iMac Matlab Integer addition using dynamic library

Supports signed and unsigned 8,16,32,64bit integers with overflow.

This project uses a dynamic librarys named to perform x64 integer register adds for 8,16,32,64bit integers with overflow.

These files were built on an iMac (Retina 5K, 27-inch, 2017) with Intel 64x (i5)

No support for M1 Macs or Windows. (as of yet)

Built using:

Xcode 14.2 (14c18). You will need to install Xcode 14.2

You will need Matlab 2017b.

You will need to download the shared libraries from my Github repository:

browdone1/db_git_imac_x64_integer_sum

NOTE: The shared library libdysum.dylib does not always work when down loaded from Github.

THEREFORE: I am including mex file versions of the shared library functions.

The mex files have been demonstrated to work when copied to your Matlab path.

IF USING dylib:

Copy libdysum.dylib (from Github) and *.m, *.h fies to your Matlab working directory.

Note: When downloading single text or m files from Github:

First select file, then when it opens select "raw".

Then when it opens again, right click and select save as.

To check for virus:

objdump -D -C libdysum.dylib

This lists code for each function, mainly a bunch of argument loads and with one "add" for each function.

IF USING MEX files:

Copy .m testdy*.m files to Matlab working directory.

Copy file in folder mex/*.* (mex files from Github) to your Matlab working path.

For example:

The testdySumInt16.m will demonstrate integer overflow issues with Matlab and demonstrate

how to make calls to shared libraries. All test files have a USE_MEX variable

that can be edited to select whether dylib or mex shared libraries are to be used.