imports.c

mesa-6.5-minigui源码

#endif
}


/**
 inv_sqrt - A single precision 1/sqrt routine for IEEE format floats.
 written by Josh Vanderhoof, based on newsgroup posts by James Van Buskirk
 and Vesa Karvonen.
*/
float
_mesa_inv_sqrtf(float n)
{
#if defined(USE_IEEE) && !defined(DEBUG)
        float r0, x0, y0;
        float r1, x1, y1;
        float r2, x2, y2;
#if 0 /* not used, see below -BP */
        float r3, x3, y3;
#endif
        union { float f; unsigned int i; } u;
        unsigned int magic;

        /*
         Exponent part of the magic number -

         We want to:
         1. subtract the bias from the exponent,
         2. negate it
         3. divide by two (rounding towards -inf)
         4. add the bias back

         Which is the same as subtracting the exponent from 381 and dividing
         by 2.

         floor(-(x - 127) / 2) + 127 = floor((381 - x) / 2)
        */

        magic = 381 << 23;

        /*
         Significand part of magic number -

         With the current magic number, "(magic - u.i) >> 1" will give you:

         for 1 <= u.f <= 2: 1.25 - u.f / 4
         for 2 <= u.f <= 4: 1.00 - u.f / 8

         This isn't a bad approximation of 1/sqrt.  The maximum difference from
         1/sqrt will be around .06.  After three Newton-Raphson iterations, the
         maximum difference is less than 4.5e-8.  (Which is actually close
         enough to make the following bias academic...)

         To get a better approximation you can add a bias to the magic
         number.  For example, if you subtract 1/2 of the maximum difference in
         the first approximation (.03), you will get the following function:

         for 1 <= u.f <= 2:    1.22 - u.f / 4
         for 2 <= u.f <= 3.76: 0.97 - u.f / 8
         for 3.76 <= u.f <= 4: 0.72 - u.f / 16
         (The 3.76 to 4 range is where the result is < .5.)

         This is the closest possible initial approximation, but with a maximum
         error of 8e-11 after three NR iterations, it is still not perfect.  If
         you subtract 0.0332281 instead of .03, the maximum error will be
         2.5e-11 after three NR iterations, which should be about as close as
         is possible.

         for 1 <= u.f <= 2:    1.2167719 - u.f / 4
         for 2 <= u.f <= 3.73: 0.9667719 - u.f / 8
         for 3.73 <= u.f <= 4: 0.7167719 - u.f / 16

        */

        magic -= (int)(0.0332281 * (1 << 25));

        u.f = n;
        u.i = (magic - u.i) >> 1;

        /*
         Instead of Newton-Raphson, we use Goldschmidt's algorithm, which
         allows more parallelism.  From what I understand, the parallelism
         comes at the cost of less precision, because it lets error
         accumulate across iterations.
        */
        x0 = 1.0f;
        y0 = 0.5f * n;
        r0 = u.f;

        x1 = x0 * r0;
        y1 = y0 * r0 * r0;
        r1 = 1.5f - y1;

        x2 = x1 * r1;
        y2 = y1 * r1 * r1;
        r2 = 1.5f - y2;

#if 1
        return x2 * r2;  /* we can stop here, and be conformant -BP */
#else
        x3 = x2 * r2;
        y3 = y2 * r2 * r2;
        r3 = 1.5f - y3;

        return x3 * r3;
#endif
#elif defined(XFree86LOADER) && defined(IN_MODULE)
        return 1.0F / xf86sqrt(n);
#else
        return (float) (1.0 / sqrt(n));
#endif
}


/**
 * Wrapper around either pow() or xf86pow().
 */
double
_mesa_pow(double x, double y)
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86pow(x, y);
#else
   return pow(x, y);
#endif
}


/**
 * Find the first bit set in a word.
 */
int
_mesa_ffs(int i)
{
#if (defined(_WIN32) && !defined(__MINGW32__) ) || defined(__IBMC__) || defined(__IBMCPP__)
   register int bit = 0;
   if (i != 0) {
      if ((i & 0xffff) == 0) {
         bit += 16;
         i >>= 16;
      }
      if ((i & 0xff) == 0) {
         bit += 8;
         i >>= 8;
      }
      if ((i & 0xf) == 0) {
         bit += 4;
         i >>= 4;
      }
      while ((i & 1) == 0) {
         bit++;
         i >>= 1;
      }
   }
   return bit;
#elif defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86ffs(i);
#else
   return ffs(i);
#endif
}


/**
 * Return number of bits set in given GLuint.
 */
unsigned int
_mesa_bitcount(unsigned int n)
{
   unsigned int bits;
   for (bits = 0; n > 0; n = n >> 1) {
      bits += (n & 1);
   }
   return bits;
}


/**
 * Convert a 4-byte float to a 2-byte half float.
 * Based on code from:
 * http://www.opengl.org/discussion_boards/ubb/Forum3/HTML/008786.html
 */
GLhalfARB
_mesa_float_to_half(float val)
{
   const int flt = *((int *) (void *) &val);
   const int flt_m = flt & 0x7fffff;
   const int flt_e = (flt >> 23) & 0xff;
   const int flt_s = (flt >> 31) & 0x1;
   int s, e, m = 0;
   GLhalfARB result;
   
   /* sign bit */
   s = flt_s;

   /* handle special cases */
   if ((flt_e == 0) && (flt_m == 0)) {
      /* zero */
      /* m = 0; - already set */
      e = 0;
   }
   else if ((flt_e == 0) && (flt_m != 0)) {
      /* denorm -- denorm float maps to 0 half */
      /* m = 0; - already set */
      e = 0;
   }
   else if ((flt_e == 0xff) && (flt_m == 0)) {
      /* infinity */
      /* m = 0; - already set */
      e = 31;
   }
   else if ((flt_e == 0xff) && (flt_m != 0)) {
      /* NaN */
      m = 1;
      e = 31;
   }
   else {
      /* regular number */
      const int new_exp = flt_e - 127;
      if (new_exp < -24) {
         /* this maps to 0 */
         /* m = 0; - already set */
         e = 0;
      }
      else if (new_exp < -14) {
         /* this maps to a denorm */
         unsigned int exp_val = (unsigned int) (-14 - new_exp); /* 2^-exp_val*/
         e = 0;
         switch (exp_val) {
            case 0:
               _mesa_warning(NULL,
                   "float_to_half: logical error in denorm creation!\n");
               /* m = 0; - already set */
               break;
            case 1: m = 512 + (flt_m >> 14); break;
            case 2: m = 256 + (flt_m >> 15); break;
            case 3: m = 128 + (flt_m >> 16); break;
            case 4: m = 64 + (flt_m >> 17); break;
            case 5: m = 32 + (flt_m >> 18); break;
            case 6: m = 16 + (flt_m >> 19); break;
            case 7: m = 8 + (flt_m >> 20); break;
            case 8: m = 4 + (flt_m >> 21); break;
            case 9: m = 2 + (flt_m >> 22); break;
            case 10: m = 1; break;
         }
      }
      else if (new_exp > 15) {
         /* map this value to infinity */
         /* m = 0; - already set */
         e = 31;
      }
      else {
         /* regular */
         e = new_exp + 15;
         m = flt_m >> 13;
      }
   }

   result = (s << 15) | (e << 10) | m;
   return result;
}


/**
 * Convert a 2-byte half float to a 4-byte float.
 * Based on code from:
 * http://www.opengl.org/discussion_boards/ubb/Forum3/HTML/008786.html
 */
float
_mesa_half_to_float(GLhalfARB val)
{
   /* XXX could also use a 64K-entry lookup table */
   const int m = val & 0x3ff;
   const int e = (val >> 10) & 0x1f;
   const int s = (val >> 15) & 0x1;
   int flt_m, flt_e, flt_s, flt;
   float result;

   /* sign bit */
   flt_s = s;

   /* handle special cases */
   if ((e == 0) && (m == 0)) {
      /* zero */
      flt_m = 0;
      flt_e = 0;
   }
   else if ((e == 0) && (m != 0)) {
      /* denorm -- denorm half will fit in non-denorm single */
      const float half_denorm = 1.0f / 16384.0f; /* 2^-14 */
      float mantissa = ((float) (m)) / 1024.0f;
      float sign = s ? -1.0f : 1.0f;
      return sign * mantissa * half_denorm;
   }
   else if ((e == 31) && (m == 0)) {
      /* infinity */
      flt_e = 0xff;
      flt_m = 0;
   }
   else if ((e == 31) && (m != 0)) {
      /* NaN */
      flt_e = 0xff;
      flt_m = 1;
   }
   else {
      /* regular */
      flt_e = e + 112;
      flt_m = m << 13;
   }

   flt = (flt_s << 31) | (flt_e << 23) | flt_m;
   result = *((float *) (void *) &flt);
   return result;
}

/*@}*/


/**********************************************************************/
/** \name Sort & Search */
/*@{*/

/**
 * Wrapper for bsearch().
 */
void *
_mesa_bsearch( const void *key, const void *base, size_t nmemb, size_t size, 
               int (*compar)(const void *, const void *) )
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86bsearch(key, base, nmemb, size, compar);
#else
   return bsearch(key, base, nmemb, size, compar);
#endif
}

/*@}*/


/**********************************************************************/
/** \name Environment vars */
/*@{*/

/**
 * Wrapper for getenv().
 */
char *
_mesa_getenv( const char *var )
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86getenv(var);
#elif defined(_XBOX)
   return NULL;
#else
   return getenv(var);
#endif
}

/*@}*/


/**********************************************************************/
/** \name String */
/*@{*/

/** Wrapper around either strstr() or xf86strstr() */
char *
_mesa_strstr( const char *haystack, const char *needle )
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86strstr(haystack, needle);
#else
   return strstr(haystack, needle);
#endif
}

/** Wrapper around either strncat() or xf86strncat() */
char *
_mesa_strncat( char *dest, const char *src, size_t n )
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86strncat(dest, src, n);
#else
   return strncat(dest, src, n);
#endif
}

/** Wrapper around either strcpy() or xf86strcpy() */
char *
_mesa_strcpy( char *dest, const char *src )
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86strcpy(dest, src);
#else
   return strcpy(dest, src);
#endif
}

/** Wrapper around either strncpy() or xf86strncpy() */
char *
_mesa_strncpy( char *dest, const char *src, size_t n )
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86strncpy(dest, src, n);
#else
   return strncpy(dest, src, n);
#endif
}

/** Wrapper around either strlen() or xf86strlen() */
size_t
_mesa_strlen( const char *s )
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86strlen(s);
#else
   return strlen(s);
#endif
}

/** Wrapper around either strcmp() or xf86strcmp() */
int
_mesa_strcmp( const char *s1, const char *s2 )
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86strcmp(s1, s2);
#else
   return strcmp(s1, s2);
#endif
}

/** Wrapper around either strncmp() or xf86strncmp() */
int
_mesa_strncmp( const char *s1, const char *s2, size_t n )
{
#if defined(XFree86LOADER) && defined(IN_MODULE)
   return xf86strncmp(s1, s2, n);
#else
   return strncmp(s1, s2, n);
#endif
}

/** Implemented using _mesa_malloc() and _mesa_strcpy */
char *
_mesa_strdup( const char *s )