platform/upstream/libav.git
11 years agoconfigure: Mention that icl does not build both static and shared libs
Diego Biurrun [Tue, 10 Sep 2013 10:28:19 +0000 (12:28 +0200)]
configure: Mention that icl does not build both static and shared libs

11 years agog2meet: Allocate cursor buffers large enough to fit the aligned width
Martin Storsjö [Wed, 11 Sep 2013 20:40:12 +0000 (23:40 +0300)]
g2meet: Allocate cursor buffers large enough to fit the aligned width

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoaic: Validate values read from the bitstream
Martin Storsjö [Wed, 11 Sep 2013 20:25:04 +0000 (23:25 +0300)]
aic: Validate values read from the bitstream

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agompc8: Make sure the first stream exists before parsing the seek table
Martin Storsjö [Wed, 11 Sep 2013 19:53:15 +0000 (22:53 +0300)]
mpc8: Make sure the first stream exists before parsing the seek table

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agompc8: Check the seek table size parsed from the bitstream
Martin Storsjö [Wed, 11 Sep 2013 19:47:06 +0000 (22:47 +0300)]
mpc8: Check the seek table size parsed from the bitstream

Limit the size to INT_MAX/2 (for simplicity) to be sure that
size + FF_INPUT_BUFFER_PADDING_SIZE won't overflow.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agozmbvdec: Check the buffer size for uncompressed data
Michael Niedermayer [Sun, 11 Nov 2012 17:08:39 +0000 (18:08 +0100)]
zmbvdec: Check the buffer size for uncompressed data

Also don't pointlessly set the buffer size to 1 after copying
one packet.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoape: Don't allow the seektable to be omitted
Martin Storsjö [Wed, 11 Sep 2013 19:29:33 +0000 (22:29 +0300)]
ape: Don't allow the seektable to be omitted

The seektable is required for filling in ape->frames[i].pos
further down.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoshorten: Break out of loop looking for fmt chunk if none is found
Martin Storsjö [Wed, 11 Sep 2013 19:19:28 +0000 (22:19 +0300)]
shorten: Break out of loop looking for fmt chunk if none is found

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoshorten: Use a checked bytestream reader for the wave header
Martin Storsjö [Wed, 11 Sep 2013 19:17:13 +0000 (22:17 +0300)]
shorten: Use a checked bytestream reader for the wave header

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agosmacker: Free memory properly if the init function fails
Martin Storsjö [Wed, 11 Sep 2013 12:55:18 +0000 (15:55 +0300)]
smacker: Free memory properly if the init function fails

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agosmacker: Make sure we don't fill in huffman codes out of range
Martin Storsjö [Wed, 11 Sep 2013 12:54:20 +0000 (15:54 +0300)]
smacker: Make sure we don't fill in huffman codes out of range

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agosmacker: Check malloc return values
Martin Storsjö [Wed, 11 Sep 2013 12:35:19 +0000 (15:35 +0300)]
smacker: Check malloc return values

Also try to free local allocations on errors.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agosmacker: Avoid integer overflow when allocating packets
Martin Storsjö [Wed, 11 Sep 2013 12:25:13 +0000 (15:25 +0300)]
smacker: Avoid integer overflow when allocating packets

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agosmacker: Don't return packets in unallocated streams
Martin Storsjö [Wed, 11 Sep 2013 12:20:01 +0000 (15:20 +0300)]
smacker: Don't return packets in unallocated streams

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agodsicin: Add some basic sanity checks for fields read from the file
Martin Storsjö [Wed, 11 Sep 2013 12:13:48 +0000 (15:13 +0300)]
dsicin: Add some basic sanity checks for fields read from the file

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoconfigure: Fix wrong reference to user support mailing list
Diego Biurrun [Tue, 10 Sep 2013 10:42:10 +0000 (12:42 +0200)]
configure: Fix wrong reference to user support mailing list

11 years agoDrop pointless directory name prefixes from #includes in the current dir
Diego Biurrun [Tue, 10 Sep 2013 06:30:54 +0000 (08:30 +0200)]
Drop pointless directory name prefixes from #includes in the current dir

11 years agodoc: Drop VDPAU from list of supported codecs
Diego Biurrun [Tue, 10 Sep 2013 06:30:42 +0000 (08:30 +0200)]
doc: Drop VDPAU from list of supported codecs

11 years agofate.sh: Run git-clone quietly
Diego Biurrun [Tue, 10 Sep 2013 06:26:27 +0000 (08:26 +0200)]
fate.sh: Run git-clone quietly

11 years agoavformat: Use av_reallocp_array() where suitable
Alexandra Khirnova [Tue, 10 Sep 2013 09:57:35 +0000 (11:57 +0200)]
avformat: Use av_reallocp_array() where suitable

Signed-off-by: Diego Biurrun <diego@biurrun.de>
11 years agomovenc: Simplify setting the fragmentation flag
Martin Storsjö [Mon, 9 Sep 2013 11:35:09 +0000 (14:35 +0300)]
movenc: Simplify setting the fragmentation flag

This makes sure the faststart vs fragmentation check works as
intended when fragmentation is enabled due to using the ismv mode.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agomovenc: Add a warning message if conflicting options have been specified
Martin Storsjö [Mon, 9 Sep 2013 11:13:55 +0000 (14:13 +0300)]
movenc: Add a warning message if conflicting options have been specified

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agotcp: Explicitly convert a pointer to a boolean integer
Martin Storsjö [Mon, 9 Sep 2013 08:28:14 +0000 (11:28 +0300)]
tcp: Explicitly convert a pointer to a boolean integer

This fixes warnings about making integers from pointers without
a cast, and avoids the theoretical case where the lower 32 bits of
the pointer would all be zero where the implicit cast wouldn't give
the right result.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agolavf: fix the comparison in an overflow check
Anton Khirnov [Wed, 4 Sep 2013 06:55:08 +0000 (08:55 +0200)]
lavf: fix the comparison in an overflow check

CC: libav-stable@libav.org
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
11 years agodv: Add a guard to not overread the ppcm array
Luca Barbato [Mon, 5 Aug 2013 20:15:24 +0000 (22:15 +0200)]
dv: Add a guard to not overread the ppcm array

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
11 years agonuv: check ff_rtjpeg_decode_frame_yuv420 return value
Luca Barbato [Sun, 11 Aug 2013 18:35:40 +0000 (20:35 +0200)]
nuv: check ff_rtjpeg_decode_frame_yuv420 return value

CC: libav-stable@libav.org
11 years agoavisynth: K&R formatting cosmetics
Diego Biurrun [Wed, 4 Sep 2013 18:01:59 +0000 (20:01 +0200)]
avisynth: K&R formatting cosmetics

11 years agoavisynth: Add missing #include for NULL_IF_CONFIG_SMALL
Diego Biurrun [Wed, 4 Sep 2013 18:03:14 +0000 (20:03 +0200)]
avisynth: Add missing #include for NULL_IF_CONFIG_SMALL

11 years agoavcodec: Stop exporting the removed audio_resample* symbols
Diego Biurrun [Thu, 5 Sep 2013 10:05:57 +0000 (12:05 +0200)]
avcodec: Stop exporting the removed audio_resample* symbols

11 years agompeg12enc: K&R formatting cosmetics
Vittorio Giovara [Wed, 4 Sep 2013 15:17:30 +0000 (17:17 +0200)]
mpeg12enc: K&R formatting cosmetics

Signed-off-by: Diego Biurrun <diego@biurrun.de>
11 years agompeg12enc: drop forward declarations
Vittorio Giovara [Wed, 4 Sep 2013 15:17:29 +0000 (17:17 +0200)]
mpeg12enc: drop forward declarations

Signed-off-by: Diego Biurrun <diego@biurrun.de>
11 years agomem: Do not check unsigned values for negative size
Diego Biurrun [Wed, 4 Sep 2013 10:28:01 +0000 (12:28 +0200)]
mem: Do not check unsigned values for negative size

11 years agomem: Improve documentation wording and spelling
Diego Biurrun [Wed, 4 Sep 2013 10:27:09 +0000 (12:27 +0200)]
mem: Improve documentation wording and spelling

11 years agomatroskaenc: Allow chapters to be written in trailer
John Stebbins [Tue, 3 Sep 2013 17:53:34 +0000 (10:53 -0700)]
matroskaenc: Allow chapters to be written in trailer

This allows creation of frame accurate chapter marks from sources like
DVD and BD where the precise chapter location is not known until the
chapter mark has been reached during reading.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
11 years agompegvideo: Avoid 32-bit wrapping of linesize multiplications
Martin Storsjö [Tue, 3 Sep 2013 22:36:51 +0000 (01:36 +0300)]
mpegvideo: Avoid 32-bit wrapping of linesize multiplications

This makes sure that linesize * start_y doesn't overflow, so that
emulated_edge_mc can get back the original value if needed.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agodoc: Describe TB option of setpts filter
Clifford Wolf [Sun, 1 Sep 2013 09:07:17 +0000 (11:07 +0200)]
doc: Describe TB option of setpts filter

Signed-off-by: Diego Biurrun <diego@biurrun.de>
11 years agomem: Document the av_realloc family of functions properly
Luca Barbato [Sun, 1 Sep 2013 17:46:59 +0000 (19:46 +0200)]
mem: Document the av_realloc family of functions properly

realloc() does not accept pointers from memalign().

11 years agobuild: Report an error message when a pc file is not found
Luca Barbato [Tue, 3 Sep 2013 20:15:23 +0000 (22:15 +0200)]
build: Report an error message when a pc file is not found

Ease tracking path problems.

11 years agomjpegb: Detect changing number of planes in interlaced video
Michael Niedermayer [Sat, 10 Mar 2012 21:02:46 +0000 (22:02 +0100)]
mjpegb: Detect changing number of planes in interlaced video

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoalac: Check that the channels fit at the given offset
Martin Storsjö [Tue, 3 Sep 2013 11:16:40 +0000 (14:16 +0300)]
alac: Check that the channels fit at the given offset

The code tries to decode a number of channels at the
offset given by the ff_alac_channel_layout_offsets table.
Even if the number of channels decoded so far doesn't
exceed the total number of channels, we need to check that
we actually can decode that number of channels at this offset
as well.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years ago4xm: Check that the read track value is non-negative
Martin Storsjö [Tue, 3 Sep 2013 10:53:23 +0000 (13:53 +0300)]
4xm: Check that the read track value is non-negative

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agomatroskadec: Check that .lang was allocated and set before reading it
Martin Storsjö [Tue, 3 Sep 2013 09:10:50 +0000 (12:10 +0300)]
matroskadec: Check that .lang was allocated and set before reading it

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoalac: Limit max_samples_per_frame
Martin Storsjö [Tue, 3 Sep 2013 08:54:03 +0000 (11:54 +0300)]
alac: Limit max_samples_per_frame

Otherwise buffer size calculations in allocate_buffers could
overflow later, making the code think a large enough buffer
actually was allocated.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agomovenc: Don't flush after each written packet
Martin Storsjö [Mon, 2 Sep 2013 13:23:17 +0000 (16:23 +0300)]
movenc: Don't flush after each written packet

This should improve write performance quite significantly.
---
Tested with both writing a normal mp4, by using the faststart
feature and writing a fragmented mp4 file; all turn out with the
same md5sum as before.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agocosmetics: Fix ATRAC codec name spelling
Diego Biurrun [Mon, 2 Sep 2013 18:21:49 +0000 (20:21 +0200)]
cosmetics: Fix ATRAC codec name spelling

11 years agoape demuxer: check for EOF in potentially long loops
Anton Khirnov [Sat, 24 Aug 2013 19:30:46 +0000 (21:30 +0200)]
ape demuxer: check for EOF in potentially long loops

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
11 years ago4xm: check that bits per sample is strictly positive
Anton Khirnov [Sat, 24 Aug 2013 19:30:46 +0000 (21:30 +0200)]
4xm: check that bits per sample is strictly positive

Avoids a divide by zero.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
11 years agolavf: avoid integer overflow when estimating bitrate
Anton Khirnov [Sat, 24 Aug 2013 19:30:46 +0000 (21:30 +0200)]
lavf: avoid integer overflow when estimating bitrate

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
11 years agolavf: move a variable declaration to the block where it's used
Anton Khirnov [Sun, 25 Aug 2013 09:53:34 +0000 (11:53 +0200)]
lavf: move a variable declaration to the block where it's used

11 years agopictordec: pass correct context to avpriv_request_sample
Anton Khirnov [Sat, 24 Aug 2013 19:30:46 +0000 (21:30 +0200)]
pictordec: pass correct context to avpriv_request_sample

Fixes invalid reads.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
11 years agopictordec: break out of both decoding loops when y drops below 0
Anton Khirnov [Sat, 24 Aug 2013 19:30:46 +0000 (21:30 +0200)]
pictordec: break out of both decoding loops when y drops below 0

Otherwise picmemset can get called with negative y, resulting in an
invalid write.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
11 years agovcr1: add sanity checks
Anton Khirnov [Sat, 24 Aug 2013 19:30:46 +0000 (21:30 +0200)]
vcr1: add sanity checks

Fixes invalid reads with corrupted files.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
11 years agopcm-dvd: Fix build on big endian
Martin Storsjö [Sat, 31 Aug 2013 15:35:33 +0000 (17:35 +0200)]
pcm-dvd: Fix build on big endian

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoconfigure: Add docdir configuration option
Vittorio Giovara [Wed, 28 Aug 2013 08:30:14 +0000 (10:30 +0200)]
configure: Add docdir configuration option

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
11 years agovf_lut: Constantize
Diego Elio Pettenò [Sat, 31 Aug 2013 10:30:16 +0000 (03:30 -0700)]
vf_lut: Constantize

The pixel format tables are never modified, mark them as constant.

Signed-off-by: Diego Elio Pettenò <flameeyes@flameeyes.eu>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
11 years agopcm_dvd: consolidate pieces from pcm.c and mpeg.c
Christian Schmidt [Fri, 30 Aug 2013 16:15:47 +0000 (18:15 +0200)]
pcm_dvd: consolidate pieces from pcm.c and mpeg.c

Remove the header decoding for PCM audio from mpeg.c and the
20/24bit parts from pcm.c and merge them into a new decoder in
pcm-dvd.c.

The decoder has added support for samples that span multiple
packets and modified 20/24bit group decoding. Both is needed to
decode samples that have been generated with DVD-Lab Pro 2. The
decoding of 16bit PCM and two channel 24bit is identical to
before. No other samples are known to verify the correctness of
the encoding this software does.
The complete list of tested formats is
48kHz/16bit/2-8 channels
48kHz/24bit/2-5 channels
96kHz/16bit/2-4 channels
96kHz/24bit/2 channels

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
11 years agopcm_bluray: cosmetics
Christian Schmidt [Fri, 30 Aug 2013 16:15:46 +0000 (18:15 +0200)]
pcm_bluray: cosmetics

11 years agopcm_bluray: rename pcm-mpeg.c to pcm-bluray.c
Christian Schmidt [Fri, 30 Aug 2013 16:15:44 +0000 (18:15 +0200)]
pcm_bluray: rename pcm-mpeg.c to pcm-bluray.c

The original idea was to collect PCM codecs that could appear in various
MPEG streams in this file. Discussion in IRC lead to the conclusion that
one codec per file would be better and stop the need for #ifdefs.

11 years agoppc: don't return a value from a function declared void
Sean McGovern [Thu, 29 Aug 2013 14:39:31 +0000 (10:39 -0400)]
ppc: don't return a value from a function declared void

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agox86: avcodec: Consistently structure CPU extension initialization
Diego Biurrun [Tue, 20 Aug 2013 13:32:00 +0000 (15:32 +0200)]
x86: avcodec: Consistently structure CPU extension initialization

11 years agox86: avcodec: Use convenience macros to check for CPU flags
Diego Biurrun [Tue, 20 Aug 2013 12:46:58 +0000 (14:46 +0200)]
x86: avcodec: Use convenience macros to check for CPU flags

11 years agox86: Add and use more convenience macros to check CPU extension availability
Diego Biurrun [Tue, 20 Aug 2013 12:39:36 +0000 (14:39 +0200)]
x86: Add and use more convenience macros to check CPU extension availability

11 years agoppc: cosmetics: Consistently format CPU flag detection invocations
Diego Biurrun [Fri, 23 Aug 2013 16:50:53 +0000 (18:50 +0200)]
ppc: cosmetics: Consistently format CPU flag detection invocations

11 years agocosmetics: Place arch initialization calls in alphabetical order
Diego Biurrun [Tue, 20 Aug 2013 16:24:27 +0000 (18:24 +0200)]
cosmetics: Place arch initialization calls in alphabetical order

11 years agoarm: fmtconvert: Split armv6 fmtconvert code off from vfp code
Diego Biurrun [Fri, 23 Aug 2013 18:01:36 +0000 (20:01 +0200)]
arm: fmtconvert: Split armv6 fmtconvert code off from vfp code

11 years agoarm: float_dsp: Propagate cpu_flags to vfp initialization function
Diego Biurrun [Fri, 23 Aug 2013 17:39:21 +0000 (19:39 +0200)]
arm: float_dsp: Propagate cpu_flags to vfp initialization function

11 years agoarm: dcadsp: Move synth filter initialization to dcadsp file
Diego Biurrun [Fri, 23 Aug 2013 16:15:32 +0000 (18:15 +0200)]
arm: dcadsp: Move synth filter initialization to dcadsp file

11 years agoswscale: cosmetics: Drop silly camelCase from swScale function pointer name
Diego Biurrun [Sun, 25 Aug 2013 15:30:05 +0000 (17:30 +0200)]
swscale: cosmetics: Drop silly camelCase from swScale function pointer name

11 years agoswscale: Add some missing av_cold to arch-specific init functions
Diego Biurrun [Tue, 20 Aug 2013 08:52:54 +0000 (10:52 +0200)]
swscale: Add some missing av_cold to arch-specific init functions

11 years agoswscale: consistent names for arch-specific acceleration functions
Diego Biurrun [Tue, 20 Aug 2013 08:48:40 +0000 (10:48 +0200)]
swscale: consistent names for arch-specific acceleration functions

11 years agoswscale: ppc: Hide arch-specific initialization details
Diego Biurrun [Wed, 10 Oct 2012 23:06:04 +0000 (01:06 +0200)]
swscale: ppc: Hide arch-specific initialization details

Also give consistent names to init functions.

11 years agoppc: Add missing AltiVec cpuflag detection invocations
Diego Biurrun [Fri, 23 Aug 2013 16:48:17 +0000 (18:48 +0200)]
ppc: Add missing AltiVec cpuflag detection invocations

11 years agoppc: fdct: Remove vim editor settings comment
Diego Biurrun [Fri, 23 Aug 2013 16:46:49 +0000 (18:46 +0200)]
ppc: fdct: Remove vim editor settings comment

11 years agompegvideo: Replace arch initialization ifdeffery by standard conditionals
Diego Biurrun [Tue, 20 Aug 2013 16:25:04 +0000 (18:25 +0200)]
mpegvideo: Replace arch initialization ifdeffery by standard conditionals

11 years agox86: rv40dsp: Move inline assembly optimizations out of YASM init section
Diego Biurrun [Tue, 20 Aug 2013 13:26:02 +0000 (15:26 +0200)]
x86: rv40dsp: Move inline assembly optimizations out of YASM init section

11 years agoswscale: x86: Hide arch-specific initialization details
Diego Biurrun [Fri, 12 Oct 2012 17:54:45 +0000 (19:54 +0200)]
swscale: x86: Hide arch-specific initialization details

Also give consistent names to init functions.

11 years agodsputil: x86: Hide arch-specific initialization details
Diego Biurrun [Fri, 12 Oct 2012 17:57:25 +0000 (19:57 +0200)]
dsputil: x86: Hide arch-specific initialization details

Also give consistent names to init functions.

11 years agoppc: Add and use convenience macro to check for AltiVec availability
Diego Biurrun [Tue, 20 Aug 2013 14:36:47 +0000 (16:36 +0200)]
ppc: Add and use convenience macro to check for AltiVec availability

11 years agoavutil: Refactor CPU extension availability macros
Diego Biurrun [Tue, 20 Aug 2013 14:36:46 +0000 (16:36 +0200)]
avutil: Refactor CPU extension availability macros

11 years agoavutil: Move internal CPU detection function declarations to private header
Diego Biurrun [Tue, 20 Aug 2013 15:49:01 +0000 (17:49 +0200)]
avutil: Move internal CPU detection function declarations to private header

11 years agoapedec: do not buffer decoded samples over AVPackets
Rafaël Carré [Tue, 27 Aug 2013 15:35:49 +0000 (17:35 +0200)]
apedec: do not buffer decoded samples over AVPackets

Only consume an AVPacket when all the samples have been read.

When the rate of samples output is limited (by the default value
of max_samples), consuming the first packet immediately will cause
timing problems:

- The first packet with PTS 0 will output 4608 samples and be
consumed entirely
- The second packet with PTS 64 will output the remaining samples
(typically, a lot, that's why max_samples exist) until the decoded
samples of the first packet have been exhausted, at which point the
samples of the second packet will be decoded and output when
av_decode_frame is called with the next packet).

That means there's a PTS jump since the first packet is 'decoded'
immediately, which can be seen with avplay or mplayer: the timing
jumps immediately to 6.2s (which is the size of a packet).

Sample: http://streams.videolan.org/issues/6348/Goldwave-MAClib.ape
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
11 years agoconfigure: x86: Fix handling of i686 and cpunop features
Diego Biurrun [Tue, 27 Aug 2013 21:33:57 +0000 (23:33 +0200)]
configure: x86: Fix handling of i686 and cpunop features

The i686 feature really is a CPU feature and should be handled as such.
The cpunop dependency on i686 should be expressed with a standard _deps
declaration instead of a manual test.

11 years agortmpproto: Fix limelight authentication with abbreviated app names
Martin Storsjö [Thu, 15 Aug 2013 10:17:15 +0000 (13:17 +0300)]
rtmpproto: Fix limelight authentication with abbreviated app names

When streaming to limelight, the app name is either a full
"appname/subaccount" or "appname/_definst_". In the latter case,
the app name can be simplified into simply "appname", but the
authentication hashing assumes the /_definst_ still to be present.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agovf_fps: use double constants for default/min/max for start_time
Hendrik Leppkes [Tue, 27 Aug 2013 20:44:14 +0000 (22:44 +0200)]
vf_fps: use double constants for default/min/max for start_time

When using AV_NOPTS_VALUE (which expands to INT64_C(0x8000000000000000))
as union initializer for a double field, the c99 converter needs to
interpret this constant when filling the union initializer, and it is
interpreted as a positive value.

When converting AV_NOPTS_VALUE to a double, MSVC 2010 ends up with
the same positive value as the c99 converter, while MSVC 2012 gets
a negative value.

This results in an infite loop in various FATE tests on MSVC 2012.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agodoc: document the asf demuxer
Luca Barbato [Tue, 27 Aug 2013 14:14:42 +0000 (16:14 +0200)]
doc: document the asf demuxer

And drop a typo from the previous FLV entry.

11 years agovf_interlace: fix file permissions
Vittorio Giovara [Tue, 27 Aug 2013 21:41:39 +0000 (23:41 +0200)]
vf_interlace: fix file permissions

Signed-off-by: Diego Biurrun <diego@biurrun.de>
11 years agoRevert "w32pthread: help compiler figure out undeeded code"
Martin Storsjö [Tue, 27 Aug 2013 11:55:06 +0000 (14:55 +0300)]
Revert "w32pthread: help compiler figure out undeeded code"

This reverts commit 4622f11f9c83db8a2e08408c71ff901826ca652c.

The compiler should be able to do the dead code elimination now
without this when the cond_* names point directly to the real
functions instead of to local function pointers.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agow32threads: Don't use function pointers when linking directly to newer APIs
Martin Storsjö [Tue, 27 Aug 2013 11:48:56 +0000 (14:48 +0300)]
w32threads: Don't use function pointers when linking directly to newer APIs

This reduces the call overhead slightly. More noticeably, it
restores the earlier (unintended?) feature that condition variable
functions work just fine even if w32thread_init() hasn't been called.
This was broken as a side effect of 4622f11f9, if explicitly targeting
Vista+.

This makes w32threading work in VP8 again, if targeting Vista+.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agosmoothstreamingenc: Write to a temp file while updating the manifest
Martin Storsjö [Sat, 17 Aug 2013 15:42:23 +0000 (16:42 +0100)]
smoothstreamingenc: Write to a temp file while updating the manifest

If a client tries to read the file while it's being updated, the client
would get an incomplete manifest. Instead write to a separate temp file
and atomically rename it to replace the previous one.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoac3: Return proper error codes
Luca Barbato [Sat, 27 Jul 2013 08:16:35 +0000 (10:16 +0200)]
ac3: Return proper error codes

11 years agoac3: Clean up the error paths
Luca Barbato [Sun, 28 Jul 2013 11:32:18 +0000 (13:32 +0200)]
ac3: Clean up the error paths

11 years agoac3: Do not clash with normal AVERROR
Luca Barbato [Sun, 28 Jul 2013 11:26:12 +0000 (13:26 +0200)]
ac3: Do not clash with normal AVERROR

The parsing function return AVERROR and AAC_AC3_PARSE_ERROR values,
make sure they are not misunderstood.

11 years agoflv: Fix the help string
Luca Barbato [Tue, 27 Aug 2013 11:04:21 +0000 (13:04 +0200)]
flv: Fix the help string

11 years agodoc: Document the flv demuxer
Luca Barbato [Mon, 26 Aug 2013 14:22:44 +0000 (16:22 +0200)]
doc: Document the flv demuxer

11 years agomatroskaenc: Fix writing TRACKDEFAULTFLAG
John Stebbins [Thu, 22 Aug 2013 15:54:59 +0000 (08:54 -0700)]
matroskaenc: Fix writing TRACKDEFAULTFLAG

The element was only being written when the value == 1.  But the default
value of this element is 1, so this has no useful effect.  This element
needs to be written when the value == 0.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
11 years agomovenc: Check for errors from mov_create_chapter_track
Martin Storsjö [Mon, 26 Aug 2013 07:45:27 +0000 (10:45 +0300)]
movenc: Check for errors from mov_create_chapter_track

On failures in the write_trailer function, we could also ignore
the errors and try to finish the file despite these errors (which
would only leave an incomplete chapters track). It's probably better
to signal the error clearly to the caller though (and if this
function failed there's no guarantee that there's enough memory to
finish the trailer either).

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agomovenc: Properly free allocated data on failures in mov_write_header
Martin Storsjö [Mon, 26 Aug 2013 07:59:46 +0000 (10:59 +0300)]
movenc: Properly free allocated data on failures in mov_write_header

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agomovenc: Check that tracks->enc exists before trying to free extradata
Martin Storsjö [Mon, 26 Aug 2013 07:56:17 +0000 (10:56 +0300)]
movenc: Check that tracks->enc exists before trying to free extradata

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agomovenc: Check for allocation failures in mov_create_chapter_track
Martin Storsjö [Sun, 25 Aug 2013 18:23:58 +0000 (21:23 +0300)]
movenc: Check for allocation failures in mov_create_chapter_track

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agoimc: support mode 1
Kostya Shishkov [Thu, 11 Jul 2013 05:04:42 +0000 (07:04 +0200)]
imc: support mode 1