Compute the number of representable half-precision floating-point values that separate two half-precision floating-point numbers along the real number line.
Difference between one and the smallest value greater than one that can be represented as a half-precision floating-point number.
Convert a half-precision floating-point number to the nearest double-precision floating-point number.
Convert a half-precision floating-point number to the nearest single-precision floating-point number.
Convert a single-precision floating-point number to the nearest half-precision floating-point number.
Convert a double-precision floating-point number to the nearest half-precision floating-point number.
16-bit half-precision floating-point number.
Number of significand bits for a half-precision floating-point number.
Mask for the sign bit of a half-precision floating-point number.
Mask for the significand of a half-precision floating-point number.
Smallest positive normalized half-precision floating-point number.
Maximum half-precision floating-point number.
Mask for the exponent of a half-precision floating-point number.
Compute the number of representable double-precision floating-point values that separate two double-precision floating-point numbers along the real number line.
Return a normal number `y` and exponent `exp` satisfying `x = y * 2^exp`.
The bias of a half-precision floating-point number's exponent.
Create a double-precision floating-point number from a higher order word and a lower order word.
Split a double-precision floating-point number into a higher order word and a lower order word.
Convert a double-precision floating-point number to the nearest single-precision floating-point number.
Compute the number of representable single-precision floating-point values that separate two single-precision floating-point numbers along the real number line.
Return an integer corresponding to the unbiased exponent of a double-precision floating-point number.
Return an unsigned 32-bit integer corresponding to the more significant 32 bits of a double-precision floating-point number.
Difference between one and the smallest value greater than one that can be represented as a double-precision floating-point number.
Convert an unsigned 32-bit integer to a signed 32-bit integer.