Defference between calling Functions defined in same unit and different units

Posted on 2004-08-05
Last Modified: 2010-04-05


    I made two projects. In first one, from the main unit I am calling 20 different functions defined in a single unit. In second project all these 20 functions made seperated in to 20 different units. The exe size of second one is found 1KB more than that of the first one.

   1) What is the exact reason for this exe size differnce?
   2) Is there any differnce in performace between the two exes (I couln't notice any. All the 20 functions were just a showmessage popup) ? Why?

  Pls give me some valuable comments.


Question by:binu_emmes

Expert Comment

ID: 11725970
Each unit that interfaces procedures necessitates making the routine "far", whereas when they are NOT interfaced, they are "near", in the same code segment. The compiler can optimize the call to near routines/procedures/functions. Interfaced procedures need some extra code, which I believe you found to be added a few more times becuase of all the extra units.

There probably is a speed difference, but it wouldn't be noticed until it was over several hundred or thousand calls to the procedure.

Just the nutshell version.
LVL 17

Accepted Solution

Wim ten Brink earned 50 total points
ID: 11727942
Go to Project|Options, tabpage "Linker" and set "Map file" to "Detailed" for any of your projects. Then build that project and next to the resulting binary, you'll see a map file. If you look inside this map file, you can see which units are used, where they are located in the SINGLE codesegment and where theeir data is stored in the data segment. You will see lines like:

 Start         Length     Name                   Class
 0001:00000000 00055794H .text                   CODE
 0002:00000000 00001684H .data                   DATA
 0002:00001684 00002F79H .bss                    BSS

And this tells you where the code and data segments are. Next, you get lines like:

 0001:00000000 000050D0 C=CODE     S=.text    G=(none)   M=System   ACBP=A9
 0001:000050D0 00000174 C=CODE     S=.text    G=(none)   M=SysInit  ACBP=A9
 0001:00005244 00000B14 C=CODE     S=.text    G=(none)   M=Windows  ACBP=A9
 0001:00005D58 00000038 C=CODE     S=.text    G=(none)   M=Messages ACBP=A9
 0001:00005D90 000002C8 C=CODE     S=.text    G=(none)   M=SysConst ACBP=A9

The first four digits tell you in which segment this part of the code is located. Next you get the start position and length of the code that's put in your project. Then some information about what kind of data (code) is stored in that location and with the M=xxx you will know which unit this piece of binary belongs to.

A bit further you'll see a list of addresses of where your procedures and variables are located in the binary. And again, these addresses start with the segment ID of four digits.
Below this you'll see an overview of where your lines of code can be found in the binary. Then a list of resources included in your executable. And at the end you'll see the start address of your program.

LRHGuy is talking about near and far jumps but no... This is the old 16-bits way of thinking, where memory was still divided in multiple segments and calls were either far or near. But in 32-bits Windows you don't use many different code segments anyway. (You can, but there's no real use for it anymore.)

The growth could actually just be related to the fact that you just added another unit with a bit of code to your application. If every additional unit adds another 50 bytes to your application for it's initialization/finalization or whatever else, you will indeed notice a 1 KB (50x20) growth with 20 such units...
There's no difference in performance, though. One unit or 20 units, all function calls are just near calls...

Author Comment

ID: 11759297


    Thanks Workshop_Alex and LRHGuy for your comments.


Featured Post

Free Tool: ZipGrep

ZipGrep is a utility that can list and search zip (.war, .ear, .jar, etc) archives for text patterns, without the need to extract the archive's contents.

One of a set of tools we're offering as a way to say thank you for being a part of the community.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Suggested Solutions

Creating an auto free TStringList The TStringList is a basic and frequently used object in Delphi. On many occasions, you may want to create a temporary list, process some items in the list and be done with the list. In such cases, you have to…
Introduction I have seen many questions in this Delphi topic area where queries in threads are needed or suggested. I know bumped into a similar need. This article will address some of the concepts when dealing with a multithreaded delphi database…
The Email Laundry PDF encryption service allows companies to send confidential encrypted  emails to anybody. The PDF document can also contain attachments that are embedded in the encrypted PDF. The password is randomly generated by The Email Laundr…
In an interesting question ( here at Experts Exchange, a member asked how to split a single image into multiple images. The primary usage for this is to place many photographs on a flatbed scanner…

821 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question