Open Issues Need Help
View All on GitHubAI Summary: This issue proposes optimizing Keccak-256 performance by implementing ARM NEON SIMD intrinsics. The goal is to widen the performance lead over `alloy.rs`, particularly for larger inputs (1kb-4kb) where the current advantage diminishes. Keccak-256 is a critical hot path for various blockchain operations like transaction hashing and address derivation.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
AI Summary: The issue aims to optimize RLP encoding for EIP-1559 transactions in Zig, which is currently 1.08x slower than a Rust counterpart. The performance gap stems from runtime dispatch for field types during encoding, even when using comptime reflection. The proposed solution is to generate specialized, struct-specific RLP encoders at comptime to eliminate this runtime dispatch and achieve performance parity.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
AI Summary: This GitHub issue proposes adding an optional variable-time secp256k1 signing path to `eth.zig` to significantly improve performance, aiming to close a 4.09x gap against `alloy/k256-rs`. The current implementation uses slower constant-time operations for security, while the proposed variable-time approach leverages precomputed tables for speed, with the trade-off of potential timing side-channels that need to be considered for security-sensitive applications.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
AI Summary: This GitHub issue proposes optimizing ABI encoding in Zig by generating specialized, monomorphic encode functions at compile time for known function signatures. The current implementation suffers a performance loss compared to Alloy's `sol!` macro for specific cases (e.g., `transfer`) due to runtime union dispatch on `AbiValue` arguments. The proposed solution aims to eliminate this runtime overhead by leveraging Zig's `comptime` features to create highly optimized, type-specific encoders.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
AI Summary: This GitHub issue proposes optimizing `u256` compound arithmetic in Zig, specifically targeting a 3.31x performance gap in the `u256_uniswapv2_amount_out` benchmark compared to a Rust implementation. The root cause is Zig's generic `u256` operations losing context in compound calculations, unlike Rust's hand-optimized limb-based approach. The solution involves implementing explicit 4x `u64` limb-based multiplication and division for these `u256` compound operations.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.
Pure Zig Ethereum client library. Zero dependencies. Faster than alloy.rs on 18/24 benchmarks.