str4d

Results 430 issues of str4d

One of the 0xPARC participants found the Poseidon chip interface tricky to work with. We should have a look over the API, as well as the wrapper they created around...

A-documentation
A-rust-api

We currently use `simple-example.rs` to show how to write circuits, but that file only uses the `MockProver` to test itself. We should have a page showing how to implement keygen,...

A-documentation
A-rust-api
A-book

Circuits need to specify the floor planner that they use, because the floor planner has explicit power to alter the structure of the circuit (and thus the verifying key). Once...

I-performance
A-rust-api
A-planning

Once #195 is implemented, floor planners will be able to reason about and control the horizontal aspects of the circuit. The selector combining optimisation is currently baked into the proving...

I-performance

At RWC I was chatting with [Wei Dai](https://twitter.com/_weidai), who mentioned (I think? I talked to a lot of people) that Anoma (was it?) are working on a specification called Vampir...

C-feature-request
A-dev-tooling
A-planning
A-compiler-api

Things I've noticed: - ~~[ ] Error types should implement `PartialEq` (so they can be checked in tests).~~ - Not doing any more (because `plonk::Error` now has a case that...

A-dev-tooling

Imagine the following circut that has two fields: a value, and a structural parameter. ```rust struct MyCircuit { value: Value, width: usize, } ``` when implementing the `Circuit` trait, it...

A-documentation

The user-facing API in #251 focuses on a single-field circuit, which is much easier for crate users to reason about (as they don't need to handle deferreds, or the way...

One of the design choices we made in `halo2_proofs` is that we want circuit synthesis and witness generation logic to be closely-coupled, to reduce chances for error (a design choice...

A-documentation
A-book

This will be the equivalent of `halo2_proofs::plonk::verifier`, but for the recursive proofs generated by `halo2`.