ffmpeg.git
3 years agoMerge commit 'b22693b06d1e5d73454a65c203b4d31c1ca5b69a'
Hendrik Leppkes [Sat, 10 Oct 2015 07:45:16 +0000 (09:45 +0200)]
Merge commit 'b22693b06d1e5d73454a65c203b4d31c1ca5b69a'

* commit 'b22693b06d1e5d73454a65c203b4d31c1ca5b69a':
  w32pthreads: Add pthread_once emulation

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit '8b830ee9a26d47b138f12a82085cdb372f407f1e'
Hendrik Leppkes [Sat, 10 Oct 2015 07:43:47 +0000 (09:43 +0200)]
Merge commit '8b830ee9a26d47b138f12a82085cdb372f407f1e'

* commit '8b830ee9a26d47b138f12a82085cdb372f407f1e':
  avconv: Do not try to configure filter outputs without streams

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit 'd7a5a178c252b625537adc046392624ad543dea7'
Hendrik Leppkes [Sat, 10 Oct 2015 07:37:51 +0000 (09:37 +0200)]
Merge commit 'd7a5a178c252b625537adc046392624ad543dea7'

* commit 'd7a5a178c252b625537adc046392624ad543dea7':
  configure: When disabling a library disable all the related components

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit '58b42345b38b46d11c32e11d9c57517f99d6a601'
Hendrik Leppkes [Sat, 10 Oct 2015 07:32:59 +0000 (09:32 +0200)]
Merge commit '58b42345b38b46d11c32e11d9c57517f99d6a601'

* commit '58b42345b38b46d11c32e11d9c57517f99d6a601':
  dcadec: reorganise context data

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agoMerge commit '3a4d369ea4ded91b1178ae6e2ff0ab9ea470e344'
Hendrik Leppkes [Sat, 10 Oct 2015 07:13:32 +0000 (09:13 +0200)]
Merge commit '3a4d369ea4ded91b1178ae6e2ff0ab9ea470e344'

* commit '3a4d369ea4ded91b1178ae6e2ff0ab9ea470e344':
  g2m: Relax resolution change constraints

Merged-by: Hendrik Leppkes <h.leppkes@gmail.com>
3 years agox86/takdsp: use arithmetic shift instructions
James Almer [Sat, 10 Oct 2015 02:17:31 +0000 (23:17 -0300)]
x86/takdsp: use arithmetic shift instructions

p1 and p2 are int32_t.

Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
3 years agolavf/mov: add support for sidx fragment indexes
Rodger Combs [Mon, 20 Jul 2015 20:00:35 +0000 (15:00 -0500)]
lavf/mov: add support for sidx fragment indexes

Fixes trac #3842

3 years agoh264_mp4toannexb: fix the pps offset when there are more than one sps in avcc
赵宇龙 [Fri, 9 Oct 2015 05:59:56 +0000 (13:59 +0800)]
h264_mp4toannexb: fix the pps offset when there are more than one sps in avcc

the pps offset is used to locate pps in the spspps_buf; however, the
current calc method is wrong because it is the offset of the original
avctx->extradata;
when there is only one sps in the avcc; the value is correct by
coincidence, however, it will fail in avcc with multi sps

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/h264_mp4toannexb_bsf: Use av_freep() to free spspps_buf
Michael Niedermayer [Fri, 9 Oct 2015 22:07:30 +0000 (00:07 +0200)]
avcodec/h264_mp4toannexb_bsf: Use av_freep() to free spspps_buf

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/xvmc: apply attribute_deprecated correctly
Ganesh Ajjanagadde [Fri, 9 Oct 2015 22:44:49 +0000 (18:44 -0400)]
avcodec/xvmc: apply attribute_deprecated correctly

This fixes a warning observed on Clang 3.7:
"warning: attribute 'deprecated' is ignored, place it after "struct" to apply attribute to type declaration [-Wignored-attributes]"
and thus enables deprecation warning for the relevant struct.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
3 years agoavcodec/mpegvideoenc: fix undefined negative left shift
Ganesh Ajjanagadde [Fri, 9 Oct 2015 18:43:04 +0000 (14:43 -0400)]
avcodec/mpegvideoenc: fix undefined negative left shift

This should fix the first undefined behavior reported in:
https://trac.ffmpeg.org/ticket/4727.

I can't reproduce the runtime behavior reported in the ticket, hence I
can't confirm that this actually fixes the exact issue reported in the
ticket.

Regardless, I can confirm that this is a genuine issue, and that
negative shifts can (and do) occur, fixed by this.

Tested with FATE.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agofate: add DNxHD/HR tests
Christophe Gisquet [Mon, 5 Oct 2015 15:37:38 +0000 (17:37 +0200)]
fate: add DNxHD/HR tests

Currently only 2 profiles are evaluated because they are the only 2
with distributed test sequences.
- CID 1260: YUV 4:2:2 10 bits with block-adaptive interlace coding,
  from ticket 4876;
- CID 1270: YUV 4:4:4 10 bits (HR), 1920x839, from ticket 4581.

They were generated from the ticket sequences by running the
following kind of command-line;
ffmpeg -i $INPUT -an -sn -vcodec copy -vframes 1 -y $OUTPUT.mov

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodoc: fix spelling errors
Andreas Cadhalpun [Tue, 29 Sep 2015 18:08:26 +0000 (20:08 +0200)]
doc: fix spelling errors

Reviewed-by: Lou Logan <lou@lrcd.com>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
3 years agoavcodec/takdec: add x86 SIMD for rest of decorrelation modes
Paul B Mahol [Mon, 5 Oct 2015 10:28:41 +0000 (12:28 +0200)]
avcodec/takdec: add x86 SIMD for rest of decorrelation modes

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoffmpeg: avoid possible undefined behavior
Ganesh Ajjanagadde [Fri, 9 Oct 2015 14:39:27 +0000 (10:39 -0400)]
ffmpeg: avoid possible undefined behavior

On lines 1633,1634 FFABS(pts) is performed. However, if av_stream_get_end_pts
returns AV_NOPTS_VALUE always, pts remains stuck at INT64_MIN, leading
to undefined behavior on FFABS.

One could conceive of a solution using FFNABS. However, such a solution
has to deal with the implementation defined rounding of integer division
with at least one negative operand in ANSI C89. C99 forces truncation to
zero, but I am not sure that all of our platforms compile with full C99
support, and in particular whether we can safely assume a fixed
rounding behavior across all platforms.

This solution is simple, and I doubt changing INT64_MIN to INT64_MIN + 1
has any practical loss - if it is stuck at its initial value, the stream
is messed up anyway.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agolavc/mjpegenc: Add an option to force outputting chroma matrix.
Carl Eugen Hoyos [Fri, 9 Oct 2015 18:58:07 +0000 (20:58 +0200)]
lavc/mjpegenc: Add an option to force outputting chroma matrix.

RFC 2435 suggests that mjpeg over rtp uses both two tables.

3 years agoavcodec/pthread_frame: update doxygen for update_context_from_thread
Ganesh Ajjanagadde [Fri, 9 Oct 2015 13:27:47 +0000 (09:27 -0400)]
avcodec/pthread_frame: update doxygen for update_context_from_thread

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
3 years agobuild system: fix dependency generation
Christophe Gisquet [Thu, 8 Oct 2015 10:30:51 +0000 (12:30 +0200)]
build system: fix dependency generation

When trying to build the binary dct-test under MSYS2/Win64, the
makefile rule does not have the SUBDIR in the target for its
object file.

Consequently, modifications to various include files (e.g. C ones)
do not trigger a recompilation.

When tracing the dependency generating, the dependency generation
has this strange content (linebreak inserted):
sed -e "/^#.*/d" -e "s,^[[:space:]]*dct\\.o,libavcodec/dct.o," \
 > libavcodec/dct-test.d

For some reason, the $(*F) has weird content. It looks simpler to
use $(@F) instead of $(*F)\\.o, although this was tested on one
single version of make.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/vf_blend: fix normal mode with opacity != 1
Paul B Mahol [Fri, 9 Oct 2015 10:12:03 +0000 (12:12 +0200)]
avfilter/vf_blend: fix normal mode with opacity != 1

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agolavf/hls: don't convert NULL options to empty strings; fixes HTTP CRLF warnings
Rodger Combs [Mon, 21 Sep 2015 02:50:22 +0000 (21:50 -0500)]
lavf/hls: don't convert NULL options to empty strings; fixes HTTP CRLF warnings

3 years agolavu/opt: add flag to return NULL when applicable in av_opt_get
Rodger Combs [Mon, 21 Sep 2015 09:16:58 +0000 (04:16 -0500)]
lavu/opt: add flag to return NULL when applicable in av_opt_get

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agolavu/opt: switch AV_OPT flags to shift-based formatting
Rodger Combs [Mon, 21 Sep 2015 09:15:56 +0000 (04:15 -0500)]
lavu/opt: switch AV_OPT flags to shift-based formatting

3 years agobuild: restore videotoolbox compilation on iOS
Clément Bœsch [Thu, 1 Oct 2015 15:58:40 +0000 (17:58 +0200)]
build: restore videotoolbox compilation on iOS

3 years agoavformat/srtdec: make sure we probe a number
Clément Bœsch [Fri, 9 Oct 2015 08:52:58 +0000 (10:52 +0200)]
avformat/srtdec: make sure we probe a number

Fixes regression since 7218352e0228028dfa009a3799ec93fd041065f1: WebVTT
files were matching the SRT probing.

3 years agolavf/matroskadec: drop indexes that appear broken
Rodger Combs [Thu, 8 Oct 2015 20:34:59 +0000 (15:34 -0500)]
lavf/matroskadec: drop indexes that appear broken

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/ass: fix doxygen typo
wm4 [Thu, 8 Oct 2015 21:28:36 +0000 (23:28 +0200)]
avcodec/ass: fix doxygen typo

3 years agodoc/faq: use https instead of http
Ganesh Ajjanagadde [Thu, 8 Oct 2015 01:40:53 +0000 (21:40 -0400)]
doc/faq: use https instead of http

Change to https for FFmpeg websites.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/af_afade: fix start of fade out
Justin Greer [Thu, 8 Oct 2015 20:26:20 +0000 (15:26 -0500)]
avfilter/af_afade: fix start of fade out

Fixes #4919

3 years agoconfigure: add message to avoid manual modification of config.texi
Ganesh Ajjanagadde [Thu, 8 Oct 2015 13:01:20 +0000 (09:01 -0400)]
configure: add message to avoid manual modification of config.texi

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agodoc/writing_filters: miscellaneous grammar and typo fixes
Ganesh Ajjanagadde [Thu, 8 Oct 2015 13:00:44 +0000 (09:00 -0400)]
doc/writing_filters: miscellaneous grammar and typo fixes

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agotests: update fate-filter-delogo test reference after commit 8bc708fcee137ead6d0773fa...
Stefano Sabatini [Thu, 8 Oct 2015 10:44:48 +0000 (12:44 +0200)]
tests: update fate-filter-delogo test reference after commit 8bc708fcee137ead6d0773fad8e24ab471ab2122

3 years agoavfilter/delogo: Set default band to 1
Jean Delvare [Wed, 7 Oct 2015 13:03:32 +0000 (15:03 +0200)]
avfilter/delogo: Set default band to 1

The original interpolation algorithm behaved poorly on the borders and
did not even guarantee continuity at the borders. For this reason, a
second interpolation/blending pass was required on the borders to make
them seamless.

However, since the interpolation algorithm was improved in June 2013,
the border issues no longer exist. The new algorithm does guarantee
continuity at the borders, making the second pass useless. A larger
band always increases the cumulated interpolation error. In most cases
it also increases the average interpolation error, even though the
samples in the band are only partially interpolated.

For this reason I would like to get rid of the "band" parameter. As a
first step, let's change its default value from 4 to 1 and document it
as deprecated.

I have benchmarked this change on a combination of input sources and
realistic logo areas. Lowering the band value from 4 to 1 resulted in
8 to 39 % less interpolation error per frame (or 1 to 34 % less
interpolation error per luma sample.)

Signed-off-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: Stefano Sabatini <stefasab@gmail.com>
3 years agodnxhdenc: mark CID 1260 encoder experimental
Christophe Gisquet [Mon, 5 Oct 2015 18:44:46 +0000 (20:44 +0200)]
dnxhdenc: mark CID 1260 encoder experimental

The MBAFF handling recently introduced on the decoder side shows that
the encoder does not support it correctly. Therefore, make the related
profile experimental.

Furthermore, current encoder logic treats it as unable to encode as
progressive, which isn't the case.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodnxhddata: introduce and use MBAFF flag
Christophe Gisquet [Mon, 5 Oct 2015 18:44:45 +0000 (20:44 +0200)]
dnxhddata: introduce and use MBAFF flag

MBAFF-like handling of interlaced content in CID 1260 is different from
the other CIDs, and in particular doesn't use the same syntax.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/vf_w3fdif: scale down coefficiends by 2
Paul B Mahol [Wed, 7 Oct 2015 21:27:14 +0000 (23:27 +0200)]
avfilter/vf_w3fdif: scale down coefficiends by 2

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agodnxhddec: better support for 4:4:4
Christophe Gisquet [Sun, 4 Oct 2015 08:06:28 +0000 (10:06 +0200)]
dnxhddec: better support for 4:4:4

Profiles 1256 & 1270 (currently) signal at the frame header and MB
levels the colorspace used, either RGB or YUV. While a MB-level
varying colorspace is not supported, whether it is constant can be
tracked so as to determine the exact colorspace.

This requires having bitdepth and the ACT and 4:4:4 flags, in turn
needing the CID. Because setting those before having validated
enough things may result in invalid/unset DSP fucntions, setting
the bitdepth in the context is delayed.

It is not tested against a true RGB sequence, though.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/x86/vf_blend.asm: hardmix: do same with two pxor instructions less
Paul B Mahol [Wed, 7 Oct 2015 21:12:09 +0000 (23:12 +0200)]
avfilter/x86/vf_blend.asm: hardmix: do same with two pxor instructions less

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavfilter/x86/vf_blend.asm: 11th register is used, update functions
Paul B Mahol [Wed, 7 Oct 2015 20:53:54 +0000 (22:53 +0200)]
avfilter/x86/vf_blend.asm: 11th register is used, update functions

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavfilter/x86/vf_blend.asm: add hardmix and phoenix sse2 SIMD
Paul B Mahol [Wed, 7 Oct 2015 08:12:26 +0000 (10:12 +0200)]
avfilter/x86/vf_blend.asm: add hardmix and phoenix sse2 SIMD

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agolavfi/vf_yadif: reindent after last commit.
Nicolas George [Fri, 2 Oct 2015 13:58:53 +0000 (15:58 +0200)]
lavfi/vf_yadif: reindent after last commit.

3 years agolavfi/vf_thumbnail: reindent after last commit.
Nicolas George [Fri, 2 Oct 2015 14:20:17 +0000 (16:20 +0200)]
lavfi/vf_thumbnail: reindent after last commit.

3 years agolavfi/vf_w3fdif: reindent after last commit.
Nicolas George [Fri, 2 Oct 2015 14:17:56 +0000 (16:17 +0200)]
lavfi/vf_w3fdif: reindent after last commit.

3 years agolavfi/vf_alphamerge: reindent after last commit.
Nicolas George [Fri, 2 Oct 2015 14:12:51 +0000 (16:12 +0200)]
lavfi/vf_alphamerge: reindent after last commit.

3 years agolavfi/avf_showspectrum: reindent after last commit.
Nicolas George [Fri, 2 Oct 2015 14:03:16 +0000 (16:03 +0200)]
lavfi/avf_showspectrum: reindent after last commit.

3 years agolavfi/vf_yadif: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 13:54:41 +0000 (15:54 +0200)]
lavfi/vf_yadif: remove looping on request_frame().

3 years agolavfi/vf_w3fdif: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 14:17:41 +0000 (16:17 +0200)]
lavfi/vf_w3fdif: remove looping on request_frame().

3 years agolavfi/vf_thumbnail: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 14:19:55 +0000 (16:19 +0200)]
lavfi/vf_thumbnail: remove looping on request_frame().

3 years agolavfi/vf_select: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 14:29:08 +0000 (16:29 +0200)]
lavfi/vf_select: remove looping on request_frame().

3 years agolavfi/vf_fps: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 14:16:10 +0000 (16:16 +0200)]
lavfi/vf_fps: remove looping on request_frame().

3 years agolavfi/vf_alphamerge: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 14:12:38 +0000 (16:12 +0200)]
lavfi/vf_alphamerge: remove looping on request_frame().

3 years agolavfi/avf_showwaves: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 14:04:47 +0000 (16:04 +0200)]
lavfi/avf_showwaves: remove looping on request_frame().

3 years agolavfi/avf_showspectrum: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 14:03:02 +0000 (16:03 +0200)]
lavfi/avf_showspectrum: remove looping on request_frame().

3 years agolavfi/avf_showcqt: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 14:00:53 +0000 (16:00 +0200)]
lavfi/avf_showcqt: remove looping on request_frame().

3 years agolavfi/af_aresample: remove looping on request_frame().
Nicolas George [Fri, 2 Oct 2015 13:54:55 +0000 (15:54 +0200)]
lavfi/af_aresample: remove looping on request_frame().

3 years agow32pthreads: Add pthread_once emulation
Hendrik Leppkes [Wed, 7 Oct 2015 13:10:59 +0000 (15:10 +0200)]
w32pthreads: Add pthread_once emulation

The emulation uses native InitOnce* APIs on Windows Vista+, and a
lock-free/allocation-free approach using atomics and spinning for
Windows XP.

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
3 years agoavconv: Do not try to configure filter outputs without streams
Luca Barbato [Mon, 5 Oct 2015 18:49:55 +0000 (20:49 +0200)]
avconv: Do not try to configure filter outputs without streams

Prevent a NULL-dereference.

CC: libav-stable@libav.org
3 years agoconfigure: When disabling a library disable all the related components
Luca Barbato [Mon, 5 Oct 2015 01:41:41 +0000 (03:41 +0200)]
configure: When disabling a library disable all the related components

This way is sufficient to use the component specific configure variable
and not guard against the global library configure variable in code
that is outside it (e.g. checkasm).

3 years agodcadec: reorganise context data
Alexandra Khirnova [Fri, 2 Oct 2015 15:53:26 +0000 (17:53 +0200)]
dcadec: reorganise context data

place primary audio coding header data into DCAAudioHeader
structure to make DCAContext clearer
and move channel related data to DCAChan structure to make
them easier to use by extensions

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
3 years agog2m: Relax resolution change constraints
Vittorio Giovara [Wed, 7 Oct 2015 13:16:44 +0000 (15:16 +0200)]
g2m: Relax resolution change constraints

Do not fail when original resolution is smaller than current one,
as the frame buffer is resized automatically.

Signed-off-by: Vittorio Giovara <vittorio.giovara at gmail.com>
3 years agoswresample/resample: manually unroll the main loop in bessel()
Michael Niedermayer [Wed, 7 Oct 2015 15:57:58 +0000 (17:57 +0200)]
swresample/resample: manually unroll the main loop in bessel()

About 10% faster

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoswresample/resample: merge first iteration into init in bessel()
Michael Niedermayer [Wed, 7 Oct 2015 15:33:00 +0000 (17:33 +0200)]
swresample/resample: merge first iteration into init in bessel()

speedup of about 1%

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/mips: build fix for MSA 64bit
Shivraj Patil [Wed, 7 Oct 2015 12:51:59 +0000 (18:21 +0530)]
avcodec/mips: build fix for MSA 64bit

Modified datatype of function argument (pitch from int32_t to ptrdiff_t)

Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/mips: build fix for MSA
Shivraj Patil [Wed, 7 Oct 2015 12:50:53 +0000 (18:20 +0530)]
avcodec/mips: build fix for MSA

Modified sps and pps access from old HEVCContext(s) structure to newly introduced HEVCParamSets(ps)

Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodoc/encoders: Fix libkvazaar documentation
Arttu Ylä-Outinen [Mon, 28 Sep 2015 08:30:06 +0000 (11:30 +0300)]
doc/encoders: Fix libkvazaar documentation

The -threads option is ignored with libkvazaar since it does not have
any of the AV_CODEC_CAP_{FRAME,SLICE,AUTO}_THREADS capabilities. This
commit removes the incorrect documentation as well as the no-op of
setting the number of threads in libkvazaar encoder.

Signed-off-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>
3 years agolibkvazaar: Fix setting framerate
Arttu Ylä-Outinen [Mon, 28 Sep 2015 08:28:46 +0000 (11:28 +0300)]
libkvazaar: Fix setting framerate

The divisor and dividend in the equation had been swapped, making the
result the inverse of the actual framerate.

Signed-off-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>
3 years agolibkvazaar: Use av_image_copy for copying pixels
Arttu Ylä-Outinen [Mon, 28 Sep 2015 08:27:16 +0000 (11:27 +0300)]
libkvazaar: Use av_image_copy for copying pixels

Replaces a for loop for copying pixels by a call to av_image_copy.

Signed-off-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>
3 years agolibkvazaar: Set pts and dts
Arttu Ylä-Outinen [Mon, 28 Sep 2015 08:11:17 +0000 (11:11 +0300)]
libkvazaar: Set pts and dts

Signed-off-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>
3 years agolibkvazaar: Replace asserts with proper errors
Arttu Ylä-Outinen [Mon, 28 Sep 2015 08:08:19 +0000 (11:08 +0300)]
libkvazaar: Replace asserts with proper errors

Changes function libkvazaar_encode to return proper error codes instead
of crashing when the video dimensions or pixel format change in the
middle of encoding.

Signed-off-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>
3 years agolibkvazaar: Remove unnecessary NULL checks
Arttu Ylä-Outinen [Mon, 28 Sep 2015 08:03:07 +0000 (11:03 +0300)]
libkvazaar: Remove unnecessary NULL checks

Signed-off-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>
3 years agoconfigure: Add version check for libkvazaar
Arttu Ylä-Outinen [Tue, 29 Sep 2015 11:07:10 +0000 (14:07 +0300)]
configure: Add version check for libkvazaar

Signed-off-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>
3 years agolibkvazaar: Update to work with the latest version
Arttu Ylä-Outinen [Mon, 28 Sep 2015 07:55:14 +0000 (10:55 +0300)]
libkvazaar: Update to work with the latest version

Function encoder_encode in Kvazaar API was changed to have new output
parameters: source picture and frame info. Frame info is used to set the
keyframe flag and source picture is ignored.

Signed-off-by: Arttu Ylä-Outinen <arttu.yla-outinen@tut.fi>
3 years agovp9: don't keep a stack pointer if we don't need it.
Ronald S. Bultje [Tue, 6 Oct 2015 12:38:17 +0000 (08:38 -0400)]
vp9: don't keep a stack pointer if we don't need it.

This saves one register in a few cases on 32bit builds with unaligned
stack (e.g. MSVC), making the code slightly easier to maintain.

(Can someone please test this on 32bit+msvc and confirm make fate-vp9
and tests/checkasm/checkasm still work after this patch?)

3 years agoavcodec/libx264: silence -Waddress
Ganesh Ajjanagadde [Wed, 7 Oct 2015 02:48:13 +0000 (22:48 -0400)]
avcodec/libx264: silence -Waddress

This patch moves the pointer validity check outside the macro,
and silences the -Waddress observed with GCC 5.2.

Note that this changes the error message slightly, from:
"bad option..." to "Error parsing option...".

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
3 years agodoc/developer: use https instead of http
Ganesh Ajjanagadde [Wed, 7 Oct 2015 03:15:49 +0000 (23:15 -0400)]
doc/developer: use https instead of http

Change to https for FFmpeg websites.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/pngenc: Check that there is at least 1 frame
Michael Niedermayer [Wed, 7 Oct 2015 02:50:16 +0000 (04:50 +0200)]
avcodec/pngenc: Check that there is at least 1 frame

Fixes null pointer dereference
Fixes CID1322330

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavcodec/pngenc: Initialize fctl_chunk to 0
Michael Niedermayer [Wed, 7 Oct 2015 00:49:34 +0000 (02:49 +0200)]
avcodec/pngenc: Initialize fctl_chunk to 0

The structure is copied around and that triggers warnings if it is uninitialized
Fixes CID1322360

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavformat/flvdec: Print stream type in case a new stream is discovered after the header
Michael Niedermayer [Wed, 7 Oct 2015 00:41:40 +0000 (02:41 +0200)]
avformat/flvdec: Print stream type in case a new stream is discovered after the header

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agocheckasm: add alacdsp tests
James Almer [Tue, 6 Oct 2015 23:25:49 +0000 (20:25 -0300)]
checkasm: add alacdsp tests

Signed-off-by: James Almer <jamrial@gmail.com>
3 years agox86/alacdsp: add simd optimized functions
James Almer [Tue, 6 Oct 2015 23:22:00 +0000 (20:22 -0300)]
x86/alacdsp: add simd optimized functions

Signed-off-by: James Almer <jamrial@gmail.com>
3 years agoavformat/httpauth: Add space after commas in HTTP/RTSP auth header
Andrey Utkin [Thu, 1 Oct 2015 10:56:31 +0000 (13:56 +0300)]
avformat/httpauth: Add space after commas in HTTP/RTSP auth header

This fixes access to Grandstream cameras, which return 401 to ffmpeg
otherwise.
VLC sends Authorization: header with spaces between parameters, and it
is known to work with Grandstream devices and broad range of other HTTP
and RTSP servers, so author considers switching to such behaviour safe.
Just for record - RFC 2617 (HTTP Auth) does not specify the need in
spaces, so this is not a bug of FFmpeg.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavfilter/vf_stereo3d: add x86 SIMD for anaglyph outputs
Paul B Mahol [Sun, 4 Oct 2015 09:34:03 +0000 (11:34 +0200)]
avfilter/vf_stereo3d: add x86 SIMD for anaglyph outputs

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoREADME: replace http with https
Ganesh Ajjanagadde [Tue, 6 Oct 2015 03:39:08 +0000 (23:39 -0400)]
README: replace http with https

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoafilter/af_extrastereo: remove dead initialization
Paul B Mahol [Tue, 6 Oct 2015 07:56:22 +0000 (09:56 +0200)]
afilter/af_extrastereo: remove dead initialization

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavfilter:audio: fix shadowed AVFrame *out
Paul B Mahol [Tue, 6 Oct 2015 07:50:30 +0000 (09:50 +0200)]
avfilter:audio: fix shadowed AVFrame *out

Signed-off-by: Paul B Mahol <onemda@gmail.com>
3 years agoavfilter/af_extrastereo: Fix shadowed variable
Michael Niedermayer [Tue, 6 Oct 2015 01:16:24 +0000 (03:16 +0200)]
avfilter/af_extrastereo: Fix shadowed variable

Fixes CID1325675

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoavformat/flvdec: Remove dead loop
Michael Niedermayer [Mon, 5 Oct 2015 22:17:40 +0000 (00:17 +0200)]
avformat/flvdec: Remove dead loop

Fixes CID1325682

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agovp9: fix msvc build by using 6 GPRs on 32bit if stack!=aligned.
Ronald S. Bultje [Mon, 5 Oct 2015 11:39:04 +0000 (07:39 -0400)]
vp9: fix msvc build by using 6 GPRs on 32bit if stack!=aligned.

3 years agoavformat/omadec: Fix { typo
Michael Niedermayer [Mon, 5 Oct 2015 19:09:40 +0000 (21:09 +0200)]
avformat/omadec: Fix { typo

Fixes CID1324299

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoffplay: use correct context for av_log
Ganesh Ajjanagadde [Mon, 5 Oct 2015 00:38:24 +0000 (20:38 -0400)]
ffplay: use correct context for av_log

Recent commits c3e8de1c248f8c742dd9e61a0c71ee56bba22c28 and
8dc6e92c3dc67a85026f3010045c7a28b1c0adc8 used av_log incorrectly.
This fixes such usage.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agoffplay: more robust condition variable creation
Ganesh Ajjanagadde [Mon, 5 Oct 2015 00:38:06 +0000 (20:38 -0400)]
ffplay: more robust condition variable creation

SDL_CreateCond can fail:
https://wiki.libsdl.org/SDL_CreateCond.
This patch makes creation more robust in one instance.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agoffplay: log SDL error messages
Ganesh Ajjanagadde [Mon, 5 Oct 2015 00:37:39 +0000 (20:37 -0400)]
ffplay: log SDL error messages

This logs the SDL error messages on failure of creation of SDL_CreateMutex,
SDL_CreateCond, and SDL_CreateThread.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Marton Balint <cus@passwd.hu>
3 years agoavutil/attributes: add av_warn_unused_result
Ganesh Ajjanagadde [Mon, 5 Oct 2015 05:45:12 +0000 (01:45 -0400)]
avutil/attributes: add av_warn_unused_result

GCC 3.4 introduced an attribute warn_unused_result to warn when a programmer
discards the return value. Applying this judiciously across the codebase can help
in fixing a lot of problems. At a high level, functions which return error codes
should always be checked. More concretely, consider the functions ff_add_format
and the like in avfilter/formats.h. A quick examination shows that a large portion
of libavfilter fails to handle the associated errors, usually AVERROR(ENOMEM).
The above example was where I observed the utility of this, but it should be
useful in many places across the code base.

Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodnxhd: add better support for CIDs 1270 to 1274
Jeremy James [Mon, 28 Sep 2015 15:28:12 +0000 (16:28 +0100)]
dnxhd: add better support for CIDs 1270 to 1274

These are DNxHR profiles with the following properties:
- Variable size in a profile (property added in a previous commit),
  requiring variable-sized macroblock table;
- Variable bitdepth, up to 12 bits.
- Better validation of buffer sizes and positions

Signed-off-by: Christophe Gisquet <christophe.gisquet@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodnxhd: add CID 1270
Christophe Gisquet [Sat, 3 Oct 2015 16:59:20 +0000 (18:59 +0200)]
dnxhd: add CID 1270

This a 4:4:4 10 bits profile, where image size is not fixed by the
profile, and which strays a bit outside the old frame header parsing
code.

Fixes ticket #4581 (DNxHR is not stricly supported, but that sequence is).

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodnxhd: add decoder support for DNxHR
Jeremy James [Sat, 3 Oct 2015 16:59:19 +0000 (18:59 +0200)]
dnxhd: add decoder support for DNxHR

Signed-off-by: Christophe Gisquet <christophe.gisquet@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agoisom: add support for DNxHR codec family
Christophe Gisquet [Sat, 3 Oct 2015 16:59:18 +0000 (18:59 +0200)]
isom: add support for DNxHR codec family

This is actually similar to DNxHD.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodnxhddec: Introduce DNXHD_VARIABLE
Christophe Gisquet [Sat, 3 Oct 2015 16:59:17 +0000 (18:59 +0200)]
dnxhddec: Introduce DNXHD_VARIABLE

Currently not used, but will be used to indicate that a CIDEntry field
is not set, because it is variable, and that checks should be adapted.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodnxhdenc: do not select 4:4:4 profiles
Christophe Gisquet [Sat, 3 Oct 2015 16:59:16 +0000 (18:59 +0200)]
dnxhdenc: do not select 4:4:4 profiles

The encoder can only deal with 4:2:2.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodnxhd: profile flags
Christophe Gisquet [Sat, 3 Oct 2015 16:59:15 +0000 (18:59 +0200)]
dnxhd: profile flags

Move the 'interlaced' flag to this element (arbitrarily set to 16bits).
This should allow better detection/selection of profiles.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
3 years agodnxhddec: cleanup frame header parsing
Christophe Gisquet [Fri, 2 Oct 2015 19:00:44 +0000 (21:00 +0200)]
dnxhddec: cleanup frame header parsing

Rely more on the actual syntax from the specs (also seen in the
encoder code).

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>