4.4.1.2 Floating-Point Numbers

A floating-point number is represented in IEEE 754 format. A floating-point number is written by writing (in order):

• an optional prefix, which consists of the digit ‘0’, followed by the letter ‘e’, ‘f ’ or ‘d’ in upper or lower case. Because floating point constants are used only with .float and .double directives, the precision of the binary representation is independent of the prefix.
• an optional sign: either ‘+’ or ‘-’.
• an optional integer part: zero or more decimal digits.
• an optional fractional part: ‘.’ followed by zero or more decimal digits.
• an optional exponent, consisting of:
  • an ‘E’ or ‘e’.
  • an optional sign: either ‘+’ or ‘-’.
  • one or more decimal digits.

At least one of the integer part or fractional part must be present. The floating-point number has the usual base-10 value.

Floating-point numbers are computed independently of any floating-point hardware in the computer running the assembler.