Track T by default: the FAILING STEP named no owner. Line 36 of 63 is grep -q 'error: __thread counter: the per-thread variable area is full' /tmp/ctls_full.log || { echo "FAIL ctls: area-fu. The job's own src (test/c_thread_local_is_per_thread.c, 4 file(s)) is NOT used here on purpose: it is what the job compiles, not what broke, and guessing a lane from it is what sent three reds in one job to the wrong lane. This is a FALLBACK, not a finding — nothing says the defect is Track T's. Re-lane it before working it.

origin/master has advanced 5 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/c_thread_local_is_per_thread.c at 5e9c8da7481e in step 36/63, grep -q 'error: __thread counter: the per-thread variable area is full' /tmp/ctls_full.log || { echo "FAIL ctls: area-f… (auto-filed by twatch)

• Type: regression (auto-filed by Track T watcher, host borg, twatch 1f06bcf02d3f). Untriaged.
• Found: 2026-09-19T10:43:37Z
• Test source: test/c_thread_local_is_per_thread.c test/c_errno_is_per_thread.c +2
• Failing step: line 36 of 63 of the job's recipe; it names no source file of its own — so it is the JOB's sources, one line up, that are unproven here, not this step's.

  grep -q 'error: __thread counter: the per-thread variable area is full' /tmp/ctls_full.log || { echo "FAIL ctls: area-full did not report as an error naming the declaration"; cat /tmp/ctls_full.log; exit 1; }
  
  

Repro

tools/testmgr.py --tier native --job 'test-core#src:test/c_thread_local_is_per_thread.c' at 5e9c8da7481e6e9e62afed990d1847f1c7d91588

Range

The named sha 5e9c8da7481e 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 5e9c8da7481e, last good 2f7bee67cc0a, 2 commit(s) in range — the watcher narrows this by idle bisect; check tstate/TSTATE.md for the current range.

Log tail

pascal26:19: error: __thread errno: the per-thread variable area is full (0 bytes). It is a fixed cap and not a shortage of memory: the block is reserved by EmitTlsMainInstall, which runs before parsing, so its size is baked in before any thread-local has been seen. Raise it on the COMMAND LINE and recompile this program -- no compiler rebuild: -dPXX_TLS_USER_4K, _8K or _16K (also _0, _1K, _2K to shrink it). lib/rtl/palthread.pas reads the real size through __pxxTlsBlockSize, so it follows without being edited.
(tail)
ok: /tmp/testmgr-scratch-418715/c_thread_local26  [code=362537B  data=15808B  bss=92476B  procs=992  codeseg=364256B]
ok: /tmp/testmgr-scratch-418715/c_errno_per_thread26  [code=361545B  data=15776B  bss=92420B  procs=993  codeseg=364256B]
FAIL ctls: area-full did not report as an error naming the declaration
pascal26:19: error: __thread errno: the per-thread variable area is full (0 bytes). It is a fixed cap and not a shortage of memory: the block is reserved by EmitTlsMainInstall, which runs before parsing, so its size is baked in before any thread-local has been seen. Raise it on the COMMAND LINE and recompile this program -- no compiler rebuild: -dPXX_TLS_USER_4K, _8K or _16K (also _0, _1K, _2K to shrink it). lib/rtl/palthread.pas reads the real size through __pxxTlsBlockSize, so it follows without being edited.
  in: /tmp/testmgr-scratch-418715/compiler/../lib/crtl/include/errno.h
  near: extern __thread  errno >>>  

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

Log

• 2026-09-19 — auto-closed by the borg watcher: test-core#src:test/c_thread_local_is_per_thread.c passes at 3dc2317b83e9 (tier native); it was red at 5e9c8da7481e. Reopening is by a fresh NEW-RED stub, since a second red is a second finding with its own range.