ffmpeg.git
2 years agoavcodec/pngdec: Fix alpha detection with skip_frame
Michael Niedermayer [Fri, 8 Apr 2016 22:47:10 +0000 (00:47 +0200)]
avcodec/pngdec: Fix alpha detection with skip_frame

Fixes Ticket4816

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agoswscale/arm: add ff_hscale_8_to_15_neon
Matthieu Bouron [Thu, 7 Apr 2016 14:52:00 +0000 (14:52 +0000)]
swscale/arm: add ff_hscale_8_to_15_neon

2 years agoavcodec/shorten: fix decoding of files with number of samples lower than max_frame_size
Paul B Mahol [Fri, 8 Apr 2016 21:35:45 +0000 (23:35 +0200)]
avcodec/shorten: fix decoding of files with number of samples lower than max_frame_size

Note that support of very big block sizes is not currently supported at all due
too flawed logic in decoder.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavcodec/shorten: remove useless if condition and comment, reindent
Paul B Mahol [Fri, 8 Apr 2016 20:50:39 +0000 (22:50 +0200)]
avcodec/shorten: remove useless if condition and comment, reindent

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavformat/nistshperedec: add support for mu-law as sample-byte-format
Paul B Mahol [Fri, 8 Apr 2016 20:26:08 +0000 (22:26 +0200)]
avformat/nistshperedec: add support for mu-law as sample-byte-format

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavcodec/shorten: fix decoding of last frame
Paul B Mahol [Fri, 8 Apr 2016 08:35:11 +0000 (10:35 +0200)]
avcodec/shorten: fix decoding of last frame

Previously it would be always discarded.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agofate: Add test for Ticket 2397 (dvdsub)
Michael Niedermayer [Sat, 2 Apr 2016 00:36:41 +0000 (02:36 +0200)]
fate: Add test for Ticket 2397 (dvdsub)

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agosws/aarch64/yuv2rgb: honor iOS calling convention
Clément Bœsch [Fri, 8 Apr 2016 11:39:22 +0000 (13:39 +0200)]
sws/aarch64/yuv2rgb: honor iOS calling convention

y_offset and y_coeff being successive 32-bit integers, they are packed
into 8 bytes instead of 2x8 bytes.

See https://developer.apple.com/library/ios/documentation/Xcode/Conceptual/iPhoneOSABIReference/Articles/ARM64FunctionCallingConventions.html

> iOS diverges from Procedure Call Standard for the ARM 64-bit
> Architecture in several ways
[...]
> In the generic procedure call standard, all function arguments passed
> on the stack consume slots in multiples of 8 bytes. In iOS, this
> requirement is dropped, and values consume only the space required.
[...]
> Padding is still inserted on the stack to satisfy arguments’ alignment
> requirements.

2 years agoavformat/nistspheredec: fix silly bug
Paul B Mahol [Fri, 8 Apr 2016 11:19:00 +0000 (13:19 +0200)]
avformat/nistspheredec: fix silly bug

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavcodec/pngenc: check return value of av_frame_copy()
Paul B Mahol [Fri, 8 Apr 2016 10:45:40 +0000 (12:45 +0200)]
avcodec/pngenc: check return value of av_frame_copy()

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavcodec/pngenc: restore image size before copying frame
Dmitriy [Fri, 8 Apr 2016 10:37:20 +0000 (12:37 +0200)]
avcodec/pngenc: restore image size before copying frame

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavcodec/shorten: if allocation fails reset max_frame_size
Paul B Mahol [Fri, 8 Apr 2016 07:59:05 +0000 (09:59 +0200)]
avcodec/shorten: if allocation fails reset max_frame_size

Otherwise crash happens.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavcodec: remove "get_buffer() failed" message
Paul B Mahol [Fri, 8 Apr 2016 07:40:32 +0000 (09:40 +0200)]
avcodec: remove "get_buffer() failed" message

It is already provided by ff_get_buffer().

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavformat: support shorten in nistshpere demuxer
Paul B Mahol [Wed, 6 Apr 2016 19:57:52 +0000 (21:57 +0200)]
avformat: support shorten in nistshpere demuxer

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agodoc/filters: add drawtext example
Mulvya [Sun, 3 Apr 2016 08:57:01 +0000 (14:27 +0530)]
doc/filters: add drawtext example

Show example which draws text at a random position, and switches positions every 30 seconds

Signed-off-by: Mulvya <mulvya@gmail.com>
Signed-off-by: Lou Logan <lou@lrcd.com>
2 years agoavfilter: add remap filter
F.Sluiter [Tue, 5 Apr 2016 07:36:37 +0000 (09:36 +0200)]
avfilter: add remap filter

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoswscale/utils: Remove unused variable
Michael Niedermayer [Wed, 6 Apr 2016 09:50:01 +0000 (11:50 +0200)]
swscale/utils: Remove unused variable

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agoavformat: add aix demuxer
Paul B Mahol [Tue, 5 Apr 2016 12:05:10 +0000 (14:05 +0200)]
avformat: add aix demuxer

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavcodec/exr: enable mipmap, ripmap decoding
Martin Vignali [Tue, 5 Apr 2016 19:17:51 +0000 (21:17 +0200)]
avcodec/exr: enable mipmap, ripmap decoding

2 years agofate: add missing filter-meta-4560-rotate0 dependencies
James Almer [Wed, 6 Apr 2016 00:17:13 +0000 (21:17 -0300)]
fate: add missing filter-meta-4560-rotate0 dependencies

Signed-off-by: James Almer <jamrial@gmail.com>
2 years agoAAC encoder: fix valgrind errors
Claudio Freire [Wed, 6 Apr 2016 02:13:44 +0000 (23:13 -0300)]
AAC encoder: fix valgrind errors

Move wi.clipping computation outside of psy_lame_window, LFE
channels don't even call that, and make the LFE path also
initialize window_type[1] which is needed by analyze_channel

2 years agofate: force fixed point aac decoder in filter-meta-4560-rotate0
Michael Niedermayer [Wed, 6 Apr 2016 01:05:36 +0000 (03:05 +0200)]
fate: force fixed point aac decoder in filter-meta-4560-rotate0

This should fix a off by 1 difference between AARCH64 and X86

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agolibxvid: Create extradata in init using a dummy frame
Derek Buitenhuis [Sun, 3 Apr 2016 16:55:25 +0000 (12:55 -0400)]
libxvid: Create extradata in init using a dummy frame

Modifying global header extradata in encode_frame is an API violation
and only happens to work currently because mov writes its header
at the end of the file.

Heavily based off of a patch from 2012.

Original-by: Nicolas George <george@nsup.org>
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 years agosws/aarch64: restore ff_hscale_8_to_15_neon()
Clément Bœsch [Tue, 5 Apr 2016 08:16:32 +0000 (10:16 +0200)]
sws/aarch64: restore ff_hscale_8_to_15_neon()

Fix final scaling and required filter alignment. Pass FATE.

2 years agoavformat: add musx demuxer
Paul B Mahol [Sun, 3 Apr 2016 10:10:09 +0000 (12:10 +0200)]
avformat: add musx demuxer

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoavcodec: add adpcm dat4 decoder
Paul B Mahol [Sun, 3 Apr 2016 10:55:17 +0000 (12:55 +0200)]
avcodec: add adpcm dat4 decoder

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agoAAC encoder: fix filling of wi.clipping array
Claudio Freire [Tue, 5 Apr 2016 07:03:05 +0000 (04:03 -0300)]
AAC encoder: fix filling of wi.clipping array

Fill all windows in all window groups instead of only the first
window of each group.

Also avoid uninitialized access of window_type.

2 years agoffplay: fix silence insertion on error or pause
Marton Balint [Sun, 3 Apr 2016 19:23:41 +0000 (21:23 +0200)]
ffplay: fix silence insertion on error or pause

Insertion of silence was a bit broken since
df34b700981de606ca4847e1ed0bfdf9ac3e9104 because the info whether or not the
source buffer supposed to be silence must be kept between callbacks. Failing to
do so causes rogue samples from the last buffer to be presented, I guess even a
crash can occur under some circumstances.

This patch uses a NULL audio_buf to keep the silence state across audio
callbacks.

Reviewed-by: Lukasz Marek <lukasz.m.luki2 at gmail.com>
Signed-off-by: Marton Balint <cus@passwd.hu>
2 years agolibavcodec/exr : add support for compression in tile
Martin Vignali [Mon, 4 Apr 2016 20:00:31 +0000 (22:00 +0200)]
libavcodec/exr : add support for compression in tile

2 years agolibavcodec/exr : cosmetic change
Martin Vignali [Mon, 4 Apr 2016 15:23:42 +0000 (17:23 +0200)]
libavcodec/exr : cosmetic change

2 years agofate: Add wav-ac3 test
Michael Niedermayer [Sat, 2 Apr 2016 11:45:41 +0000 (13:45 +0200)]
fate: Add wav-ac3 test

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agofate: add fate-flv-demux
Michael Niedermayer [Sat, 2 Apr 2016 20:02:52 +0000 (22:02 +0200)]
fate: add fate-flv-demux

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agofate: Add fate-ts-opus-demux
Michael Niedermayer [Sat, 2 Apr 2016 20:00:41 +0000 (22:00 +0200)]
fate: Add fate-ts-opus-demux

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agofate: add demux test for OggOpus
James Almer [Sat, 2 Apr 2016 20:31:36 +0000 (17:31 -0300)]
fate: add demux test for OggOpus

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>
2 years agoAAC encoder: new regression test
Claudio Freire [Sun, 3 Apr 2016 18:28:17 +0000 (15:28 -0300)]
AAC encoder: new regression test

Test to catch the recently fixed minsf bug

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agoAAC encoder: fix initialization of minsf
Claudio Freire [Sun, 3 Apr 2016 18:03:53 +0000 (15:03 -0300)]
AAC encoder: fix initialization of minsf

In some situations (exactly zeroed DC coeffs) minsf would
be initialized with garbage

2 years agoavformat/brstm: lower magic number, fixes decoding of some files
Paul B Mahol [Sun, 3 Apr 2016 17:12:07 +0000 (19:12 +0200)]
avformat/brstm: lower magic number, fixes decoding of some files

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agolavc/utils: Introduce ff_bprint_to_codecpar_extradata for avformat
Hendrik Leppkes [Thu, 31 Mar 2016 20:49:02 +0000 (21:49 +0100)]
lavc/utils: Introduce ff_bprint_to_codecpar_extradata for avformat

It will be used by text subtitle demuxers to construct format instructions
straight into extradata. They all currently a similar function that accepts
an AVCodecContext instead.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 years agoavformat/wavdec: fix typo with len
Michael Niedermayer [Sun, 3 Apr 2016 15:34:15 +0000 (17:34 +0200)]
avformat/wavdec: fix typo with len

Found-by: carl
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agolibavcodec/exr: add tile support
Martin Vignali [Sat, 2 Apr 2016 22:21:30 +0000 (00:21 +0200)]
libavcodec/exr: add tile support

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agolibavcodec/exr: fix PRX24 Float decompression
Martin Vignali [Sun, 3 Apr 2016 12:12:44 +0000 (14:12 +0200)]
libavcodec/exr: fix PRX24 Float decompression

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2 years agofate/exr : add test for b44/b44a compression
Martin Vignali [Sat, 26 Mar 2016 20:02:07 +0000 (21:02 +0100)]
fate/exr : add test for b44/b44a compression

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agoavformat/wavdec: Remove direct s->pb->buffer access
Michael Niedermayer [Sun, 3 Apr 2016 00:15:06 +0000 (02:15 +0200)]
avformat/wavdec: Remove direct s->pb->buffer access

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agowavdec: Only set the codec ID in read_header
Derek Buitenhuis [Sat, 2 Apr 2016 21:43:17 +0000 (22:43 +0100)]
wavdec: Only set the codec ID in read_header

WAV is not a NOHEADER format, and thus should not be changing
stream codec IDs and probing in read_packet.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agoavcodec/mjpegdec: Do not permute quantization tables
Michael Niedermayer [Sat, 2 Apr 2016 23:28:07 +0000 (01:28 +0200)]
avcodec/mjpegdec: Do not permute quantization tables

This fixes issues if the permutation changes, as quantizations tables would need to be reread

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agoavfilter/src_movie: fix how we check for overflows with seek_point
Marios Titas [Sat, 2 Apr 2016 18:11:44 +0000 (21:11 +0300)]
avfilter/src_movie: fix how we check for overflows with seek_point

Currently, if the movie source filter is used and a seek_point is
specified on a file that has a negative start time, ffmpeg will fail.

An easy way to reproduce this is as follows:
$ ffmpeg -vsync passthrough -filter_complex 'color=d=10,setpts=PTS-1/TB' test.mp4
$ ffmpeg -filter_complex 'movie=filename=test.mp4:seek_point=2' -f null -

The problem is caused by checking for int64_t overflow the wrong way.
In general, to check whether a + b overflows, it is not enough to do:
    a > INT64_MAX - b
because b might be negative; the correct way is:
    b > 0 && > a > INT64_MAX - b

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agoexif: take a generic log context
Hendrik Leppkes [Fri, 1 Apr 2016 11:13:03 +0000 (13:13 +0200)]
exif: take a generic log context

The AVCodecContext is only used for logging, so instead take any valid log context.
This allows reusing the exif functions more easily in avformat.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 years agoavcodec: properly initialize AVCodecParameters profile/level
Hendrik Leppkes [Fri, 1 Apr 2016 11:13:02 +0000 (13:13 +0200)]
avcodec: properly initialize AVCodecParameters profile/level

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 years agolavf/segment: add option to write empty filler segments as needed
Rodger Combs [Mon, 7 Mar 2016 02:48:57 +0000 (20:48 -0600)]
lavf/segment: add option to write empty filler segments as needed

2 years agolavf/segment: slight refactor to seg_write_packet
Rodger Combs [Mon, 7 Mar 2016 02:43:37 +0000 (20:43 -0600)]
lavf/segment: slight refactor to seg_write_packet

This reduces some code duplication, and ensures that cur_entry.last_duration is
always set.

2 years agolavf/segment: style nit
Rodger Combs [Mon, 7 Mar 2016 01:38:39 +0000 (19:38 -0600)]
lavf/segment: style nit

2 years agolavc/videotoolboxenc: Workaround encoder error
Rick Kern [Thu, 24 Mar 2016 17:56:04 +0000 (01:56 +0800)]
lavc/videotoolboxenc: Workaround encoder error

CMVideoFormatDescriptionGetH264ParameterSetAtIndex() fails on some
hardware/OS versions when retrieving the parameter set count alone.

Signed-off-by: Rick Kern <kernrj@gmail.com>
Signed-off-by: wm4 <nfxjfg@googlemail.com>
2 years agolavc/videotoolboxenc: Fix crash when closing codec after error
Rick Kern [Sun, 20 Mar 2016 16:55:56 +0000 (00:55 +0800)]
lavc/videotoolboxenc: Fix crash when closing codec after error

Fixes crash in #5352. VTCompressionSessionInvalidate() crashes if the
internal encoder hasn't completed, but hasn't experienced an error.
The function call isn't needed since the encoder is invalidated when
the reference count reaches 0 anyway.

Signed-off-by: Rick Kern <kernrj@gmail.com>
Signed-off-by: wm4 <nfxjfg@googlemail.com>
2 years agolavc/utils: use pkt_timebase for legacy subtitles timing code
Clément Bœsch [Mon, 28 Mar 2016 12:15:42 +0000 (14:15 +0200)]
lavc/utils: use pkt_timebase for legacy subtitles timing code

This is consistent with other AVSubtitle timing adjustments.

2 years agolavc/ccaption_dec: remove usage of avctx->time_base
Clément Bœsch [Mon, 28 Mar 2016 12:10:08 +0000 (14:10 +0200)]
lavc/ccaption_dec: remove usage of avctx->time_base

lavc/utils already rescales avpkt->pts to sub->pts in AV_TIME_BASE_Q
before calling the decode callback. This prevents from rescaling again
into the decoder, and avoid the use of avctx->time_base which will
disappear in the incoming codecpar merge.

This commit also replaces the use of "20 centisecond" (ass time base)
with "200 ms".

2 years agolavu/base64: add AV_BASE64_DECODE_SIZE() macro
Stefano Sabatini [Sun, 20 Mar 2016 14:59:47 +0000 (15:59 +0100)]
lavu/base64: add AV_BASE64_DECODE_SIZE() macro

This is consistent with the AV_BASE64_SIZE macro and avoids the literal
use of constants in the code.

2 years agolavc/audiotoolboxdec: add eac3 decoder
Rodger Combs [Thu, 31 Mar 2016 01:55:32 +0000 (20:55 -0500)]
lavc/audiotoolboxdec: add eac3 decoder

This is added in 10.11, so we add a #define when building against older SDKs.

The decoder actually supports 7.1-channel eac3, but since the parser only
reports 6 channels, we end up decoding the 5.1 downmix (same as the internal
decoder) for now.

2 years agolavc/audiotoolboxdec: fix a number of config and timestamp issues
Rodger Combs [Thu, 24 Mar 2016 05:49:51 +0000 (00:49 -0500)]
lavc/audiotoolboxdec: fix a number of config and timestamp issues

- ADTS-formatted AAC didn't work
- Channel layouts were never exported
- Channel mappings were incorrect beyond stereo
- Channel counts weren't updated after packets were decoded
- Timestamps were exported incorrectly

2 years agolavc/audiotoolboxdec: support ADTS AAC input
Rodger Combs [Thu, 24 Mar 2016 02:50:41 +0000 (21:50 -0500)]
lavc/audiotoolboxdec: support ADTS AAC input

2 years agolavc/audiotoolboxenc: allow setting maxrate with pre-10.9 deployment targets
Rodger Combs [Sun, 27 Mar 2016 17:17:25 +0000 (12:17 -0500)]
lavc/audiotoolboxenc: allow setting maxrate with pre-10.9 deployment targets

The build failure here is caused by the enum value not being defined, but
as long as we're on a newer SDK that has it, it's safe to use it even
when our deployment target is older. Setting the property will error, but
we're not failing on errors there.

2 years agolavc/audiotoolboxenc: fix iOS build
Rodger Combs [Thu, 24 Mar 2016 15:17:42 +0000 (10:17 -0500)]
lavc/audiotoolboxenc: fix iOS build

2 years agolavc/audiotoolboxenc: fix a number of config issues
Rodger Combs [Wed, 23 Mar 2016 21:46:39 +0000 (16:46 -0500)]
lavc/audiotoolboxenc: fix a number of config issues

- size variables were used in a confusing way
- incorrect size var use led to channel layouts not being set properly
- channel layouts were incorrectly mapped for >2-channel AAC
- bitrates not accepted by the encoder were discarded instead of being clamped
- some minor style/indentation fixes

2 years agolavc/audiotoolboxenc: remove unneeded packet metadata
Rodger Combs [Wed, 23 Mar 2016 21:29:50 +0000 (16:29 -0500)]
lavc/audiotoolboxenc: remove unneeded packet metadata

This isn't necessary here, and for some reason broke only multichannel
AAC encoding when a channel layout tag was set.

2 years agolavf/segment: support automatic bitstream filtering
Rodger Combs [Wed, 23 Mar 2016 14:24:22 +0000 (09:24 -0500)]
lavf/segment: support automatic bitstream filtering

Most useful for MPEG-TS. Works by having the underlying muxer configure the
bitstream filters, then moving them to our own AVStreams.

2 years agoswscale: add bt2020 yuv2rgb coefficients.
Ronald S. Bultje [Thu, 31 Mar 2016 20:01:09 +0000 (16:01 -0400)]
swscale: add bt2020 yuv2rgb coefficients.

2 years agoswscale: fix bt709 yuv2rgb coefficients.
Ronald S. Bultje [Thu, 31 Mar 2016 20:00:58 +0000 (16:00 -0400)]
swscale: fix bt709 yuv2rgb coefficients.

They were derived from incorrect values cr=0.2125 (should be cr=0.2126)
and cb=0.0721 (should be cb=0.0722).

2 years agofate: add demux test for TS with AC3 (Ticket 4864)
Michael Niedermayer [Wed, 30 Mar 2016 19:29:01 +0000 (21:29 +0200)]
fate: add demux test for TS with AC3 (Ticket 4864)

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agofate: Test for Ticket4560
Michael Niedermayer [Thu, 31 Mar 2016 22:57:14 +0000 (00:57 +0200)]
fate: Test for Ticket4560

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agofate: Add regression test for Ticket 1239
Michael Niedermayer [Thu, 31 Mar 2016 22:26:46 +0000 (00:26 +0200)]
fate: Add regression test for Ticket 1239

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agoavcodec/parser: assert that the codec id is not NONE in av_parser_parse2()
Michael Niedermayer [Fri, 1 Apr 2016 16:50:45 +0000 (18:50 +0200)]
avcodec/parser: assert that the codec id is not NONE in av_parser_parse2()

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agoavcodec/j2kenc: Add attribution to OpenJPEG project:
Aaron Boxer [Thu, 31 Mar 2016 20:02:14 +0000 (16:02 -0400)]
avcodec/j2kenc: Add attribution to OpenJPEG project:

http://ghostscript.com/~tor/gs-browse/gs/openjpeg/libopenjpeg/t1.c

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2 years agosws/aarch64: disable ff_hscale_8_to_15_neon temporarly
Clément Bœsch [Fri, 1 Apr 2016 15:32:16 +0000 (17:32 +0200)]
sws/aarch64: disable ff_hscale_8_to_15_neon temporarly

Looks broken.

2 years agoswscale/arm/yuv2rgb: make the code bitexact with its aarch64 counter part
Matthieu Bouron [Mon, 28 Mar 2016 14:17:58 +0000 (14:17 +0000)]
swscale/arm/yuv2rgb: make the code bitexact with its aarch64 counter part

2 years agoswscale/arm/yuv2rgb: save a few instructions by processing the luma line interleaved
Matthieu Bouron [Mon, 28 Mar 2016 11:33:38 +0000 (11:33 +0000)]
swscale/arm/yuv2rgb: save a few instructions by processing the luma line interleaved

2 years agoswscale/arm/yuv2rgb: re-order compute_rgba macro arguments
Matthieu Bouron [Mon, 28 Mar 2016 12:24:44 +0000 (12:24 +0000)]
swscale/arm/yuv2rgb: re-order compute_rgba macro arguments

2 years agoswscale/arm/yuv2rgb: rename {process_1l,process_2l,compute}_16px{,_internal} macros
Matthieu Bouron [Thu, 31 Mar 2016 12:01:43 +0000 (14:01 +0200)]
swscale/arm/yuv2rgb: rename {process_1l,process_2l,compute}_16px{,_internal} macros

2 years agoswscale/arm/yuv2rgb: macro-ify
Matthieu Bouron [Sun, 27 Mar 2016 18:20:36 +0000 (18:20 +0000)]
swscale/arm/yuv2rgb: macro-ify

2 years agoswscale/arm/yuv2rgb: factorize lsl in load_args_nvx
Matthieu Bouron [Thu, 24 Mar 2016 16:33:28 +0000 (17:33 +0100)]
swscale/arm/yuv2rgb: factorize lsl in load_args_nvx

2 years agoswscale/arm/yuv2rgb: factorize lsl in load_args_yuv420p
Matthieu Bouron [Wed, 23 Mar 2016 13:00:33 +0000 (13:00 +0000)]
swscale/arm/yuv2rgb: factorize lsl in load_args_yuv420p

2 years agoswscale/arm/yuv2rgb: remove unused store of dst + linesize in load_args_yuv422p
Matthieu Bouron [Thu, 24 Mar 2016 16:29:39 +0000 (17:29 +0100)]
swscale/arm/yuv2rgb: remove unused store of dst + linesize in load_args_yuv422p

2 years agoswscale/arm/yuv2rgb: factorize lsl in load_args_yuv422p
Matthieu Bouron [Thu, 31 Mar 2016 11:51:21 +0000 (13:51 +0200)]
swscale/arm/yuv2rgb: factorize lsl in load_args_yuv422p

2 years agoswscale/arm/yuv2rgb: fix comments in load_args_yuv422p
Matthieu Bouron [Thu, 31 Mar 2016 11:50:29 +0000 (13:50 +0200)]
swscale/arm/yuv2rgb: fix comments in load_args_yuv422p

2 years agoswscale/arm/yuv2rgb: remove 32bit code path
Matthieu Bouron [Tue, 22 Mar 2016 16:46:10 +0000 (17:46 +0100)]
swscale/arm/yuv2rgb: remove 32bit code path

2 years agolavc: factor apply_param_change() AV_EF_EXPLODE handling
wm4 [Fri, 1 Apr 2016 10:52:56 +0000 (12:52 +0200)]
lavc: factor apply_param_change() AV_EF_EXPLODE handling

Remove the duplicated code for handling failure of apply_param_change().

2 years agolavu: improve documentation of some AVFrame functions
wm4 [Fri, 1 Apr 2016 10:52:12 +0000 (12:52 +0200)]
lavu: improve documentation of some AVFrame functions

2 years agopsxstr: Remove some commented out code
Derek Buitenhuis [Thu, 31 Mar 2016 17:37:21 +0000 (18:37 +0100)]
psxstr: Remove some commented out code

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 years agolavc/utils: transfer width/height for subs in codecpar
Clément Bœsch [Mon, 28 Mar 2016 17:02:02 +0000 (19:02 +0200)]
lavc/utils: transfer width/height for subs in codecpar

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 years agocodecpar: Add video delay field
Derek Buitenhuis [Wed, 30 Mar 2016 16:15:04 +0000 (17:15 +0100)]
codecpar: Add video delay field

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 years agoAdd frame_size to AVCodecParameters
Hendrik Leppkes [Sat, 5 Mar 2016 10:22:11 +0000 (11:22 +0100)]
Add frame_size to AVCodecParameters

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 years agolavc: Add seek_preroll to AVCodecParameters
James Almer [Fri, 4 Mar 2016 18:51:41 +0000 (15:51 -0300)]
lavc: Add seek_preroll to AVCodecParameters

Signed-off-by: James Almer <jamrial@gmail.com>
2 years agoMerge commit 'a8068346e48e123f8d3bdf4d64464d81e53e5fc7'
Derek Buitenhuis [Thu, 31 Mar 2016 20:19:49 +0000 (21:19 +0100)]
Merge commit 'a8068346e48e123f8d3bdf4d64464d81e53e5fc7'

* commit 'a8068346e48e123f8d3bdf4d64464d81e53e5fc7':
  lavc: add a variant of av_get_audio_frame_duration working with AVCodecParameters

Fixes from jamrial incorporated.

Merged-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 years agoMerge commit '998e1b8f521b73e1ed3a13caaabcf79eb401cf0d'
Derek Buitenhuis [Thu, 31 Mar 2016 20:18:53 +0000 (21:18 +0100)]
Merge commit '998e1b8f521b73e1ed3a13caaabcf79eb401cf0d'

* commit '998e1b8f521b73e1ed3a13caaabcf79eb401cf0d':
  lavc: add codec parameters API

  Fixes added in:
    - bit_rate has been made int64_t to match.
    - profile and level are properly initialize.

Merged-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2 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

2 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>
2 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>
2 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>
2 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>
2 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

2 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)

2 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>