Track guessed as N from the FAILING STEP — line 2 of 17, /tmp/test_nilpy_zeroctor26 | diff -u test/test_nilpy_zero_argument_builtin_constructors.expected -, which names test/test_nilpy_zero_argument_builtin_constructors.expected. Not from the job's name or its src: those describe what the job is ABOUT, and this job's recipe spans 2 source file(s). The ranker reads frontmatter, so this line — not the body — decides who works it; correct it if the guess is wrong.

origin/master has advanced 4 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: test-core#src:test/test_nilpy_zero_argument_builtin_constructors.npy at d876977f4bb8 in step 2/17, /tmp/test_nilpy_zeroctor26 | diff -u test/test_nilpy_zero_argument_builtin_constructors.expected - (auto-filed by twatch)

• Type: regression (auto-filed by Track T watcher, host borg, twatch f35167cd4e55). Untriaged.
• Found: 2026-09-27T18:54:54Z
• Test source: test/test_nilpy_zero_argument_builtin_constructors.npy test/test_nilpy_zero_argument_builtin_constructors.expected
• Failing step: line 2 of 17 of the job's recipe; it names test/test_nilpy_zero_argument_builtin_constructors.expected.

  /tmp/test_nilpy_zeroctor26 | diff -u test/test_nilpy_zero_argument_builtin_constructors.expected -
  
  

Repro

tools/testmgr.py --tier native --job 'test-core#src:test/test_nilpy_zero_argument_builtin_constructors.npy' at d876977f4bb8be3ba799d0ee179a2b391aef4e0b

Range

The named sha d876977f4bb8 CANNOT be the cause — it touches no buildable file (docs / tickets / tstate only). It is the sha that was TESTED, i.e. the upper bound of an untested range; the cause is somewhere below it.

bad d876977f4bb8, last good 25ea38ef930d, 2 commit(s) in range — the watcher narrows this by idle bisect; check tstate/TSTATE.md for the current range.

Log tail

ok: /tmp/testmgr-scratch-119220/test_nilpy_zeroctor26  [code=479494B  data=103652B  bss=66444B  procs=2275  codeseg=483040B]
--- test/test_nilpy_zero_argument_builtin_constructors.expected	2026-09-11 21:30:22.035069937 +0200
+++ -	2026-09-27 20:50:39.124523110 +0200
@@ -8,8 +8,8 @@
 True True True True
 400
 [1, 2] 2 {'k': 0}
-('', 0, 1)
-('x', 3, 1)
+()
+()
 '' 0 []
 1 2 [0, 1]
 [0, 0, 0]

Stub ticket: signal only. Track T agent (face 2) enriches or a dev track takes it from the repro line.

• 2026-09-27 — fixed by frankB: list() wrapped its literal temp in pylist_mark_list, an identity marker whose result the store treats as owned and does not retain, so the temp held the list's only reference. 31d314dfa8 released that temp at the statement end, freeing the list c = list() had bound, and return (5, len(c)) printed (). The stamp is now a hoisted statement and the value the bare temp (as the set comprehension does), so the store retains. A return's temps are also no longer released. Row: test/test_nilpy_a_returned_container_survives_the_statement_end_release.npy.