ffmpeg.git
6 years agovf_setdar: make config_props work properly when called multiple times.
Anton Khirnov [Mon, 18 Mar 2013 20:31:54 +0000 (21:31 +0100)]
vf_setdar: make config_props work properly when called multiple times.

Do not overwrite the variable set through AVOptions.

6 years agovf_drawbox: make config_props work properly when called multiple times.
Anton Khirnov [Mon, 18 Mar 2013 20:31:54 +0000 (21:31 +0100)]
vf_drawbox: make config_props work properly when called multiple times.

Do not overwrite the variables set through AVOptions.

6 years agovf_drawtext: do not reset the frame number in config_input.
Anton Khirnov [Mon, 18 Mar 2013 21:07:10 +0000 (22:07 +0100)]
vf_drawtext: do not reset the frame number in config_input.

Frame number should be incremented normally even if the link properties
change.

6 years agovf_fps: move initializing pts from config_props to init.
Anton Khirnov [Mon, 18 Mar 2013 21:32:39 +0000 (22:32 +0100)]
vf_fps: move initializing pts from config_props to init.

It should not be reinitialized if the link properties change.

6 years agovf_frei0r: make config_props work properly when called multiple times.
Anton Khirnov [Mon, 18 Mar 2013 20:31:54 +0000 (21:31 +0100)]
vf_frei0r: make config_props work properly when called multiple times.

Do not leak the initialized filter instance.

6 years agovf_gradfun: make config_props work properly when called multiple times.
Anton Khirnov [Mon, 18 Mar 2013 20:31:54 +0000 (21:31 +0100)]
vf_gradfun: make config_props work properly when called multiple times.

6 years agovf_lut: make config_props work properly when called multiple times.
Anton Khirnov [Mon, 18 Mar 2013 20:31:54 +0000 (21:31 +0100)]
vf_lut: make config_props work properly when called multiple times.

Do not leak the expressions.

6 years agoApple Intermediate Codec decoder
Kostya Shishkov [Thu, 16 May 2013 05:21:08 +0000 (07:21 +0200)]
Apple Intermediate Codec decoder

6 years agomovenc: Don't add base_data_offset in tfhd for PIFF files
Jeremy Hinegardner [Sun, 24 Mar 2013 05:15:31 +0000 (23:15 -0600)]
movenc: Don't add base_data_offset in tfhd for PIFF files

According to the PIFF specification[1] the base_data_offset field MUST be
omitteed. See section 5.2.17. Since the ISMV files created by libavformat
state that they are 'piff' compatible via 'ftyp' box, this needs to be
corrected.

[1] http://www.iis.net/learn/media/smooth-streaming/protected-interoperable-file-format

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Add support for interlaced B-frames
Sebastian Sandberg [Thu, 31 Jan 2013 17:04:22 +0000 (18:04 +0100)]
vc1dec: Add support for interlaced B-frames

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agosparc: Eliminate dead code in VIS acceleration macros
Michael Kostylev [Wed, 15 May 2013 00:25:30 +0000 (20:25 -0400)]
sparc: Eliminate dead code in VIS acceleration macros

Signed-off-by: Diego Biurrun <diego@biurrun.de>
6 years agoflacdec: drop unnecessary assert
Luca Barbato [Wed, 15 May 2013 16:16:18 +0000 (18:16 +0200)]
flacdec: drop unnecessary assert

The condition cannot happen anymore.

6 years agomjpegdec: properly report unsupported disabled features
Luca Barbato [Tue, 14 May 2013 13:27:26 +0000 (15:27 +0200)]
mjpegdec: properly report unsupported disabled features

When JPEG-LS support is disabled the decoder would feed the
data to the JPEG Lossless decode_*_scan function resulting in
faulty decoding.

CC: libav-stable@libav.org
6 years agovf_pad: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_pad: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_overlay: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_overlay: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_lut: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_lut: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_libopencv: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_libopencv: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_hqdn3d: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_hqdn3d: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_hflip: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_hflip: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_gradfun: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_gradfun: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_frei0r: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_frei0r: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_format: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_format: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_fieldorder: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_fieldorder: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_fieldorder: cosmetics, remove an extra linebreak
Anton Khirnov [Mon, 18 Mar 2013 21:26:49 +0000 (22:26 +0100)]
vf_fieldorder: cosmetics, remove an extra linebreak

6 years agovf_fade: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_fade: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_drawtext: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_drawtext: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_drawbox: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_drawbox: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_delogo: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_delogo: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_cropdetect: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_cropdetect: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_crop: cosmetics, break lines
Anton Khirnov [Mon, 18 Mar 2013 20:48:18 +0000 (21:48 +0100)]
vf_crop: cosmetics, break lines

6 years agovf_crop: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_crop: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_boxblur: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_boxblur: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_blackframe: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_blackframe: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agovf_aspect: use the name 's' for the pointer to the private context
Anton Khirnov [Mon, 18 Mar 2013 19:44:36 +0000 (20:44 +0100)]
vf_aspect: use the name 's' for the pointer to the private context

This is shorter and consistent across filters.

6 years agoRemove commented-out debug #define cruft
Diego Biurrun [Wed, 15 May 2013 13:16:12 +0000 (15:16 +0200)]
Remove commented-out debug #define cruft

6 years agojpegls: K&R formatting cosmetics
Luca Barbato [Wed, 15 May 2013 17:03:53 +0000 (19:03 +0200)]
jpegls: K&R formatting cosmetics

Signed-off-by: Diego Biurrun <diego@biurrun.de>
6 years agojpegls: cosmetics: Drop some unnecessary parentheses
Diego Biurrun [Wed, 15 May 2013 17:55:09 +0000 (19:55 +0200)]
jpegls: cosmetics: Drop some unnecessary parentheses

6 years agompegvideo: Remove commented-out PARANOID debug cruft
Diego Biurrun [Wed, 15 May 2013 13:17:08 +0000 (15:17 +0200)]
mpegvideo: Remove commented-out PARANOID debug cruft

6 years agoRemove commented-out #includes
Diego Biurrun [Wed, 15 May 2013 13:09:20 +0000 (15:09 +0200)]
Remove commented-out #includes

6 years agoh263dec: Remove broken and disabled debug cruft
Diego Biurrun [Wed, 15 May 2013 12:49:31 +0000 (14:49 +0200)]
h263dec: Remove broken and disabled debug cruft

6 years agovc1: Reindent INIT_LUT(), align backslashes
Martin Storsjö [Wed, 15 May 2013 08:56:04 +0000 (11:56 +0300)]
vc1: Reindent INIT_LUT(), align backslashes

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agoconfigure: Use linker hardening flags on mingw
Alex Smith [Tue, 14 May 2013 14:05:52 +0000 (10:05 -0400)]
configure: Use linker hardening flags on mingw

This makes it consistent with the msvc builds which automatically set
the DEP and ASLR flags by default.  There really is no good reason why
they shouldn't be set.

The fact that binutils does not set them on by default boggles the mind.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1: Make INIT_LUT() self contained
Michael Niedermayer [Wed, 24 Apr 2013 13:33:11 +0000 (15:33 +0200)]
vc1: Make INIT_LUT() self contained

Move the local variables it uses into the macro, enclosing them
in a do {} while (0) scope.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1: Simplify code by using INIT_LUT()
Michael Niedermayer [Wed, 24 Apr 2013 10:51:28 +0000 (12:51 +0200)]
vc1: Simplify code by using INIT_LUT()

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1: Move INIT_LUT() further up to allow using it in more places
Michael Niedermayer [Wed, 24 Apr 2013 10:51:28 +0000 (12:51 +0200)]
vc1: Move INIT_LUT() further up to allow using it in more places

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Remove some now unused variables
Martin Storsjö [Wed, 15 May 2013 07:32:49 +0000 (10:32 +0300)]
vc1dec: Remove some now unused variables

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Do not allow field_mode to change after the first header
Michael Niedermayer [Thu, 15 Nov 2012 23:48:15 +0000 (00:48 +0100)]
vc1dec: Do not allow field_mode to change after the first header

This fixes out of array accesses.

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1, mss2: Check for any negative return value from ff_vc1_parse_frame_header
Martin Storsjö [Wed, 15 May 2013 08:51:03 +0000 (11:51 +0300)]
vc1, mss2: Check for any negative return value from ff_vc1_parse_frame_header

This is required if we return other error codes than explicitly
-1, which so far has been the only other possible return value
besides 0.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Fix current ptr selection in vc1_mc_4mv_chroma()
Michael Niedermayer [Wed, 24 Apr 2013 13:01:08 +0000 (15:01 +0200)]
vc1dec: Fix current ptr selection in vc1_mc_4mv_chroma()

No sample tried shows a difference.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Factorize picture pointer selection in vc1_mc_4mv_chroma()
Michael Niedermayer [Wed, 24 Apr 2013 12:59:49 +0000 (14:59 +0200)]
vc1dec: Factorize picture pointer selection in vc1_mc_4mv_chroma()

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Factorize picture pointer selection code
Michael Niedermayer [Wed, 24 Apr 2013 12:49:31 +0000 (14:49 +0200)]
vc1dec: Factorize picture pointer selection code

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Factorize srcU/V offseting
Michael Niedermayer [Mon, 12 Nov 2012 00:34:44 +0000 (01:34 +0100)]
vc1dec: Factorize srcU/V offseting

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Fix tff == 0 handling in init_block_index()
Michael Niedermayer [Sat, 20 Apr 2013 00:15:13 +0000 (02:15 +0200)]
vc1dec: Fix tff == 0 handling in init_block_index()

This fixes several files from VLC ticket 5887.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Update destination pointers in context for fields
Michael Niedermayer [Fri, 1 Feb 2013 18:58:26 +0000 (19:58 +0100)]
vc1dec: Update destination pointers in context for fields

This replaces a large number of checks for the second field by
fixing the pointers when they are setup.

This should also fix I/BI field pictures.

Changes checksums for vc1_sa10143, the file becomes slightly closer
to what the reference decoder outputs.

Based on "vc1dec: the second field is written wrong to the picture"
by Sebastian Sandberg <sebastiand.sandberg@gmail.com>.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agopixdesc: rename PIX_FMT_* flags to AV_PIX_FMT_FLAG_*
Anton Khirnov [Sun, 12 May 2013 13:41:49 +0000 (15:41 +0200)]
pixdesc: rename PIX_FMT_* flags to AV_PIX_FMT_FLAG_*

6 years agojpeg2000dec: don't use deprecated PIX_FMT values
Anton Khirnov [Sun, 12 May 2013 13:38:17 +0000 (15:38 +0200)]
jpeg2000dec: don't use deprecated PIX_FMT values

6 years agolavfi: fix compatibility code for old vf_scale options syntax
Anton Khirnov [Tue, 14 May 2013 06:14:21 +0000 (08:14 +0200)]
lavfi: fix compatibility code for old vf_scale options syntax

Currently it would incorrectly trigger on a string that contains a '='
but does not contain a ':', e.g. flags=<flags>.

6 years agoavfiltergraph: simplify inserting conversion filters.
Anton Khirnov [Tue, 14 May 2013 06:16:37 +0000 (08:16 +0200)]
avfiltergraph: simplify inserting conversion filters.

There is now no need to explicitly pass 0:0 as width/height to scale,
those are the defaults.

6 years agoClarify output of av_get_bits_per_pixel
Kieran Kunhya [Tue, 14 May 2013 19:23:40 +0000 (14:23 -0500)]
Clarify output of av_get_bits_per_pixel

Signed-off-by: Anton Khirnov <anton@khirnov.net>
6 years agofate: update the mmf reference
Luca Barbato [Tue, 14 May 2013 16:03:43 +0000 (18:03 +0200)]
fate: update the mmf reference

Using Optional Data chunk introduced changes the hash.

6 years agommf: Write metadata into an Optional Data chunk
Yoshihisa Uchida [Sat, 5 Jan 2013 22:11:09 +0000 (23:11 +0100)]
mmf: Write metadata into an Optional Data chunk

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
6 years agojpeg2000dec: Use correct printf length modifier for pointer differences
Diego Biurrun [Mon, 13 May 2013 19:40:53 +0000 (21:40 +0200)]
jpeg2000dec: Use correct printf length modifier for pointer differences

Also eliminate an unnecessary cast. Fixes the following warning:
jpeg2000dec.c:1200:20: warning: format ‘%lX’ expects argument of type ‘long unsigned int’, but argument 5 has type ‘long long unsigned int’

6 years agovc1dec: Don't apply the loop filter on fields
Martin Storsjö [Tue, 14 May 2013 11:36:00 +0000 (14:36 +0300)]
vc1dec: Don't apply the loop filter on fields

Fixes read of uninitialized memory.

Based on a patch by Michael Niedermayer.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1: implement frame interlaced b-frame header parsing
Hendrik Leppkes [Mon, 23 Jan 2012 21:17:46 +0000 (22:17 +0100)]
vc1: implement frame interlaced b-frame header parsing

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: Don't attempt error concealment on field pictures
Michael Niedermayer [Tue, 24 Jul 2012 02:05:18 +0000 (04:05 +0200)]
vc1dec: Don't attempt error concealment on field pictures

This is not implemented and doesn't work.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agovc1dec: fieldtx is only valid for interlaced frame pictures
Sebastian Sandberg [Thu, 31 Jan 2013 17:04:21 +0000 (18:04 +0100)]
vc1dec: fieldtx is only valid for interlaced frame pictures

The fieldtx_plane is not cleared for interlaced fields.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agoaacenc: Fix erasure of surround channels
Claudio Freire [Sun, 12 May 2013 07:38:40 +0000 (09:38 +0200)]
aacenc: Fix erasure of surround channels

This was due to a miscomputation of s->cur_channel, which led to
psy-based encoders using the psy coefficients for the wrong channel.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agoaacenc: Fix target bitrate for twoloop quantiser search
Claudio Freire [Sat, 4 May 2013 21:36:37 +0000 (18:36 -0300)]
aacenc: Fix target bitrate for twoloop quantiser search

This fixes a case where multichannel bitrate isn't accurately
targetted by psy model alone, never achieving the target bitrate.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agoaacenc: Fix a rounding bug in aacpsy channel bitrate computation
Claudio Freire [Sat, 4 May 2013 21:35:49 +0000 (18:35 -0300)]
aacenc: Fix a rounding bug in aacpsy channel bitrate computation

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agoyadif: correct strides in filter_edges_16bit
James Darnley [Fri, 15 Mar 2013 15:37:33 +0000 (16:37 +0100)]
yadif: correct strides in filter_edges_16bit

The C code treats the data as arrays of uint16_t so strides must not
be in bytes but in pixels.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
6 years agoyadif: restore speed of the C filtering code
James Darnley [Sun, 10 Mar 2013 14:08:50 +0000 (15:08 +0100)]
yadif: restore speed of the C filtering code

Always use the special filter for the first and last 3 columns (only).

Changes made in 64ed397 slowed the filter to just under 3/4 of what it
was.  This commit restores the speed while maintaining identical output.

For reference, on my Athlon64:
1733222 decicycles in old
2358563 decicycles in new
1727558 decicycles in this

Signed-off-by: Anton Khirnov <anton@khirnov.net>
6 years agolavc: Fix msvc compilation of utils.c with -WX (warnings treated as errors)
Matt Wolenetz [Tue, 14 May 2013 02:29:26 +0000 (19:29 -0700)]
lavc: Fix msvc compilation of utils.c with -WX (warnings treated as errors)

This fixes https://bugzilla.libav.org/show_bug.cgi?id=514.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agoconfigure: Disable PIC for cygwin just as for win32/mingw
Martin Storsjö [Mon, 13 May 2013 09:06:40 +0000 (12:06 +0300)]
configure: Disable PIC for cygwin just as for win32/mingw

If this is explicitly disabled for win32/mingw, it should also
be disabled for cygwin, for consistency and for the same reasons
as for win32/mingw.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agox86: dsputil: Rename dsputil_mmx.h --> dsputil_x86.h
Diego Biurrun [Tue, 7 May 2013 23:52:57 +0000 (01:52 +0200)]
x86: dsputil: Rename dsputil_mmx.h --> dsputil_x86.h

The header is not (anymore) MMX-specific.

6 years agox86: dsputil: Split inline assembly from init code
Diego Biurrun [Tue, 7 May 2013 23:45:39 +0000 (01:45 +0200)]
x86: dsputil: Split inline assembly from init code

Also remove some pointless comments.

6 years agox86: dsputil: Refactor pixels16 wrapper functions with a macro
Diego Biurrun [Sat, 20 Apr 2013 18:28:28 +0000 (20:28 +0200)]
x86: dsputil: Refactor pixels16 wrapper functions with a macro

6 years agoconfigure: x86: Only enable cpunop on i686
Diego Biurrun [Fri, 26 Apr 2013 14:49:03 +0000 (16:49 +0200)]
configure: x86: Only enable cpunop on i686

The assembler may insert nopl instructions for cpunop, which are not
(universally) supported on i586 CPUs.

6 years agoconfigure: Rename cmov processor capability to i686
Diego Biurrun [Fri, 26 Apr 2013 14:49:02 +0000 (16:49 +0200)]
configure: Rename cmov processor capability to i686

The goal is to make the capapility slightly more general and have it
cover the availability of the nopl instruction in addition to cmov.

6 years agoconfigure: Explicitly disable PIC when targeting win32/mingw
Martin Storsjö [Tue, 7 May 2013 16:25:03 +0000 (19:25 +0300)]
configure: Explicitly disable PIC when targeting win32/mingw

These platforms do not have any notion of PIC. On some compilers,
enabling pic produces a number of warnings.

This avoids trying to produce PIC loads in the ARM assembly - there
are no relocation types in PE/COFF that correspond to
BFD_RELOC_32_PCREL (R_ARM_REL32 in ELF).

As a side-effect, this avoids enabling PIC on mingw64, getting rid
of the warnings about PIC not having any effect on that platform.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agoconfigure: Use enable_weak when enabling pic
Martin Storsjö [Thu, 9 May 2013 12:39:25 +0000 (15:39 +0300)]
configure: Use enable_weak when enabling pic

This allows it to be overridden, either by the user on the command
line, or by other sections of the configure script.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agoindeo3: avoid writes without necessary alignment in copy_cell()
Janne Grunau [Thu, 9 May 2013 12:53:03 +0000 (14:53 +0200)]
indeo3: avoid writes without necessary alignment in copy_cell()

Cells starting at a position aligned to 8 pixels but wider than
4 blocks are copied with 3 blocks per loop. This creates problems on the
next loop iterations since the routine copying 2 blocks requires the
same alignment on some architectures like ARM NEON.

6 years agox86: sbrdsp: implement SSE2 qmf_pre_shuffle
Christophe Gisquet [Sun, 25 Nov 2012 08:10:36 +0000 (09:10 +0100)]
x86: sbrdsp: implement SSE2 qmf_pre_shuffle

From 253 to 51 cycles on Arrandale and Win64.
44 cycles on SandyBridge.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
6 years agoaf_asyncts: fix offset calculation
Anton Khirnov [Wed, 8 May 2013 19:44:20 +0000 (21:44 +0200)]
af_asyncts: fix offset calculation

delta is in samples, not bytes. Also the sample format is not guaranteed
to be planar.

CC:libav-stable@libav.org

6 years agoavconv: use current syntax for the scale filter
Luca Barbato [Fri, 10 May 2013 00:37:14 +0000 (02:37 +0200)]
avconv: use current syntax for the scale filter

6 years agommf: K&R formatting cosmetics
Luca Barbato [Thu, 9 May 2013 08:37:31 +0000 (10:37 +0200)]
mmf: K&R formatting cosmetics

Signed-off-by: Diego Biurrun <diego@biurrun.de>
6 years agox86: dsputil: Remove unused argument from QPEL_OP macro
Diego Biurrun [Tue, 7 May 2013 23:55:21 +0000 (01:55 +0200)]
x86: dsputil: Remove unused argument from QPEL_OP macro

6 years agox86: vf_yadif: Remove stray dsputil_mmx #include
Diego Biurrun [Tue, 7 May 2013 23:51:31 +0000 (01:51 +0200)]
x86: vf_yadif: Remove stray dsputil_mmx #include

6 years agox86: dsputil: Move TRANSPOSE4 macro to the only place it is used
Diego Biurrun [Tue, 7 May 2013 22:57:57 +0000 (00:57 +0200)]
x86: dsputil: Move TRANSPOSE4 macro to the only place it is used

6 years agox86: dsputil: Move constant declarations into separate header
Diego Biurrun [Tue, 7 May 2013 22:50:17 +0000 (00:50 +0200)]
x86: dsputil: Move constant declarations into separate header

6 years agox86: dsputil: Group all assembly constants together in constants.c
Diego Biurrun [Mon, 6 May 2013 23:49:02 +0000 (01:49 +0200)]
x86: dsputil: Group all assembly constants together in constants.c

6 years agox86: dsputil: Move ff_pd assembly constants to the only place they are used
Diego Biurrun [Mon, 6 May 2013 23:42:08 +0000 (01:42 +0200)]
x86: dsputil: Move ff_pd assembly constants to the only place they are used

6 years agomsvc: Move linker pragma from config.h to libavutil/internal.h
Martin Storsjö [Mon, 6 May 2013 20:08:41 +0000 (23:08 +0300)]
msvc: Move linker pragma from config.h to libavutil/internal.h

This makes linking succeed for tools that include config.h but
don't link to libavutil.

Signed-off-by: Martin Storsjö <martin@martin.st>
6 years agofate-run: Drop unnecessary regtest() indirection
Diego Biurrun [Mon, 6 May 2013 16:35:34 +0000 (18:35 +0200)]
fate-run: Drop unnecessary regtest() indirection

The indirection was used to abstract from the different -regression.sh
scripts. Now that only one remains, this indirection is pointless.

6 years agox86: dsputil: Remove unused ff_pb_3F constant
Diego Biurrun [Mon, 6 May 2013 23:37:03 +0000 (01:37 +0200)]
x86: dsputil: Remove unused ff_pb_3F constant

6 years agox86: dsputil: Remove unused MOVQ_BONE macro
Diego Biurrun [Mon, 6 May 2013 23:34:53 +0000 (01:34 +0200)]
x86: dsputil: Remove unused MOVQ_BONE macro

6 years agox86: dsputil: Move rv40-specific functions where they belong
Diego Biurrun [Sat, 27 Apr 2013 19:30:06 +0000 (21:30 +0200)]
x86: dsputil: Move rv40-specific functions where they belong

6 years agox86: dsputil hpeldsp: Move shared template functions into separate object
Diego Biurrun [Mon, 6 May 2013 22:42:22 +0000 (00:42 +0200)]
x86: dsputil hpeldsp: Move shared template functions into separate object

6 years agox86: rnd_template: Eliminate pointless OP_AVG macro indirection
Diego Biurrun [Mon, 6 May 2013 22:15:41 +0000 (00:15 +0200)]
x86: rnd_template: Eliminate pointless OP_AVG macro indirection

6 years agooma: properly forward errors in oma_read_packet
Luca Barbato [Tue, 7 May 2013 07:39:10 +0000 (09:39 +0200)]
oma: properly forward errors in oma_read_packet

Prevent spurios EIO on EOF.

CC:libav-stable@libav.org

6 years agooma: return meaningful errors
Luca Barbato [Wed, 17 Apr 2013 09:41:54 +0000 (11:41 +0200)]
oma: return meaningful errors