Mercurial > libavcodec.hg
graph
-
aacenc: Favor log2f() and sqrtf() over log2() and sqrt(). libavcodecSat, 24 Jul 2010 02:10:59 +0000, by alexc
-
VP8: Use vp56_rac_get_prob_branchy when the bit is only used by an if() libavcodecFri, 23 Jul 2010 21:46:20 +0000, by conrad
-
Decode DCT tokens by branching to a different code path for each branch libavcodecFri, 23 Jul 2010 21:46:17 +0000, by conrad
-
Move renormalization of the VP56 arith decoder to before decoding a bit libavcodecFri, 23 Jul 2010 21:46:14 +0000, by conrad
-
Split renorm of vp56 arith decoder to its own function libavcodecFri, 23 Jul 2010 21:46:08 +0000, by conrad
-
vp56's arith decoder's code_word is only 16 bits, no need for unsigned long libavcodecFri, 23 Jul 2010 21:46:01 +0000, by conrad
-
Add myself to VP8 copyright and maintainers. libavcodecFri, 23 Jul 2010 21:42:35 +0000, by darkshikari
-
VP8: always_inline some things to force gcc to do the right thing libavcodecFri, 23 Jul 2010 21:36:21 +0000, by darkshikari
-
VP8: use AV_RL24 instead of defining a new RL24. libavcodecFri, 23 Jul 2010 21:17:18 +0000, by darkshikari
-
VP8: use AV_ZERO32 instead of AV_WN32A where relevant libavcodecFri, 23 Jul 2010 10:42:19 +0000, by darkshikari
-
VP8: shave a few clocks off check_intra_pred_mode libavcodecFri, 23 Jul 2010 10:24:38 +0000, by darkshikari
-
VP8: fix broken sign bias code in MV pred libavcodecFri, 23 Jul 2010 06:41:35 +0000, by darkshikari
-
VP8: optimize DC-only chroma case in the same way as luma. libavcodecFri, 23 Jul 2010 06:02:52 +0000, by darkshikari
-
In h264 parser, return immediately if buf_size is 0, avoid printing libavcodecFri, 23 Jul 2010 00:34:09 +0000, by bcoudurier
-
VP8: clear DCT blocks in iDCT instead of using clear_blocks. libavcodecFri, 23 Jul 2010 00:07:16 +0000, by darkshikari
-
VP8: avoid a memset for non-i4x4 blocks with no coefficients libavcodecFri, 23 Jul 2010 00:05:44 +0000, by darkshikari
-
Get rid of more unnecessary dereferences in VP8 deblocking libavcodecThu, 22 Jul 2010 23:11:40 +0000, by darkshikari
-
Eliminate a LUT in escape decoding in VP8 decode_block_coeffs libavcodecThu, 22 Jul 2010 22:08:09 +0000, by darkshikari
-
Eliminate some repeated dereferences in VP8 inter_predict libavcodecThu, 22 Jul 2010 21:05:30 +0000, by darkshikari
-
Use pextrw for SSE4 mbedge filter result writing, speedup 5-10cycles on libavcodecThu, 22 Jul 2010 19:59:34 +0000, by rbultje
-
Eliminate a pointless memset for intra blocks in P-frames in VP8 libavcodecThu, 22 Jul 2010 12:15:29 +0000, by darkshikari
-
VP8: Don't store segment in macroblock struct anymore. libavcodecThu, 22 Jul 2010 11:55:55 +0000, by darkshikari
-
Convert VP8 macroblock structures to a ring buffer. libavcodecThu, 22 Jul 2010 11:45:18 +0000, by darkshikari
-
Calculate deblock strength per-MB instead of per-row libavcodecThu, 22 Jul 2010 07:24:22 +0000, by darkshikari
-
Avoid tracking i4x4 modes in P-frames in VP8 libavcodecThu, 22 Jul 2010 07:04:45 +0000, by darkshikari
-
Avoid useless fill_rectangle in P-frames in VP8 libavcodecThu, 22 Jul 2010 06:39:54 +0000, by darkshikari
-
Fix and enable horizontal >=SSE2 mbedge loopfilter. libavcodecThu, 22 Jul 2010 01:35:26 +0000, by rbultje
-
Eliminate one instruction in VP8 dc_add_sse4 libavcodecWed, 21 Jul 2010 22:41:37 +0000, by darkshikari
-
Make ff_inverse stay with libavutil, and optional copy it to libavcodec. libavcodecWed, 21 Jul 2010 12:37:37 +0000, by flameeyes
-
Add header declarations for mmx/sse constants missing them libavcodecWed, 21 Jul 2010 10:02:07 +0000, by conrad
-
Move ff_pw_* from vc1dsp_mmx.c to dsputil_mmx.c libavcodecWed, 21 Jul 2010 10:02:03 +0000, by conrad
-
VP8 MBedge loopfilter MMX/MMX2/SSE2 functions for both luma (width=16) libavcodecTue, 20 Jul 2010 22:58:56 +0000, by rbultje
-
Chroma (width=8) inner loopfilter MMX/MMX2/SSE2 for VP8 decoder. libavcodecTue, 20 Jul 2010 22:04:18 +0000, by rbultje
-
vp8: add do { } while(0) around XCHG() macro to avoid confusing if/else libavcodecTue, 20 Jul 2010 17:54:25 +0000, by mru
-
Revert r24339 (it causes fate failures on x86-64) - I'll figure out what's libavcodecMon, 19 Jul 2010 23:57:09 +0000, by rbultje
-
Remove FF_MM_SSE2/3 flags for CPUs where this is generally not faster than libavcodecMon, 19 Jul 2010 22:38:23 +0000, by rbultje
-
Implement chroma (width=8) inner loopfilter MMX/MMX2/SSE2 functions. libavcodecMon, 19 Jul 2010 21:53:28 +0000, by rbultje
-
Be more efficient with registers or stack memory. Saves 8/16 bytes stack libavcodecMon, 19 Jul 2010 21:45:36 +0000, by rbultje