virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2026-06-19 19:03:00 +02:00
Code Issues Releases Activity
60,468 Commits 39 Branches 425 Tags
0d4d223353b746b2e898a137322acc4bfeaab7d7
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Ronald S. BultjeandClément Bœsch 0d4d223353 vp9: invert order in l[] intra prediction array. 
The directional intra predictors either don't care about order (dc, h,
dc_left, tm), or they prefer inverted order (vr, dr, hd). This allows
more efficient SIMD implementations.
 2014-02-09 18:07:15 +01:00
Ronald S. BultjeandClément Bœsch 37c6eac296 vp9: remove memset in loopfilter level setting loop. 
Use same function for segmentation map writing also.
 2014-02-09 18:07:14 +01:00
Ronald S. BultjeandClément Bœsch 45245e202e vp9: coef decoding context-splat speedup. 2014-02-09 18:07:14 +01:00
Ronald S. BultjeandClément Bœsch fccd7d81b6 vp9: decode_mode context optimizations. 2014-02-09 18:07:14 +01:00
Ronald S. BultjeandClément Bœsch 1dc8559e25 vp9: more aligned zeroes. 2014-02-09 18:07:14 +01:00
Ronald S. BultjeandClément Bœsch 0c30c8ddc1 vp9: faster skip-block coefficient blanking. 2014-02-09 18:07:14 +01:00
Ronald S. BultjeandClément Bœsch cdfcd06126 vp9: coef context reading optimization. 2014-02-09 18:07:14 +01:00
Ronald S. BultjeandClément Bœsch 9aeca1c572 vp9: just disallow temporal or no-update segmentation on size-change. 
The spec doesn't describe how it should be decoded so this is probably
the safest thing to do. Fixes valgrind errors on fuzzed11.ivf and fixes
valgrind errors on fuzzed10.ivf differently.
 2014-02-09 18:07:14 +01:00
Ronald S. BultjeandClément Bœsch 0c67864a37 vp9: don't allow retaining old segmentation maps after a size change. 
Fixes valgrind warnings on fuzzed10.ivf.
 2014-02-08 22:08:20 +01:00
Ronald S. BultjeandClément Bœsch 7f0f47b3df vp9: some variable re-arrangements for alignment. 2014-02-08 18:30:17 +01:00
Ronald S. BultjeandClément Bœsch af63ea7078 vp9: re-allocate block buffers on uses_2pass change w/o size change. 
Fixes valgrind errors and crashes in fuzzed9.ivf.
 2014-02-08 18:30:17 +01:00
Ronald S. BultjeandClément Bœsch c72f587353 vp9/parser: change size type to unsigned. 
Fixes valgrind errors in fuzzed8.ivf because size >= 0x80000000.
 2014-02-08 12:53:59 +01:00
Ronald S. BultjeandClément Bœsch bbc3425fa2 vp9: fix mix-up of last-frame/cur-frame in frame size checks. 
Fixes invalid reads in fuzzed7.ivf.
 2014-02-08 11:18:37 +01:00
Ronald S. BultjeandJanne Grunau 5351964a2b vp8: fix bilinear C code to work if src_stride != dst_stride. 
Signed-off-by: Anton Khirnov <anton@khirnov.net>
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2014-02-06 22:45:29 +01:00
Ronald S. BultjeandJanne Grunau 9ee9c679a7 x86: videodsp: Fix a bug in a %if statement where we used '%%' instead of '&&'. 
Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2014-01-30 15:33:23 +01:00
Ronald S. BultjeandJanne Grunau 51daafb02e x86: videodsp: Properly mark sse2 instructions in emulated_edge_mc as such. 
Should fix crashes or corrupt output on pre-SSE2 CPUs when they were
using SSE2-code (e.g. AMD Athlon XP 2400+ or Intel Pentium III) in
hfix or hvar single-edge (left/right) extension functions.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
 2014-01-30 15:30:01 +01:00
Ronald S. BultjeandMichael Niedermayer c2871568cf vp9: fix invalid ref frame w/h on size change. 
Fixes invalid reads and crashes in vp90-2-05-resize.webm and fuzzed6.ivf.
The output is still not identical to what libvpx does (because we don't
actually scale in MC).

Reviewed-by: ubitux
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-01-26 20:16:01 +01:00
Ronald S. BultjeandMichael Niedermayer d9343c3484 vp9: disable use_last_frame_mvs on resolution change (scalable). 
Prevents some invalid memory accesses after resolution change in
vp90-2-05-resize.webm, and libvpx does this too.

Reviewed-by: ubitux
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-01-26 20:15:45 +01:00
Ronald S. Bultje 4147b337c1 vp9: fix memory corruption if header decoding fails after size change. 2014-01-24 19:25:26 -05:00
Ronald S. Bultje c9e6325ed9 vp9/x86: use explicit register for relative stack references. 
Before this patch, we explicitly modify rsp, which isn't necessarily
universally acceptable, since the space under the stack pointer might
be modified in things like signal handlers. Therefore, use an explicit
register to hold the stack pointer relative to the bottom of the stack
(i.e. rsp). This will also clear out valgrind errors about the use of
uninitialized data that started occurring after the idct16x16/ssse3
optimizations were first merged.
 2014-01-24 19:25:25 -05:00
Ronald S. Bultje 97474d527f vp9/x86: iwht4x4 (lossless) mmx. 2014-01-24 19:25:25 -05:00
Ronald S. Bultje d43efa68bd vp9/x86: 4x4 iadst SIMD (ssse3) variants. 
Cycle measurements for intra itxfm_4x4_add on ped1080p.webm:
idct_idct:    66 -> 67 cycles (noise measurement)
idct_iadst:  199 -> 79 cycles
iadst_idct:  165 -> 70 cycles
iadst_iadst: 183 -> 82 cycles
 2014-01-24 19:25:25 -05:00
Ronald S. Bultje baf47020cd vp9/x86: 8x8 iadst SIMD (ssse3/avx) variants. 
Cycle measurements for intra itxfm_8x8_add on ped1080p.webm:
idct_idct:   133 -> 135 cycles (noise measurement)
idct_iadst:  900 -> 241 cycles
iadst_idct:  864 -> 215 cycles
iadst_iadst: 973 -> 310 cycles
 2014-01-24 19:25:25 -05:00
Ronald S. BultjeandAnton Khirnov bd01412313 vp9: fix mvref finding to adhere to bug in libvpx. 
Fixes a particular youtube video that I unfortunately can't share.

Signed-off-by: Anton Khirnov <anton@khirnov.net>
 2014-01-24 07:02:56 +01:00
Ronald S. BultjeandClément Bœsch 8173d1ffc0 vp9/x86: 16x16 iadst_idct, idct_iadst and iadst_iadst (ssse3+avx). 
Sample timings on ped1080p.webm (of the ssse3 functions):
iadst_idct:  4672 -> 1175 cycles
idct_iadst:  4736 -> 1263 cycles
iadst_iadst: 4924 -> 1438 cycles
Total decoding time changed from 6.565s to 6.413s.
 2014-01-16 13:49:31 +01:00
Ronald S. BultjeandMichael Niedermayer 0065d2d520 vp9: fix mt-related hang a parser infinite loop. 
Fixes trac ticket 3274.

Looked-at-by: ubitux
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-01-11 22:35:14 +01:00
Ronald S. BultjeandMichael Niedermayer c6fe984f2f vp9/x86: make STORE_2X2 macro local. 
Prevents this assembler warning:
libavcodec/x86/vp9itxfm.asm:1208: warning: (VP9_IDCT32_1D:309)
redefining multi-line macro `STORE_2X2'

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-01-08 14:07:15 +01:00
Ronald S. Bultje 024fac5cd4 vp9: make mv bounds 32bit. 
Fixes an assert in file from trac ticket 3188.
 2014-01-07 20:43:35 -05:00
Ronald S. Bultje 5b0fc07819 vp9: reset contextual caches on frame size change with mt enabled. 
Fixes crash/valgrind errors in trac ticket 3188 and hang in 3274.
 2014-01-07 20:43:35 -05:00
Ronald S. Bultje 04a187fb2a vp9/x86: idct_32x32_add_ssse3 sub-8x8-idct. 
Runtime of the full 32x32 idct goes from 2446 to 2441 cycles (intra) or
from 1425 to 1306 cycles (inter). Overall runtime is not significantly
affected.
 2014-01-07 20:43:35 -05:00
Ronald S. Bultje 37b001d14d vp9/x86: idct_32x32_add_ssse3 sub-16x16-idct. 
Runtime of all IDCTs together goes from 3327 to 2473 cycles (intra, i.e.
~35% faster) or from 2312 to 1448 cycles (inter, i.e. ~60% faster). Total
decode time of ped1080p.webm goes from 8.086sec to 7.974sec (1.4% faster).
 2014-01-07 20:43:34 -05:00
Ronald S. Bultje e84d14df10 vp9/x86: idct_32x32_add_ssse3. 
Sub-IDCTs will follow later. ped1080.webm goes from 9.295s to 8.191s
(13.5% faster). The IDCT itself goes from 4372 (intra) or 4337 (inter)
to 403 (intra) or 329 (inter) cycles for the DC-only form, 23755 (intra)
or 23723 (inter) to 3497 (intra) or 3607 (inter) cycles for the no-DC
form, which averages from 23393 (intra) or 16612 (inter) to 3449 (intra)
or 2392 (inter) for all 32x32s together, i.e. about ~7x faster (all
tests done on ped1080p.webm).
 2014-01-07 20:43:30 -05:00
Ronald S. BultjeandMichael Niedermayer 847072873c vp9_parse: don't use AVCodecParserContext as av_log context. 
Fixes trac ticket 3273.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2014-01-05 21:45:24 +01:00
Ronald S. Bultje 18175baa54 vp9/x86: 16px MC functions (64bit only). 
Cycle counts for large MCs (old -> new on ped1080p.webm, mx!=0&&my!=0):
16x8:    876 ->   870  (0.7%)
16x16:  1444 ->  1435  (0.7%)
16x32:  2784 ->  2748  (1.3%)
32x16:  2455 ->  2349  (4.5%)
32x32:  4641 ->  4084 (13.6%)
32x64:  9200 ->  7834 (17.4%)
64x32:  8980 ->  7197 (24.8%)
64x64: 17330 -> 13796 (25.6%)
Total decoding time goes from 9.326sec to 9.182sec.
 2013-12-26 21:05:10 -05:00
Ronald S. Bultje 0d9375fc90 vp9/x86: 16x16 sub-IDCT for top-left 8x8 subblock (eob <= 38). 
Sub8x8 speed (w/o dc-only case) goes from ~750 cycles (inter) or ~735
cycles (intra) to ~415 cycles (inter) or ~430 cycles (intra). Average
overall 16x16 idct speed goes from ~635 cycles (inter) or ~720 cycles
(intra) to ~415 cycles (inter) or ~545 (intra) - all measurements done
using ped1080p.webm.
 2013-12-26 07:40:25 -05:00
Ronald S. BultjeandMichael Niedermayer acafbb4dd2 vp9: fix crash if segmentation=1, keyframe/intraonly=1 and updatemap=0. 
The reference map is never used in such cases, but we accidently copied
it anyway. This could cause crashes if this map has not yet been
allocated. Fixes trac ticket 3188.

Reviewed-by: Clément Bœsch <u@pkh.me>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-12-24 11:43:53 +01:00
Ronald S. BultjeandMichael Niedermayer 4a55bffad3 vp9: fix bug in updating of coef probabilities with parallelmode=1. 
Fixes trac ticket 3228.

Reviewed-by: Clément Bœsch <u@pkh.me>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-12-24 11:43:08 +01:00
Ronald S. Bultje 8d4c616fc0 vp9/x86: idct_add_16x16_ssse3. 
Currently only dc-only and full 16x16. Other subforms will follow in the
near future. Total decoding time of ped1080p.webm goes from 9.7 to 9.3
seconds. DC-only goes from 957 -> 131 cycles, and the full IDCT goes
from ~4050 to ~745 cycles.
 2013-12-14 12:13:26 -05:00
Ronald S. Bultje 92436e8ad9 vp9: implement top/left half (4x4) sub-8x8-IDCT. 
For that specific case (eob>3&&eob<=12), runtime of idct8x8 goes from
668 to 477 cycles. For all idct8x8, runtime goes from 521 to 490 cycles.
 2013-12-07 12:39:36 -05:00
Ronald S. Bultje b2045c44a9 vp9: split pre-load of 11585x2 out of 1d idct macro. 
This allows us to load it only once, instead of twice, in this function.
 2013-12-07 12:39:36 -05:00
Ronald S. Bultje f9a0d4c6e0 vp9: minor refactorings in idct ssse3 assembly. 
Make register usage in macros explicit; change mulsub_2w_4x to use 2
instead of 3 temp registers.
 2013-12-07 12:39:35 -05:00
Ronald S. Bultje 8729964b99 vp9: split x86 assembly in two files. 
(And in future, loopfilter or intra pred could be put in their own
respective files also.)
 2013-12-07 12:39:35 -05:00
Ronald S. BultjeandMichael Niedermayer 9695fb2622 vp9_parser: handle zero-sized packets. 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-12-03 14:06:34 +01:00
Ronald S. Bultje f068aed7b8 vp9: make decode_coeffs() return value void. 
It was previously int and would return error if decode_coeffs_b()
returns an error; however, that can never happen, so refactor all
that code to make all dependent functions return void also (all the
way up to decode_coeffs_sb()).
 2013-11-30 16:26:14 -05:00
Ronald S. Bultje 65f41b5c5c vp9: add a new segmentation sample. 
The old one didn't use segmentation. One uses segmentation in all frame
types (--aq-mode=1), and the other uses all segmentation features, but
only in inter frames (mbgraph).
 2013-11-30 16:26:14 -05:00
Ronald S. Bultje 1d6bb21348 vp9: add fate sample for parallelmode. 
This disables backward probability updates, which makes the codec more
friendly for frame-level multi-threading.
 2013-11-30 16:26:14 -05:00
Ronald S. Bultje 47c6d9403d vp9: cosmetics. 2013-11-30 16:26:14 -05:00
Ronald S. Bultje 76bd878d95 vp9: add a 2-pass decoding mode, and add frame-mt support. 
For a random 1080p sample, decoding time went from 9.7sec (1 threads)
to 6.0sec (2 threads) and 5.2sec (4 threads) in 2-pass decoding mode.
I don't have any samples that use the parallelmode feature, but the
gains should be higher.
 2013-11-30 16:26:13 -05:00
Ronald S. Bultje 46955ae430 vp9: allocate 'b', 'block/uvblock' and 'eob/uveob' dynamically. 2013-11-30 16:26:13 -05:00
Ronald S. Bultje fc7d910b2e vp9: split last/cur_frame from the reference buffers. 
We need more information from last/cur_frame than from reference
buffers, so we can use a simplified structure for reference buffers,
and then store mvs and segmentation map information in last/cur.
 2013-11-30 16:26:13 -05:00
Ronald S. Bultje 39b244002a vp9: move some entries out of VP9Block into VP9Context. 
They aren't really block-related variables in the sense that they are
not block-coded, rather they are state trackers.
 2013-11-30 16:26:12 -05:00
Ronald S. Bultje 84d362f020 vp9: split packet parsing into an AVParser. 2013-11-30 16:26:12 -05:00
Ronald S. Bultje 816737ea5d vp9: use proper refcounting. 
Based on something similar in libav. Author is likely Anton Khirnov
<anton@khirnov.net> but I'm not sure.
 2013-11-30 16:26:12 -05:00
Ronald S. BultjeandLuca Barbato 72ca830f51 lavc: VP9 decoder 
Originally written by Ronald S. Bultje <rsbultje@gmail.com> and
Clément Bœsch <u@pkh.me>

Further contributions by:
Anton Khirnov <anton@khirnov.net>
Diego Biurrun <diego@biurrun.de>
Luca Barbato <lu_zero@gentoo.org>
Martin Storsjö <martin@martin.st>

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
 2013-11-15 10:16:28 +01:00
Ronald S. BultjeandClément Bœsch dd3d0f46bd vp9: fix mvref finding to adhere to bug in libvpx. 
Fixes a particular youtube video that I unfortunately can't share.
 2013-11-09 21:31:06 +01:00
Ronald S. BultjeandMichael Niedermayer 6cf0c4107f Put vp9_scans and vp9_scans_nb in ro_data. 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-11-04 02:32:18 +01:00
Ronald S. BultjeandMichael Niedermayer efc5a54cab vp9: skip itxfm_add if the whole block has no coefficients. 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-10-27 15:02:48 +01:00
Ronald S. BultjeandMichael Niedermayer 960490c0b2 avcodec/x86/videodsp: Small speedups in ff_emulated_edge_mc x86 SIMD. 
Don't use word-size multiplications if size == 2, and if we're using
SIMD instructions (size >= 8), complete leftover 4byte sets using movd,
not mov. Both of these changes lead to minor speedups.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-10-27 15:02:48 +01:00
Ronald S. BultjeandMichael Niedermayer cd86eb265f avcodec/x86/videodsp: fix a bug in a %if statement where we used '%%' instead of '&&'. 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-10-27 15:02:48 +01:00
Ronald S. BultjeandMichael Niedermayer 1b3a7e1f42 avcodec/x86/videodsp: Properly mark sse2 instructions in emulated_edge_mc x86 simd as such. 
Should fix crashes or corrupt output on pre-SSE2 CPUs when they were
using SSE2-code (e.g. AMD Athlon XP 2400+ or Intel Pentium III) in
hfix or hvar single-edge (left/right) extension functions.

Tested-by: Ingo Brückl <ib@wupperonline.de>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-10-24 13:36:55 +02:00
Ronald S. BultjeandMichael Niedermayer 20d78a8606 libavcodec/x86: Fix emulated_edge_mc SSE code to not contain SSE2 instructions on x86-32. 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-10-10 13:36:06 +02:00
Ronald S. BultjeandMichael Niedermayer ad75d2b590 x86: Fix compilation with nasm on PPC & OS/2 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-10-08 12:36:19 +02:00
Ronald S. BultjeandMichael Niedermayer ba9c557b92 avcodec/x86/vp9dsp: Fix compilation with nasm. 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-10-08 02:27:12 +02:00
Ronald S. Bultje f1548c008f Full-pixel MC functions. 
Decoding time of ped1080p.webm goes from 11.3sec to 11.1sec.
 2013-10-02 21:03:15 -04:00
Ronald S. Bultje c07ac8d467 VP9 MC (ssse3) optimizations. 
Decoding time of ped1080p.webm goes from 20.7sec to 11.3sec.
 2013-10-02 21:03:15 -04:00
Ronald S. Bultje 848826f527 Native VP9 decoder. 
Authors: Ronald S. Bultje <rsbultje gmail com>,
         Clement Boesch <u pkh me>
 2013-10-02 21:03:11 -04:00
Ronald S. Bultje face578d56 Rewrite emu_edge functions to have separate src/dst_stride arguments. 
This allows supporting files for which the image stride is smaller than
the max. block size + number of subpel mc taps, e.g. a 64x64 VP9 file
or a 16x16 VP8 file with -fflags +emu_edge.
 2013-09-28 20:28:08 -04:00
Ronald S. Bultje f574b4da56 vp8: use 2 registers for dst_stride and src_stride in neon bilin filter. 2013-09-28 20:25:29 -04:00
Ronald S. BultjeandMichael Niedermayer c341f734e5 Convert multiplier for MV from int to ptrdiff_t. 
This prevents emulated_edge_mc from not undoing mvy*stride-related
integer overflows.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-09-28 11:28:09 +02:00
Ronald S. BultjeandMichael Niedermayer 34b429d5ba vp8: fix PPC assembly and bilinear C code to work if src_stride != dst_stride. 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-09-23 15:23:53 +02:00
Ronald S. BultjeandMartin Storsjö 46d208e1e0 vp8: Wait for prev_frame to parse segment_map before reading it 
This fixes occasional failures of vp8-test-vector-010 with frame-level
multithreading enabled.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-07-07 13:30:29 +03:00
Ronald S. BultjeandMartin Storsjö 6c86a63bad yuv4mpeg: Correctly round chroma up for odd luma sizes 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-07-07 13:30:22 +03:00
Ronald S. BultjeandDerek Buitenhuis cea8a0077f yuv4mpeg: correctly handle chroma for odd luma sizes. 
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
 2013-06-24 19:13:58 -04:00
Ronald S. BultjeandMichael Niedermayer c329713de7 vp8: wait for prev_frame to parse segment_map before reading it. 
This fixes occasional failures of vp8-test-vector-010 with frame-level
multithreading enabled.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
 2013-06-23 17:13:58 +02:00
Ronald S. BultjeandMartin Storsjö d4d186d185 dsputil: Remove non-8bpp draw_edge 
It is never used.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:29:22 +03:00
Ronald S. BultjeandMartin Storsjö c9f5fcd08c dsputil: Merge 9-10 bpp functions for get_pixels and draw_edge 
These only care about pixel storage unit size, not actual bits
used (i.e. they don't clip).

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:29:22 +03:00
Ronald S. BultjeandMartin Storsjö 619e0da191 dsputil: Remove unused 32-bit functions 
Previously, if dct_bits was set to 32, we used separate 32-bit
versions of these functions. Since dct_bits now is removed,
remove the unused 32-bit versions of the functions.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:29:22 +03:00
Ronald S. BultjeandMartin Storsjö c443117f25 dsputil: Remove dct_bits 
dct_bits is never set except in h264, where it is never used,
thus remove it.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:29:22 +03:00
Ronald S. BultjeandMartin Storsjö 54cd5e4f92 dsputil: Remove hpel functions (moved to hpeldsp) 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:29:22 +03:00
Ronald S. BultjeandMartin Storsjö 2957d29f05 alpha: hpeldsp: Move half-pel assembly from dsputil to hpeldsp 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:29:22 +03:00
Ronald S. BultjeandMartin Storsjö 78ce568e43 sparc: hpeldsp: Move vis half-pel assembly from dsputil to hpeldsp 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:29:22 +03:00
Ronald S. BultjeandMartin Storsjö bfb41b5039 bfin: hpeldsp: Move half-pel assembly from dsputil to hpeldsp 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:19:13 +03:00
Ronald S. BultjeandMartin Storsjö 7384b7a713 arm: hpeldsp: Move half-pel assembly from dsputil to hpeldsp 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:19:08 +03:00
Ronald S. BultjeandMartin Storsjö 47e5a98174 ppc: hpeldsp: Move half-pel assembly from dsputil to hpeldsp 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:18:59 +03:00
Ronald S. BultjeandMartin Storsjö 8db00081a3 x86: hpeldsp: Move half-pel assembly from dsputil to hpeldsp 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:18:53 +03:00
Ronald S. BultjeandMartin Storsjö 28bc406c84 mjpeg: Use hpeldsp instead of dsputil for half-pel functions 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:18:38 +03:00
Ronald S. BultjeandMartin Storsjö 1277dc07fb svq1enc: Use hpeldsp instead of dsputil for half-pel functions 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:18:32 +03:00
Ronald S. BultjeandMartin Storsjö 2f6bc5f7c1 svq3: Use hpeldsp instead of dsputil for half-pel functions 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:18:27 +03:00
Ronald S. BultjeandMartin Storsjö f4fed5a2f9 mpegvideo: Use hpeldsp instead of dsputil for half-pel functions 
This also converts vc1, since that is mpegvideo-based.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:18:21 +03:00
Ronald S. BultjeandMartin Storsjö 6caa44aa7d svq1: Use hpeldsp instead of dsputil for half-pel functions 
This makes svq1 independent of dsputil.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:18:14 +03:00
Ronald S. BultjeandMartin Storsjö c10470035e mimic: Use hpeldsp instead of dsputil for half-pel functions 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:18:08 +03:00
Ronald S. BultjeandMartin Storsjö 8071264f21 interplayvideo: Use hpeldsp instead of dsputil for half-pel functions 
This makes interplayvideo independent of dsputil.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:18:02 +03:00
Ronald S. BultjeandMartin Storsjö 0f0a11d576 bink: Use hpeldsp instead of dsputil for half-pel functions 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:17:56 +03:00
Ronald S. BultjeandMartin Storsjö 8f992dc8c7 indeo3: Use hpeldsp instead of dsputil for half-pel functions 
This makes the Indeo 3 decoder independent of dsputil.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:17:50 +03:00
Ronald S. BultjeandMartin Storsjö cb7ecb7563 vp56: Use hpeldsp instead of dsputil for half-pel functions 
This makes vp5 and vp6 independent of dsputil.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:17:33 +03:00
Ronald S. BultjeandMartin Storsjö 3bd062bf7f vp3: Use hpeldsp instead of dsputil for half-pel functions 
This makes vp3 independent of dsputil.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:17:26 +03:00
Ronald S. BultjeandMartin Storsjö 68d8238cca hpeldsp: Add half-pel functions (currently copies of dsputil) 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-19 23:17:17 +03:00
Ronald S. BultjeandMartin Storsjö 015821229f vp3: Use full transpose for all IDCTs 
This way, the special IDCT permutations are no longer needed. This
is similar to how H264 does it, and removes the dsputil dependency
imposed by the scantable code.

Also remove the unused type == 0 cases from the plain C version
of the idct.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-15 12:32:05 +03:00
Ronald S. BultjeandMartin Storsjö c46819f229 x86: Move constants to the only place where they are used 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-15 12:17:39 +03:00
Ronald S. BultjeandMartin Storsjö b93b27edb0 dsputil: Make dsputil selectable 
Signed-off-by: Martin Storsjö <martin@martin.st>
 2013-04-10 11:04:05 +03:00