ci: Scope vLLM build caches by ref - #214
Open
alexeldeib wants to merge 2 commits into
Open
Conversation
Let branch builds read the canonical shared cache while writing only to a ref-scoped cache. Queue runs per ref so every pushed SHA builds serially without cross-branch cache races or duplicate ARM contention. Co-authored-by: Codex <codex@openai.com> Signed-off-by: Ace Eldeib <aeldeib@coreweave.com>
Scope main cache writes separately from the legacy shared ref. Branch builds now read branch, main, and legacy caches in order while writing only to their branch cache, preserving warm starts without cross-branch manifest races. Co-authored-by: Codex <codex@openai.com> Signed-off-by: Ace Eldeib <aeldeib@coreweave.com>
|
@alexeldeib Build complete, success: https://github.com/coreweave/ml-containers/actions/runs/32306688845 |
Contributor
Author
|
mehh this logic is kind of tricky to get the right behavior...using branch alone leads to a cold first build...so ideally you read the shared cache once, then write to branch, then read from branch...mmm edit: think that kind of works actually? |
|
@alexeldeib Build complete, success: https://github.com/coreweave/ml-containers/actions/runs/32306688816 |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Let branch builds read the canonical shared cache while writing only to a ref-scoped cache. Queue runs per ref so every pushed SHA builds serially without cross-branch cache races or duplicate ARM contention. This avoids auto-cancellation which I recall Eta preferred, while permitting concurrency groups to prevent same-branch cache collisions.