Module float
Floating-point limit constants (C89/C99/C11/C17/C23) All values assume IEEE 754 binary32 (float) and binary64 (double).
Macros
| FLT_RADIX | Radix of floating-point representation (always 2 for IEEE 754). |
| FLT_MANT_DIG | float — mantissa bits (including implicit leading 1 bit) |
| FLT_DIG | float — significant decimal digits |
| FLT_MIN_EXP | float — minimum exponent (unbiased) |
| FLT_MAX_EXP | float — maximum exponent |
| FLT_MIN_10_EXP | float — minimum decimal exponent |
| FLT_MAX_10_EXP | float — maximum decimal exponent |
| FLT_EPSILON | float — machine epsilon (2^-23) |
| FLT_MIN | float — smallest positive normalised value |
| FLT_MAX | float — largest finite value |
| FLT_TRUE_MIN | float — smallest positive value (subnormal) |
| DBL_MANT_DIG | double — mantissa bits |
| DBL_DIG | double — significant decimal digits |
| DBL_MIN_EXP | double — minimum exponent (unbiased) |
| DBL_MAX_EXP | double — maximum exponent |
| DBL_MIN_10_EXP | double — minimum decimal exponent |
| DBL_MAX_10_EXP | double — maximum decimal exponent |
| DBL_EPSILON | double — machine epsilon (2^-52) |
| DBL_MIN | double — smallest positive normalised value |
| DBL_MAX | double — largest finite value |
| DBL_TRUE_MIN | double — smallest positive value (subnormal) |
| LDBL_MANT_DIG | long double — treated as double on LP64 / LLP64 without 80-bit FPU extension |
| LDBL_DIG | |
| LDBL_MIN_EXP | |
| LDBL_MAX_EXP | |
| LDBL_EPSILON | |
| LDBL_MIN | |
| LDBL_MAX | |
| FLT_ROUNDS | Decimal rounding mode: 1 = round to nearest (IEEE 754 default). |
| FLT_EVAL_METHOD | Evaluation method: 0 = each operation evaluated in its declared type. |
| DECIMAL_DIG | Decimal digits needed to represent any long double value uniquely. |