Skip to content

Commit 063e446

Browse files
luisleo526claude
andcommitted
The floors count the rows the integrated tree runs (INT14, wave E)
No lane's number is taken: every lane measured its floor against `origin/main` e9ad37d alone, so five of them would each have claimed "+1" over the same base and the sum would have been wrong twice over. These are counted on this tree. Kernel, 200 -> 202. Configured exactly as `ci_verify kernel` does (`PINEFORGE_BUILD_SOURCE_LAYER=OFF`, live runner and examples ON, tutorial and corpus OFF), `ctest -N` registers 203 and 202 run; the one row that does not is `test_native_live_websocket`, which still skips on a system libcurl, as it did at 199. The three rows are one per source-free lane: 199 main e9ad37d +1 E16 test_native_trail_stop_ladder +1 E23 test_native_continuation_portable +1 E24 test_native_continuation_digest_tail --- 202 run (203 registered) E19 and E25 add a row each as well, but `test_e19_excursion_path_order` and `test_session_islastbar_aggregation` both reach `pineforge/source/`, so the include-closure filter in tests/CMakeLists.txt drops them here — they are in the release count below. E17 has no commit at all: it ruled the engine not guilty of the slot-181 divergence (UB in the codegen-emitted TU). E18, E20, E21 and E22 add no row: every witness of theirs is a scenario inside a TU that already existed, which is why E18's three suites grew 472 -> 490, 603 -> 624 and 787 -> 815 checks instead. Release, 567 -> 571. `ctest --test-dir build-on` ran 571 of 571 with the six ABI providers prepared, so nothing skipped: 566 main e9ad37d +1 E16 test_native_trail_stop_ladder +1 E19 test_e19_excursion_path_order +1 E23 test_native_continuation_portable +1 E24 test_native_continuation_digest_tail +1 E25 test_session_islastbar_aggregation --- 571 run E24's own commit had already raised the pair to 200/567 in its branch; both numbers are superseded here rather than added to, which is what "no lane's number is taken" means. Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
1 parent fea2167 commit 063e446

3 files changed

Lines changed: 27 additions & 11 deletions

File tree

‎CONTRIBUTING.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -149,8 +149,8 @@ What each gate refuses:
149149
### The floors
150150

151151
`ci_verify.py` counts the CTest rows that actually **ran** and fails below a
152-
floor — `KERNEL_MIN_TESTS` ci_verify.py:110 and `RELEASE_MIN_TESTS`
153-
ci_verify.py:141.
152+
floor — `KERNEL_MIN_TESTS` ci_verify.py:122 and `RELEASE_MIN_TESTS`
153+
ci_verify.py:157.
154154
A deleted or silently skipped row is a failure, not a quieter run. If your
155155
change adds rows, raise the floor in the same commit and say by how much; if it
156156
legitimately removes one, lower it deliberately and say why. `--min-tests`

‎docs/pages/contributing-llm.md‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -103,8 +103,8 @@ finding to report, not a step to take.
103103
coverage claim.
104104

105105
10. **A test row never silently disappears.** Each profile counts the rows that
106-
*ran* against a floor: `KERNEL_MIN_TESTS` ci_verify.py:110 and
107-
`RELEASE_MIN_TESTS` ci_verify.py:141. Adding rows means raising the floor
106+
*ran* against a floor: `KERNEL_MIN_TESTS` ci_verify.py:122 and
107+
`RELEASE_MIN_TESTS` ci_verify.py:157. Adding rows means raising the floor
108108
in the same commit.
109109

110110
## The recipe for a lane
@@ -249,7 +249,7 @@ measurement that produced it, so a later change to it is visible as a change to
249249
the record, not as an edit to a literal.
250250

251251
**floor** — the minimum number of CTest rows a profile must actually run
252-
(`KERNEL_MIN_TESTS` ci_verify.py:110, `RELEASE_MIN_TESTS` ci_verify.py:141). It
252+
(`KERNEL_MIN_TESTS` ci_verify.py:122, `RELEASE_MIN_TESTS` ci_verify.py:157). It
253253
counts rows that ran, so a skipped row does not pad it.
254254

255255
**receipt** — the recorded evidence an ABI-comparison row needs (a prepared

‎scripts/ci_verify.py‎

Lines changed: 22 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -105,9 +105,21 @@
105105
# E12, E13 and E15 add no row: every one of their witnesses is a scenario
106106
# inside a TU that already existed.
107107
# 200 registered, 199 run: the WebSocket row still skips on a system libcurl.
108-
# 200 run = those 199 plus gap lane E24's test_native_continuation_digest_tail
109-
# (source-free: it drives NativeStrategyHost only), so 201 register.
110-
KERNEL_MIN_TESTS = 200
108+
# 202 run = those 199 plus the three TUs of the R5 follow-up wave (INT14,
109+
# wave E) that this profile builds -- measured on the integrated tree, not
110+
# summed from the lanes' own bases:
111+
# +1 E16 test_native_trail_stop_ladder
112+
# +1 E23 test_native_continuation_portable
113+
# +1 E24 test_native_continuation_digest_tail
114+
# All three are source-free: each drives NativeStrategyHost alone, so the
115+
# include-closure filter above keeps them. Lanes E19 and E25 add a row each
116+
# (test_e19_excursion_path_order, test_session_islastbar_aggregation), but both
117+
# TUs reach the source layer, so they register in release only -- see
118+
# RELEASE_MIN_TESTS. E17 adds nothing: it ruled the engine not guilty and has
119+
# no commit. E18, E20, E21 and E22 add no row either: every one of their
120+
# witnesses is a scenario inside a TU that already existed.
121+
# 203 registered, 202 run: the WebSocket row still skips on a system libcurl.
122+
KERNEL_MIN_TESTS = 202
111123
# Release-row floor, the same gate for the default profile. Before lane P7
112124
# only the kernel profile had one, so a row that left release alone (a
113125
# source-bound TU dropped from TEST_SOURCES, a deleted twin or ABI row) left a
@@ -136,9 +148,13 @@
136148
# +1 E5 test_offset_trail_quantized_arm
137149
# +1 E9 test_trail_activation_tick_reach
138150
# No release row skips, so 566 registered is 566 run.
139-
# 567 = those 566 plus gap lane E24's test_native_continuation_digest_tail,
140-
# which registers in every profile the tests build.
141-
RELEASE_MIN_TESTS = 567
151+
# 571 = those 566 plus the five TUs of the R5 follow-up wave (INT14, wave E):
152+
# the three KERNEL_MIN_TESTS lists above, which register here too, plus the two
153+
# source-bound ones the kernel profile does not build:
154+
# +1 E19 test_e19_excursion_path_order
155+
# +1 E25 test_session_islastbar_aggregation
156+
# No release row skips, so 571 registered is 571 run.
157+
RELEASE_MIN_TESTS = 571
142158
# CTest's closing summary: '100% tests passed out of N' when nothing failed,
143159
# '97% tests passed, 3 tests failed out of N' otherwise. N includes a skipped
144160
# row (counted as passed) and a row CTest could not start (counted as

0 commit comments

Comments
 (0)