generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
60,986 Commits 24 Branches 241 Tags
Commit Graph

27562 Commits

Author SHA1 Message Date
Michael Niedermayer
fe6603745e Merge commit '6e4009d4cdf5927bdaedf58fcfc5e813b14c366b' 
* commit '6e4009d4cdf5927bdaedf58fcfc5e813b14c366b':
  arm: dcadsp: implement decode_hf as external NEON asm

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 21:52:25 +01:00
Michael Niedermayer
fb3c33f3cd Merge commit '4cb6964244fd6c099383d8b7e99731e72cc844b9' 
* commit '4cb6964244fd6c099383d8b7e99731e72cc844b9':
  dcadec: simplify decoding of VQ high frequencies

Conflicts:
	configure
	libavcodec/dcadec.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 21:41:19 +01:00
Michael Niedermayer
ffb7d7195b avcodec/dcadec: use brackets to ensure that no slow division is used 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 21:32:24 +01:00
Michael Niedermayer
747b0337e7 Merge commit '7686afd049be98d18663682b92d983340fa2c305' 
* commit '7686afd049be98d18663682b92d983340fa2c305':
  dca: factorize scaling in inverse ADPCM

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 21:32:14 +01:00
Michael Niedermayer
baf3adc621 Merge commit '08e3ea60ff4059341b74be04a428a38f7c3630b0' 
* commit '08e3ea60ff4059341b74be04a428a38f7c3630b0':
  x86: synth filter float: implement SSE2 version

Conflicts:
	libavcodec/x86/dcadsp.asm
	libavcodec/x86/dcadsp_init.c

See: 2cdbcc004837ce092a14f326f24d97a29512a2c3
Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 20:38:39 +01:00
Christophe Gisquet
2cdbcc0048 x86: synth filter float: implement SSE2 version 
Timings for Arrandale:
          C    SSE
win32:  2108   334
win64:  1152   322

Factorizing the inner loop with a call/jmp is a >15 cycles cost, even with
the jmp destination being aligned.

Unrolling for ARCH_X86_64 is a 20 cycles gain.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 20:34:40 +01:00
Michael Niedermayer
5333e0dd66 Merge commit '57b1eb9f75b04571063ddec316e290c216c114ac' 
* commit '57b1eb9f75b04571063ddec316e290c216c114ac':
  dcadsp: scan coefficients linearly in dca_lfe_fir

Conflicts:
	libavcodec/dcadsp.c

See: 9ae8e23188fc2e533eea74757c9060557941d3d9
Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 19:40:40 +01:00
Michael Niedermayer
e346a59383 Merge commit 'ad507d7907457e678900bac132122ba7be4644cb' 
* commit 'ad507d7907457e678900bac132122ba7be4644cb':
  x86: dcadsp: implement SSE lfe_dir

Conflicts:
	libavcodec/x86/dcadsp.asm

See: 169243112c1e310d90c030fb258092f6d2e46117
Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 19:22:00 +01:00
Christophe Gisquet
169243112c x86: dcadsp: implement SSE lfe_dir 
Results for Arrandale/Windows:
32: 1670 -> 316
64:  728 -> 298

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 19:20:03 +01:00
Michael Niedermayer
90f674d55b Merge commit '87ec849fe9acba075c843e67bcd01f256f481a18' 
* commit '87ec849fe9acba075c843e67bcd01f256f481a18':
  dcadec: remove scaling in lfe_interpolation_fir

Conflicts:
	libavcodec/dcadec.c
	libavcodec/dcadsp.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 18:14:12 +01:00
Michael Niedermayer
810eb285e3 Merge commit 'a55546f48d55e3d1155840541b2be5f4f8cf18ab' 
* commit 'a55546f48d55e3d1155840541b2be5f4f8cf18ab':
  proresenc: Reuse proper dsputil infrastructure for FDCT

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 18:06:12 +01:00
Michael Niedermayer
2e88f82a8a Merge commit '92e598a57a7ce4b8ac9ea56274af39f5fd888311' 
* commit '92e598a57a7ce4b8ac9ea56274af39f5fd888311':
  prores: Drop DSP infrastructure for prores encoder bits

Conflicts:
	libavcodec/Makefile
	libavcodec/proresdsp.c
	libavcodec/proresenc_kostya.c

Note, these changes only affect one of the 2 prores encoders we have
If someone wants to add optimizations to the affected encoder, or needs/wants
this infrastructure, then iam happy to revert this

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 18:02:00 +01:00
Michael Niedermayer
18d870da83 Merge commit 'd6acefe05862af244fd5a30ae946ed507c063994' 
* commit 'd6acefe05862af244fd5a30ae946ed507c063994':
  proresenc: Drop unnecessary DCT permutation bits

Conflicts:
	libavcodec/proresenc_kostya.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 17:26:37 +01:00
Michael Niedermayer
5ba1648318 Merge commit 'b23650491fbd579a4365f42bd42575afb7b53f7e' 
* commit 'b23650491fbd579a4365f42bd42575afb7b53f7e':
  prores: Use consistent names for DSP arch initialization functions

Conflicts:
	libavcodec/proresdsp.c
	libavcodec/proresdsp.h
	libavcodec/x86/proresdsp_init.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 17:13:00 +01:00
Michael Niedermayer
f3eef02746 avcodec/msvideo1: Fix palette in case of seek before decode 
Fixes Ticket3212

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 13:20:22 +01:00
Janne Grunau
6e4009d4cd arm: dcadsp: implement decode_hf as external NEON asm 2014-02-28 13:12:19 +01:00
Christophe Gisquet
4cb6964244 dcadec: simplify decoding of VQ high frequencies 
The vector dequantization has a test in a loop preventing effective SIMD
implementation. By moving it out of the loop, this loop can be DSPized.

Therefore, modify the current DSP implementation. In particular, the
DSP implementation no longer has to handle null loop sizes.

The decode_hf implementations have following timings:

For x86 Arrandale:
        C  SSE SSE2 SSE4
win32: 260 162  119  104
win64: 242 N/A   89   72

The arm NEON optimizations follow in a later patch as external asm. The
now unused check for the y modifier in arm inline asm is removed from
configure.
 2014-02-28 13:03:22 +01:00
Janne Grunau
7686afd049 dca: factorize scaling in inverse ADPCM 
Based on a patch from Christophe Gisquet.

Unrolling of the m == 0 case avoids a possible use of the uninitilized
value sum when s->predictor_history is not set. I failed to find a
sample for it. It also reduced the cycle count from 220 to 150 on
sandy bridge, x86_64 linux, gcc 4.8.2 compared to his patch.
 2014-02-28 13:00:48 +01:00
Christophe Gisquet
08e3ea60ff x86: synth filter float: implement SSE2 version 
Timings for Arrandale:
          C    SSE
win32:  2108   334
win64:  1152   322

Factorizing the inner loop with a call/jmp is a >15 cycles cost, even with
the jmp destination being aligned.

Unrolling for ARCH_X86_64 is a 20 cycles gain.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2014-02-28 13:00:48 +01:00
Christophe Gisquet
57b1eb9f75 dcadsp: scan coefficients linearly in dca_lfe_fir 
This change is inspired by x86 asm where it frees a register.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2014-02-28 13:00:47 +01:00
Christophe Gisquet
ad507d7907 x86: dcadsp: implement SSE lfe_dir 
Results for Arrandale/Windows:
32: 1670 -> 316
64:  728 -> 298

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2014-02-28 13:00:47 +01:00
Christophe Gisquet
87ec849fe9 dcadec: remove scaling in lfe_interpolation_fir 
The scaling factor is constant so it is faster to scale the
FIR coefficients in the tables during compilation.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2014-02-28 13:00:47 +01:00
Diego Biurrun
a55546f48d proresenc: Reuse proper dsputil infrastructure for FDCT 2014-02-28 11:19:47 +01:00
Diego Biurrun
92e598a57a prores: Drop DSP infrastructure for prores encoder bits 
None of the encoder bits are arch-optimized.
 2014-02-28 11:17:25 +01:00
Diego Biurrun
d6acefe058 proresenc: Drop unnecessary DCT permutation bits 
No permutation is necessary for the FDCT.
 2014-02-28 11:00:24 +01:00
Diego Biurrun
b23650491f prores: Use consistent names for DSP arch initialization functions 2014-02-28 10:34:55 +01:00
Michael Niedermayer
5c634cbeb7 Merge remote-tracking branch 'qatar/master' 
* qatar/master:
  Give IDCT matrix transpose macro a more descriptive name

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-28 02:08:11 +01:00
James Almer
2163a40a46 x86/imdct36: use sse3 instructions in the last BUTTERF step when possible 
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 23:28:15 +01:00
James Almer
fbf98375e4 x86/imdct36: don't build imdct36_float_sse on x86_64 targets 
There's an SSE2 version as well, and x86_64 guarantees that
instruction set is present.

Signed-off-by: James Almer <jamrial@gmail.com>
Reviewed-by: "Ronald S. Bultje" <rsbultje@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 22:54:03 +01:00
Diego Biurrun
f2408ec9d7 Give IDCT matrix transpose macro a more descriptive name 
This also avoids a macro name clash and related warning on ARM.
 2014-02-27 13:38:00 -08:00
Michael Niedermayer
a05635ee01 avcodec/mjpegdec: convert CMYK to GBRAP 
Fixes Ticket2799

This should be moved into swscale once we have a CMYK pixel format

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 22:18:34 +01:00
Michael Niedermayer
501beae6f9 avcodec/mjpegdec: fix decoding 4th plane 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 21:49:46 +01:00
Michael Niedermayer
6904168c79 avcodec/mjpegdec: Print error in case of CMYK 
Also fail if AV_EF_EXPLODE is set.
We do not fail by default, but rather return some image as it may be usefull to the
end user to see what is on the image, for example text could be read quite fine and
objects recognized.

Possibly fixes Ticket3424

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 20:24:39 +01:00
Michael Niedermayer
681e72a668 avcodec/mjpegdec: parse adobe_transform 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 20:13:48 +01:00
Michael Niedermayer
7e8be7081f avcodec/mjpegdec: Print human readable string for APPx 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 19:35:37 +01:00
Michael Niedermayer
4f4cc43fd8 avcodec/h264: allow mixing idr and non idr slices with frame threading again 
This combination exists in the wild
Fixes Ticket3131

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 15:49:25 +01:00
Michael Niedermayer
649686d89b avcodec/h264_refs: remove lost frames instead of disfavoring them 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 11:16:23 +01:00
Michael Niedermayer
64bb64f704 avcodec/h264: fix droped frame handling also for threads > 1 
Seems i mistakely tested just with threads=1
Fixes part of Ticket3386

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 08:07:46 +01:00
Michael Niedermayer
b5005def8a avcodec/h264: avoid using lost frames as references 
Fixes Ticket3386

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-27 03:26:03 +01:00
Michael Niedermayer
c4c5351f08 Merge remote-tracking branch 'qatar/master' 
* qatar/master:
  build: Do not redundantly specifiy H.263-related object files for MSMPEG4v*

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-26 22:02:23 +01:00
Diego Biurrun
a63ac1106d build: Do not redundantly specifiy H.263-related object files for MSMPEG4v* 
These are already covered through dependencies specified in configure.
 2014-02-26 19:44:55 +01:00
Peter Ross
1524b0fa68 libavcodec/rawdec: avoid memcpy when performing 16-bit samples shift 
Signed-off-by: Peter Ross <pross@xvid.org>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-26 19:44:34 +01:00
Michael Niedermayer
bdadf05ec8 avcodec/parser: put lost comments back 
Found-by: ubitux
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-26 18:57:43 +01:00
Michael Niedermayer
2673357048 Merge remote-tracking branch 'qatar/master' 
* qatar/master:
  parser: cosmetics: Drop some unnecessary parentheses

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-25 20:52:45 +01:00
Michael Niedermayer
72d580f819 Merge commit 'a1c699659d56b76c0bf399307f642c6fd6d28281' 
* commit 'a1c699659d56b76c0bf399307f642c6fd6d28281':
  parser: K&R formatting cosmetics

Conflicts:
	libavcodec/parser.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-25 19:12:30 +01:00
Michael Niedermayer
0306436416 Merge commit 'ed61f3ca8a0664a697782253b354055136c5d303' 
* commit 'ed61f3ca8a0664a697782253b354055136c5d303':
  parser: Remove commented-out cruft

Merged-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-25 19:03:45 +01:00
Hendrik Leppkes
bc249bd673 mpegvideo: re-indent buffer clearing code 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-25 13:47:16 +01:00
Hendrik Leppkes
fa84231ee8 mpegvideo: fix overwriting hwaccel surface objects 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-25 13:47:10 +01:00
Diego Biurrun
4ec336484d parser: cosmetics: Drop some unnecessary parentheses 2014-02-25 13:40:47 +01:00
Peter Ross
bef6b27f10 avcodec/vp8dsp: use AV_ZERO64 to clear idct coefficient rows 
Signed-off-by: Peter Ross <pross@xvid.org>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-02-25 12:49:35 +01:00