matter-labs / era-boojum Goto Github PK

🌟 Feature Request

📝 Description

According the papre, it's faster than Poseidon2

https://eprint.iacr.org/2023/1025.pdf

🤔 Rationale

Explain why this feature is important and how it benefits the project.

📋 Additional Context

Add any other context or information about the feature request here.

Boojum violates terms of MIT / Apache license.

This line says you've used plonky2 source code:

But it is not enough. You must include their copyright, even if it's just 5% of code reuse.

🐛 Bug Report

era-boojum uses the currently deprecated https://github.com/rust-lang/packed_simd which can no longer be compiled with recent rust versions.

📝 Description

era-boojum uses the currently deprecated https://github.com/rust-lang/packed_simd which had its last update some 10 months ago. The project maintainer has deprecated it in favor of std::simd. During this time several of the features have been removed from the rust compiler, (std_simd, platform_intrinsics, etc.). Thus the package can no longer be compiled with latest rust stable (v1.78) or any subsequent nightlies.

Related errors
rust-lang/packed_simd#359
rust-lang/packed_simd#360

🔄 Reproduction Steps

Compile with rustc 1.78.0 (9b00956e5 2024-04-29) to observe compilation failures citing deprecated features

🤔 Expected Behavior

Module should compile with latest stable rust.

😯 Current Behavior

Compilation fails

🐛 Bug Report

📝 Description

pad_and_shrink function fails, if the circuit doesn't have any tables.

🔄 Reproduction Steps

Create a circuit without any tables - and execute pad_and_shink.

🤔 Expected Behavior

It should handle circruits without lookup tables correctly.

😯 Current Behavior

Currently it fails.

📎 Log Output

boojum/src/cs/implementations/cs.rs:1042:18:
table 0 must exist when queried

Hi zksync team,

I have a problem when I run sha256_bench_non_recursive.sh in linux system:

Hello zkSync Team,

I am currently conducting research using the era-boojum proof system and have encountered an issue during the execution of the SHA-256 benchmark, the program appears to stall at the CSReferenceAssembly::wait_for_witness() function, both run_sha256_prover_non_recursive and run_sha256_prover_recursive_mode_poseidon2.