OSDN Git Service
Michael Niedermayer [Tue, 3 Jul 2012 03:24:26 +0000 (05:24 +0200)]
jvdemux: correct size for truncated packets
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Paul B Mahol [Tue, 3 Jul 2012 00:52:06 +0000 (00:52 +0000)]
fate: add test for Bitmap Brothers JV demuxer
Signed-off-by: Paul B Mahol <onemda@gmail.com>
Paul B Mahol [Tue, 3 Jul 2012 00:44:55 +0000 (00:44 +0000)]
fate: add test for Bitmap Brothers JV video
Signed-off-by: Paul B Mahol <onemda@gmail.com>
Michael Niedermayer [Tue, 3 Jul 2012 01:21:24 +0000 (03:21 +0200)]
h264pred: assert that depth is supported
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Tue, 3 Jul 2012 01:20:51 +0000 (03:20 +0200)]
h264dsp: assert that depth is supported
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Tue, 3 Jul 2012 01:19:55 +0000 (03:19 +0200)]
dsputil: assert that depth is supported
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Stefano Sabatini [Mon, 25 Jun 2012 22:41:33 +0000 (00:41 +0200)]
lavfi: remove vsrc_buffer.h API
The vsrc_buffer.h is not installed since commit
0b3b95813576, so there is
no point into keeping it around.
The buffersrc.h API, which is more powerful, should be used instead.
Stefano Sabatini [Mon, 25 Jun 2012 16:56:46 +0000 (18:56 +0200)]
lavc/rawdec: add assertion check in raw_decode()
Check on an implicit assumption done on the number of coded bits for
paletted rawvideo with coded bits == 2.
Stefano Sabatini [Mon, 25 Jun 2012 15:59:01 +0000 (17:59 +0200)]
lavc/rawdec: prefer constant AVPALETTE_SIZE over 256*4
Improve readability.
Michael Niedermayer [Mon, 2 Jul 2012 22:40:26 +0000 (00:40 +0200)]
h264: sps: handle more profiles
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 22:36:08 +0000 (00:36 +0200)]
h264: add support for constraint_set4_flag / constraint_set5_flag
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 21:49:54 +0000 (23:49 +0200)]
x86/gradfun: fix compilation failure on open solaris
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 21:43:06 +0000 (23:43 +0200)]
Revert "alsdec: remove dead assignments"
This reverts commit
4ca6d206d1b5beea42c4290d2ee801aaf5cd31f0.
The assignment is not dead, this should fix fate failures on BSD
Michael Niedermayer [Mon, 2 Jul 2012 21:23:33 +0000 (23:23 +0200)]
has_decode_delay_been_guessed: tighten up the heuristic.
This adds the minimum delay needed with the current decoder to
recognize the reorder buffer size for the reference bitstreams.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 21:18:32 +0000 (23:18 +0200)]
has_decode_delay_been_guessed: skip guessing when sps.num_reorder_frames is available
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 21:18:04 +0000 (23:18 +0200)]
fate: drop strict 1 for MR4_TANDBERG_C.264 and MR5_TANDBERG_C.264
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 21:16:59 +0000 (23:16 +0200)]
h264: add avpriv_h264_has_num_reorder_frames()
This function exports the exact sps.num_reorder_frames value
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 20:59:42 +0000 (22:59 +0200)]
has_decode_delay_been_guessed: improve heuristic
This allows MR4_TANDBERG_C.264 and MR5_TANDBERG_C.264 to be decoded without -strict 1
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 19:24:47 +0000 (21:24 +0200)]
doc/developer: refer to av_malloc() instead of malloc()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 19:19:41 +0000 (21:19 +0200)]
Merge remote-tracking branch 'qatar/master'
* qatar/master:
asfdec: read attached pictures.
apetag: reindent
apetag: export attached covers as video streams.
apetag: fix the amount of data read from binary tags.
apetag: make sure avio_get_str() doesn't read more than it should.
mov: read itunes cover art.
snow: remove VLA in mc_block()
intfloat: Don't use designated initializers in the public headers
snow: remove a VLA.
doc: Remind devs to check return values, especially for malloc() et al
MS ATC Screen (aka MSS3) decoder
vf_yadif: move x86 init code to x86/yadif.c
vf_gradfun: move x86 init code to x86/gradfun.c
roqvideo: Remove a totally unused dspcontext
smacker: remove some unused code
dsicin: remove dead assignment
aacdec: remove dead assignment
rl2: remove dead assignment
proresenc: make a variable local to the loop where it is used
alsdec: remove dead assignments
Conflicts:
Changelog
doc/developer.texi
libavcodec/allcodecs.c
libavcodec/avcodec.h
libavcodec/version.h
libavfilter/gradfun.h
libavfilter/x86/gradfun.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Clément Bœsch [Mon, 2 Jul 2012 17:31:35 +0000 (19:31 +0200)]
mem: heap memory poisoning.
Enable it by default with FATE.
limitation: not random, and not supported with realloc.
Michael Niedermayer [Mon, 2 Jul 2012 18:52:36 +0000 (20:52 +0200)]
vc1dsp: use av_assert2
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 18:52:22 +0000 (20:52 +0200)]
resample2: use av_assert()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 15:33:27 +0000 (17:33 +0200)]
swr: fix warning: passing argument 2 of s->mix_any_f from incompatible pointer type
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 15:32:23 +0000 (17:32 +0200)]
swr: fix warning: passing argument 1 of s->mix_any_f from incompatible pointer type
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 15:31:16 +0000 (17:31 +0200)]
swr: fix warning: passing argument 2 of ctx->simd_f from incompatible pointer type
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 2 Jul 2012 15:23:22 +0000 (17:23 +0200)]
swr: fix mix* related function pointer warnings
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Anton Khirnov [Thu, 21 Jun 2012 20:17:43 +0000 (22:17 +0200)]
asfdec: read attached pictures.
Anton Khirnov [Thu, 21 Jun 2012 17:02:40 +0000 (19:02 +0200)]
apetag: reindent
Anton Khirnov [Tue, 26 Jun 2012 16:58:39 +0000 (18:58 +0200)]
apetag: export attached covers as video streams.
Anton Khirnov [Tue, 26 Jun 2012 16:54:00 +0000 (18:54 +0200)]
apetag: fix the amount of data read from binary tags.
Substract the filename size from the data size.
Anton Khirnov [Tue, 26 Jun 2012 16:49:04 +0000 (18:49 +0200)]
apetag: make sure avio_get_str() doesn't read more than it should.
Anton Khirnov [Thu, 21 Jun 2012 16:24:27 +0000 (18:24 +0200)]
mov: read itunes cover art.
Clément Bœsch [Mon, 2 Jul 2012 06:17:40 +0000 (08:17 +0200)]
lavf/wtvdec: add missing { } around if.
This should fix the current failures spotted by FATE.
Mans Rullgard [Sun, 1 Jul 2012 14:39:22 +0000 (15:39 +0100)]
snow: remove VLA in mc_block()
Signed-off-by: Mans Rullgard <mans@mansr.com>
Ronald S. Bultje [Mon, 2 Jul 2012 07:22:30 +0000 (10:22 +0300)]
intfloat: Don't use designated initializers in the public headers
intfloat.h is a public header, and is now (since
a1245d5ca) included
by mathematics.h, which many external callers include.
This fixes building third party applications that include
mathematics.h in a language that doesn't support designated
initalizers.
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Mon, 2 Jul 2012 10:47:08 +0000 (12:47 +0200)]
wmaenc: dont mess with the bitrate.
The bitrate is not writeable by an encoder.
Fixes generation of invalid wma
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Ronald S. Bultje [Fri, 15 Jun 2012 16:59:57 +0000 (09:59 -0700)]
snow: remove a VLA.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Sat, 30 Jun 2012 13:35:57 +0000 (15:35 +0200)]
doc: Remind devs to check return values, especially for malloc() et al
Kostya Shishkov [Thu, 28 Jun 2012 19:18:23 +0000 (21:18 +0200)]
MS ATC Screen (aka MSS3) decoder
Peter Ross [Mon, 2 Jul 2012 00:28:31 +0000 (10:28 +1000)]
wtvdec: return error when filetime_to_iso8601/crazytime_to_iso8601 conversion fails
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Mans Rullgard [Sun, 1 Jul 2012 12:08:17 +0000 (13:08 +0100)]
vf_yadif: move x86 init code to x86/yadif.c
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sun, 1 Jul 2012 11:51:30 +0000 (12:51 +0100)]
vf_gradfun: move x86 init code to x86/gradfun.c
Signed-off-by: Mans Rullgard <mans@mansr.com>
Michael Niedermayer [Sun, 1 Jul 2012 23:58:30 +0000 (01:58 +0200)]
configure: make fast_unaligned configureable
Fixes Ticket1481
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Martin Storsjö [Sun, 1 Jul 2012 20:46:28 +0000 (23:46 +0300)]
roqvideo: Remove a totally unused dspcontext
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Sun, 1 Jul 2012 20:12:38 +0000 (22:12 +0200)]
Merge remote-tracking branch 'qatar/master'
* qatar/master:
opt/eval: Include mathematics.h for NAN/INFINITY
mathematics: Don't use division by zero in NAN/INFINITY macros
wma: Lower the maximum number of channels to 2
x86: cpu: clean up check for cpuid instruction support
ARM: generate position independent code to access data symbols
Conflicts:
libavutil/opt.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sun, 1 Jul 2012 19:41:06 +0000 (21:41 +0200)]
mov: add workaround for incorrect 0 time point.
Fixes Ticket1471
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Mans Rullgard [Sun, 1 Jul 2012 13:56:16 +0000 (14:56 +0100)]
smacker: remove some unused code
This removes some code apparently left over from vlc reader
debugging.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sun, 1 Jul 2012 13:28:50 +0000 (14:28 +0100)]
dsicin: remove dead assignment
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sun, 1 Jul 2012 12:50:09 +0000 (13:50 +0100)]
aacdec: remove dead assignment
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sun, 1 Jul 2012 12:43:44 +0000 (13:43 +0100)]
rl2: remove dead assignment
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sun, 1 Jul 2012 12:38:12 +0000 (13:38 +0100)]
proresenc: make a variable local to the loop where it is used
This moves the mbs_per_slice declaration inside the only loop
where it is used. Fixes a dead assignment.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sun, 1 Jul 2012 12:36:30 +0000 (13:36 +0100)]
alsdec: remove dead assignments
Signed-off-by: Mans Rullgard <mans@mansr.com>
Michael Niedermayer [Sun, 1 Jul 2012 19:21:07 +0000 (21:21 +0200)]
RELEASE: update, seem this has been forgotten
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Matthieu Bouron [Sat, 30 Jun 2012 17:05:05 +0000 (19:05 +0200)]
lavf/aiffdec: handle ID3 chunk
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sun, 1 Jul 2012 18:29:52 +0000 (20:29 +0200)]
h264: clear the ref lists when the reference are cleared.
Fixes reading of freed arrays (Ticket1498)
Found-by: Piotr Bandurski <ami_stuff@o2.pl>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sun, 1 Jul 2012 17:25:17 +0000 (19:25 +0200)]
h264: check log2_max_poc_lsb for validity
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Stefano Sabatini [Fri, 13 Jan 2012 23:01:05 +0000 (00:01 +0100)]
lavf/segment: rename variable "size" to "list_size"
The new name is more meaningful/less confusing.
Stefano Sabatini [Fri, 13 Jan 2012 22:24:13 +0000 (23:24 +0100)]
lavf/segment: pass the top level chained muxer format context to segment_end()
Consistent with segment_start() and less confusing.
Martin Storsjö [Sat, 30 Jun 2012 15:49:33 +0000 (18:49 +0300)]
opt/eval: Include mathematics.h for NAN/INFINITY
These files use NAN/INFINITY but didn't include mathematics.h to get
the fallback definitions if the system lacks the macros.
Signed-off-by: Martin Storsjö <martin@martin.st>
Ronald S. Bultje [Fri, 22 Jun 2012 10:05:21 +0000 (13:05 +0300)]
mathematics: Don't use division by zero in NAN/INFINITY macros
Some compilers, MSVC among them, don't recognize the divisions by
zero as meaning infinity/nan.
These macros should, according to the standard, expand to constant
expressions, but this shouldn't matter for our usage.
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Sun, 1 Jul 2012 16:04:08 +0000 (18:04 +0200)]
ituh263dec: use av_assert
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sun, 1 Jul 2012 16:02:17 +0000 (18:02 +0200)]
wmaenc: switch all remaining asserts to av_asserts
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Martin Storsjö [Sat, 30 Jun 2012 18:30:28 +0000 (21:30 +0300)]
wma: Lower the maximum number of channels to 2
ff_wma_init is used only by wmadec and wmaenc, and neither of them
can handle more than 2 channels.
This fixes crashes with invalid files.
Based on patch by Piotr Bandurski and Michael Niedermayer.
Signed-off-by: Martin Storsjö <martin@martin.st>
Mans Rullgard [Sun, 24 Jun 2012 11:29:28 +0000 (12:29 +0100)]
x86: cpu: clean up check for cpuid instruction support
This adds macros for accessing the EFLAGS register and uses
these instead of coding the entire check in inline asm.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Fri, 29 Jun 2012 12:35:08 +0000 (13:35 +0100)]
ARM: generate position independent code to access data symbols
This creates proper position independent code when accessing
data symbols if CONFIG_PIC is set.
References to external symbols should now use the movrelx macro.
Some additional code changes are required since this macro may
need a register to hold the GOT pointer.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Michael Niedermayer [Sun, 1 Jul 2012 00:44:22 +0000 (02:44 +0200)]
Merge remote-tracking branch 'cus/stable'
* cus/stable:
ffplay: only configure video filters after we got the first frame
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 30 Jun 2012 20:44:18 +0000 (22:44 +0200)]
Merge remote-tracking branch 'qatar/master'
* qatar/master:
dsputilenc_mmx: split assignment of ff_sse16_sse2 to SSE2 section.
dnxhdenc: add space between function argument type and comment.
x86: fmtconvert: add special asm for float_to_int16_interleave_misc_*
attributes: Add a definition of av_always_inline for MSVC
cmdutils: Pass the actual chosen encoder to filter_codec_opts
os_support: Add fallback definitions for stat flags
os_support: Rename the poll fallback function to ff_poll
network: Check for struct pollfd
os_support: Don't compare a negative number against socket descriptors
os_support: Include all the necessary headers for the win32 open function
x86: vc1: fix and enable optimised loop filter
Conflicts:
cmdutils.c
cmdutils.h
ffmpeg.c
ffplay.c
libavformat/os_support.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Nicolas George [Sat, 30 Jun 2012 19:33:48 +0000 (21:33 +0200)]
ffmpeg: re-set frame_size after filter graph reset.
The filter contexts are completely cleared,
so the frame size can not be kept.
Michael Niedermayer [Sat, 30 Jun 2012 19:27:39 +0000 (21:27 +0200)]
swr: fix MMX resample code, add emms
Fixes Ticket1495
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 30 Jun 2012 18:43:18 +0000 (20:43 +0200)]
avcodec_encode_audio2: print error messages for errors
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 30 Jun 2012 17:47:00 +0000 (19:47 +0200)]
wmaenc: rewrite 2nd stage quantization code
this is faster and more correct
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 30 Jun 2012 17:42:01 +0000 (19:42 +0200)]
wmaenc: remove alternative quantization loop
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 30 Jun 2012 17:38:11 +0000 (19:38 +0200)]
wmaenc: fix off by 1 error in quantization loop
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 30 Jun 2012 17:36:38 +0000 (19:36 +0200)]
wmaenc: change another assert to av_assert
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 30 Jun 2012 16:58:52 +0000 (18:58 +0200)]
nutdec: set duration_estimation_method
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Ronald S. Bultje [Thu, 14 Jun 2012 22:05:10 +0000 (22:05 +0000)]
dsputilenc_mmx: split assignment of ff_sse16_sse2 to SSE2 section.
Ronald S. Bultje [Thu, 14 Jun 2012 22:57:26 +0000 (15:57 -0700)]
dnxhdenc: add space between function argument type and comment.
Signed-off-by: Martin Storsjö <martin@martin.st>
Ronald S. Bultje [Thu, 14 Jun 2012 14:03:08 +0000 (15:03 +0100)]
x86: fmtconvert: add special asm for float_to_int16_interleave_misc_*
This gets rid of a variable-length array and a for loop in C code.
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Sat, 30 Jun 2012 15:19:43 +0000 (17:19 +0200)]
nutdec: estimate duration from last syncpoint
Previously nut used the fallback of estimation by bitrate.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 30 Jun 2012 15:18:46 +0000 (17:18 +0200)]
nutdec: remove assert that is wrong for some new duration code.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 30 Jun 2012 15:01:59 +0000 (17:01 +0200)]
nutdec: check that filesize is valid before using it.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Marton Balint [Thu, 7 Jun 2012 22:16:49 +0000 (00:16 +0200)]
ffplay: only configure video filters after we got the first frame
Otherwise the codec width, height and pixel format values may not be set.
Signed-off-by: Marton Balint <cus@passwd.hu>
Ronald S. Bultje [Sun, 24 Jun 2012 17:57:14 +0000 (20:57 +0300)]
attributes: Add a definition of av_always_inline for MSVC
Signed-off-by: Martin Storsjö <martin@martin.st>
Stefano Sabatini [Fri, 29 Jun 2012 22:02:58 +0000 (00:02 +0200)]
ffplay: factorize expression in audio_decode_frame()
Possibly improve readability.
Martin Storsjö [Fri, 29 Jun 2012 22:28:02 +0000 (01:28 +0300)]
cmdutils: Pass the actual chosen encoder to filter_codec_opts
This allows passing the right options to encoders when there's more
than one encoder for a certain codec id.
Signed-off-by: Martin Storsjö <martin@martin.st>
Ronald S. Bultje [Sun, 24 Jun 2012 18:21:07 +0000 (21:21 +0300)]
os_support: Add fallback definitions for stat flags
Mingw headers provide similar defines already (unconditional #defines,
without any #undef or #ifdef around it), while MSVC doesn't have
them.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Mon, 25 Jun 2012 09:27:37 +0000 (12:27 +0300)]
os_support: Rename the poll fallback function to ff_poll
The fallback function is a non-static function, we shouldn't be
defining non-static functions outside of the proper ff/av prefix
namespaces.
This is especially important for a function like poll, which
other parties (other libraries, or executables linking these
libraries) also might provide similar but incompatible fallbacks for.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Mon, 25 Jun 2012 09:08:44 +0000 (12:08 +0300)]
network: Check for struct pollfd
We need to include winsock2.h here, to make sure we have the
real pollfd struct definition, if one exists, before defining the
fallback poll function.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Sun, 24 Jun 2012 19:36:37 +0000 (22:36 +0300)]
os_support: Don't compare a negative number against socket descriptors
The fds are unsigned integers in the windows definition of struct
sockfds. Due to this, the comparison if (fds[i].fd > n) was always
false.
Signed-off-by: Martin Storsjö <martin@martin.st>
Ronald S. Bultje [Sun, 24 Jun 2012 18:29:14 +0000 (21:29 +0300)]
os_support: Include all the necessary headers for the win32 open function
io.h is required for open and _wopen, and fcntl.h is required for
the O_CREAT flag. On mingw, fcntl.h is included by os_support.h (and
the mingw fcntl.h includes io.h), but include it explicitly here
since this implementation requires it.
Also move the #undef open up. open must not be defined to ff_win32_open
while including the headers that declare the open function. On mingw,
this happened in os_support.h before open was redirected.
Signed-off-by: Martin Storsjö <martin@martin.st>
Nicolas George [Sun, 24 Jun 2012 22:32:49 +0000 (00:32 +0200)]
ffmpeg: use the more powerful version of abuffersink.
Nicolas George [Sun, 24 Jun 2012 22:30:15 +0000 (00:30 +0200)]
sink_buffer: implement fixed frame size.
Nicolas George [Sat, 30 Jun 2012 08:11:22 +0000 (10:11 +0200)]
lavfi: implement samples framing on links.
Links can be set up to group samples into buffers of
specified minimum and maximum size.
Nicolas George [Sat, 30 Jun 2012 10:15:54 +0000 (12:15 +0200)]
fate: add a test for the -force_key_frames option.
It broke twice in the last three months due to other changes.
Nicolas George [Sat, 30 Jun 2012 10:09:24 +0000 (12:09 +0200)]
ffmpeg: fix parsing of -force_key_frames option.
Identical to commit
4f6a1c9 by Reimar Döffinger,
that was lost after the merge of
19ad567.
Clément Bœsch [Sat, 30 Jun 2012 09:21:53 +0000 (11:21 +0200)]
swr: fix make checkheaders.
Clément Bœsch [Sat, 30 Jun 2012 09:18:45 +0000 (11:18 +0200)]
swr: small align cosmetic.
Clément Bœsch [Sat, 30 Jun 2012 09:18:05 +0000 (11:18 +0200)]
swr: use __asm__ instead of __asm.
For consistency only.
Mans Rullgard [Fri, 22 Jun 2012 20:40:28 +0000 (21:40 +0100)]
x86: vc1: fix and enable optimised loop filter
The problem is that the ssse3 psign instruction does the wrong
thing here. Commit
ea60dfe incorrectly removed a macro emulating
this instruction for pre-ssse3 code. However, the emulation is
incorrect, and the code relies on the behaviour of the macro.
Specifically, the psign sets destination elements to zero where
the corresponding source element is zero, whereas the emulation
only negates destination elements where the source is negative.
Furthermore, the PSIGNW_MMX macro in x86util.asm is totally bogus,
which is why the original VC-1 code had an additional right shift
when using it. Since the psign instruction cannot be used here,
skip all the macro hell and use the working instruction sequence
directly.
None of this was noticed due a stray return statement in
ff_vc1dsp_init_mmx() which meant that only the mmx version of the
loop filter was ever used (before being removed in
ea60dfe).
Signed-off-by: Mans Rullgard <mans@mansr.com>