Age | Commit message (Expand) | Author |
2010-01-28 | factorize first filter call out, this makes the code somewhat | Michael Niedermayer |
2010-01-28 | Change wraper functions to always inline, they are faster now that way. | Michael Niedermayer |
2010-01-28 | indent | Michael Niedermayer |
2010-01-28 | Restructure check_mv() | Michael Niedermayer |
2010-01-28 | Restructure if() in check_mv() | Michael Niedermayer |
2010-01-28 | Unroll loops in check_mv() | Michael Niedermayer |
2010-01-28 | Factor mv/ref compare code out. | Michael Niedermayer |
2010-01-28 | Simplify first edge filter condition. | Michael Niedermayer |
2010-01-28 | Cosmetics, mostly indention, 2 or so new fixme comments that i was to lazy | Michael Niedermayer |
2010-01-28 | Make the fast loop filter path work with unavailable left MBs. | Michael Niedermayer |
2010-01-28 | get rid of the start variable. | Michael Niedermayer |
2010-01-28 | Unroll main loop so the edge==0 case is seperate. | Michael Niedermayer |
2010-01-27 | Update comment. | Michael Niedermayer |
2010-01-27 | Use table to speedup access to non_zero_count in MBAFF with differing interla... | Michael Niedermayer |
2010-01-27 | Optimize loop filtering of the left edge in MBAFF. | Michael Niedermayer |
2010-01-26 | remove unneeded check | Michael Niedermayer |
2010-01-26 | Use left_mb_xy from fill_caches instead of recalculating it. | Michael Niedermayer |
2010-01-26 | Simplify loop filter a little by using top/left_type. | Michael Niedermayer |
2010-01-24 | Remove all uses of slice_type* from the loop filter, also remove its | Michael Niedermayer |
2010-01-23 | Move +52 from the loop filter to the alpha/beta offsets in the context. | Michael Niedermayer |
2010-01-23 | Set edges based on cbp and mv partitioning, not just skiped MBs. | Michael Niedermayer |
2010-01-23 | Optimize filter_mb_mbaff_edge*() | Michael Niedermayer |
2010-01-23 | Optmize 8x8dct check used to skip some borders in the loop filter. | Michael Niedermayer |
2010-01-22 | Move array specifiers outside DECLARE_ALIGNED() invocations | Måns Rullgård |
2010-01-22 | Gcc idiocy fixes related to filter_mb_edge*. | Michael Niedermayer |
2010-01-21 | Make calculation of mask_edge free of branches, faster of course but probably | Michael Niedermayer |
2010-01-20 | H.264: Declare bS with DECLARE_ALIGNED_8 for uint64_t casts. | Alexander Strange |
2010-01-20 | Simplify/Optimize another of the mbaff loop filter cases. | Michael Niedermayer |
2010-01-20 | Only calculate the second chroma qp if it differs from the firstin the main | Michael Niedermayer |
2010-01-20 | Set bS with 64bits at a time. | Michael Niedermayer |
2010-01-20 | Merge multiple IS_* macro uses where possible. | Michael Niedermayer |
2010-01-20 | Simplify and optimize intra code in h264_loopfilter.c | Michael Niedermayer |
2010-01-20 | Sightly simplify initialization of int start. | Michael Niedermayer |
2010-01-19 | Reenable ff_h264_filter_mb_fast() for all slices it supported before. | Michael Niedermayer |
2010-01-19 | Fix compilation with -O0. | Michael Niedermayer |
2010-01-19 | Rather call filter_mb_mbaff_edge*v() more often than do extra calculations | Michael Niedermayer |
2010-01-18 | Use h->slice_num where possible. | Michael Niedermayer |
2010-01-18 | Enable filter_mb_fast for CAVLC P slices. | Michael Niedermayer |
2010-01-18 | PAFF CABAC P slices seem to work as well, so enable them for ff_h264_filter_m... | Michael Niedermayer |
2010-01-18 | Reenable filter_mb_fast for I slices and progressive CABAC P slices. | Michael Niedermayer |
2010-01-18 | Move CAVLC 8x8 DCT special case from ff_h264_filter_mb() to fill_caches | Michael Niedermayer |
2010-01-18 | Perform reference remapping at fill_cache() time instead of in the | Michael Niedermayer |
2010-01-18 | Move the qp check to skip the loop filter up. | Michael Niedermayer |
2010-01-18 | Reorganize how values are stored in h->non_zero_count. | Michael Niedermayer |
2010-01-17 | Rearchitecturing the stiched up goose part 1 | Michael Niedermayer |
2010-01-16 | Comment for() ; out | Michael Niedermayer |
2010-01-16 | Mark a few functions as noinline, this makes ff_h264_filter_mb() a bit smaller | Michael Niedermayer |
2010-01-16 | Apply last 2 optimizations to similar code i forgot. | Michael Niedermayer |
2010-01-16 | Another microopt, 4 cpu cycles for avoidance of FFABS(). | Michael Niedermayer |
2010-01-16 | Minor (2 cpu cycles) optimization ||->|. | Michael Niedermayer |