i made a random console application and loaded into IDA. The start of .text section shown is 00BA1000. Then i loaded the application into CFF explorer. The address of entry point was 000110AA. Is it the offset to the function? If yes, is it the raw offset (in the exe file) or virtual offset (after the exe is loaded in memory) ? Then i saw there was a key named 'base of code'. Why isnt addressofentrypoint the same as base of code? I mean, they are start of the text section too right? With IDA , i can know the start of .text section is 00BA1000. But, how can i know it with CFF explorer? With CFF, i can know the virtual address of text section is 00011000. So when the program is loaded into memory, the start of the text section should be imagebase + 00011000 right? With ollydbg i can know its wrong. The calculated address isnt the correct one, but the one shown in IDA is. These questions messed me up. Hope you can help me !!!

CFF explorer (basically helps you look into PE header) : https://ntcore.com/?page_id=388

Screenshots: https://i.sstatic.net/Fj8vy.jpg



Thanks a lot for reading!

  • When you open an application in IDA, it is automatically loaded to an address which is usually random. you can rebase it to a different address(edit->segments->rebase). To look at different segments go to view->open subviews->segments or press Shift+F7. I don't know if there's a rebase feature in CFF explorer.
    – sudhackar
    Commented Jul 19, 2018 at 5:41
  • 3
    there's a bit of chaos in your question. Would you mind rewriting it so it's more clear? Commented Jul 19, 2018 at 8:31

1 Answer 1


Base of code is where the code starts
Address of Entrypoint is Where the executable starts executing (the main function's first instruction)

they both can be same or different

here is an example where they both are same

:\>dumpbin /headers funadd.exe | grep -iE "base of code|entry point"
            1000 entry point (00401000) _WinMain@16
            1000 base of code

that is becasue the main function does not refer any other function

suppose you have code like this

int main (void) { printf ( "%x\n" , dosomecrap() ); }   

then when compiling the main function needs to refer to the dosomecrap() function so the compiler will compile the dosomecrap() first and put the code from base of code

in this case main () comes later in the code section
so they both wont be same

Address of Entry Point will point to main()
Base of code may point to start of dosomecrap()

in normal compiled executables main is not the first code that is exceuted

it is cruntime init code like maincrtstartup() or WinMainCrtStartup()

that is executed first

import section can be merged into text section

in that case the resolved imports start at base of code

you can write code like this and merge data section into .text section in this case the embedded bytes would be the first bytes at base of code

#include <windows.h>

int p[] = { '\x90\x90\x90\x90' };

int funAdd() {
    int myvar = 2; 
    return myvar + p[0];

int CALLBACK  WinMain(_In_ HINSTANCE,_In_opt_ HINSTANCE,_In_ LPSTR,_In_ int) {

if you compile this with

cl /Zi /W4 /analyze /EHsc /Od /nologo funadd.cpp /link /release /subsystem:windows /entry:WinMain /nologo /merge:.data=.text

and check you can see bith base of code an address of entry point differ

Address of entry point has been shifted 16 bytes or 0x10 bytes (alignment requirement says code should be aligned to 16 byte boundary )

:\>dumpbin /headers funadd.exe | grep -iE "base of code|entry point"
            1010 entry point (00401010) _WinMain@16
            1000 base of code

you can see the 0x90 as below

:\>dumpbin /disasm /range:0x401000,0x401016  funadd.exe
Microsoft (R) COFF/PE Dumper Version 14.14.26430.0
Copyright (C) Microsoft Corporation.  All rights reserved.

Dump of file funadd.exe


  00401000: 90 90 90 90                                      ....
  00401004: 00 00 00 00 00 00 00 00 00 00 00 00              ............
  00401010: 55                 push        ebp
  00401011: 8B EC              mov         ebp,esp
  00401013: E8 08 00 00 00     call        ?funAdd@@YAHXZ


        1000 .text
  • great explanation! but i have 2 questions here: 1. as you mentioned , data of variable p (\x90\x90\x90\x90) is located in .text section. But what should we do make it appear in the .idata section? i mean it should be more preferable to place in data section right? :) 2. Why should there be 400000 bytes of image base. Is there any purpose for it? According to msdn, it is the preferred address of the first byte of image when loaded into memory. But why is it needed to do that? Does it improve loading efficiency?
    – Mod Coc
    Commented May 14, 2019 at 14:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.