Difference between a function and a macro

Question

I would like to know what is the advantage of using a macro instead of declaring a function.

For example I found this line in a header file of an USB library for an ATMEL microcontroller:

#define  udd_detach_device()                 ( Clr_bits(UDP->UDP_TXVC, UDP_TXVC_PUON))

Why is it defined like this, and not like a function?

Thank you for your help.

Macro: used for text substitution in source; Function: used to aid in calculating values, setting up data structures, ... you cannot really replace one with the other — pmg, Nov 08 '19 at 13:22
Also, functions are called at run-time while macros are expanded (substituted) at compile-time. — Some programmer dude, Nov 08 '19 at 13:23
For your specific case, maybe this is a common operation and the writers simply wanted a simple way to write the expansion that doesn't result in the extra run-time overhead of calling another function. — Some programmer dude, Nov 08 '19 at 13:24
Yes I know that macro are text substition for the compiler, but I wonder what is the advantage (or disadvantage) of doing that instead of declaring a void -> voind function that performs the same operation? — Wheatley, Nov 08 '19 at 13:27

meaning-matters · Accepted Answer · 2019-11-08T13:32:25.030

2

All this #define does is replace every textual occurrence of udd_detach_device() in your source with the following text ( Clr_bits(UDP->UDP_TXVC, UDP_TXVC_PUON)).

This is done before your code is compiled, in the so called pre-processing phase.

You should see this C pre-processor (CPP) phase as a text-editing phase in which parts of the code can be replaced or left out. The CPP recognises a limited set of editing commands that all start with #.

Read this Wikipedia page for more background.

edited Nov 08 '19 at 13:32

answered Nov 08 '19 at 13:23

meaning-matters

21,929
10
82
142

What is the advantage of doing that instead of declaring a void -> void function that does the same operation? – Wheatley Nov 08 '19 at 13:25
1

@Wheatley it really depends on the use case. – Jabberwocky Nov 08 '19 at 13:26
1

@Jabberwocky in this case the macro's operation is clearing a bit in a register. – Wheatley Nov 08 '19 at 13:29
1

@Wheatley then using a macro rather than a function is probably a good idea, because if you call a function you get quite some overhead, which might be a good idea to avoid especially on micro controllers – Jabberwocky Nov 08 '19 at 13:41
@Jabberwocky Depends on compiler optimisation. Many compilers are capable of inlining simple functions. – meaning-matters Nov 08 '19 at 16:49

Difference between a function and a macro

1 Answers1