rpi-vk-driver/external/include/broadcom/v3d_packet_helpers.h

/*
 * Copyright (C) 2016 Intel Corporation
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice (including the next
 * paragraph) shall be included in all copies or substantial portions of the
 * Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
 * IN THE SOFTWARE.
 */

#include <stdio.h>
#include <stdint.h>
#include <stdbool.h>
#include <assert.h>
#include <math.h>

#ifdef HAVE_VALGRIND
#include <valgrind.h>
#include <memcheck.h>
#define VG(x) x
#define __gen_validate_value(x) VALGRIND_CHECK_MEM_IS_DEFINED(&(x), sizeof(x))
#else
#define VG(x)
#endif

#ifndef __gen_validate_value
#define __gen_validate_value(x)
#endif
/*
#ifndef __gen_address_type
#error #define __gen_address_type before including this file
#endif

#ifndef __gen_user_data
#error #define __gen_combine_address before including this file
#endif
*/
union __gen_value {
   float f;
   uint32_t dw;
};

static inline uint64_t
__gen_mbo(uint32_t start, uint32_t end)
{
   return (~0ull >> (64 - (end - start + 1))) << start;
}

static inline uint64_t
__gen_uint(uint64_t v, uint32_t start, uint32_t end)
{
   __gen_validate_value(v);

#if DEBUG
   const int width = end - start + 1;
   if (width < 64) {
      const uint64_t max = (1ull << width) - 1;
      assert(v <= max);
   }
#endif

   return v << start;
}

static inline uint64_t
__gen_sint(int64_t v, uint32_t start, uint32_t end)
{
   const int width = end - start + 1;

   __gen_validate_value(v);

#if DEBUG
   if (width < 64) {
      const int64_t max = (1ll << (width - 1)) - 1;
      const int64_t min = -(1ll << (width - 1));
      assert(min <= v && v <= max);
   }
#endif

   const uint64_t mask = ~0ull >> (64 - width);

   return (v & mask) << start;
}

static inline uint64_t
__gen_offset(uint64_t v, uint32_t start, uint32_t end)
{
   __gen_validate_value(v);
#if DEBUG
   uint64_t mask = (~0ull >> (64 - (end - start + 1))) << start;

   assert((v & ~mask) == 0);
#endif

   return v;
}

static inline uint32_t
__gen_float(float v)
{
   __gen_validate_value(v);
   return ((union __gen_value) { .f = (v) }).dw;
}

static inline uint64_t
__gen_sfixed(float v, uint32_t start, uint32_t end, uint32_t fract_bits)
{
   __gen_validate_value(v);

   const float factor = (1 << fract_bits);

#if DEBUG
   const float max = ((1 << (end - start)) - 1) / factor;
   const float min = -(1 << (end - start)) / factor;
   assert(min <= v && v <= max);
#endif

   const int64_t int_val = llroundf(v * factor);
   const uint64_t mask = ~0ull >> (64 - (end - start + 1));

   return (int_val & mask) << start;
}

static inline uint64_t
__gen_ufixed(float v, uint32_t start, uint32_t end, uint32_t fract_bits)
{
   __gen_validate_value(v);

   const float factor = (1 << fract_bits);

#if DEBUG
   const float max = ((1 << (end - start + 1)) - 1) / factor;
   const float min = 0.0f;
   assert(min <= v && v <= max);
#endif

   const uint64_t uint_val = llroundf(v * factor);

   return uint_val << start;
}

static inline uint64_t
__gen_unpack_uint(const uint8_t *restrict cl, uint32_t start, uint32_t end)
{
   uint64_t val = 0;
   const int width = end - start + 1;
   const uint32_t mask = (width == 32 ? ~0 : (1 << width) - 1 );

   for (int byte = start / 8; byte <= end / 8; byte++) {
      val |= cl[byte] << ((byte - start / 8) * 8);
   }

   return (val >> (start % 8)) & mask;
}

static inline uint64_t
__gen_unpack_sint(const uint8_t *restrict cl, uint32_t start, uint32_t end)
{
   int size = end - start + 1;
   int64_t val = __gen_unpack_uint(cl, start, end);

   /* Get the sign bit extended. */
   return (val << (64 - size)) >> (64 - size);
}

static inline float
__gen_unpack_sfixed(const uint8_t *restrict cl, uint32_t start, uint32_t end,
                    uint32_t fractional_size)
{
        int32_t bits = __gen_unpack_sint(cl, start, end);
        return (float)bits / (1 << fractional_size);
}

static inline float
__gen_unpack_ufixed(const uint8_t *restrict cl, uint32_t start, uint32_t end,
                    uint32_t fractional_size)
{
        int32_t bits = __gen_unpack_uint(cl, start, end);
        return (float)bits / (1 << fractional_size);
}

static inline float
__gen_unpack_float(const uint8_t *restrict cl, uint32_t start, uint32_t end)
{
   assert(start % 8 == 0);
   assert(end - start == 31);

   struct PACKED { float f; } *f = (void *)(cl + (start / 8));

   return f->f;
}