读了老罗的“仅通过崩溃地址找出源代码的出错行”(下称"罗文")一文后，感觉该文还是可以学到不少东西的。不过文中尚存在有些说法不妥，以及有些操作太繁琐的地方 。为此，本人在学习了此文后，在多次实验实践基础上，把该文中的一些内容进行补充与改进，希望对大家调试程序，尤其是release版本的程序有帮助 。欢迎各位朋友批评指正-old Romanian read the "only by the collapse of addresses to identify the source code error" (hereinafter referred to as "Luo"), a text, the feeling of the text can still learn a lot of things. But the text of the surviving some of the comments inappropriate, and too cumbersome to operate in some areas. Therefore, I studied in the article, the practice of several experiments, based on the text of some elements to complement and improve, we hope to debug procedures, particularly release versions of the program would be helpful. Welcome ladies and correct criticism