When the compiler finds a “D” immediately after a floating point constant value, it treats that value as data of type double.
When the pragma float_constants is enabled, floating point constants should end with a “D” so that the compiler does not treat them as values of type float.
For related information, see “float_constants”.