🥅 netlistx-rs

netlistx-rs is a Rust library for working with netlists (hypergraphs) in electronic design automation (EDA). It provides efficient data structures and algorithms for representing, analyzing, and partitioning circuit netlists.

📖 What is a Netlist?

A netlist is a representation of an electronic circuit that describes the connections between components. In EDA, netlists are typically modeled as hypergraphs where:

• Modules (or cells) represent the circuit components
• Nets represent the wires that connect multiple modules together
• Each net can connect multiple modules, making it a hyperedge

✨ Features

• Efficient Data Structures: Fast graph-based representation using petgraph
• Hypergraph Support: True hypergraph semantics for multi-terminal nets
• Partitioning Algorithms: Kernighan-Lin and Fiduccia-Mattheyses partitioning
• Connectivity Analysis: Compute connectivity metrics and analyze circuit structure
• Statistics: Comprehensive netlist statistics and metrics
• Serde Support: Optional serialization/deserialization via serde feature
• Parallel Processing: Optional parallel operations via rayon feature

🚀 Installation

Cargo

Add netlistx-rs to your Cargo.toml:

[dependencies]
netlistx-rs = "0.1"

For optional features:

[dependencies]
netlistx-rs = { version = "0.1", features = ["serde", "rayon"] }

💡 Usage Examples

Basic Netlist Creation

use netlistx_rs::netlist::Netlist;

// Create a new netlist
let mut netlist = Netlist::new();

// Add modules
netlist.add_module("cell_a".to_string());
netlist.add_module("cell_b".to_string());
netlist.add_module("cell_c".to_string());

// Add nets
netlist.add_net("net1".to_string());
netlist.add_net("net2".to_string());

// Connect modules to nets
netlist.add_edge("net1", "cell_a").unwrap();
netlist.add_edge("net1", "cell_b").unwrap();
netlist.add_edge("net2", "cell_b").unwrap();
netlist.add_edge("net2", "cell_c").unwrap();

println!("Modules: {}", netlist.num_modules);
println!("Nets: {}", netlist.num_nets);

Using the Builder Pattern

use netlistx_rs::netlist::NetlistBuilder;

let netlist = NetlistBuilder::new()
    .add_module("cell_a")
    .add_module("cell_b")
    .add_module("cell_c")
    .add_net("net1")
    .add_net("net2")
    .add_edge("net1", "cell_a")
    .add_edge("net1", "cell_b")
    .add_edge("net2", "cell_b")
    .add_edge("net2", "cell_c")
    .build()
    .unwrap();

Netlist Partitioning

use netlistx_rs::partitioning::FiducciaMattheyses;

let partitioner = FiducciaMattheyses::new();
let result = partitioner.partition(&netlist, 0.5); // 50% balance

Statistics and Metrics

use netlistx_rs::statistics::NetlistStats;

let stats = NetlistStats::analyze(&netlist);
println!("Average net degree: {}", stats.avg_net_degree());
println!("Maximum module degree: {}", stats.max_module_degree());

📚 API Documentation

Full API documentation is available at docs.rs/netlistx-rs.

Main Modules

• netlist: Core netlist data structure and operations
• partitioning: Circuit partitioning algorithms
• statistics: Netlist statistics and metrics
• io: File I/O for various netlist formats

🔧 Optional Features

• serde: Enable serialization/deserialization support
• rayon: Enable parallel processing for large netlists

🗺️ Roadmap

• Additional file format support (Verilog, ISPD, LEF/DEF)
• More partitioning algorithms (Multilevel partitioning)
• Placement algorithms
• Timing analysis utilities
• Power analysis tools
• Python bindings via PyO3

🤝 Contributing

Contributions are welcome! Please see CONTRIBUTING.md for details.

📄 License

Licensed under either of

• Apache License, Version 2.0 (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
• MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)

at your option.

License

Licensed under either of

• Apache License, Version 2.0 (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
• MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)

at your option.

Contribution

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.

See CONTRIBUTING.md.