git clone https://github.com/tcsiwula/parity -b tims_benchmarks
cd parity
rustup override set nightly
Next command takes 5/10 minutes or so to compile: cargo bench -p ethcore
Should yield similar results: test views::block::tests::test_block_view ... ignored
test views::body::tests::test_block_view ... ignored
test views::header::tests::test_header_view ... ignored
test views::transaction::tests::test_transaction_view ... ignored
test result: ok. 0 passed; 0 failed; 428 ignored; 0 measured; 0 filtered out
Running target/release/deps/evm-951fe081a6b8fa68
running 20 tests
test bn_128_add ... bench: 14,738 ns/iter (+/- 6,839)
test bn_128_mul ... bench: 906,528 ns/iter (+/- 47,898)
test bn_128_pairing ... bench: 9,389,759 ns/iter (+/- 238,981)
test ecrecover ... bench: 171,418 ns/iter (+/- 6,510)
test modexp_1_pow0x10001 ... bench: 84,416 ns/iter (+/- 3,198)
test modexp_1_qube ... bench: 15,915 ns/iter (+/- 818)
test modexp_1_square ... bench: 10,918 ns/iter (+/- 362)
test modexp_2_pow0x10001 ... bench: 209,112 ns/iter (+/- 18,417)
test modexp_2_qube ... bench: 36,232 ns/iter (+/- 2,928)
test modexp_2_square ... bench: 24,709 ns/iter (+/- 4,128)
test modexp_3_pow0x10001 ... bench: 655,461 ns/iter (+/- 200,168)
test modexp_3_qube ... bench: 110,525 ns/iter (+/- 7,614)
test modexp_3_square ... bench: 74,690 ns/iter (+/- 4,092)
test modexp_4_pow0x10001 ... bench: 2,079,294 ns/iter (+/- 711,949)
test modexp_4_qube ... bench: 343,332 ns/iter (+/- 6,962)
test modexp_4_square ... bench: 232,197 ns/iter (+/- 106,911)
test modexp_5_pow0x10001 ... bench: 7,126,826 ns/iter (+/- 1,425,467)
test modexp_5_qube ... bench: 1,195,532 ns/iter (+/- 48,562)
test modexp_5_square ... bench: 792,480 ns/iter (+/- 46,469)
test sha256 ... bench: 1,826 ns/iter (+/- 203)
test result: ok. 0 passed; 0 failed; 0 ignored; 20 measured; 0 filtered out
=>
Yes, I could follow the procedure. Your machine is 10 percent faster than mine. I envy you. (Only on bn_128_add, my machine is slightly faster. Weird.)