Commit 7a64649d authored by Vitor Sessak's avatar Vitor Sessak
Browse files

Create a function for float dot product instead of duplicating all over the file

Originally committed as revision 14079 to svn://svn.ffmpeg.org/ffmpeg/trunk
parent a4ce7248
......@@ -37,6 +37,16 @@ typedef struct {
float lhist[10];
} Real288_internal;
static inline float scalar_product_float(float * v1, float * v2, int size)
{
float res = 0.;
while (size--)
res += *v1++ * *v2++;
return res;
}
/* Decode and produce output */
static void decode(Real288_internal *glob, float gain, int cb_coef)
{
......@@ -48,19 +58,11 @@ static void decode(Real288_internal *glob, float gain, int cb_coef)
glob->sb[x+5] = glob->sb[x];
for (x=4; x >= 0; x--) {
float *p1 = glob->sb+x;
float *p2 = glob->pr1;
sum = 0;
for (y=0; y < 36; y++)
sum -= (*(++p1))*(*(p2++));
glob->sb[x] = sum;
glob->sb[x] = -scalar_product_float(glob->sb + x + 1, glob->pr1, 36);
}
/* convert log and do rms */
sum = 32;
for (x=0; x < 10; x++)
sum -= glob->pr2[x] * glob->lhist[x];
sum = 32. - scalar_product_float(glob->pr2, glob->lhist, 10);
if (sum < 0)
sum = 0;
......@@ -148,17 +150,9 @@ static int pred(float *in, float *tgt, int n)
/* product sum (lsf) */
static void prodsum(float *tgt, float *src, int len, int n)
{
unsigned int x;
double sum;
for (; n >= 0; n--)
tgt[n] = scalar_product_float(src, src - n, len);
while (n >= 0) {
float *p2 = src;
float *p1 = p2 - n;
sum = 0;
for (x=0; x < len; x++)
sum += (*p1++) * (*p2++);
tgt[n--] = sum;
}
}
static void co(int n, int i, int j, float *in, float *out, float *st1,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment