Enforce floating arithmetic prerequisites for polynomial certificates

experiments/POLYNOMIAL_ENTHALPY_ADMISSIBILITY_2026-09-17.md

@@ -19,3 +19,5 @@ Tests cover nodal-positive/interior-negative cubics, a positive polynomial with
 `PreparedStellarEnthalpyStrategy<State>` resolves the enthalpy block from the compiled form type and captures the initial physical enthalpy as its anchor. Every full-state check denormalizes the state; pointwise positivity is not invariant under a general diagonal Riesz normalization. The strategy minimizes the full normalized linear-residual norm over the sufficient one-dimensional repair interval, verifies the actual normalized root candidate after rounding, and rejects a non-descent result. Unrelated blocks of the normalized correction are retained bit for bit. It does not prepare an operator or change an accepted state. A valid unaltered full candidate costs one Jacobian action for its true merit slope; an altered direction costs three actions (family model plus the actual rounded chosen direction). The anchor's certificate is cached only when coefficients agree exactly on every rank.
 
 A standalone extraction of the scalar/tensor certification engine was compiled with the system C++20 compiler and exercised before module integration. It certified a strictly positive quadratic with negative coarse coefficients and a three-dimensional zero-face bubble; it proved a negative pocket of depth 1e-10; it returned unresolved for an exact tangential zero. Production module compilation and MPI execution remain part of the integration campaign.
+
+Arithmetic requirements are enforced at compilation: `__FAST_MATH__` and a positive `__FINITE_MATH_ONLY__` are rejected, and static assertions require IEC 559 binary64 with subnormal support. Runtime floating-point modes must retain gradual underflow; the application must not enable flush-to-zero after construction. The existing Release build was inspected: `-O3 -g -DNDEBUG -std=c++23 -flto=thin -arch arm64`; no unsafe floating-point flags were present in the generated build commands.

libmeanfield/interface/solver/enthalpy_admissibility.cppm

@@ -15,6 +15,12 @@ module;
 #include <mfem.hpp>
 #include <mpi.h>
 
+// Outward enclosure depends on ordinary floating-point operations. Optimizers
+// that discard NaNs or reassociate expressions invalidate the certificate.
+#if defined(__FAST_MATH__) || (defined(__FINITE_MATH_ONLY__) && __FINITE_MATH_ONLY__ > 0)
+#error "Polynomial enthalpy certification does not support fast-math or finite-math-only compilation."
+#endif
+
 export module mean_field:solver.enthalpy_admissibility;
 export import :field.mfem;
 
@@ -54,6 +60,11 @@ export namespace mean_field::solver {
 }
 
 namespace mean_field::solver::enthalpy_detail {
+    static_assert(std::numeric_limits<double>::is_iec559 && std::numeric_limits<double>::radix == 2 &&
+                  std::numeric_limits<double>::digits == 53 && std::numeric_limits<double>::max_exponent == 1024 &&
+                  std::numeric_limits<double>::has_denorm == std::denorm_present,
+                  "Polynomial enthalpy certification requires IEEE 754 binary64 with subnormal support.");
+
     // Bounds concern the represented polynomial, not a tolerance-relaxed EOS.
     // Exact zero is retained algebraically so a prescribed zero trace does not
     // become an unresolved negative interval merely through outward rounding.