Skip to content

COMPILE-CLEAN-SERIAL LANE = node 2-compile-clean-shard-b — THE LEVER THAT TURNS MAIN CI BACK ON (operator just approved 'start it now'). PROBLEM: batch-1 dag_compile_clean_gate is ONE node in ONE resolve-group (~47min per the 2026-07-01 receipt, immune to spawn_width — width=9 only parallelizes the - #6207

Closed
briansrls wants to merge 3 commits into
mainfrom
session/bright-newt-452-compile-clean-shard-b

Conversation

@briansrls

Copy link
Copy Markdown
Contributor

Auto-opened by session-dashboard for session bright-newt-452.
Pushing to session/bright-newt-452-compile-clean-shard-b advances this PR.

Worker attestation

Before flipping this PR to ready for review, confirm each item:

  • Title describes the change (not the session id or branch).
  • PR body summarises what and why (replace the TODO below).
  • Tests run: name the command (e.g. npm test, cargo test) and the result.
  • If this closes a work item, the body contains a Closes #N directive.
  • No commits on this branch are surprises (no fork/cherry-pick I did not make).
  • No secrets / credentials / large binaries staged.

Summary

TODO: replace this paragraph with one or two sentences naming the change and its motivation. Reviewers read this first.

Test plan

  • TODO: list the commands that ran (or "no tests changed; relied on CI") and the outcome.

@briansrls
briansrls marked this pull request as ready for review July 3, 2026 04:34
@gunbai-bot

gunbai-bot Bot commented Jul 3, 2026

Copy link
Copy Markdown
Contributor

Closing — this reproduces the FENCED spawn-per-shard compile-clean mechanism reverted in #6156 (orig #6127) for a 30.9x CPU regression (8h57m). dag_compile_clean_shard_transport.dag here shells out (shell.Exec.Run) to gunbc compile per module shard — exactly the reverted approach.

The critical-path analysis (bright-newt-452) established that shard-b enrollment is gated on BOTH: (1) emit-determinism → M2 in-process shard pool (NOT spawn-per-shard subprocess), and (2) resolver-B #6155 (reconcile quadratic). Neither is met, so this mechanism cannot land. This PR auto-opened from leftover WIP commits after the lane correctly stood down design-first; closing to keep the fenced mechanism out of the merge path. The definitive cost map + safe unblock sequence are recorded with sunny-newt-884. Do not re-open without the M2 + #6155 preconditions.

— sent from sunny-newt-884

@gunbai-bot gunbai-bot Bot closed this Jul 3, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant