- 07 1月, 2017 4 次提交
-
-
由 Paul B Mahol 提交于
Signed-off-by: NPaul B Mahol <onemda@gmail.com>
-
由 Paul B Mahol 提交于
Signed-off-by: NPaul B Mahol <onemda@gmail.com>
-
由 Andreas Cadhalpun 提交于
Signed-off-by: NAndreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
-
由 Kevin Wheatley 提交于
looks like there is a bug in commit 1a08758e relating to the handling of ptr in decode_frame after decode_block is called, before this commit ptr would have been incremented for each line in the data window, now after the commit it is left at the start of the first included line rather than the line after the data window then the code sets the remaining lines to 0 and thus the whole image is over written. Fix by adjusting ptr to the correct line after decode_block returns Signed-off-by: NKevin Wheatley <kevin.j.wheatley@gmail.com>
-
- 06 1月, 2017 8 次提交
-
-
由 Michael Niedermayer 提交于
Fixes CID1244189 Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Paul B Mahol 提交于
Signed-off-by: NPaul B Mahol <onemda@gmail.com>
-
由 Steven Liu 提交于
CID: 1396257 Reviewed-by: NMichael Niedermayer <michael@niedermayer.cc> Signed-off-by: NSteven Liu <lq@chinaffmpeg.org>
-
由 Rostislav Pehlivanov 提交于
This commit replaces the current inefficient non-power-of-two FFT with a much faster FFT based on the Prime Factor Algorithm. Although it is already much faster than the old algorithm without SIMD, the new algorithm makes use of the already very throughouly SIMD'd power of two FFT, which improves performance even more across all platforms which we have SIMD support for. Most of the work was done by Peter Barfuss, who passed the code to me to implement into the iMDCT and the current codebase. The code for a 5-point and 15-point FFT was derived from the previous implementation, although it was optimized and simplified, which will make its future SIMD easier. The 15-point FFT is currently using 6% of the current overall decoder overhead. The FFT can now easily be used as a forward transform by simply not multiplying the 5-point FFT's imaginary component by -1 (which comes from the fact that changing the complex exponential's angle by -1 also changes the output by that) and by multiplying the "theta" angle of the main exptab by -1. Hence the deliberately left multiplication by -1 at the end. FATE passes, and performance reports on other platforms/CPUs are welcome. Performance comparisons: iMDCT, PFA: 101127 decicycles in speed, 32765 runs, 3 skips iMDCT, Old: 211022 decicycles in speed, 32768 runs, 0 skips Standalone FFT, 300000 transforms of size 960: PFA Old FFT kiss_fft libfftw3f 3.659695s, 15.726912s, 13.300789s, 1.182222s Being only 3x slower than libfftw3f is a big achievement by itself. There appears to be something capping the performance in the iMDCT side of things, possibly during the pre-stage reindexing. However, it is certainly fast enough for now. Signed-off-by: NRostislav Pehlivanov <atomnuker@gmail.com>
-
由 Rostislav Pehlivanov 提交于
Prep work for the next commit, which will add a new FFT algorithm which makes the iMDCT over 3x faster than it is currently (standalone, the FFT is with some framesizes over 10x faster). The new FFT algorithm uses the already thouroughly SIMD'd power of two FFT which already has SIMD for AArch64, so users of that platform will still see an improvement. The previous FFT+SIMD was barely 2.5x faster than the C versions on these platforms. Signed-off-by: NRostislav Pehlivanov <atomnuker@gmail.com>
-
由 Steven Liu 提交于
CID: 1398228 Passing null pointer dirname to strlen, which dereferences it. Reviewed-by: NMichael Niedermayer <michael@niedermayer.cc> Signed-off-by: NSteven Liu <lq@chinaffmpeg.org>
-
由 Steve Lhomme 提交于
Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Bela Bodecs 提交于
Signed-off-by: NBela Bodecs <bodecsb@vivanet.hu> Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
- 05 1月, 2017 7 次提交
-
-
由 Bela Bodecs 提交于
in get_default_pattern_localtime_fmt the default pattern contains %Y%m%d%H%I%S but the original intention was %Y%m%d%H%M%S Signed-off-by: NBela Bodecs <bodecsb@vivanet.hu> Signed-off-by: NSteven Liu <lq@chinaffmpeg.org>
-
由 Carl Eugen Hoyos 提交于
-
由 Michael Niedermayer 提交于
Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Michael Niedermayer 提交于
Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Carl Eugen Hoyos 提交于
Also fixes compilation with --disable-error-resilience. Fixes ticket #6060.
-
由 Steven Liu 提交于
fix CID: 1398364 Resource leak refine the code of the new options Reviewed-by: NBodecs Bela <bodecsb@vivanet.hu> Signed-off-by: NSteven Liu <lq@chinaffmpeg.org>
-
由 Bela Bodecs 提交于
A wrong, unitialized variable is used for testing. This patch fixes this typo. Signed-off-by: NBela Bodecs <bodecsb@vivanet.hu> Signed-off-by: NSteven Liu <lq@chinaffmpeg.org>
-
- 04 1月, 2017 2 次提交
-
-
由 Paul B Mahol 提交于
Signed-off-by: NPaul B Mahol <onemda@gmail.com>
-
由 Carl Eugen Hoyos 提交于
When the mapping was originally added AV_CODEC_ID_MP1 did not exist.
-
- 03 1月, 2017 9 次提交
-
-
由 Bela Bodecs 提交于
1st: This patch makes it possible to put actual segment file size (measured in bytes) and/or duration (calculated in microseconds) into segment filenames. This feature is useful when post-processing live streaming access log files. New behaviour works only when -use_localtime option is set and second_level_segment_size or/and second_level_segment_duration new hls_flags are specified. %%s is the placeholder for size and %%t for duration in hls_segment_filename option. Fix sized trailing zeropadding also works eg. %%09s or %%023t. A command to test new features: ./ffmpeg -loglevel info -y -f lavfi -i color=c=red:size=640x480:r=25 -f lavfi -i sine=f=440:b=4:r=44100 -c:v mpeg2video -g 25 -acodec aac -cutoff 20000 -ac 2 -ar 44100 -ab 192k -f hls -hls_time 3 -hls_list_size 5 -hls_flags second_level_segment_index+second_level_segment_size+second_level_segment_duration -use_localtime 1 -use_localtime_mkdir 1 -hls_segment_filename "segment_%Y%m%d%H%M%S_%%04d_%%08s_%%013t.ts" stream.m3u8 2nd: doc/muxers: beside second_level_segment_duration and second_level_segment_size, added some more details and example to hls_segment_filename, use_localtime, use_localtime_mkdir, hls_flags. hls_flags option list reformatted to table Signed-off-by: NBela Bodecs <bodecsb@vivanet.hu> Signed-off-by: NSteven Liu <lq@chinaffmpeg.org>
-
由 Tobias Rapp 提交于
Signed-off-by: NTobias Rapp <t.rapp@noa-archive.com> Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Michael Niedermayer 提交于
Fixes CID1396539 Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Andreas Cadhalpun 提交于
This prevents a division by zero crash in wmavoice_decode_packet. The problem was introduced by commit 3deb4b54. Reviewed-by: NRonald S. Bultje <rsbultje@gmail.com> Signed-off-by: NAndreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
-
由 Andreas Cadhalpun 提交于
This fixes triggering the av_assert0(ret <= tmp.size). The problem was reintroduced by commit 7b27dd5c and originally fixed by 2a4700a4. Reviewed-by: NRonald S. Bultje <rsbultje@gmail.com> Signed-off-by: NAndreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
-
由 Michael Niedermayer 提交于
make fate passes Reviewed-by: N"Ronald S. Bultje" <rsbultje@gmail.com> Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 John Comeau 提交于
fixes `operation size not specified` errors as described here: http://stackoverflow.com/questions/36854583/compiling-ffmpeg-for-kali-linux-2 I rebuilt again with yasm and made sure it didn't break that. Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Steven Liu 提交于
if the http server don't response the http command, then the thread will be blocked and never be interrupted. Reported-by: Nyinyunjiang <yinyunjiang1991@qq.com> Signed-off-by: NSteven Liu <lq@chinaffmpeg.org> Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Michael Niedermayer 提交于
Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
- 02 1月, 2017 6 次提交
-
-
由 Carl Eugen Hoyos 提交于
Fixes ticket #6045.
-
由 James Almer 提交于
Happy new year!
-
由 Thomas Turner 提交于
Signed-off-by: NThomas Turner <thomastdt@googlemail.com> Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Michael Niedermayer 提交于
Fixes CID1396253 Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Carl Eugen Hoyos 提交于
Also increases the score for large jpeg files. Fixes autodetection for the file from mpv issue 3973.
-
由 Carl Eugen Hoyos 提交于
Fixes ticket #6055.
-
- 01 1月, 2017 4 次提交
-
-
由 Thomas Turner 提交于
Signed-off-by: NThomas Turner <thomastdt@googlemail.com> Signed-off-by: NMichael Niedermayer <michael@niedermayer.cc>
-
由 Miroslav Slugen 提交于
Signed-off-by: NTimo Rothenpieler <timo@rothenpieler.org>
-
由 Miroslav Slugen 提交于
Signed-off-by: NTimo Rothenpieler <timo@rothenpieler.org>
-
由 Miroslav Slugeň 提交于
Round qpIntra and qpInter calculation instead of old floor behavior. Adopted from vaapi_encode_h264.c Signed-off-by: NTimo Rothenpieler <timo@rothenpieler.org>
-