Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/FFmpeg/FFmpeg.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
AgeCommit message (Expand)Author
2011-01-21Set gray (128) U/V planes for chroma-less samples. Fixes two fate samplesRonald S. Bultje
2011-01-21Revert 2a1f431d38ea9c05abb215d70c7dc09cdb6888ab, it broke H264 lossless.Ronald S. Bultje
2011-01-15Remove evil timers that snuck their way into r26375.Jason Garrett-Glaser
2011-01-15Fix r26375 on non-x86.Jason Garrett-Glaser
2011-01-15H.264: Partially inline CABAC residual decodingJason Garrett-Glaser
2011-01-15H.264/SVQ3: make chroma DC work the same way as luma DCJason Garrett-Glaser
2011-01-15H.264: switch to x264-style tracking of luma/chroma DC NNZJason Garrett-Glaser
2011-01-15H.264: split luma dc idct out and implement MMX/SSE2 versionsJason Garrett-Glaser
2010-04-20Remove explicit filename from Doxygen @file commands.Diego Biurrun
2010-03-30Replace @returns by @return.Benoit Fouet
2010-03-26h264: Use + instead of | in some placesAlexander Strange
2010-03-26h264: Remove unused function argumentAlexander Strange
2010-03-26h264: Simplify decode_cabac_residual() specializationAlexander Strange
2010-03-01Remove some unneeded fill_rectangle() for 16x16 blocks.Michael Niedermayer
2010-02-27Optimize (amvd>2)+(amvd>32), about 1 cpu cycles faster.Zhou Zongyi
2010-02-24Change mvd_cache & mvd_table to 8bit, this is overall a bit fasterMichael Niedermayer
2010-02-24Calculate mvd without abs()Michael Niedermayer
2010-02-24switch back to (amvd>2)+(amvd>32), its 5 cpu cycles faster now.Michael Niedermayer
2010-02-24Factorize common code from the top of decode_cabac_mb_mvd()Michael Niedermayer
2010-02-24Replace mvd>2 + mvd>32 by MIN((mvd+28)*17>>9, 2)Michael Niedermayer
2010-02-24Replace ad-hoc fill rectangle by fill_rectangle().Michael Niedermayer
2010-02-19get rid of an if() 1 cpu cycle faster.Michael Niedermayer
2010-02-19Get rid of a local variable, 10 cpu cycles faster.Michael Niedermayer
2010-02-19Move abs() from decode_cabac_mb_mvd() to the code that writes mvd_cache.Michael Niedermayer
2010-02-18Speedup decode_cabac_field_decoding_flag() by 9 cpu cycles.Michael Niedermayer
2010-02-17Move check for and call of predict_field_decoding_flag() from the mb code toMichael Niedermayer
2010-02-172x faster ff_h264_init_cabac_states(), 4k cpu cycles less.Michael Niedermayer
2010-02-162 cpu cycles faster context calculation for decode_cabac_intra_mb_type()Michael Niedermayer
2010-02-16Drop a few redundant slice_num checks.Michael Niedermayer
2010-02-16Drop compute_mb_neighbors() and move fill_decode_neighbors() up to take itsMichael Niedermayer
2010-02-16Split setting neighboring MBs from fill_decode_caches()Michael Niedermayer
2010-02-15Simplify decode_cabac_mb_intra4x4_pred_mode().Michael Niedermayer
2010-02-15Merge decode_cabac_mb_type_b() into calling code.Michael Niedermayer
2010-02-15Merge the single line function decode_cabac_mb_transform_size()Michael Niedermayer
2010-02-15indentMichael Niedermayer
2010-02-15Merge decode_cabac_mb_dqp() with surronding code.Michael Niedermayer
2010-02-14Set sub_mb_type in direct_cache instead of just the direct flag.Michael Niedermayer
2010-02-14Store sub_mb_type in direct_cache/direct_table.Michael Niedermayer
2010-02-14Remove slice_table checks from decode_cabac_mb_cbp_luma() and set left/top_cb...Michael Niedermayer
2010-01-25Optimize decode_cabac_field_decoding_flag().Michael Niedermayer
2010-01-22Move array specifiers outside DECLARE_ALIGNED() invocationsMåns Rullgård
2010-01-19Cosmetic, get rid of &x[0]Michael Niedermayer
2010-01-18Split fill_caches() between filter and decoder.Michael Niedermayer
2010-01-17Rearchitecturing the stiched up goose part 1Michael Niedermayer
2010-01-16Replace cabac checks in inline functions from h264.h with constants.Michael Niedermayer
2010-01-13Split cabac decoding code out of h264.c.Michael Niedermayer