[rorupuntou]

Description

Extends test coverage for EIP-2537 MAP_FP_TO_G1 (precompile 0x10) and MAP_FP2_TO_G2 (precompile 0x11), focusing on input padding, boundary validation, and coordinate modulus checks:

1. Top-16 Byte Big-Endian Zero Padding Violations:
   • Tests non-zero bytes across byte indices 0..15 of the 64-byte field representations (e.g. byte 15 0x01, high bit 0x80, mid-padding byte 8, and dirty padding paired with a valid Spec.P - 1 coordinate), verifying that implementations do not ignore dirty upper padding before coordinate validation.
   • For MAP_FP2_TO_G2, covers dirty padding individually on $c_0$ (0..15), individually on $c_1$ (64..79), and simultaneously on both $c_0$ and $c_1$.
2. Modulus Overflows:
   • Values $\ge P$ including $P$, $P + 1$, $2^{381} - 1$, $2^{384} - 1$, and $P + 2^{128}$ for single coordinate inputs and simultaneous $(c_0, c_1)$ coordinates.
3. Calldata Length Violations:
   • Truncated inputs: 32 bytes (single word) and 64 bytes (single Fp for Fp2).

All 66 test fixtures generate cleanly against the Prague fork (uv run fill).

Related Issues or PRs

Follow-up to #3641 (rebased onto forks/bogota from #3821).

Checklist

• Ran fast static checks to avoid CI fails, see Code Standards & Verifying Changes: just static
• PR title has the form <type>(<area>): <description>, where <type> and <area> come from an appropriate C-<type>, respectively A-<area>, label. The title should match the target squash commit message.

Cute Animal Picture

[rorupuntou]

Hi @spencer-tb @marioevz! Whenever you have a moment, could you approve the pending CI workflow run and take a quick look at these negative boundary test cases for MAP_FP_TO_G1 (0x10) and MAP_FP2_TO_G2 (0x11) on forks/bogota? All 66 Prague fixtures fill cleanly locally and pass just static. Thanks!