JVM源码分析之Hotspot VM中的Thread创建

本文首发于公众号：javaadu

在面试候选人的时候，我有时候会提出这样的一个问题：说说你对Java线程的理解？从这个问题开始，可以考察候选人对并发基础的掌握、对操作系统基本概念的理解，如果遇到对底层有浓厚兴趣的同学，我会抛出另一个问题：Java里的线程和操作系统的线程是什么关系？它们是如何对应的？这两个问题，就是今天这篇文章想讲述的。

基础知识

JVM中的线程是和OS中的线程一一对应的，操作系统负责调度所有的线程，因此在不同的平台上，Java线程的优先级有所不同。

在JVM中除了应用线程，还有其他的一些线程用于支持JVM的运行，这些线程可以被划分为以下几类：

VM Thread：负责JVM在安全点内的各种操作，这些操作（诸如自动内存管理、取消偏向锁、线程dump、线程挂起等等）在执行过程中需要JVM处于这样一个状态——堆的内容不会被改变，这种状态在JVM里叫做安全点（safe-point）。
Periodic task thread：这个线程负责响应定时触发的事件（例如：中断），用来执行一些定时操作。
GC thread：这些线程负责JVM里的垃圾收集活动；
Compiler threads：这些线程负责在运行时将字节码编译为本地代码；
Singal dispatcher thread：这些线程负责响应外部发给当前JVM进程的信号，并通过调用JVM内的其他线程。

JVM中的线程模型

我们现在写一个简单的hello word程序，代码如下：

    public class GcExample {
        private static class E {
            public static final int[] a = new int[1024 * 10];
        }

        public static void main(String[] args) {
            System.out.println("hello world");
            while (true) {
                new E();
            }
        }
    }

然后使用jmc（Java Mission Control）attach到这个程序上，展现为如下的情况： jmc中看到的线程

RMI开头的线程，负责JVM跟JMC客户端通信，吐出JVM内的运行信息；
Attach Listener和Single Dispatcher两个线程，属于信号处理线程，负责接收外部到当前JVM的attach信号，并建立通信用的文件socket；
Finalizer线程，用于处理Finalizer队列的线程，在Java中，如果一个对象重写了finalize()方法，那么JVM会为之创建一个对应的Finalizer对象，所有的Finzlizer对象会构成一个列表，由Finalizer线程统一处理
Reference Handler，负责JVM中的引用处理
main，我们例子中的业务线程。

我想你现在也有这个疑问——跟上面说的那个分类对不上，有些线程没看到，是的，可能是由于JMC的实现机制，这些线程没有被展示出来，我们再用jstack命令做一次线程dump，就可以得到如下内容：

Full thread dump Java HotSpot(TM) 64-Bit Server VM (25.151-b12 mixed mode):

"RMI TCP Connection(5)-192.168.0.139" #17 daemon prio=9 os_prio=31 tid=0x00007fba7c830800 nid=0x5c03 runnable [0x000070000f740000]
   java.lang.Thread.State: RUNNABLE
    at java.net.SocketInputStream.socketRead0(Native Method)
    at java.net.SocketInputStream.socketRead(SocketInputStream.java:116)
    at java.net.SocketInputStream.read(SocketInputStream.java:171)
    at java.net.SocketInputStream.read(SocketInputStream.java:141)
    at java.io.BufferedInputStream.fill(BufferedInputStream.java:246)
    at java.io.BufferedInputStream.read(BufferedInputStream.java:265)
    - locked <0x000000076f590708> (a java.io.BufferedInputStream)
    at java.io.FilterInputStream.read(FilterInputStream.java:83)
    at sun.rmi.transport.tcp.TCPTransport.handleMessages(TCPTransport.java:550)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run0(TCPTransport.java:826)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.lambda$run$0(TCPTransport.java:683)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler$$Lambda$3/745402377.run(Unknown Source)
    at java.security.AccessController.doPrivileged(Native Method)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run(TCPTransport.java:682)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at java.lang.Thread.run(Thread.java:748)

"RMI TCP Connection(2)-192.168.0.139" #16 daemon prio=9 os_prio=31 tid=0x00007fba7e0b0000 nid=0xa403 in Object.wait() [0x000070000f63c000]
   java.lang.Thread.State: TIMED_WAITING (on object monitor)
    at java.lang.Object.wait(Native Method)
    - waiting on <0x000000076f580878> (a com.sun.jmx.remote.internal.ArrayNotificationBuffer)
    at com.sun.jmx.remote.internal.ArrayNotificationBuffer.fetchNotifications(ArrayNotificationBuffer.java:449)
    - locked <0x000000076f580878> (a com.sun.jmx.remote.internal.ArrayNotificationBuffer)
    at com.sun.jmx.remote.internal.ArrayNotificationBuffer$ShareBuffer.fetchNotifications(ArrayNotificationBuffer.java:227)
    at com.sun.jmx.remote.internal.ServerNotifForwarder.fetchNotifs(ServerNotifForwarder.java:274)
    at javax.management.remote.rmi.RMIConnectionImpl$4.run(RMIConnectionImpl.java:1270)
    at javax.management.remote.rmi.RMIConnectionImpl$4.run(RMIConnectionImpl.java:1268)
    at javax.management.remote.rmi.RMIConnectionImpl.fetchNotifications(RMIConnectionImpl.java:1274)
    at sun.reflect.GeneratedMethodAccessor62.invoke(Unknown Source)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at sun.rmi.server.UnicastServerRef.dispatch(UnicastServerRef.java:357)
    at sun.rmi.transport.Transport$1.run(Transport.java:200)
    at sun.rmi.transport.Transport$1.run(Transport.java:197)
    at java.security.AccessController.doPrivileged(Native Method)
    at sun.rmi.transport.Transport.serviceCall(Transport.java:196)
    at sun.rmi.transport.tcp.TCPTransport.handleMessages(TCPTransport.java:568)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run0(TCPTransport.java:826)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.lambda$run$0(TCPTransport.java:683)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler$$Lambda$3/745402377.run(Unknown Source)
    at java.security.AccessController.doPrivileged(Native Method)
    at sun.rmi.transport.tcp.TCPTransport$ConnectionHandler.run(TCPTransport.java:682)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at java.lang.Thread.run(Thread.java:748)

"JMX server connection timeout 15" #15 daemon prio=9 os_prio=31 tid=0x00007fba80014800 nid=0xa503 in Object.wait() [0x000070000f53b000]
   java.lang.Thread.State: TIMED_WAITING (on object monitor)
    at java.lang.Object.wait(Native Method)
    - waiting on <0x000000076f5887f8> (a [I)
    at com.sun.jmx.remote.internal.ServerCommunicatorAdmin$Timeout.run(ServerCommunicatorAdmin.java:168)
    - locked <0x000000076f5887f8> (a [I)
    at java.lang.Thread.run(Thread.java:748)

"RMI Scheduler(0)" #14 daemon prio=9 os_prio=31 tid=0x00007fba7c805800 nid=0xa603 waiting on condition [0x000070000f438000]
   java.lang.Thread.State: TIMED_WAITING (parking)
    at sun.misc.Unsafe.park(Native Method)
    - parking to wait for  <0x000000076f598188> (a java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject)
    at java.util.concurrent.locks.LockSupport.parkNanos(LockSupport.java:215)
    at java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.awaitNanos(AbstractQueuedSynchronizer.java:2078)
    at java.util.concurrent.ScheduledThreadPoolExecutor$DelayedWorkQueue.take(ScheduledThreadPoolExecutor.java:1093)
    at java.util.concurrent.ScheduledThreadPoolExecutor$DelayedWorkQueue.take(ScheduledThreadPoolExecutor.java:809)
    at java.util.concurrent.ThreadPoolExecutor.getTask(ThreadPoolExecutor.java:1074)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1134)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at java.lang.Thread.run(Thread.java:748)

"RMI TCP Accept-0" #12 daemon prio=9 os_prio=31 tid=0x00007fba7d906000 nid=0xa803 runnable [0x000070000f232000]
   java.lang.Thread.State: RUNNABLE
    at java.net.PlainSocketImpl.socketAccept(Native Method)
    at java.net.AbstractPlainSocketImpl.accept(AbstractPlainSocketImpl.java:409)
    at java.net.ServerSocket.implAccept(ServerSocket.java:545)
    at java.net.ServerSocket.accept(ServerSocket.java:513)
    at sun.management.jmxremote.LocalRMIServerSocketFactory$1.accept(LocalRMIServerSocketFactory.java:52)
    at sun.rmi.transport.tcp.TCPTransport$AcceptLoop.executeAcceptLoop(TCPTransport.java:400)
    at sun.rmi.transport.tcp.TCPTransport$AcceptLoop.run(TCPTransport.java:372)
    at java.lang.Thread.run(Thread.java:748)

"Attach Listener" #10 daemon prio=9 os_prio=31 tid=0x00007fba7d865800 nid=0xa903 waiting on condition [0x0000000000000000]
   java.lang.Thread.State: RUNNABLE

"Service Thread" #9 daemon prio=9 os_prio=31 tid=0x00007fba7d803000 nid=0x3903 runnable [0x0000000000000000]
   java.lang.Thread.State: RUNNABLE

"C1 CompilerThread3" #8 daemon prio=9 os_prio=31 tid=0x00007fba7e002000 nid=0x3803 waiting on condition [0x0000000000000000]
   java.lang.Thread.State: RUNNABLE

"C2 CompilerThread2" #7 daemon prio=9 os_prio=31 tid=0x00007fba80000000 nid=0x3703 waiting on condition [0x0000000000000000]
   java.lang.Thread.State: RUNNABLE

"C2 CompilerThread1" #6 daemon prio=9 os_prio=31 tid=0x00007fba7d82d800 nid=0x3e03 waiting on condition [0x0000000000000000]
   java.lang.Thread.State: RUNNABLE

"C2 CompilerThread0" #5 daemon prio=9 os_prio=31 tid=0x00007fba7c020000 nid=0x3f03 waiting on condition [0x0000000000000000]
   java.lang.Thread.State: RUNNABLE

"Signal Dispatcher" #4 daemon prio=9 os_prio=31 tid=0x00007fba7c01e800 nid=0x3403 runnable [0x0000000000000000]
   java.lang.Thread.State: RUNNABLE

"Finalizer" #3 daemon prio=8 os_prio=31 tid=0x00007fba7f022000 nid=0x4903 in Object.wait() [0x000070000e917000]
   java.lang.Thread.State: WAITING (on object monitor)
    at java.lang.Object.wait(Native Method)
    - waiting on <0x000000076f5a0600> (a java.lang.ref.ReferenceQueue$Lock)
    at java.lang.ref.ReferenceQueue.remove(ReferenceQueue.java:143)
    - locked <0x000000076f5a0600> (a java.lang.ref.ReferenceQueue$Lock)
    at java.lang.ref.ReferenceQueue.remove(ReferenceQueue.java:164)
    at java.lang.ref.Finalizer$FinalizerThread.run(Finalizer.java:209)

"Reference Handler" #2 daemon prio=10 os_prio=31 tid=0x00007fba7c01d800 nid=0x4b03 in Object.wait() [0x000070000e814000]
   java.lang.Thread.State: WAITING (on object monitor)
    at java.lang.Object.wait(Native Method)
    - waiting on <0x000000076f5983e8> (a java.lang.ref.Reference$Lock)
    at java.lang.Object.wait(Object.java:502)
    at java.lang.ref.Reference.tryHandlePending(Reference.java:191)
    - locked <0x000000076f5983e8> (a java.lang.ref.Reference$Lock)
    at java.lang.ref.Reference$ReferenceHandler.run(Reference.java:153)

"main" #1 prio=5 os_prio=31 tid=0x00007fba7f000000 nid=0x2303 runnable [0x000070000ddf6000]
   java.lang.Thread.State: RUNNABLE
    at GcExample.main(GcExample.java:9)

"VM Thread" os_prio=31 tid=0x00007fba7c01b000 nid=0x2f03 runnable 

"GC task thread#0 (ParallelGC)" os_prio=31 tid=0x00007fba7c007000 nid=0x2007 runnable 

"GC task thread#1 (ParallelGC)" os_prio=31 tid=0x00007fba7d804800 nid=0x2a03 runnable 

"GC task thread#2 (ParallelGC)" os_prio=31 tid=0x00007fba7c007800 nid=0x5303 runnable 

"GC task thread#3 (ParallelGC)" os_prio=31 tid=0x00007fba7f800000 nid=0x2c03 runnable 

"GC task thread#4 (ParallelGC)" os_prio=31 tid=0x00007fba7d805000 nid=0x5103 runnable 

"GC task thread#5 (ParallelGC)" os_prio=31 tid=0x00007fba7c013800 nid=0x5003 runnable 

"GC task thread#6 (ParallelGC)" os_prio=31 tid=0x00007fba7f001000 nid=0x4e03 runnable 

"GC task thread#7 (ParallelGC)" os_prio=31 tid=0x00007fba7c014000 nid=0x4c03 runnable 

"VM Periodic Task Thread" os_prio=31 tid=0x00007fba7d858800 nid=0x3a03 waiting on condition 

JNI global references: 295

OK，从上面这个dump文件中，可以找到Periodic Task Thread、GC Thread、VM Thread、Compiler Thread的身影了。

JVM源码分析

前面从概念和分类两个角度观察了JVM中的线程，现在我们从源码角度看下另一个问题，JVM是如何实现Java线程的。

java.lang.Thread类的start接口，用来启动一个Java线程，然后JVM会执行run()方法中的内容，run()方法是Runnable接口定义然后在java.lang.Thread中提供了实现方法，start()方法的内容如下：

    /**
     * Causes this thread to begin execution; the Java Virtual Machine
     * calls the <code>run</code> method of this thread.
     * <p>
     * The result is that two threads are running concurrently: the
     * current thread (which returns from the call to the
     * <code>start</code> method) and the other thread (which executes its
     * <code>run</code> method).
     * <p>
     * It is never legal to start a thread more than once.
     * In particular, a thread may not be restarted once it has completed
     * execution.
     *
     * @exception  IllegalThreadStateException  if the thread was already
     *               started.
     * @see        #run()
     * @see        #stop()
     */
    public synchronized void start() {
        /**
         * This method is not invoked for the main method thread or "system"
         * group threads created/set up by the VM. Any new functionality added
         * to this method in the future may have to also be added to the VM.
         *
         * A zero status value corresponds to state "NEW".
         */
        if (threadStatus != 0)
            throw new IllegalThreadStateException();

        /* Notify the group that this thread is about to be started
         * so that it can be added to the group's list of threads
         * and the group's unstarted count can be decremented. */
        group.add(this);

        boolean started = false;
        try {
            start0();
            started = true;
        } finally {
            try {
                if (!started) {
                    group.threadStartFailed(this);
                }
            } catch (Throwable ignore) {
                /* do nothing. If start0 threw a Throwable then
                  it will be passed up the call stack */
            }
        }
    }

    private native void start0();

根据注释中说的，一个线程退出后是再次start是非法的，会抛出异常，我们可以用下面的代码验证下：

package org.java.learn.concurrent;

public class ThreadRestartExample {

    public static void main(String[] args) throws InterruptedException {
        Thread thread = new Thread(() -> {
            System.out.println("hello");
        });

        thread.start();

        Thread.sleep(1000);

        thread.start();
    }
}

运行这个代码的结果是：

hello
Exception in thread "main" java.lang.IllegalThreadStateException
    at java.lang.Thread.start(Thread.java:708)
    at org.java.learn.concurrent.ThreadRestartExample.main(ThreadRestartExample.java:18)

Process finished with exit code 1

start方法调用了start0方法，这是一个JNI接口，在Java中通过JNI接口可以实现Java调用本地方法；通过JVMTI接口可以实现在C++空间调用Java对象的方法。start0方法的实现在jdk/src/share/native/java/lang/Thread.c中定义，代码如下所示：

“`c++
static JNINativeMethod methods[] = {
{"start0", "()V", (void *)&JVM_StartThread},
{"stop0", "(" OBJ ")V", (void *)&JVM_StopThread},
{"isAlive", "()Z", (void *)&JVM_IsThreadAlive},
{"suspend0", "()V", (void *)&JVM_SuspendThread},
{"resume0", "()V", (void *)&JVM_ResumeThread},
{"setPriority0", "(I)V", (void *)&JVM_SetThreadPriority},
{"yield", "()V", (void *)&JVM_Yield},
{"sleep", "(J)V", (void *)&JVM_Sleep},
{"currentThread", "()" THD, (void *)&JVM_CurrentThread},
{"countStackFrames", "()I", (void *)&JVM_CountStackFrames},
{"interrupt0", "()V", (void *)&JVM_Interrupt},
{"isInterrupted", "(Z)Z", (void *)&JVM_IsInterrupted},
{"holdsLock", "(" OBJ ")Z", (void *)&JVM_HoldsLock},
{"getThreads", "()[" THD, (void *)&JVM_GetAllThreads},
{"dumpThreads", "([" THD ")[[" STE, (void *)&JVM_DumpThreads},
{"setNativeName", "(" STR ")V", (void *)&JVM_SetNativeThreadName},
};

<pre><code class="line-numbers"><br />JVM_StartThread的接口定义在jvm.h中，JDK中用到的jni接口，最终都会在jvm.h文件中定义，并在jvm.cpp中作为C++实现的入口，也就是说jvm.cpp是Java世界和JVM中C++世界沟通的桥梁。

“`c++
/*
* java.lang.Thread
*/
JNIEXPORT void JNICALL
JVM_StartThread(JNIEnv *env, jobject thread);

JVM_StartThread的具体实现在jvm.cpp中，主要代码逻辑列举如下（本文主要是要看线程创建的逻辑，因此一些分支代码没有展示）：

“`c++
JVM_ENTRY(void, JVM_StartThread(JNIEnv* env, jobject jthread))
JVMWrapper("JVM_StartThread");
JavaThread *native_thread = NULL;
//……
//获取栈的大小
jlong size = java_lang_Thread::stackSize(JNIHandles::resolve_non_null(jthread));

//……
//栈的大小不能为负数
size_t sz = size > 0 ? (size_t) size : 0;
//通过new JavaThread新建os线程对象，这里thread_entry就是runnable的run方法。
native_thread = new JavaThread(&thread_entry, sz);

//……
Thread::start(native_thread);

JVM_END


JavaThread的构造方法实现时在thread.cpp文件中，做一些准备工作后，会通过`os::create_thread(this, thr_type, stack_sz);`创建线程，os::create_thread的实现时跟具体平台有关的，如下图所示：
![os::create_thread有不同平台的实现](http://img.cdn.javaadu.online/jvm-thread-3.png)

这里我们选择os_linux.cpp这个文件。os::create_thread的主要动作有几个：

1. 通过`pthread_attr_init(&attr);`初始化线程的属性
2. 通过`int ret = pthread_create(&tid, &attr, (void* (*)(void*)) java_start, thread);`创建os线程，这里最重要了，参见[pthread_create](https://www.systutorials.com/docs/linux/man/3-pthread_create/)手册，可以知道，第三个参数表示启动这个线程后要执行的方法的入口，第四个参数表示要给这个方法传入的参数。 

这里我们看下java_start方法的实现（该方法也在thread.cpp中），在这个方法的入参是Thread指针；

```c++
// Thread start routine for all newly created threads
static void *java_start(Thread *thread) {
  // Try to randomize the cache line index of hot stack frames.
  // This helps when threads of the same stack traces evict each other's
  // cache lines. The threads can be either from the same JVM instance, or
  // from different JVM instances. The benefit is especially true for
  // processors with hyperthreading technology.
  static int counter = 0;
  int pid = os::current_process_id();
  alloca(((pid ^ counter++) & 7) * 128);

  ThreadLocalStorage::set_thread(thread);

  OSThread* osthread = thread->osthread();
  Monitor* sync = osthread->startThread_lock();

  // non floating stack LinuxThreads needs extra check, see above
  if (!_thread_safety_check(thread)) {
    // notify parent thread
    MutexLockerEx ml(sync, Mutex::_no_safepoint_check_flag);
    osthread->set_state(ZOMBIE);
    sync->notify_all();
    return NULL;
  }

  // thread_id is kernel thread id (similar to Solaris LWP id)
  osthread->set_thread_id(os::Linux::gettid());

  if (UseNUMA) {
    int lgrp_id = os::numa_get_group_id();
    if (lgrp_id != -1) {
      thread->set_lgrp_id(lgrp_id);
    }
  }
  // initialize signal mask for this thread
  os::Linux::hotspot_sigmask(thread);

  // initialize floating point control register
  os::Linux::init_thread_fpu_state();

  // handshaking with parent thread
  {
    MutexLockerEx ml(sync, Mutex::_no_safepoint_check_flag);

    // notify parent thread
    osthread->set_state(INITIALIZED);
    sync->notify_all();

    // wait until os::start_thread()
    // 这里说明，新创建的os线程不会立即执行，会等os::start_thread()的通知，在后面我们马上会分析到。
    while (osthread->get_state() == INITIALIZED) {
      sync->wait(Mutex::_no_safepoint_check_flag);
    }
  }

  // call one more level start routine
  thread->run();

  return 0;
}
</code></pre>

在方法的最后，会通过<code>thread->run();</code>调用JavaThread的run方法，然后再到JavaThread的thread_main_inner方法，

```c++
void JavaThread::thread_main_inner() {
  assert(JavaThread::current() == this, "sanity check");
  assert(this->threadObj() != NULL, "just checking");

// Execute thread entry point unless this thread has a pending exception
  // or has been stopped before starting.
  // Note: Due to JVM_StopThread we can have pending exceptions already!
  if (!this->has_pending_exception() &&
      !java_lang_Thread::is_stillborn(this->threadObj())) {
    {
      ResourceMark rm(this);
      this->set_native_thread_name(this->get_thread_name());
    }
    HandleMark hm(this);
    //注意：这里就是Java线程要执行的run方法
    this->entry_point()(this, this);
  }

DTRACE_THREAD_PROBE(stop, this);

this->exit(false);
  delete this;
}


`this->entry_point()(this, this);`这行的调用，就会执行java.lang.Thread中的run方法，那么这个entry_point是在哪里被设置到JavaThread对象中的呢，回顾上文，在jvm.cpp里有一个new JavaThread(&thread_entry,sz)的调用，是的，就是这里，thread_entry的具体实现是：

```c++
static void thread_entry(JavaThread* thread, TRAPS) {
  HandleMark hm(THREAD);
  Handle obj(THREAD, thread->threadObj());
  JavaValue result(T_VOID);
  JavaCalls::call_virtual(&result,
                          obj,
                          KlassHandle(THREAD, SystemDictionary::Thread_klass()),
                          vmSymbols::run_method_name(),
                          vmSymbols::void_method_signature(),
                          THREAD);
}
</code></pre>

这段代码要做的事情就是在JVM的c++世界里，获取到对应的java.lang.Thread的对象，然后调用它的run方法。

再看下JVM_StartThread的逻辑，native_thread被创建后并不会立即被执行，而是出于初始化状态，后面还会执行<code>Thread::start(native_thread);</code>代码，这是做了什么工作呢？

```c++
void Thread::start(Thread* thread) {
  trace("start", thread);
  // Start is different from resume in that its safety is guaranteed by context or
  // being called from a Java method synchronized on the Thread object.
  if (!DisableStartThread) {
    if (thread->is_Java_thread()) {
      // Initialize the thread state to RUNNABLE before starting this thread.
      // Can not set it after the thread started because we do not know the
      // exact thread state at that time. It could be in MONITOR_WAIT or
      // in SLEEPING or some other state.
      java_lang_Thread::set_thread_status(((JavaThread*)thread)->threadObj(),
                                          java_lang_Thread::RUNNABLE);
    }
    os::start_thread(thread);
  }
}


根据代码可知道，这个方法先将thread的状态设置为RUNNABLE，然后再调用`os::start_thread(thread)；`通知刚刚创建的os线程开始运行，具体的代码如下：

```c++
void os::start_thread(Thread* thread) {
  // guard suspend/resume
  MutexLockerEx ml(thread->SR_lock(), Mutex::_no_safepoint_check_flag);
  OSThread* osthread = thread->osthread();
  osthread->set_state(RUNNABLE);
  pd_start_thread(thread);
}

在这里pd_start_thread(具体实现在os_linux.cpp)就负责通知刚刚被创建的但是处于初始化状态的线程，代码如下：
c++

void os::pd_start_thread(Thread* thread) {

  OSThread * osthread = thread->osthread();

  assert(osthread->get_state() != INITIALIZED, &quot;just checking&quot;);

  Monitor* sync_with_child = osthread->startThread_lock();

  MutexLockerEx ml(sync_with_child, Mutex::_no_safepoint_check_flag);

  sync_with_child->notify();

}
sync_with_child->notify();这行代码就是用来通知线程开始运行的。
总结
这篇文章主要梳理了三个问题：（1）JVM中的线程模型是怎么样的，跟os中的线程一一对应；（2）JVM里常见的几类线程都有哪些？VM Thread、周期线程、Compiler 线程、GC线程、信号量处理线程；（3）当我们在java代码中执行start()方法的时候，JVM内部做了哪些事情？
参考资料

HotSpot JVM internal threads
JVM原理与实现——Thread
How Java thread maps to OS thread?
Java并发的官方文档
JVM attach机制的实现
HOTSPOTVM线程实现浅析


本号专注于后端技术、JVM问题排查和优化、Java面试题、个人成长和自我管理等主题，为读者提供一线开发者的工作和成长经验，期待你能在这里有所收获。


JVM源码分析之Hotspot VM中的Thread创建