Machine parameters (double precision floating-point arithmetic)
- Purpose
- Determines the machine parameters on double precision floating-point arithmetic.
- Returns
- float
The value specified by cmach. (0 will be returned if cmach is invalid.)
- Parameters
-
| [in] | cmach | Specifies the value to be returned.
= 'E' or 'e': Relative machine precision (eps)
= 'S' or 's': Safe minimum, such that 1/sfmin does not overflow
= 'B' or 'b': Base of the machine (base)
= 'P' or 'p': eps*base
= 'N' or 'n': Number of (base) digits in the mantissa
= 'R' or 'r': 1 when rounding occurs in addition, 0 otherwise
= 'M' or 'm': Minimum exponent before (gradual) underflow
= 'U' or 'u': Underflow threshold - base^(emin-1)
= 'L' or 'l': Largest exponent before overflow
= 'O' or 'o': Overflow threshold - (base^emax)*(1-eps) |
- Reference
- LAPACK
1.9.6