Thread start 源码揭秘

Hi~ o(￣▽￣)ブ ，小伙伴们早上好呀~

赶紧搭上 冒险家 4ye 的航船叭 ，一起来探索下 hotspot 源码大陆~ ?

这期和大家带来了这个 线程 Thread start 的奥秘

Thread start 源码揭秘

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 {
        // 看这里~
        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 */
        }
    }
}
复制代码

从上面的代码中，我们可以发现它会先去 判断 threadStatus 是不是 0， 不是的话会抛出异常 。

在 Thread 源码中我们可以发现 threadStatus 默认值就是 0

• 那什么时候会被改变呢？

• 这个 threadStatus 都有哪些值呢？

嘿嘿 带着小小的疑问继续往下看叭~ ?

---

start0

状态的改变肯定伴随着线程的启动，所以我们直接来到下面这个 start0 方法

private native void start0();
复制代码

可以发现它是一个 native 函数，我们直接

OpenJDK 的 JDK 源码中全局搜索，就可以找到它了

path： jdk\src\share\native\java\lang\Thread.c

JNINativeMethod

这个 JNINativeMethod 是一个结构体

path： jdk\src\share\javavm\export\jni.h

/*
 * used in RegisterNatives to describe native method name, signature,
 * and function pointer.
 */
typedef struct {
    char *name;
    char *signature;
    void *fnPtr;
} JNINativeMethod;
复制代码

JVM_StartThread

顺着思路，我们找到这个 JVM_StartThread 方法

path: jdk\src\share\javavm\export\jvm.h

/*
 * java.lang.Thread
 */
JNIEXPORT void JNICALL
JVM_StartThread(JNIEnv *env, jobject thread);
复制代码

这里 JNIEXPORT 和 JNICALL 都是 JNI 的关键字，表示此函数是要被 JNI 调用的

看到它是 JVM 开头的文件，我们来到下面这个 hotspot 的源码中找找看?

发现这里有使用到?

path: hotspot\src\share\vm\prims\jvm.cpp

代码：

JVM_ENTRY(void, JVM_StartThread(JNIEnv* env, jobject jthread))
  JVMWrapper("JVM_StartThread");
  JavaThread *native_thread = NULL;

  // We cannot hold the Threads_lock when we throw an exception,
  // due to rank ordering issues. Example:  we might need to grab the
  // Heap_lock while we construct the exception.
  bool throw_illegal_thread_state = false;

  // We must release the Threads_lock before we can post a jvmti event
  // in Thread::start.
  {
    // Ensure that the C++ Thread and OSThread structures aren't freed before
    // we operate.
    MutexLocker mu(Threads_lock);

    // Since JDK 5 the java.lang.Thread threadStatus is used to prevent
    // re-starting an already started thread, so we should usually find
    // that the JavaThread is null. However for a JNI attached thread
    // there is a small window between the Thread object being created
    // (with its JavaThread set) and the update to its threadStatus, so we
    // have to check for this
    if (java_lang_Thread::thread(JNIHandles::resolve_non_null(jthread)) != NULL) {
      throw_illegal_thread_state = true;
    } else {
      // We could also check the stillborn flag to see if this thread was already stopped, but
      // for historical reasons we let the thread detect that itself when it starts running

      jlong size =
             java_lang_Thread::stackSize(JNIHandles::resolve_non_null(jthread));
      // Allocate the C++ Thread structure and create the native thread.  The
      // stack size retrieved from java is signed, but the constructor takes
      // size_t (an unsigned type), so avoid passing negative values which would
      // result in really large stacks.
      size_t sz = size > 0 ? (size_t) size : 0;
      native_thread = new JavaThread(&thread_entry, sz);

      // At this point it may be possible that no osthread was created for the
      // JavaThread due to lack of memory. Check for this situation and throw
      // an exception if necessary. Eventually we may want to change this so
      // that we only grab the lock if the thread was created successfully -
      // then we can also do this check and throw the exception in the
      // JavaThread constructor.
      if (native_thread->osthread() != NULL) {
        // Note: the current thread is not being used within "prepare".
        native_thread->prepare(jthread);
      }
    }
  }

  if (throw_illegal_thread_state) {
    THROW(vmSymbols::java_lang_IllegalThreadStateException());
  }

  assert(native_thread != NULL, "Starting null thread?");

  if (native_thread->osthread() == NULL) {
    // No one should hold a reference to the 'native_thread'.
    delete native_thread;
    if (JvmtiExport::should_post_resource_exhausted()) {
      JvmtiExport::post_resource_exhausted(
        JVMTI_RESOURCE_EXHAUSTED_OOM_ERROR | JVMTI_RESOURCE_EXHAUSTED_THREADS,
        "unable to create new native thread");
    }
    THROW_MSG(vmSymbols::java_lang_OutOfMemoryError(),
              "unable to create new native thread");
  }

 // 看这里~
  Thread::start(native_thread);

JVM_END
复制代码

这里 JVM_ENTRY 和 JVM_END是两个宏定义，定义了函数体的头和尾。 ?

这里太多细节了，啃不下呀 ?

讲几个大概的~ 咳咳 也就看看作者的注释翻译翻译?

比如：

创建线程

native_thread = new JavaThread(&thread_entry, sz);
复制代码

来到构造器中

JavaThread::JavaThread(ThreadFunction entry_point, size_t stack_sz) :
  Thread()
#if INCLUDE_ALL_GCS
  , _satb_mark_queue(&_satb_mark_queue_set),
  _dirty_card_queue(&_dirty_card_queue_set)
#endif // INCLUDE_ALL_GCS
{
  if (TraceThreadEvents) {
    tty->print_cr("creating thread %p", this);
  }
  initialize();
  _jni_attach_state = _not_attaching_via_jni;
  set_entry_point(entry_point);
  // Create the native thread itself.
  // %note runtime_23
  os::ThreadType thr_type = os::java_thread;
  thr_type = entry_point == &compiler_thread_entry ? os::compiler_thread :
                                                     os::java_thread;
  // 看这里~
   os::create_thread(this, thr_type, stack_sz);
  _safepoint_visible = false;
 
}
复制代码

创建内核线程

发现这里是去 创建内核线程~ ?

os::create_thread(this, thr_type, stack_sz);
复制代码

顺着思路继续探索~

针对 linux 的场景

有如下的代码 ?

path: hotspot\src\os\linux\vm\os_linux.cpp

终于看到有点眼熟的函数了~ 哈哈哈

int ret = pthread_create(&tid, &attr, (void* (*)(void*)) java_start, thread);
复制代码

底层内容真是让人抓狂~ 无止境的.. 哈哈哈

这里截下这个指针函数 java_start ，有兴趣的小伙伴继续往下冲冲冲！?

启动线程

另一个重点 **Thread::start(native_thread); **

Thread::start

来到下面这里~

path: hotspot\src\share\vm\runtime\thread.cpp

第一个红框的作用：

在启动该线程之前，将线程状态初始化为 RUNNABLE。

不能在线程启动后设置，因为我们不知道 正确的线程状态，它可能在 MONITOR_WAIT 或 在睡眠或其他状态。

第二个红框的作用，启动内核线程！

// The INITIALIZED state is distinguished from the SUSPENDED state because the
// conditions in which a thread is first started are different from those in which
// a suspension is resumed.  These differences make it hard for us to apply the
// tougher checks when starting threads that we want to do when resuming them.
// However, when start_thread is called as a result of Thread.start, on a Java
// thread, the operation is synchronized on the Java Thread object.  So there
// cannot be a race to start the thread and hence for the thread to exit while
// we are working on it.  Non-Java threads that start Java threads either have
// to do so in a context in which races are impossible, or should do appropriate
// locking.

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);
}
复制代码

嘿嘿 这里就直接搜搜看 pd_start_thread 啦，毕竟一看就知道是它去启动线程的 哈哈哈 ?

结果如图~

最后记录下这个 全局函数 notify ?

bool Monitor::notify() {
  assert (_owner == Thread::current(), "invariant") ;
  assert (ILocked(), "invariant") ;
  if (_WaitSet == NULL) return true ;
  NotifyCount ++ ;

  // Transfer one thread from the WaitSet to the EntryList or cxq.
  // Currently we just unlink the head of the WaitSet and prepend to the cxq.
  // And of course we could just unlink it and unpark it, too, but
  // in that case it'd likely impale itself on the reentry.
  Thread::muxAcquire (_WaitLock, "notify:WaitLock") ;
  ParkEvent * nfy = _WaitSet ;
  if (nfy != NULL) {                  // DCL idiom
    _WaitSet = nfy->ListNext ;
    assert (nfy->Notified == 0, "invariant") ;
    // push nfy onto the cxq
    for (;;) {
      const intptr_t v = _LockWord.FullWord ;
      assert ((v & 0xFF) == _LBIT, "invariant") ;
      nfy->ListNext = (ParkEvent *)(v & ~_LBIT);
      if (CASPTR (&_LockWord, v, UNS(nfy)|_LBIT) == v) break;
      // interference - _LockWord changed -- just retry
    }
    // Note that setting Notified before pushing nfy onto the cxq is
    // also legal and safe, but the safety properties are much more
    // subtle, so for the sake of code stewardship ...
    OrderAccess::fence() ;
    nfy->Notified = 1;
  }
  Thread::muxRelease (_WaitLock) ;
  if (nfy != NULL && (NativeMonitorFlags & 16)) {
    // Experimental code ... light up the wakee in the hope that this thread (the owner)
    // will drop the lock just about the time the wakee comes ONPROC.
    nfy->unpark() ;
  }
  assert (ILocked(), "invariant") ;
  return true ;
}
复制代码

惊喜，居然在这个代码中看到这段注释，这里提到 WaitSet ， EntryList ，cxq

这是 隐式锁? [[Synchronized 内部的实现原理呀]] 或者 称它为 [[Monitor机制]] ?

• WaitSet 是一个等待队列，存放进入等待状态的线程
• cxq 是一个竞争队列，所有请求锁?的线程会先到这里
• EntryList 存放 cxq 中有资格成为候选资源去竞争锁的线程

// Transfer one thread from the WaitSet to the EntryList or cxq.
// Currently we just unlink the head of the WaitSet and prepend to the cxq.
// And of course we could just unlink it and unpark it, too, but
// in that case it'd likely impale itself on the reentry.
Thread::muxAcquire (_WaitLock, "notify:WaitLock") ;
复制代码

嘿嘿 后面深入锁的部分再分享?

---

threadStatus

回到最上面，那还剩一个问题没解决~

这个 threadStatus 都有哪些值呢？

这里我们可以从 jvm.h 源码中获取到 ?

/*
 * Java thread state support
 */
enum {
    JAVA_THREAD_STATE_NEW           = 0,
    JAVA_THREAD_STATE_RUNNABLE      = 1,
    JAVA_THREAD_STATE_BLOCKED       = 2,
    JAVA_THREAD_STATE_WAITING       = 3,
    JAVA_THREAD_STATE_TIMED_WAITING = 4,
    JAVA_THREAD_STATE_TERMINATED    = 5,
    JAVA_THREAD_STATE_COUNT         = 6
};
复制代码

总结

本次的 Thread start 深入虚拟机探索之旅 就先告一段落啦~ 太难啃了?（咳咳，实力差太多了！被 Hotspot 源码 Boss 暴打一顿后，4ye 冒险家只能带着一点点经验跑开了ε=ε=ε=(￣▽￣) 哈哈哈? ）

收获：

1. 学了一点 C 和 C++ 语法 哈哈哈

2. 了解到 线程创建的本质 ，针对 linux 场景，有如下的感悟

3. 发现每个 java 线程，都会对应到一个内核线程 ；

4. 可以更好地体会 用户态 和 内核态 的存在 ；

5. 了解到启动一个线程时，线程状态的改变是在开启线程之前 ；

6. 启动内核线程的实际操作是调用了 Monitor::notify 方法去唤醒它 ?

7. 同时，在这个方法中，窥探到了一丝丝 Monitor 机制 的存在 ；

8. 看到了无处不在的 mutex ；

9. 更能体会到 java为能一次编译到处运行了 哈哈， JVM 已经封装好了对应的 OS

10. 源码阅读经验+1 哈哈哈

11. 开启了新大陆 ： 后面还可以尝试编译下，做一个自己的 JVM ?