platform/upstream/libav.git
11 years agoavformat: Drop pointless "format" from container long names
Diego Biurrun [Tue, 24 Jul 2012 21:51:41 +0000 (23:51 +0200)]
avformat: Drop pointless "format" from container long names

11 years agoswscale: bury one more piece of inline asm under HAVE_INLINE_ASM.
Ronald S. Bultje [Thu, 26 Jul 2012 02:27:03 +0000 (19:27 -0700)]
swscale: bury one more piece of inline asm under HAVE_INLINE_ASM.

11 years agowv: K&R formatting cosmetics
Anton Khirnov [Sat, 28 Jul 2012 10:42:25 +0000 (12:42 +0200)]
wv: K&R formatting cosmetics

11 years agoconfigure: Add missing descriptions to help output
Diego Biurrun [Fri, 27 Jul 2012 11:01:04 +0000 (13:01 +0200)]
configure: Add missing descriptions to help output

11 years agoh264_ps: declare array of colorspace strings on its own line.
Ronald S. Bultje [Sun, 29 Jul 2012 04:45:24 +0000 (21:45 -0700)]
h264_ps: declare array of colorspace strings on its own line.

11 years agofate: amix: specify f32 sample format for comparison
Mans Rullgard [Sun, 29 Jul 2012 17:11:27 +0000 (18:11 +0100)]
fate: amix: specify f32 sample format for comparison

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agotiny_psnr: support 32-bit float samples
Mans Rullgard [Sun, 29 Jul 2012 17:10:36 +0000 (18:10 +0100)]
tiny_psnr: support 32-bit float samples

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agoeamad/eatgq/eatqi: call special EA IDCT directly
Mans Rullgard [Sun, 29 Jul 2012 12:09:10 +0000 (13:09 +0100)]
eamad/eatgq/eatqi: call special EA IDCT directly

These decoders use a special non-MPEG2 IDCT.  Call it directly
instead of going through dsputil.  There is never any reason
to use a regular IDCT with these decoders or to use the EA IDCT
with other codecs.

This also fixes the bizarre situation of eamad and eatqi decoding
incorrectly if eatgq is disabled.

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agoeamad: remove use of MpegEncContext
Mans Rullgard [Fri, 27 Jul 2012 16:00:38 +0000 (17:00 +0100)]
eamad: remove use of MpegEncContext

There is no sense in pulling in this monster struct just for
a handful of fields.  The code does not call any functions
expecting an MpegEncContext.

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agompegvideo: remove unnecessary inclusions of faandct.h
Mans Rullgard [Sun, 29 Jul 2012 12:20:54 +0000 (13:20 +0100)]
mpegvideo: remove unnecessary inclusions of faandct.h

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agoaf_asyncts: avoid overflow in out_size with large delta values
Justin Ruggles [Sat, 28 Jul 2012 17:12:38 +0000 (13:12 -0400)]
af_asyncts: avoid overflow in out_size with large delta values

11 years agoaf_asyncts: add first_pts option
Justin Ruggles [Sat, 28 Jul 2012 16:57:05 +0000 (12:57 -0400)]
af_asyncts: add first_pts option

This allows for padding/trimming at the start of stream. By default, no
assumption is made about the first frame's expected pts, so no padding or
trimming is done.

11 years agoflvdec: remove spurious use of stream id
Luca Barbato [Thu, 26 Jul 2012 14:18:11 +0000 (16:18 +0200)]
flvdec: remove spurious use of stream id

We match streams by codec id now.

11 years agolavf: deprecate r_frame_rate.
Anton Khirnov [Tue, 26 Jun 2012 11:10:01 +0000 (13:10 +0200)]
lavf: deprecate r_frame_rate.

According to its description, it is supposed to be the LCM of all the
frame durations. The usability of such a thing is vanishingly small,
especially since we cannot determine it with any amount of reliability.
Therefore get rid of it after the next bump.

Replace it with the average framerate where it makes sense.

FATE results for the wtv and xmv demux tests change. In the wtv case
this is caused by the file being corrupted (or possibly badly cut) and
containing invalid timestamps. This results in lavf estimating the
framerate wrong and making up wrong frame durations.
In the xmv case the file contains pts jumps, so again the estimated
framerate is far from anything sane and lavf again makes up different
frame durations.

In some other tests lavf starts making up frame durations from different
frame.

11 years agolavf: round estimated average fps to a "standard" fps.
Anton Khirnov [Fri, 27 Jul 2012 12:04:07 +0000 (14:04 +0200)]
lavf: round estimated average fps to a "standard" fps.

11 years agolavf: use dts difference instead of AVPacket.duration in find_stream_info()
Anton Khirnov [Thu, 28 Jun 2012 13:49:51 +0000 (15:49 +0200)]
lavf: use dts difference instead of AVPacket.duration in find_stream_info()

AVPacket.duration is mostly made up and thus completely useless, this is
especially true for video streams.
Therefore use dts difference for framerate estimation and
the max_analyze_duration check.

The asyncts test now needs -analyzeduration, because the default is 5
seconds and the audio stream in the sample appears at ~10 seconds.

11 years agoavf: introduce nobuffer option
Luca Barbato [Thu, 26 Jul 2012 23:35:12 +0000 (16:35 -0700)]
avf: introduce nobuffer option

Useful in cases where a significant analyzeduration is
still needed, while minimizing buffering before output.

An example is processing low-latency streams where all
media types won't necessarily come in if the
analyzeduration is small.

Additional changes by Josh Allmann <joshua.allmann@gmail.com>

Signed-off-by: Anton Khirnov <anton@khirnov.net>
11 years agofate: make yadif tests consistent across systems
Mans Rullgard [Sat, 28 Jul 2012 17:15:56 +0000 (18:15 +0100)]
fate: make yadif tests consistent across systems

MMX-enabled systems by default use some dsputil functions differing
from the C versions.  Adding these flags ensures accurate ones are
used everywhere.

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agovf_hqdn3d: support 9 and 10bit colordepth
Loren Merritt [Fri, 27 Jul 2012 01:42:05 +0000 (01:42 +0000)]
vf_hqdn3d: support 9 and 10bit colordepth

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
11 years agovf_hqdn3d: reduce intermediate precision
Loren Merritt [Thu, 26 Jul 2012 22:51:15 +0000 (22:51 +0000)]
vf_hqdn3d: reduce intermediate precision

11% faster on penryn, 7% on sandybridge, 5% on bulldozer
Negligible change to output.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
11 years agovf_hqdn3d: simplify and optimize
Loren Merritt [Thu, 26 Jul 2012 22:51:14 +0000 (22:51 +0000)]
vf_hqdn3d: simplify and optimize

14% faster on penryn, 2% on sandybridge, 9% on bulldozer

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
11 years agofactor identical ff_inplace_start_frame out of two filters
Loren Merritt [Thu, 26 Jul 2012 22:51:12 +0000 (22:51 +0000)]
factor identical ff_inplace_start_frame out of two filters

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
11 years agovf_hqdn3d: cosmetics
Loren Merritt [Thu, 26 Jul 2012 22:51:13 +0000 (22:51 +0000)]
vf_hqdn3d: cosmetics

Change code style to match the rest of libav.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
11 years agoavprobe/avconv: fix tentative declaration compile errors on MSVS.
Ronald S. Bultje [Thu, 26 Jul 2012 04:23:23 +0000 (04:23 +0000)]
avprobe/avconv: fix tentative declaration compile errors on MSVS.

11 years agoh264_idct_10bit: port x86 assembly to cpuflags.
Ronald S. Bultje [Sat, 28 Jul 2012 15:01:12 +0000 (08:01 -0700)]
h264_idct_10bit: port x86 assembly to cpuflags.

11 years agox86inc: clip num_args to 7 on x86-32.
Loren Merritt [Fri, 27 Jul 2012 16:33:41 +0000 (09:33 -0700)]
x86inc: clip num_args to 7 on x86-32.

This allows us to unconditionally set the cglobal num_args
parameter to a bigger value, thus making writing yasm code
even easier than before.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
11 years agox86inc: sync to latest version from x264.
Ronald S. Bultje [Fri, 27 Jul 2012 03:26:17 +0000 (20:26 -0700)]
x86inc: sync to latest version from x264.

11 years agofft: rename "z" to "zc" to prevent name collision.
Ronald S. Bultje [Sat, 28 Jul 2012 15:20:19 +0000 (08:20 -0700)]
fft: rename "z" to "zc" to prevent name collision.

Without this, cglobal will expand "z" to "zh" to access the high byte
in a register's word, which causes a name collision with the ZH(x) macro
further up in this file.

11 years agowv: return meaningful error codes.
Anton Khirnov [Sat, 28 Jul 2012 10:28:05 +0000 (12:28 +0200)]
wv: return meaningful error codes.

11 years agowv: return AVERROR_EOF on EOF, not EIO.
Anton Khirnov [Sat, 28 Jul 2012 10:23:04 +0000 (12:23 +0200)]
wv: return AVERROR_EOF on EOF, not EIO.

11 years agomp3dec: forward errors for av_get_packet().
Anton Khirnov [Sat, 28 Jul 2012 10:21:21 +0000 (12:21 +0200)]
mp3dec: forward errors for av_get_packet().

Don't invent a bogus EIO error.

The code now doesn't check for ret == 0, but that check is redundant,
av_get_packet() never returns 0.

11 years agomp3dec: remove a pointless local variable.
Anton Khirnov [Sat, 28 Jul 2012 10:20:08 +0000 (12:20 +0200)]
mp3dec: remove a pointless local variable.

11 years agomp3dec: remove commented out cruft.
Anton Khirnov [Sat, 28 Jul 2012 10:18:51 +0000 (12:18 +0200)]
mp3dec: remove commented out cruft.

11 years agolavfi: bump minor to mark stabilizing the ABI.
Anton Khirnov [Wed, 25 Jul 2012 06:34:44 +0000 (08:34 +0200)]
lavfi: bump minor to mark stabilizing the ABI.

11 years agoFATE: add tests for yadif.
Anton Khirnov [Wed, 25 Jul 2012 06:25:38 +0000 (08:25 +0200)]
FATE: add tests for yadif.

11 years agoFATE: add a test for delogo video filter.
Anton Khirnov [Mon, 9 Jul 2012 10:09:58 +0000 (12:09 +0200)]
FATE: add a test for delogo video filter.

11 years agoFATE: add a test for amix audio filter.
Anton Khirnov [Tue, 29 May 2012 10:08:58 +0000 (12:08 +0200)]
FATE: add a test for amix audio filter.

11 years agoaudiogen: allow specifying random seed as a commandline parameter.
Anton Khirnov [Wed, 4 Jul 2012 04:32:30 +0000 (06:32 +0200)]
audiogen: allow specifying random seed as a commandline parameter.

11 years agovc1dec: Override invalid macroblock quantizer
Michael Niedermayer [Sat, 28 Jul 2012 12:07:45 +0000 (18:07 +0600)]
vc1dec: Override invalid macroblock quantizer

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
11 years agovc1: avoid reading beyond the last line in vc1_draw_sprites()
Michael Niedermayer [Sat, 28 Jul 2012 10:27:55 +0000 (16:27 +0600)]
vc1: avoid reading beyond the last line in vc1_draw_sprites()

Fixes overread

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
11 years agovc1dec: check that coded slice positions and interlacing match.
Michael Niedermayer [Sat, 28 Jul 2012 11:14:50 +0000 (17:14 +0600)]
vc1dec: check that coded slice positions and interlacing match.

This fixes out of array writes

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
11 years agovc1dec: Do not ignore ff_vc1_parse_frame_header_adv return value
Michael Niedermayer [Sat, 28 Jul 2012 11:14:48 +0000 (17:14 +0600)]
vc1dec: Do not ignore ff_vc1_parse_frame_header_adv return value

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
11 years agoconfigure: Move parts that should not be user-selectable to CONFIG_EXTRA
Diego Biurrun [Fri, 27 Jul 2012 00:06:29 +0000 (02:06 +0200)]
configure: Move parts that should not be user-selectable to CONFIG_EXTRA

11 years agolavf: remove commented out cruft in avformat_find_stream_info()
Anton Khirnov [Thu, 28 Jun 2012 11:33:12 +0000 (13:33 +0200)]
lavf: remove commented out cruft in avformat_find_stream_info()

11 years agoAdd missing libavutil/time.h includes.
Anton Khirnov [Fri, 27 Jul 2012 14:28:36 +0000 (16:28 +0200)]
Add missing libavutil/time.h includes.

11 years agovp3: don't compile mmx IDCT functions on x86-64.
Ronald S. Bultje [Fri, 27 Jul 2012 05:16:37 +0000 (22:16 -0700)]
vp3: don't compile mmx IDCT functions on x86-64.

64-bit CPUs always have SSE2, and a SSE2 version exists, thus the MMX
version will never be used.

11 years agoh264_loopfilter: port x86 simd to cpuflags.
Ronald S. Bultje [Sat, 28 Jul 2012 00:45:30 +0000 (17:45 -0700)]
h264_loopfilter: port x86 simd to cpuflags.

11 years agobuild: support non-standard replacements for -c flag
Mans Rullgard [Fri, 20 Jul 2012 22:43:27 +0000 (23:43 +0100)]
build: support non-standard replacements for -c flag

This allows non-standard replacements for the -c compiler flag.
Some compilers use other flags or no flag at all in place of
the usual one.

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agobuild: support non-standard replacements for -E flag
Mans Rullgard [Fri, 20 Jul 2012 21:22:41 +0000 (22:22 +0100)]
build: support non-standard replacements for -E flag

This allows using non-standard flags for running the C preprocessor.
The -o flag must be included in this setting due to strange syntax
required by some compilers.

Set the correct flags for tms470.

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agoh264_chromamc_10bit: port x86 simd to cpuflags.
Ronald S. Bultje [Fri, 27 Jul 2012 03:43:50 +0000 (20:43 -0700)]
h264_chromamc_10bit: port x86 simd to cpuflags.

11 years agovp3: port x86 SIMD to cpuflags.
Ronald S. Bultje [Fri, 27 Jul 2012 22:17:27 +0000 (15:17 -0700)]
vp3: port x86 SIMD to cpuflags.

11 years agobuild: Use portable compiler flag constructs in header compilation rule
Diego Biurrun [Fri, 27 Jul 2012 10:06:01 +0000 (12:06 +0200)]
build: Use portable compiler flag constructs in header compilation rule

11 years agox11grab: cosmetics: consistent naming for x11grab-related things
Diego Biurrun [Sat, 21 Jul 2012 19:28:54 +0000 (21:28 +0200)]
x11grab: cosmetics: consistent naming for x11grab-related things

11 years agobuild: Rename YASMDEP variable to DEPYASM for consistency
Diego Biurrun [Wed, 25 Jul 2012 11:25:58 +0000 (13:25 +0200)]
build: Rename YASMDEP variable to DEPYASM for consistency

11 years agohls: Proceed to the next segment at any error code
Martin Storsjö [Fri, 27 Jul 2012 17:24:53 +0000 (20:24 +0300)]
hls: Proceed to the next segment at any error code

Previously, we returned any error code except AVERROR_EOF to the
caller - only if AVERROR_EOF or 0 was returned, we proceeded to
the next segment.

With some setups of web servers, using Connection: close in https
and GnuTLS, we don't get a clean error code at the end of segments.
In those cases, just proceed to the next segment.

Tested-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agotls: Return AVERROR_EOF if the TLS_read/write functions return 0
Martin Storsjö [Fri, 27 Jul 2012 17:22:51 +0000 (20:22 +0300)]
tls: Return AVERROR_EOF if the TLS_read/write functions return 0

OpenSSL returns 0 when the peer has closed the connection. GnuTLS
doesn't return that though, but returns
GNUTLS_E_UNEXPECTED_PACKET_LENGTH if the connection simply is closed
without a clean close notify packet.

Tested-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agorv34: port x86 SIMD to cpuflags.
Ronald S. Bultje [Fri, 27 Jul 2012 05:19:19 +0000 (22:19 -0700)]
rv34: port x86 SIMD to cpuflags.

11 years agovp56: only compile MMX SIMD on x86-32.
Ronald S. Bultje [Fri, 27 Jul 2012 05:09:46 +0000 (22:09 -0700)]
vp56: only compile MMX SIMD on x86-32.

All x86-64 CPUs have SSE2, so the MMX version will never be used. This
leads to smaller binaries.

11 years agovp56: port x86 simd to cpuflags.
Ronald S. Bultje [Fri, 27 Jul 2012 05:07:29 +0000 (22:07 -0700)]
vp56: port x86 simd to cpuflags.

11 years agoproresdsp: port x86 assembly to cpuflags.
Ronald S. Bultje [Fri, 27 Jul 2012 03:38:27 +0000 (20:38 -0700)]
proresdsp: port x86 assembly to cpuflags.

11 years agolavr: x86: improve non-SSE4 version of S16_TO_S32_SX macro
Justin Ruggles [Tue, 26 Jun 2012 20:50:10 +0000 (16:50 -0400)]
lavr: x86: improve non-SSE4 version of S16_TO_S32_SX macro

Removes a false dependency on existing contents of the 2nd dst register,
giving better performance for OOE.

11 years agolavfi: better channel layout negotiation
Justin Ruggles [Sat, 14 Jul 2012 00:22:08 +0000 (20:22 -0400)]
lavfi: better channel layout negotiation

Allow substitution of channel pairs in the input for nearby channel pairs in
the output in order to get a closer match. Also weigh LFE channel mismatch
differently to favor matching the same layout without LFE over one less
channel with LFE.

11 years agoalac: check for truncated packets
Justin Ruggles [Fri, 20 Jul 2012 00:04:40 +0000 (20:04 -0400)]
alac: check for truncated packets

This will give a clearer error message when the error is caused by a
truncated packet.

11 years agoalac: reverse lpc coeff order, simplify filter
Justin Ruggles [Thu, 19 Jul 2012 18:08:22 +0000 (14:08 -0400)]
alac: reverse lpc coeff order, simplify filter

Reversing the lpc coefficient order simplifies indexing in the filter.

11 years agolavr: add x86-optimized mixing functions
Justin Ruggles [Mon, 7 May 2012 13:56:33 +0000 (09:56 -0400)]
lavr: add x86-optimized mixing functions

Adds optimized functions for mixing 3 through 8 input channels to 1 and 2
output channels in fltp or s16p format with flt coeffs.

11 years agox86: add support for fmaddps fma4 instruction with abstraction to avx/sse
Justin Ruggles [Tue, 19 Jun 2012 03:39:14 +0000 (23:39 -0400)]
x86: add support for fmaddps fma4 instruction with abstraction to avx/sse

11 years agotscc2: fix typo in array index
Kostya Shishkov [Thu, 26 Jul 2012 17:42:22 +0000 (19:42 +0200)]
tscc2: fix typo in array index

11 years agobuild: use COMPILE template for HOSTOBJS
Mans Rullgard [Fri, 20 Jul 2012 10:01:19 +0000 (11:01 +0100)]
build: use COMPILE template for HOSTOBJS

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agobuild: do full flag handling for all compiler-type tools
Mans Rullgard [Fri, 20 Jul 2012 12:43:07 +0000 (13:43 +0100)]
build: do full flag handling for all compiler-type tools

This adds a full identification probe of CC, AS, LD and HOSTCC,
and sets up correct flags and dependency tracking for each.

Signed-off-by: Mans Rullgard <mans@mansr.com>
11 years agoeval: fix printing of NaN in eval fate test.
Ronald S. Bultje [Thu, 26 Jul 2012 03:42:04 +0000 (20:42 -0700)]
eval: fix printing of NaN in eval fate test.

This fixes "make fate-eval" on MSVC builds. Without this, the test outputs
"-1.#NaN" instead of "nan" on MSVS 2010.

11 years agobuild: Rename aandct component to more descriptive aandcttables
Diego Biurrun [Tue, 24 Jul 2012 17:16:29 +0000 (19:16 +0200)]
build: Rename aandct component to more descriptive aandcttables

11 years agompegaudio: bury inline asm under HAVE_INLINE_ASM.
Ronald S. Bultje [Thu, 26 Jul 2012 03:32:58 +0000 (20:32 -0700)]
mpegaudio: bury inline asm under HAVE_INLINE_ASM.

11 years agox86inc: automatically insert vzeroupper for YMM functions.
Ronald S. Bultje [Thu, 26 Jul 2012 03:52:16 +0000 (20:52 -0700)]
x86inc: automatically insert vzeroupper for YMM functions.

11 years agortmp: Check the buffer length of ping packets
Samuel Pitoiset [Thu, 26 Jul 2012 18:45:42 +0000 (20:45 +0200)]
rtmp: Check the buffer length of ping packets

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Allow having more unknown data at the end of a chunk size packet without failing
Samuel Pitoiset [Thu, 26 Jul 2012 12:05:58 +0000 (14:05 +0200)]
rtmp: Allow having more unknown data at the end of a chunk size packet without failing

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Prevent reading outside of an allocate buffer when receiving server bandwidth...
Samuel Pitoiset [Thu, 26 Jul 2012 12:05:18 +0000 (14:05 +0200)]
rtmp: Prevent reading outside of an allocate buffer when receiving server bandwidth packets

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agolibopenjpeg: support YUV and deep RGB pixel formats
Michael Bradshaw [Tue, 24 Jul 2012 17:43:06 +0000 (19:43 +0200)]
libopenjpeg: support YUV and deep RGB pixel formats

Based on FFmpeg version from
commit 3275981207e30e140cffaea334ac390f1a04266a

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
11 years agoFix typo in v410 decoder.
Carl Eugen Hoyos [Mon, 23 Jul 2012 22:45:17 +0000 (18:45 -0400)]
Fix typo in v410 decoder.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
11 years agovf_yadif: unset cur_buf on the input link.
Anton Khirnov [Wed, 25 Jul 2012 06:42:27 +0000 (08:42 +0200)]
vf_yadif: unset cur_buf on the input link.

The buffer is stored internally, so this prevents it from being unreffed
automatically.

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
11 years agovf_overlay: ensure the overlay frame does not get leaked.
Anton Khirnov [Wed, 25 Jul 2012 06:42:26 +0000 (08:42 +0200)]
vf_overlay: ensure the overlay frame does not get leaked.

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
11 years agovf_overlay: prevent premature freeing of cur_buf
Michael Niedermayer [Wed, 25 Jul 2012 06:42:25 +0000 (08:42 +0200)]
vf_overlay: prevent premature freeing of cur_buf

Signed-off-by: Anton Khirnov <anton@khirnov.net>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
11 years agoSupport urlencoded http authentication credentials
Antti Seppälä [Wed, 25 Jul 2012 09:43:39 +0000 (12:43 +0300)]
Support urlencoded http authentication credentials

It should be possible to specify usernames in http requests containing
urlencoded characters. This patch adds support for decoding the auth
strings.

Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Return an error when the client bandwidth is incorrect
Samuel Pitoiset [Wed, 25 Jul 2012 18:51:11 +0000 (20:51 +0200)]
rtmp: Return an error when the client bandwidth is incorrect

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Return proper error code in handle_server_bw
Samuel Pitoiset [Wed, 25 Jul 2012 18:51:10 +0000 (20:51 +0200)]
rtmp: Return proper error code in handle_server_bw

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Return proper error code in handle_client_bw
Samuel Pitoiset [Wed, 25 Jul 2012 18:51:09 +0000 (20:51 +0200)]
rtmp: Return proper error code in handle_client_bw

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Return proper error codes in handle_chunk_size
Samuel Pitoiset [Wed, 25 Jul 2012 18:51:08 +0000 (20:51 +0200)]
rtmp: Return proper error codes in handle_chunk_size

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agolavr: x86: add missing vzeroupper in ff_mix_1_to_2_fltp_flt()
Justin Ruggles [Wed, 25 Jul 2012 18:49:40 +0000 (14:49 -0400)]
lavr: x86: add missing vzeroupper in ff_mix_1_to_2_fltp_flt()

11 years agovp8: Replace x*155/100 by x*101581>>16.
Ronald S. Bultje [Tue, 24 Jul 2012 21:23:12 +0000 (21:23 +0000)]
vp8: Replace x*155/100 by x*101581>>16.

Idea stolen from webp (by Pascal Massimino) - because it's Cool.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
11 years agovp3: don't use calls to inline asm in yasm code.
Ronald S. Bultje [Sun, 22 Jul 2012 20:38:56 +0000 (20:38 +0000)]
vp3: don't use calls to inline asm in yasm code.

Mixing yasm and inline asm is a bad idea, since if either yasm or inline
asm is not supported by your toolchain, all of the asm stops working.
Thus, better to use either one or the other alone.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
11 years agox86/dsputil: put inline asm under HAVE_INLINE_ASM.
Ronald S. Bultje [Sun, 22 Jul 2012 21:14:20 +0000 (21:14 +0000)]
x86/dsputil: put inline asm under HAVE_INLINE_ASM.

This allows compiling with compilers that don't support gcc-style
inline assembly.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
11 years agodsputil_mmx: fix incorrect assembly code
Yang Wang [Mon, 23 Jul 2012 22:51:10 +0000 (00:51 +0200)]
dsputil_mmx: fix incorrect assembly code

In ff_put_pixels_clamped_mmx(), there are two assembly code blocks.
In the first block (in the unrolled loop), the instructions
"movq 8%3, %%mm1 \n\t", and so forth, have problems.

From above instruction, it is clear what the programmer wants: a load from
p + 8. But this assembly code doesn’t guarantee that. It only works if the
compiler puts p in a register to produce an instruction like this:
"movq 8(%edi), %mm1". During compiler optimization, it is possible that the
compiler will be able to constant propagate into p. Suppose p = &x[10000].
Then operand 3 can become 10000(%edi), where %edi holds &x. And the instruction
becomes "movq 810000(%edx)". That is, it will stride by 810000 instead of 8.

This will cause a segmentation fault.

This error was fixed in the second block of the assembly code, but not in
the unrolled loop.

How to reproduce:
    This error is exposed when we build using Intel C++ Compiler, with
    IPO+PGO optimization enabled. Crashed when decoding an MJPEG video.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
11 years agortmp: Factorize the code by adding handle_invoke
Samuel Pitoiset [Sat, 21 Jul 2012 10:59:58 +0000 (12:59 +0200)]
rtmp: Factorize the code by adding handle_invoke

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Factorize the code by adding handle_chunk_size
Samuel Pitoiset [Sat, 21 Jul 2012 10:59:52 +0000 (12:59 +0200)]
rtmp: Factorize the code by adding handle_chunk_size

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Factorize the code by adding handle_ping
Samuel Pitoiset [Sat, 21 Jul 2012 10:59:51 +0000 (12:59 +0200)]
rtmp: Factorize the code by adding handle_ping

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Factorize the code by adding handle_client_bw
Samuel Pitoiset [Sat, 21 Jul 2012 10:59:50 +0000 (12:59 +0200)]
rtmp: Factorize the code by adding handle_client_bw

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Factorize the code by adding handle_server_bw
Samuel Pitoiset [Sat, 21 Jul 2012 10:59:49 +0000 (12:59 +0200)]
rtmp: Factorize the code by adding handle_server_bw

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Add a new option 'rtmp_pageurl'
Samuel Pitoiset [Tue, 24 Jul 2012 14:29:40 +0000 (16:29 +0200)]
rtmp: Add a new option 'rtmp_pageurl'

This option specifies the URL of the web page in which the media
was embedded.

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agodoc: Update the description of the rtmp_tcurl option
Samuel Pitoiset [Tue, 24 Jul 2012 14:29:39 +0000 (16:29 +0200)]
doc: Update the description of the rtmp_tcurl option

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agortmp: Make the description of the rtmp_tcurl option more generic
Samuel Pitoiset [Tue, 24 Jul 2012 14:29:38 +0000 (16:29 +0200)]
rtmp: Make the description of the rtmp_tcurl option more generic

Signed-off-by: Martin Storsjö <martin@martin.st>
11 years agolibfdk-aacenc: add LATM/LOAS encapsulation support
Kieran Kunhya [Mon, 23 Jul 2012 16:20:04 +0000 (11:20 -0500)]
libfdk-aacenc: add LATM/LOAS encapsulation support

Signed-off-by: Martin Storsjö <martin@martin.st>