ffmpeg.git
3 years agoswscale: cleanup unused code
Pedro Arthur [Mon, 28 Mar 2016 16:25:18 +0000 (13:25 -0300)]
swscale: cleanup unused code

Removed previous swscale code under '#ifndef NEW_FILTER'
and removed unused fields of SwsContext

3 years agoswscale: Deprecate vector functions which are unused outside swscale
Michael Niedermayer [Thu, 31 Mar 2016 15:20:58 +0000 (17:20 +0200)]
swscale: Deprecate vector functions which are unused outside swscale

There are no known users of these functions within debian
It should be thus possible to remove these functions without recommandition of a
replacement

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoconfigure: Fix debugging on mingw-w64 with gdb
Alex Smith [Tue, 29 Mar 2016 05:30:50 +0000 (01:30 -0400)]
configure: Fix debugging on mingw-w64 with gdb

The relocation hack broke debugging on mingw-w64 when using gdb.  This
makes the reloc hack dependent on --disable-debug so it's still enabled
for release builds.

This is simply an immediate fix for the issue of broken debugging, we
should probably still look at the possibility of reverting it outright
if it proves to be more trouble than it's worth.  For now keeping it
enabled for release builds is a reasonable trade off.

Signed-off-by: Alex Smith <theryuu@warpsharp.info>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodoc/filters: add stereotools examples
Paul B Mahol [Thu, 31 Mar 2016 09:12:40 +0000 (11:12 +0200)]
doc/filters: add stereotools examples

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agodoc/filters: add sofalizer examples
Paul B Mahol [Thu, 31 Mar 2016 09:04:42 +0000 (11:04 +0200)]
doc/filters: add sofalizer examples

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agosws/aarch64: add ff_hscale_8_to_15_neon
Clément Bœsch [Fri, 18 Mar 2016 15:39:34 +0000 (16:39 +0100)]
sws/aarch64: add ff_hscale_8_to_15_neon

./ffmpeg -nostats -f lavfi -i testsrc2=4k:d=2 -vf bench=start,scale=1024x1024,bench=stop -f null -

    before: t:0.489726 avg:0.489883 max:0.491852 min:0.489482
    after:  t:0.256515 avg:0.256458 max:0.256999 min:0.253755

3 years agoAAC encoder: fix undefined behavior
Claudio Freire [Wed, 30 Mar 2016 21:34:08 +0000 (18:34 -0300)]
AAC encoder: fix undefined behavior

Fix uninitialized access of minsf in short windows
Fix potential invocation of coef2minsf(0)

3 years agoavformat/latmenc: auto-insert aac_adtstoasc bitstream filter when needed
James Almer [Sat, 26 Mar 2016 04:42:35 +0000 (01:42 -0300)]
avformat/latmenc: auto-insert aac_adtstoasc bitstream filter when needed

Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
3 years agofate: Add filter-metadata-cropdetect
Michael Niedermayer [Thu, 24 Mar 2016 16:55:11 +0000 (17:55 +0100)]
fate: Add filter-metadata-cropdetect

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoAAC encoder: fix signed integer overflow
Claudio Freire [Wed, 30 Mar 2016 13:05:03 +0000 (10:05 -0300)]
AAC encoder: fix signed integer overflow

Clamp scalefactors by coef2minsf to avoid undefined behavior
caused by signed integer overflow. It also avoids clipping of
coefficients so it should avoid artifacts as well, on very
rare corner cases.

3 years agoavcodec/nvenc: Generate AUD NAL units for better compatiblity
Timo Rothenpieler [Wed, 30 Mar 2016 10:03:59 +0000 (12:03 +0200)]
avcodec/nvenc: Generate AUD NAL units for better compatiblity

3 years agotests/audiomatch: Reduce search window
Michael Niedermayer [Tue, 22 Mar 2016 11:21:27 +0000 (12:21 +0100)]
tests/audiomatch: Reduce search window

This avoids unneeded computations

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agotests/audiomatch: also print signal vs reference length
Michael Niedermayer [Tue, 22 Mar 2016 08:59:07 +0000 (09:59 +0100)]
tests/audiomatch: also print signal vs reference length

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodoc/filters: remove false claim in sofalizer description
Paul B Mahol [Tue, 29 Mar 2016 20:29:31 +0000 (22:29 +0200)]
doc/filters: remove false claim in sofalizer description

Remove false claim that processing in freq domain is
worse quality than processing in time domain.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoRevert "aacenc: use av_clip() instead of av_clip() during quantization"
Rostislav Pehlivanov [Wed, 30 Mar 2016 01:31:22 +0000 (02:31 +0100)]
Revert "aacenc: use av_clip() instead of av_clip() during quantization"

This reverts commit c0918613a0ecaac6819409c64107583eebc0ccc2, since it
triggered a new assertion with gcc.
This bug definitely needs a true proper fix.

3 years agoRevert "doc/filters: remove false claim in sofalizer description"
Rostislav Pehlivanov [Wed, 30 Mar 2016 01:30:19 +0000 (02:30 +0100)]
Revert "doc/filters: remove false claim in sofalizer description"

This reverts commit bf1495d9a991dd13e382a4c1871d96708fdec4e7.

3 years agoaacenc: use av_clip() instead of av_clip() during quantization
Rostislav Pehlivanov [Tue, 29 Mar 2016 23:58:28 +0000 (00:58 +0100)]
aacenc: use av_clip() instead of av_clip() during quantization

Seems like clang might be miscompiling it and causing a signed integer overflow,
making a FATE test fail.
Doesn't seem to affect performance, it only runs on the ESC codebook.

Reviewed-by: Claudio Freire <klaussfreire@gmail.com>
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
3 years agodoc/filters: remove false claim in sofalizer description
Paul B Mahol [Tue, 29 Mar 2016 20:29:31 +0000 (22:29 +0200)]
doc/filters: remove false claim in sofalizer description

Remove false claim that processing in freq domain is
worse quality than processing in time domain.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavfilter/vf_waveform: optimize lowpass filter even more
Paul B Mahol [Tue, 29 Mar 2016 13:00:45 +0000 (15:00 +0200)]
avfilter/vf_waveform: optimize lowpass filter even more

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavcodec/mips: Optimization synced to the newest code base.
Jovan Zelincevic [Tue, 29 Mar 2016 07:43:35 +0000 (09:43 +0200)]
avcodec/mips: Optimization synced to the newest code base.

FFT expanded to 2^17.

Signed-off-by: Jovan Zelincevic <jovan.zelincevic@imgtec.com>
Reviewed-by: Nedeljko Babic <Nedeljko.Babic@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agocfhd: Do not initialize context size
Vittorio Giovara [Tue, 29 Mar 2016 14:03:44 +0000 (16:03 +0200)]
cfhd: Do not initialize context size

Otherwise probing and stream analisys will report a correct coded size
but an empty visible size.

Approved by: kieran
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavformat/svag: fix division by zero
James Almer [Tue, 29 Mar 2016 00:34:08 +0000 (21:34 -0300)]
avformat/svag: fix division by zero

Fixes ticket #5386

Signed-off-by: James Almer <jamrial@gmail.com>
3 years agofix some a/an typos
Lou Logan [Mon, 28 Mar 2016 22:07:47 +0000 (14:07 -0800)]
fix some a/an typos

Signed-off-by: Lou Logan <lou@lrcd.com>
3 years agoavcodec/utils: fix packet duration of frames with discarded paddings
Marton Balint [Sun, 20 Mar 2016 13:48:08 +0000 (14:48 +0100)]
avcodec/utils: fix packet duration of frames with discarded paddings

Reviewed-by: Hendrik Leppkes <h.leppkes@gmail.com>
Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agolavd/dshow_crossbar: remove trailing whitespace
Lou Logan [Sun, 27 Mar 2016 04:58:52 +0000 (20:58 -0800)]
lavd/dshow_crossbar: remove trailing whitespace

Signed-off-by: Lou Logan <lou@lrcd.com>
3 years agoavfilter/af_sofalizer: allow user to setup custom virtual speakers positions
Paul B Mahol [Mon, 28 Mar 2016 16:07:45 +0000 (18:07 +0200)]
avfilter/af_sofalizer: allow user to setup custom virtual speakers positions

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agolavc/mediacodec: fix zero stride for OMX.allwinner.video.decoder.avc
Kirill Gavrilov [Mon, 28 Mar 2016 16:51:24 +0000 (19:51 +0300)]
lavc/mediacodec: fix zero stride for OMX.allwinner.video.decoder.avc

3 years agoavfilter/vf_waveform: optimize 16bit lowpass filter
Paul B Mahol [Mon, 28 Mar 2016 10:28:03 +0000 (12:28 +0200)]
avfilter/vf_waveform: optimize 16bit lowpass filter

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavcodec: add dca core extraction bsf
Paul B Mahol [Sun, 27 Mar 2016 11:02:33 +0000 (13:02 +0200)]
avcodec: add dca core extraction bsf

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agolavc/mediacodec: use ternary operator to set slice-height value
Matthieu Bouron [Mon, 28 Mar 2016 08:10:21 +0000 (10:10 +0200)]
lavc/mediacodec: use ternary operator to set slice-height value

3 years agoavcodec/diracdec: check bitstream size related fields for overflows
Michael Niedermayer [Mon, 28 Mar 2016 02:01:08 +0000 (04:01 +0200)]
avcodec/diracdec: check bitstream size related fields for overflows

Fixes segfault
Fixes Ticket5333

Regression since bfc8a4dabe5a0154b31128b59dca575010176441
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoconfigure: Fail if CUDA enabled but not found
Timo Rothenpieler [Fri, 25 Mar 2016 23:13:39 +0000 (00:13 +0100)]
configure: Fail if CUDA enabled but not found

Without this patch, configure still passes and enables CUDA, no matter
if it was actually found, breaking the build in case it was not.

3 years agoavfilter/vf_waveform: optimize lowpass 8bit filter
Paul B Mahol [Sun, 27 Mar 2016 21:29:38 +0000 (23:29 +0200)]
avfilter/vf_waveform: optimize lowpass 8bit filter

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavfilter/vf_waveform: move mirror variable into function argument
Paul B Mahol [Sun, 27 Mar 2016 20:50:04 +0000 (22:50 +0200)]
avfilter/vf_waveform: move mirror variable into function argument

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavcodec: Remove libdcadec, we already have it merged internally
Kieran Kunhya [Fri, 25 Mar 2016 21:32:26 +0000 (21:32 +0000)]
avcodec: Remove libdcadec, we already have it merged internally

3 years agoavcodec/h264_slice: Check PPS more extensively when its not copied
Michael Niedermayer [Fri, 25 Mar 2016 14:29:26 +0000 (15:29 +0100)]
avcodec/h264_slice: Check PPS more extensively when its not copied

Fixes Ticket5371
Fixes null pointer dereference

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavformat/segment: set format options for all segments
Marton Balint [Tue, 15 Mar 2016 18:19:20 +0000 (19:19 +0100)]
avformat/segment: set format options for all segments

Fixes ticket #5318.

Reviewed-by: Stefano Sabatini <stefasab@gmail.com>
Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agofate: Remove vsync drop from some h264 tests
Michael Niedermayer [Sun, 20 Mar 2016 19:10:40 +0000 (20:10 +0100)]
fate: Remove vsync drop from some h264 tests

Note some tests need vsync drop to produce exact timestamps, these seem not to
need it. quite likely many more dont need it either, ive not checked beyond finding
one that needs it and the ones which have it removed

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agofate: Modify a random h264 test to also test the -framerate option
Michael Niedermayer [Sun, 20 Mar 2016 19:07:14 +0000 (20:07 +0100)]
fate: Modify a random h264 test to also test the -framerate option

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/audiotoolboxenc: Fix compile error on OS X 10.8.
Dan Dennedy [Fri, 25 Mar 2016 04:13:18 +0000 (04:13 +0000)]
avcodec/audiotoolboxenc: Fix compile error on OS X 10.8.

Fixes error "libavcodec/audiotoolboxenc.c:294:50: error: use of undeclared
identifier 'kAudioCodecPropertyPacketSizeLimitForVBR'."

That was added to 10.9:
https://developer.apple.com/library/mac/releasenotes/General/APIDiffsMacOSX10_9/AudioUnit.html

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agolavc/flicvideo: Implement padding in COPY chunks.
Carl Eugen Hoyos [Sat, 26 Mar 2016 13:56:55 +0000 (14:56 +0100)]
lavc/flicvideo: Implement padding in COPY chunks.

Reviewed-by: Reimar
3 years agoavformat/rtpdec: Remove stray debug av_log()
Michael Niedermayer [Sat, 26 Mar 2016 09:50:43 +0000 (10:50 +0100)]
avformat/rtpdec: Remove stray debug av_log()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/exr: add support for B44 and B44A compression
Martin Vignali [Sun, 20 Mar 2016 10:53:27 +0000 (11:53 +0100)]
avcodec/exr: add support for B44 and B44A compression

3 years agoavcodec/jpeg2000dec: add slice threading support
Paul B Mahol [Fri, 25 Mar 2016 13:33:15 +0000 (14:33 +0100)]
avcodec/jpeg2000dec: add slice threading support

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavcodec/jpeg2000dec: account two last bytes from end of bytestream as EOC marker
Paul B Mahol [Fri, 25 Mar 2016 10:22:39 +0000 (11:22 +0100)]
avcodec/jpeg2000dec: account two last bytes from end of bytestream as EOC marker

This silences missing EOC marker when decoding j2k files with Psot set to 0.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agolavfi: Add coreimage filter for GPU based image filtering on OSX.
Thilo Borgmann [Sat, 19 Mar 2016 15:28:40 +0000 (16:28 +0100)]
lavfi: Add coreimage filter for GPU based image filtering on OSX.

3 years agolavu/dict: Add new flag to allow multiple equal keys.
Thilo Borgmann [Sat, 12 Mar 2016 13:52:17 +0000 (14:52 +0100)]
lavu/dict: Add new flag to allow multiple equal keys.

3 years agosws/aarch64/yuv2rgb: save a few mul and add
Clément Bœsch [Fri, 25 Mar 2016 14:49:09 +0000 (15:49 +0100)]
sws/aarch64/yuv2rgb: save a few mul and add

27ms to 26ms with UHD 2160 input.

3 years agoavcodec/libopenjpegenc: fix mixed declarations and code
James Almer [Fri, 25 Mar 2016 03:05:05 +0000 (00:05 -0300)]
avcodec/libopenjpegenc: fix mixed declarations and code

Signed-off-by: James Almer <jamrial@gmail.com>
3 years agoavcodec/libopenjpegdec: fix mixed declarations and code
James Almer [Fri, 25 Mar 2016 03:04:57 +0000 (00:04 -0300)]
avcodec/libopenjpegdec: fix mixed declarations and code

Signed-off-by: James Almer <jamrial@gmail.com>
3 years agoavcodec/libutvideodec: copy frame so it has reference counters when refcounted_frames...
Michael Niedermayer [Thu, 24 Mar 2016 19:55:30 +0000 (20:55 +0100)]
avcodec/libutvideodec: copy frame so it has reference counters when refcounted_frames is set

Reviewed-by: maintainer
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavutil/parseutils: mark args as static const
Michael Niedermayer [Thu, 24 Mar 2016 18:31:02 +0000 (19:31 +0100)]
avutil/parseutils: mark args as static const

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoAdded more tests to libavutil/parseutils.c
Petru Rares Sincraian [Tue, 22 Mar 2016 15:54:09 +0000 (16:54 +0100)]
Added more tests to libavutil/parseutils.c

    - Added tests for av_find_info_tag().
    - Added test for av_get_known_color_name()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavformat/rtpdec_jpeg: fix low contrast image on low quality setting
Ico Doornekamp [Thu, 24 Mar 2016 13:31:38 +0000 (14:31 +0100)]
avformat/rtpdec_jpeg: fix low contrast image on low quality setting

Original mail and my own followup on ffmpeg-user earlier today:

I have a device sending out a MJPEG/RTP stream on a low quality setting.
Decoding and displaying the video with libavformat results in a washed
out, low contrast, greyish image. Playing the same stream with VLC results
in proper color representation.

Screenshots for comparison:

  http://zevv.nl/div/libav/shot-ffplay.jpg
  http://zevv.nl/div/libav/shot-vlc.jpg

A pcap capture of a few seconds of video and SDP file for playing the
stream are available at

  http://zevv.nl/div/libav/mjpeg.pcap
  http://zevv.nl/div/libav/mjpeg.sdp

I believe the problem might be in the calculation of the quantization
tables in the function create_default_qtables(), the attached patch
solves the issue for me.

The problem is that the argument 'q' is of the type uint8_t. According to the
JPEG standard, if 1 <= q <= 50, the scale factor 'S' should be 5000 / Q.
Because the create_default_qtables() reuses the variable 'q' to store the
result of this calculation, for small values of q < 19, q wil subsequently
overflow and give wrong results in the calculated quantization tables. The
patch below uses a new variable 'S' (same name as in RFC2435) with the proper
range to store the result of the division.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agovc2enc: optimize and simplify quantization
Rostislav Pehlivanov [Thu, 24 Mar 2016 13:02:59 +0000 (13:02 +0000)]
vc2enc: optimize and simplify quantization

Everything except ORing the sign is now done in the LUT.

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
3 years agoavformat/mpegtsenc: Fix used service
Michael Niedermayer [Wed, 23 Mar 2016 01:58:52 +0000 (02:58 +0100)]
avformat/mpegtsenc: Fix used service

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavformat/mpegtsenc: Keep track of the program for each service
Michael Niedermayer [Wed, 23 Mar 2016 01:52:03 +0000 (02:52 +0100)]
avformat/mpegtsenc: Keep track of the program for each service

Simplifies code

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoRefactor libavutil/parseutils.c
Petru Rares Sincraian [Wed, 23 Mar 2016 15:32:56 +0000 (15:32 +0000)]
Refactor libavutil/parseutils.c

All tests were in the main method which produces a long main. Now, each test
is in his own method.

I think this produces a more clear code and follows more with the main
priority of FFmpeg "simplicity and small code size"

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agofate: Add test that also decodes packed_bframes.avi
Michael Niedermayer [Wed, 23 Mar 2016 18:24:43 +0000 (19:24 +0100)]
fate: Add test that also decodes packed_bframes.avi

No new reference samples are needed for this as the file already exists
for testing the bitstream filter

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agovc2enc: simplify calc_slice_sizes()
Rostislav Pehlivanov [Wed, 23 Mar 2016 18:04:05 +0000 (18:04 +0000)]
vc2enc: simplify calc_slice_sizes()

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
3 years agovc2enc: correctly zero out the slice size cache
Rostislav Pehlivanov [Wed, 23 Mar 2016 17:48:05 +0000 (17:48 +0000)]
vc2enc: correctly zero out the slice size cache

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
3 years agovc2enc: macro out slice size rounding
Rostislav Pehlivanov [Wed, 23 Mar 2016 17:46:23 +0000 (17:46 +0000)]
vc2enc: macro out slice size rounding

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
3 years agolavc/psymodel: check for av_malloc failure
Ganesh Ajjanagadde [Thu, 3 Mar 2016 01:14:08 +0000 (20:14 -0500)]
lavc/psymodel: check for av_malloc failure

No idea why in commit 01ecb7172b684f1c4b3e748f95c5a9a494ca36ec the
checks were removed; this can lead to NULL pointer dereferences. This
effectively reverts that portion of the commit.

Reviewed-by: Benoit Fouet <benoit.fouet@free.fr>
Reviewed-by: Rostislav Pehlivanov <atomnuker@gmail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanag@gmail.com>
3 years agoavcodec/dvaudiodec: now that we got samples, fix 12bit case
Paul B Mahol [Wed, 23 Mar 2016 08:26:35 +0000 (09:26 +0100)]
avcodec/dvaudiodec: now that we got samples, fix 12bit case

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agolavc: add AudioToolbox encoders
Rodger Combs [Wed, 24 Feb 2016 03:01:24 +0000 (21:01 -0600)]
lavc: add AudioToolbox encoders

Fixes trac #4828

3 years agolavc: add AudioToolbox decoders
Rodger Combs [Fri, 19 Feb 2016 03:38:37 +0000 (21:38 -0600)]
lavc: add AudioToolbox decoders

Part of trac #4828

3 years agoall: move ff_exp10, ff_exp10f, ff_fast_powf to lavu/ffmath.h
Ganesh Ajjanagadde [Tue, 15 Mar 2016 01:28:56 +0000 (21:28 -0400)]
all: move ff_exp10, ff_exp10f, ff_fast_powf to lavu/ffmath.h

The idea is to use ffmath.h for internal implementations of math functions.
Currently, it is used for variants of libm functions, but is by no means
limited to such things.

Note that this is not exported; use lavu/mathematics for such purposes.

Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanag@gmail.com>
3 years agoAdded a selftest to libavutil/display.c
Petru Rares Sincraian [Sun, 20 Mar 2016 08:54:07 +0000 (09:54 +0100)]
Added a selftest to libavutil/display.c

    - Check if av_display_rotation_get() gets the correct degrees
    - Check if av_display_rotation_set() sets the correct matrix
    - Check if av_display_matrix_flip() changes correct the matrix

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agovc2enc: add non-experimental support for all video formats from spec
Rostislav Pehlivanov [Mon, 21 Mar 2016 17:32:55 +0000 (17:32 +0000)]
vc2enc: add non-experimental support for all video formats from spec

Until now, for formats which were in the spec but not in the encoder's
list of supported formats required the -strict -1 flag. This enables
support for all video formats which are specified, all the way from
QSIF525 to 8K.

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
3 years agoswresample/swresample: Remove "less than" comparissions of enums
Michael Niedermayer [Tue, 22 Mar 2016 13:11:03 +0000 (14:11 +0100)]
swresample/swresample: Remove "less than" comparissions of enums

Found-by: wm4
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agolavf/img2dec: Skip COM when auto-detecting jpeg.
Carl Eugen Hoyos [Tue, 22 Mar 2016 10:13:24 +0000 (11:13 +0100)]
lavf/img2dec: Skip COM when auto-detecting jpeg.

It could theoretically contain invalid data that gets ignored by decoders.

3 years agofate: add audiomatch
Michael Niedermayer [Mon, 21 Mar 2016 03:18:48 +0000 (04:18 +0100)]
fate: add audiomatch

Testset provided by Justin Greer <justin@zencoder.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoconfigure: Remove (b)zlib and iconv dependencies for videoltoolbox encoder.
Carl Eugen Hoyos [Tue, 22 Mar 2016 00:30:33 +0000 (01:30 +0100)]
configure: Remove (b)zlib and iconv dependencies for videoltoolbox encoder.

3 years agortpdec: support for VC-2 HQ RTP payload format (draft v1)
Thomas Volkert [Mon, 21 Mar 2016 23:17:12 +0000 (00:17 +0100)]
rtpdec: support for VC-2 HQ RTP payload format (draft v1)

3 years agoffplay: remove redundant silence buffer
Lukasz Marek [Thu, 17 Mar 2016 17:26:09 +0000 (18:26 +0100)]
ffplay: remove redundant silence buffer

Signed-off-by: Lukasz Marek <lukasz.m.luki2@gmail.com>
3 years agoavformat/utils: scan a bit farther for a keyframe in mpeg/mpegts (7 sec instead of...
Michael Niedermayer [Sat, 19 Mar 2016 20:11:58 +0000 (21:11 +0100)]
avformat/utils: scan a bit farther for a keyframe in mpeg/mpegts (7 sec instead of 5, we already scan 90sec in some cases by default)

Fixes Ticket5305

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agovc2enc: simplify count_hq_slice() and caching
Rostislav Pehlivanov [Mon, 21 Mar 2016 11:04:49 +0000 (11:04 +0000)]
vc2enc: simplify count_hq_slice() and caching

The count_hq_slice() function is always used with a SliceArgs struct

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
3 years agovc2enc: fix segfault
Rostislav Pehlivanov [Mon, 21 Mar 2016 11:01:01 +0000 (11:01 +0000)]
vc2enc: fix segfault

Fixes trac bug #5353

Uninitialized memory for the initial quantization index

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
3 years agolavf/matroskaenc.c: add early support for colour elements
Neil Birkbeck [Mon, 7 Mar 2016 22:42:47 +0000 (14:42 -0800)]
lavf/matroskaenc.c: add early support for colour elements

Adding early support for a subset of the proposed colour elements
according to the latest version of spec:
https://mailarchive.ietf.org/arch/search/?email_list=cellar&gbt=1&index=hIKLhMdgTMTEwUTeA4ct38h0tmE

Like matroskadec, I've left out elements for pix_fmt related things
 as there still seems to be some discussion around these.

The new elements are exposed under strict experimental mode.

Signed-off-by: Neil Birkbeck <neil.birkbeck@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/vf_vectorscope: always flip output vertically
Paul B Mahol [Sun, 20 Mar 2016 20:03:43 +0000 (21:03 +0100)]
avfilter/vf_vectorscope: always flip output vertically

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavformat/mov: read start_pad from edit list start time if codec is aac
Marton Balint [Sat, 20 Feb 2016 19:09:42 +0000 (20:09 +0100)]
avformat/mov: read start_pad from edit list start time if codec is aac

Related to ticket #2324, #2325.

Stream duration still need to be fixed...

Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agoavformat/utils: increase detected start_time with skip_samples
Marton Balint [Sat, 20 Feb 2016 21:39:07 +0000 (22:39 +0100)]
avformat/utils: increase detected start_time with skip_samples

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agotests/gapless: add gapless aac tests
Marton Balint [Mon, 7 Mar 2016 21:57:51 +0000 (22:57 +0100)]
tests/gapless: add gapless aac tests

Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agolavf/dv: use c->sys->frame_size in dv_frame_offset()
Clément Bœsch [Sun, 20 Mar 2016 15:14:22 +0000 (16:14 +0100)]
lavf/dv: use c->sys->frame_size in dv_frame_offset()

dv_frame_offset() is static and called only from dv_read_seek(), where
c->sys->frame_size is already used.

This simplifies the incoming codecpar merge where
avctx->{coded_width,coded_height,time_base} are not accessible anymore.

3 years agolavf/dv: reindent after previous commit
Clément Bœsch [Sun, 20 Mar 2016 18:28:07 +0000 (19:28 +0100)]
lavf/dv: reindent after previous commit

3 years agolavf/dv: do not check for c->sys
Clément Bœsch [Sun, 20 Mar 2016 13:24:53 +0000 (14:24 +0100)]
lavf/dv: do not check for c->sys

dv_extract_video_info() is a static function called only from
avpriv_dv_produce_packet(), where c->sys is made sure to be set.

3 years agoffplay: Remove "&& 0" from already disabled debug code
Michael Niedermayer [Fri, 18 Mar 2016 10:27:57 +0000 (11:27 +0100)]
ffplay: Remove "&& 0" from already disabled debug code

Reviewed-by: Marton Balint <cus@passwd.hu>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/vf_waveform: set color range for output frames
Paul B Mahol [Sun, 20 Mar 2016 09:00:53 +0000 (10:00 +0100)]
avfilter/vf_waveform: set color range for output frames

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavformat/utils: Do not wait for more than 1 frame on attachments
Michael Niedermayer [Sat, 19 Mar 2016 21:11:34 +0000 (22:11 +0100)]
avformat/utils: Do not wait for more than 1 frame on attachments

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavformat/wtvdec: Set AVFMTCTX_NOHEADER
Michael Niedermayer [Sat, 19 Mar 2016 21:37:52 +0000 (22:37 +0100)]
avformat/wtvdec: Set AVFMTCTX_NOHEADER

Needed for noStreams.wtv unless something else forces continued parsing (like looking for more than 1
frame in attachments)

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/vf_waveform: add graticule to aflat filter
Paul B Mahol [Sat, 19 Mar 2016 20:47:15 +0000 (21:47 +0100)]
avfilter/vf_waveform: add graticule to aflat filter

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavfilter/vf_waveform: add subsampling input support for remaining filters
Paul B Mahol [Sat, 19 Mar 2016 19:44:59 +0000 (20:44 +0100)]
avfilter/vf_waveform: add subsampling input support for remaining filters

Remove achroma filter, as same output can be done with lowpass filter
and multiple components with overlay display.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavformat/file: Add crypto to default whitelist
Michael Niedermayer [Sat, 19 Mar 2016 17:37:53 +0000 (18:37 +0100)]
avformat/file: Add crypto to default whitelist

Fixes Ticket5287

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/mjpegenc_common: Store approximate aspect if exact cannot be stored
Michael Niedermayer [Sat, 19 Mar 2016 14:41:30 +0000 (15:41 +0100)]
avcodec/mjpegenc_common: Store approximate aspect if exact cannot be stored

Fixes Ticket5244

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/vf_detelecine: Remove redundant declaration
Michael Niedermayer [Sat, 19 Mar 2016 11:26:14 +0000 (12:26 +0100)]
avfilter/vf_detelecine: Remove redundant declaration

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agolavc/hevc: Allow arbitrary garbage in bytestream as long as at least one NAL unit...
Mark Thompson [Thu, 17 Mar 2016 13:41:02 +0000 (13:41 +0000)]
lavc/hevc: Allow arbitrary garbage in bytestream as long as at least one NAL unit is found.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoFix start_frame handling in detelecine filter
Benjamin Steffes [Thu, 17 Mar 2016 22:10:00 +0000 (23:10 +0100)]
Fix start_frame handling in detelecine filter

Signed-off-by: Benjamin Steffes <benjaminst123@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoFix detelecine filter for patterns containing 1
Benjamin Steffes [Thu, 17 Mar 2016 22:09:59 +0000 (23:09 +0100)]
Fix detelecine filter for patterns containing 1

Signed-off-by: Benjamin Steffes <benjaminst123@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agolavf/srtdec: do not be strict wrt timing digit lengths
Clément Bœsch [Fri, 18 Mar 2016 15:41:20 +0000 (16:41 +0100)]
lavf/srtdec: do not be strict wrt timing digit lengths

Fixes a sample with 3-length digits for the seconds reported by wm4.

3 years agolavc/aacenc_utils: replace powf(x,y) by expf(logf(x), y)
Ganesh Ajjanagadde [Tue, 8 Mar 2016 02:16:29 +0000 (21:16 -0500)]
lavc/aacenc_utils: replace powf(x,y) by expf(logf(x), y)

This is ~2x faster for y not an integer on Haswell+GCC, and should
generally be faster due to the fact that anyway powf essentially does
this under the hood. Made an inline function in lavu/internal.h for this
purpose.

Note that there are some accuracy differences, that should generally be
negligible. In particular, FATE still passes on this platform.

Results in ~ 7% speedup in aac encoding with -march=native, Haswell+GCC.
before:
ffmpeg -i sin.flac -acodec aac -y sin_new.aac  6.05s user 0.06s system 104% cpu 5.821 total

after:
ffmpeg -i sin.flac -acodec aac -y sin_new.aac  5.67s user 0.03s system 105% cpu 5.416 total

This is also faster than an alternative approach that pulls in powf, gets rid of
the crufty NaN checks and other special cases, exploits knowledge about the intervals, etc.
This of course does not exclude smarter approaches; just suggests that
there would need to be significant work on this front of lower utility than
searches for hotspots elsewhere.

Reviewed-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanag@gmail.com>