har0ke/FFmpeg
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

optimize decode_subframe_lpc()

Browse Source 
50%/67%/43% faster on core2/k8/p4, making flac decoding overall 24%/25%/11% faster

Originally committed as revision 10586 to svn://svn.ffmpeg.org/ffmpeg/trunk
This commit is contained in:
Loren Merritt 2007-09-25 23:31:13 +00:00
parent f74471e043
commit bd49d4fd49
1 changed files with 18 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
libavcodec/flac.c
View File

@ -359,10 +359,25 @@ static int decode_subframe_lpc(FLACContext *s, int channel, int pred_order)
s->decoded[channel][i] += sum >> qlevel; s->decoded[channel][i] += sum >> qlevel;
} }
} else { } else {
int sum; for (i = pred_order; i < s->blocksize-1; i += 2)
for (i = pred_order; i < s->blocksize; i++)
{ {
sum = 0; int c = coeffs[pred_order-1];
int s0 = c * s->decoded[channel][i-pred_order];
int s1 = 0;
for (j = pred_order-1; j > 0; j--)
{
int d = s->decoded[channel][i-j];
s1 += c*d;
c = coeffs[j-1];
s0 += c*d;
}
s0 = s->decoded[channel][i] += s0 >> qlevel;
s1 += c * s0;
s->decoded[channel][i+1] += s1 >> qlevel;
}
if (i < s->blocksize)
{
int sum = 0;
for (j = 0; j < pred_order; j++) for (j = 0; j < pred_order; j++)
sum += coeffs[j] * s->decoded[channel][i-j-1]; sum += coeffs[j] * s->decoded[channel][i-j-1];
s->decoded[channel][i] += sum >> qlevel; s->decoded[channel][i] += sum >> qlevel;