Code Monkey home page Code Monkey logo

fastcrypto's Introduction

fastcrypto

crate Docs Build status Update the Fastcrypto pointer Sui Apache2/MIT licensed Rust Version codecov

fastcrypto is a common cryptography library used in software at Mysten Labs. It is published as an independent crate to encourage reusability across different applications and domains. It is a wrapper library around several carefully selected crates with the following considerations:

  • Security: Whether the libraries are vulnerable to known attack vectors or possible misuses.
  • Performance: Whether the crate performs cryptographic operations with speed after extensive benchmarking. This is critical for the Sui Network to be performant when signing and verifying large amounts of transactions and certificates.
  • Determinism: Whether the signature is non-malleable.
  • Popularity: Whether the library is used by other consensus critical systems.

Furthermore, we extend the selected libraries with additional features:

  • Robust testing framework: Wycheproof tests and prop tests are added when possible to protect against arbitrary inputs and crafted edge cases.
  • Zeroization: Sensitive private key materials are cleared from memory securely and proactively when it goes out of scope using zeroize trait.
  • Serialization: Effective and standardized serialization are required.

This library will be continuously updated with more schemes and faster implementations based on benchmarking results, RFC updates, new research and auditor inputs.

This crate contains:

  • Traits that should be implemented by concrete types representing digital cryptographic materials.

    • [SigningKey]: Trait implemented by the private key with associated types of its public key and signature.
    • [VerifyingKey]: Trait implemented by the public key with associated types of its private key and signature. It also includes a default implementation of batch verification that fails on empty batch verification.
    • [Authenticator]: Trait implemented by the signature with associated types of its public key and private key.
    • [AggregateAuthenticator]: Trait implemented by the aggregated signature, which allows adding signatures to the aggregated signature and verifying against the public keys with the corresponding messages.
    • [KeyPair]: Trait that represents a public/private keypair, which includes the common get priv/pub key functions and a keypair generation function with seeded randomness.
    • [ToFromBytes]: Trait that aims to minimize the number of steps involved in obtaining a serializable key.
    • [EncodeDecodeBase64]: Trait that extends ToFromBytes for immediate conversion to/from Base64 strings. This is the format in which cryptographic materials are stored.
  • Concrete signature schemes of type that implement the recommended traits required for cryptographic agility.

    • Ed25519: Backed by ed25519-consensus crate. Compliant to ZIP-215 that defines the signature validity that is lacking from RFC8032 but critical for consensus algorithms. ed25519-dalek is fully deprecated due to the recently discovered Chalkias double pub-key api vulnerability.
    • Secp256k1: ECDSA signatures over the secp256k1 curve. Backed by Secp256k1 FFI wrapper that binds to C library and provides performance faster than the native Rust implementation k256 library by ~30% on verification. Produces either a standard ECDSA signature or a 65-byte recoverable signature of shape [r, s, v] where v can be 0 or 1 representing the recovery Id. Produces deterministic signatures using the pseudo-random deterministic nonce generation according to RFC6979, without the strong requirement to generate randomness for nonce protection. Uses sha256 as the default hash function for sign and verify. An interface for verify_hashed is provided to accept a pre-hashed message and its signature for verification. Supports public key recovery by providing the Secp256k1 recoverable signature with the corresponding pre-hashed message.
    • BLS12-381: Backed by blst crate written in Assembly and C that optimizes for performance and security. G1 and G2 points are serialized following ZCash specification in compressed format. Provides methods for verifying signatures in the G1 group against public keys in the G2 group. Provides methods for aggregating signatures and fast verifying aggregated signatures, where public keys are assumed to be verified for proof of possession.
    • Secp256r1: ECDSA signatures over the secp256r1 curve backed by the p256 crate which is a pure rust implementation of the Secp256r1 (aka NIST P-256 and prime256v1) curve. The functionality from p256 is extended such, besides standard ECDSA signatures, our implementation can also produce and verify 65 byte recoverable signatures of the form [r, s, v] where v is the recoveryID. Signatures are produced deterministically using the pseudo-random deterministic nonce generation according to RFC6979, without the strong requirement to generate randomness for nonce protection. Uses sha256 as the default hash function for sign and verify. Supports public key recovery by providing the Secp256r1 ECDSA recoverable signature with the corresponding pre-hashed message.
  • Utility functions that serve as the underlying RUST implementation for the Move smart contract api.

    • HKDF: An HMAC-based key derivation function based on RFC-5869, to derive keypairs with a salt and an optional domain for the given keypair. This requires choosing an HMAC function that expands precisely to the byte length of a private key for the chosen KeyPair parameter.
    • Pedersen Commitment: Function to create a Pedersen commitment with a value and a blinding factor. Add or subtract Ristretto points that represent Pedersen commitments.
    • Bulletproofs Range Proof: Function to prove that a committed value is an unsigned integer that is within the range [0, 2^bits). Function to verify that the commitment is a Pedersen commitment of some value with an unsigned bit length, a value is an integer within the range [0, 2^bits)
  • A asynchronous signature service is provided for testing and benchmarking.

Tests

There exist tests for all the three schemes, which can be run by:

$ cargo test

Benchmarks

One can compare all currently implemented schemes for sign, verify, verify_batch and key-generation by running:

$ cargo bench

A report of the benchmarks is generated for each release, allowing easy comparison of the performance of the different cryptographic primitives and schemes available in fastcrypto. As an example, we get these timings for signing messages and verifying the signature for the different schemes in fastcrypto as of revision 375ab9:

Signature schemes comparison.

Below is another plot made using data from the benchmark report, showing benchmarks for batched signature verification where all signatures are on the same message:

Batched signature verification with all signatures on same message.

License

All crates licensed under either of

fastcrypto's People

Contributors

jonas-lj avatar dependabot[bot] avatar joyqvq avatar benr-ml avatar huitseeker avatar kchalkias avatar erwanor avatar punwai avatar jnaulty avatar andll avatar mskd12 avatar rex4539 avatar bajpai244 avatar natalipatel8 avatar patrickkuo avatar aschran avatar mwtian avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.