test(rtx-fsi): RTX_FSI{2,3}_UMEAN inflow override for the TWIN-1 parameter sweep
Performance Benchmarks / Run Benchmarks (push) Canceled after 0s
CI / Format Check (push) Canceled after 0s
CI / Clippy Check (push) Canceled after 0s
CI / Build (macos-latest) (push) Canceled after 0s
CI / Build (ubuntu-latest) (push) Canceled after 0s
CI / Test (macos-latest) (push) Canceled after 0s
CI / Test (ubuntu-latest) (push) Canceled after 0s
CI / Build CPU-Only (Explicit) (push) Canceled after 0s
CI / Python Bindings (maturin) (macos-latest) (push) Canceled after 0s
CI / Python Bindings (maturin) (ubuntu-latest) (push) Canceled after 0s
CI / WASM Build + Size Check (push) Canceled after 0s
CI / Distributed Training Tests (push) Canceled after 0s
CI / CI Success (push) Canceled after 0s
Documentation / Build API Documentation (push) Canceled after 0s
Documentation / Build User Guide (push) Canceled after 0s

case_from_env applies the knob over the case's benchmark inflow; unset,
the same f64 flows and both committed defaults are verified
digit-identical in vivo (physics lines diff-clean vs same-day
baselines; only wall-split timing percentages moved). Physics bands in
both tests are guarded to benchmark inflow — an off-benchmark march
(the sweep) asserts machinery invariants only; its bands live in the
TWIN-1 composition harness (omni-cortex
docs/twin_composition_campaign.md). The override prints loudly so
sweep logs are self-describing.

Co-Authored-By: Claude Fable 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01X2GmJXeQ2njUecEKiJZ1G2
This commit is contained in:
Omar Sobh
2026-08-31 21:25:30 -07:00
co-authored by Claude Fable 5
parent 4aec4e589d
commit c08abbdc4f
3 changed files with 42 additions and 7 deletions
@@ -79,6 +79,26 @@ pub const FSI3: BenchmarkCase = BenchmarkCase {
rigid_drag_reference: 439.45,
};
/// `RTX_{prefix}_UMEAN` over the case's benchmark inflow — TWIN-1's
/// sweep parameter (`docs/twin_composition_campaign.md` in omni-cortex).
/// Unset, the case comes back unchanged: the same f64 flows and the
/// march is digit-identical by construction (and verified in vivo on
/// both committed defaults). When set, the march's printed rigid-drag
/// reference still names the BENCHMARK value, which only applies at the
/// case's own inflow — the override line below keeps logs honest.
pub fn case_from_env(prefix: &str, case: BenchmarkCase) -> BenchmarkCase {
let u_mean = env_or(&format!("RTX_{prefix}_UMEAN"), case.u_mean);
if u_mean.to_bits() != case.u_mean.to_bits() {
println!(
" {} u_mean OVERRIDDEN to {u_mean} (benchmark {}; Re = {:.0})",
case.name,
case.u_mean,
100.0 * u_mean
);
}
BenchmarkCase { u_mean, ..case }
}
pub fn circle_sdf(x: f64, y: f64) -> f64 {
((x - 0.2).powi(2) + (y - 0.2).powi(2)).sqrt() - 0.05
}