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_loopfilter.c
AgeCommit message (Collapse)Author
2012-01-14cosmetics: Move static and inline attributes to more standard places.Diego Biurrun
Fixes several "‘static’ is not at beginning of declaration" warnings.
2011-10-21h264: 4:2:2 intra decoding supportBaptiste Coudurier
Signed-off-by: Diego Biurrun <diego@biurrun.de> Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-07-12H.264: add filter_mb_fast support for >8-bit decodingJason Garrett-Glaser
Much faster high bit depth deblocking.
2011-07-11H.264: avoid redundant alpha/beta calculations in loopfilterJason Garrett-Glaser
2011-07-11H.264: optimize intra/inter loopfilter decisionJason Garrett-Glaser
2011-07-11Eliminate FF_COMMON_FRAME macro.Diego Biurrun
FF_COMMON_FRAME holds the contents of the AVFrame structure and is also copied to struct Picture. Replace by an embedded AVFrame structure in struct Picture.
2011-07-10H.264: fix filter_mb_fast with 4:4:4 + 8x8dctJason Garrett-Glaser
2011-07-09H.264: improve qp_thresh checkJason Garrett-Glaser
Eliminate redundant check in filter_mb_fast, consider bit depth in calculating qp_thresh.
2011-07-06H.264: make loopfilter bS const where applicableJason Garrett-Glaser
2011-07-04H.264: template left MB handlingJason Garrett-Glaser
Faster H.264 decoding with ALLOW_INTERLACE off.
2011-07-04H.264: make filter_mb_fast support the case of unavailable top mbJason Garrett-Glaser
Significantly faster deblocking in streams with lots of slices.
2011-06-27H.264: make filter_mb_fast support 4:4:4Jason Garrett-Glaser
2011-06-22H.264: fix 4:4:4 + deblocking + 8x8dct + cavlc + MBAFFJason Garrett-Glaser
2011-06-22H.264: fix 4:4:4 + deblocking + MBAFFJason Garrett-Glaser
2011-06-14H.264: fix CODEC_FLAG_GRAYJason Garrett-Glaser
It was broken in 4:4:4, and still did chroma deblocking for no reason in 4:2:0.
2011-06-144:4:4 H.264 decoding supportJason Garrett-Glaser
Note: this is 4:4:4 from the 2007 spec revision, not the previous (now deprecated) 4:4:4 mode in H.264.
2011-06-14Roll back 4:4:4 H.264 for nowJason Garrett-Glaser
Needs some ARM/PPC asm modifications.
2011-06-13H.264: fix CODEC_FLAG_GRAYJason Garrett-Glaser
It was broken in 4:4:4, and still did chroma deblocking for no reason in 4:2:0.
2011-06-134:4:4 H.264 decoding supportJason Garrett-Glaser
Note: this is 4:4:4 from the 2007 spec revision, not the previous (now deprecated) 4:4:4 mode in H.264.
2011-05-10Add the notion of pixel size in h264 related functions.Oskar Arvidsson
In high bit depth the pixels will not be stored in uint8_t like in the normal case, but in uint16_t. The pixel size is thus 1 in normal bit depth and 2 in high bit depth. Preparatory patch for high bit depth h264 decoding support. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-05-10Make the h264 loop filter bit depth aware.Oskar Arvidsson
Preparatory patch for high bit depth h264 decoding support. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-05-10h264: DSP'ize MBAFF loopfilter.Ronald S. Bultje
2011-03-19Replace FFmpeg with Libav in licence headersMans Rullgard
Signed-off-by: Mans Rullgard <mans@mansr.com>
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
2010-03-16Move H264 dsputil functions into their own structMåns Rullgård
This moves the H264-specific functions from DSPContext to the new H264DSPContext. The code is made conditional on CONFIG_H264DSP which is set by the codecs requiring it. The qpel and chroma MC functions are not moved as these are used by non-h264 code. Originally committed as revision 22565 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-06Remove DECLARE_ALIGNED_{8,16} macrosMåns Rullgård
These macros are redundant. All uses are replaced with the generic DECLARE_ALIGNED macro instead. Originally committed as revision 22233 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-18H264: use alias-safe macrosMåns Rullgård
This eliminates all aliasing violation warnings in h264 code. No measurable speed difference with gcc-4.4.3 on i7. Originally committed as revision 21881 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-17Use LOCAL_ALIGNED macro for local arraysMåns Rullgård
Originally committed as revision 21866 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-14h264: Remove unused variables.Alexander Strange
Originally committed as revision 21815 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-07Fix CAVLC+8x8DCT+MBAFF loopfiltering.Michael Niedermayer
Fixes issue1250 Originally committed as revision 21665 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-31Get rid of a check in one direction that cant be true in it in that partMichael Niedermayer
of the code. No meassureable speed change. Originally committed as revision 21566 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-30Split first reference list comparission from mv comparission.Michael Niedermayer
about 0.5% faster MBAFF loop filtering Originally committed as revision 21552 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-30Replace h->left_type[0] by the local variable for it we have.Michael Niedermayer
No meassureable speed effect. Originally committed as revision 21541 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-30slightly faster bit trickery.Michael Niedermayer
Originally committed as revision 21540 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-30Replace ?: by branchless code.Michael Niedermayer
about 0.5% faster loop filtering Originally committed as revision 21539 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28factorize first filter call out, this makes the code somewhatMichael Niedermayer
smaller without any speed loss. Originally committed as revision 21514 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28Change wraper functions to always inline, they are faster now that way.Michael Niedermayer
1% faster MBAFF decoding overall, maybe ~0.1% faster for the cathedral sample. Originally committed as revision 21507 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28indentMichael Niedermayer
Originally committed as revision 21506 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28Restructure check_mv()Michael Niedermayer
~20 cpu cycles faster loopfilter Originally committed as revision 21505 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28Restructure if() in check_mv()Michael Niedermayer
quite a bit faster Originally committed as revision 21504 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28Unroll loops in check_mv()Michael Niedermayer
~6% faster (slow path) loopfilter (should be ~2% overall) Originally committed as revision 21503 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28Factor mv/ref compare code out.Michael Niedermayer
This is a hair slower (0.15% maybe) but i really dont want to have the identical code duplicated 3 times because gcc adds odd threaded jumps with register reshuffling and register safe/restore. Originally committed as revision 21502 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28Simplify first edge filter condition.Michael Niedermayer
Originally committed as revision 21497 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28Cosmetics, mostly indention, 2 or so new fixme comments that i was to lazyMichael Niedermayer
to split out Originally committed as revision 21496 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28Make the fast loop filter path work with unavailable left MBs.Michael Niedermayer
This prevents the issue with having to switch between slow and fast code paths in each row. 0.5% faster loopfilter for cathedral Originally committed as revision 21495 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28get rid of the start variable.Michael Niedermayer
a few cycles faster Originally committed as revision 21494 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-28Unroll main loop so the edge==0 case is seperate.Michael Niedermayer
This allows many things to be simplified away. h264 decoder is overall 1% faster with a mbaff sample and 0.1% slower with the cathedral sample, probably because the slow loop filter code must be loaded into the code cache for each first MB of each row but isnt used for the following MBs. Originally committed as revision 21493 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-27Update comment.Michael Niedermayer
Originally committed as revision 21479 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-27Use table to speedup access to non_zero_count in MBAFF with differing ↵Michael Niedermayer
interlacing. ~4 cpu cycles speedup Originally committed as revision 21474 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-01-27Optimize loop filtering of the left edge in MBAFF.Michael Niedermayer
60 cpu cycles speedup Originally committed as revision 21467 to svn://svn.ffmpeg.org/ffmpeg/trunk
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-19 08:17:16 +0300