Benchmark Runner

Set up and run criterion benchmarks for an Aion crate. The crate name is provided as $ARGUMENTS (without the aion_ prefix, e.g., /bench sim).

Steps

1. •

   Validate the crate exists:

   bash

   ls crates/aion_<name>/Cargo.toml
   

2. •

   Check if benchmarks already exist:

   bash

   ls crates/aion_<name>/benches/ 2>/dev/null
   

3. •

   If no benchmarks exist, scaffold them:

   a. Add criterion to Cargo.toml:

   toml

   [dev-dependencies]
   criterion = { version = "0.5", features = ["html_reports"] }
   
   [[bench]]
   name = "<name>_bench"
   harness = false
   

   b. Create benches/<name>_bench.rs:

   rust

   use criterion::{criterion_group, criterion_main, Criterion};
   
   fn bench_<relevant_operation>(c: &mut Criterion) {
       c.bench_function("<operation>", |b| {
           b.iter(|| {
               // TODO: Add benchmark body
           });
       });
   }
   
   criterion_group!(benches, bench_<relevant_operation>);
   criterion_main!(benches);
   

   c. Choose benchmarks based on crate purpose:

   • •Parsers (vhdl_parser, verilog_parser, sv_parser): Parse a representative source file
   • •sim: Simulate N delta cycles on a standard testbench
   • •elaborate: Elaborate a multi-module design
   • •synth: Run optimization passes on a netlist
   • •pnr: Placement + routing on a small design
   • •timing: STA on a timing graph with N nodes
   • •bitstream: Generate bitstream for a configured design
4. •

   Run the benchmarks:

   bash

   cd "$CLAUDE_PROJECT_DIR" && cargo bench -p aion_<name>
   

5. •

   Report results in this format:

## Benchmark Results: aion_<name>

### Benchmarks Run
- <benchmark_name>: <time> (± <std_dev>)

### Comparison (if baseline exists)
- <benchmark_name>: <change%> vs previous

### HTML Report
- Open `target/criterion/<group>/report/index.html` for detailed analysis

Notes

• •Do NOT modify existing benchmarks without asking first.
• •If benchmarks already exist, just run them and report results.
• •Use criterion::black_box() to prevent dead code elimination.
• •Keep benchmark inputs realistic — use patterns from the conformance test suite where possible.