From 34619892e49a0f1a3119bb14137fb17380c64e9f Mon Sep 17 00:00:00 2001
From: bong-water-water-bong
 <277547417+bong-water-water-bong@users.noreply.github.com>
Date: Tue, 10 Nov 2026 23:50:23 -0400
Subject: [PATCH] tests: register the last baseline entry, so the
 unreferenced-test sweep is zero-tolerance (Fixes #2018)

The baseline recorded tests/hyperloom_benchmark_test.py as one that needs the benchmark harness. It does not: it stubs
llama-bench and runs stdlib-only (hashlib, importlib.util, json, pathlib, platform, sys) -- 'hyperloom_benchmark_test: OK',
rc=0, in under a second. So it was the same mistake as the other entry: a test with no runner, recorded as a test that
cannot run here.

With this registered the baseline list is empty and the sweep stops being a record of known gaps: any future tests/*.py
referenced by nothing that can run it now fails on the spot.
---
 CMakeLists.txt                     | 5 +++++
 tools/ci/unreferenced_py_tests.txt | 5 ++---
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 610142905..cb2fdfaba 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -311,6 +311,11 @@ add_python_test(ci_targets_self ${CMAKE_SOURCE_DIR}/tools/ci/ci_targets_check.py
 # baseline records a test as never-run, which for a passing test is the wrong outcome -- so it is registered and the line
 # is removed, which is the shrink the baseline's own comment asks for.
 add_python_test(moe_expert_chunks_accum_test ${CMAKE_SOURCE_DIR}/tests/moe_expert_chunks_accum_test.py)
+# engine#2018: the last baseline entry. The file recorded it as "needs the benchmark harness", but the test stubs
+# llama-bench (stdlib only: hashlib/importlib.util/json/pathlib/platform/sys) and passes on a plain interpreter, so it is a
+# test with no runner rather than a test that cannot run here. Registered, and the baseline is now empty: from here the
+# sweep is zero-tolerance instead of a record of known gaps.
+add_python_test(hyperloom_benchmark_test ${CMAKE_SOURCE_DIR}/tests/hyperloom_benchmark_test.py)
 # engine#1820: the other half of ci_targets_check -- a standalone shell test that nothing references cannot fail.
 # `need` is every tests/*.sh, `have` is every file that mentions its basename somewhere that can run it, and
 # tools/ci/unreferenced_tests.txt carries the deliberate exclusions WITH their reasons, so the list can shrink.
diff --git a/tools/ci/unreferenced_py_tests.txt b/tools/ci/unreferenced_py_tests.txt
index e27a98d88..742700096 100644
--- a/tools/ci/unreferenced_py_tests.txt
+++ b/tools/ci/unreferenced_py_tests.txt
@@ -24,8 +24,8 @@
 # A SET baseline, not an allow-list: `tools/ci/tests_referenced_check.py` FAILS when a name appears here that is not in
 # the tree's current set, and REPORTS a line whose file is registered now, so this can only shrink.
 #
-# The number is 1 (was 2: moe_expert_chunks_accum_test.py is portable and passing, so it is registered rather than
-# baselined -- engine#2018), and getting it right took a correction worth recording: the first version of the rule scanned
+# The number is 0. Both original entries were portable tests with no runner, not tests that cannot run here, so both are
+# registered (moe_expert_chunks_accum_test.py, then hyperloom_benchmark_test.py -- engine#2018), and getting it right took a correction worth recording: the first version of the rule scanned
 # `RUNNERS`, which contains one FILE (CMakeLists.txt) and two directories, and tested `is_dir()` on all three -- so every
 # `add_python_test(...)` registration in CMakeLists.txt was invisible and 35 of 42 tests looked unregistered. The arm
 # "a cleared baseline entry is reported" caught it: registering a test in CMakeLists.txt did not clear its line. The true
@@ -33,4 +33,3 @@
 #
 # Each line is a real test whose runner was never added; the fix per file is one `add_python_test(...)`.
 #
-tests/hyperloom_benchmark_test.py