Skip to content

Latest commit

 

History

History
108 lines (82 loc) · 4.12 KB

README.md

File metadata and controls

108 lines (82 loc) · 4.12 KB
Raw

StdLogicVector

An easy to use C++ class, offering arbitrary-precision integer arithmetic for basic operations often required during the development of hardware designs. The StdLogicVector class provides a wrapper around the GMP library, which should simplify the development of software models for hardware designers in the field of cryptographic applications.

Motivation

When developing hardware designs in the field of cryptographic applications, it is pretty common to work with large integers (128bits and way beyond). While hardware description languages (HDLs), like VHDL, provide arbitrary-precision arithmetic based on data types such as std_logic_vector and the like, writing software models for these designs often requires custom-made solutions or the use of third party libraries.

In order to simplify the development of software models, I have put together the StdLogicVector class, which is a small wrapper around the GMP library, offering arbitrary-precision arithmetic for operations frequently used in the design of hardware architectures (bit-wise operations such as those of logic gates, arithmetic functions, etc.).

Example

Everything, what can be done with the StdLogicVector class, can, of course, also be done with plain GMP data types and its functions. However, StdLogicVector offers a more intuitive application (somewhat similar to the BigInteger class of Java), which allows to concatenate operations like the following:

// Create a four-bit integer 'a' with value 5 and an eight-bit integer 'b' with
// value 210, initialized using binary representation.
StdLogicVector a = StdLogicVector(5, 4);
StdLogicVector b = StdLogicVector("11010010", 2, 8);

// Perform some operations on 'b'.
b.ShiftRight(2).TruncateAfter(4).Xor(a);

Usage

In order to clearly separate the StdLogicVector class from any other project I am using it in, I normally build a shared library from the sources and link my software models against that library. If you want to do something similar, you may stick to the following steps (tested on a CentOS 6.4 installation):

Prerequisites

  1. Make sure that you have available a GMP library against which you can link the StdLogicVector sources later on (although it should work with earlier versions as well, the version I usually use is 6.0.0a).

  2. Make sure that you have Google's C++ Testing Framework in order to run the unit tests, which come with the StdLogicVector class.

Building the library

  1. Download the latest version of the sources.

  2. Take a look at the provided Makefile and adapt it according to your needs (at least the paths to the GMP and the Google Test files will have to be adapted).

  3. Finally, build the shared library, as well as the executable running the unit tests, and test the library.

make
make test
make run

Documentation

A documentation of the StdLogicVector class, created using Doxygen, is available here.

##FAQ

A couple of questions, which came to my mind when setting up this readme file.

Is the StdLogicVector class intended for any special purpose?

No, I have written the StdLogicVector class to simplify working with arbitrary-precision integers in general, without having any special purpose in mind. Therefore, you will most-likely miss some operations when using it for your own needs.

Which is the "final" version of the StdLogicVector class?

There is nothing like a "final version" of this class. I am extending its functionality depending on my personal needs (which again depend on the projects I am currently working on), which is why there exists (and will most-likely never exist) something like a "final" version.