The arithmetic type specifiers are built up from the following keywords: void, char, int, float and double, together with the prefixes short, long, signed and unsigned. From these keywords you can build both integral and floating-point types.

Due to the code compatibility with other compilers, the size of some integral types and floating point types is user selectable via Edit Project window. Please see the tables below for the corresponding arithmetic types size.

Integral Types

The types char and int, together with their variants, are considered to be integral data types. Variants are created by using one of the prefix modifiers short, long, signed and unsigned.

In the table below is an overview of the integral types – keywords in parentheses can be (and often are) omitted.

The modifiers signed and unsigned can be applied to both char and int. In the absence of the unsigned prefix, signed is automatically assumed for integral types. The only exception is char, which is unsigned by default. The keywords signed and unsigned, when used on their own, mean signed int and unsigned int, respectively.

The modifiers short and long can only be applied to int. The keywords short and long, used on their own, mean short int and long int, respectively.

An overview of the integral types is shown in the table below:

Floating-point Types

The types float and double, together with the long double variant, are considered to be floating-point types. Floating point in the mikroC PRO for ARM is implemented using the 32-bit IEEE 754 format type.

An overview of the floating-point types is shown in the table below:

Want more examples and libraries? 
Find them on