Make calculation of mask_edge free of branches, faster of course but probably
[ffmpeg.git] / libavcodec / h264_loopfilter.c
2010-01-21 Michael NiedermayerMake calculation of mask_edge free of branches, faster...
2010-01-20 Alexander StrangeH.264: Declare bS with DECLARE_ALIGNED_8 for uint64_t...
2010-01-20 Michael NiedermayerSimplify/Optimize another of the mbaff loop filter...
2010-01-20 Michael NiedermayerOnly calculate the second chroma qp if it differs from...
2010-01-20 Michael NiedermayerSet bS with 64bits at a time.
2010-01-20 Michael NiedermayerMerge multiple IS_* macro uses where possible.
2010-01-20 Michael NiedermayerSimplify and optimize intra code in h264_loopfilter.c
2010-01-20 Michael NiedermayerSightly simplify initialization of int start.
2010-01-19 Michael NiedermayerReenable ff_h264_filter_mb_fast() for all slices it...
2010-01-18 Michael NiedermayerFix compilation with -O0.
2010-01-18 Michael NiedermayerRather call filter_mb_mbaff_edge*v() more often than...
2010-01-18 Michael NiedermayerUse h->slice_num where possible.
2010-01-18 Michael NiedermayerEnable filter_mb_fast for CAVLC P slices.
2010-01-18 Michael NiedermayerPAFF CABAC P slices seem to work as well, so enable...
2010-01-18 Michael NiedermayerReenable filter_mb_fast for I slices and progressive...
2010-01-18 Michael NiedermayerMove CAVLC 8x8 DCT special case from ff_h264_filter_mb...
2010-01-18 Michael NiedermayerPerform reference remapping at fill_cache() time instea...
2010-01-18 Michael NiedermayerMove the qp check to skip the loop filter up.
2010-01-17 Michael NiedermayerReorganize how values are stored in h->non_zero_count.
2010-01-17 Michael NiedermayerRearchitecturing the stiched up goose part 1
2010-01-16 Michael NiedermayerComment for() ; out
2010-01-16 Michael NiedermayerMark a few functions as noinline, this makes ff_h264_fi...
2010-01-16 Michael NiedermayerApply last 2 optimizations to similar code i forgot.
2010-01-16 Michael NiedermayerAnother microopt, 4 cpu cycles for avoidance of FFABS().
2010-01-16 Michael NiedermayerMinor (2 cpu cycles) optimization ||->|.
2010-01-16 Michael NiedermayerAvoid wasting 4 cpu cycles per MB in redundantly calcul...
2010-01-12 Michael NiedermayerSplit h264 loop filter off h264.c.