ffmpeg.git
8 years agoffplay: dont drop frames by default when there is no audio stream
Michael Niedermayer [Sun, 19 Jun 2011 03:55:12 +0000 (05:55 +0200)]
ffplay: dont drop frames by default when there is no audio stream

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agolavfi: add layout negotiation fields and helper functions.
Mina Nagy Zaki [Sun, 19 Jun 2011 18:31:24 +0000 (20:31 +0200)]
lavfi: add layout negotiation fields and helper functions.

8 years agolavfi: use int64_t lists in AVFilteFormats
Mina Nagy Zaki [Tue, 7 Jun 2011 18:17:23 +0000 (21:17 +0300)]
lavfi: use int64_t lists in AVFilteFormats

The list type was changed to int64_t to be able to hold
channel layouts.

avfilter_make_format_list() still takes a int32_t array and converts
it to int64_t. A new function, avfilter_make_format64_list, that
takes int64_t arrays has been added.

8 years agolavfi: fix realloc size computation in avfilter_add_format()
Stefano Sabatini [Sun, 19 Jun 2011 20:07:18 +0000 (22:07 +0200)]
lavfi: fix realloc size computation in avfilter_add_format()

Replace sizeof((*avff)->formats)
with    sizeof(*(*avff)->formats)

as the size of the array element is given by the pointed element
rather than by its pointer.

In particular fix computation with the pending patch when
sizeof(int64_t) != sizeof(int64_t *).

8 years agolavfi: fix Makefile HEADERS
Stefano Sabatini [Sun, 19 Jun 2011 20:42:09 +0000 (22:42 +0200)]
lavfi: fix Makefile HEADERS

Add a missing space between vsink_buffer.h and vsrc_buffer.h.
1000l.

8 years agolavfi: add vsink_buffer, and use it in ff* tools
Stefano Sabatini [Sat, 11 Jun 2011 16:43:11 +0000 (18:43 +0200)]
lavfi: add vsink_buffer, and use it in ff* tools

Also add the public interface libavfilter/vsink_buffer.h.

8 years agovsrc_color: use internal timebase
Stefano Sabatini [Fri, 17 Jun 2011 23:52:40 +0000 (01:52 +0200)]
vsrc_color: use internal timebase

Avoid timescale conversion, simplify.

8 years agoopt: do not crash in av_set_options_string() if opts == NULL
Stefano Sabatini [Sun, 19 Jun 2011 00:49:04 +0000 (02:49 +0200)]
opt: do not crash in av_set_options_string() if opts == NULL

Add missing NULL check, and update documentation accordingly.

8 years agolavfi: add negate filter
Stefano Sabatini [Sat, 28 May 2011 20:00:26 +0000 (22:00 +0200)]
lavfi: add negate filter

This filter is a simple wrapper around the LUT filter.

8 years agolavfi: add LUT (LookUp Table) generic filters
Stefano Sabatini [Sat, 28 May 2011 19:51:03 +0000 (21:51 +0200)]
lavfi: add LUT (LookUp Table) generic filters

8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Sun, 19 Jun 2011 02:02:06 +0000 (04:02 +0200)]
Merge remote-tracking branch 'qatar/master'

* qatar/master:
  lavf: prevent crash in av_open_input_file() if ap == NULL.
  more Changelog additions
  lavf: add a forgotten NULL check in convert_format_parameters().
  Fix build if yasm is not available.
  H.264: Add x86 assembly for 10-bit MC Chroma H.264 functions.

Conflicts:
Changelog

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoswscale: fix mono input for width%8
Michael Niedermayer [Sun, 19 Jun 2011 01:33:15 +0000 (03:33 +0200)]
swscale: fix mono input for width%8

Fixes ticket193

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoavidec: Ignore unknown stream types
ami_stuff [Sat, 18 Jun 2011 23:47:25 +0000 (01:47 +0200)]
avidec: Ignore unknown stream types

Improves ticket131

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoh264: print actual slice number in "Too many slices" warning
Michael Niedermayer [Sat, 18 Jun 2011 22:02:55 +0000 (00:02 +0200)]
h264: print actual slice number in "Too many slices" warning

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agojpegdec: include mjpb_skiptosod in debug output
Michael Niedermayer [Sat, 18 Jun 2011 22:00:52 +0000 (00:00 +0200)]
jpegdec: include mjpb_skiptosod in debug output

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agojpegdec: fix CJPG decoding.
Michael Niedermayer [Sat, 18 Jun 2011 21:59:27 +0000 (23:59 +0200)]
jpegdec: fix CJPG decoding.

Fixes ticket133

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agow32threads: with only 1 thread there cant be a active threading mode.
Michael Niedermayer [Sat, 18 Jun 2011 19:59:11 +0000 (21:59 +0200)]
w32threads: with only 1 thread there cant be a active threading mode.

Fixes ticket284

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoffmpeg: warn if samples need to be converted in a lossy way for lossless encoders.
Michael Niedermayer [Sat, 18 Jun 2011 19:54:11 +0000 (21:54 +0200)]
ffmpeg: warn if samples need to be converted in a lossy way for lossless encoders.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoflacenc: mark as lossless codec
Michael Niedermayer [Sat, 18 Jun 2011 19:53:54 +0000 (21:53 +0200)]
flacenc: mark as lossless codec

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agolibavcodec: add CODEC_CAP_LOSSLESS
Michael Niedermayer [Sat, 18 Jun 2011 19:53:20 +0000 (21:53 +0200)]
libavcodec: add CODEC_CAP_LOSSLESS

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agolavf: prevent crash in av_open_input_file() if ap == NULL.
Ronald S. Bultje [Sat, 18 Jun 2011 19:33:49 +0000 (15:33 -0400)]
lavf: prevent crash in av_open_input_file() if ap == NULL.

Needed for proper behaviour in our old API compatibility code.

8 years agomore Changelog additions
Reinhard Tartler [Thu, 16 Jun 2011 17:59:47 +0000 (19:59 +0200)]
more Changelog additions

8 years agolavf: add a forgotten NULL check in convert_format_parameters().
Anton Khirnov [Sat, 18 Jun 2011 15:26:40 +0000 (17:26 +0200)]
lavf: add a forgotten NULL check in convert_format_parameters().

8 years agoFix build if yasm is not available.
Ronald S. Bultje [Sat, 18 Jun 2011 12:34:14 +0000 (08:34 -0400)]
Fix build if yasm is not available.

8 years agoH.264: Add x86 assembly for 10-bit MC Chroma H.264 functions.
Daniel Kang [Sun, 5 Jun 2011 22:33:23 +0000 (18:33 -0400)]
H.264: Add x86 assembly for 10-bit MC Chroma H.264 functions.

Mainly ported from 8-bit H.264 MC Chroma.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
8 years agodcaenc: small quantization simplification.
Clément Bœsch [Wed, 15 Jun 2011 18:26:57 +0000 (20:26 +0200)]
dcaenc: small quantization simplification.

8 years agoFix -s documentation.
Carl Eugen Hoyos [Sat, 18 Jun 2011 07:36:07 +0000 (09:36 +0200)]
Fix -s documentation.

8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Sat, 18 Jun 2011 02:40:18 +0000 (04:40 +0200)]
Merge remote-tracking branch 'qatar/master'

* qatar/master:
  lavc: add opt_find to AVCodecContext class.
  h264: Complexify frame num gap shortening code
  intreadwrite.h: fix AV_RL32/AV_RB32 signedness.
  Fix decoding of mpegts streams with h264 video that does *NOT* have b frames
  Add minor bumps and APIChanges entries for lavf private options.
  ffmpeg: deprecate -vc and -tvstd
  ffmpeg: use new avformat_open_* API.
  ffserver: use new avformat_open_* API.
  ffprobe: use new avformat_open_* API.
  ffplay: use new avformat_open_* API.
  cmdutils: add opt_default2().
  dict: add AV_DICT_APPEND flag.
  lavf: add avformat_write_header() as a replacement for av_write_header().
  Deprecate av_open_input_* and remove their uses.
  lavf: add avformat_open_input() as a replacement for av_open_input_*
  AVOptions: add av_opt_find() as a replacement for av_find_opt.
  AVOptions: add av_opt_set_dict() mapping a dictionary struct to a context.
  ffmpeg: don't abuse a global for passing frame size from input to output
  ffmpeg: don't abuse a global for passing pixel format from input to output
  ffmpeg: initialise encoders earlier.

Conflicts:
cmdutils.c
doc/APIchanges
ffmpeg.c
ffplay.c
ffprobe.c
libavcodec/h264.c
libavformat/avformat.h
libavformat/utils.c
libavformat/version.h
libavutil/avutil.h

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoqdm2: Fix alignment of local array.
Michael Niedermayer [Sat, 18 Jun 2011 00:44:17 +0000 (02:44 +0200)]
qdm2: Fix alignment of local array.

Fixes ticket270

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agovsrc_color: set output pos values to -1
Stefano Sabatini [Fri, 17 Jun 2011 23:47:37 +0000 (01:47 +0200)]
vsrc_color: set output pos values to -1

-1 is more correct than 0, as the position in the file is undefined.

8 years agovsrc_color: add @file doxy
Stefano Sabatini [Fri, 17 Jun 2011 23:46:27 +0000 (01:46 +0200)]
vsrc_color: add @file doxy

Also remove outdated reference to color in vf_pad.c.

8 years agographparser: add missing NULL check in avfilter_graph_parse()
Stefano Sabatini [Sat, 18 Jun 2011 00:02:31 +0000 (02:02 +0200)]
graphparser: add missing NULL check in avfilter_graph_parse()

Fix a crash occurring when open_inputs is NULL and *open_inputs is
checked, the crash was introduced by the recent avfilter_graph_parse()
syntax change.

In particular, fix graph2dot crash.

8 years agoSupport DTS in mp4/mov.
Carl Eugen Hoyos [Thu, 16 Jun 2011 21:45:38 +0000 (23:45 +0200)]
Support DTS in mp4/mov.

8 years agolavc: add opt_find to AVCodecContext class.
Anton Khirnov [Fri, 17 Jun 2011 04:39:42 +0000 (06:39 +0200)]
lavc: add opt_find to AVCodecContext class.

Fixes encoder private options, which are broken now in ffmpeg, because
opt_default2() can't find them, thus making opt_default() to fail.

8 years agoMerge commit '9446d75941d639f19cfa9ae007eb4c5ca041f200'
Michael Niedermayer [Fri, 17 Jun 2011 03:49:41 +0000 (05:49 +0200)]
Merge commit '9446d75941d639f19cfa9ae007eb4c5ca041f200'

* commit '9446d75941d639f19cfa9ae007eb4c5ca041f200':
  ffmpeg: merge output_codecs array into AVOutputStream members.
  rtpenc_latm: Consistently use "Libav" in license boilerplate.
  rtsp: Don't pass string pointer as format string to ff_url_join
  mmsh: fixed printf injection bug in mmsh request
  codec-regression: remove pointless -r options for dnxhd

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoDo not crash if SDL_SetVideoMode() fails.
Carl Eugen Hoyos [Thu, 16 Jun 2011 21:21:19 +0000 (23:21 +0200)]
Do not crash if SDL_SetVideoMode() fails.

8 years agoMove do_exit() and stream_close() up.
Carl Eugen Hoyos [Thu, 16 Jun 2011 21:20:29 +0000 (23:20 +0200)]
Move do_exit() and stream_close() up.

8 years agoFix documentation for -ar and -ac.
Carl Eugen Hoyos [Thu, 16 Jun 2011 21:16:58 +0000 (23:16 +0200)]
Fix documentation for -ar and -ac.

8 years agoFix compilation with old yasm.
Carl Eugen Hoyos [Thu, 16 Jun 2011 21:14:17 +0000 (23:14 +0200)]
Fix compilation with old yasm.

8 years agoffmpeg: Force 420 with target dvd and svcd.
Michael Niedermayer [Thu, 16 Jun 2011 21:06:27 +0000 (23:06 +0200)]
ffmpeg: Force 420 with target dvd and svcd.

Fixes ticket283

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoh264: Complexify frame num gap shortening code
Alexander Strange [Sun, 12 Jun 2011 20:40:00 +0000 (20:40 +0000)]
h264: Complexify frame num gap shortening code

By observation it did not seem to handle prev_frame_num > frame_num.
This does not affect any files I have.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agointreadwrite.h: fix AV_RL32/AV_RB32 signedness.
Uoti Urpala [Tue, 14 Jun 2011 21:53:30 +0000 (21:53 +0000)]
intreadwrite.h: fix AV_RL32/AV_RB32 signedness.

The output type of the AV_RL32/AV_RB32 macros was signed int. The
resulting overflow broke at least some ASF streams with large
timestamps. Fix by adding a cast to uint32_t.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agoFix decoding of mpegts streams with h264 video that does *NOT* have b frames
Mike Scheutzow [Thu, 28 Apr 2011 14:14:26 +0000 (10:14 -0400)]
Fix decoding of mpegts streams with h264 video that does *NOT* have b frames

One of the causes of this bug is that the h264 parser defaults low_delay
to 1, but the h264 codec defaults low_delay to 0. Really Ugly.

After many hours of looking at this, I'm still not sure how has_b_frames
is *intended* to behave, but to me the implementation appears way more
complicated than it ought to be.

My patch relies on the encoder to set an optional field in the SPS. This
works for libx264 streams, but I'm not sure that all h264 encoders will
set it.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agoAdd minor bumps and APIChanges entries for lavf private options.
Anton Khirnov [Thu, 16 Jun 2011 15:21:46 +0000 (17:21 +0200)]
Add minor bumps and APIChanges entries for lavf private options.

8 years agoffmpeg: deprecate -vc and -tvstd
Anton Khirnov [Wed, 15 Jun 2011 05:34:12 +0000 (07:34 +0200)]
ffmpeg: deprecate -vc and -tvstd

They've been replaced by demuxer private options.

8 years agoffmpeg: use new avformat_open_* API.
Anton Khirnov [Thu, 9 Jun 2011 08:58:23 +0000 (10:58 +0200)]
ffmpeg: use new avformat_open_* API.

8 years agoffserver: use new avformat_open_* API.
Anton Khirnov [Thu, 9 Jun 2011 08:58:23 +0000 (10:58 +0200)]
ffserver: use new avformat_open_* API.

8 years agoffprobe: use new avformat_open_* API.
Anton Khirnov [Thu, 9 Jun 2011 08:58:23 +0000 (10:58 +0200)]
ffprobe: use new avformat_open_* API.

8 years agoffplay: use new avformat_open_* API.
Anton Khirnov [Thu, 9 Jun 2011 08:58:23 +0000 (10:58 +0200)]
ffplay: use new avformat_open_* API.

8 years agocmdutils: add opt_default2().
Anton Khirnov [Wed, 25 May 2011 14:42:41 +0000 (16:42 +0200)]
cmdutils: add opt_default2().

It stores options in a dictionary to be passed to new open calls.

It will replace opt_default once all the pieces are in place.

8 years agodict: add AV_DICT_APPEND flag.
Anton Khirnov [Sat, 4 Jun 2011 16:43:05 +0000 (18:43 +0200)]
dict: add AV_DICT_APPEND flag.

8 years agolavf: add avformat_write_header() as a replacement for av_write_header().
Anton Khirnov [Sun, 22 May 2011 11:53:33 +0000 (13:53 +0200)]
lavf: add avformat_write_header() as a replacement for av_write_header().

It supports more convenient setting of AVOptions.

8 years agoDeprecate av_open_input_* and remove their uses.
Anton Khirnov [Sat, 4 Jun 2011 15:36:30 +0000 (17:36 +0200)]
Deprecate av_open_input_* and remove their uses.

Deprecate the last remaining member of AVFormatParameters.

8 years agolavf: add avformat_open_input() as a replacement for av_open_input_*
Anton Khirnov [Sun, 22 May 2011 06:37:25 +0000 (08:37 +0200)]
lavf: add avformat_open_input() as a replacement for av_open_input_*

Add support for demuxer private options.

8 years agoAVOptions: add av_opt_find() as a replacement for av_find_opt.
Anton Khirnov [Sun, 22 May 2011 11:25:19 +0000 (13:25 +0200)]
AVOptions: add av_opt_find() as a replacement for av_find_opt.

8 years agoAVOptions: add av_opt_set_dict() mapping a dictionary struct to a context.
Anton Khirnov [Sun, 22 May 2011 10:52:50 +0000 (12:52 +0200)]
AVOptions: add av_opt_set_dict() mapping a dictionary struct to a context.

This way the caller can pass all the options in one nice package.

8 years agoffmpeg: don't abuse a global for passing frame size from input to output
Anton Khirnov [Wed, 15 Jun 2011 06:00:03 +0000 (08:00 +0200)]
ffmpeg: don't abuse a global for passing frame size from input to output

It's broken with multiple files or video streams.

8 years agoffmpeg: don't abuse a global for passing pixel format from input to output
Anton Khirnov [Wed, 15 Jun 2011 06:00:03 +0000 (08:00 +0200)]
ffmpeg: don't abuse a global for passing pixel format from input to output

It's broken with multiple files or video streams.

8 years agommsh: fix 400 bad request
Kirill Zorin [Thu, 16 Jun 2011 18:15:53 +0000 (20:15 +0200)]
mmsh: fix 400 bad request

There is no need to write two HTTP newlines (\r\n) into "headers",
because http_connect (in http.c) already appends one HTTP newline at
the end of the given headers chunk, which would result in sending
three HTTP newlines after the headers. Most of the time it's okay
(although not RFC-conforming), but many proxy servers and the
occasional strict httpd will puke with a "400 bad request".

8 years agoffmpeg: merge output_codecs array into AVOutputStream members.
Anton Khirnov [Thu, 16 Jun 2011 16:04:56 +0000 (18:04 +0200)]
ffmpeg: merge output_codecs array into AVOutputStream members.

There's no point in keeping them separate.

8 years agoffmpeg: initialise encoders earlier.
Anton Khirnov [Thu, 16 Jun 2011 04:24:54 +0000 (06:24 +0200)]
ffmpeg: initialise encoders earlier.

Fixes choosing supported samplerate and framerate.

8 years agortpenc_latm: Consistently use "Libav" in license boilerplate.
Diego Biurrun [Thu, 16 Jun 2011 15:04:06 +0000 (17:04 +0200)]
rtpenc_latm: Consistently use "Libav" in license boilerplate.

8 years agortsp: Don't pass string pointer as format string to ff_url_join
Martin Storsjö [Thu, 16 Jun 2011 08:04:27 +0000 (11:04 +0300)]
rtsp: Don't pass string pointer as format string to ff_url_join

In this case, the string that was passed couldn't contain
user-defined data and thus there was no risk for injection
bugs, but it's safer this way, if we later change the
content of the options string.

Signed-off-by: Martin Storsjö <martin@martin.st>
8 years agommsh: fixed printf injection bug in mmsh request
Kirill Zorin [Wed, 15 Jun 2011 17:18:29 +0000 (19:18 +0200)]
mmsh: fixed printf injection bug in mmsh request

Signed-off-by: Martin Storsjö <martin@martin.st>
8 years agocodec-regression: remove pointless -r options for dnxhd
Anton Khirnov [Thu, 16 Jun 2011 05:24:14 +0000 (07:24 +0200)]
codec-regression: remove pointless -r options for dnxhd

8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Thu, 16 Jun 2011 01:53:58 +0000 (03:53 +0200)]
Merge remote-tracking branch 'qatar/master'

* qatar/master:
  ac3enc: use correct alignment and length in channel coupling dsp functions.
  ffmpeg: don't abuse a global for passing framerate from input to output
  ffmpeg: don't abuse a global for passing channels from input to output
  ffmpeg: don't abuse a global for passing samplerate from input to output
  ARM: update ff_h264_idct8_add4_neon for 4:4:4 changes
  swscale: use SwsContext for av_log when available
  swscale: Remove HAVE_MMX from files that are only compiled with MMX enabled.
  swscale: Fix compilation with --disable-mmx2.

Conflicts:
ffmpeg.c
libswscale/utils.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agommsh: fixed printf injection bug in mmsh request
Kirill Zorin [Wed, 15 Jun 2011 17:18:29 +0000 (19:18 +0200)]
mmsh: fixed printf injection bug in mmsh request

8 years agoac3enc: use correct alignment and length in channel coupling dsp functions.
Justin Ruggles [Tue, 14 Jun 2011 17:45:38 +0000 (13:45 -0400)]
ac3enc: use correct alignment and length in channel coupling dsp functions.

This fixes a segfault when using the C version of ac3dsp.float_to_fixed24().

8 years agoffmpeg: don't abuse a global for passing framerate from input to output
Anton Khirnov [Wed, 15 Jun 2011 06:00:03 +0000 (08:00 +0200)]
ffmpeg: don't abuse a global for passing framerate from input to output

It's broken with multiple files or video streams.

8 years agoffmpeg: don't abuse a global for passing channels from input to output
Anton Khirnov [Wed, 15 Jun 2011 06:00:03 +0000 (08:00 +0200)]
ffmpeg: don't abuse a global for passing channels from input to output

It's broken with multiple files or audio streams.

8 years agoffmpeg: don't abuse a global for passing samplerate from input to output
Anton Khirnov [Wed, 15 Jun 2011 06:00:03 +0000 (08:00 +0200)]
ffmpeg: don't abuse a global for passing samplerate from input to output

It's broken with multiple files or audio streams.

This removes the default samplerate of 44100 for raw input, hence all
the FATE changes.

8 years agoMake buffer size check consistent and avoid a possible overflow.
Reimar Döffinger [Wed, 15 Jun 2011 18:21:33 +0000 (20:21 +0200)]
Make buffer size check consistent and avoid a possible overflow.

8 years agoFix spelling.
Reimar Döffinger [Wed, 15 Jun 2011 18:20:47 +0000 (20:20 +0200)]
Fix spelling.

8 years agoFull support for sending H.264 in RTP
Luca Abeni [Wed, 15 Jun 2011 15:59:03 +0000 (17:59 +0200)]
Full support for sending H.264 in RTP

This implements support for the "MP4" syntax of H.264 bitstreams.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoARM: update ff_h264_idct8_add4_neon for 4:4:4 changes
Mans Rullgard [Wed, 15 Jun 2011 11:58:00 +0000 (12:58 +0100)]
ARM: update ff_h264_idct8_add4_neon for 4:4:4 changes

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agoswscale: use SwsContext for av_log when available
Janne Grunau [Mon, 25 Apr 2011 22:27:48 +0000 (00:27 +0200)]
swscale: use SwsContext for av_log when available

Signed-off-by: Diego Biurrun <diego@biurrun.de>
8 years agoSupport reading chan atoms with empty channel descriptions.
Andrew Wason [Wed, 15 Jun 2011 09:09:17 +0000 (11:09 +0200)]
Support reading chan atoms with empty channel descriptions.

Fixes ticket 263.

8 years agoReindent after last commit.
Carl Eugen Hoyos [Wed, 15 Jun 2011 08:58:25 +0000 (10:58 +0200)]
Reindent after last commit.

8 years agoFix multi-channel AAC encoding.
Carl Eugen Hoyos [Wed, 15 Jun 2011 08:57:26 +0000 (10:57 +0200)]
Fix multi-channel AAC encoding.

Fixes ticket 55.

8 years agoFix "redundant redeclaration" warning.
Carl Eugen Hoyos [Wed, 15 Jun 2011 08:53:18 +0000 (10:53 +0200)]
Fix "redundant redeclaration" warning.

8 years agoFix compilation with --disable-everything --enable-encoder=ac3/ac3_fixed.
Carl Eugen Hoyos [Wed, 15 Jun 2011 08:47:46 +0000 (10:47 +0200)]
Fix compilation with --disable-everything --enable-encoder=ac3/ac3_fixed.

8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Wed, 15 Jun 2011 00:15:25 +0000 (02:15 +0200)]
Merge remote-tracking branch 'qatar/master'

* qatar/master:
  swscale: remove misplaced comment.
  ffmpeg: fix streaming to ffserver.
  swscale: split out RGB48 output functions from yuv2packed[12X]_c().
  build: move vpath directives to main Makefile
  swscale: fix JPEG-range YUV scaling artifacts.
  build: move ALLFFLIBS to a more logical place
  ARM: factor some repetitive code into macros
  Fix SVQ3 after adding 4:4:4 H.264 support
  H.264: fix CODEC_FLAG_GRAY
  4:4:4 H.264 decoding support
  ac3enc: fix allocation of floating point samples.

Conflicts:
ffmpeg.c
libavcodec/dsputil_template.c
libavcodec/h264.c
libavcodec/mpegvideo.c
libavcodec/snow.c
libswscale/swscale.c
libswscale/swscale_internal.h

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agovf_mp: Fix large memleak.
Michael Niedermayer [Tue, 14 Jun 2011 23:36:12 +0000 (01:36 +0200)]
vf_mp: Fix large memleak.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoswscale: Remove HAVE_MMX from files that are only compiled with MMX enabled.
Diego Biurrun [Tue, 14 Jun 2011 22:56:31 +0000 (00:56 +0200)]
swscale: Remove HAVE_MMX from files that are only compiled with MMX enabled.

8 years agoswscale: Fix compilation with --disable-mmx2.
Diego Biurrun [Tue, 14 Jun 2011 22:52:43 +0000 (00:52 +0200)]
swscale: Fix compilation with --disable-mmx2.

Some MMX2 functions were being referenced without proper #ifdefs.

8 years agomjpegenc: Fix JFIF version
William Yu [Tue, 14 Jun 2011 15:16:51 +0000 (17:16 +0200)]
mjpegenc: Fix JFIF version

8 years agoswscale: remove misplaced comment.
Ronald S. Bultje [Tue, 14 Jun 2011 20:55:24 +0000 (16:55 -0400)]
swscale: remove misplaced comment.

The comment should have been placed only in
yuv2rgb48_X_c_template, not yuv2rgb48_1_c_template.

8 years agoffmpeg: fix streaming to ffserver.
Ronald S. Bultje [Tue, 14 Jun 2011 19:35:05 +0000 (15:35 -0400)]
ffmpeg: fix streaming to ffserver.

8 years agoswscale: split out RGB48 output functions from yuv2packed[12X]_c().
Ronald S. Bultje [Tue, 14 Jun 2011 19:02:24 +0000 (15:02 -0400)]
swscale: split out RGB48 output functions from yuv2packed[12X]_c().

This is part of the Great Evil Plan to simplify swscale. Note that
you'll see some code duplication between the output functions for
different RGB variants, and even between packed-YUV and RGB
variants. This is intentional because it improves readability.

8 years agobuild: move vpath directives to main Makefile
Mans Rullgard [Sun, 5 Jun 2011 21:00:54 +0000 (22:00 +0100)]
build: move vpath directives to main Makefile

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agoswscale: fix JPEG-range YUV scaling artifacts.
Michael Niedermayer [Mon, 13 Jun 2011 01:53:22 +0000 (21:53 -0400)]
swscale: fix JPEG-range YUV scaling artifacts.

YUV planes were marked as uint16_t, but they contained signed data.
Fixes issue 1108 and 675.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
8 years agobuild: move ALLFFLIBS to a more logical place
Mans Rullgard [Sun, 5 Jun 2011 21:18:06 +0000 (22:18 +0100)]
build: move ALLFFLIBS to a more logical place

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agoARM: factor some repetitive code into macros
Mans Rullgard [Sun, 12 Jun 2011 23:33:19 +0000 (00:33 +0100)]
ARM: factor some repetitive code into macros

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agoCrystalHD: Use mp4toannexb bitstream filter.
Philip Langdale [Mon, 13 Jun 2011 03:22:20 +0000 (20:22 -0700)]
CrystalHD: Use mp4toannexb bitstream filter.

The H.264 parser that we use to detect interlacing can only handle
an Annex B stream, so we need to actually use the filter. This is
unfortunate as the crystalhd library is already doing this conversion
internally. A future change will reorganise the decode path more
completely so that we can feed the converted stream into libcrystalhd
and avoid the second conversion.

Signed-off-by: Philip Langdale <philipl@overt.org>
8 years agoCrystalHD: Keep mp4toannexb filter around for entire decoder lifetime.
Philip Langdale [Mon, 13 Jun 2011 03:21:28 +0000 (20:21 -0700)]
CrystalHD: Keep mp4toannexb filter around for entire decoder lifetime.

In preparation for using the filter on the actual bitstream, we need
to extend it's lifetime to match that of the decoder.

Signed-off-by: Philip Langdale <philipl@overt.org>
8 years agoFix SVQ3 after adding 4:4:4 H.264 support
Jason Garrett-Glaser [Mon, 13 Jun 2011 17:21:46 +0000 (10:21 -0700)]
Fix SVQ3 after adding 4:4:4 H.264 support

8 years agoH.264: fix CODEC_FLAG_GRAY
Jason Garrett-Glaser [Thu, 9 Jun 2011 23:17:41 +0000 (16:17 -0700)]
H.264: fix CODEC_FLAG_GRAY

It was broken in 4:4:4, and still did chroma deblocking for no reason in 4:2:0.

8 years ago4:4:4 H.264 decoding support
Jason Garrett-Glaser [Fri, 3 Jun 2011 08:12:28 +0000 (01:12 -0700)]
4:4:4 H.264 decoding support

Note: this is 4:4:4 from the 2007 spec revision, not the previous (now deprecated) 4:4:4 mode in H.264.

8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Tue, 14 Jun 2011 02:55:27 +0000 (04:55 +0200)]
Merge remote-tracking branch 'qatar/master'

* qatar/master: (24 commits)
  utils: Drop pointless '#if 1' preprocessor directive.
  ac3enc: remove empty ac3_float function that is never called
  ac3enc: split templated float vs. fixed functions into a separate file.
  ac3enc: dynamically allocate AC3EncodeContext fields windowed_samples and mdct
  ac3enc: use function pointer to choose between AC-3 and E-AC-3 header output functions.
  Roll back 4:4:4 H.264 for now Needs some ARM/PPC asm modifications.
  Fix SVQ3 after adding 4:4:4 H.264 support
  H.264: fix CODEC_FLAG_GRAY
  4:4:4 H.264 decoding support
  h264_parser: Fix whitespace after previous change.
  h264_parser: Fix behaviour when PARSER_FLAG_COMPLETE_FRAMES is set.
  wav: remove an invalid free().
  lavf: initialise reference_dts in av_estimate_timings_from_pts.
  h264: don't be so picky on decoding pps in extradata.
  avcodec.h: add or elaborate on some documentation comments.
  h264: change a few comments into error messages
  ac3dec: fix doxy-style for comment ("///>" should be "///<" instead).
  img2: add .dpx to the list of supported file extensions.
  ffv1: fix undefined behavior with insane widths.
  ARM: jrevdct_arm: simplify stack usage
  ...

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agomatroskadec: properly decode color space in an endian neutral way
Aurelien Jacobs [Tue, 14 Jun 2011 00:00:06 +0000 (02:00 +0200)]
matroskadec: properly decode color space in an endian neutral way