Mans Rullgard [Sun, 9 Sep 2012 18:25:43 +0000 (19:25 +0100)]
ARM: libavresample: NEON optimised flat float to s16 conversion
Signed-off-by: Mans Rullgard <mans@mansr.com>
Martin Storsjö [Thu, 13 Sep 2012 11:26:10 +0000 (14:26 +0300)]
os_support: Choose between direct.h and io.h using a configure check
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Thu, 13 Sep 2012 09:34:17 +0000 (12:34 +0300)]
os_support: Include io.h instead of direct.h on mingw32ce
Windows CE doesn't have neither mkdir nor _mkdir officially (only
CreateDirectoryW), but mingw32ce has compat wrappers with these names
(declared in io.h since direct.h is unavailable).
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Thu, 13 Sep 2012 08:18:25 +0000 (11:18 +0300)]
x86: ac3dsp: Only refer to the ac3_downmix_sse symbol if it has been declared
This fixes building without inline assembly.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Tue, 11 Sep 2012 20:11:25 +0000 (22:11 +0200)]
swscale: Remove two bogus asserts
Mans Rullgard [Tue, 11 Sep 2012 16:55:11 +0000 (17:55 +0100)]
ac3: move ac3_downmix() from dsputil to ac3dsp
Signed-off-by: Mans Rullgard <mans@mansr.com>
Tim Walker [Wed, 12 Sep 2012 15:09:33 +0000 (17:09 +0200)]
lavr/audio_mix_matrix: acknowledge the existence of LFE2.
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
Tim Walker [Wed, 12 Sep 2012 15:09:32 +0000 (17:09 +0200)]
mlp_parser: avoid mapping multiple disctinct TrueHD channels to the same Libav channel.
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
Tim Walker [Wed, 12 Sep 2012 15:09:31 +0000 (17:09 +0200)]
lavu/audioconvert: add a second low frequency channel.
Can be used by DTS-HD, TrueHD and E-AC-3, among others.
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
Christian Schmidt [Thu, 6 Sep 2012 14:43:24 +0000 (16:43 +0200)]
mpegts: check substreams before discarding
Allow to extract the AC3 core from TrueHD with the "copy" codec.
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Martin Storsjö [Mon, 19 Mar 2012 21:12:54 +0000 (23:12 +0200)]
Add a smooth streaming segmenter muxer
This muxer splits the output from the ismv muxer into individual
files, in realtime.
The same can also be done by the standalone tool ismindex, but this
muxer is needed for doing it in realtime (especially for live
streams that need extra handling for updating the lookahead fields
in the fragment headers).
Using this muxer, one can deliver live smooth streaming from a
normal static file web server. (Using ismindex, one can deliver
premade smooth streaming files from a static file web server,
or prepare files for serving with IIS.)
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Sun, 9 Sep 2012 19:42:56 +0000 (22:42 +0300)]
file: Add an avoption for disabling truncating existing files on open
Signed-off-by: Martin Storsjö <martin@martin.st>
Tomas Härdin [Fri, 7 Sep 2012 11:28:48 +0000 (13:28 +0200)]
img2dec: always close AVIOContexts
Zero sized files would cause the contexts to leak away.
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Martin Storsjö [Tue, 11 Sep 2012 11:42:22 +0000 (14:42 +0300)]
rtpdec_jpeg: Error out on other unsupported type values as well
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 11:41:42 +0000 (14:41 +0300)]
rtpdec_jpeg: Disallow using the reserved q values
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 11:39:58 +0000 (14:39 +0300)]
rtpdec_jpeg: Fold the default qtables case into an existing if statement
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 11:29:54 +0000 (14:29 +0300)]
rtpdec_jpeg: Store and reuse old qtables for q values 128-254
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 11:43:10 +0000 (14:43 +0300)]
rtpdec_jpeg: Simplify the calculation of the number of qtables
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 09:54:06 +0000 (12:54 +0300)]
rtpdec_jpeg: Add more comments about the fields in the SOF0 section
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 09:52:55 +0000 (12:52 +0300)]
rtpdec_jpeg: Clarify where the subsampling magic numbers come from
Write out the numbers the way they are constructed, not just the
final values.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 09:46:44 +0000 (12:46 +0300)]
rtpdec_jpeg: Don't use a bitstream writer for the EOI marker
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 09:44:25 +0000 (12:44 +0300)]
rtpdec_jpeg: Don't needlessly use a bitstream writer for the header
Everything written with this bitstream writer is 8/16 bit units
(except for a pair of 4 bit values), so using a bitstream writer
isn't necessary.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 09:32:37 +0000 (12:32 +0300)]
rtpdec_jpeg: Simplify writing of the jpeg header
Generalize writing of any number of qtables. Don't manually write
16 bit values in two separate calls.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 09:07:56 +0000 (12:07 +0300)]
rtpdec_jpeg: Merge two if statements
This makes the code more readable and robust.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 11 Sep 2012 08:03:05 +0000 (11:03 +0300)]
rtpdec_jpeg: Write the DHT section properly
Currently the size header of the generated DHT section is
incorrect, making the mjpeg decoder just skip it. Since the
written huffman tables are the default ones, this failure had
gone undetected.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Tue, 11 Sep 2012 15:08:05 +0000 (17:08 +0200)]
utvideoenc: Add missing AV_ prefix to codec ID
Diego Biurrun [Tue, 11 Sep 2012 15:07:37 +0000 (17:07 +0200)]
avserver: Add missing #include for av_gettime()
Martin Storsjö [Tue, 11 Sep 2012 08:10:12 +0000 (11:10 +0300)]
configure: Only disable aligned stack on MSVC on 32 bit
On 64 bit, the stack seems to be aligned enough for our needs.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Sun, 19 Aug 2012 19:35:30 +0000 (21:35 +0200)]
configure: indentation cosmetics
Samuel Pitoiset [Mon, 10 Sep 2012 19:15:09 +0000 (21:15 +0200)]
rtpdec_jpeg: Add support for default quantizers
Generate quantization tables when they are not present in the
first chunk.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Wed, 5 Sep 2012 14:42:57 +0000 (16:42 +0200)]
x86: dsputil: Move specific optimization settings out of global init function
They belong in the init functions specific to each CPU capability.
Anton Khirnov [Mon, 10 Sep 2012 12:42:24 +0000 (14:42 +0200)]
avplay: get rid of ugly casts in the options table
Anton Khirnov [Mon, 10 Sep 2012 12:37:46 +0000 (14:37 +0200)]
avplay: fix prototypes for option callbacks.
They have been wrong since
11d957fbd81288e64408e79ed369446346000b29
Anton Khirnov [Sat, 1 Sep 2012 14:20:58 +0000 (16:20 +0200)]
flvdec: always set AVFMTCTX_NOHEADER.
New streams may be created at any time, e.g. on codec change.
Martin Storsjö [Sun, 9 Sep 2012 19:35:50 +0000 (22:35 +0300)]
file: Use a normal private context for storing the file descriptor
Previously the file descriptor was stored in the priv_data pointer.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Mon, 10 Sep 2012 14:35:39 +0000 (17:35 +0300)]
configure: Adjust the xgetbv instrinsic check
The 64 bit cl.exe version 16.00.30319.01 crashes with an internal
compiler error on the current check (and thus deduces it isn't
supported, even if the actual usage in libavuil/x86/cpu.c works
fine), but by assigning the value from the intrinsic to a variable,
or returning it, it works fine.
This error is fixed in cl.exe version 16.00.40219.01.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Tue, 4 Sep 2012 16:23:25 +0000 (18:23 +0200)]
configure: Add --disable-inline-asm command line option
This can come in handy for testing and possibly other purposes.
Martin Storsjö [Mon, 10 Sep 2012 12:34:51 +0000 (15:34 +0300)]
configure: Don't try to enable the log2 function on msvcrt
Some msvcrt versions (the static 64 bit libc in MSVC 10) have
a log2 function, but there is no declaration for it in the headers.
Therefore, the normal configure check might find it, but it can fail
during build or at runtime, depending on whether implicit function
declarations are an error or not.
Therefore simply ignore this function on this platform.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Sun, 9 Sep 2012 19:19:50 +0000 (21:19 +0200)]
x86: dsputil: Only compile motion_est code when encoders are enabled
Mans Rullgard [Sun, 9 Sep 2012 21:47:15 +0000 (22:47 +0100)]
mem: fix typo in check for __ICC
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sun, 9 Sep 2012 15:49:54 +0000 (16:49 +0100)]
fate: mp3: drop redundant CMP setting
Signed-off-by: Mans Rullgard <mans@mansr.com>
Samuel Pitoiset [Fri, 7 Sep 2012 17:30:23 +0000 (19:30 +0200)]
rtp: Depacketization of JPEG (RFC 2435)
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Sun, 9 Sep 2012 13:03:52 +0000 (16:03 +0300)]
Rename ff_put_string to avpriv_put_string
This allows using it from libavformat as well. This will be used
by the RTP/JPEG depacketizer.
Signed-off-by: Martin Storsjö <martin@martin.st>
Samuel Pitoiset [Thu, 6 Sep 2012 13:50:20 +0000 (15:50 +0200)]
mjpeg: Rename some symbols to avpriv_* instead of ff_*
These symbols will be used from the RTP/JPEG depacketizer.
Signed-off-by: Martin Storsjö <martin@martin.st>
Luca Barbato [Sat, 1 Sep 2012 14:04:42 +0000 (16:04 +0200)]
yadif: cosmetics
Martin Storsjö [Sat, 8 Sep 2012 22:43:53 +0000 (01:43 +0300)]
swscale: Provide the right alignment for external mmx asm
This reverts parts of
e0c6cce4472. There is external mmx asm that
requires this alignment.
This fixes crashes when using swscale in builds with external mmx,
without inline assembly.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Wed, 29 Aug 2012 17:01:05 +0000 (19:01 +0200)]
x86: Replace checks for CPU extensions and flags by convenience macros
This separates code relying on inline from that relying on external
assembly and fixes instances where the coalesced check was incorrect.
Mans Rullgard [Fri, 7 Sep 2012 10:13:36 +0000 (11:13 +0100)]
configure: msvc: fix/simplify setting of flags for hostcc
Signed-off-by: Mans Rullgard <mans@mansr.com>
Hendrik Leppkes [Sat, 8 Sep 2012 06:40:19 +0000 (08:40 +0200)]
x86: mlpdsp: mlp_filter_channel_x86 requires inline asm
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Fri, 7 Sep 2012 07:45:16 +0000 (10:45 +0300)]
mov_chan: Only set the channel_layout if setting it to a nonzero value
If regularly parsing new chan atoms (as in rtpdec_qt), but the
chan atoms don't actually contain any channel layout, don't reset
the value that the caller has filled in (by guessing).
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Fri, 7 Sep 2012 07:41:37 +0000 (10:41 +0300)]
mov_chan: Reindent an incorrectly indented line
Signed-off-by: Martin Storsjö <martin@martin.st>
Carl Eugen Hoyos [Sun, 28 Aug 2011 17:43:04 +0000 (19:43 +0200)]
mp2 muxer: mark as AVFMT_NOTIMESTAMPS.
Signed-off-by: Alex Converse <alex.converse@gmail.com>
Justin Ruggles [Fri, 3 Aug 2012 21:38:49 +0000 (17:38 -0400)]
x86: float_dsp: fix ff_vector_fmac_scalar_avx() on Win64
The SWAP macro does not work for explicit xmm/ymm usage, so instead just move
the scalar value from xmm2 to xmm0.
Diego Biurrun [Tue, 4 Sep 2012 06:30:16 +0000 (08:30 +0200)]
x86: more specific checks for availability of required assembly capabilities
Diego Biurrun [Sat, 11 Aug 2012 20:35:27 +0000 (22:35 +0200)]
x86: avcodec: Drop silly "_mmx" suffix from dsputil template names
Diego Biurrun [Sun, 26 Aug 2012 11:02:40 +0000 (13:02 +0200)]
fate: Drop redundant setting of FUZZ to 1
Mans Rullgard [Tue, 4 Sep 2012 12:52:01 +0000 (13:52 +0100)]
cavsdsp: set idct permutation independently of dsputil
CAVS uses its own idct so using dsputil to set the permutation
is fragile.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Mans Rullgard [Sun, 2 Sep 2012 15:05:56 +0000 (16:05 +0100)]
x86: allow using add_hfyu_median_prediction_cmov on any cpu with cmov
For some reason add_hfyu_median_prediction_cmov is only selected
on 3Dnow-capable CPUs, even though it uses no 3Dnow instructions.
This patch allows it to be selected on any cpu with cmov with the
possibility of being overridden by the mmxext version.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Samuel Pitoiset [Thu, 6 Sep 2012 10:46:37 +0000 (12:46 +0200)]
mov_chan: Pass a separate AVIOContext for reading
This fixes crashes when called from rtpdec_qt, where
AVFormatContext->pb is null, a crash present since
3bab7cd128.
Signed-off-by: Martin Storsjö <martin@martin.st>
Anton Khirnov [Tue, 4 Sep 2012 20:01:38 +0000 (22:01 +0200)]
af_asyncts: check return value from lavr when flushing.
Fixes an infinite loop on flush when avresample_get_delay() still
reports some samples but avresample_convert() doesn't return any data.
Alberto Delmás [Mon, 3 Sep 2012 15:32:01 +0000 (17:32 +0200)]
mss2: simplify loop in decode_rle()
It calculates the sum of power of two series, which can be done in one step.
Suggested by Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
Alberto Delmás [Mon, 3 Sep 2012 15:32:01 +0000 (17:32 +0200)]
mss12: avoid unnecessary division in arith*_get_bit()
That division can be replaced with a comparison:
((c->value - c->low) << 1) + 1 >= range
By expanding 'range' definition and simplifying this inequation we obtain
the final expression.
Suggested by Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
Alberto Delmás [Mon, 3 Sep 2012 15:28:01 +0000 (17:28 +0200)]
mss2: do not try to read too many palette entries
Reported by Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
Janne Grunau [Wed, 5 Sep 2012 18:25:48 +0000 (20:25 +0200)]
mpegvideo: set AVFrame fields to NULL after freeing the base memory
Prevents dangling pointers and makes access after free more obvious.
Setting AVFrame.qscale_table to NULL is required for successfully
allocating a previously freed Picture with ff_alloc_picture().
Martin Storsjö [Thu, 6 Sep 2012 11:44:07 +0000 (14:44 +0300)]
configure: Set the right cc_e flags for msvc
The default ones work, but outputs the preprocessed file on stdout
(into config.log).
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Wed, 5 Sep 2012 19:57:40 +0000 (22:57 +0300)]
fate: Allow setting the ld parameter from the config file
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Wed, 5 Sep 2012 14:02:17 +0000 (16:02 +0200)]
x86: dsputil: Do not redundantly check for CPU caps before calling init funcs
The init functions check for CPU capabilities on their own already.
Martin Storsjö [Wed, 5 Sep 2012 18:39:20 +0000 (21:39 +0300)]
configure: Disable some warnings in MSVC
This disables the following warnings:
C4100: unreferenced formal parameter
(1035 occurrances)
C4214: nonstandard extension used : bit field types other than int
(609 occurances)
C4996: 'avpriv_snprintf': This function or variable may be unsafe.
Consider using _snprintf_s instead. To disable deprecation,
use _CRT_SECURE_NO_WARNINGS. See online help for details.
(351 occurrances)
Signed-off-by: Martin Storsjö <martin@martin.st>
Hendrik Leppkes [Wed, 5 Sep 2012 17:43:18 +0000 (19:43 +0200)]
x86: vp56: cmov version of vp56_rac_get_prob requires inline asm
Signed-off-by: Diego Biurrun <diego@biurrun.de>
Michael Niedermayer [Wed, 5 Sep 2012 16:40:21 +0000 (18:40 +0200)]
avopt: fix examples to match the same style about default values as the actual code.
Signed-off-by: Martin Storsjö <martin@martin.st>
Ronald S. Bultje [Fri, 22 Jun 2012 09:24:59 +0000 (12:24 +0300)]
configure: Add support for MSVC cl.exe/link.exe
Signed-off-by: Martin Storsjö <martin@martin.st>
Ronald S. Bultje [Tue, 4 Sep 2012 13:30:37 +0000 (14:30 +0100)]
lavu: add snprintf(), vsnprint() and strtod() replacements for MS runtime.
Signed-off-by: Martin Storsjö <martin@martin.st>
Luca Barbato [Wed, 25 Jul 2012 14:22:34 +0000 (16:22 +0200)]
dsputil: workaround __VA_ARGS__ missing tokenization for MSVC
A second expansion forces the preprocessor to tokenize properly.
Signed-off-by: Martin Storsjö <martin@martin.st>
Mans Rullgard [Tue, 4 Sep 2012 12:57:37 +0000 (13:57 +0100)]
configure: add section for libc-specific hacks
Signed-off-by: Martin Storsjö <martin@martin.st>
Mans Rullgard [Tue, 4 Sep 2012 14:20:45 +0000 (15:20 +0100)]
build: disable ranlib on mingw
Using ranlib is not required but prevents using the libraries with
msvc.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 4 Sep 2012 11:45:00 +0000 (14:45 +0300)]
parser: Don't use pc as context for av_dlog
The ParserContext class doesn't have an AVClass, required for
using it as a logging class.
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Mon, 6 Aug 2012 22:18:59 +0000 (00:18 +0200)]
h264: Remove an assert on current_picture_ptr being null
It is possible in various error paths as well as gap handling
that this has already been allocated. It is not clear why that
would be a problem with the current code, thus disable the
assert to avoid a common assert failure when asserts are enabled.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Tue, 4 Sep 2012 15:01:06 +0000 (17:01 +0200)]
build: eamad: Add missing dependency on mpegvideo code
Diego Biurrun [Tue, 4 Sep 2012 14:57:29 +0000 (16:57 +0200)]
build: utvideoenc: Add missing dependency on Huffman code
Martin Storsjö [Fri, 31 Aug 2012 10:35:47 +0000 (13:35 +0300)]
avopt: Reorder the default_val struct, making i64 the first field
Also mention this change in APIchanges.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Thu, 30 Aug 2012 21:42:30 +0000 (00:42 +0300)]
avopt: Explicitly store rational option defaults in .dbl
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Fri, 31 Aug 2012 10:34:23 +0000 (13:34 +0300)]
avopt: Explicitly store float/double option defaults in .dbl
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Fri, 31 Aug 2012 10:22:31 +0000 (13:22 +0300)]
avopt: Store defaults for AV_OPT_TYPE_INT in the i64 union member
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Fri, 31 Aug 2012 09:52:18 +0000 (12:52 +0300)]
avopt: Store defaults for AV_OPT_TYPE_FLAGS in the i64 union member
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Fri, 31 Aug 2012 09:45:52 +0000 (12:45 +0300)]
avopt: Store defaults for AV_OPT_TYPE_CONST in the i64 union member
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Mon, 6 Aug 2012 14:28:13 +0000 (16:28 +0200)]
svq1enc: Set picture_structure correctly
This fixes assert failures when running in debug mode.
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Mon, 6 Aug 2012 14:49:49 +0000 (16:49 +0200)]
mpegvideo: remove last_picture_ptr / h264 assert.
This assert is no longer true since h264 error concealment needs
last_picture_ptr to be set.
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Mon, 6 Aug 2012 08:26:04 +0000 (10:26 +0200)]
mpegvideo_enc: remove assert that has become obsolete with the new API
It now just checks uninitialized and unused data.
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Sat, 18 Aug 2012 19:53:32 +0000 (21:53 +0200)]
elbg: Fix an assert
It seems the condition was flipped from what was intended.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Elio Pettenò [Mon, 3 Sep 2012 12:20:44 +0000 (05:20 -0700)]
configure: add support for bdver1 and bdver2 CPU types.
Signed-off-by: Diego Elio Pettenò <flameeyes@flameeyes.eu>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Luca Barbato [Tue, 4 Sep 2012 13:04:46 +0000 (15:04 +0200)]
avio: make avio_close NULL the freed buffer
Luca Barbato [Fri, 31 Aug 2012 09:22:20 +0000 (11:22 +0200)]
pixdesc: cosmetics
Martin Storsjö [Tue, 4 Sep 2012 11:57:45 +0000 (14:57 +0300)]
proresenc: Don't free a buffer not owned by the codec
The data in coded_frame isn't allocated using get_buffer, but
is copied from the input frame to the encoder, so we should
not try to free it ourselves.
This fixes an assert failure when running in debug mode.
Signed-off-by: Martin Storsjö <martin@martin.st>
Martin Storsjö [Tue, 4 Sep 2012 11:31:52 +0000 (14:31 +0300)]
proresenc: Write the full value in one put_bits call
Previously, the put_bits call writing the value wrote a value
larger than the number of bits specified, failing asserts
in debug mode. There was no actual bitstream writer corruption,
since the overwritten bit already always was set to 1.
Signed-off-by: Martin Storsjö <martin@martin.st>
Michael Niedermayer [Tue, 4 Sep 2012 11:02:30 +0000 (14:02 +0300)]
adpcmenc: Calculate the IMA_QT predictor without overflow
Previously, the value given to put_bits was 10 bits long for positive
predictors, even though 9 bits were to be written. The extra bit could
in some cases overwrite existing bits in the bitstream writer cache.
This fixes a failed assert in put_bits.h, when running a version
built with -DDEBUG.
The fate test result gets slightly improved, thanks to getting rid
of the overwritten bits in the bitstream writer cache.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diego Biurrun [Wed, 29 Aug 2012 16:53:14 +0000 (18:53 +0200)]
x86: Add convenience macros to check for CPU extensions and flags
Diego Biurrun [Sun, 2 Sep 2012 15:03:41 +0000 (17:03 +0200)]
x86: h264dsp: drop some unnecessary ifdefs around prototype declarations
Alberto Delmás [Fri, 31 Aug 2012 09:24:26 +0000 (11:24 +0200)]
mss12: merge decode_pixel() and decode_top_left_pixel()
No meaningful generated code differences using gcc -O3.
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
Alberto Delmás [Sun, 2 Sep 2012 10:44:21 +0000 (12:44 +0200)]
mss12: reduce SliceContext size from 1067 to 164 KB
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
Alberto Delmás [Sun, 2 Sep 2012 10:42:01 +0000 (12:42 +0200)]
mss12: move SliceContexts out of the common context into the codec contexts
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>