chore: pin celeris main (9aa258c) in the refapps, debugvars and the bench server - #381
Merged
Merged
Conversation
…ench server Brings celeris#647 into the validated code: the epoll standby loop no longer parks on top of its own unfinished transplant hand-off, which is the root cause of celeris#624. The checkptr tier measured that defect as transplant detached minus adopted equal to the I-CONN-2 drift, exactly, on both architectures. Also celeris#646, which splits ErrorCount into eleven cause buckets plus a standby half so the adaptive engine's post-promotion error rate can be attributed rather than only bounded (celeris#645). The race tier and the release soak run on this pin, so the long runs measure the fixed engine rather than the one the drift was found on.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Brings two engine changes into the validated code, ahead of the race tier and the release soak so the long runs measure the fixed engine.
celeris#647 fixes celeris#624. An epoll standby loop was parking on top of its own unfinished transplant hand-off: the suspend gate reads
connCount == 0as nothing left to do, but the count is decremented the instant a connection is detached for transplant, while the hand-off is completed later by that same loop. So the last connection a standby loop drained parked the loop on top of it, leaving the descriptor open and owned by no engine.The checkptr tier measured it as transplant detached minus adopted equal to the connection-count drift, exactly, on both architectures, and the fix reproduces deterministically from residual 8 to residual 0.
celeris#646 splits
ErrorCountinto eleven cause buckets plus a standby half, so the adaptive engine's post-promotion error rate can be attributed rather than only bounded (celeris#645).Ten modules repinned,
go build ./...and the report and checker suites clean.