/*
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * FFmpeg is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#ifndef AVUTIL_INTREADWRITE_H
#define AVUTIL_INTREADWRITE_H

#include <stdint.h>

#ifdef __GNUC__

struct unaligned_64 { uint64_t l; } __attribute__((packed));
struct unaligned_32 { uint32_t l; } __attribute__((packed));
struct unaligned_16 { uint16_t l; } __attribute__((packed));

#define LD16(a) (((const struct unaligned_16 *) (a))->l)
#define LD32(a) (((const struct unaligned_32 *) (a))->l)
#define LD64(a) (((const struct unaligned_64 *) (a))->l)

#define ST16(a, b) (((struct unaligned_16 *) (a))->l) = (b)
#define ST32(a, b) (((struct unaligned_32 *) (a))->l) = (b)

#else /* __GNUC__ */

#define LD16(a) (*((uint16_t*)(a)))
#define LD32(a) (*((uint32_t*)(a)))
#define LD64(a) (*((uint64_t*)(a)))

#define ST16(a, b) *((uint16_t*)(a)) = (b)
#define ST32(a, b) *((uint32_t*)(a)) = (b)

#endif /* !__GNUC__ */

#ifndef AV_RB16
#   define AV_RB16(x)                           \
    ((((const uint8_t*)(x))[0] << 8) |          \
      ((const uint8_t*)(x))[1])
#endif
#ifndef AV_WB16
#   define AV_WB16(p, val) do {                 \
        uint16_t d = (val);                     \
        ((uint8_t*)(p))[1] = (d);               \
        ((uint8_t*)(p))[0] = (d)>>8;            \
    } while(0)
#endif

#ifndef AV_RL16
#   define AV_RL16(x)                           \
    ((((const uint8_t*)(x))[1] << 8) |          \
      ((const uint8_t*)(x))[0])
#endif
#ifndef AV_WL16
#   define AV_WL16(p, val) do {                 \
        uint16_t d = (val);                     \
        ((uint8_t*)(p))[0] = (d);               \
        ((uint8_t*)(p))[1] = (d)>>8;            \
    } while(0)
#endif

#ifndef AV_RB32
#   define AV_RB32(x)                                \
    (((uint32_t)((const uint8_t*)(x))[0] << 24) |    \
               (((const uint8_t*)(x))[1] << 16) |    \
               (((const uint8_t*)(x))[2] <<  8) |    \
                ((const uint8_t*)(x))[3])
#endif
#ifndef AV_WB32
#   define AV_WB32(p, val) do {                 \
        uint32_t d = (val);                     \
        ((uint8_t*)(p))[3] = (d);               \
        ((uint8_t*)(p))[2] = (d)>>8;            \
        ((uint8_t*)(p))[1] = (d)>>16;           \
        ((uint8_t*)(p))[0] = (d)>>24;           \
    } while(0)
#endif

#ifndef AV_RL32
#   define AV_RL32(x)                                \
    (((uint32_t)((const uint8_t*)(x))[3] << 24) |    \
               (((const uint8_t*)(x))[2] << 16) |    \
               (((const uint8_t*)(x))[1] <<  8) |    \
                ((const uint8_t*)(x))[0])
#endif
#ifndef AV_WL32
#   define AV_WL32(p, val) do {                 \
        uint32_t d = (val);                     \
        ((uint8_t*)(p))[0] = (d);               \
        ((uint8_t*)(p))[1] = (d)>>8;            \
        ((uint8_t*)(p))[2] = (d)>>16;           \
        ((uint8_t*)(p))[3] = (d)>>24;           \
    } while(0)
#endif

#ifndef AV_RB64
#   define AV_RB64(x)                                   \
    (((uint64_t)((const uint8_t*)(x))[0] << 56) |       \
     ((uint64_t)((const uint8_t*)(x))[1] << 48) |       \
     ((uint64_t)((const uint8_t*)(x))[2] << 40) |       \
     ((uint64_t)((const uint8_t*)(x))[3] << 32) |       \
     ((uint64_t)((const uint8_t*)(x))[4] << 24) |       \
     ((uint64_t)((const uint8_t*)(x))[5] << 16) |       \
     ((uint64_t)((const uint8_t*)(x))[6] <<  8) |       \
      (uint64_t)((const uint8_t*)(x))[7])
#endif
#ifndef AV_WB64
#   define AV_WB64(p, val) do {                 \
        uint64_t d = (val);                     \
        ((uint8_t*)(p))[7] = (d);               \
        ((uint8_t*)(p))[6] = (d)>>8;            \
        ((uint8_t*)(p))[5] = (d)>>16;           \
        ((uint8_t*)(p))[4] = (d)>>24;           \
        ((uint8_t*)(p))[3] = (d)>>32;           \
        ((uint8_t*)(p))[2] = (d)>>40;           \
        ((uint8_t*)(p))[1] = (d)>>48;           \
        ((uint8_t*)(p))[0] = (d)>>56;           \
    } while(0)
#endif

#ifndef AV_RL64
#   define AV_RL64(x)                                   \
    (((uint64_t)((const uint8_t*)(x))[7] << 56) |       \
     ((uint64_t)((const uint8_t*)(x))[6] << 48) |       \
     ((uint64_t)((const uint8_t*)(x))[5] << 40) |       \
     ((uint64_t)((const uint8_t*)(x))[4] << 32) |       \
     ((uint64_t)((const uint8_t*)(x))[3] << 24) |       \
     ((uint64_t)((const uint8_t*)(x))[2] << 16) |       \
     ((uint64_t)((const uint8_t*)(x))[1] <<  8) |       \
      (uint64_t)((const uint8_t*)(x))[0])
#endif
#ifndef AV_WL64
#   define AV_WL64(p, val) do {                 \
        uint64_t d = (val);                     \
        ((uint8_t*)(p))[0] = (d);               \
        ((uint8_t*)(p))[1] = (d)>>8;            \
        ((uint8_t*)(p))[2] = (d)>>16;           \
        ((uint8_t*)(p))[3] = (d)>>24;           \
        ((uint8_t*)(p))[4] = (d)>>32;           \
        ((uint8_t*)(p))[5] = (d)>>40;           \
        ((uint8_t*)(p))[6] = (d)>>48;           \
        ((uint8_t*)(p))[7] = (d)>>56;           \
    } while(0)
#endif

#if AV_HAVE_BIGENDIAN
#   define AV_RN(s, p)    AV_RB##s(p)
#   define AV_WN(s, p, v) AV_WB##s(p, v)
#else
#   define AV_RN(s, p)    AV_RL##s(p)
#   define AV_WN(s, p, v) AV_WL##s(p, v)
#endif

#ifndef AV_RN16
#   define AV_RN16(p) AV_RN(16, p)
#endif

#ifndef AV_RN32
#   define AV_RN32(p) AV_RN(32, p)
#endif

#ifndef AV_RN64
#   define AV_RN64(p) AV_RN(64, p)
#endif

#ifndef AV_WN16
#   define AV_WN16(p, v) AV_WN(16, p, v)
#endif

#ifndef AV_WN32
#   define AV_WN32(p, v) AV_WN(32, p, v)
#endif

#ifndef AV_WN64
#   define AV_WN64(p, v) AV_WN(64, p, v)
#endif

#if AV_HAVE_BIGENDIAN
#   define AV_RB(s, p)    AV_RN##s(p)
#   define AV_WB(s, p, v) AV_WN##s(p, v)
#   define AV_RL(s, p)    av_bswap##s(AV_RN##s(p))
#   define AV_WL(s, p, v) AV_WN##s(p, av_bswap##s(v))
#else
#   define AV_RB(s, p)    av_bswap##s(AV_RN##s(p))
#   define AV_WB(s, p, v) AV_WN##s(p, av_bswap##s(v))
#   define AV_RL(s, p)    AV_RN##s(p)
#   define AV_WL(s, p, v) AV_WN##s(p, v)
#endif

#define AV_RB8(x)     (((const uint8_t*)(x))[0])
#define AV_WB8(p, d)  do { ((uint8_t*)(p))[0] = (d); } while(0)

#define AV_RL8(x)     AV_RB8(x)
#define AV_WL8(p, d)  AV_WB8(p, d)

#ifndef AV_RB16
#   define AV_RB16(p)    AV_RB(16, p)
#endif
#ifndef AV_WB16
#   define AV_WB16(p, v) AV_WB(16, p, v)
#endif

#ifndef AV_RL16
#   define AV_RL16(p)    AV_RL(16, p)
#endif
#ifndef AV_WL16
#   define AV_WL16(p, v) AV_WL(16, p, v)
#endif

#ifndef AV_RB32
#   define AV_RB32(p)    AV_RB(32, p)
#endif
#ifndef AV_WB32
#   define AV_WB32(p, v) AV_WB(32, p, v)
#endif

#ifndef AV_RL32
#   define AV_RL32(p)    AV_RL(32, p)
#endif
#ifndef AV_WL32
#   define AV_WL32(p, v) AV_WL(32, p, v)
#endif

#ifndef AV_RB64
#   define AV_RB64(p)    AV_RB(64, p)
#endif
#ifndef AV_WB64
#   define AV_WB64(p, v) AV_WB(64, p, v)
#endif

#ifndef AV_RL64
#   define AV_RL64(p)    AV_RL(64, p)
#endif
#ifndef AV_WL64
#   define AV_WL64(p, v) AV_WL(64, p, v)
#endif

#ifndef AV_RB24
#   define AV_RB24(x)                           \
    ((((const uint8_t*)(x))[0] << 16) |         \
     (((const uint8_t*)(x))[1] <<  8) |         \
      ((const uint8_t*)(x))[2])
#endif
#ifndef AV_WB24
#   define AV_WB24(p, d) do {                   \
        ((uint8_t*)(p))[2] = (d);               \
        ((uint8_t*)(p))[1] = (d)>>8;            \
        ((uint8_t*)(p))[0] = (d)>>16;           \
    } while(0)
#endif

#ifndef AV_RL24
#   define AV_RL24(x)                           \
    ((((const uint8_t*)(x))[2] << 16) |         \
     (((const uint8_t*)(x))[1] <<  8) |         \
      ((const uint8_t*)(x))[0])
#endif
#ifndef AV_WL24
#   define AV_WL24(p, d) do {                   \
        ((uint8_t*)(p))[0] = (d);               \
        ((uint8_t*)(p))[1] = (d)>>8;            \
        ((uint8_t*)(p))[2] = (d)>>16;           \
    } while(0)
#endif

#ifndef AV_RB48
#   define AV_RB48(x)                                     \
    (((uint64_t)((const uint8_t*)(x))[0] << 40) |         \
     ((uint64_t)((const uint8_t*)(x))[1] << 32) |         \
     ((uint64_t)((const uint8_t*)(x))[2] << 24) |         \
     ((uint64_t)((const uint8_t*)(x))[3] << 16) |         \
     ((uint64_t)((const uint8_t*)(x))[4] <<  8) |         \
      (uint64_t)((const uint8_t*)(x))[5])
#endif
#ifndef AV_WB48
#   define AV_WB48(p, darg) do {                \
        uint64_t d = (darg);                    \
        ((uint8_t*)(p))[5] = (d);               \
        ((uint8_t*)(p))[4] = (d)>>8;            \
        ((uint8_t*)(p))[3] = (d)>>16;           \
        ((uint8_t*)(p))[2] = (d)>>24;           \
        ((uint8_t*)(p))[1] = (d)>>32;           \
        ((uint8_t*)(p))[0] = (d)>>40;           \
    } while(0)
#endif

#ifndef AV_RL48
#   define AV_RL48(x)                                     \
    (((uint64_t)((const uint8_t*)(x))[5] << 40) |         \
     ((uint64_t)((const uint8_t*)(x))[4] << 32) |         \
     ((uint64_t)((const uint8_t*)(x))[3] << 24) |         \
     ((uint64_t)((const uint8_t*)(x))[2] << 16) |         \
     ((uint64_t)((const uint8_t*)(x))[1] <<  8) |         \
      (uint64_t)((const uint8_t*)(x))[0])
#endif
#ifndef AV_WL48
#   define AV_WL48(p, darg) do {                \
        uint64_t d = (darg);                    \
        ((uint8_t*)(p))[0] = (d);               \
        ((uint8_t*)(p))[1] = (d)>>8;            \
        ((uint8_t*)(p))[2] = (d)>>16;           \
        ((uint8_t*)(p))[3] = (d)>>24;           \
        ((uint8_t*)(p))[4] = (d)>>32;           \
        ((uint8_t*)(p))[5] = (d)>>40;           \
    } while(0)
#endif

/*
 * The AV_[RW]NA macros access naturally aligned data
 * in a type-safe way.
 */

#define AV_RNA(s, p)    (((const av_alias##s*)(p))->u##s)
#define AV_WNA(s, p, v) (((av_alias##s*)(p))->u##s = (v))

#ifndef AV_RN16A
#   define AV_RN16A(p) AV_RNA(16, p)
#endif

#ifndef AV_RN32A
#   define AV_RN32A(p) AV_RNA(32, p)
#endif

#ifndef AV_RN64A
#   define AV_RN64A(p) AV_RNA(64, p)
#endif

#ifndef AV_WN16A
#   define AV_WN16A(p, v) AV_WNA(16, p, v)
#endif

#ifndef AV_WN32A
#   define AV_WN32A(p, v) AV_WNA(32, p, v)
#endif

#ifndef AV_WN64A
#   define AV_WN64A(p, v) AV_WNA(64, p, v)
#endif

#if AV_HAVE_BIGENDIAN
#   define AV_RLA(s, p)    av_bswap##s(AV_RN##s##A(p))
#   define AV_WLA(s, p, v) AV_WN##s##A(p, av_bswap##s(v))
#else
#   define AV_RLA(s, p)    AV_RN##s##A(p)
#   define AV_WLA(s, p, v) AV_WN##s##A(p, v)
#endif

#ifndef AV_RL64A
#   define AV_RL64A(p) AV_RLA(64, p)
#endif
#ifndef AV_WL64A
#   define AV_WL64A(p, v) AV_WLA(64, p, v)
#endif

/*
 * The AV_COPYxxU macros are suitable for copying data to/from unaligned
 * memory locations.
 */

#define AV_COPYU(n, d, s) AV_WN##n(d, AV_RN##n(s));

#ifndef AV_COPY16U
#   define AV_COPY16U(d, s) AV_COPYU(16, d, s)
#endif

#ifndef AV_COPY32U
#   define AV_COPY32U(d, s) AV_COPYU(32, d, s)
#endif

#ifndef AV_COPY64U
#   define AV_COPY64U(d, s) AV_COPYU(64, d, s)
#endif

#ifndef AV_COPY128U
#   define AV_COPY128U(d, s)                                    \
    do {                                                        \
        AV_COPY64U(d, s);                                       \
        AV_COPY64U((char *)(d) + 8, (const char *)(s) + 8);     \
    } while(0)
#endif

/* Parameters for AV_COPY*, AV_SWAP*, AV_ZERO* must be
 * naturally aligned. They may be implemented using MMX,
 * so emms_c() must be called before using any float code
 * afterwards.
 */

#define AV_COPY(n, d, s) \
    (((av_alias##n*)(d))->u##n = ((const av_alias##n*)(s))->u##n)

#ifndef AV_COPY16
#   define AV_COPY16(d, s) AV_COPY(16, d, s)
#endif

#ifndef AV_COPY32
#   define AV_COPY32(d, s) AV_COPY(32, d, s)
#endif

#ifndef AV_COPY64
#   define AV_COPY64(d, s) AV_COPY(64, d, s)
#endif

#ifndef AV_COPY128
#   define AV_COPY128(d, s)                    \
    do {                                       \
        AV_COPY64(d, s);                       \
        AV_COPY64((char*)(d)+8, (char*)(s)+8); \
    } while(0)
#endif

#define AV_SWAP(n, a, b) FFSWAP(av_alias##n, *(av_alias##n*)(a), *(av_alias##n*)(b))

#ifndef AV_SWAP64
#   define AV_SWAP64(a, b) AV_SWAP(64, a, b)
#endif

#define AV_ZERO(n, d) (((av_alias##n*)(d))->u##n = 0)

#ifndef AV_ZERO16
#   define AV_ZERO16(d) AV_ZERO(16, d)
#endif

#ifndef AV_ZERO32
#   define AV_ZERO32(d) AV_ZERO(32, d)
#endif

#ifndef AV_ZERO64
#   define AV_ZERO64(d) AV_ZERO(64, d)
#endif

#ifndef AV_ZERO128
#   define AV_ZERO128(d)         \
    do {                         \
        AV_ZERO64(d);            \
        AV_ZERO64((char*)(d)+8); \
    } while(0)
#endif

#endif /* AVUTIL_INTREADWRITE_H */