/*
 * ARM generic helpers for various arithmetical operations.
 *
 * This code is licensed under the GNU GPL v2 or later.
 *
 * SPDX-License-Identifier: GPL-2.0-or-later
 */
#include "qemu/osdep.h"
#include "cpu.h"
#include "exec/helper-proto.h"
#include "qemu/crc32c.h"
#include <zlib.h> /* for crc32 */

/*
 * Note that signed overflow is undefined in C.  The following routines are
 * careful to use unsigned types where modulo arithmetic is required.
 * Failure to do so _will_ break on newer gcc.
 */

/* Signed saturating arithmetic.  */

/* Perform 16-bit signed saturating addition.  */
static inline uint16_t add16_sat(uint16_t a, uint16_t b)
{
    uint16_t res;

    res = a + b;
    if (((res ^ a) & 0x8000) && !((a ^ b) & 0x8000)) {
        if (a & 0x8000) {
            res = 0x8000;
        } else {
            res = 0x7fff;
        }
    }
    return res;
}

/* Perform 8-bit signed saturating addition.  */
static inline uint8_t add8_sat(uint8_t a, uint8_t b)
{
    uint8_t res;

    res = a + b;
    if (((res ^ a) & 0x80) && !((a ^ b) & 0x80)) {
        if (a & 0x80) {
            res = 0x80;
        } else {
            res = 0x7f;
        }
    }
    return res;
}

/* Perform 16-bit signed saturating subtraction.  */
static inline uint16_t sub16_sat(uint16_t a, uint16_t b)
{
    uint16_t res;

    res = a - b;
    if (((res ^ a) & 0x8000) && ((a ^ b) & 0x8000)) {
        if (a & 0x8000) {
            res = 0x8000;
        } else {
            res = 0x7fff;
        }
    }
    return res;
}

/* Perform 8-bit signed saturating subtraction.  */
static inline uint8_t sub8_sat(uint8_t a, uint8_t b)
{
    uint8_t res;

    res = a - b;
    if (((res ^ a) & 0x80) && ((a ^ b) & 0x80)) {
        if (a & 0x80) {
            res = 0x80;
        } else {
            res = 0x7f;
        }
    }
    return res;
}

#define ADD16(a, b, n) RESULT(add16_sat(a, b), n, 16);
#define SUB16(a, b, n) RESULT(sub16_sat(a, b), n, 16);
#define ADD8(a, b, n)  RESULT(add8_sat(a, b), n, 8);
#define SUB8(a, b, n)  RESULT(sub8_sat(a, b), n, 8);
#define PFX q

#include "op_addsub.c.inc"

/* Unsigned saturating arithmetic.  */
static inline uint16_t add16_usat(uint16_t a, uint16_t b)
{
    uint16_t res;
    res = a + b;
    if (res < a) {
        res = 0xffff;
    }
    return res;
}

static inline uint16_t sub16_usat(uint16_t a, uint16_t b)
{
    if (a > b) {
        return a - b;
    } else {
        return 0;
    }
}

static inline uint8_t add8_usat(uint8_t a, uint8_t b)
{
    uint8_t res;
    res = a + b;
    if (res < a) {
        res = 0xff;
    }
    return res;
}

static inline uint8_t sub8_usat(uint8_t a, uint8_t b)
{
    if (a > b) {
        return a - b;
    } else {
        return 0;
    }
}

#define ADD16(a, b, n) RESULT(add16_usat(a, b), n, 16);
#define SUB16(a, b, n) RESULT(sub16_usat(a, b), n, 16);
#define ADD8(a, b, n)  RESULT(add8_usat(a, b), n, 8);
#define SUB8(a, b, n)  RESULT(sub8_usat(a, b), n, 8);
#define PFX uq

#include "op_addsub.c.inc"

/* Signed modulo arithmetic.  */
#define SARITH16(a, b, n, op) do { \
    int32_t sum; \
    sum = (int32_t)(int16_t)(a) op (int32_t)(int16_t)(b); \
    RESULT(sum, n, 16); \
    if (sum >= 0) \
        ge |= 3 << (n * 2); \
    } while (0)

#define SARITH8(a, b, n, op) do { \
    int32_t sum; \
    sum = (int32_t)(int8_t)(a) op (int32_t)(int8_t)(b); \
    RESULT(sum, n, 8); \
    if (sum >= 0) \
        ge |= 1 << n; \
    } while (0)


#define ADD16(a, b, n) SARITH16(a, b, n, +)
#define SUB16(a, b, n) SARITH16(a, b, n, -)
#define ADD8(a, b, n)  SARITH8(a, b, n, +)
#define SUB8(a, b, n)  SARITH8(a, b, n, -)
#define PFX s
#define ARITH_GE

#include "op_addsub.c.inc"

/* Unsigned modulo arithmetic.  */
#define ADD16(a, b, n) do { \
    uint32_t sum; \
    sum = (uint32_t)(uint16_t)(a) + (uint32_t)(uint16_t)(b); \
    RESULT(sum, n, 16); \
    if ((sum >> 16) == 1) \
        ge |= 3 << (n * 2); \
    } while (0)

#define ADD8(a, b, n) do { \
    uint32_t sum; \
    sum = (uint32_t)(uint8_t)(a) + (uint32_t)(uint8_t)(b); \
    RESULT(sum, n, 8); \
    if ((sum >> 8) == 1) \
        ge |= 1 << n; \
    } while (0)

#define SUB16(a, b, n) do { \
    uint32_t sum; \
    sum = (uint32_t)(uint16_t)(a) - (uint32_t)(uint16_t)(b); \
    RESULT(sum, n, 16); \
    if ((sum >> 16) == 0) \
        ge |= 3 << (n * 2); \
    } while (0)

#define SUB8(a, b, n) do { \
    uint32_t sum; \
    sum = (uint32_t)(uint8_t)(a) - (uint32_t)(uint8_t)(b); \
    RESULT(sum, n, 8); \
    if ((sum >> 8) == 0) \
        ge |= 1 << n; \
    } while (0)

#define PFX u
#define ARITH_GE

#include "op_addsub.c.inc"

/* Halved signed arithmetic.  */
#define ADD16(a, b, n) \
  RESULT(((int32_t)(int16_t)(a) + (int32_t)(int16_t)(b)) >> 1, n, 16)
#define SUB16(a, b, n) \
  RESULT(((int32_t)(int16_t)(a) - (int32_t)(int16_t)(b)) >> 1, n, 16)
#define ADD8(a, b, n) \
  RESULT(((int32_t)(int8_t)(a) + (int32_t)(int8_t)(b)) >> 1, n, 8)
#define SUB8(a, b, n) \
  RESULT(((int32_t)(int8_t)(a) - (int32_t)(int8_t)(b)) >> 1, n, 8)
#define PFX sh

#include "op_addsub.c.inc"

/* Halved unsigned arithmetic.  */
#define ADD16(a, b, n) \
  RESULT(((uint32_t)(uint16_t)(a) + (uint32_t)(uint16_t)(b)) >> 1, n, 16)
#define SUB16(a, b, n) \
  RESULT(((uint32_t)(uint16_t)(a) - (uint32_t)(uint16_t)(b)) >> 1, n, 16)
#define ADD8(a, b, n) \
  RESULT(((uint32_t)(uint8_t)(a) + (uint32_t)(uint8_t)(b)) >> 1, n, 8)
#define SUB8(a, b, n) \
  RESULT(((uint32_t)(uint8_t)(a) - (uint32_t)(uint8_t)(b)) >> 1, n, 8)
#define PFX uh

#include "op_addsub.c.inc"

static inline uint8_t do_usad(uint8_t a, uint8_t b)
{
    if (a > b) {
        return a - b;
    } else {
        return b - a;
    }
}

/* Unsigned sum of absolute byte differences.  */
uint32_t HELPER(usad8)(uint32_t a, uint32_t b)
{
    uint32_t sum;
    sum = do_usad(a, b);
    sum += do_usad(a >> 8, b >> 8);
    sum += do_usad(a >> 16, b >> 16);
    sum += do_usad(a >> 24, b >> 24);
    return sum;
}

/* For ARMv6 SEL instruction.  */
uint32_t HELPER(sel_flags)(uint32_t flags, uint32_t a, uint32_t b)
{
    uint32_t mask;

    mask = 0;
    if (flags & 1) {
        mask |= 0xff;
    }
    if (flags & 2) {
        mask |= 0xff00;
    }
    if (flags & 4) {
        mask |= 0xff0000;
    }
    if (flags & 8) {
        mask |= 0xff000000;
    }
    return (a & mask) | (b & ~mask);
}

/*
 * CRC helpers.
 * The upper bytes of val (above the number specified by 'bytes') must have
 * been zeroed out by the caller.
 */
uint32_t HELPER(crc32)(uint32_t acc, uint32_t val, uint32_t bytes)
{
    uint8_t buf[4];

    stl_le_p(buf, val);

    /* zlib crc32 converts the accumulator and output to one's complement.  */
    return crc32(acc ^ 0xffffffff, buf, bytes) ^ 0xffffffff;
}

uint32_t HELPER(crc32c)(uint32_t acc, uint32_t val, uint32_t bytes)
{
    uint8_t buf[4];

    stl_le_p(buf, val);

    /* Linux crc32c converts the output to one's complement.  */
    return crc32c(acc, buf, bytes) ^ 0xffffffff;
}