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

Factorize 3 multiplications out, code becomes 3 cpu cycles faster.

Browse Source 
(not significant as thats just per MB)

Originally committed as revision 16174 to svn://svn.ffmpeg.org/ffmpeg/trunk
This commit is contained in:
Michael Niedermayer 2008-12-16 21:08:16 +00:00
parent 1dd488e955
commit 6120a343aa
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
libavcodec/h264.c
View File

@ -2371,9 +2371,9 @@ static av_always_inline void hl_decode_mb_internal(H264Context *h, int simple){
void (*idct_add)(uint8_t *dst, DCTELEM *block, int stride); void (*idct_add)(uint8_t *dst, DCTELEM *block, int stride);
void (*idct_dc_add)(uint8_t *dst, DCTELEM *block, int stride); void (*idct_dc_add)(uint8_t *dst, DCTELEM *block, int stride);
dest_y = s->current_picture.data[0] + (mb_y * 16* s->linesize ) + mb_x * 16; dest_y = s->current_picture.data[0] + (mb_x + mb_y * s->linesize ) * 16;
dest_cb = s->current_picture.data[1] + (mb_y * 8 * s->uvlinesize) + mb_x * 8; dest_cb = s->current_picture.data[1] + (mb_x + mb_y * s->uvlinesize) * 8;
dest_cr = s->current_picture.data[2] + (mb_y * 8 * s->uvlinesize) + mb_x * 8; dest_cr = s->current_picture.data[2] + (mb_x + mb_y * s->uvlinesize) * 8;
s->dsp.prefetch(dest_y + (s->mb_x&3)*4*s->linesize + 64, s->linesize, 4); s->dsp.prefetch(dest_y + (s->mb_x&3)*4*s->linesize + 64, s->linesize, 4);
s->dsp.prefetch(dest_cb + (s->mb_x&7)*s->uvlinesize + 64, dest_cr - dest_cb, 2); s->dsp.prefetch(dest_cb + (s->mb_x&7)*s->uvlinesize + 64, dest_cr - dest_cb, 2);