cache by specified device #15003

merged 3 commits into from
Mar 19, 2025

Conversation

maobaolong
Copy link
Contributor

@maobaolong maobaolong commented Mar 18, 2025 •
edited by github-actions bot
Loading

This PR aimed to supply a way to reset specified Device prefix cache.

The existing device is CPU or GPU

@github-actions GitHub Actions
Copy link

👋 Hi! Thank you for contributing to the vLLM project.

💬 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.

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 either: Add ready label to the PR or enable auto-merge.

🚀

@mergify mergify bot added the frontend label Mar 18, 2025
comaniac
comaniac reviewed Mar 18, 2025
View reviewed changes
@mergify mergify bot added the v1 label Mar 18, 2025
comaniac
comaniac reviewed Mar 18, 2025
View reviewed changes
@maobaolong maobaolong force-pushed the resetPrefixCacheByDevice branch from 17766d8 to 1769998 Compare March 19, 2025 01:02
@maobaolong maobaolong force-pushed the resetPrefixCacheByDevice branch from d570519 to faf1586 Compare March 19, 2025 03:54
@maobaolong maobaolong force-pushed the resetPrefixCacheByDevice branch from 27a8599 to 1f590cf Compare March 19, 2025 12:11
@comaniac comaniac added the ready ONLY add when PR is ready to merge/full CI is needed label Mar 19, 2025
@maobaolong
Copy link
Contributor Author

@comaniac Thanks for your review and help, now almost all the ci checks are passed except some of them, but it seems have no relevant with this PR?

@maobaolong maobaolong requested a review from comaniac March 19, 2025 17:09
@DarkLight1337
Copy link
Member

@comaniac can you approve the PR?

@comaniac
Copy link
Collaborator

@comaniac can you approve the PR?

Oops sorry I missed that

@vllm-bot vllm-bot merged commit 26dd972 into vllm-project:main Mar 19, 2025
44 of 47 checks passed
njhill
njhill reviewed Mar 19, 2025
View reviewed changes
vllm/entrypoints/openai/api_server.py
device_str = raw_request.query_params.get("device")
if device_str is not None:
device = Device[device_str.upper()]
logger.info("Resetting prefix cache with specific %s...", str(device))
Copy link
Member

Choose a reason for hiding this comment

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

logger.info("Resetting prefix cache with specific device: %s...", str(device))

vllm/v1/engine/llm_engine.py

def reset_prefix_cache(self):
def reset_prefix_cache(self, device: Optional[Device] = None):
self.engine_core.reset_prefix_cache()
Copy link
Member

Choose a reason for hiding this comment

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

It looks like this isn't implemented for V1... should that be included? (we are aiming to deprecate v0 now)

Copy link
Contributor Author

Choose a reason for hiding this comment

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

I will implement this for v1 after #13377 merged.

@maobaolong
Copy link
Contributor Author

@comaniac @DarkLight1337 @njhill Thanks for your review!

lulmer pushed a commit to lulmer/vllm that referenced this pull request Apr 7, 2025
@maobaolong @lulmer
[FEAT]Support reset prefix cache by specified device (vllm-project#15003
c841d46 
)

Signed-off-by: Louis Ulmer <[email protected]>
@ckhordiasma ckhordiasma mentioned this pull request Apr 17, 2025
Closed
shreyankg pushed a commit to shreyankg/vllm that referenced this pull request May 3, 2025
@maobaolong @shreyankg
[FEAT]Support reset prefix cache by specified device (vllm-project#15003
cec8f20 
)
RichardoMrMu pushed a commit to RichardoMrMu/vllm that referenced this pull request May 12, 2025
@maobaolong
[FEAT]Support reset prefix cache by specified device (vllm-project#15003
444bd01 
)

Signed-off-by: Mu Huai <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@njhill njhill njhill left review comments

@comaniac comaniac comaniac approved these changes

@zhuohan123 zhuohan123 Awaiting requested review from zhuohan123

@youkaichao youkaichao Awaiting requested review from youkaichao

@alexm-redhat alexm-redhat Awaiting requested review from alexm-redhat alexm-redhat is a code owner

@WoosukKwon WoosukKwon Awaiting requested review from WoosukKwon WoosukKwon is a code owner

@robertgshaw2-redhat robertgshaw2-redhat Awaiting requested review from robertgshaw2-redhat robertgshaw2-redhat is a code owner

@ywang96 ywang96 Awaiting requested review from ywang96 ywang96 is a code owner

Assignees

No one assigned

Labels

force-merge frontend ready ONLY add when PR is ready to merge/full CI is needed v1

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

5 participants

@maobaolong @DarkLight1337 @comaniac @njhill @vllm-bot