summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* lavfi: make request_frame() non-recursive.Nicolas George2015-12-225-5/+99
| | | | | | | Instead of calling the input filter request_frame() method, ff_request_frame() now marks the link and returns immediately. buffersink is changed to activate the marked filters until a frame is obtained.
* lavfi: replace link.closed by link.status.Nicolas George2015-12-228-21/+49
| | | | | | The status field can carry any error code instead of just EOF. Also only update it through a wrapper function and provide a timestamp. Update the few filters that used it directly.
* lavfi: deprecate avfilter_link_set_closed().Nicolas George2015-12-223-1/+10
| | | | | | | Applications are not supposed to mess with links, they should close the sinks. Furthermore, this function does not distinguish what end of the link caused the close and does not have a timestamp.
* lavfi: add link.current_pts field.Nicolas George2015-12-222-0/+8
|
* lavfi: rename link.current_pts to current_pts_us.Nicolas George2015-12-223-6/+6
| | | | | | | This field is used for fast comparison between link ages, it is in AV_TIME_BASE units, in other words microseconds, µs =~ us. Renaming it allows a second field in link time base units.
* lavfi/vf_mpdecimate: remove request_frame().Nicolas George2015-12-221-14/+0
| | | | It is no longer needed since looping is not necessary.
* avcodec/ass: check for av_mallocz() failureMichael Niedermayer2015-12-221-0/+2
| | | | Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* MAINTAINERS: add Eran Kornblau for aes_ctr and movenccencerankor2015-12-221-0/+2
| | | | Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* vaapi: Add VP9 hwaccell supportTimo Rothenpieler2015-12-227-2/+179
| | | | Signed-off-by: Timo Rothenpieler <timo@rothenpieler.org>
* lavf/rmdec: Use correct format specifier for int64_t.Carl Eugen Hoyos2015-12-221-1/+1
| | | | Fixes ticket #5100.
* AAC encoder: fix possible assertion failure in PNSClaudio Freire2015-12-221-2/+3
| | | | | | | Fix possible SF delta violation that would cause an eventual assertion failure in some corner cases (esp on very low bitrates) when marking bands for PNS due to misuse of the sf_delta utilities
* aacenc_is: rename variableRostislav Pehlivanov2015-12-211-5/+5
| | | | | | | | 'erf' is far from the best name for a variable and is not very descriptive since the actual variable points to the comparitively best IS phase. Therefore rename it to 'best'. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
* lavu/libm: add erf hack and make dynaudnorm available everywhereGanesh Ajjanagadde2015-12-212-1/+201
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Source code is from Boost: http://www.boost.org/doc/libs/1_46_1/boost/math/special_functions/erf.hpp with appropriate modifications for FFmpeg. Tested on interval -6 to 6 (beyond which it saturates), +/-NAN, +/-INFINITY under -fsanitize=undefined on clang to test for possible undefined behavior. This function turns out to actually be essentially as accurate and faster than the libm (GNU/BSD's/Mac OS X), and I can think of 3 reasons why upstream does not use this: 1. They are not aware of it. 2. They are concerned about licensing - this applies especially to GNU libm. 3. They do not know and/or appreciate the benefits of rational approximations over polynomial approximations. Boost uses them to great effect, see e.g swr/resample for bessel derived from them, which is also similarly superior to libm variants. First, performance. sample benchmark (clang -O3, Haswell, GNU/Linux): 3e8 values evenly spaced from 0 to 6 time (libm): ./test 13.39s user 0.00s system 100% cpu 13.376 total time (boost based): ./test 9.20s user 0.00s system 100% cpu 9.190 total Second, accuracy. 1e8 eval pts from 0 to 6 maxdiff (absolute): 2.2204460492503131e-16 occuring at point where libm erf is correctly rounded, this is not. Illustration of superior rounding of this function: arg : 0.83999999999999997 erf : 0.76514271145499457 boost : 0.76514271145499446 real : 0.76514271145499446 i.e libm is actually incorrectly rounded. Note that this is clear from: https://github.com/JuliaLang/openlibm/blob/master/src/s_erf.c (the Sun implementation used by both BSD and GNU libm's), where only 1 ulp is guaranteed. Reasons it is not easy/worthwhile to create a "correctly rounded" variant of this function (i.e 0.5ulp): 1. Upstream libm's don't do it anyway, so we can't guarantee this unless we force this implementation on all platforms. This is not easy, as the linker would complain unless measures are taken. 2. Nothing in FFmpeg cares or can care about such things, due to the above and FFmpeg's nature. 3. Creating a correctly rounded function will in practice need some use of long double/fma. long double, although C89/C90, unfortunately has problems on ppc. This needs fixing of toolchain flags/configure. In any case this will be slower for miniscule gain. Reviewed-by: James Almer <jamrial@gmail.com> Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavf/avformat: add av_warn_unused_result to avformat_write_headerGanesh Ajjanagadde2015-12-211-0/+1
| | | | | | May be useful as a defense, see e.g c62d1780fff8a1997dd1707bbc557efc8fe41e3c. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavd/pulse_audio_enc: replace lround by lrintGanesh Ajjanagadde2015-12-211-1/+2
| | | | | | | Here it is mostly a cosmetic change, but there might be benefits in that there are no compat hacks for lround, while there are for lrint. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavfi/vf_histogram: replace round by lrintGanesh Ajjanagadde2015-12-211-1/+1
| | | | | | | lrint is at least as fast, uses a superior rounding mode, and avoids an implicit cast. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavfi/af_dynaudnorm: replace round by lrintGanesh Ajjanagadde2015-12-211-1/+1
| | | | | | | lrint is at least as fast, uses a superior rounding mode, and avoids an implicit cast. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavfi/vf_crop: replace round by lrintGanesh Ajjanagadde2015-12-211-1/+1
| | | | | | | lrint is at least as fast, avoids an implicit cast, and uses a superior rounding mode. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavc/libvpxenc: replace round by lrintGanesh Ajjanagadde2015-12-211-2/+3
| | | | | | Mostly cosmetic here. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavfi/vf_drawtext: replace round by llrintGanesh Ajjanagadde2015-12-211-1/+1
| | | | | | llrint is at least as fast, and avoids an implicit cast. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavfi/vf_colorlevels: replace round by lrintGanesh Ajjanagadde2015-12-211-8/+8
| | | | | | | lrint avoids an implicit cast, and is not slower on non-broken libm's. Thus this represents a Pareto improvement. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavfi/vf_colorchannelmixer: replace round by lrintGanesh Ajjanagadde2015-12-211-19/+19
| | | | | | | | | | | | | lrint is faster here on -ftree-vectorize with GCC. This is likely simply an artifact of GCC's rather terrible auto-vectorizer, since as per the instruction set manuals cvtsd2si and cvttsd2si (or their vector equivalents) have identical cycle timings. Anyway, regardless of above, lrint is superior to round accuracy wise. Safety guaranteed as long int has at least 32 bits. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* lavfi/drawtext: fix shadow[xy] descriptionsClément Bœsch2015-12-211-2/+2
|
* lavfi/drawtext: hide first font load warning when fontconfig is presentClément Bœsch2015-12-211-0/+2
|
* lavfi/drawtext: fix crash when no text, file or timecode providedClément Bœsch2015-12-211-6/+6
|
* avfilter/af_ladspa: fix av_assert0()Paul B Mahol2015-12-211-2/+2
| | | | Signed-off-by: Paul B Mahol <onemda@gmail.com>
* lavf/img2enc: add atomic_writing optionClément Bœsch2015-12-211-1/+3
| | | | | This behaviour change caused a regression on our side recently, we might want to disable the option by default.
* lavc: add text encoderClément Bœsch2015-12-216-5/+253
|
* lavfi/scale: add nb_slices debug optionClément Bœsch2015-12-211-0/+12
|
* avcodec/indeo2: use init_get_bits8Paul B Mahol2015-12-201-1/+2
| | | | Signed-off-by: Paul B Mahol <onemda@gmail.com>
* x86/hevc_sao: add ff_hevc_sao_edge_filter_{8,16}_{10,12}James Almer2015-12-202-7/+10
| | | | | Reviewed-by: Christophe Gisquet <christophe.gisquet@gmail.com> Signed-off-by: James Almer <jamrial@gmail.com>
* x86/hevc_sao: simplify sao_edge_filter 10/12bitJames Almer2015-12-201-96/+54
| | | | | | Reviewed-by: Michael Niedermayer <michaelni@gmx.at> Reviewed-by: Christophe Gisquet <christophe.gisquet@gmail.com> Signed-off-by: James Almer <jamrial@gmail.com>
* x86/hevc_sao: simplify sao_band_filter 10/12bitJames Almer2015-12-201-85/+57
| | | | | | Reviewed-by: Michael Niedermayer <michaelni@gmx.at> Reviewed-by: Christophe Gisquet <christophe.gisquet@gmail.com> Signed-off-by: James Almer <jamrial@gmail.com>
* avfilter/avf_showfreqs: make it possible to split channelsPaul B Mahol2015-12-202-2/+28
| | | | Signed-off-by: Paul B Mahol <onemda@gmail.com>
* avcodec/flacenc: use designated initializers for AVClassPaul B Mahol2015-12-201-4/+4
| | | | Signed-off-by: Paul B Mahol <onemda@gmail.com>
* avcodec/s302menc: check if buf_size can actually be put into 16bit sizePaul B Mahol2015-12-201-0/+5
| | | | | | This disallows creating unplayable audio. Signed-off-by: Paul B Mahol <onemda@gmail.com>
* avcodec/s302menc: set supported channel layouts by codecPaul B Mahol2015-12-201-0/+5
| | | | Signed-off-by: Paul B Mahol <onemda@gmail.com>
* nuv: sanitize negative fps rateAndreas Cadhalpun2015-12-201-0/+9
| | | | Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
* rawdec: only exempt BIT0 with need_copy from buffer sanity checkAndreas Cadhalpun2015-12-201-1/+1
| | | | | | | | Otherwise the too small buffer is directly used in the frame, causing segmentation faults, when trying to use the frame. Reviewed-by: Michael Niedermayer <michael@niedermayer.cc> Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
* mlvdec: check that index_entries existAndreas Cadhalpun2015-12-201-0/+5
| | | | | | | This fixes NULL pointer dereferencing. Reviewed-by: Michael Niedermayer <michael@niedermayer.cc> Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
* avcodec/mpeg4videodec: also for empty partitioned slicesMichael Niedermayer2015-12-191-1/+1
| | | | | | | | Fixes assertion failure Fixes: id_acf3e47f864e1ee4c7b86c0653e0ff31e5bde56e.m4v Found-by: Andreas Cadhalpun <andreas.cadhalpun@googlemail.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* lavu/libm: add copysign hackGanesh Ajjanagadde2015-12-192-1/+10
| | | | | | | | | | | | For systems with broken libms. Tested with NAN, -NAN, INFINITY, -INFINITY, +/-x for regular double x and combinations of these. Old versions of MSVC need some UINT64_C hackery. Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com> Reviewed-by: Hendrik Leppkes <h.leppkes@gmail.com> Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* avcodec/Makefile: add missing dep for g723_1 encoderJames Almer2015-12-191-1/+1
| | | | Signed-off-by: James Almer <jamrial@gmail.com>
* avfilter/af_dynaudnorm: use av_malloc_array()Paul B Mahol2015-12-191-5/+5
| | | | Signed-off-by: Paul B Mahol <onemda@gmail.com>
* avcodec/h264_refs: Fix long_idx checkMichael Niedermayer2015-12-191-1/+1
| | | | | | | | Fixes out of array read Fixes mozilla bug 1233606 Found-by: Tyson Smith Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* swscale/arm/yuv2rgb: add ff_yuv420p_to_{argb,rgba,abgr,bgra}_neon_{16,32}Matthieu Bouron2015-12-192-11/+118
|
* swscale/arm/yuv2rgb: disable neon if accurate_rnd is enabledMatthieu Bouron2015-12-191-4/+3
| | | | | This disables the 32bit precision neon code path in favor of the default C one and avoids breaking fate.
* avcodec/ffv1enc: Fix 2 pass mode with the default rc tableMichael Niedermayer2015-12-191-2/+13
| | | | Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* avfilter/vf_stereo3d: add interleave columns input supportPaul B Mahol2015-12-192-51/+172
| | | | Signed-off-by: Paul B Mahol <onemda@gmail.com>
* lavu/intmath: add faster clz supportGanesh Ajjanagadde2015-12-191-0/+18
| | | | | | | | This should be useful for the sofalizer filter. Reviewed-by: Kieran Kunhya <kierank@ob-encoder.com> Reviewed-by: Clément Bœsch <u@pkh.me> Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
OpenPOWER on IntegriCloud