claude
bd7585b609
Cycle-close audit for the self-description cycle. Architect: drift_found (mild), core holds -- the gate is pure string shape in core imported by every seam (C28), the engine never evaluates description text (C17/inv 10), determinism pins untouched (C1); the store boundary is gated by construction with the raw write private; Op::Doc is a closed typed construct (C25). Regression: aura-bench green (exit 0) after one cycle-caused infra fix and one ratified re-pin. Infra fix: the bench campaign seeds (BP_SMA_A/B) predate C29 and were refused by the store gate (exit 2, infrastructure-first rule) -- both thread the authored sma_signal meaning line. Ratify: cli_fixed_cost's run_line_fnv moved because bench_sma_a.json now carries its doc -- the blueprint's content id shifts exactly as the spec anticipates for fixtures gaining docs; campaign result fingerprints are unchanged (descriptions stay out of execution, C29). Baselines re-pinned via aura-bench pin (throughput baselines re-pinned in the same sweep, all deltas within tolerance at pin time). Drift disposition: - [high] C29 guarantee overclaim vs emit/binding-column vocabularies -> contract now carries an explicit Scope paragraph naming the residue, tracked as #321 (also covers the [medium] non-roster schema walk gap). - [low] Delay's meaning line cited the stale C5 for the delay register; corrected to C9. refs #316, refs #321
aura-bench
Wall-clock benchmarks with committed baselines (issue #251, phase 1). Catches hot-path regressions the E2E suite structurally cannot see: every surface pins its inputs and asserts a determinism fingerprint, so a fast-but-wrong binary fails the bench instead of pinning a bogus baseline.
Running
cargo run --release -p aura-bench -- run # measure + compare, report-only
cargo run --release -p aura-bench -- run --surface engine_throughput
cargo run --release -p aura-bench -- pin # rewrite committed baselines
Debug builds refuse to measure (exit 2) — wall-clock numbers from a debug build measure the profile, not the code.
Surfaces
| surface | workload | metrics |
|---|---|---|
engine_throughput |
10M synthetic bars through an SMA-cross harness (library) | wall_s, bars_per_s |
ingest_throughput |
24 synthetic archive months x 20 fresh drains via the ingest window API (library) | wall_s, bars_per_s |
campaign_sweep |
8-cell sweep-only campaign, release aura binary on a scratch project |
wall_s, cpu_percent, peak_rss_mb |
campaign_heavy |
same cells; sweep → gate → walk-forward → 1000-resample bootstrap | wall_s, cpu_percent, peak_rss_mb |
cli_fixed_cost |
spawn floor (aura --help) + minimal data-only project run |
help_ms, run_ms |
Discipline
- Quiet box: the driver warns at 1-min loadavg > 2.0. Close heavy jobs first.
- 1 discarded warmup + 3 measured reps (median). Reps must agree on the fingerprint — cross-rep drift is reported as nondeterminism, not averaged.
- Baselines are per-host context (
hostblock inside each JSON): comparison on a different host prints an informational banner and stays report-only. - Timing drift NEVER fails the run (report-only phase, #251 cadence decision); a NOTICE line flags drift >= 10%. A fingerprint mismatch exits 1.
- Re-pin (
pin) deliberately, after an accepted behaviour or performance change; the baseline diff is reviewed like any other commit.
Deferred (follow-up on the tracker)
- Instruction-count backend (iai-callgrind) once valgrind is installed on the measuring host — slots in as a sibling backend without reshaping baselines.
- Native-crate project-load sub-cost (dylib SHA-256) — needs a build step in the scratch project.
- Thresholded failure, once recorded variance data justifies thresholds.