Statistics
| Branch: | Revision:

ffmpeg / libavcodec / dsputil.c @ a24c4a2a

History | View | Annotate | Download (144 KB)

# Date Author Comment
ef9d1d15 02/10/2006 06:55 AM Loren Merritt

h264: special case dc-only idct. ~1% faster overall

Originally committed as revision 4971 to svn://svn.ffmpeg.org/ffmpeg/trunk

68b51e58 02/05/2006 01:35 PM Steve L'Homme

MSVC-compatible _align8/_align16 declaration
patch by Steve Lhomme, steve .dot. lhomme .at. free .dot. fr

Originally committed as revision 4942 to svn://svn.ffmpeg.org/ffmpeg/trunk

5509bffa 01/12/2006 10:43 PM Diego Biurrun

Update licensing information: The FSF changed postal address.

Originally committed as revision 4842 to svn://svn.ffmpeg.org/ffmpeg/trunk

e8b56208 01/09/2006 03:38 AM Loren Merritt

tweak h264_biweight

Originally committed as revision 4835 to svn://svn.ffmpeg.org/ffmpeg/trunk

80e44bc3 01/08/2006 01:50 AM Michael Niedermayer

use h264 MC functions for 2xX Xx2 blocks in snow too

Originally committed as revision 4824 to svn://svn.ffmpeg.org/ffmpeg/trunk

60900991 01/05/2006 06:58 AM Mike Melanson

ugly, but necessary

Originally committed as revision 4814 to svn://svn.ffmpeg.org/ffmpeg/trunk

27c61ac5 01/04/2006 04:31 PM Michael Niedermayer

8x8 integer dct from x264 as cmp function (under CONFIG_GPL)
if this gives better quality then SATD then someone should port the x86 code too or maybe we could even just call it from libx264
the 4x4 one could be tried too ...

Originally committed as revision 4811 to svn://svn.ffmpeg.org/ffmpeg/trunk

bb270c08 12/22/2005 01:10 AM Diego Biurrun

COSMETICS: tabs --> spaces, some prettyprinting

Originally committed as revision 4764 to svn://svn.ffmpeg.org/ffmpeg/trunk

115329f1 12/17/2005 06:14 PM Diego Biurrun

COSMETICS: Remove all trailing whitespace.

Originally committed as revision 4749 to svn://svn.ffmpeg.org/ffmpeg/trunk

bf4e3bd2 08/14/2005 03:42 PM Måns Rullgård

kill a bunch of compiler warnings

Originally committed as revision 4522 to svn://svn.ffmpeg.org/ffmpeg/trunk

36940eca 06/17/2005 08:24 AM Loren Merritt

H.264 lossless mode

Originally committed as revision 4381 to svn://svn.ffmpeg.org/ffmpeg/trunk

43efd19a 06/02/2005 09:15 PM Loren Merritt

decode H.264 with 8x8 transform.
deblocking is still incorrect with 8x8+cavlc

Originally committed as revision 4339 to svn://svn.ffmpeg.org/ffmpeg/trunk

5cf08f23 05/18/2005 09:17 AM Loren Merritt

H.264 deblocking optimizations (mmx for chroma_bS4 case, convert existing cases to 8-bit math)

Originally committed as revision 4271 to svn://svn.ffmpeg.org/ffmpeg/trunk

5773a746 05/17/2005 06:28 PM Michael Niedermayer

porting the mmx&sse2 (sse2 untested) vp3 idcts to the lavc idct API

Originally committed as revision 4260 to svn://svn.ffmpeg.org/ffmpeg/trunk

8b6103da 05/17/2005 11:12 AM Michael Niedermayer

porting vp3 idct over to lavc idct api

Originally committed as revision 4257 to svn://svn.ffmpeg.org/ffmpeg/trunk

bda1c56c 05/07/2005 04:35 PM Michael Niedermayer

minor optimization of the h264 loop filter

Originally committed as revision 4198 to svn://svn.ffmpeg.org/ffmpeg/trunk

8b975b7c 05/02/2005 10:14 PM Michael Niedermayer

put most codecs under ifdefs

Originally committed as revision 4173 to svn://svn.ffmpeg.org/ffmpeg/trunk

42251a2a 04/25/2005 01:01 AM Loren Merritt

MMX for H.264 deblocking filter

Originally committed as revision 4158 to svn://svn.ffmpeg.org/ffmpeg/trunk

88730be6 02/24/2005 07:08 PM Måns Rullgård

kill warnings patch by (Måns Rullgård <mru inprovide com>)

Originally committed as revision 3977 to svn://svn.ffmpeg.org/ffmpeg/trunk

91c56db6 01/23/2005 11:08 AM Michael Niedermayer

use clip_uint8()

Originally committed as revision 3867 to svn://svn.ffmpeg.org/ffmpeg/trunk

9f2d1b4f 01/07/2005 05:56 AM Loren Merritt

H.264 weighted prediction.
Bidirectional weighting has not been tested for bitwise accuracy, but looks correct.

Originally committed as revision 3811 to svn://svn.ffmpeg.org/ffmpeg/trunk

0fd6aea1 12/08/2004 06:02 PM Michael Niedermayer

more flexible frame skip decission

Originally committed as revision 3743 to svn://svn.ffmpeg.org/ffmpeg/trunk

0fa8158d 09/27/2004 07:47 PM Michael Niedermayer

move h264 idct to its own file and call via function pointer in DspContext
allow h264 idct to be used for lowres=1

Originally committed as revision 3524 to svn://svn.ffmpeg.org/ffmpeg/trunk

1aa8c57b 09/26/2004 10:01 AM Michael Niedermayer

1/8 resolution decoding

Originally committed as revision 3511 to svn://svn.ffmpeg.org/ffmpeg/trunk

9ca358b9 09/26/2004 12:18 AM Michael Niedermayer

1/4 resolution decoding

Originally committed as revision 3509 to svn://svn.ffmpeg.org/ffmpeg/trunk

178fcca8 09/25/2004 11:18 PM Michael Niedermayer

1/2 resolution decoding

Originally committed as revision 3507 to svn://svn.ffmpeg.org/ffmpeg/trunk

0c62d343 09/06/2004 10:32 AM Michael Niedermayer

useless

Originally committed as revision 3436 to svn://svn.ffmpeg.org/ffmpeg/trunk

26efc54e 08/22/2004 05:16 PM Michael Niedermayer

4x4 SSE compare function
wavelet based compare functions
make epzs_motion_search() more flexible so it can be used for a wider range of block sizes
make get_penalty_factor() independant of MpegEncContext

Originally committed as revision 3410 to svn://svn.ffmpeg.org/ffmpeg/trunk

8b69867f 08/13/2004 03:30 PM Michael Niedermayer

Minor Patch for shared libs on Mac OSX by (Bill May <wmay at cisco dot com>)

Originally committed as revision 3387 to svn://svn.ffmpeg.org/ffmpeg/trunk

d4c5d2ad 06/07/2004 05:30 PM Michael Niedermayer

nsse weight

Originally committed as revision 3205 to svn://svn.ffmpeg.org/ffmpeg/trunk

e6a2ac34 06/07/2004 03:23 AM Michael Niedermayer

noise preserving sum of squares comparission function

Originally committed as revision 3204 to svn://svn.ffmpeg.org/ffmpeg/trunk

fdbbf2e0 05/30/2004 09:21 PM Michael Niedermayer

rewrite h261 loop filter
no malloc(64) memcpy free stuff
no filter1 A->B then filter2 A->B (yes not B->A)
no incorrect rouding after the 1d filter

Originally committed as revision 3177 to svn://svn.ffmpeg.org/ffmpeg/trunk

c6148de2 05/30/2004 08:37 PM Michael Niedermayer

h261 decoder by (Maarten Daniels <maarten.daniels at student dot luc dot ac dot be>)

Originally committed as revision 3176 to svn://svn.ffmpeg.org/ffmpeg/trunk

f9ed9d85 04/27/2004 03:58 AM Mike Melanson

separate out put_signed_pixels_clamped() into its own function and
implement an optimized MMX version of the function

Originally committed as revision 3082 to svn://svn.ffmpeg.org/ffmpeg/trunk

116824d0 04/26/2004 12:20 AM Mike Melanson

reorganize and simplify the VP3 IDCT stuff

Originally committed as revision 3071 to svn://svn.ffmpeg.org/ffmpeg/trunk

44f54ceb 04/23/2004 11:38 PM Michael Niedermayer

VIS optimized motion compensation code. by (David S. Miller <davem at redhat dot com>)
ported to ffmpeg by (ja2morri at csclub dot uwaterloo dot ca (james morrison))
useable under LGPL with their agreement

Originally committed as revision 3048 to svn://svn.ffmpeg.org/ffmpeg/trunk

44cb64ee 03/08/2004 02:13 AM Mike Melanson

seperated out the C-based VP3 DSP functions into a different file; also
ported the MMX-optimized versions of those functions

Originally committed as revision 2855 to svn://svn.ffmpeg.org/ffmpeg/trunk

c0a0170c 03/05/2004 06:44 PM Michael Niedermayer

fix motion compensation with (x+1/2,y+1/2) MVs

Originally committed as revision 2849 to svn://svn.ffmpeg.org/ffmpeg/trunk

364a1797 02/06/2004 02:12 AM Michael Niedermayer

quantizer noise shaping optimization

Originally committed as revision 2750 to svn://svn.ffmpeg.org/ffmpeg/trunk

8f2ab833 01/10/2004 04:04 PM Michael Niedermayer

copyright year update of the files i touched and remembered, things look annoyingly unmaintained otherwise

Originally committed as revision 2686 to svn://svn.ffmpeg.org/ffmpeg/trunk

622348f9 01/05/2004 10:57 PM Michael Niedermayer

interlaced dct decision cleanup
function moved to dspcontext
mmx&mmx2 optimized
change SSE -> SAD as default (better quality)
vbv buffer size command line option in kbyte

Originally committed as revision 2669 to svn://svn.ffmpeg.org/ffmpeg/trunk

bb198e19 12/30/2003 04:07 PM Michael Niedermayer

interlaced motion estimation
interlaced mpeg2 encoding
P & B frames
rate distored interlaced mb decission
alternate scantable support
4mv encoding fixes (thats also why the regression tests change)
passing height to most dsp functions
interlaced mpeg4 encoding (no direct mode MBs yet)...

d50635cd 12/14/2003 02:57 AM Michael Niedermayer

split intra / inter dequantization

Originally committed as revision 2607 to svn://svn.ffmpeg.org/ffmpeg/trunk

5b5404e3 12/01/2003 03:49 PM Michael Niedermayer

10l

Originally committed as revision 2550 to svn://svn.ffmpeg.org/ffmpeg/trunk

332f9ac4 12/01/2003 03:23 PM Michael Niedermayer

h263 loop filter
fixed h263 modified quantization
CODEC_FLAG_OBMC

Originally committed as revision 2549 to svn://svn.ffmpeg.org/ffmpeg/trunk

9b879566 11/03/2003 01:26 PM Michel Bardiaux

av_log() patch by (Michel Bardiaux <mbardiaux at peaktime dot be>)

Originally committed as revision 2469 to svn://svn.ffmpeg.org/ffmpeg/trunk

48b1f800 10/25/2003 12:32 AM Roman Shaposhnik
  • adding integer/floating point AAN implementations for DCT 2-4-8

Originally committed as revision 2430 to svn://svn.ffmpeg.org/ffmpeg/trunk

10acc479 10/24/2003 06:28 PM Roman Shaposhnik
  • introducing dct248 into the DSP context. * simple/accurate implementation of dct248 * DV encoding now supports 2-4-8 DCT * DV encoding gets a bit faster (but still miles away
    from what I think it could do) * misc. DV codec cleanups

Originally committed as revision 2425 to svn://svn.ffmpeg.org/ffmpeg/trunk

65e4c8c9 10/22/2003 06:26 PM Michael Niedermayer

floating point AAN DCT

Originally committed as revision 2415 to svn://svn.ffmpeg.org/ffmpeg/trunk

84705403 10/13/2003 05:27 PM Michael Niedermayer

mmx2 optimization of huffyuv median encoding

Originally committed as revision 2372 to svn://svn.ffmpeg.org/ffmpeg/trunk

89ebf4e8 06/27/2003 03:03 PM Mike Melanson

Fix third pixel motion compensation

Originally committed as revision 1990 to svn://svn.ffmpeg.org/ffmpeg/trunk

4fb518c3 06/21/2003 01:11 AM Michael Niedermayer

the return of the idct with 16bit output by ("Ivan Kalvachev" <ivan at cacad dot com>)

Originally committed as revision 1983 to svn://svn.ffmpeg.org/ffmpeg/trunk

da3b9756 06/19/2003 01:44 AM Mike Melanson

added support for B-frames and multiple slices

Originally committed as revision 1974 to svn://svn.ffmpeg.org/ffmpeg/trunk

3d2e8cce 05/19/2003 01:30 PM Michael Niedermayer

ASV1 codec
with postprocessing support :)

Originally committed as revision 1891 to svn://svn.ffmpeg.org/ffmpeg/trunk

669ac79c 05/15/2003 11:30 PM Michael Niedermayer

moving the svq3 motion compensation stuff to dsputil (this also means that existing optimized halfpel code is used now ...)

Originally committed as revision 1885 to svn://svn.ffmpeg.org/ffmpeg/trunk

d8085ea7 05/14/2003 11:08 PM Michael Niedermayer

cleanup

Originally committed as revision 1882 to svn://svn.ffmpeg.org/ffmpeg/trunk

0c6bd2ea 05/14/2003 12:18 PM BERO

sh4 optimized idct & bswap patch by (BERO <bero at geocities dot co dot jp>)

Originally committed as revision 1877 to svn://svn.ffmpeg.org/ffmpeg/trunk

59cf08ce 04/20/2003 04:16 PM Fabrice Bellard

fixed static init

Originally committed as revision 1799 to svn://svn.ffmpeg.org/ffmpeg/trunk

0da71265 04/04/2003 02:42 PM Michael Niedermayer

H264 decoder & demuxer

Originally committed as revision 1732 to svn://svn.ffmpeg.org/ffmpeg/trunk

983e3246 03/06/2003 11:32 AM Michael Niedermayer

per file doxy

Originally committed as revision 1634 to svn://svn.ffmpeg.org/ffmpeg/trunk

2a5700de 03/06/2003 02:04 AM Michael Niedermayer

more idiot proof dox

Originally committed as revision 1629 to svn://svn.ffmpeg.org/ffmpeg/trunk

a9badb51 03/05/2003 11:55 PM Michael Niedermayer

ff_block_permute() dox

Originally committed as revision 1628 to svn://svn.ffmpeg.org/ffmpeg/trunk

b0368839 03/03/2003 02:54 PM Michael Niedermayer

MpegEncContext.(i)dct_* -> DspContext.(i)dct_*
bitexact cleanup

Originally committed as revision 1617 to svn://svn.ffmpeg.org/ffmpeg/trunk

0c1a9eda 02/11/2003 04:35 PM Zdenek Kabelac
  • UINTX -> uintx_t INTX -> intx_t

Originally committed as revision 1578 to svn://svn.ffmpeg.org/ffmpeg/trunk

76fbb024 01/19/2003 08:04 PM Michael Niedermayer

align

Originally committed as revision 1477 to svn://svn.ffmpeg.org/ffmpeg/trunk

67725183 01/19/2003 05:55 PM Michael Niedermayer

finetuneing thresholds/factors
nicer mb decission
a few minor improvements & fixes

Originally committed as revision 1472 to svn://svn.ffmpeg.org/ffmpeg/trunk

6b026927 01/19/2003 12:06 PM Falk Hüffner

sse16_c is totally fucked up (unaligned loads, LONG_MAX is undefined,
uint32 array index -> segv), so let's just use a nice plain
unobfuscated version, which also happens to be faster for me.

Originally committed as revision 1470 to svn://svn.ffmpeg.org/ffmpeg/trunk

1d0eab1d 01/18/2003 01:37 AM Michael Niedermayer

bugs

Originally committed as revision 1469 to svn://svn.ffmpeg.org/ffmpeg/trunk

0e15384d 01/15/2003 07:21 PM Michael Niedermayer

fixing DCTELEM != short

Originally committed as revision 1462 to svn://svn.ffmpeg.org/ffmpeg/trunk

3a87ac94 01/15/2003 06:05 PM Michael Niedermayer

more compare functions (rd & bit)

Originally committed as revision 1461 to svn://svn.ffmpeg.org/ffmpeg/trunk

2a006cd3 01/11/2003 10:32 PM Felix von Leitner

sse16 & pix_norm1 optimization patch by (Felix von Leitner <felix-ffmpeg at fefe dot de>) (with some modifications)

Originally committed as revision 1450 to svn://svn.ffmpeg.org/ffmpeg/trunk

d32ac509 01/11/2003 09:04 PM Felix von Leitner

add/diff_bytes bugfix patch by (Felix von Leitner <felix-ffmpeg at fefe dot de>)

Originally committed as revision 1449 to svn://svn.ffmpeg.org/ffmpeg/trunk

db794953 01/09/2003 08:42 PM Michael Niedermayer

qpel fix

Originally committed as revision 1426 to svn://svn.ffmpeg.org/ffmpeg/trunk

826f429a 01/05/2003 03:57 PM Michael Niedermayer

qpel in mmx2/3dnow
qpel refinement quality parameter

Originally committed as revision 1393 to svn://svn.ffmpeg.org/ffmpeg/trunk

477ab036 12/31/2002 05:23 PM Michael Niedermayer

trellis quantization

Originally committed as revision 1380 to svn://svn.ffmpeg.org/ffmpeg/trunk

1457ab52 12/27/2002 11:51 PM Michael Niedermayer

qpel encoding
4mv+b frames encoding finally fixed
chroma ME
5 comparission functions for ME
b frame encoding speedup
wmv2 codec (unfinished)
user specified diamond size for EPZS

Originally committed as revision 1365 to svn://svn.ffmpeg.org/ffmpeg/trunk

140cb663 12/04/2002 09:13 PM Michael Niedermayer

new PSNR code (now works with chroma, b frames, ...)
rename TYPE to FF_TYPE for the external API
allow user specified pict_type

Originally committed as revision 1308 to svn://svn.ffmpeg.org/ffmpeg/trunk

9c76bd48 11/22/2002 07:53 AM Brian Foley
  • altivec and pix_norm patch by Brian Foley

Originally committed as revision 1269 to svn://svn.ffmpeg.org/ffmpeg/trunk

9a7b310d 11/22/2002 07:24 AM Zdenek Kabelac
  • remaining part of John Ryland's patch

Originally committed as revision 1267 to svn://svn.ffmpeg.org/ffmpeg/trunk

11f18faf 11/14/2002 07:20 PM Michael Niedermayer

huffyuv

Originally committed as revision 1211 to svn://svn.ffmpeg.org/ffmpeg/trunk

92ddb692 11/12/2002 10:05 AM Zdenek Kabelac
  • init for inv_zigzag_direct16 moved to init block

Originally committed as revision 1204 to svn://svn.ffmpeg.org/ffmpeg/trunk

5abd509a 11/12/2002 10:01 AM Zdenek Kabelac
  • init table once

Originally committed as revision 1203 to svn://svn.ffmpeg.org/ffmpeg/trunk

45553457 11/12/2002 09:47 AM Zdenek Kabelac
  • using macros to shorten init part
  • removing some unused/unneeded code
  • unifing some function names

Originally committed as revision 1202 to svn://svn.ffmpeg.org/ffmpeg/trunk

34dfe896 11/11/2002 09:44 PM Zdenek Kabelac
  • quick fix for bit exact setting
    (FIXME - add this into context flags)

Originally committed as revision 1200 to svn://svn.ffmpeg.org/ffmpeg/trunk

c9a2ebc4 11/11/2002 04:17 PM Michael Niedermayer

100l for kabi (fixing qpel)

Originally committed as revision 1198 to svn://svn.ffmpeg.org/ffmpeg/trunk

eb4b3dd3 11/11/2002 09:40 AM Zdenek Kabelac
  • using DSPContext - so each codec could use its local (sub)set of CPU extension

Originally committed as revision 1194 to svn://svn.ffmpeg.org/ffmpeg/trunk

7801d21d 10/23/2002 03:11 PM Michael Niedermayer

optimize block_permute()
optimize dct_quantize_c()
dont permute s->q_inter/intra_matrix

Originally committed as revision 1067 to svn://svn.ffmpeg.org/ffmpeg/trunk

073b013d 10/16/2002 07:55 PM Michael Niedermayer

complete mpeg4 GMC decoding support

Originally committed as revision 1046 to svn://svn.ffmpeg.org/ffmpeg/trunk

d2975f8d 10/15/2002 03:24 PM Michael Niedermayer

warning fixes

Originally committed as revision 1044 to svn://svn.ffmpeg.org/ffmpeg/trunk

2ad1516a 09/29/2002 10:44 PM Michael Niedermayer

idct permutation cleanup, idct can be selected per context now
fixing some threadunsafe code

Originally committed as revision 980 to svn://svn.ffmpeg.org/ffmpeg/trunk

d46aba26 09/23/2002 08:44 AM Leon van Stuivenberg

ps2 idct patch by (Leon van Stuivenberg <leonvs at iae dot nl>)

Originally committed as revision 963 to svn://svn.ffmpeg.org/ffmpeg/trunk

5596c60c 09/14/2002 10:07 PM Michael Niedermayer

dont store version for bit-exact tests

Originally committed as revision 948 to svn://svn.ffmpeg.org/ffmpeg/trunk

b3184779 09/11/2002 12:39 PM Michael Niedermayer

put/avg_pixels16
fixing 2 small qpel bugs

Originally committed as revision 915 to svn://svn.ffmpeg.org/ffmpeg/trunk

ab6c65f6 09/02/2002 08:48 AM Brian Foley

altivec build tidyup patch by (Brian Foley <bfoley at compsoc dot nuigalway dot ie>)

Originally committed as revision 898 to svn://svn.ffmpeg.org/ffmpeg/trunk

a43bd1d7 08/30/2002 12:05 AM Heliodoro Tammaro

patch by Heliodoro Tammaro <helio at interactives dot org>
- configures/builds on all darwin versions
- ffserver builds on darwin (if you have the extra libraries)
- builds correctly (again) on non-darwin ppc (no -faltivec in gcc-3.2)

Originally committed as revision 880 to svn://svn.ffmpeg.org/ffmpeg/trunk

28db7fce 08/29/2002 11:55 PM Michael Niedermayer

slow but accurate integer dct from IJG (should be ok with the LGPL as the old DCT is the fast integer DCT from IJG)
per context DCT selection

Originally committed as revision 878 to svn://svn.ffmpeg.org/ffmpeg/trunk

59925ef2 08/28/2002 01:14 PM Brian Foley

first cut at altivec support on darwin patch by (Brian Foley <bfoley at compsoc dot nuigalway dot ie>)

Originally committed as revision 875 to svn://svn.ffmpeg.org/ffmpeg/trunk

3aa102be 08/25/2002 09:19 PM Michael Niedermayer

new ratecontrol code

Originally committed as revision 862 to svn://svn.ffmpeg.org/ffmpeg/trunk

fc2bb4f4 08/02/2002 11:58 AM Michael Niedermayer

fixing normal mmx idct

Originally committed as revision 835 to svn://svn.ffmpeg.org/ffmpeg/trunk

adc09b2e 07/10/2002 09:01 PM Marko Kreen

adding ff prefix to avoid global name conficts with xvid (patch by Marko Kreen <marko at l-t.ee>)

Originally committed as revision 739 to svn://svn.ffmpeg.org/ffmpeg/trunk