Repository navigation
ci(perf): loosen the HTTPS churn-ratio backstop for Intel runners - #316
Merged
Merged
Conversation
The first gate run on dev after c7f71d3 landed on an Intel Xeon Platinum 8573C and failed on healthy code: churn ratio 0.071 against the 0.08 backstop (RTT 0.08 ms, every other gate fine). Plaintext accepts are relatively faster there, so the ratio is not as hardware-neutral across CPU vendors as the AMD-only data suggested: a regressed AMD tree measures 0.052 and a healthy Intel one 0.071, so no absolute backstop separates them. Drop the churn-ratio backstop to 0.04 (large drops only) and leave the rest to the same-CPU check. #307-class regressions stay caught by the RTT gate on any CPU (0.08 -> 43 ms). Record the Intel run in the history; tests cover it and the same-CPU churn-ratio path.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Follow-up to #314. The first dev gate run after the merge (37301931499) landed on a new Intel Xeon Platinum 8573C runner and failed on healthy code. Its churn ratio was 0.071 against the 0.08 backstop. RTT was 0.08 ms and every other gate passed.
The AMD runs measured 0.11–0.13, so I assumed the ratio was hardware-neutral. Across CPU vendors it isn't: a regressed AMD tree measures 0.052 and a healthy Intel one 0.071, so no absolute backstop can separate them.
All 8 recorded CI runs pass 10/10. The tree with the #307 fix disabled still fails (RTT, churn).
scripts/ciunit tests pass.