ffmpeg.git
8 years agoAllow running make in subdirs again.
Reimar Döffinger [Fri, 24 Jun 2011 07:10:25 +0000 (09:10 +0200)]
Allow running make in subdirs again.

8 years agoRevert "build: move basic rules and variables to main Makefile"
Reimar Döffinger [Fri, 24 Jun 2011 07:02:57 +0000 (09:02 +0200)]
Revert "build: move basic rules and variables to main Makefile"

This reverts commit bb5249244ddb22df1cfb8dee1c369e6cd653d00d.
This is necessary to allow running "make" in subdirs again.

Conflicts:

Makefile

8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Fri, 24 Jun 2011 00:54:12 +0000 (02:54 +0200)]
Merge remote-tracking branch 'qatar/master'

* qatar/master:
  doxygen: Consistently use '@' instead of '\' for Doxygen markup.
  Use av_printf_format to check the usage of printf style functions
  Add av_printf_format, for marking printf style format strings and their parameters
  ARM: enable thumb for Cortex-M* CPUs
  nsvdec: Propagate error values instead of returning 0 in nsv_read_header().
  build: remove SRC_PATH_BARE variable
  build: move basic rules and variables to main Makefile
  build: move special targets to end of main Makefile
  lavdev: improve feedback in case of invalid frame rate/size
  vfwcap: prefer "framerate_q" over "fps" in vfw_read_header()
  v4l2: prefer "framerate_q" over "fps" in v4l2_set_parameters()
  fbdev: prefer "framerate_q" over "fps" in device context
  bktr: prefer "framerate" over "fps" for grab_read_header()
  ALSA: implement channel layout for playback.
  alsa: support unsigned variants of already supported signed formats.
  alsa: add support for more formats.
  ARM: allow building in Thumb2 mode

Conflicts:
common.mak
doc/APIchanges
libavcodec/vdpau.h
libavdevice/alsa-audio-common.c
libavdevice/fbdev.c
libavdevice/libdc1394.c
libavutil/avutil.h

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodoxygen: Consistently use '@' instead of '\' for Doxygen markup.
Diego Biurrun [Thu, 23 Jun 2011 20:41:54 +0000 (22:41 +0200)]
doxygen: Consistently use '@' instead of '\' for Doxygen markup.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
8 years agoUse av_printf_format to check the usage of printf style functions
Martin Storsjö [Fri, 17 Jun 2011 07:31:11 +0000 (10:31 +0300)]
Use av_printf_format to check the usage of printf style functions

This helps catching cases where the format string doesn't
match what is passed in, or injection bugs where user data
is passed in as format string.

Signed-off-by: Martin Storsjö <martin@martin.st>
8 years agoAdd av_printf_format, for marking printf style format strings and their parameters
Martin Storsjö [Fri, 17 Jun 2011 07:30:15 +0000 (10:30 +0300)]
Add av_printf_format, for marking printf style format strings and their parameters

Signed-off-by: Martin Storsjö <martin@martin.st>
8 years agoARM: enable thumb for Cortex-M* CPUs
Mans Rullgard [Thu, 23 Jun 2011 14:42:19 +0000 (15:42 +0100)]
ARM: enable thumb for Cortex-M* CPUs

These CPUs are thumb-only and thus require this option.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agoSupport u8, s8, alaw and mulaw channel reordering in alsa.
Anton Khirnov [Thu, 23 Jun 2011 14:49:51 +0000 (16:49 +0200)]
Support u8, s8, alaw and mulaw channel reordering in alsa.

8 years agonsvdec: Propagate error values instead of returning 0 in nsv_read_header().
Diego Biurrun [Thu, 23 Jun 2011 11:27:21 +0000 (13:27 +0200)]
nsvdec: Propagate error values instead of returning 0 in nsv_read_header().

This eliminates a warning about a set-but-unused variable.

8 years agobuild: remove SRC_PATH_BARE variable
Mans Rullgard [Mon, 6 Jun 2011 21:38:38 +0000 (22:38 +0100)]
build: remove SRC_PATH_BARE variable

Setting SRC_PATH to "." when building in-tree removes the need
for a quoted version of the source path since out-of-tree builds
are not possible if the pathname contains spaces.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agobuild: move basic rules and variables to main Makefile
Mans Rullgard [Sun, 5 Jun 2011 21:29:37 +0000 (22:29 +0100)]
build: move basic rules and variables to main Makefile

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agobuild: move special targets to end of main Makefile
Mans Rullgard [Mon, 6 Jun 2011 14:21:12 +0000 (15:21 +0100)]
build: move special targets to end of main Makefile

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agolavdev: improve feedback in case of invalid frame rate/size
Stefano Sabatini [Fri, 3 Jun 2011 08:50:28 +0000 (10:50 +0200)]
lavdev: improve feedback in case of invalid frame rate/size

Show the invalid string in the error message.

While at it also prefer "Could not" over "Couldn't", plain forms are
preferred over contractions (simplify readability, especially for non
English-savvy people).

Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agovfwcap: prefer "framerate_q" over "fps" in vfw_read_header()
Stefano Sabatini [Fri, 3 Jun 2011 08:41:47 +0000 (10:41 +0200)]
vfwcap: prefer "framerate_q" over "fps" in vfw_read_header()

The variable is used for containing the parsed value of framerate,
using a lexically consistent name eases readability/understanding.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agov4l2: prefer "framerate_q" over "fps" in v4l2_set_parameters()
Stefano Sabatini [Fri, 3 Jun 2011 08:39:54 +0000 (10:39 +0200)]
v4l2: prefer "framerate_q" over "fps" in v4l2_set_parameters()

The variable is used for containing the parsed value of framerate,
using a lexically consistent name eases readability/understanding.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agofbdev: prefer "framerate_q" over "fps" in device context
Stefano Sabatini [Fri, 3 Jun 2011 08:31:42 +0000 (10:31 +0200)]
fbdev: prefer "framerate_q" over "fps" in device context

The variable is used for containing the parsed value of framerate,
using a lexically consistent name eases readability/understanding.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agobktr: prefer "framerate" over "fps" for grab_read_header()
Stefano Sabatini [Fri, 3 Jun 2011 08:54:10 +0000 (10:54 +0200)]
bktr: prefer "framerate" over "fps" for grab_read_header()

The variable is used for containing the parsed value of s1->framerate,
using a lexically consistent name ease readability/understanding.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agoALSA: implement channel layout for playback.
Nicolas George [Thu, 10 Mar 2011 11:10:34 +0000 (12:10 +0100)]
ALSA: implement channel layout for playback.

Currently quad, 5.0, 5.1 and 7.1 are implemented.
Implementing support for other formats/layouts and capture should be
straightforward.

5.0 and 7.1 support by Carl Eugen Hoyos.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agoalsa: support unsigned variants of already supported signed formats.
Carl Eugen Hoyos [Wed, 13 Apr 2011 23:04:18 +0000 (01:04 +0200)]
alsa: support unsigned variants of already supported signed formats.

8 years agoalsa: add support for more formats.
Carl Eugen Hoyos [Mon, 13 Jun 2011 19:16:30 +0000 (21:16 +0200)]
alsa: add support for more formats.

Specifically, f32, f64, s32, s24, a-law and mu-law.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
8 years agoARM: allow building in Thumb2 mode
Mans Rullgard [Tue, 14 Jun 2011 10:29:48 +0000 (11:29 +0100)]
ARM: allow building in Thumb2 mode

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Thu, 23 Jun 2011 02:49:04 +0000 (04:49 +0200)]
Merge remote-tracking branch 'qatar/master'

* qatar/master:
  configure: add --optflags option
  build: move documentation rules to doc/Makefile
  build: move test rules to tests/Makefile
  ac3enc: remove unneeded local variable in asym_quant()
  ac3enc: remove a branch in asym_quant() by doing 2 shifts
  ac3enc: avoid masking output in asym_quant() by using signed values for quantized mantissas.
  H.264: fix 4:4:4 + deblocking + 8x8dct + cavlc + MBAFF
  H.264: fix 4:4:4 + deblocking + MBAFF
  H.264: fix 4:4:4 cropping warning
  H.264: reference the correct SPS in decode_scaling_matrices
  H.264: fix bug in lossless 4:4:4 decoding

Conflicts:
Makefile

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agogeneral.texi: use "the FFmpeg libraries" in place of enumerating them
Stefano Sabatini [Wed, 22 Jun 2011 19:16:40 +0000 (21:16 +0200)]
general.texi: use "the FFmpeg libraries" in place of enumerating them

Avoid an incomplete listing of the libav* libraries, and the need to
update them again.

8 years agogeneral.texi: apply minor formatting fix in the MinGW section
Stefano Sabatini [Wed, 22 Jun 2011 19:17:31 +0000 (21:17 +0200)]
general.texi: apply minor formatting fix in the MinGW section

8 years agoconfigure: add --optflags option
Mans Rullgard [Sun, 19 Jun 2011 17:47:06 +0000 (18:47 +0100)]
configure: add --optflags option

This allows overriding the default optimisation flags selected by
configure.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agobuild: move documentation rules to doc/Makefile
Mans Rullgard [Mon, 6 Jun 2011 17:03:22 +0000 (18:03 +0100)]
build: move documentation rules to doc/Makefile

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agobuild: move test rules to tests/Makefile
Mans Rullgard [Mon, 6 Jun 2011 18:44:29 +0000 (19:44 +0100)]
build: move test rules to tests/Makefile

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agoThis is to match gcc's {athlon64,k8,opteron}-sse3 -march flags.
Alexis Ballier [Wed, 22 Jun 2011 20:25:14 +0000 (22:25 +0200)]
This is to match gcc's {athlon64,k8,opteron}-sse3 -march flags.

8 years agoUse HAVE_TEN_OPERANDS for new decode_significance* functions.
Carl Eugen Hoyos [Wed, 22 Jun 2011 19:42:09 +0000 (21:42 +0200)]
Use HAVE_TEN_OPERANDS for new decode_significance* functions.

8 years agoRevert needless API change in 05e84c95.
Reimar Döffinger [Wed, 22 Jun 2011 16:25:49 +0000 (18:25 +0200)]
Revert needless API change in 05e84c95.

When providing a custom AVIOContex for a AVFMT_NOFILE format
only print a warning instead of erroring out.
This allows the code to work with older MPlayer versions that
just always set pb out of laziness.

Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
8 years agoffplay: use new avformat_open_* API.
Anton Khirnov [Thu, 9 Jun 2011 08:58:23 +0000 (10:58 +0200)]
ffplay: use new avformat_open_* API.

8 years agovf_mp: do not add duplicated pixel formats.
Clément Bœsch [Wed, 22 Jun 2011 08:33:01 +0000 (10:33 +0200)]
vf_mp: do not add duplicated pixel formats.

This avoid a crash with in avfilter_merge_formats() in case one of the
filter formats list has multiple time the same entry.

Thanks to Mina Nagy Zaki for helping figuring out the issue.

8 years agolavfi: minor style fixes in formats.c
Stefano Sabatini [Tue, 21 Jun 2011 08:30:41 +0000 (10:30 +0200)]
lavfi: minor style fixes in formats.c

8 years agoac3enc: remove unneeded local variable in asym_quant()
Justin Ruggles [Tue, 21 Jun 2011 20:49:16 +0000 (16:49 -0400)]
ac3enc: remove unneeded local variable in asym_quant()

8 years agoac3enc: remove a branch in asym_quant() by doing 2 shifts
Justin Ruggles [Tue, 21 Jun 2011 20:14:19 +0000 (16:14 -0400)]
ac3enc: remove a branch in asym_quant() by doing 2 shifts

8 years agoac3enc: avoid masking output in asym_quant() by using signed values for
Justin Ruggles [Tue, 21 Jun 2011 17:39:02 +0000 (13:39 -0400)]
ac3enc: avoid masking output in asym_quant() by using signed values for
quantized mantissas.

8 years agoH.264: fix 4:4:4 + deblocking + 8x8dct + cavlc + MBAFF
Jason Garrett-Glaser [Wed, 22 Jun 2011 09:34:02 +0000 (02:34 -0700)]
H.264: fix 4:4:4 + deblocking + 8x8dct + cavlc + MBAFF

8 years agoH.264: fix 4:4:4 + deblocking + MBAFF
Jason Garrett-Glaser [Wed, 22 Jun 2011 09:05:14 +0000 (02:05 -0700)]
H.264: fix 4:4:4 + deblocking + MBAFF

8 years agoH.264: fix 4:4:4 cropping warning
Jason Garrett-Glaser [Tue, 21 Jun 2011 11:16:33 +0000 (04:16 -0700)]
H.264: fix 4:4:4 cropping warning

8 years agoH.264: reference the correct SPS in decode_scaling_matrices
Jason Garrett-Glaser [Tue, 21 Jun 2011 08:10:37 +0000 (01:10 -0700)]
H.264: reference the correct SPS in decode_scaling_matrices

8 years agoH.264: fix bug in lossless 4:4:4 decoding
Jason Garrett-Glaser [Mon, 20 Jun 2011 22:56:35 +0000 (15:56 -0700)]
H.264: fix bug in lossless 4:4:4 decoding

Coefficient test for i16x16 add_pixels4 assumed luma plane.

8 years agoCrystalHD: Always identify H.264 streams as Annex B.
Philip Langdale [Sat, 18 Jun 2011 17:58:42 +0000 (10:58 -0700)]
CrystalHD: Always identify H.264 streams as Annex B.

Now that we're converting all streams to Annex B format, we
can identify them as such to the hardware.

Signed-off-by: Philip Langdale <philipl@overt.org>
8 years agoCrystalHD: Always send filtered H.264 stream to hardware.
Philip Langdale [Sat, 18 Jun 2011 18:00:37 +0000 (11:00 -0700)]
CrystalHD: Always send filtered H.264 stream to hardware.

As we're now always running mp4 format streams through the annex b
filter, it makes sense to pass the filtered stream down, as
libcrystalhd would be doing the conversion internally anyway.

Signed-off-by: Philip Langdale <philipl@overt.org>
8 years agoCrystalHD: Simplfy extradata handling for non-Annex B content.
Philip Langdale [Mon, 13 Jun 2011 03:13:18 +0000 (20:13 -0700)]
CrystalHD: Simplfy extradata handling for non-Annex B content.

Originally, we needed to restore the original extradata after
initialising the mp4toannexb filter because mplayer would end up
taking two passes through the init sequence for the same stream
and end up miscategorising the stream. This doesn't seem to happen
anymore, making the backup/restore process unnecessary.

Signed-off-by: Philip Langdale <philipl@overt.org>
8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Wed, 22 Jun 2011 03:16:40 +0000 (05:16 +0200)]
Merge remote-tracking branch 'qatar/master'

* qatar/master:
  rawdec: Fix decoding of QT WRAW files.
  configure: report optimization for size separately
  mov: Support Digital Voodoo SD 8 Bit and DTS codec identifiers.
  mov: Support R10g codec identifier.
  riff/img2: Add JPEG 2000 codec IDs.
  riff: Add DAVC fourcc.
  riff: Add M263, XVIX, MMJP, CDV5 fourccs.
  rawvideo: Support auv2 fourcc.
  swscale: Remove unused variable from ff_bfin_get_unscaled_swscale().
  h264: Fix assert that failed to compile with -DDEBUG.
  h264: Add x86 assembly for 10-bit weight/biweight H.264 functions.
  fate: remove output redirections from old regtest scripts

Conflicts:
configure
libavcodec/rawdec.c
libavformat/isom.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agorawdec: Fix decoding of QT WRAW files.
ami_stuff [Tue, 24 May 2011 21:38:01 +0000 (23:38 +0200)]
rawdec: Fix decoding of QT WRAW files.

From some tests it results that:
1. All of the AVI/MOV WRAW files need to be flipped.
2. MOV WRAW files need to use AVI color modes.
3. Assigning PAL8 mode by default to WRAW codec is not correct.

8 years agoconfigure: report optimization for size separately
Mans Rullgard [Sun, 19 Jun 2011 17:41:14 +0000 (18:41 +0100)]
configure: report optimization for size separately

This removes an unsightly override of the 'optimizations' setting
only to make the configure report print 'small' when --enable-small
is used.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agoffplay: do not init SDL audio if -an is specified. n0.8
Clément Bœsch [Tue, 21 Jun 2011 09:25:25 +0000 (11:25 +0200)]
ffplay: do not init SDL audio if -an is specified.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoFix zero-length gnu_printf format string warning.
Clément Bœsch [Wed, 15 Jun 2011 09:24:00 +0000 (11:24 +0200)]
Fix zero-length gnu_printf format string warning.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoA cmp instruction with two constants is invalid, thus "g" constraint
Reimar Döffinger [Tue, 21 Jun 2011 19:41:04 +0000 (21:41 +0200)]
A cmp instruction with two constants is invalid, thus "g" constraint
is not correct but must be "rm" instead.

Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agomov: Support Digital Voodoo SD 8 Bit and DTS codec identifiers.
Carl Eugen Hoyos [Thu, 12 May 2011 23:08:01 +0000 (01:08 +0200)]
mov: Support Digital Voodoo SD 8 Bit and DTS codec identifiers.

8 years agomov: Support R10g codec identifier.
ami_stuff [Thu, 12 May 2011 22:49:29 +0000 (00:49 +0200)]
mov: Support R10g codec identifier.

8 years agoriff/img2: Add JPEG 2000 codec IDs.
Kamil Nowosad [Mon, 23 May 2011 21:13:34 +0000 (23:13 +0200)]
riff/img2: Add JPEG 2000 codec IDs.

8 years agoriff: Add DAVC fourcc.
ami_stuff [Wed, 11 May 2011 20:18:15 +0000 (22:18 +0200)]
riff: Add DAVC fourcc.

This fourcc is used by the "mpegable AVC" codec and files encoded with
this codec decode correctly with our H.264 decoder.

8 years agoriff: Add M263, XVIX, MMJP, CDV5 fourccs.
Carl Eugen Hoyos [Sun, 24 Apr 2011 22:27:03 +0000 (00:27 +0200)]
riff: Add M263, XVIX, MMJP, CDV5 fourccs.

8 years agorawvideo: Support auv2 fourcc.
ami_stuff [Sun, 24 Apr 2011 22:20:27 +0000 (00:20 +0200)]
rawvideo: Support auv2 fourcc.

8 years agorelease_notes: document not fully understood mingw-sdl issue
Michael Niedermayer [Tue, 21 Jun 2011 18:58:39 +0000 (20:58 +0200)]
release_notes: document not fully understood mingw-sdl issue

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoswscale: Remove unused variable from ff_bfin_get_unscaled_swscale().
Diego Biurrun [Wed, 15 Jun 2011 11:45:24 +0000 (13:45 +0200)]
swscale: Remove unused variable from ff_bfin_get_unscaled_swscale().

8 years agoh264: Fix assert that failed to compile with -DDEBUG.
Diego Biurrun [Mon, 20 Jun 2011 23:11:28 +0000 (01:11 +0200)]
h264: Fix assert that failed to compile with -DDEBUG.

The assert referenced a variable that no longer exists since 4:4:4 support.

8 years agorelease_notes: some updates
Michael Niedermayer [Tue, 21 Jun 2011 16:16:16 +0000 (18:16 +0200)]
release_notes: some updates

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agopresets: forgotten libvpx presets
Michael Niedermayer [Tue, 21 Jun 2011 15:52:16 +0000 (17:52 +0200)]
presets: forgotten libvpx presets

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agorelease_notes: fix version
Michael Niedermayer [Tue, 21 Jun 2011 15:08:01 +0000 (17:08 +0200)]
release_notes: fix version

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agorelease_notes: mention more codecs
Michael Niedermayer [Tue, 21 Jun 2011 15:06:35 +0000 (17:06 +0200)]
release_notes: mention more codecs
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agorelease_notes: there will be 2 releases each for one ABI/API.
Michael Niedermayer [Tue, 21 Jun 2011 15:02:14 +0000 (17:02 +0200)]
release_notes: there will be 2 releases each for one ABI/API.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agorelease_notes: suggest git log instead of the poorly maintained APIChanges
Michael Niedermayer [Tue, 21 Jun 2011 14:43:45 +0000 (16:43 +0200)]
release_notes: suggest git log instead of the poorly maintained APIChanges

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agorelease_notes: we do support releases
Michael Niedermayer [Tue, 21 Jun 2011 14:37:11 +0000 (16:37 +0200)]
release_notes: we do support releases

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agobuild system: disable memalign on haiku, its not reliable there.
Michael Niedermayer [Tue, 21 Jun 2011 14:36:23 +0000 (16:36 +0200)]
build system: disable memalign on haiku, its not reliable there.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoh264: Add x86 assembly for 10-bit weight/biweight H.264 functions.
Daniel Kang [Tue, 21 Jun 2011 04:30:23 +0000 (00:30 -0400)]
h264: Add x86 assembly for 10-bit weight/biweight H.264 functions.

Mainly ported from 8-bit H.264 weight/biweight.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
8 years agoffprobe: remove duplicate avformat_alloc_context()
Michael Niedermayer [Tue, 21 Jun 2011 12:38:58 +0000 (14:38 +0200)]
ffprobe: remove duplicate avformat_alloc_context()

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoFix segmentation fault in ffprobe
Gavin Kinsey [Tue, 21 Jun 2011 12:13:37 +0000 (13:13 +0100)]
Fix segmentation fault in ffprobe

8 years agofate: remove output redirections from old regtest scripts
Mans Rullgard [Sun, 19 Jun 2011 15:19:06 +0000 (16:19 +0100)]
fate: remove output redirections from old regtest scripts

All tests are run through the fate-run.sh script which already
sets up redirections.  Using the outputs set there simplifies
things somewhat.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agowma: fix infinite loop
Michael Niedermayer [Tue, 21 Jun 2011 11:17:10 +0000 (13:17 +0200)]
wma: fix infinite loop

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoFix H.264 4:4:4 lossless decoding.
Jason Garrett-Glaser [Tue, 21 Jun 2011 08:45:22 +0000 (10:45 +0200)]
Fix H.264 4:4:4 lossless decoding.

The coefficient test for i16x16 add_pixels4 assumed luma plane.

8 years agolibx264: fix open gop default.
Michael Niedermayer [Tue, 21 Jun 2011 02:45:29 +0000 (04:45 +0200)]
libx264: fix open gop default.
Please use -x264opts to force open gop
This fixes Ticket268

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoavfilter picture pool: double free hotfix
Michael Niedermayer [Tue, 21 Jun 2011 02:24:03 +0000 (04:24 +0200)]
avfilter picture pool: double free hotfix

This fix introduces a small memleak

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoMerge remote-tracking branch 'qatar/master'
Michael Niedermayer [Tue, 21 Jun 2011 01:38:25 +0000 (03:38 +0200)]
Merge remote-tracking branch 'qatar/master'

* qatar/master: (28 commits)
  Replace usages of av_get_bits_per_sample_fmt() with av_get_bytes_per_sample().
  x86: cabac: fix register constraints for 32-bit mode
  cabac: move x86 asm to libavcodec/x86/cabac.h
  x86: h264: cast pointers to intptr_t rather than int
  x86: h264: remove hardcoded edi in decode_significance_8x8_x86()
  x86: h264: remove hardcoded esi in decode_significance[_8x8]_x86()
  x86: h264: remove hardcoded edx in decode_significance[_8x8]_x86()
  x86: h264: remove hardcoded eax in decode_significance[_8x8]_x86()
  x86: cabac: change 'a' constraint to 'r' in get_cabac_inline()
  x86: cabac: remove hardcoded esi in get_cabac_inline()
  x86: cabac: remove hardcoded edx in get_cabac_inline()
  x86: cabac: remove unused macro parameter
  x86: cabac: remove hardcoded ebx in inline asm
  x86: cabac: remove hardcoded struct offsets from inline asm
  cabac: remove inline asm under #if 0
  cabac: remove BRANCHLESS_CABAC_DECODER switch
  cabac: remove #if 0 cascade under never-set #ifdef ARCH_X86_DISABLED
  document libswscale bump
  error_resilience: skip last-MV predictor step if MVs are not available.
  error_resilience: actually add counter when adding a MV predictor.
  ...

Conflicts:
Changelog
libavcodec/error_resilience.c
libavfilter/defaults.c
libavfilter/vf_drawtext.c
libswscale/swscale.h
tests/ref/vsynth1/error
tests/ref/vsynth2/error

Merged-by: Michael Niedermayer <michaelni@gmx.at>
8 years agompegaudio_parser: be less picky on the start position
Michael Niedermayer [Tue, 21 Jun 2011 00:49:20 +0000 (02:49 +0200)]
mpegaudio_parser: be less picky on the start position

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agoppc32: Fix movrel
Kim Nguyen [Mon, 20 Jun 2011 22:18:54 +0000 (00:18 +0200)]
ppc32: Fix movrel

Fixes ticket272

8 years agoReplace usages of av_get_bits_per_sample_fmt() with av_get_bytes_per_sample().
Justin Ruggles [Tue, 7 Jun 2011 17:40:22 +0000 (13:40 -0400)]
Replace usages of av_get_bits_per_sample_fmt() with av_get_bytes_per_sample().

av_get_bits_per_sample_fmt() is deprecated.

8 years agox86: cabac: fix register constraints for 32-bit mode
Mans Rullgard [Mon, 20 Jun 2011 22:10:26 +0000 (23:10 +0100)]
x86: cabac: fix register constraints for 32-bit mode

Some operands need to be accessed in byte mode, which restricts the
available registers in 32-bit mode.  Using the 'q' constraint selects
a suitable register.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agocabac: move x86 asm to libavcodec/x86/cabac.h
Mans Rullgard [Mon, 20 Jun 2011 09:53:41 +0000 (10:53 +0100)]
cabac: move x86 asm to libavcodec/x86/cabac.h

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: h264: cast pointers to intptr_t rather than int
Mans Rullgard [Mon, 20 Jun 2011 09:24:26 +0000 (10:24 +0100)]
x86: h264: cast pointers to intptr_t rather than int

Only the low-order bits are used here so the type is not important,
but this avoids a compiler warning.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: h264: remove hardcoded edi in decode_significance_8x8_x86()
Mans Rullgard [Mon, 20 Jun 2011 09:20:15 +0000 (10:20 +0100)]
x86: h264: remove hardcoded edi in decode_significance_8x8_x86()

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: h264: remove hardcoded esi in decode_significance[_8x8]_x86()
Mans Rullgard [Mon, 20 Jun 2011 09:12:29 +0000 (10:12 +0100)]
x86: h264: remove hardcoded esi in decode_significance[_8x8]_x86()

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: h264: remove hardcoded edx in decode_significance[_8x8]_x86()
Mans Rullgard [Mon, 20 Jun 2011 09:06:07 +0000 (10:06 +0100)]
x86: h264: remove hardcoded edx in decode_significance[_8x8]_x86()

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: h264: remove hardcoded eax in decode_significance[_8x8]_x86()
Mans Rullgard [Mon, 20 Jun 2011 08:23:26 +0000 (09:23 +0100)]
x86: h264: remove hardcoded eax in decode_significance[_8x8]_x86()

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: cabac: change 'a' constraint to 'r' in get_cabac_inline()
Mans Rullgard [Mon, 20 Jun 2011 08:28:19 +0000 (09:28 +0100)]
x86: cabac: change 'a' constraint to 'r' in get_cabac_inline()

Nothing requires this value in %eax.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: cabac: remove hardcoded esi in get_cabac_inline()
Mans Rullgard [Mon, 20 Jun 2011 08:19:27 +0000 (09:19 +0100)]
x86: cabac: remove hardcoded esi in get_cabac_inline()

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: cabac: remove hardcoded edx in get_cabac_inline()
Mans Rullgard [Mon, 20 Jun 2011 02:39:45 +0000 (03:39 +0100)]
x86: cabac: remove hardcoded edx in get_cabac_inline()

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: cabac: remove unused macro parameter
Mans Rullgard [Mon, 20 Jun 2011 02:19:20 +0000 (03:19 +0100)]
x86: cabac: remove unused macro parameter

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: cabac: remove hardcoded ebx in inline asm
Mans Rullgard [Mon, 20 Jun 2011 01:31:53 +0000 (02:31 +0100)]
x86: cabac: remove hardcoded ebx in inline asm

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agox86: cabac: remove hardcoded struct offsets from inline asm
Mans Rullgard [Mon, 20 Jun 2011 00:54:32 +0000 (01:54 +0100)]
x86: cabac: remove hardcoded struct offsets from inline asm

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agocabac: remove inline asm under #if 0
Mans Rullgard [Mon, 20 Jun 2011 00:58:36 +0000 (01:58 +0100)]
cabac: remove inline asm under #if 0

A comment says it's not faster than the C code.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agocabac: remove BRANCHLESS_CABAC_DECODER switch
Mans Rullgard [Mon, 20 Jun 2011 00:49:38 +0000 (01:49 +0100)]
cabac: remove BRANCHLESS_CABAC_DECODER switch

The code does not compile without this set.

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agocabac: remove #if 0 cascade under never-set #ifdef ARCH_X86_DISABLED
Mans Rullgard [Sun, 19 Jun 2011 23:59:35 +0000 (00:59 +0100)]
cabac: remove #if 0 cascade under never-set #ifdef ARCH_X86_DISABLED

Signed-off-by: Mans Rullgard <mans@mansr.com>
8 years agoFix bink audio playback outside of FFmpeg.
Ivan Kalvachev [Thu, 16 Jun 2011 16:35:33 +0000 (19:35 +0300)]
Fix bink audio playback outside of FFmpeg.

There are 2 known Bink audio codecs. Additionally they have
a different flavor if they are found inside Bink-b "BIKb" file.
In order to guess the correct flavor, the demuxer sets the audio
codec_tag to be the same as the file format tag.
This causes problem because same tag is used for both audio codecs.
The hack works in FFmpeg because audio codecs are identified by their
codec_id, but other players rely on standard behavior.

This fix removes the codec_tag hack and instead uses artificial
extradata format to signal the codec flavor. This would also
allow proper embedding of Bink audio in other containers.

Signed-off-by: Ivan Kalvachev <ikalvachev@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
8 years agodocument libswscale bump
Reinhard Tartler [Mon, 20 Jun 2011 20:40:31 +0000 (22:40 +0200)]
document libswscale bump

8 years agoerror_resilience: skip last-MV predictor step if MVs are not available.
Ronald Bultje [Mon, 20 Jun 2011 15:07:56 +0000 (15:07 +0000)]
error_resilience: skip last-MV predictor step if MVs are not available.

Fixes crashes when playing broken MPEG2-TS streams.

8 years agoerror_resilience: actually add counter when adding a MV predictor.
Ronald Bultje [Mon, 20 Jun 2011 15:07:55 +0000 (15:07 +0000)]
error_resilience: actually add counter when adding a MV predictor.

Without, the predictor isn't actually used.

8 years agoqdm2: Fix alignment of local array.
Michael Niedermayer [Sat, 18 Jun 2011 00:44:17 +0000 (02:44 +0200)]
qdm2: Fix alignment of local array.

Fixes ticket270

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Anton Khirnov <anton@khirnov.net>