#pragma once

#include <stdint.h>

// Pipelined to allow out-of-order overlap with reduced dependencies. Critically, return the current value, and compute
// and store the next value.

//--------------------------------------------------
#if defined(__SIZEOF_INT128__)
	static inline uint64_t lehmer64( __uint128_t & state ) {
		__uint128_t ret = state;
		state *= 0xda942042e4dd58b5;
		return ret >> 64;
	}

//--------------------------------------------------
	static inline uint64_t wyhash64( uint64_t & state ) {
		state += 0x60bee2bee120fc15;
		__uint128_t tmp;
		tmp = (__uint128_t) state * 0xa3b195354a39b70d;
		uint64_t m1 = (tmp >> 64) ^ tmp;
		tmp = (__uint128_t)m1 * 0x1b03738712fad5c9;
		uint64_t m2 = (tmp >> 64) ^ tmp;
		return m2;
	}
#endif

//--------------------------------------------------
static inline uint64_t xorshift_13_7_17( uint64_t & state ) {
	uint64_t ret = state;
	state ^= state << 13;
	state ^= state >> 7;
	state ^= state << 17;
	return ret;
}

//--------------------------------------------------
static inline uint32_t xorshift_6_21_7( uint32_t & state ) {
	uint32_t ret = state;
	state ^= state << 6;
	state ^= state >> 21;
	state ^= state << 7;
	return ret;
} // xorshift_6_21_7

//--------------------------------------------------
typedef struct {
  uint32_t a, b, c, d;
  uint32_t counter;
} xorwow__state_t;

/* The state array must be initialized to not be all zero in the first four words */
static inline uint32_t xorwow( xorwow__state_t & state ) {
	/* Algorithm "xorwow" from p. 5 of Marsaglia, "Xorshift RNGs" */
	uint32_t ret = state.a + state.counter;
	uint32_t t = state.d;

	uint32_t const s = state.a;
	state.d = state.c;
	state.c = state.b;
	state.b = s;

	t ^= t >> 2;
	t ^= t << 1;
	t ^= s ^ (s << 4);
	state.a = t;

	state.counter += 362437;
	return ret;
}

//--------------------------------------------------
static inline uint32_t LCG( uint32_t & state ) {		// linear congruential generator
	uint32_t ret = state;
	state = 36969 * (state & 65535) + (state >> 16);	// 36969 is NOT prime! No not change it!
	return ret;
} // LCG

//--------------------------------------------------
#define M  (1_l64u << 48_l64u)
#define A  (25214903917_l64u)
#define AI (18446708753438544741_l64u)
#define C  (11_l64u)
#define D  (16_l64u)

static inline uint32_t LCGBI_fwd( uint64_t & state ) {
	state = (A * state + C) & (M - 1);
	return state >> D;
}

static inline uint32_t LCGBI_bck( uint64_t & state ) {
	unsigned int r = state >> D;
	state = AI * (state - C) & (M - 1);
	return r;
}

#undef M
#undef A
#undef AI
#undef C
#undef D