最近發現項目中的日志用得不大規范，于是準備做些改造，在我們項目中用的是 slf4j1.5.10 的版本，順便看了下 slf4j 的源碼，看到 LoggerFactory 初始化的邏輯如下

  public static ILoggerFactory getILoggerFactory() {
    if (INITIALIZATION_STATE == UNINITIALIZED) {
      INITIALIZATION_STATE = ONGOING_INITILIZATION;
      performInitialization();
    }
    switch (INITIALIZATION_STATE) {
    case SUCCESSFUL_INITILIZATION:
      return getSingleton().getLoggerFactory();
    case FAILED_INITILIZATION:
      throw new IllegalStateException(UNSUCCESSFUL_INIT_MSG);
    case ONGOING_INITILIZATION:
      // support re-entrant behavior.
      // See also http://bugzilla.slf4j.org/show_bug.cgi?id=106
      return TEMP_FACTORY;
    }
    throw new IllegalStateException("Unreachable code");
  }

有發現什么問題嗎？再看看 INITIALIZATION_STATE 這個變量是怎么聲明的，如下

  static int INITIALIZATION_STATE = UNINITIALIZED;

現在很明確了，如果在多線程環境下，多個線程調用 LoggerFactory.getLogger() 方法的時候初始化會出現問題，這簡直有點顛覆世界觀，說好的大神代碼呢？懷著極度懷疑自己的心情翻看了最新的 slf4j1.7.21 代碼，看到初始化邏輯如下:

     public static ILoggerFactory getILoggerFactory() {
        if (INITIALIZATION_STATE == UNINITIALIZED) {
            synchronized (LoggerFactory.class) {
                if (INITIALIZATION_STATE == UNINITIALIZED) {
                    INITIALIZATION_STATE = ONGOING_INITIALIZATION;
                    performInitialization();
                }
            }
        }
        switch (INITIALIZATION_STATE) {
        case SUCCESSFUL_INITIALIZATION:
            return StaticLoggerBinder.getSingleton().getLoggerFactory();
        case NOP_FALLBACK_INITIALIZATION:
            return NOP_FALLBACK_FACTORY;
        case FAILED_INITIALIZATION:
            throw new IllegalStateException(UNSUCCESSFUL_INIT_MSG);
        case ONGOING_INITIALIZATION:
            // support re-entrant behavior.
            // See also http://jira.qos.ch/browse/SLF4J-97
            return SUBST_FACTORY;
        }
        throw new IllegalStateException("Unreachable code");
    }

再看看 INITIALIZATION_STATE 變量的聲明

        static volatile int INITIALIZATION_STATE = UNINITIALIZED;

頓時淚流滿面，這才像傳說中的大神代碼啊