| 1 | /* |
| 2 | * Copyright (C) 2014 Apple Inc. All rights reserved. |
| 3 | * |
| 4 | * Redistribution and use in source and binary forms, with or without |
| 5 | * modification, are permitted provided that the following conditions |
| 6 | * are met: |
| 7 | * 1. Redistributions of source code must retain the above copyright |
| 8 | * notice, this list of conditions and the following disclaimer. |
| 9 | * 2. Redistributions in binary form must reproduce the above copyright |
| 10 | * notice, this list of conditions and the following disclaimer in the |
| 11 | * documentation and/or other materials provided with the distribution. |
| 12 | * |
| 13 | * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY |
| 14 | * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| 15 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR |
| 16 | * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR |
| 17 | * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, |
| 18 | * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, |
| 19 | * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR |
| 20 | * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY |
| 21 | * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
| 22 | * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
| 23 | * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| 24 | */ |
| 25 | |
| 26 | #include "Benchmark.h" |
| 27 | #include "CPUCount.h" |
| 28 | #include "alloc_free.h" |
| 29 | #include "balloon.h" |
| 30 | #include "big.h" |
| 31 | #include "churn.h" |
| 32 | #include "facebook.h" |
| 33 | #include "flickr.h" |
| 34 | #include "fragment.h" |
| 35 | #include "list.h" |
| 36 | #include "medium.h" |
| 37 | #include "message.h" |
| 38 | #include "nimlang.h" |
| 39 | #include "reddit.h" |
| 40 | #include "realloc.h" |
| 41 | #include "stress.h" |
| 42 | #include "stress_aligned.h" |
| 43 | #include "theverge.h" |
| 44 | #include "tree.h" |
| 45 | #include <algorithm> |
| 46 | #include <iostream> |
| 47 | #include <map> |
| 48 | #include <stdio.h> |
| 49 | #include <string> |
| 50 | #include <strings.h> |
| 51 | #include <sys/time.h> |
| 52 | #include <thread> |
| 53 | #include <vector> |
| 54 | |
| 55 | #ifdef __APPLE__ |
| 56 | #include <dispatch/dispatch.h> |
| 57 | #endif |
| 58 | |
| 59 | #include "mbmalloc.h" |
| 60 | |
| 61 | using namespace std; |
| 62 | |
| 63 | struct BenchmarkPair { |
| 64 | const char* const name; |
| 65 | const BenchmarkFunction function; |
| 66 | }; |
| 67 | |
| 68 | static const BenchmarkPair benchmarkPairs[] = { |
| 69 | { "alloc_free" , benchmark_alloc_free }, |
| 70 | { "balloon" , benchmark_balloon }, |
| 71 | { "big" , benchmark_big }, |
| 72 | { "churn" , benchmark_churn }, |
| 73 | { "facebook" , benchmark_facebook }, |
| 74 | { "flickr" , benchmark_flickr }, |
| 75 | { "flickr_memory_warning" , benchmark_flickr_memory_warning }, |
| 76 | { "fragment" , benchmark_fragment }, |
| 77 | { "fragment_iterate" , benchmark_fragment_iterate }, |
| 78 | { "list_allocate" , benchmark_list_allocate }, |
| 79 | { "list_traverse" , benchmark_list_traverse }, |
| 80 | { "medium" , benchmark_medium }, |
| 81 | { "message_many" , benchmark_message_many }, |
| 82 | { "message_one" , benchmark_message_one }, |
| 83 | { "nimlang" , benchmark_nimlang }, |
| 84 | { "realloc" , benchmark_realloc }, |
| 85 | { "reddit" , benchmark_reddit }, |
| 86 | { "reddit_memory_warning" , benchmark_reddit_memory_warning }, |
| 87 | { "stress" , benchmark_stress }, |
| 88 | { "stress_aligned" , benchmark_stress_aligned }, |
| 89 | { "theverge" , benchmark_theverge }, |
| 90 | { "theverge_memory_warning" , benchmark_theverge_memory_warning }, |
| 91 | { "tree_allocate" , benchmark_tree_allocate }, |
| 92 | { "tree_churn" , benchmark_tree_churn }, |
| 93 | { "tree_traverse" , benchmark_tree_traverse }, |
| 94 | }; |
| 95 | |
| 96 | static const size_t benchmarksPairsCount = sizeof(benchmarkPairs) / sizeof(BenchmarkPair); |
| 97 | |
| 98 | static inline bool operator==(const BenchmarkPair& benchmarkPair, const string& string) |
| 99 | { |
| 100 | return string == benchmarkPair.name; |
| 101 | } |
| 102 | |
| 103 | static void*** allocateHeap(size_t heapSize, size_t chunkSize, size_t objectSize) |
| 104 | { |
| 105 | if (!heapSize) |
| 106 | return 0; |
| 107 | |
| 108 | size_t chunkCount = heapSize / chunkSize; |
| 109 | size_t objectCount = chunkSize / objectSize; |
| 110 | void*** chunks = (void***)mbmalloc(chunkCount * sizeof(void**)); |
| 111 | for (size_t i = 0; i < chunkCount; ++i) { |
| 112 | chunks[i] = (void**)mbmalloc(objectCount * sizeof(void*)); |
| 113 | for (size_t j = 0; j < objectCount; ++j) { |
| 114 | chunks[i][j] = (void*)mbmalloc(objectSize); |
| 115 | bzero(chunks[i][j], objectSize); |
| 116 | } |
| 117 | } |
| 118 | return chunks; |
| 119 | } |
| 120 | |
| 121 | static void deallocateHeap(void*** chunks, size_t heapSize, size_t chunkSize, size_t objectSize) |
| 122 | { |
| 123 | if (!heapSize) |
| 124 | return; |
| 125 | |
| 126 | size_t chunkCount = heapSize / chunkSize; |
| 127 | size_t objectCount = chunkSize / objectSize; |
| 128 | for (size_t i = 0; i < chunkCount; ++i) { |
| 129 | for (size_t j = 0; j < objectCount; ++j) |
| 130 | mbfree(chunks[i][j], objectSize); |
| 131 | mbfree(chunks[i], objectCount * sizeof(void*)); |
| 132 | } |
| 133 | mbfree(chunks, chunkCount * sizeof(void**)); |
| 134 | } |
| 135 | |
| 136 | Benchmark::Benchmark(CommandLine& commandLine) |
| 137 | : m_commandLine(commandLine) |
| 138 | { |
| 139 | const BenchmarkPair* benchmarkPair = std::find( |
| 140 | benchmarkPairs, benchmarkPairs + benchmarksPairsCount, m_commandLine.benchmarkName()); |
| 141 | if (benchmarkPair == benchmarkPairs + benchmarksPairsCount) |
| 142 | return; |
| 143 | |
| 144 | m_benchmarkPair = benchmarkPair; |
| 145 | } |
| 146 | |
| 147 | void Benchmark::printBenchmarks() |
| 148 | { |
| 149 | cout << "Benchmarks: " << endl; |
| 150 | for (size_t i = 0; i < benchmarksPairsCount; ++i) |
| 151 | cout << "\t" << benchmarkPairs[i].name << endl; |
| 152 | } |
| 153 | |
| 154 | void Benchmark::runOnce() |
| 155 | { |
| 156 | if (!m_commandLine.isParallel()) { |
| 157 | m_benchmarkPair->function(m_commandLine); |
| 158 | return; |
| 159 | } |
| 160 | |
| 161 | #ifdef __APPLE__ |
| 162 | dispatch_group_t group = dispatch_group_create(); |
| 163 | |
| 164 | for (size_t i = 0; i < cpuCount(); ++i) { |
| 165 | dispatch_group_async(group, dispatch_get_global_queue(DISPATCH_QUEUE_PRIORITY_HIGH, 0), ^{ |
| 166 | m_benchmarkPair->function(m_commandLine); |
| 167 | }); |
| 168 | } |
| 169 | |
| 170 | dispatch_group_wait(group, DISPATCH_TIME_FOREVER); |
| 171 | |
| 172 | dispatch_release(group); |
| 173 | #else |
| 174 | std::vector<std::thread> threads; |
| 175 | for (size_t i = 0; i < cpuCount(); ++i) { |
| 176 | threads.emplace_back([&] { |
| 177 | m_benchmarkPair->function(m_commandLine); |
| 178 | }); |
| 179 | } |
| 180 | for (auto& thread : threads) |
| 181 | thread.join(); |
| 182 | #endif |
| 183 | } |
| 184 | |
| 185 | void Benchmark::run() |
| 186 | { |
| 187 | static const size_t objectSize = 32; |
| 188 | static const size_t chunkSize = 1024 * 1024; |
| 189 | |
| 190 | void*** heap = allocateHeap(m_commandLine.heapSize(), chunkSize, objectSize); |
| 191 | |
| 192 | if (m_commandLine.warmUp()) |
| 193 | runOnce(); // Warmup run. |
| 194 | |
| 195 | size_t runs = m_commandLine.runs(); |
| 196 | |
| 197 | for (size_t i = 0; i < runs; ++i) { |
| 198 | double start = currentTimeMS(); |
| 199 | runOnce(); |
| 200 | double end = currentTimeMS(); |
| 201 | double elapsed = end - start; |
| 202 | m_elapsedTime += elapsed; |
| 203 | } |
| 204 | m_elapsedTime /= runs; |
| 205 | |
| 206 | deallocateHeap(heap, m_commandLine.heapSize(), chunkSize, objectSize); |
| 207 | |
| 208 | mbscavenge(); |
| 209 | m_memory = currentMemoryBytes(); |
| 210 | } |
| 211 | |
| 212 | void Benchmark::printReport() |
| 213 | { |
| 214 | size_t kB = 1024; |
| 215 | |
| 216 | cout << "Time: \t" << m_elapsedTime << "ms" << endl; |
| 217 | cout << "Peak Memory:\t" << m_memory.residentMax / kB << "kB" << endl; |
| 218 | cout << "Memory at End: \t" << m_memory.resident / kB << "kB" << endl; |
| 219 | } |
| 220 | |
| 221 | double Benchmark::currentTimeMS() |
| 222 | { |
| 223 | struct timeval now; |
| 224 | gettimeofday(&now, 0); |
| 225 | return (now.tv_sec * 1000.0) + now.tv_usec / 1000.0; |
| 226 | } |
| 227 | |
| 228 | |