summaryrefslogtreecommitdiffstats
path: root/libavcodec
Commit message (Expand)AuthorAgeFilesLines
* Deprecate av_parse_video_frame_size() and av_parse_video_frame_rate()Stefano Sabatini2010-07-263-134/+13
* Use pmaddubsw for the mbedge_filter (>=ssse3), 6-10 cycles faster.Ronald S. Bultje2010-07-261-2/+78
* VP8: Much faster SSE2 MCJason Garrett-Glaser2010-07-261-88/+78
* Define static functions fill_image_linesize() andS.N. Hemanth Meenakshisundaram2010-07-261-34/+45
* Enable no-loop memory/register saving for ssse3/sse4 also.Ronald S. Bultje2010-07-261-2/+2
* Save a register (or regsize of stackspace for x86-32) for the no-loopRonald S. Bultje2010-07-261-16/+24
* Use nested ifs instead of &&, which appears to not work with %ifidn (i.e. thisRonald S. Bultje2010-07-261-3/+9
* Rename pow variable to pwr.Axel Holzinger2010-07-261-4/+4
* Split pextrw macro-spaghetti into several opt-specific macros, this will makeRonald S. Bultje2010-07-261-30/+49
* VP8: add missing freeJason Garrett-Glaser2010-07-261-0/+1
* Fix r24445: Instead of needlessly initialising a variable, silence the warning.Carl Eugen Hoyos2010-07-251-1/+1
* Only 4-bit ADPCM IMA WAV files are supported.Carl Eugen Hoyos2010-07-251-0/+6
* Fix obvious bug in assignment. Somehow, the test vectors don't test this...Ronald S. Bultje2010-07-251-1/+1
* add SubRip muxer and demuxerAurelien Jacobs2010-07-241-0/+1
* Fix SPLATB_REG mess. Used to be a if/elseif/elseif/elseif spaghetti, so thisRonald S. Bultje2010-07-241-33/+52
* aacenc: TLS: Save maximum values for each swb in a table.Alex Converse2010-07-241-6/+14
* 10l: Remove some commented out code that slipped in.Alex Converse2010-07-241-1/+0
* aacenc: TLS: Try to preserve some energy in each non-zero band.Alex Converse2010-07-241-1/+6
* Use "const" qualifier for pointers that point to input data ofReimar Döffinger2010-07-2410-25/+25
* aacenc: Prevent premature termination of the two loop search.Alex Converse2010-07-241-3/+1
* aacdec: Only load and write each predictor variable once.Alex Converse2010-07-241-9/+12
* aacdec: 4% faster main profile decoding.Alex Converse2010-07-241-4/+7
* aacenc: Favor log2f() and sqrtf() over log2() and sqrt().Alex Converse2010-07-241-7/+7
* aacenc: Factorize some scalefactor utilities.Alex Converse2010-07-231-4/+14
* Inline asm for VP56 arith coderEli Friedman2010-07-232-5/+61
* VP8: Inline traversing vp8_small_mvtreeDavid Conrad2010-07-232-13/+12
* VP8: Use vp56_rac_get_prob_branchy when the bit is only used by an if()David Conrad2010-07-231-5/+5
* Decode DCT tokens by branching to a different code path for each branchDavid Conrad2010-07-233-40/+67
* Move renormalization of the VP56 arith decoder to before decoding a bitDavid Conrad2010-07-231-15/+11
* Split renorm of vp56 arith decoder to its own functionDavid Conrad2010-07-231-13/+17
* vp56's arith decoder's code_word is only 16 bits, no need for unsigned longDavid Conrad2010-07-231-2/+2
* Add myself to VP8 copyright and maintainers.Jason Garrett-Glaser2010-07-231-0/+1
* VP8: always_inline some things to force gcc to do the right thingJason Garrett-Glaser2010-07-231-32/+43
* VP8: use AV_RL24 instead of defining a new RL24.Jason Garrett-Glaser2010-07-231-6/+4
* VP8: Slightly faster MV selectionJason Garrett-Glaser2010-07-231-8/+9
* VP8: use AV_ZERO32 instead of AV_WN32A where relevantJason Garrett-Glaser2010-07-231-3/+3
* VP8: eliminate redundant code in r24458Jason Garrett-Glaser2010-07-231-1/+1
* VP8: shave a few clocks off check_intra_pred_modeJason Garrett-Glaser2010-07-231-6/+5
* VP8: fix broken sign bias code in MV predJason Garrett-Glaser2010-07-231-1/+2
* VP8: optimize DC-only chroma case in the same way as luma.Jason Garrett-Glaser2010-07-235-43/+89
* VP8: Clean up some variable shadowing.Jason Garrett-Glaser2010-07-231-6/+6
* VP8 asm: cosmetics (spacing)Jason Garrett-Glaser2010-07-231-2/+2
* VP8: 30% faster idct_mbJason Garrett-Glaser2010-07-235-87/+190
* VP8: smarter prefetchingJason Garrett-Glaser2010-07-231-8/+8
* In h264 parser, return immediately if buf_size is 0, avoid printingBaptiste Coudurier2010-07-231-0/+3
* VP8: clear DCT blocks in iDCT instead of using clear_blocks.Jason Garrett-Glaser2010-07-234-6/+29
* VP8: avoid a memset for non-i4x4 blocks with no coefficientsJason Garrett-Glaser2010-07-231-4/+3
* Get rid of more unnecessary dereferences in VP8 deblockingJason Garrett-Glaser2010-07-221-33/+36
* Shut up an uninitialized variable GCC warning in VP8.Jason Garrett-Glaser2010-07-221-1/+1
* Smarter VP8 prefetchingJason Garrett-Glaser2010-07-221-12/+21
OpenPOWER on IntegriCloud