fix(ci): no new assertionless spec test, green on master - #5502
Merged
Merged
Conversation
Corpus Ratchet failed its first step on master: three files gained a test that cannot fail. specs/igla/coder/pipeline.t27 added one whose body is 'assert true' (#4796); ternary_mac_synth.t27 (on_clock_reset) and gf16_matmul_top.t27 (led_r23_toggles) added ones whose bodies are only comments, the second ending "we will skip and hope". None of the three can be given an assertion that runs: t27c test-report reports all three specs BLOCKED in Zig, so no test in them executes. The three bodies are removed, as the gate asks; the pilot measurement in pipeline.t27 stays as written, with its invariant line. The ledger drops the one row that was already slack: specs/igla/race/cordic.t27 144 -> 0 (total 4049 -> 3905). The pipeline seal, stale on master since #4796, is resealed; --verify reports all MATCH. All four backends generate the three specs before and after, and the corpus suite summary is identical to master's. Refs #5497 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Contributor
Contributor
|
📓 NotebookLM Notebook linked to this PR
This notebook contains session context, decisions, and artifacts for this work. |
This was referenced Oct 2, 2026
This was referenced Oct 2, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
The first step of
Corpus ratchet (expected-failure ledger)fails on master: three files gained a test that cannot fail.specs/igla/coder/pipeline.t27igla_coder_pilot_checkpoint_measured_2026_09_24(#4796)assert truespecs/port/fpga/verilog/ternary_mac_synth.t27on_clock_resetspecs/port/fpga/vivado/gf16_matmul_top.t27led_r23_togglesWhy remove rather than add an assertion
t27c test-reportreports all three specs BLOCKED in Zig (they do not compile), so no test in them executes today. An assertion added now would be as untested as the empty body. The gate's own instruction is "give it an assertion, or do not add it"; this takes the second branch. Whaton_clock_resetmeant to check is the declared value ofacc_out, whichinitial_valuealready asserts. The pilot measurement inpipeline.t27stays as written, with itsinvariantline.Ledger and seal
tools/assertionless_spec_tests_baseline.txt: the gate also fails when a file drops below its row, so the one stale row moves down in the same commit:specs/igla/race/cordic.t27144 -> 0 (total 4049 -> 3905). No row rises.coder_igla-coder-pipeline.jsonwas already stale on master (spec hash and all four gen hashes, since feat(igla-coder): record the pilot's measured t27 result (Closes #1041) #4796). Resealed with thet27cbuilt from this tree;t27c seal --verify: all MATCH. The two port specs have no seal.Checked
check_assertionless_spec_tests.py:ok: no file gained an assertionless test, and none silently lost one;--self-checkpasses.t27c suite --ratchet --corpus-onlysummary is identical to master (pipeline.t27 still discards 517 tokens, as pinned).The job will still stop later, at
tri types ratchet(15 new type-name conflicts, #5497). That is the next pull request.Refs #5497
🤖 Generated with Claude Code