Skip to content

Gas / resource benchmark harness for enter() #3

Open

Description

@samjay8

Capture the CPU-instruction + footprint budget for enter (mock verifier now, real verifier after M3) into docs/BENCHMARKS.md. Baseline for deciding if UltraHonk verification is affordable for a payments use case vs. a Groth16 fallback via pairing_check.

Activity

  1. samjay8 commented on Oct 6, 2026

    @samjay8
    ContributorAuthor

    Done by M3 step 2 — closing. decode is fuzz-covered: abi_fuzz tests (decode_never_panics_and_matches_length, word_decoders_reject_random_non_canonical_words, u32/u64 word round-trips over many values) run in CI on every push, plus property boundaries for the canonical-word rejection (R2-M7 fix, retained). 54 host tests green on PR #16.

  2. samjay8 commented on Oct 6, 2026

    @samjay8
    ContributorAuthor

    Reopening — closed in error (wrong issue number; the fuzzing work maps to #7). The gas/resource benchmark harness remains open, milestone M6.

  3. reopened this on Oct 6, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions