* @copyright 2017 Jim Wigginton * @license http://www.opensource.org/licenses/mit-license.html MIT License * @link http://pear.php.net/package/Math_BigInteger */ namespace phpseclib3\Math\BigInteger\Engines\PHP\Reductions; use phpseclib3\Math\BigInteger\Engines\PHP\Montgomery as Progenitor; /** * PHP Montgomery Modular Exponentiation Engine * * @author Jim Wigginton */ abstract class Montgomery extends Progenitor { /** * Prepare a number for use in Montgomery Modular Reductions * * @param array $x * @param array $n * @param string $class * @return array */ protected static function prepareReduce(array $x, array $n, $class) { $lhs = new $class(); $lhs->value = array_merge(self::array_repeat(0, count($n)), $x); $rhs = new $class(); $rhs->value = $n; list(, $temp) = $lhs->divide($rhs); return $temp->value; } /** * Montgomery Multiply * * Interleaves the montgomery reduction and long multiplication algorithms together as described in * {@link http://www.cacr.math.uwaterloo.ca/hac/about/chap14.pdf#page=13 HAC 14.36} * * @param array $x * @param array $n * @param string $class * @return array */ protected static function reduce(array $x, array $n, $class) { static $cache = [ self::VARIABLE => [], self::DATA => [] ]; if (($key = array_search($n, $cache[self::VARIABLE])) === false) { $key = count($cache[self::VARIABLE]); $cache[self::VARIABLE][] = $x; $cache[self::DATA][] = self::modInverse67108864($n, $class); } $k = count($n); $result = [self::VALUE => $x]; for ($i = 0; $i < $k; ++$i) { $temp = $result[self::VALUE][$i] * $cache[self::DATA][$key]; $temp = $temp - $class::BASE_FULL * ($class::BASE === 26 ? intval($temp / 0x4000000) : ($temp >> 31)); $temp = $class::regularMultiply([$temp], $n); $temp = array_merge(self::array_repeat(0, $i), $temp); $result = $class::addHelper($result[self::VALUE], false, $temp, false); } $result[self::VALUE] = array_slice($result[self::VALUE], $k); if (self::compareHelper($result, false, $n, false) >= 0) { $result = $class::subtractHelper($result[self::VALUE], false, $n, false); } return $result[self::VALUE]; } /** * Modular Inverse of a number mod 2**26 (eg. 67108864) * * Based off of the bnpInvDigit function implemented and justified in the following URL: * * {@link http://www-cs-students.stanford.edu/~tjw/jsbn/jsbn.js} * * The following URL provides more info: * * {@link http://groups.google.com/group/sci.crypt/msg/7a137205c1be7d85} * * As for why we do all the bitmasking... strange things can happen when converting from floats to ints. For * instance, on some computers, var_dump((int) -4294967297) yields int(-1) and on others, it yields * int(-2147483648). To avoid problems stemming from this, we use bitmasks to guarantee that ints aren't * auto-converted to floats. The outermost bitmask is present because without it, there's no guarantee that * the "residue" returned would be the so-called "common residue". We use fmod, in the last step, because the * maximum possible $x is 26 bits and the maximum $result is 16 bits. Thus, we have to be able to handle up to * 40 bits, which only 64-bit floating points will support. * * Thanks to Pedro Gimeno Fortea for input! * * @param array $x * @param string $class * @return int */ protected static function modInverse67108864(array $x, $class) // 2**26 == 67,108,864 { $x = -$x[0]; $result = $x & 0x3; // x**-1 mod 2**2 $result = ($result * (2 - $x * $result)) & 0xF; // x**-1 mod 2**4 $result = ($result * (2 - ($x & 0xFF) * $result)) & 0xFF; // x**-1 mod 2**8 $result = ($result * ((2 - ($x & 0xFFFF) * $result) & 0xFFFF)) & 0xFFFF; // x**-1 mod 2**16 $result = $class::BASE == 26 ? fmod($result * (2 - fmod($x * $result, $class::BASE_FULL)), $class::BASE_FULL) : // x**-1 mod 2**26 ($result * (2 - ($x * $result) % $class::BASE_FULL)) % $class::BASE_FULL; return $result & $class::MAX_DIGIT; } }