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
path: root/libavcodec/h264_cabac.c
AgeCommit message (Collapse)Author
2011-01-21Set gray (128) U/V planes for chroma-less samples. Fixes two fate samplesRonald S. Bultje
when played with -flags emu_edge. (cherry picked from commit 8bcfe7f7fd71e5d548c2cd96d33295ff06a5f6c8)
2011-01-21Revert 2a1f431d38ea9c05abb215d70c7dc09cdb6888ab, it broke H264 lossless.Ronald S. Bultje
(cherry picked from commit 66c6b5e2a5e6376c4a5c65b1d96fd7a1580bdfcf)
2011-01-15Remove evil timers that snuck their way into r26375.Jason Garrett-Glaser
Originally committed as revision 26377 to svn://svn.ffmpeg.org/ffmpeg/trunk
2011-01-15Fix r26375 on non-x86.Jason Garrett-Glaser
Originally committed as revision 26376 to svn://svn.ffmpeg.org/ffmpeg/trunk
2011-01-15H.264: Partially inline CABAC residual decodingJason Garrett-Glaser
Improves CABAC performance about ~1.2%. Trick originates from x264 and has also been used in ffvp8. It's useful because coded block flags are usually zero, so it helps to have the early termination inlined into the main function. Originally committed as revision 26375 to svn://svn.ffmpeg.org/ffmpeg/trunk
2011-01-15H.264/SVQ3: make chroma DC work the same way as luma DCJason Garrett-Glaser
No speed improvement, but necessary for some future stuff. Also opens up the possibility of asm chroma dc idct/dequant. Originally committed as revision 26349 to svn://svn.ffmpeg.org/ffmpeg/trunk
2011-01-15H.264: switch to x264-style tracking of luma/chroma DC NNZJason Garrett-Glaser
Useful so that we don't have to run the hierarchical DC iDCT if there aren't any coefficients. Opens up some future opportunities for optimization as well. Originally committed as revision 26337 to svn://svn.ffmpeg.org/ffmpeg/trunk
2011-01-15H.264: split luma dc idct out and implement MMX/SSE2 versionsJason Garrett-Glaser
About 2.5x the speed. NOTE: the way that the asm code handles large qmuls is a bit suboptimal. If x264-style dequant was used (separate shift and qmul values), it might be possible to get some extra speed. Originally committed as revision 26336 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-04-20Remove explicit filename from Doxygen @file commands.Diego Biurrun
Passing an explicit filename to this command is only necessary if the documentation in the @file block refers to a file different from the one the block resides in. Originally committed as revision 22921 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-30Replace @returns by @return.Benoit Fouet
Originally committed as revision 22729 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-26h264: Use + instead of | in some placesAlexander Strange
6 insns less on x86-64/gcc 4.2. Originally committed as revision 22692 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-26h264: Remove unused function argumentAlexander Strange
Originally committed as revision 22690 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-26h264: Simplify decode_cabac_residual() specializationAlexander Strange
Gives more consistent inlining with some compilers (such as llvm). Originally committed as revision 22689 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-01Remove some unneeded fill_rectangle() for 16x16 blocks.Michael Niedermayer
Originally committed as revision 22124 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-27Optimize (amvd>2)+(amvd>32), about 1 cpu cycles faster.Zhou Zongyi
patch by Zhou Zongyi @ zhouzy () os punkt pku dot edu speck cn Originally committed as revision 22084 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24Change mvd_cache & mvd_table to 8bit, this is overall a bit fasterMichael Niedermayer
for high resolution videos. about 20cycles faster per MB for cathederal. Originally committed as revision 22038 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24Calculate mvd without abs()Michael Niedermayer
same speed (ask gcc why, i dont know) Originally committed as revision 22035 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24switch back to (amvd>2)+(amvd>32), its 5 cpu cycles faster now.Michael Niedermayer
Originally committed as revision 22032 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24Factorize common code from the top of decode_cabac_mb_mvd()Michael Niedermayer
10-15 cpu cycles faster. Originally committed as revision 22029 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24Replace mvd>2 + mvd>32 by MIN((mvd+28)*17>>9, 2)Michael Niedermayer
same speed as far as i can meassure but it might have fewer branches on some archs. Idea from x264 / jason Originally committed as revision 22027 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24Replace ad-hoc fill rectangle by fill_rectangle().Michael Niedermayer
Originally committed as revision 22025 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-19get rid of an if() 1 cpu cycle faster.Michael Niedermayer
Originally committed as revision 21889 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-19Get rid of a local variable, 10 cpu cycles faster.Michael Niedermayer
Originally committed as revision 21888 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-19Move abs() from decode_cabac_mb_mvd() to the code that writes mvd_cache.Michael Niedermayer
4-8 cycles faster Originally committed as revision 21887 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-18Speedup decode_cabac_field_decoding_flag() by 9 cpu cycles.Michael Niedermayer
Originally committed as revision 21875 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-17Move check for and call of predict_field_decoding_flag() from the mb code toMichael Niedermayer
the row code. This function would only be needed on a MB basis for MBAFF+FMO Originally committed as revision 21860 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-172x faster ff_h264_init_cabac_states(), 4k cpu cycles less.Michael Niedermayer
Sadly this is just per slice so the speedup with normal files should be negligible. Originally committed as revision 21859 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-162 cpu cycles faster context calculation for decode_cabac_intra_mb_type()Michael Niedermayer
Originally committed as revision 21845 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-16Drop a few redundant slice_num checks.Michael Niedermayer
Originally committed as revision 21844 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-16Drop compute_mb_neighbors() and move fill_decode_neighbors() up to take itsMichael Niedermayer
role. Should be faster as this is a strict code removial. Originally committed as revision 21843 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-16Split setting neighboring MBs from fill_decode_caches()Michael Niedermayer
no speed change. Originally committed as revision 21842 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-15Simplify decode_cabac_mb_intra4x4_pred_mode().Michael Niedermayer
same speed Originally committed as revision 21839 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-15Merge decode_cabac_mb_type_b() into calling code.Michael Niedermayer
This avoids a conditional branch and is about 3 cpu cyclues faster. Originally committed as revision 21838 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-15Merge the single line function decode_cabac_mb_transform_size()Michael Niedermayer
into the calling code. 8 cpu cycles faster Originally committed as revision 21828 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-15indentMichael Niedermayer
Originally committed as revision 21827 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-15Merge decode_cabac_mb_dqp() with surronding code.Michael Niedermayer
~20 cpu cycles faster Originally committed as revision 21826 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-14Set sub_mb_type in direct_cache instead of just the direct flag.Michael Niedermayer
Simpler, cleaner and faster. Originally committed as revision 21822 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-14Store sub_mb_type in direct_cache/direct_table.Michael Niedermayer
This is equal complexity but could be more usefull. Originally committed as revision 21821 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-14Remove slice_table checks from decode_cabac_mb_cbp_luma() and set ↵Michael Niedermayer
left/top_cbp so these checks arent needed. Originally committed as revision 21819 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-25Optimize decode_cabac_field_decoding_flag().Michael Niedermayer
~4 cpu cycles faster Originally committed as revision 21447 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-22Move array specifiers outside DECLARE_ALIGNED() invocationsMåns Rullgård
Originally committed as revision 21377 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-19Cosmetic, get rid of &x[0]Michael Niedermayer
Originally committed as revision 21309 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-18Split fill_caches() between filter and decoder.Michael Niedermayer
Originally committed as revision 21271 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-17Rearchitecturing the stiched up goose part 1Michael Niedermayer
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. Originally committed as revision 21270 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-16Replace cabac checks in inline functions from h264.h with constants.Michael Niedermayer
No benchmark because its just replacing variables with litteral constants (so no risk for slowdown outside gcc silliness) and i need sleep. Originally committed as revision 21237 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-13Split cabac decoding code out of h264.c.Michael Niedermayer
not slower according to benchmarks. Originally committed as revision 21181 to svn://svn.ffmpeg.org/ffmpeg/trunk
    generated by cgit v1.2.3 (git 2.25.1) at 2024-08-30 23:41:44 +0300