B.6 Numerical Limits (§2.2.4.2)

DEC C's numerical limits are defined in the limits.h and float.h header files. These header files contain the implementation-defined values so that the following descriptions hold:

• There are 8 bits in a character of the execution character set.

• The representation and set of values for the type char are the same as that of type signed char . This equivalence can be changed from signed char to unsigned char with a command-line option.

• On OpenVMS systems, the representation and set of values for the types int and signed int are the same as that for type long (32 bits).

• On OpenVMS systems, the representation and set of values for the type unsigned int are the same as that for type unsigned long (32 bits).

• On Digital UNIX systems, the long int and unsigned long int types are 64 bits, while int and unsigned int are 32 bits.

• The representation and set of values for the type long double are the same as that for type double (64 bits).

Any limits not found in the previous list are defined as shown in the Standard.