Mercurial > libavcodec.hg
comparison h264_cavlc.c @ 10906:1b5fba731e24 libavcodec
Rearchitecturing the stiched up goose part 1
Run loop filter per row instead of per MB, this also should make it
much easier to switch to per frame filtering and also doing so in a
seperate thread in the future if some volunteer wants to try.
Overall decoding speedup of 1.7% (single thread on pentium dual / cathedral sample)
This change also allows some optimizations to be tried that would not have
been possible before.
author | michael |
---|---|
date | Sun, 17 Jan 2010 20:35:55 +0000 |
parents | 2aafcafbe1f0 |
children | 2d82b73b12ef |
comparison
equal
deleted
inserted
replaced
10905:06d20a468d1e | 10906:1b5fba731e24 |
---|---|
618 } | 618 } |
619 | 619 |
620 // In deblocking, the quantizer is 0 | 620 // In deblocking, the quantizer is 0 |
621 s->current_picture.qscale_table[mb_xy]= 0; | 621 s->current_picture.qscale_table[mb_xy]= 0; |
622 // All coeffs are present | 622 // All coeffs are present |
623 memset(h->non_zero_count[mb_xy], 16, 16); | 623 memset(h->non_zero_count[mb_xy], 16, 32); |
624 | 624 |
625 s->current_picture.mb_type[mb_xy]= mb_type; | 625 s->current_picture.mb_type[mb_xy]= mb_type; |
626 return 0; | 626 return 0; |
627 } | 627 } |
628 | 628 |