[Misc] Remove cache stream and cache events #3461
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
cadedaniel
approved these changes
Mar 18, 2024
zhuohan123
approved these changes
Mar 18, 2024
There was a problem hiding this comment.
LGTM! Please consider adding a test as suggested by @cadedaniel
|
@cadedaniel @zhuohan123 Added the test. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR removes
cache_streamandcache_events. Previously, these objects were introduced for fine-grained overlapping between swapping and model execution. However, the overlapping has been disabled since vLLM introduced CUDA graphs. Therefore, we don't need a separate CUDA stream and CUDA events for swapping.This PR removes such unused objects by launching the swapping ops in the default CUDA stream (same stream used in model execution).