what's a difference between return and ret retn retf?
nice day

harry

return is a marco in fasm if I am not wrong. Intel specifies ret to be both retn and retf though retn and retf does not exist in Intel's manual. retn is return near which is a return to a call from a calling procedure within the current code segment, while retf is return far which is a return to a call from a calling procedure located in a different segment from the current segment.

Like Petzold said :there is no near and far addresses in 32bit modes...
so what's a difference???




what does exactly 'return' do? and why return instead of ret ?

Return is a macro that together with proc macro permits structured procedures that uses the stack.
Example: This procedure is going to use 2 dword arguments and 2 dword locals from stack:

LEAVE OPCODE C9 -> Set ESP to EBP, then pop EBP
Description: Releases the stack frame set up by an earlier ENTER instruction. The LEAVE instruction copies the frame pointer (in the EBP register) into the stack pointer register (ESP), which releases the stack space allocated to the stack frame. The old frame pointer (the frame pointer for the calling procedure that was saved by the ENTER instruction) is then popped from the stack into the EBP register, restoring the calling procedure?s stack frame.
A RET instruction is commonly executed following a LEAVE instruction to return program control to the calling procedure.

See "Procedure Calls for Block-Structured Languages" in Chapter 6 of the Intel® Architecture Software Developer's Manual, Volume 1, for detailed information on the use of the ENTER and LEAVE instructions.

Hi.

At firtst about retn/retf differences:

retn - (return from near call) loads IP (or EIP) from the stack.
retf - (return from far call) loads from the stack not only IP (EIP) but also CS:

All this is because near call changes only IP (EIP), but far call changes IP(EIP) and CS simultaneously.

ret instructions allow one imediate parameter that should be added to ESP after return address is extracted from the stack.

Here is some pseudo code describing retn/retf action:


BTW: Using of retf in Windows is very limited. You have to use it only in DOS and if you want to write your own OS.

You know that procedure arguments in stdcall calling convention should be pushed in the stack and procedure use them via [ebp+offset]. So after finishing procedure should remove these arguments from the stack - this is a purpose of imm constant in ret instruction. It simply moves esp to point below arguments. This imm constant should be as great as the number of bytes passed as arguments to the procedure.

return macro is "inteligent" return instruction for use with "proc" macro. Proc macro calculate the size of arguments and return macro use this number as imm argument of retn instruction. It simply frees the programmer from counting how many bytes he use as procedure arguments.

Regards.

now I know(thank You) what for is the return macro but still have some lack of understanding
,why there is the code like this:



why someone push and pop the registers ?

WindowProc procedure is called in the most cases from Windows. So, Windows needs these registers to be preserved, because it uses them for it's internal needs. If you never change esi, edi or ebx you don't need to preserve them in the stack. But in the most cases it is hard to track whole big WindowProc whether some message handler uses or not these registers, so the most programmers simply push them on the begin ant pop them on the end of WindowProc. It is important to preserve registers only in the code that will be called from Windows. For the code that only your program call yoo may preserve/not preserve the registers you want/need.

Regards.

Now I am fully satisfied and thx you all for explanation.
I very thank you JohnFound.

this is my curse that I am very nosy and always want to know all the details I'm interested in...


Once more THX!,

harry