/*
Written in 2018 by David Blackman and Sebastiano Vigna (vigna@acm.org)
Modified for his own use by kio in 2020.

To the extent possible under law, the author has dedicated all copyright
and related and neighboring rights to this software to the public domain
worldwide. This software is distributed without any warranty.

See <http://creativecommons.org/publicdomain/zero/1.0/>.
See	<http://prng.di.unimi.it>
*/

#include "Xoshiro256.h"
#include <math.h>
#include <random>


namespace kio
{

static inline uint64 rotl(const uint64 x, int k) noexcept { return (x << k) | (x >> (64 - k)); }

static inline int msbit(uint32 n) noexcept // 0 .. 31
{
	int b = 0, i = 16;
	do {
		if (n >> i)
		{
			n >>= i;
			b += i;
		}
	}
	while (i >>= 1);
	return b;
}

static inline int msbit(uint64 n) noexcept // 0 .. 63
{
	int b = 0, i = 32;
	do {
		if (n >> i)
		{
			n >>= i;
			b += i;
		}
	}
	while (i >>= 1);
	return b;
}

/*	This is xoshiro256+ 1.0, our best and fastest generator for floating-point
	numbers. We suggest to use its upper bits for floating-point
	generation, as it is slightly faster than xoshiro256++/xoshiro256**. It
	passes all tests we are aware of except for the lowest three bits,
	which might fail linearity tests (and just those), so if low linear
	complexity is not considered an issue (as it is usually the case) it
	can be used to generate 64-bit outputs, too.

	Use a sign test to extract a random Boolean value,
	and	right shifts to extract subsets of bits.
	The state must be seeded so that it is not everywhere zero.
*/
uint64 xoshiro256p(uint64* s) noexcept
{
	const uint64 result = s[0] + s[3];
	const uint64 t		= s[1] << 17;

	s[2] ^= s[0];
	s[3] ^= s[1];
	s[1] ^= s[2];
	s[0] ^= s[3];

	s[2] ^= t;

	s[3] = rotl(s[3], 45);

	return result;
}

/*	This is xoshiro256++ 1.0, one of our all-purpose, rock-solid generators.
	It has excellent (sub-ns) speed, a state (256 bits) that is large
	enough for any parallel application, and it passes all tests we are
	aware of.

	For generating just floating-point numbers, xoshiro256+ is even faster.
	The state must be seeded so that it is not everywhere zero.
*/
uint64 xoshiro256pp(uint64* s) noexcept
{
	const uint64 result = rotl(s[0] + s[3], 23) + s[0];
	const uint64 t		= s[1] << 17;

	s[2] ^= s[0];
	s[3] ^= s[1];
	s[1] ^= s[2];
	s[0] ^= s[3];

	s[2] ^= t;

	s[3] = rotl(s[3], 45);

	return result;
}

/*	This is xoshiro256** 1.0, one of our all-purpose, rock-solid
	generators. It has excellent (sub-ns) speed, a state (256 bits) that is
	large enough for any parallel application, and it passes all tests we
	are aware of.

	For generating just floating-point numbers, xoshiro256+ is even faster.
	The state must be seeded so that it is not everywhere zero.
*/
uint64 xoshiro256ss(uint64* s) noexcept
{
	const uint64 result = rotl(s[1] * 5, 7) * 9;
	const uint64 t		= s[1] << 17;

	s[2] ^= s[0];
	s[3] ^= s[1];
	s[1] ^= s[2];
	s[0] ^= s[3];

	s[2] ^= t;

	s[3] = rotl(s[3], 45);

	return result;
}


inline double random(uint64 rand) noexcept
{
	// interval [0 .. [1

	return ldexp(double(rand), -64);
}

inline double random(uint64 rand, double max) noexcept
{
	// interval [0 .. [max

	return ldexp(double(rand) * max, -64);
}

uint32 Xoshiro256::random(uint32 max) noexcept
{
	// interval [0 .. [max

	if (max == 0u) return uint32(next());

	const int n = 31 - msbit(max);

	for (;;)
	{
		uint32 r = uint32(next()) >> n;
		if (r < max) return r;
	}
}

uint64 Xoshiro256::random(uint64 max) noexcept
{
	// interval [0 .. [max

	if (max == 0u) return next();

	const int n = 63 - msbit(max);

	for (;;)
	{
		uint64 r = next() >> n;
		if (r < max) return r;
	}
}

void Xoshiro256::_skip(const uint64 bits[4]) noexcept
{
	// helper for jump() and jump_long():

	uint64 s0 = 0;
	uint64 s1 = 0;
	uint64 s2 = 0;
	uint64 s3 = 0;

	for (uint i = 0; i < 256 / 64; i++)
		for (uint b = 0; b < 64; b++)
		{
			if (bits[i] & UINT64_C(1) << b)
			{
				s0 ^= s[0];
				s1 ^= s[1];
				s2 ^= s[2];
				s3 ^= s[3];
			}
			next();
		}

	s[0] = s0;
	s[1] = s1;
	s[2] = s2;
	s[3] = s3;
}

void Xoshiro256::jump() noexcept
{
	// This is the short jump function for the generator. It is equivalent
	// to 2^128 calls to next(); it can be used to generate 2^128
	// non-overlapping subsequences for parallel computations.

	static const uint64 bits[4] = {0x180ec6d33cfd0aba, 0xd5a61266f0c9392c, 0xa9582618e03fc9aa, 0x39abdc4529b1661c};

	_skip(bits);
}

void Xoshiro256::jump_long() noexcept
{
	// This is the long jump function for the generator. It is equivalent to
	// 2^192 calls to next(); it can be used to generate 2^64 starting points,
	// from each of which skip128() will generate 2^64 non-overlapping
	// subsequences for parallel distributed computations.

	static const uint64 bits[4] = {0x76e15d3efefdcbbf, 0xc5004e441c522fb3, 0x77710069854ee241, 0x39109bb02acbe635};

	_skip(bits);
}

Xoshiro256::Xoshiro256(const void* q) noexcept
{
	memcpy(s, q, sizeof s);
	next();
}

Xoshiro256::Xoshiro256(uint32 seed) noexcept
{
	srand(seed);
	s[0] = unsigned(rand());
	s[1] = unsigned(rand());
	s[2] = unsigned(rand());
	s[3] = unsigned(rand());
	next();
}

Xoshiro256::Xoshiro256() noexcept : Xoshiro256(std::random_device {}()) {}

} // namespace kio