Diego Biurrun [Tue, 28 Aug 2012 15:47:39 +0000 (17:47 +0200)]
x86: Always compile files with functions that are called unconditionally
Diego Biurrun [Tue, 28 Aug 2012 15:45:24 +0000 (17:45 +0200)]
x86: mpegvideoenc: fix linking with --disable-mmx
The optimized dct_quantize template functions reference optimized
fdct symbols, so these functions must only be enabled if the relevant
optimizations have been enabled by configure.
Diego Biurrun [Tue, 28 Aug 2012 15:37:54 +0000 (17:37 +0200)]
x86: mpegvideoenc: Do not abuse HAVE_ variables for template instantiation
This avoids trouble if HAVE_ variables are used elsewhere in the file.
Brad Smith [Mon, 27 Aug 2012 18:57:52 +0000 (14:57 -0400)]
configure: support Bitrig OS
Signed-off-by: Diego Biurrun <diego@biurrun.de>
Ronald S. Bultje [Tue, 28 Aug 2012 00:32:50 +0000 (17:32 -0700)]
yuv2rgb: handle line widths that are not a multiple of 4.
This introduces support for width%4==2 in addition to width%4==0. For
odd widths, some more checks are needed, since the current code always
handles two luma items in a row, thus there is a possibility of an
overread by one.
Martin Storsjö [Tue, 28 Aug 2012 16:35:39 +0000 (19:35 +0300)]
graph2dot: Use the fallback getopt implementation if needed
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 28 Aug 2012 19:54:43 +0000 (22:54 +0300)]
tools: Include io.h for open/read/write/close if unistd.h doesn't exist
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 28 Aug 2012 19:44:31 +0000 (22:44 +0300)]
testprogs: Remove unused includes
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 28 Aug 2012 19:40:27 +0000 (22:40 +0300)]
qt-faststart: Use other seek/tell functions on MSVC than on mingw
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 28 Aug 2012 19:30:36 +0000 (22:30 +0300)]
ismindex: Include direct.h for _mkdir on windows
The Windows SDK in MSVC doesn't have mkdir, only _mkdir, and
MSDN says one should include direct.h to use it.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 28 Aug 2012 16:24:14 +0000 (19:24 +0300)]
sdp: Use static const char arrays instead of pointers to strings
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Wed, 8 Aug 2012 00:12:17 +0000 (02:12 +0200)]
x86: avcodec: Drop silly "_mmx" suffixes from filenames
Diego Biurrun [Tue, 7 Aug 2012 23:49:46 +0000 (01:49 +0200)]
x86: avcodec: Drop silly "_sse" suffixes from filenames
Martin Storsjö [Tue, 28 Aug 2012 11:57:26 +0000 (14:57 +0300)]
sdp: Include profile-level-id for H264
This is required for playback with the Stagefright RTSP framework
on Android.
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Wed, 22 Aug 2012 02:02:38 +0000 (04:02 +0200)]
utvideoenc: use ff_huff_gen_len_table
Avoid code duplication and provide faster and better compression.
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Michael Niedermayer [Mon, 27 Aug 2012 14:00:47 +0000 (16:00 +0200)]
huffman: add ff_huff_gen_len_table
The function will be used by utvideo as well.
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Reimar Döffinger [Mon, 27 Aug 2012 20:11:08 +0000 (20:11 +0000)]
cllc: simplify/fix swapped data buffer allocation.
Using the malloc variant avoids pointless memcpy on size
increase and simplifies handling allocation failure.
Also change code to ensure that allocation, bswap and bitstream
reader all use the same size, even when the packet size is odd
for example.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Samuel Pitoiset [Sun, 26 Aug 2012 15:35:08 +0000 (17:35 +0200)]
rtpdec_h264: Don't set the pixel format
There is no need for this depacketizer to set the pixel format,
the decoder can do that just fine.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 28 Aug 2012 11:27:10 +0000 (14:27 +0300)]
h264: Check that the codec isn't null before accessing it
This fixes crashes introduced by
2e8f3cbcda5, the codec can be null
when called from parsers.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 28 Aug 2012 14:23:16 +0000 (17:23 +0300)]
audio_frame_queue: Define af_queue_log_state before using it
This fixes building with DEBUG defined after the function was made
static and the prototype removed in
d7f9786cbc.
Signed-off-by: Martin Storsjö <martin@martin.st>
Kostya Shishkov [Mon, 27 Aug 2012 17:21:00 +0000 (19:21 +0200)]
vc1: export some functions
This is a preparatory step for the MSS2 decoder which needs to use
the WMV9 decoder to decode some kinds of frames.
From the patch by Alberto Delmás <adelmas@gmail.com>
Mans Rullgard [Mon, 27 Aug 2012 21:19:42 +0000 (22:19 +0100)]
configure: use HOSTCC_C/O in check_host_cc
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Mon, 27 Aug 2012 21:19:06 +0000 (22:19 +0100)]
configure: use AS_O setting in check_as
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Mon, 27 Aug 2012 21:04:25 +0000 (22:04 +0100)]
configure: use LD_O setting in check_ld()
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sat, 25 Aug 2012 11:34:55 +0000 (12:34 +0100)]
Revert "dsputil: make {add/put/put_signed}_pixels_clamped() non-static."
This reverts commit
484a337cd7cd8bb180c4a1bd3321881f1c874a92.
These functions were used in f8bed30 "VC1: merge idct8x8, coeff
adjustments and put_pixels" which was reverted in 18b6a69.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Diego Biurrun [Mon, 27 Aug 2012 19:05:52 +0000 (21:05 +0200)]
build: Restore dependency of acelp_filters.o on celp_math.o
Diego Biurrun [Sun, 26 Aug 2012 09:29:39 +0000 (11:29 +0200)]
celp_math: Replace duplicate ff_dot_productf() by ff_scalarproduct_c()
Diego Biurrun [Sun, 26 Aug 2012 15:27:12 +0000 (17:27 +0200)]
celp_math: Move ff_cos() to the only place it is used
Diego Biurrun [Mon, 27 Aug 2012 03:09:29 +0000 (05:09 +0200)]
build: Use portable abstraction for linker/hostcc output file syntax
Brad Smith [Sun, 26 Aug 2012 20:23:06 +0000 (16:23 -0400)]
configure: Fix shared library creation for OpenBSD
OpenBSD does not expect or require symlinked shared libraries with
just the major version in the name.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
Derek Buitenhuis [Mon, 27 Aug 2012 00:57:08 +0000 (20:57 -0400)]
vp56: Don't use DECLARE_ALIGN on a typedef name
Instead, use it on the first member, since by definition, if
any member is aligned, the whole struct must be, in order to
maintain that alignment.
Fixes compilation with some finicky compilers.
Idea for fix from Måns Rullgård.
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Kostya Shishkov [Sun, 26 Aug 2012 12:16:07 +0000 (14:16 +0200)]
mss1: move code that will be reused by MSS2 decoder into separate file
Kostya Shishkov [Sun, 26 Aug 2012 11:46:57 +0000 (13:46 +0200)]
mss1: merge decode_intra() and decode_inter()
They have most of their code in common.
Martin Storsjö [Mon, 27 Aug 2012 11:07:12 +0000 (14:07 +0300)]
avprobe: Get rid of ugly casts in the options table
This also makes sure the function pointers are stored via the same
union member as they are accessed via.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Mon, 27 Aug 2012 12:55:43 +0000 (15:55 +0300)]
vf_hqdn3d: Remove a duplicate inline declaration
Compilation seems to fail on GCC 3.4 due to this duplicate
declaration.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Mon, 27 Aug 2012 09:33:42 +0000 (12:33 +0300)]
vf_hqdn3d: Don't declare the loop variable within the for loop
This C99 feature is generally not used in the libav codebase,
since it breaks building with some fringe compilers (GCC 2.95,
MSVC).
Signed-off-by: Martin Storsjö <martin@martin.st>
Luca Barbato [Mon, 27 Aug 2012 10:30:13 +0000 (12:30 +0200)]
huffyuv: update to current coding style
Luca Barbato [Mon, 27 Aug 2012 10:29:34 +0000 (12:29 +0200)]
huffman: update to current coding style
Martin Storsjö [Fri, 17 Aug 2012 19:22:17 +0000 (22:22 +0300)]
rtsp: Free the rtpdec context properly
The condition for calling the rtpdec cleanup was broken in
df8cf076c86.
This fixes a memory leak.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Sun, 26 Aug 2012 16:13:59 +0000 (18:13 +0200)]
build: fft: x86: Drop unused YASM-OBJS-FFT- variable
Diego Biurrun [Sun, 26 Aug 2012 10:44:05 +0000 (12:44 +0200)]
audio_frame_queue: Clean up ff_af_queue_log_state debug function
The function is debug-only, so only compile it in debug mode.
Make it static as it has no uses outside of the file.
Change av_log() to av_dlog().
Diego Biurrun [Sun, 3 Jul 2011 14:56:01 +0000 (16:56 +0200)]
dwt: Remove unused code.
Mans Rullgard [Sun, 26 Aug 2012 07:59:41 +0000 (09:59 +0200)]
cavs: convert cavsdata.h to a .c file
Defining tables in header files is ugly and prone to duplication.
Signed-off-by: Diego Biurrun <diego@biurrun.de>
Diego Biurrun [Sun, 26 Aug 2012 07:57:19 +0000 (09:57 +0200)]
cavs: Move inline functions only used in one file out of the header
Diego Biurrun [Sun, 26 Aug 2012 07:19:33 +0000 (09:19 +0200)]
cavs: Move data tables used in only one place to that file
Jan Ekström [Fri, 24 Aug 2012 11:42:41 +0000 (14:42 +0300)]
fate: Add a single symbol Ut Video decoder test
Signed-off-by: Diego Biurrun <diego@biurrun.de>
Loren Merritt [Sun, 26 Aug 2012 10:26:42 +0000 (10:26 +0000)]
vf_hqdn3d: x86 asm
13% faster on penryn, 16% on sandybridge, 15% on bulldozer
Not simd; a compiler should have generated this, but gcc didn't.
Loren Merritt [Sun, 26 Aug 2012 10:26:00 +0000 (10:26 +0000)]
vf_hqdn3d: support 16bit colordepth
Anton Khirnov [Fri, 24 Aug 2012 15:42:46 +0000 (17:42 +0200)]
avconv: prefer user-forced input framerate when choosing output framerate
Mans Rullgard [Sat, 25 Aug 2012 14:57:19 +0000 (15:57 +0100)]
x86: fft: remove unused fft_dispatch* functions
These functions are not used since the yasm conversion.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sat, 25 Aug 2012 10:41:13 +0000 (11:41 +0100)]
avconv: remove unused variable opt_shortest
This was replaced with a per-file value in 3c0df90.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Derek Buitenhuis [Thu, 23 Aug 2012 18:17:35 +0000 (18:17 +0000)]
FATE: Add Canopus Lossless tests
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Derek Buitenhuis [Sat, 25 Aug 2012 02:10:19 +0000 (22:10 -0400)]
cllc: Pad swapped buffer
The bitstream buffer must be padded, or the bitstream reader might
read over the end.
Fixes the following valgrind warning:
Use of uninitialised value of size 8 at 0x591BAE: cllc_decode_frame (cllc.c:166)
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Jan Ekström [Thu, 23 Aug 2012 20:32:06 +0000 (23:32 +0300)]
utvideodec: Fix single symbol mode decoding
Put the zero length check in place of code that was never used
during decoding, as zero-length slices were generally refused
in decode_frame().
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
Mans Rullgard [Thu, 23 Aug 2012 18:47:29 +0000 (19:47 +0100)]
truespeech: drop useless casts
These values already have the correct types.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Thu, 23 Aug 2012 17:48:13 +0000 (18:48 +0100)]
libavcodec: drop bogus dependencies from mpc[78] and qdm2
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Thu, 23 Aug 2012 17:47:49 +0000 (18:47 +0100)]
mpegaudio: move ff_mpa_enwindow to a separate file
This table is used only by mpegaudiodsp and mpegaudioenc. Separating
it allows dropping some dependencies from mpc[78] and qdm2.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Anton Khirnov [Wed, 22 Aug 2012 08:07:54 +0000 (10:07 +0200)]
AVOptions: store defaults for INT64 options in int64 union member.
Double does not have enough precision to represent all int64 numbers
exactly.
Anton Khirnov [Thu, 23 Aug 2012 07:18:06 +0000 (09:18 +0200)]
cmdutils: fix a memleak when specifying an option twice.
Diego Biurrun [Mon, 13 Aug 2012 11:13:57 +0000 (13:13 +0200)]
x86: mpegvideo: more sensible names for optimization file and init function
Diego Biurrun [Mon, 13 Aug 2012 11:11:38 +0000 (13:11 +0200)]
x86: mpegvideoenc: Split optimizations off into a separate file
Diego Biurrun [Mon, 13 Aug 2012 10:20:07 +0000 (12:20 +0200)]
dnxhdenc: x86: more sensible names for optimization file and init function
Diego Biurrun [Sun, 19 Aug 2012 20:53:28 +0000 (22:53 +0200)]
svq1/svq3: Move common code out of SVQ1 decoder-specific file
Jordi Ortiz [Wed, 22 Aug 2012 11:24:34 +0000 (13:24 +0200)]
dirac: add Comments and references to the standard
Signed-off-by: Diego Biurrun <diego@biurrun.de>
Justin Ruggles [Wed, 2 May 2012 22:27:30 +0000 (18:27 -0400)]
lavr: x86: optimized 6-channel flt to fltp conversion
Justin Ruggles [Wed, 2 May 2012 22:26:01 +0000 (18:26 -0400)]
lavr: x86: optimized 2-channel flt to fltp conversion
Justin Ruggles [Wed, 2 May 2012 22:23:20 +0000 (18:23 -0400)]
lavr: x86: optimized 6-channel flt to s16p conversion
Justin Ruggles [Wed, 2 May 2012 22:21:26 +0000 (18:21 -0400)]
lavr: x86: optimized 2-channel flt to s16p conversion
Justin Ruggles [Wed, 2 May 2012 22:03:11 +0000 (18:03 -0400)]
lavr: x86: optimized 6-channel s16 to fltp conversion
Justin Ruggles [Wed, 2 May 2012 21:57:56 +0000 (17:57 -0400)]
lavr: x86: optimized 2-channel s16 to fltp conversion
Justin Ruggles [Wed, 2 May 2012 21:54:56 +0000 (17:54 -0400)]
lavr: x86: optimized 6-channel s16 to s16p conversion
Justin Ruggles [Wed, 2 May 2012 21:53:15 +0000 (17:53 -0400)]
lavr: x86: optimized 2-channel s16 to s16p conversion
Justin Ruggles [Wed, 2 May 2012 20:32:08 +0000 (16:32 -0400)]
lavr: x86: optimized 2-channel fltp to flt conversion
Justin Ruggles [Wed, 2 May 2012 20:27:11 +0000 (16:27 -0400)]
lavr: x86: optimized 6-channel fltp to s16 conversion
Justin Ruggles [Wed, 2 May 2012 19:50:46 +0000 (15:50 -0400)]
lavr: x86: optimized 2-channel fltp to s16 conversion
Justin Ruggles [Wed, 2 May 2012 19:45:29 +0000 (15:45 -0400)]
lavr: x86: optimized 6-channel s16p to flt conversion
Justin Ruggles [Wed, 2 May 2012 19:38:11 +0000 (15:38 -0400)]
lavr: x86: optimized 2-channel s16p to flt conversion
Justin Ruggles [Wed, 2 May 2012 19:36:43 +0000 (15:36 -0400)]
lavr: x86: optimized 6-channel s16p to s16 conversion
Justin Ruggles [Wed, 2 May 2012 19:32:08 +0000 (15:32 -0400)]
lavr: x86: optimized 2-channel s16p to s16 conversion
Anton Khirnov [Wed, 22 Aug 2012 08:37:35 +0000 (10:37 +0200)]
doc/APIchanges: add an entry for codec descriptors.
Also fill in missing dates and hashes.
Justin Ruggles [Mon, 20 Aug 2012 17:35:48 +0000 (13:35 -0400)]
vorbisenc: set AVCodecContext.bit_rate to 0
The Vorbis encoder is always VBR.
Justin Ruggles [Mon, 20 Aug 2012 17:30:44 +0000 (13:30 -0400)]
vorbisenc: fix quality parameter
This generates output with bitrates similar to libvorbis for a given quality
value.
Justin Ruggles [Tue, 21 Aug 2012 16:48:23 +0000 (12:48 -0400)]
FATE: add ALAC encoding tests
Justin Ruggles [Tue, 21 Aug 2012 16:29:11 +0000 (12:29 -0400)]
lpc: fix alignment of windowed samples for odd maximum LPC order
Fixes crash on x86 due to alignment requirements for w_data in
lpc_apply_welch_window_sse2().
Justin Ruggles [Tue, 21 Aug 2012 13:21:56 +0000 (09:21 -0400)]
alacenc: use s16p sample format as input
Justin Ruggles [Tue, 21 Aug 2012 15:27:34 +0000 (11:27 -0400)]
alacenc: remove unneeded sample_fmt check
Justin Ruggles [Tue, 21 Aug 2012 15:29:37 +0000 (11:29 -0400)]
alacenc: fix max_frame_size calculation for the final frame
Justin Ruggles [Mon, 20 Aug 2012 16:18:58 +0000 (12:18 -0400)]
adpcm_swf: Use correct sample offsets when using trellis.
Fixes invalid reads when encoding mono streams when trellis is used.
Samuel Pitoiset [Sun, 19 Aug 2012 11:02:23 +0000 (13:02 +0200)]
rtmp: support strict rtmp servers
In order to send or receive a stream FCPublish, FCSubscribe and _checkbw
are completely optional and often not implemented. releaseStream over a
non-existen stream might report an error instead of being silent.
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Michael Niedermayer [Fri, 10 Aug 2012 13:47:07 +0000 (15:47 +0200)]
mjpegdec: support AVRn interlaced
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Mans Rullgard [Sat, 18 Aug 2012 19:22:10 +0000 (20:22 +0100)]
x86: remove FASTDIV inline asm
GCC 4.3 and later do the right thing with the plain C code. Earlier
versions in 32-bit mode generate one extra instruction, needlessly
zeroing what would be the high half of the shifted value. At least
two gcc configurations miscompile the inline asm in some situations.
In 64-bit mode, all gcc versions generate imul r64, r64 followed by
shr. On Intel i7 and later, this imul is faster 32-bit mul. On
older Intel and all AMD, it is slightly slower. On Atom it is much
slower.
Considering where the FASTDIV macro is used, any overall negative
performance impact of this change should be negligible. If anyone
cares, they should file a bug against gcc and get the instruction
selection fixed.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Diego Biurrun [Mon, 13 Aug 2012 11:02:38 +0000 (13:02 +0200)]
build: x86: Only compile mpegvideo optimizations when necessary
Diego Biurrun [Mon, 20 Aug 2012 22:36:16 +0000 (00:36 +0200)]
configure: Drop fastdiv option
There is no point in having the user disable any fastdiv macros.
Besides the condition implementation was broken and only disabled
the C implementation, but no platform specific assembly versions.
Diego Biurrun [Sun, 19 Aug 2012 21:19:38 +0000 (23:19 +0200)]
build: Make the E-AC-3 encoder select the AC-3 encoder
The E-AC-3 encoder depends on almost all of the code of the AC-3
encoder, so it makes no sense to enable one without the other.
Diego Biurrun [Tue, 21 Aug 2012 22:45:28 +0000 (00:45 +0200)]
fate: flac: Only run tests requiring samples when samples are available
Jordi Ortiz [Tue, 21 Aug 2012 13:09:40 +0000 (15:09 +0200)]
dirac: use meaningful return values
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Mans Rullgard [Mon, 2 Jul 2012 00:59:04 +0000 (01:59 +0100)]
flacdec: simplify sample buffer handling
Pass pointer to sample buffer instead of channel number to various
functions called from decode_subframe(). Also simplify a few
expressions within this function.
Mans Rullgard [Mon, 2 Jul 2012 21:39:34 +0000 (22:39 +0100)]
flacdec: simplify loop in decode_residuals()
Jan Ekström [Mon, 20 Aug 2012 18:16:42 +0000 (21:16 +0300)]
fate: make Ut Video encoder tests use bitexact swscale flags
The failures on various architectures and compilers on the RGB(A)
tests seem to have been because of one-off YCbCr->RGB conversion
results. This should make the conversion results match on most if
not all code paths.
Signed-off-by: Anton Khirnov <anton@khirnov.net>
Diego Biurrun [Mon, 13 Aug 2012 23:51:15 +0000 (01:51 +0200)]
build: amrwb: Drop redundant lsp dependency declaration