Analysing Dump Files
Data Flow Analysis
This week I continued in my search for reference in dump files analysis. In the beginning, I found something about Data Flow Analysis and when I look inside the content, it looks like exactly like Mathematics, probability all over again. There are a lot of symbols and expressions that I have not seen before, probably because I did not learn about data flow analysis before. At first I was quite shocked to see data flow will involve things like Maths in it, I thought it would be more on coding and stuff like that, before that’s the things that I’m looking in the past few weeks. I have not gone in-depth with data flow analysis as I was mainly analysing if it is related and helpful in my case. However, as the name data flow suggests, it is a process of deriving information about the run time behaviour of a program. This is related to what I need to find, which is the data parsed by plugin events.
Remember the sequential number that I mentioned in the last post, such as D. 3369, apparently I have that it is actually the internal UID for temporary variables. They represent the parameters of the calls made in program.
Comparison of dump files
Some important tree dump files were chosen to be studied such as
.cfg. These files are more important to be studied than others as I have read from online post and some slides. I have made some brief difference analysis. I noticed that there is difference in terms of the variable sequence. For example in
.GIMPLE dump file, the variable sequence is in ascending order, then in another dump file
.lower, it is in reverse sequence. I believe there is some reason for this which is yet to find out.