更新文档

Author: hevake

README.md

@@ -52,6 +52,7 @@ One or more of the three sink can be selected in the startup parameters, and can
 The log content includes: level, time (accurate to microseconds), thread number, module name, function name, text, file name, line number.  
 It is convenient and quick to locate the problem. 
 ![log show](documents/images/0002-log-show.png)  
+As shown in the figure above, different log levels are clearly distinguished by colors. In particular, for anomalies, red and yellow are used together with highlighting, making them very eye-catching.
 
 **3) Flexible log output filter**  
 The log level can be set separately for different modules when the program is running, as follows:  
@@ -71,15 +72,17 @@ The main thread entrusts the child thread to execute:
 
 ## 7. Graceful exit process
 When receiving signals: SIGINT, SIGTERM, SIGQUIT, SIGPWR, it will execute the exit process in an orderly manner and release resources. Do a clean exit.  
-![graceful exit](documents/images/0002-exit-friendly.gif)  
+![graceful exit](documents/images/0002-exit-friendly.gif)   
+This is crucial for ensuring program integrity and conducting memory analysis.  
 
 ## 8. Comprehensive exception capture mechanism
 When various program exceptions occur in the program, such as: segment fault, assertion, bus error, exception not caught, etc., the framework will capture and print the complete call stack in the log system. Facing program crashes, no longer look blank. The effect is as follows:  
 ![stack print](documents/images/0006-error-dump.png)  
+With this mechanism in place, analyzing program crash issues of online devices becomes very simple, and there is basically no need to use gdb to analyze coredumps.  
 
 ## 9. Visualized trace module
-The trace module can record the time and duration of each execution of the marked function, and can export the flame graph for display:  
-![trace fire graph](documents/images/0011-trace-view.png)  
+The trace module can record the time and duration of each execution of the marked function, and can export the icicle diagram for display:  
+![icicle diagram](documents/images/0011-trace-view.png)  
 It is a great tool for performance analysis and event blocking problem troubleshooting. 
 
 ## 10. Simple and visualized state machine

README_CN.md

@@ -52,6 +52,7 @@
 日志内容包含了:等级、时间(精确到微秒)、线程号、模块名、函数名、正文、文件名、行号。  
 方便快速定位问题。  
 ![日志打印展示](documents/images/0002-log-show.png)  
+如上图所示，不同的日志等级颜色分明，特别是针对异常，采用红黄两色配合高亮进行显示，非常醒目。
 
 **3) 灵活的日志输出过滤器,且能运行时修改**  
 可在程序运行时针对不同的模块单独设置日志等级,如下:  
@@ -70,16 +71,19 @@
 ![ThreadPool示例](documents/images/0004-run-thread-pool.png)  
 
 ## 7. 支持优雅的退出流程
-在接收到信号:SIGINT, SIGTERM, SIGQUIT, SIGPWR 时,会有序地执行退出流程,释放资源。做到干净地退出。   
-![友好地退出](documents/images/0002-exit-friendly.gif)  
+在接收到信号:SIGINT, SIGTERM, SIGQUIT, SIGPWR 时，会有序地执行退出流程，释放资源。做到干净地退出。   
+![友好地退出](documents/images/0002-exit-friendly.gif)   
+这对保证程序的完整性与开展内存分析非常重要。
 
 ## 8. 有全面的异常捕获机制
-当程序出现各种程序异常,如:段错误、断言、总线错误、异常未捕获等,架框会捕获并在日志系统中打印完整的调用栈。面对程序崩溃,不再一脸茫然。效果如下:  
+当程序出现各种程序异常，如:段错误、断言、总线错误、异常未捕获等，架框会捕获并在日志系统中打印完整的调用栈。面对程序崩溃，不再一脸茫然。  
+效果如下:  
 ![异常栈打印](documents/images/0006-error-dump.png)  
+有该机制后，分析线上设备的程序崩馈问题便非常简单，基本上不需要使用gdb对coredump进行分析。
 
 ## 9. 有可视化的运行跟踪模块
-trace模块能记录被标记的函数每次执行的时间点与时长，可导出火焰图进行展示：  
-![trace火焰图](documents/images/0011-trace-view.png)  
+trace模块能记录被标记的函数每次执行的时间点与时长，可导出冰锥图进行展示：  
+![trace冰锥图](documents/images/0011-trace-view.png)  
 它是性能分析、事件阻塞问题排查了一大利器。
 
 ## 10. 有简单易用的状态机