MFMathLib

A mathematical library providing basic math operations on data types from 8-bits to 1024-bits with overflow/underflow tracking.

Standard C operations such as multiply and divide do not properly track overflow, underflow and remainders. As an example, multiplying two unsigned char values of 100 together will produce a result larger than could be stored in a resulting unsigned char without any warning at runtime.

All functions defined in this library provide all the parameters and returned values necessary to ensure mathematical accuracy for all supported data types.

Licensing

Public Domain

by Freshcode, Cutting edge Mac, iPhone, iPad & Android software development. http://madefresh.ca/

Originally created by Dave Poirier on 2013-02-01.

Compatibility

C compiler with support for 32-bit unsigned integers

Performance

This library has NOT yet been optimized for performance. While its development is in its early stage priority has been put into ensuring mathematical accuracy rather than execution speed. All performance improvement contributions are welcome.

Dependencies

For the library files mfmathlib.c/.h:

• none

For the test framework:

• standard OSX environment
• Accelerate framework
• 'bc' command line tool.

Limitations

The library currently only support unsigned integer operations