You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
winderica 1322767a1e
Protogalaxy based IVC (#123)
3 months ago
.config feat: port CI from `halo2curves` (#4) 1 year ago
.github Enable WASM-compat and monitor it in the CI (#142) 4 months ago
cli Add solidity verifier of the nova+cyclefold (#87) 8 months ago
examples implement HyperNova's DeciderEth (#156) 3 months ago
folding-schemes Protogalaxy based IVC (#123) 3 months ago
solidity-verifiers implement HyperNova's DeciderEth (#156) 3 months ago
.gitignore feat: fold noir circuits, add an e2e example, tests, a `compile.sh` script and update CI (#131) 4 months ago
Cargo.toml Add the digest of the Relaxed R1CS instance for CycleFold as a public input to `AugmentedFCircuit` (#84) 8 months ago
LICENSE Initial commit 1 year ago
README.md Enable WASM-compat and monitor it in the CI (#142) 4 months ago
rust-toolchain Circuit compute_c reduce constraints (#97) 7 months ago

README.md

sonobe

Experimental folding schemes library implemented jointly by 0xPARC and PSE.

Sonobe is a modular library to fold arithmetic circuit instances in an Incremental Verifiable computation (IVC) style. It features multiple folding schemes and decider setups, allowing users to pick the scheme which best fits their needs.

Sonobe is conceived as an exploratory effort with the aim to push forward the practical side of folding schemes and advancing towards onchain (EVM) verification.

"The Sonobe module is one of the many units used to build modular origami. The popularity of Sonobe modular origami models derives from the simplicity of folding the modules, the sturdy and easy assembly, and the flexibility of the system."


Warning: experimental code, do not use in production.
The code has not been audited. Several optimizations are also pending. Our focus so far has been on implementing the Nova and CycleFold schemes and achieving onchain (EVM) verification.

Schemes implemented

Folding schemes implemented:

Work in progress:

Available frontends

Available frontends to define the folded circuit:

Usage

Build & test

You can test the library for both, WASM-targets and regular ones.

Regular targets

Tier-S targets allow the user to simply run cargo test or cargo build without needing to worry about anything. We strongly recommend to test using the light-test feature. Which will omit the computationally intensive parts of the tests such as generating a SNARK of 4~5M constraints to then verify it.

WASM targets

In order to build the lib for WASM-targets, use the following command: cargo build -p folding-schemes --no-default-features --target wasm32-unknown-unknown --features "wasm, parallel". Where the target can be any WASM one and the parallel feature is optional.

Trying to build for a WASM-target without the wasm feature or viceversa will end up in a compilation error.

Docs

Detailed usage and design documentation can be found at Sonobe docs.

WASM-compatibility & features

The sonobe/folding-schemes crate is the only workspace member that supports WASM-target compilation. But, to have it working, getrandom/js needs to be imported in the Cargo.toml of the crate that uses it as dependency.

[dependencies]
folding-schemes = { version = "0.1.0", default-features = false, features = ["parallel", "wasm"] }
getrandom = { version = "0.2", features = ["js"] }

See more details about getrandom here: https://docs.rs/getrandom/latest/getrandom/#webassembly-support.

Also, notice that:

  • wasm feature IS MANDATORY if compilation to WASM targets is desired.
  • parallel feature enables some parallelization optimizations available in the crate.
  • light-test feature runs the computationally-intensive parts of the testing such as the full proof generation for the Eth-decider circuit of Nova which is approximately 4-5M constraints. This feature only matters when it comes to running Sonobe's tests.

Folding Schemes introduction

Folding schemes efficiently achieve incrementally verifiable computation (IVC), where the prover recursively proves the correct execution of the incremental computations. Once the IVC iterations are completed, the IVC proof is compressed into the Decider proof, a zkSNARK proof which proves that applying $n$ times the $F$ function (the circuit being folded) to the initial state ($z_0$) results in the final state ($z_n$).

Where $w_i$ are the external witnesses used at each iterative step.

In other words, it allows to prove efficiently that $z_n = F(...~F(F(F(F(z_0, w_0), w_1), w_2), ...), w_{n-1})$.

Overview of sonobe

Sonobe is a folding schemes modular library to fold arithmetic circuit instances in an incremental verifiable computation (IVC) style. It also provides the tools required to generate a zkSNARK proof out of an IVC proof and to verify it on Ethereum's EVM.

The development flow using Sonobe looks like:

  1. Define a circuit to be folded
  2. Set which folding scheme to be used (eg. Nova with CycleFold)
  3. Set a final decider to generate the final proof (eg. Spartan over Pasta curves)
  4. Generate the decider verifier

The folding scheme and decider used can be swapped with a few lines of code (eg. switching from a Decider that uses two Spartan proofs over a cycle of curves, to a Decider that uses a single Groth16 proof over the BN254 to be verified in an Ethereum smart contract).

The Sonobe docs contain more details about the usage and design of the library.

Complete examples can be found at folding-schemes/examples

License

Sonobe is MIT Licensed.

Acknowledgments

This project builds on top of multiple arkworks libraries. It uses Espresso system's virtual polynomial abstraction and its SumCheck implementation.

The Solidity templates used in nova_cyclefold_verifier.sol, use iden3's Groth16 implementation and a KZG10 Solidity template adapted from weijiekoh/libkzg.

In addition to the direct code contributors who make this repository possible, this project has been made possible by many conversations with Srinath Setty, Lev Soukhanov, Matej Penciak, Adrian Hamelink, François Garillot, Daniel Marin, Han Jian, Wyatt Benno, Nikkolas Gailly and Nalin Bhardwaj, to whom we are grateful.