NUTS (128-bit floats)

[mjm]

NUTS - Numerical tools - v0.1.1 alpha.

Currently the library contains only 128-bit floating-point numbers with the basic operations QAdd, QSub, QMul, QDiv, QRem, QSqrt, and QFMA. There are also string routines but there are no transcendental functions yet.

There are no signaling NaN's (all NaN's are regarded as quiet NaN's).
Any other difference, except the number of implemented functions, with the 754-2008 standard would be a bug since it would not be intentional.

As they come:
It was written with and for Free Pascal 64-bit, version 3.0.0 or higher..
There is no documentation yet (because all could still be modified) but the code is commented.
It was tested with x86_64 processors only.
It was tested with 64-bit Linux only. It could work with an other OS but I did not try.
It cannot run on big endian systems. Adapting the Pascal code should not be too difficult but I have no big endian machine in order to check the code.

License: 3-Clause BSD.

mm
http://www.ellipsa.eu