GNU Multiple Precision (GMP) Arithmetic Library for Arduino.

Port of the mini-gmp library with support for arbitrary-precision integers and rational numbers. No support for floats.

Currently based on GMP v6.2.0.

Compared to vanilla mini-gmp, this library has the following modifications:

1. In case of error, no error is printed to stderr before aborting
2. The internal limb type defaults to unsigned int (16 bit) instead of unsigned long (32 bit), since AVR microcontrollers do not natively support 32-bit multiplication.

You can install this library via the built-in library manager in your Arduino IDE (Sketch > Include Library > Manage Libraries...), searching for gmp-ino and clicking "Install".

Sample usage:

#include <gmp-ino.h>

mpz_t counter;

void setup() {
  mpz_init(counter);
}

void loop() {
  mpz_add_ui(counter, counter, 1); // equivalent to `counter++`
}

Important things to keep in mind:

• always call mp*_init before using a variable for the first time
• always call mp*_clear when the contents of a variable are not needed anymore (if you don't, you will quickly exhaust the memory as memory used by the variables is not freed until the call to mp*_clear)
• mini-gmp uses malloc/realloc/free to allocate memory for the arbitrary-precision numbers
• any internal error during any call (e.g. division by 0, invalid arguments, memory exhausted) will hang your program: validate the arguments before performing calls

• GMP website
• GMP documentation - note that mini-gmp supports only a subset of the full GMP library
• mini-gmp

2020 Carlo Alberto Ferraris.

gmp-ino is licensed under the GNU LGPL v3.

GMP is dual licensed under the GNU GPL v2 or the GNU LGPL v3.