How to better and more accurate analysis of samba source codes

[杨昌玉]

      Dear everybody:
       Currently, I am analyzing  the source codes of samba. As we all know, samba source codes are very complicated, it has several million lines of code. I personally feel, the samba's help documentation write more chaos, the structure is not clear, for example: www.samba.org/samba/docs/man
       Because samba has over 30 years of history, historical documents such as samba3 documents are very large and very redundant. It is very hard to get what we want from the reference document information.
      Currently, I study the source code while reading the samba code while finding relevant online information and documentation of official website . I also use a single-step debugging tool of gdb, and strace tool to tracking the source code.
       My first question: Are there other  better ways to study the samba source code?
       For instance, my main goal is to narrow the range in order to find kerberos authentication function call interfaces in the samba source codes, and the  ultimate goal  is to find the Service_key.
      My second question: Are there any good ideas, suggestions or implementation methods to study the samba source codes?  The codes base are very large, I am very confused.
      Please provide some materials such as links to me if possible. Any help would be appreciated sincerely.
      Best wishes. 

                                                                                                                                     Yours sincerely, 
                                                                                                                                      Changyu Yang
                                                                                                                                Dalian University Of China