Mercurial > libavcodec.hg
graph
-
factorize first filter call out, this makes the code somewhat libavcodecThu, 28 Jan 2010 19:44:13 +0000, by michael
-
Simplify. See "[PATCH] simplify ff_set_fixed_vector()" thread on mailinglist. libavcodecThu, 28 Jan 2010 14:29:53 +0000, by rbultje
-
Initialize thread_context[0] with h264 parser context. libavcodecThu, 28 Jan 2010 13:04:25 +0000, by benoit
-
Change wraper functions to always inline, they are faster now that way. libavcodecThu, 28 Jan 2010 11:37:35 +0000, by michael
-
Cosmetics, mostly indention, 2 or so new fixme comments that i was to lazy libavcodecThu, 28 Jan 2010 02:20:31 +0000, by michael
-
Make the fast loop filter path work with unavailable left MBs. libavcodecThu, 28 Jan 2010 02:15:25 +0000, by michael
-
Unroll main loop so the edge==0 case is seperate. libavcodecThu, 28 Jan 2010 01:24:25 +0000, by michael
-
Fix possible crashes in mlp parser, it tries to go back 7 bytes after libavcodecWed, 27 Jan 2010 20:04:19 +0000, by reimar
-
Reduce log level of "Extracting .. channel downmix" to AV_LOG_DEBUG, the libavcodecWed, 27 Jan 2010 20:02:18 +0000, by reimar
-
Use table to speedup access to non_zero_count in MBAFF with differing interlacing. libavcodecWed, 27 Jan 2010 11:14:29 +0000, by michael
-
Split fill_caches() between loopfilter & decode, the 2 no longer where common libavcodecTue, 26 Jan 2010 23:54:11 +0000, by michael
-
Optimize loop filtering of the left edge in MBAFF. libavcodecTue, 26 Jan 2010 22:59:19 +0000, by michael
-
use left_xy[1] in mbaff QP loop filter check, this improves the amount that can libavcodecTue, 26 Jan 2010 20:28:58 +0000, by michael
-
Use left_mb_xy from fill_caches instead of recalculating it. libavcodecTue, 26 Jan 2010 14:57:53 +0000, by michael
-
Simplify left_xy content for the loop filter, this also makes it closer to libavcodecTue, 26 Jan 2010 14:55:19 +0000, by michael
-
Simplify loop filter a little by using top/left_type. libavcodecTue, 26 Jan 2010 13:39:26 +0000, by michael
-
Set top & left types for deblock in fill_caches(). libavcodecTue, 26 Jan 2010 13:38:18 +0000, by michael
-
100l typo fix, mixed up +-1 forcing the loop filter skip to never skip. libavcodecTue, 26 Jan 2010 13:30:33 +0000, by michael
-
Optimize mb neighbor initialization for MBAFF in fill_caches(). libavcodecTue, 26 Jan 2010 10:35:36 +0000, by michael
-
Metasoft MJPEG codec has Cb and Cr swapped, fixes issue 1611. libavcodecTue, 26 Jan 2010 10:30:24 +0000, by cehoyos
-
Fix heap overflow due to lack of nb_components check. libavcodecMon, 25 Jan 2010 13:26:10 +0000, by michael
-
100l, correct wrong H.264+adaptive MBAFF decoding libavcodecMon, 25 Jan 2010 01:56:13 +0000, by astrange
-
H.264: Use 64-/128-bit write-combining macros for copies libavcodecMon, 25 Jan 2010 00:30:44 +0000, by astrange
-
Mark index as unsigend so gcc knows it doesnt need a sign extension on x86_64. libavcodecSun, 24 Jan 2010 19:19:09 +0000, by michael
-
Optimize suffix_length computation, 1 cpu cycle speedup. libavcodecSun, 24 Jan 2010 18:23:46 +0000, by michael
-
Simplify suffix_length computation, same speed. libavcodecSun, 24 Jan 2010 18:18:08 +0000, by michael
-
Optimize level_code computation, 6cpu cycles speedup. libavcodecSun, 24 Jan 2010 18:17:01 +0000, by michael
-
Fix crash in MLP decoder due to integer overflow. libavcodecSun, 24 Jan 2010 18:07:29 +0000, by reimar
-
1 cpu cycle faster suffix_length calculation. libavcodecSun, 24 Jan 2010 18:05:02 +0000, by michael
-
Remove is_complex from loop_filter() its useless there in its current form. libavcodecSun, 24 Jan 2010 17:02:44 +0000, by michael
-
Added a missing const to ff_h264_get_slice_type(). libavcodecSun, 24 Jan 2010 16:37:12 +0000, by fenrir
-
Remove all uses of slice_type* from the loop filter, also remove its libavcodecSun, 24 Jan 2010 13:20:17 +0000, by michael
-
Remove unneeded reset of non_zero_count_cache for deblock. libavcodecSat, 23 Jan 2010 22:11:46 +0000, by michael
-
Remove useless things from the deblock side of fill_caches(). libavcodecSat, 23 Jan 2010 21:57:36 +0000, by michael
-
Moved reusable functions from dxva2_h264.c to dxva2.c libavcodecSat, 23 Jan 2010 18:17:57 +0000, by fenrir
-
Move +52 from the loop filter to the alpha/beta offsets in the context. libavcodecSat, 23 Jan 2010 18:05:30 +0000, by michael
-
Use qstride = 0 and allocate only a single line for qscale_table since libavcodecSat, 23 Jan 2010 18:01:38 +0000, by reimar
-
Added ff_dxva2_ prefix to get_surface(_index) and commit_buffer functions. libavcodecSat, 23 Jan 2010 18:01:28 +0000, by fenrir
-
Reindent the content of one if(){} in ff_dxva2_common_end_frame. libavcodecSat, 23 Jan 2010 18:00:51 +0000, by fenrir
-
Moved reusable code from dxva2_h264.c:end_frame to ff_dxva2_common_end_frame. libavcodecSat, 23 Jan 2010 17:59:08 +0000, by fenrir
-
Set edges based on cbp and mv partitioning, not just skiped MBs. libavcodecSat, 23 Jan 2010 15:28:34 +0000, by michael
-
Use AVPALETTE_COUNT instead of hard-coding 256. libavcodecSat, 23 Jan 2010 15:22:48 +0000, by reimar
-
Remove redundant initialization of the palette, it is part of the context libavcodecSat, 23 Jan 2010 15:19:34 +0000, by reimar
-
Optmize 8x8dct check used to skip some borders in the loop filter. libavcodecSat, 23 Jan 2010 13:54:02 +0000, by michael
-
Use av_freep intead of av_free to free pointers in vp56 context. libavcodecSat, 23 Jan 2010 13:49:09 +0000, by reimar
-
Zero the frame data on allocation for VB codec, e.g. the FATE sample seems to libavcodecSat, 23 Jan 2010 10:25:22 +0000, by reimar
-
Export wma_critical_freqs as ff_wma_critical_freqs libavcodecSat, 23 Jan 2010 08:39:50 +0000, by daniel
-
The SSSE3 version of dct_quantize in mpegvideo_mmx_template.c needs libavcodecFri, 22 Jan 2010 23:40:18 +0000, by reimar
-
Implement an sse version of scalarproduct_float(). libavcodecFri, 22 Jan 2010 23:07:58 +0000, by alexc
-
Check for chunk boundaries when decoding VB codec data libavcodecFri, 22 Jan 2010 19:07:44 +0000, by kostya
-
Zero palette in case not all entries are initialized later libavcodecFri, 22 Jan 2010 18:28:41 +0000, by kostya
-
Move array specifiers outside DECLARE_ALIGNED() invocations libavcodecFri, 22 Jan 2010 03:25:11 +0000, by mru
-
Move dquant check into qscale overflow check. libavcodecThu, 21 Jan 2010 21:01:26 +0000, by michael
-
Make calculation of mask_edge free of branches, faster of course but probably libavcodecThu, 21 Jan 2010 16:50:31 +0000, by michael
-
Use two separate memory arguments since 8+() is invalid gas syntax libavcodecThu, 21 Jan 2010 09:46:57 +0000, by conrad
-
H.264: Declare bS with DECLARE_ALIGNED_8 for uint64_t casts. libavcodecWed, 20 Jan 2010 03:28:57 +0000, by astrange