Benchmark prover

[recmo]

No description provided.

[codspeed-hq]

CodSpeed Performance Report

Merging #46 will improve performances by 18.17%

_{Comparing remco/bench-prover (6d32046) with main (0db3025)}

Summary

⚡ 1 improvements
✅ 25 untouched benchmarks
🆕 4 new benchmarks
⁉️ 4 dropped benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
🆕	prove_poseidon_1000	N/A	10.3 s	N/A
🆕	read_poseidon_1000	N/A	118.7 ms	N/A
🆕	verify_poseidon_1000	N/A	19.3 ms	N/A
⁉️	bits_05	589.1 µs	N/A	N/A
⁉️	bits_10	1.6 ms	N/A	N/A
⁉️	bits_15	22.4 ms	N/A	N/A
⁉️	bits_20	2.6 s	N/A	N/A
🆕	bits_20_fixed	N/A	14.2 s	N/A
⚡	reduce	189.7 ns	160.6 ns	+18.17%

[recmo]

The PoW bits benchmark is too noisy. Maybe replace with deterministic challenges.

[codspeed-hq]

CodSpeed Walltime Performance Report

Congrats! CodSpeed is installed 🎉

🆕 30 new benchmarks were detected.

You will start to see performance impacts in the reports once the benchmarks are run from your default branch.

Detected benchmarks

• ark_ff (3.2 µs)
• block_mul (2.1 µs)
• montgomery_interleaved_3 (1.9 µs)
• montgomery_interleaved_4 (3.1 µs)
• scalar_mul (3.1 µs)
• simd_mul (1.8 µs)
• ark_ff (2.6 µs)
• block_sqr (2.9 µs)
• scalar_sqr (2.5 µs)
• simd_sqr (2.7 µs)
• wrm_overhead (1.7 µs)
• prove_poseidon_1000 (3.6 s)
• read_poseidon_1000 (42.5 ms)
• verify_poseidon_1000 (13.8 ms)
• block3 (188.7 µs)
• block4 (232.7 µs)
• reference (1.3 ms)
• simple (382.3 µs)
• v1 (181.6 µs)
• bar (2.8 µs)
• sbox (1.8 µs)
• sbox_16 (3.3 µs)
• sbox_8 (2.5 µs)
• square (2.5 µs)
• bits_20_fixed (221.6 ms)
• reduce (2.6 µs)
• reduce_1 (2.5 µs)
• reduce_1_partial (3 µs)
• reduce_add_rc (2.9 µs)
• reduce_partial (2.5 µs)