summaryrefslogtreecommitdiff
path: root/libavcodec/vp8.c
Commit message (Collapse)AuthorAge
...
* VP8: slightly faster loopfilter sharpness logicJason Garrett-Glaser2011-02-04
|
* VP8: faster deblock strength calculationJason Garrett-Glaser2011-02-04
| | | | Convert hev_thresh logic to a LUT, simplify mbedge_lim calculation.
* VP8: faster filter_level clipJason Garrett-Glaser2011-02-03
|
* VP8: simplify lf_delta mb mode logicJason Garrett-Glaser2011-02-03
|
* VP8: merge chroma MC callsJason Garrett-Glaser2011-01-31
| | | | | Adds some duplicated code, but avoids duplicate edge checks and similar. ~0.5% faster overall on Parkjoy test sample.
* Slightly simplify VP8 inter_predictJason Garrett-Glaser2011-01-30
| | | | Merge an if and a switch.
* Move ff_emulated_edge_mc() into DSPContext.Ronald S. Bultje2011-01-28
|
* Fix VP8 aliasing problems.Ronald S. Bultje2011-01-28
| | | | Replace * (uint32_t *) buf accesses with AV_WN32A/AV_COPY32.
* Add ff_ prefix to data symbols of encoders, decoders, hwaccel, parsers, bsf.Diego Elio Pettenò2011-01-26
| | | | | | | None of these symbols should be accessed directly, so declare them as hidden. Signed-off-by: Mans Rullgard <mans@mansr.com>
* Don't do edge emulation unless the edge pixels will be used in MC.Ronald S. Bultje2011-01-25
| | | | | | Do not emulate larger edges than we will actually use for this round of MC. Decoding goes from avg+SE 29.972+/-0.023sec to 29.856+/-0.023, i.e. 0.12sec or ~0.4% faster.
* Fix valgrind invalid read on top MB rows with CODEC_FLAG_EMU_EDGE set.Ronald S. Bultje2010-12-30
| | | | Originally committed as revision 26168 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Support CODEC_FLAG_EMU_EDGE in VP8 decoder.Ronald S. Bultje2010-12-28
| | | | Originally committed as revision 26117 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use new imgutils.h API names, fix deprecation warnings.Stefano Sabatini2010-09-07
| | | | Originally committed as revision 25058 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove some stray +s in VP8Jason Garrett-Glaser2010-08-13
| | | | Originally committed as revision 24791 to svn://svn.ffmpeg.org/ffmpeg/trunk
* remove b4_stride/mb_stride.Pascal Massimino2010-08-11
| | | | | | | | correct mb_xy to use mb_width. tighten allocations. reduce the amount of zeroing. Originally committed as revision 24760 to svn://svn.ffmpeg.org/ffmpeg/trunk
* fix over-allocation. confused b4_stride with mb_width.Pascal Massimino2010-08-11
| | | | Originally committed as revision 24758 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove use of the deprecated function avcodec_check_dimensions(), useStefano Sabatini2010-08-06
| | | | | | av_check_image_size() instead. Originally committed as revision 24711 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: fix bug in prefetchJason Garrett-Glaser2010-08-05
| | | | | | Motion vectors in VP8 are qpel, not fullpel. Originally committed as revision 24707 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP5/6/8: eliminate CABAC dependencyJason Garrett-Glaser2010-08-04
| | | | | | | Create a custom table for VP5/6/8's renorm to avoid depending on H.264's. Saves one instruction in the arithmetic decoder as well. Originally committed as revision 24701 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: partially inline decode_block_coeffsJason Garrett-Glaser2010-08-04
| | | | | | Avoids a function call in the case of empty DCT blocks (most of the time). Originally committed as revision 24691 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Fix 100L in r24689Jason Garrett-Glaser2010-08-04
| | | | | | Accidentally committed some timing code. Originally committed as revision 24690 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: simplify decode_block_coeffs to avoid having to track nonzero coeffsJason Garrett-Glaser2010-08-04
| | | | | | Slightly faster. Originally committed as revision 24689 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: slightly faster DCT coefficient probability updateJason Garrett-Glaser2010-08-03
| | | | Originally committed as revision 24687 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: make another RAC call branchyJason Garrett-Glaser2010-08-03
| | | | | | 1-2 clocks faster. Originally committed as revision 24683 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: unroll partition type decoding treeJason Garrett-Glaser2010-08-03
| | | | | | ~34% faster partition type decoding. Originally committed as revision 24681 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: unroll splitmv decoding treeJason Garrett-Glaser2010-08-03
| | | | | | Much faster splitmv mode decoding. Originally committed as revision 24680 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: unroll MB mode decoding treeJason Garrett-Glaser2010-08-03
| | | | | | ~50% faster MB mode decoding, plus eliminate a costly switch. Originally committed as revision 24679 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: eliminate a dereference in coefficient decodingJason Garrett-Glaser2010-08-02
| | | | Originally committed as revision 24671 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: much faster DC transform handlingJason Garrett-Glaser2010-08-02
| | | | | | | | | | A lot of the time the DC block is empty: don't do the WHT in this case. A lot of the rest of the time, there's only one coefficient: make a special DC-only transform for that case. When the block is empty, don't incorrectly mark luma DCT blocks as having DC coefficients. Originally committed as revision 24670 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: move zeroing of luma DC block into the WHTJason Garrett-Glaser2010-08-02
| | | | | | | Lets us do the zeroing in asm instead of C. Also makes it consistent with the way the regular iDCT code does it. Originally committed as revision 24668 to svn://svn.ffmpeg.org/ffmpeg/trunk
* only store intra prediction modes on the boundary for keyframes, not as a plane.Pascal Massimino2010-08-02
| | | | | | inter-frame behaviour unchanged. Originally committed as revision 24664 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: simplify token_prob handlingJason Garrett-Glaser2010-08-02
| | | | | | ~1.5% faster decode_block_coeffs Originally committed as revision 24659 to svn://svn.ffmpeg.org/ffmpeg/trunk
* prevent access to vp8_coeff_band[16]Pascal Massimino2010-08-01
| | | | Originally committed as revision 24656 to svn://svn.ffmpeg.org/ffmpeg/trunk
* b0rk3d FATE + black helicopters hissing -> rolling back to r24556 and sleepingPascal Massimino2010-07-27
| | | | Originally committed as revision 24559 to svn://svn.ffmpeg.org/ffmpeg/trunk
* perform the clipping on luma_dc_qmul[1] and chroma_qmul[0] earlierPascal Massimino2010-07-27
| | | | Originally committed as revision 24558 to svn://svn.ffmpeg.org/ffmpeg/trunk
* save some copies by moving some fields out of proba[2]Pascal Massimino2010-07-27
| | | | Originally committed as revision 24557 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: add missing freeJason Garrett-Glaser2010-07-26
| | | | | | Fixes a tiny memory leak. Originally committed as revision 24504 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Fix r24445: Instead of needlessly initialising a variable, silence the warning.Carl Eugen Hoyos2010-07-25
| | | | Originally committed as revision 24498 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: Inline traversing vp8_small_mvtreeDavid Conrad2010-07-23
| | | | | | Much faster read_mv_component, slightly faster overall Originally committed as revision 24470 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: Use vp56_rac_get_prob_branchy when the bit is only used by an if()David Conrad2010-07-23
| | | | Originally committed as revision 24469 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Decode DCT tokens by branching to a different code path for each branchDavid Conrad2010-07-23
| | | | | | | | | | on the huffman tree, instead of traversing the tree in a while loop. Based on the similar optimization in libvpx's detokenize.c 10% faster at normal bitrates, and 30% faster for high-bitrate intra-only Originally committed as revision 24468 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Add myself to VP8 copyright and maintainers.Jason Garrett-Glaser2010-07-23
| | | | | | Also add Ronald to maintainers. Originally committed as revision 24464 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: always_inline some things to force gcc to do the right thingJason Garrett-Glaser2010-07-23
| | | | | | Mostly seems to help in the MC code, which gets a hundred cycles faster. Originally committed as revision 24463 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: use AV_RL24 instead of defining a new RL24.Jason Garrett-Glaser2010-07-23
| | | | Originally committed as revision 24462 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: Slightly faster MV selectionJason Garrett-Glaser2010-07-23
| | | | | | Don't clamp best mv unless it's actually used. Originally committed as revision 24461 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: use AV_ZERO32 instead of AV_WN32A where relevantJason Garrett-Glaser2010-07-23
| | | | Originally committed as revision 24460 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: eliminate redundant code in r24458Jason Garrett-Glaser2010-07-23
| | | | Originally committed as revision 24459 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: shave a few clocks off check_intra_pred_modeJason Garrett-Glaser2010-07-23
| | | | Originally committed as revision 24458 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: fix broken sign bias code in MV predJason Garrett-Glaser2010-07-23
| | | | | | | Apparently the official conformance test vectors don't test this feature, even though libvpx uses it. Originally committed as revision 24456 to svn://svn.ffmpeg.org/ffmpeg/trunk
* VP8: optimize DC-only chroma case in the same way as luma.Jason Garrett-Glaser2010-07-23
| | | | | | | Add MMX idct_dc_add4uv function for this case. ~40% faster chroma idct. Originally committed as revision 24455 to svn://svn.ffmpeg.org/ffmpeg/trunk
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-07 12:30:06 +0000