origin/master has advanced 3 commit(s) since this sha. Re-verify at current HEAD before acting — the callback is tagged to the sha that was tested, which may no longer be the state of the tree.

regression CASCADE: 17 jobs newly red at 343a52551808 (auto-filed by twatch)

• Type: regression cascade (auto-filed by Track T watcher, host plexus). Untriaged. 17 jobs went red in ONE sweep — treat as ONE root cause until triage proves otherwise; do NOT fan out per-job tickets.
• Found: 2026-08-15T22:24:34Z
• Root-cause suspects in the red set: none of the known root jobs — likely a broken build or harness event

Repro (start with a suspect, or any listed job)

tools/testmgr.py --tier full --job '<job>' at 343a52551808e14463db6ff315dc6d79b8624bfd

Newly red jobs

• test-nilpy#src:test/test_nilpy_encode.npy
• test-nilpy#src:test/test_nilpy_encode_decode_codecs.npy
• test-nilpy#src:test/test_nilpy_intrinsic_result_chain.npy
• test-nilpy#src:test/test_nilpy_math_domain_errors.npy
• test-nilpy#src:test/test_nilpy_math_log.npy
• test-uforth#core
• test-uforth#coreexttest
• test-uforth#coreplustest
• test-uforth#doubletest
• test-uforth#exceptiontest
• test-uforth#facilitytest
• test-uforth#filetest
• test-uforth#localstest
• test-uforth#memorytest
• test-uforth#searchordertest
• test-uforth#stringtest
• test-uforth#toolstest

Cascade stub: one signal for one event. Track T agent (face 2) or the owning dev track triages the root; individual tickets only for whatever remains red after the root is fixed.

TRIAGE 2026-08-16 — two roots, both named; 15 of 17 already green

Not one root cause but two, and the split falls exactly on "does it reproduce on the PINNED binary".

Root 1 (mine, FIXED at e10590243) — 15 jobs. The three nilpy encode / intrinsic_result_chain rows and all twelve uforth shards. Two defects, both introduced by adding the Python method set to TPyBytes:

• a pylib name collision — my absolute-index PyBytesPut shadowed the existing cursor-advancing one at every call site of pystr_encode_enc_err, so "hi".encode("latin-1") came back b'i\x00'. Renamed to PyBytesSet.
• a parse gap the method set made universal — [[bug-nilpy-a-chained-method-on-a-runtime-dispatched-result-is-refused]]: a second .method() on a runtime-dispatched result was unexpected token, because the dispatcher returns an AN_TERNARY and the suffix loops required AN_CALL. uforth.py chains string methods everywhere, which is why twelve shards went with it.

Verified at HEAD: uforth.py compiles, the smoke run answers 3 3, and the stringtest word set is byte-identical to CPython again.

Root 2 (NOT mine) — 2 jobs. test_nilpy_math_domain_errors and test_nilpy_math_log reproduce identically on the PINNED binary and bisect to Track B's 11321a09c (Power and LogN, 26x each). Filed as [[bug-b-power-lost-an-ulp-on-a-half-integer-exponent]] — one genuine ulp regression on math.pow(2.0, 0.5) plus one .expected row that is stale because the rewrite CORRECTED it. Those two rows stay red until Track B lands that ticket; no further cascade signal is needed for them.

Closed: the cascade is triaged and its larger root is fixed and pushed. The residual is one owned ticket in another lane.

Log

• 2026-08-16 — resolved, commit 1d9f627c7.