From 78c312c983255e15fc274de2368a2ec13ce81cbf Mon Sep 17 00:00:00 2001 From: Martin Ankerl Date: Sat, 13 Jun 2020 09:37:27 +0200 Subject: 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 * Adds support for asymptotes This adds support to calculate asymptotic complexity of a benchmark. This is similar to #17375, but currently only one asymptote is supported, and I have added support in the benchmark `ComplexMemPool` as an example. Usage is e.g. like this: ``` ./bench_bitcoin -filter=ComplexMemPool -asymptote=25,50,100,200,400,600,800 ``` This runs the benchmark `ComplexMemPool` several times but with different complexityN settings. The benchmark can extract that number and use it accordingly. Here, it's used for `childTxs`. The output is this: | complexityN | ns/op | op/s | err% | ins/op | cyc/op | IPC | total | benchmark |------------:|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|----------:|:---------- | 25 | 1,064,241.00 | 939.64 | 1.4% | 3,960,279.00 | 2,829,708.00 | 1.400 | 0.01 | `ComplexMemPool` | 50 | 1,579,530.00 | 633.10 | 1.0% | 6,231,810.00 | 4,412,674.00 | 1.412 | 0.02 | `ComplexMemPool` | 100 | 4,022,774.00 | 248.58 | 0.6% | 16,544,406.00 | 11,889,535.00 | 1.392 | 0.04 | `ComplexMemPool` | 200 | 15,390,986.00 | 64.97 | 0.2% | 63,904,254.00 | 47,731,705.00 | 1.339 | 0.17 | `ComplexMemPool` | 400 | 69,394,711.00 | 14.41 | 0.1% | 272,602,461.00 | 219,014,691.00 | 1.245 | 0.76 | `ComplexMemPool` | 600 | 168,977,165.00 | 5.92 | 0.1% | 639,108,082.00 | 535,316,887.00 | 1.194 | 1.86 | `ComplexMemPool` | 800 | 310,109,077.00 | 3.22 | 0.1% |1,149,134,246.00 | 984,620,812.00 | 1.167 | 3.41 | `ComplexMemPool` | coefficient | err% | complexity |--------------:|-------:|------------ | 4.78486e-07 | 4.5% | O(n^2) | 6.38557e-10 | 21.7% | O(n^3) | 3.42338e-05 | 38.0% | O(n log n) | 0.000313914 | 46.9% | O(n) | 0.0129823 | 114.4% | O(log n) | 0.0815055 | 133.8% | O(1) The best fitting curve is O(n^2), so the algorithm seems to scale quadratic with `childTxs` in the range 25 to 800. --- src/bench/bench.cpp | 140 ++++++++++++++-------------------------------------- 1 file changed, 36 insertions(+), 104 deletions(-) (limited to 'src/bench/bench.cpp') diff --git a/src/bench/bench.cpp b/src/bench/bench.cpp index 7b93ef688d..01466d0b6f 100644 --- a/src/bench/bench.cpp +++ b/src/bench/bench.cpp @@ -8,141 +8,73 @@ #include #include -#include -#include -#include -#include -#include #include const std::function G_TEST_LOG_FUN{}; -void benchmark::ConsolePrinter::header() -{ - std::cout << "# Benchmark, evals, iterations, total, min, max, median" << std::endl; -} +namespace { -void benchmark::ConsolePrinter::result(const State& state) +void GenerateTemplateResults(const std::vector& benchmarkResults, const std::string& filename, const char* tpl) { - auto results = state.m_elapsed_results; - std::sort(results.begin(), results.end()); - - double total = state.m_num_iters * std::accumulate(results.begin(), results.end(), 0.0); - - double front = 0; - double back = 0; - double median = 0; - - if (!results.empty()) { - front = results.front(); - back = results.back(); - - size_t mid = results.size() / 2; - median = results[mid]; - if (0 == results.size() % 2) { - median = (results[mid] + results[mid + 1]) / 2; - } + if (benchmarkResults.empty() || filename.empty()) { + // nothing to write, bail out + return; } - - std::cout << std::setprecision(6); - std::cout << state.m_name << ", " << state.m_num_evals << ", " << state.m_num_iters << ", " << total << ", " << front << ", " << back << ", " << median << std::endl; -} - -void benchmark::ConsolePrinter::footer() {} -benchmark::PlotlyPrinter::PlotlyPrinter(std::string plotly_url, int64_t width, int64_t height) - : m_plotly_url(plotly_url), m_width(width), m_height(height) -{ -} - -void benchmark::PlotlyPrinter::header() -{ - std::cout << "" - << "" - << "
" - << ""; + std::cout << "Created '" << filename << "'" << std::endl; } +} // namespace benchmark::BenchRunner::BenchmarkMap& benchmark::BenchRunner::benchmarks() { - static std::map benchmarks_map; + static std::map benchmarks_map; return benchmarks_map; } -benchmark::BenchRunner::BenchRunner(std::string name, benchmark::BenchFunction func, uint64_t num_iters_for_one_second) +benchmark::BenchRunner::BenchRunner(std::string name, benchmark::BenchFunction func) { - benchmarks().insert(std::make_pair(name, Bench{func, num_iters_for_one_second})); + benchmarks().insert(std::make_pair(name, func)); } -void benchmark::BenchRunner::RunAll(Printer& printer, uint64_t num_evals, double scaling, const std::string& filter, bool is_list_only) +void benchmark::BenchRunner::RunAll(const Args& args) { - if (!std::ratio_less_equal::value) { - std::cerr << "WARNING: Clock precision is worse than microsecond - benchmarks may be less accurate!\n"; - } -#ifdef DEBUG - std::cerr << "WARNING: This is a debug build - may result in slower benchmarks.\n"; -#endif - - std::regex reFilter(filter); + std::regex reFilter(args.regex_filter); std::smatch baseMatch; - printer.header(); - + std::vector benchmarkResults; for (const auto& p : benchmarks()) { if (!std::regex_match(p.first, baseMatch, reFilter)) { continue; } - uint64_t num_iters = static_cast(p.second.num_iters_for_one_second * scaling); - if (0 == num_iters) { - num_iters = 1; - } - State state(p.first, num_evals, num_iters, printer); - if (!is_list_only) { - p.second.func(state); + if (args.is_list_only) { + std::cout << p.first << std::endl; + continue; } - printer.result(state); - } - - printer.footer(); -} - -bool benchmark::State::UpdateTimer(const benchmark::time_point current_time) -{ - if (m_start_time != time_point()) { - std::chrono::duration diff = current_time - m_start_time; - m_elapsed_results.push_back(diff.count() / m_num_iters); - if (m_elapsed_results.size() == m_num_evals) { - return false; + Bench bench; + bench.name(p.first); + if (args.asymptote.empty()) { + p.second(bench); + } else { + for (auto n : args.asymptote) { + bench.complexityN(n); + p.second(bench); + } + std::cout << bench.complexityBigO() << std::endl; } + benchmarkResults.push_back(bench.results().back()); } - m_num_iters_left = m_num_iters - 1; - return true; + GenerateTemplateResults(benchmarkResults, args.output_csv, "# Benchmark, evals, iterations, total, min, max, median\n" + "{{#result}}{{name}}, {{epochs}}, {{average(iterations)}}, {{sumProduct(iterations, elapsed)}}, {{minimum(elapsed)}}, {{maximum(elapsed)}}, {{median(elapsed)}}\n" + "{{/result}}"); + GenerateTemplateResults(benchmarkResults, args.output_json, ankerl::nanobench::templates::json()); } -- cgit v1.2.3