Integer types and constants

Signed integers can be of these types (the int after short, or long is optional):

signed char c = 127; /* required to be 1 byte, see remarks for further information. */
signed short int si = 32767; /* required to be at least 16 bits. */
signed int i = 32767; /* required to be at least 16 bits */
signed long int li = 2147483647; /* required to be at least 32 bits. */

signed long long int li = 2147483647; /* required to be at least 64 bits */

Each of these signed integer types has an unsigned version.

unsigned int i = 65535;
unsigned short = 2767;
unsigned char = 255;

For all types but char the signed version is assumed if the signed or unsigned part is omitted. The type char constitutes a third character type, different from signed char and unsigned char and the signedness (or not) depends on the platform.

Different types of integer constants (called literals in C jargon) can be written in different bases, and different width, based on their prefix or suffix.

/* the following variables are initialized to the same value: */
int d = 42;   /* decimal constant (base10) */
int o = 052;  /* octal constant (base8) */
int x = 0xaf; /* hexadecimal constants (base16) */
int X = 0XAf; /* (letters 'a' through 'f' (case insensitive) represent 10 through 15) */

Decimal constants are always signed. Hexadecimal constants start with 0x or 0X and octal constants start just with a 0. The latter two are signed or unsigned depending on whether the value fits into the signed type or not.

/* suffixes to describe width and signedness : */
long int i = 0x32; /* no suffix represent int, or long int */
unsigned int ui = 65535u; /* u or U represent unsigned int, or long int */
long int li = 65536l; /* l or L represent long int */

Without a suffix the constant has the first type that fits its value, that is a decimal constant that is larger than INT_MAX is of type long if possible, or long long otherwise.

The header file <limits.h> describes the limits of integers as follows. Their implementation-defined values shall be equal or greater in magnitude (absolute value) to those shown below, with the same sign.

Macro | Type | Value —— | —— | —— CHAR_BIT | smallest object that is not a bit-field (byte) | 8 SCHAR_MIN | signed char | -127 / -(27 - 1) SCHAR_MAX | signed char | +127 / 27 - 1 UCHAR_MAX | unsigned char | 255 / 28 - 1 CHAR_MIN | char | see below CHAR_MAX | char | see below SHRT_MIN | short int | -32767 / -(215 - 1) SHRT_MAX | short int | +32767 / 215 - 1 USHRT_MAX | unsigned short int | 65535 / 216 - 1 INT_MIN | int | -32767 / -(215 - 1) INT_MAX | int | +32767 / 215 - 1 UINT_MAX | unsigned int | 65535 / 216 - 1 LONG_MIN | long int | -2147483647 / -(231 - 1) LONG_MAX | long int | +2147483647 / 231 - 1 ULONG_MAX | unsigned long int | 4294967295 / 232 - 1

Macro | Type | Value —— | —— | —— LLONG_MIN | long long int | -9223372036854775807 / -(263 - 1) LLONG_MAX | long long int | +9223372036854775807 / 263 - 1 ULLONG_MAX | unsigned long long int | 18446744073709551615 / 264 - 1

If the value of an object of type char sign-extends when used in an expression, the value of CHAR_MIN shall be the same as that of SCHAR_MIN and the value of CHAR_MAX shall be the same as that of SCHAR_MAX . If the value of an object of type char does not sign-extend when used in an expression, the value of CHAR_MIN shall be 0 and the value of CHAR_MAX shall be the same as that of UCHAR_MAX.

The C99 standard added a new header, <stdint.h>, which contains definitions for fixed width integers. See the fixed width integer example for a more in-depth explanation.