Add thread local destroy benchmark

Summary: Add thread local destroy benchmark

Reviewed By: yfeldblum

Differential Revision: D7892479

fbshipit-source-id: f16462820354a6cf27d9348a76b28be027cef20c

folly/test/ThreadLocalDestroyBenchmark.cpp

+/*
+ * Copyright 2018-present Facebook, Inc.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include <folly/Benchmark.h>
+#include <folly/ThreadLocal.h>
+#include <condition_variable>
+#include <mutex>
+#include <thread>
+
+using namespace folly;
+
+template <typename Tag1, typename Tag2>
+void runTestTag(int iters, int numThreads) {
+  BenchmarkSuspender susp;
+
+  ThreadLocalPtr<int, Tag1> t1;
+
+  t1.reset(new int(1));
+
+  std::mutex m, mw;
+  std::condition_variable cv, cvw;
+  bool running = true;
+  int numRunning = 0;
+
+  std::vector<std::thread> threads;
+  for (int i = 0; i < numThreads; i++) {
+    threads.push_back(std::thread([&]() mutable {
+      t1.reset(new int(1));
+
+      // notify if all the threads have created the t1
+      bool notify = false;
+      {
+        std::lock_guard<std::mutex> lk(m);
+        if (++numRunning == numThreads) {
+          notify = true;
+        }
+      }
+
+      if (notify) {
+        cv.notify_one();
+      }
+
+      // now wait
+      {
+        std::unique_lock<std::mutex> lk(mw);
+        cvw.wait(lk, [&]() { return !running; });
+      }
+    }));
+  }
+
+  // wait for the threads to create the t1
+  {
+    std::unique_lock<std::mutex> lk(m);
+    cv.wait(lk, [&]() { return numRunning == numThreads; });
+  }
+
+  ThreadLocalPtr<int, Tag2> t3;
+  t3.reset(new int(2));
+
+  susp.dismiss();
+
+  // run the test loop
+  for (int i = 0; i < iters; i++) {
+    ThreadLocalPtr<int, Tag2> t2;
+    t2.reset(new int(2));
+  }
+
+  susp.rehire();
+
+  {
+    std::lock_guard<std::mutex> lk(mw);
+    running = false;
+  }
+
+  cvw.notify_all();
+
+  for (auto& t : threads) {
+    t.join();
+  }
+}
+
+void runTestSameTag(int iters, int numThreads) {
+  runTestTag<void, void>(iters, numThreads);
+}
+
+void runTestDiffTag(int iters, int numThreads) {
+  runTestTag<void, int>(iters, numThreads);
+}
+
+BENCHMARK_DRAW_LINE();
+BENCHMARK_PARAM(runTestSameTag, 1)
+BENCHMARK_PARAM(runTestSameTag, 128)
+BENCHMARK_PARAM(runTestSameTag, 256)
+BENCHMARK_PARAM(runTestSameTag, 512)
+BENCHMARK_PARAM(runTestSameTag, 1024)
+BENCHMARK_PARAM(runTestSameTag, 2048)
+BENCHMARK_DRAW_LINE();
+BENCHMARK_PARAM(runTestDiffTag, 1)
+BENCHMARK_PARAM(runTestDiffTag, 128)
+BENCHMARK_PARAM(runTestDiffTag, 256)
+BENCHMARK_PARAM(runTestDiffTag, 512)
+BENCHMARK_PARAM(runTestDiffTag, 1024)
+BENCHMARK_PARAM(runTestDiffTag, 2048)
+BENCHMARK_DRAW_LINE();
+
+int main(int argc, char* argv[]) {
+  gflags::ParseCommandLineFlags(&argc, &argv, true);
+  folly::runBenchmarks();
+
+  return 0;
+}