Most compilers under Minix, such as Turbo C / C , Borland C / C , M $ C / C , GCC, VC , and compilation processes are all "advanced languages" -> "asm" -> "OBJ". (Of course, TurBopascal is an exception, it can skip the generation of the ASM file and directly generate the target code).
The CC under Minix is no exception.
In general, the first code executed when Minix starts in MPX386.S, but # $% # $! @, How can I not understand, although the few MOVs inside it, JMP are old friends. , I can see the ancient quasial pseudo inside, I can not find the north. I have been in the Internet for a long time. Although I didn't find anything, I feel that ENGLISH has improved a lot: P.
No way, it seems that only liy (learn it by yourself, self-study).
The "self-study" said here is not the kind of self-study that is about what books holding a book. It is called "learning". Self-study here is "self-study". But now the question is from where to start? My first reaction is "Hello World!", I don't know what you are: P.
First, two paragraphs mentioned that the CC's compilation process is: SRC-> ASM-> OBJ. Yes, start from here, convert the .C file into .s file with CC.
Enter Minix first, use Man to check the CC's instructions, because it is a lookup and assembly option, so I use commands: #man cc | grep assem. So the screen shows a few lines with Assem, of course, only this line is the most useful: "- S Transform The Input Files To Assembly Files with SUFFIX .s.".
Then try it now, first come to hello.c, the code is simple: int main () {Printf ("Hello World!");}. Then convert it into .s file: "cc -s hello.c". OK, now a standard .s framework is coming out:
1] .sect .text; .sect.ROM; .sect .data; .sect .bss2] .extern_main3] .sect .text4] _main: 5] Push EBP6] MOV EBP, ESP7] Push i_18] call _printf9] POP ECX10] RET12] .SECT.ROM13] i_1: 14] .data4 181904314415] .data4 187007806316] .data4 56022949017] .data4 018] .sect .text .text .text .Text
Ok, now analyze this .s file structure:
1st: line: SECT should be the meaning of section, then.sect .text; explains a code segment ,.SECT .DATA is an explanation of a data segment (initialization) ,. subject .bss is an explanation of a data segment (initialization at the end There is also one. Sect .rom unclear, it doesn't matter, then look down: 2nd] line: It is easy to understand, indicating a label named _main.
Section 3: Row: OK, the code begins.
Sequence 4] to 11] Row: Do not skip it within the scope of this article.
12th] to 17: line: It seems that these lines are the value of "Hello World!". Oh, understand ,.rom represents constant meaning.
OK, here, a simple .s file framework is finished, if you are not satisfied, you can do it yourself: "Do Not try to memorize, learn by usage.". However, in the process of analysis, don't put a "#include
