diff options
author | Wladimir J. van der Laan <laanwj@protonmail.com> | 2020-07-30 15:20:19 +0200 |
---|---|---|
committer | Wladimir J. van der Laan <laanwj@protonmail.com> | 2020-07-30 15:34:17 +0200 |
commit | 4ebe2f6e752c453ff572eda4a108e747d6586c97 (patch) | |
tree | 59fdc087589ab55e8ade08e91fadeb150bd9733f /src/bench/chacha_poly_aead.cpp | |
parent | 2a784723f0c0e642353dc74ec6aef4d5f8345044 (diff) | |
parent | 78c312c983255e15fc274de2368a2ec13ce81cbf (diff) | |
download | bitcoin-4ebe2f6e752c453ff572eda4a108e747d6586c97.tar.xz |
Merge #18011: Replace current benchmarking framework with nanobench
78c312c983255e15fc274de2368a2ec13ce81cbf Replace current benchmarking framework with nanobench (Martin Ankerl)
Pull request description:
Replace current benchmarking framework with nanobench
This replaces the current benchmarking framework with nanobench [1], an
MIT licensed single-header benchmarking library, of which I am the
autor. This has in my opinion several advantages, especially on Linux:
* fast: Running all benchmarks takes ~6 seconds instead of 4m13s on
an Intel i7-8700 CPU @ 3.20GHz.
* accurate: I ran e.g. the benchmark for SipHash_32b 10 times and
calculate standard deviation / mean = coefficient of variation:
* 0.57% CV for old benchmarking framework
* 0.20% CV for nanobench
So the benchmark results with nanobench seem to vary less than with
the old framework.
* It automatically determines runtime based on clock precision, no need
to specify number of evaluations.
* measure instructions, cycles, branches, instructions per cycle,
branch misses (only Linux, when performance counters are available)
* output in markdown table format.
* Warn about unstable environment (frequency scaling, turbo, ...)
* For better profiling, it is possible to set the environment variable
NANOBENCH_ENDLESS to force endless running of a particular benchmark
without the need to recompile. This makes it to e.g. run "perf top"
and look at hotspots.
Here is an example copy & pasted from the terminal output:
| ns/byte | byte/s | err% | ins/byte | cyc/byte | IPC | bra/byte | miss% | total | benchmark
|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|---------------:|--------:|----------:|:----------
| 2.52 | 396,529,415.94 | 0.6% | 25.42 | 8.02 | 3.169 | 0.06 | 0.0% | 0.03 | `bench/crypto_hash.cpp RIPEMD160`
| 1.87 | 535,161,444.83 | 0.3% | 21.36 | 5.95 | 3.589 | 0.06 | 0.0% | 0.02 | `bench/crypto_hash.cpp SHA1`
| 3.22 | 310,344,174.79 | 1.1% | 36.80 | 10.22 | 3.601 | 0.09 | 0.0% | 0.04 | `bench/crypto_hash.cpp SHA256`
| 2.01 | 496,375,796.23 | 0.0% | 18.72 | 6.43 | 2.911 | 0.01 | 1.0% | 0.00 | `bench/crypto_hash.cpp SHA256D64_1024`
| 7.23 | 138,263,519.35 | 0.1% | 82.66 | 23.11 | 3.577 | 1.63 | 0.1% | 0.00 | `bench/crypto_hash.cpp SHA256_32b`
| 3.04 | 328,780,166.40 | 0.3% | 35.82 | 9.69 | 3.696 | 0.03 | 0.0% | 0.03 | `bench/crypto_hash.cpp SHA512`
[1] https://github.com/martinus/nanobench
ACKs for top commit:
laanwj:
ACK 78c312c983255e15fc274de2368a2ec13ce81cbf
Tree-SHA512: 9e18770b18b6f95a7d0105a4a5497d31cf4eb5efe6574f4482f6f1b4c88d7e0946b9a4a1e9e8e6ecbf41a3f2d7571240677dcb45af29a6f0584e89b25f32e49e
Diffstat (limited to 'src/bench/chacha_poly_aead.cpp')
-rw-r--r-- | src/bench/chacha_poly_aead.cpp | 65 |
1 files changed, 33 insertions, 32 deletions
diff --git a/src/bench/chacha_poly_aead.cpp b/src/bench/chacha_poly_aead.cpp index df10f27d03..30d7851b7f 100644 --- a/src/bench/chacha_poly_aead.cpp +++ b/src/bench/chacha_poly_aead.cpp @@ -21,7 +21,7 @@ static const unsigned char k2[32] = {0}; static ChaCha20Poly1305AEAD aead(k1, 32, k2, 32); -static void CHACHA20_POLY1305_AEAD(benchmark::State& state, size_t buffersize, bool include_decryption) +static void CHACHA20_POLY1305_AEAD(benchmark::Bench& bench, size_t buffersize, bool include_decryption) { std::vector<unsigned char> in(buffersize + CHACHA20_POLY1305_AEAD_AAD_LEN + POLY1305_TAGLEN, 0); std::vector<unsigned char> out(buffersize + CHACHA20_POLY1305_AEAD_AAD_LEN + POLY1305_TAGLEN, 0); @@ -29,7 +29,7 @@ static void CHACHA20_POLY1305_AEAD(benchmark::State& state, size_t buffersize, b uint64_t seqnr_aad = 0; int aad_pos = 0; uint32_t len = 0; - while (state.KeepRunning()) { + bench.batch(buffersize).unit("byte").run([&] { // encrypt or decrypt the buffer with a static key assert(aead.Crypt(seqnr_payload, seqnr_aad, aad_pos, out.data(), out.size(), in.data(), buffersize, true)); @@ -53,70 +53,71 @@ static void CHACHA20_POLY1305_AEAD(benchmark::State& state, size_t buffersize, b seqnr_aad = 0; aad_pos = 0; } - } + }); } -static void CHACHA20_POLY1305_AEAD_64BYTES_ONLY_ENCRYPT(benchmark::State& state) +static void CHACHA20_POLY1305_AEAD_64BYTES_ONLY_ENCRYPT(benchmark::Bench& bench) { - CHACHA20_POLY1305_AEAD(state, BUFFER_SIZE_TINY, false); + CHACHA20_POLY1305_AEAD(bench, BUFFER_SIZE_TINY, false); } -static void CHACHA20_POLY1305_AEAD_256BYTES_ONLY_ENCRYPT(benchmark::State& state) +static void CHACHA20_POLY1305_AEAD_256BYTES_ONLY_ENCRYPT(benchmark::Bench& bench) { - CHACHA20_POLY1305_AEAD(state, BUFFER_SIZE_SMALL, false); + CHACHA20_POLY1305_AEAD(bench, BUFFER_SIZE_SMALL, false); } -static void CHACHA20_POLY1305_AEAD_1MB_ONLY_ENCRYPT(benchmark::State& state) +static void CHACHA20_POLY1305_AEAD_1MB_ONLY_ENCRYPT(benchmark::Bench& bench) { - CHACHA20_POLY1305_AEAD(state, BUFFER_SIZE_LARGE, false); + CHACHA20_POLY1305_AEAD(bench, BUFFER_SIZE_LARGE, false); } -static void CHACHA20_POLY1305_AEAD_64BYTES_ENCRYPT_DECRYPT(benchmark::State& state) +static void CHACHA20_POLY1305_AEAD_64BYTES_ENCRYPT_DECRYPT(benchmark::Bench& bench) { - CHACHA20_POLY1305_AEAD(state, BUFFER_SIZE_TINY, true); + CHACHA20_POLY1305_AEAD(bench, BUFFER_SIZE_TINY, true); } -static void CHACHA20_POLY1305_AEAD_256BYTES_ENCRYPT_DECRYPT(benchmark::State& state) +static void CHACHA20_POLY1305_AEAD_256BYTES_ENCRYPT_DECRYPT(benchmark::Bench& bench) { - CHACHA20_POLY1305_AEAD(state, BUFFER_SIZE_SMALL, true); + CHACHA20_POLY1305_AEAD(bench, BUFFER_SIZE_SMALL, true); } -static void CHACHA20_POLY1305_AEAD_1MB_ENCRYPT_DECRYPT(benchmark::State& state) +static void CHACHA20_POLY1305_AEAD_1MB_ENCRYPT_DECRYPT(benchmark::Bench& bench) { - CHACHA20_POLY1305_AEAD(state, BUFFER_SIZE_LARGE, true); + CHACHA20_POLY1305_AEAD(bench, BUFFER_SIZE_LARGE, true); } // Add Hash() (dbl-sha256) bench for comparison -static void HASH(benchmark::State& state, size_t buffersize) +static void HASH(benchmark::Bench& bench, size_t buffersize) { uint8_t hash[CHash256::OUTPUT_SIZE]; std::vector<uint8_t> in(buffersize,0); - while (state.KeepRunning()) + bench.batch(in.size()).unit("byte").run([&] { CHash256().Write(in.data(), in.size()).Finalize(hash); + }); } -static void HASH_64BYTES(benchmark::State& state) +static void HASH_64BYTES(benchmark::Bench& bench) { - HASH(state, BUFFER_SIZE_TINY); + HASH(bench, BUFFER_SIZE_TINY); } -static void HASH_256BYTES(benchmark::State& state) +static void HASH_256BYTES(benchmark::Bench& bench) { - HASH(state, BUFFER_SIZE_SMALL); + HASH(bench, BUFFER_SIZE_SMALL); } -static void HASH_1MB(benchmark::State& state) +static void HASH_1MB(benchmark::Bench& bench) { - HASH(state, BUFFER_SIZE_LARGE); + HASH(bench, BUFFER_SIZE_LARGE); } -BENCHMARK(CHACHA20_POLY1305_AEAD_64BYTES_ONLY_ENCRYPT, 500000); -BENCHMARK(CHACHA20_POLY1305_AEAD_256BYTES_ONLY_ENCRYPT, 250000); -BENCHMARK(CHACHA20_POLY1305_AEAD_1MB_ONLY_ENCRYPT, 340); -BENCHMARK(CHACHA20_POLY1305_AEAD_64BYTES_ENCRYPT_DECRYPT, 500000); -BENCHMARK(CHACHA20_POLY1305_AEAD_256BYTES_ENCRYPT_DECRYPT, 250000); -BENCHMARK(CHACHA20_POLY1305_AEAD_1MB_ENCRYPT_DECRYPT, 340); -BENCHMARK(HASH_64BYTES, 500000); -BENCHMARK(HASH_256BYTES, 250000); -BENCHMARK(HASH_1MB, 340); +BENCHMARK(CHACHA20_POLY1305_AEAD_64BYTES_ONLY_ENCRYPT); +BENCHMARK(CHACHA20_POLY1305_AEAD_256BYTES_ONLY_ENCRYPT); +BENCHMARK(CHACHA20_POLY1305_AEAD_1MB_ONLY_ENCRYPT); +BENCHMARK(CHACHA20_POLY1305_AEAD_64BYTES_ENCRYPT_DECRYPT); +BENCHMARK(CHACHA20_POLY1305_AEAD_256BYTES_ENCRYPT_DECRYPT); +BENCHMARK(CHACHA20_POLY1305_AEAD_1MB_ENCRYPT_DECRYPT); +BENCHMARK(HASH_64BYTES); +BENCHMARK(HASH_256BYTES); +BENCHMARK(HASH_1MB); |