generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

SBR DSP: unroll sum_square

Browse Source 
The length is even, so some unrolling can be performed. Timings are for x86:
- 32bits: 102c -> 82c
- 64bits:  82c -> 69c

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
This commit is contained in:
Christophe GISQUET 2012-02-22 17:48:59 +01:00 committed by Ronald S. Bultje
parent 294c05ce8a
commit dabf8dd34a
1 changed files with 9 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
libavcodec/sbrdsp.c
View File

@ -35,13 +35,18 @@ static void sbr_sum64x5_c(float *z)
static float sbr_sum_square_c(float (*x)[2], int n) static float sbr_sum_square_c(float (*x)[2], int n)
{ {
float sum = 0.0f; float sum0 = 0.0f, sum1 = 0.0f;
int i; int i;
for (i = 0; i < n; i++) for (i = 0; i < n; i += 2)
sum += x[i][0] * x[i][0] + x[i][1] * x[i][1]; {
sum0 += x[i + 0][0] * x[i + 0][0];
sum1 += x[i + 0][1] * x[i + 0][1];
sum0 += x[i + 1][0] * x[i + 1][0];
sum1 += x[i + 1][1] * x[i + 1][1];
}
return sum; return sum0 + sum1;
} }
static void sbr_neg_odd_64_c(float *x) static void sbr_neg_odd_64_c(float *x)