Commit Graph

74 Commits

Author SHA1 Message Date
Dmitry Kovalev
ef003078e8 Renaming "onyx" to "encoder".
Actual renames:
  vp9_onyx_if.c -> vp9_encoder.c
  vp9_onyx_int.h -> vp9_encoder.h

Change-Id: I80532a80b118d0060518e6c6a0d640e3f411783c
2014-04-22 14:57:05 -07:00
Jim Bankoski
e890c2579b add a context tree structure to encoder
This patch sets up a quad_tree structure (pc_tree) for holding all of
pick_mode_context data we use at any square block size during encoding
or picking modes.  That includes contexts for 2 horizontal and 2 vertical
splits, one none, and pointers to 4 sub pc_tree nodes corresponding
to split.  It also includes a pointer to the current chosen partitioning.

This replaces code that held an index for every level in the pick
modes array including:  sb_index, mb_index,
b_index, ab_index.

These were used as stateful indexes that pointed to the current pick mode
contexts you had at each level stored in the following arrays

array ab4x4_context[][][],
sb8x4_context[][][], sb4x8_context[][][], sb8x8_context[][][],
sb8x16_context[][][], sb16x8_context[][][], mb_context[][], sb32x16[][],
sb16x32[],  sb32_context[], sb32x64_context[], sb64x32_context[],
sb64_context

and the partitioning that had been stored in the following:
b_partitioning, mb_partitioning, sb_partitioning, and sb64_partitioning.

Prior to this patch before doing an encode you had to set the appropriate
index for your block size ( switch statement),  update it ( up to 3
lookups for the index array value) and then make your call into a recursive
function at which point you'd have to call get_context which then
had to do a switch statement based on the blocksize,  and then up to 3
lookups based upon the block size to find the context to use.

With the new code the context for the block size is passed around directly
avoiding the extraneous switch statements and multi dimensional array
look ups that were listed above.   At any level in the search all of the
contexts are local to the pc_tree you are working on (in?).

In addition in most places code that used to call sub functions and
then check if the block size was 4x4 and index was > 0 and return
now don't preferring instead to call the right none function on the inside.



Change-Id: I06e39318269d9af2ce37961b3f95e181b57f5ed9
2014-04-17 07:30:55 -07:00
Dmitry Kovalev
2fc3a18653 Removing unused vp9_mcomp_x86.h file.
We don't use declarations from this file. The real declarations
(differently named) are in vp9_rtcd_defs.pl, e.g. vp9_full_search_sad.

Change-Id: I73cbf064305710ba20747233cfdbe67366f069a0
2014-04-14 11:32:58 -07:00
Dmitry Kovalev
129cb23c14 Adding vp9_ssim.h file.
Change-Id: Ib3b3864a6018c62ac1ea18e30795af74464596cd
2014-04-08 16:08:39 -07:00
Dmitry Kovalev
0a6d5547e2 Adding vp9_speed_features.{h, c}.
Change-Id: I7d9874da8ff78a2d7e0cf11073af9c30538bc9a6
2014-03-28 10:30:28 -07:00
Marco Paniconi
2b06bf20ce Move aq_mode=2 (complexity_aq) to separate file.
Change-Id: Iffa45b9b04196c1ded6037622a8644a2500a62de
2014-03-26 18:01:59 -07:00
Jim Bankoski
7d76cc36df Merge "vp9_write_bit_buffer.h header statics converted to globals" 2014-03-24 14:18:24 -07:00
Yunqing Wang
b458bb7c20 Merge "AVX2 SAD Optimization:" 2014-03-24 10:52:32 -07:00
Jim Bankoski
423590aa63 vp9_write_bit_buffer.h header statics converted to globals
Change-Id: I12c29a630da1fbc5508f11b61d182f9b527b3a35
2014-03-24 09:56:06 -07:00
Marco Paniconi
03a9e5edb6 Rename the aq_mode files.
Change-Id: Id76a628495c822e23825b66a7589b4a3279680e2
2014-03-21 15:20:59 -07:00
levytamar82
0fa8b668c1 AVX2 SAD Optimization:
2 functions were optimized for avx2 by using full 256 bit register
In order to handle 32 elements in parallel instead of only 16 in parallel:
1. vp9_sad32x32x4d
2. vp9_sad64x64x4d

The function level gain is 66% and the user level gain is ~1%.

Change-Id: I4efbb3bc7d8bc03b64b6c98f5cd5c4a9dd3212cb
2014-03-21 13:53:32 -07:00
Marco Paniconi
6b83884ba9 In-frame q adjustment for cyclic background refresh.
Activated using aq_mode=3.

Change-Id: Ied628b9e7bd0e88b0c75790276bca75b19eb5c07
2014-03-18 10:59:21 -07:00
Marco Paniconi
78664081d1 Move svc layer_context to separate file.
Change-Id: Ie47c139d48cb18409d71f98f6a5b9eeb9f9437a9
2014-03-13 14:39:45 -07:00
Dmitry Kovalev
021073fd5f Adding vp9_cost.{h, c} files.
Change-Id: If90c1bc822873156d4e38fca1938e4907f6c95f0
2014-03-05 11:57:57 -08:00
Dmitry Kovalev
7d5bffc452 Adding vpx_sse_to_psnr() function.
Removing all copies of identical vp8_mse2psnr/vp9_mse2psnr functions.
Using vpx_sse_to_psnr() instead in all places.

Change-Id: I15beef9834d43d8fc8a8a7a2d1fc5de3d658fed8
2014-02-26 16:21:12 -08:00
levytamar82
52dac5d1cb AVX2 SubPixel Variance Optimization
Optimizing 2 functions to process 32 elements in parallel instead of 16:
1. vp9_sub_pixel_variance64x64
2. vp9_sub_pixel_variance32x32
both of those function were calling vp9_sub_pixel_variance16xh_ssse3
instead of calling that function, it calls vp9_sub_pixel_variance32xh_avx2
that is written in avx2 and process 32 elements in parallel.
This Optimization gave 70% function level gain and 2% user level gain

Change-Id: I4f5cb386b346ff6c878a094e1c3b37e418e50bde
2014-02-14 16:59:11 -07:00
James Zern
66bfc69bfc Merge "*.mk: s/\bUSE_X86INC/CONFIG_USE_X86INC/" 2014-02-10 15:39:28 -08:00
Dmitry Kovalev
005fc6970b Finally removing "short" from transform names.
Change-Id: I5259b68dc1bcceb153e3ffe638a79a59a3019e9d
2014-02-06 11:54:15 -08:00
Dmitry Kovalev
8b53947a42 Renaming vp9_sad_c.c to vp9_sad.c.
Change-Id: I0beb01b0209cf4ae849b4c67d72107b631f46c0d
2014-02-05 11:31:15 -08:00
James Zern
7cf0c783c1 *.mk: s/\bUSE_X86INC/CONFIG_USE_X86INC/
CONFIG_USE_X86INC is available to every makefile, there's no need to
duplicate its value with USE_X86INC

Change-Id: Id12bd5f09cba78abba56ab5a8f56351562e5b8b6
2014-02-04 20:04:38 -08:00
Dmitry Kovalev
9d6d35c5ef Renaming vp9_variance_c.c to vp9_variance.c.
Change-Id: I7b29cb18ad36d79e1c6329c7de88496059f49db4
2014-02-04 14:49:43 -08:00
Deb Mukherjee
3cd37dfeb5 Adds a non-normative resize library to vp9 encoder
Adds an arbitrary-size resize library for use in scaling of input
frames in a non-normative manner in the vp9 encoder. The method
used is as follows:
Downsampling - Uses a 8 tap filter for factor of 2 decimation upto
a size just higher than the desired size. Then interpolates pixels
at a precision of 1/32 pel using a set of 8-tap filters.
Upsampling - Interpolates pixels at a precision of 1/32 pel using
a set of 8-tap filters.

There is no assembly optimization yet.

Change-Id: Ib5b81e174fc139da322bb97c8214d52289d60d8a
2014-01-21 16:50:00 -08:00
Dmitry Kovalev
c2b5a39345 Removing duplicated SAD calculation code.
Change-Id: I8d693371a29103769d5bed9d5f9cfe4f58ca3189
2014-01-21 14:24:37 -08:00
Jingning Han
2f52decd22 Inter-frame non-RD mode decision
This commit setups a test framework for real-time coding. It enables
a light motion search for non-RD mode decision purpose.

Change-Id: I8bec656331539e963c2b685a70e43e0ae32a6e9d
2014-01-16 12:35:04 -08:00
levytamar82
357b65369f AVX2 Variance Optimization
Optimizing the variance functions: vp9_variance16x16, vp9_variance32x32,
vp9_variance64x64, vp9_variance32x16, vp9_variance64x32,
vp9_mse16x16 by migrating to AVX2
some of the functions were optimized by processing 32 elements instead of 16.
some of the functions were optimized by processing 2 loop strides of 16
elements in a single 256 bit register
This optimization gives between 2.4% - 2.7% user level performance gain
and 42% function level gain.

Change-Id: I265ae08a2b0196057a224a86450153ef3aebd85d
2014-01-08 12:05:53 -07:00
Dmitry Kovalev
4084566554 Renaming vp9_boolcoder.{h, c} to vp9_writer.{h, c}.
Change-Id: I9b9a5fcce8530284df0f270706ee060a0edc1517
2013-12-20 11:10:24 -08:00
Yaowu Xu
e9c19617bf Merge "vp9_short_fdct32x32_rd vp9_short_fdct32x32 optimized for AVX2" 2013-11-27 10:27:32 -08:00
Dmitry Kovalev
204ff1c868 Removing vp9_modecosts.{c, h} files.
Renaming vp9_init_mode_costs() to fill_mode_costs() and moving it to
vp9_rdopt.c.

Change-Id: Ib2542d216458f6dced9f4b7ccbdd2cd98176aa5a
2013-11-25 12:44:05 -08:00
levytamar82
8def766de2 vp9_short_fdct32x32_rd vp9_short_fdct32x32 optimized for AVX2
Change-Id: I6366e84490883b72362f762369d7e5bccb64f02f
2013-11-21 14:19:49 -08:00
Yaowu Xu
30b03050a2 Move vp9_sadmxn.h from common to encoder
Change-Id: I6f6ba91b1b8b280902b171472314d665aa0baf0b
2013-11-19 12:46:08 -08:00
Yaowu Xu
1c61e1960d Move vp9_extend.{h,c} from common to encoder
Since they used in encoder only. This commit also re-order includes
for the files that include vp9_extend.h

Change-Id: I929fc113f2135d3198cd1fc6a17434e5a2f8a459
2013-11-18 12:43:36 -08:00
Dmitry Kovalev
5380739a87 Removing vp9_encodeintra.{h, c} files.
There was only one function in *.c file, so moving it to vp9_encodemb.c.

Change-Id: I728859d08b3d6c05c33c1c5b21f0ea1d0e0f83af
2013-11-15 12:17:16 -08:00
Dmitry Kovalev
ae2f732e8c Adding fht{4x4, 8x8, 16x16} functions.
Adding these functions to encapsulate tx_type check. Changing TX_TYPE to
int to match the declaration in vo9_rtch.h.

Change-Id: I6f3a2df6e35595ca73b6aaa9e3909ee7bc3fd16f
2013-10-25 17:55:07 -07:00
Guillaume Martres
e55f60240a Implement variance-based adaptive quantization
This should be similar to what x264 does with --aq-mode 1.
It works well with clips like parkjoy and touhou
(http://x264.nl/developers/Dark_Shikari/LosslessTouhou.mkv).
At low bitrates, the segmentation signaling overhead may negate the
benefits of this feature.

(PGW) Default changed to feature OFF to allow provisional merge.
Change-Id: I938abf9bb487e1d4ad3b0264ea03d9826275c70b
2013-10-16 11:55:13 +01:00
Jim Bankoski
79401542f7 make vp9 postproc a config option
Vp9 postproc is disabled for now as its not been shown to help and
may be merged with vp8.

Change-Id: I25620d6cd34c6e10331b18c7b5ef7482e39c6057
2013-09-04 10:02:08 -07:00
Jim Bankoski
5b307886fb variance x86inc guards
also fixed bug in sad calcs

Change-Id: I6571fcbe37556c16ae32be66dc0fd879852aac1d
2013-08-06 14:17:13 -07:00
Jim Bankoski
c9126e0b30 sad + miscellaneous updates
Enable use_x86inc as a commandline option.  Fix Bug with sse2 when
x86inc is disabled. Adds Sad asm protection to x86inc protection

Change-Id: Iee0f9dd235ea10e8ace512eb362ba9bebe8c9df6
2013-08-06 12:16:04 -07:00
Jingning Han
7d61f8fe53 Merge "Move fdct32x32 SSE2 implementation in separate file." 2013-08-06 10:46:41 -07:00
Christian Duvivier
3d98205fce Move fdct32x32 SSE2 implementation in separate file.
This is in preparation for the SSE2 version of the high-precision
32x32 forward DCT which will share a lot of code with the existing
low precision version used for rate-distortion search.

Change-Id: I7084b6bdfb480b1fabb8493fb14e3f7fcc7888c0
2013-08-06 10:17:11 -07:00
Jim Bankoski
62c6aa884d block error / x86inc mods
Change-Id: Icb607745634e10b9bac5019d06661ece09fcdb40
2013-08-06 06:23:38 -07:00
Jim Bankoski
a93b115cd6 reworked config for use_x86_inc
Support enabling it or disabling it.  Moved read out to configure.sh
so that its done once instead of in make and in config.

Change-Id: I73a9190cf31de9f03e8a577f478fa522f8c01c8b
2013-08-05 17:35:25 -07:00
Ronald S. Bultje
c13e0bcb52 Remove unused fwalsh/fdct x86 SIMD implementations.
Change-Id: Ia942e56cf322821d42ba06178672791eeee2847e
2013-07-10 18:22:51 -07:00
Yaowu Xu
ba3b2604f0 Merge "Quantize (64-bit only, for now) SSSE3 SIMD." 2013-07-01 15:58:57 -07:00
Ronald S. Bultje
7353ceab9d Quantize (64-bit only, for now) SSSE3 SIMD.
Total encoding time for first 50 frames of bus (speed 0) @ 1500kbps
goes 2min34.8 to 2min14.4, i.e. a 10.4% overall speedup. The code is
x86-64 only, it needs some minor modifications to be 32bit compatible,
because it uses 15 xmm registers, whereas 32bit only has 8.

Change-Id: I2df53770c2e850813ffa713e1a91b45b0082b904
2013-07-01 11:36:07 -07:00
Dmitry Kovalev
bb8ccf1caf Moving encoder subexp encoding functions to subexp.{h, c}.
Change-Id: I83ca53bf6def871f199a382a671f26ad7cbecbca
2013-06-29 11:50:45 -07:00
Ronald S. Bultje
54b2a59623 Implement SSE2 block_error.
Change vp9_block_error() to return a 64bit error variable, change all
callers to expect a 64bit return value (this will prevent overflows,
which we basically don't check for at all right now). Remove duplicate
block_error() function, which fixed that through truncation. Remove
old (incompatible) mmx/sse2 block_error SIMD versions and replace with
a new one that returns a 64bit value.

Encoding time of first 50 frames of bus @ 1500kbps goes from 3min29 to
3min23, i.e. a 3% overall speedup.

Change-Id: Ib71ac5508b5ee8a80f1753cd85d72df1629abe68
2013-06-21 12:54:52 -07:00
Ronald S. Bultje
25c588b1e4 Add subtract_block SSE2 version and unit test.
3% faster overall (3min35.0 to 3min28.5).

Change-Id: I5ff8a5c2c91586b6632ca5009ad1ea51ce94af5e
2013-06-21 09:35:37 -07:00
Ronald S. Bultje
8fb6c58191 Implement sse2 and ssse3 versions for all sub_pixel_variance sizes.
Overall speedup around 5% (bus @ 1500kbps first 50 frames 4min10 ->
3min58). Specific changes to timings for each function compared to
original assembly-optimized versions (or just new version timings if
no previous assembly-optimized version was available):

sse2   4x4:    99 ->   82 cycles
sse2   4x8:           128 cycles
sse2   8x4:           121 cycles
sse2   8x8:   149 ->  129 cycles
sse2   8x16:  235 ->  245 cycles (?)
sse2  16x8:   269 ->  203 cycles
sse2  16x16:  441 ->  349 cycles
sse2  16x32:          641 cycles
sse2  32x16:          643 cycles
sse2  32x32: 1733 -> 1154 cycles
sse2  32x64:         2247 cycles
sse2  64x32:         2323 cycles
sse2  64x64: 6984 -> 4442 cycles

ssse3  4x4:           100 cycles (?)
ssse3  4x8:           103 cycles
ssse3  8x4:            71 cycles
ssse3  8x8:           147 cycles
ssse3  8x16:          158 cycles
ssse3 16x8:   188 ->  162 cycles
ssse3 16x16:  316 ->  273 cycles
ssse3 16x32:          535 cycles
ssse3 32x16:          564 cycles
ssse3 32x32:          973 cycles
ssse3 32x64:         1930 cycles
ssse3 64x32:         1922 cycles
ssse3 64x64:         3760 cycles

Change-Id: I81ff6fe51daf35a40d19785167004664d7e0c59d
2013-06-20 09:34:25 -07:00
Ronald S. Bultje
d9fc451666 Move subpixel variance function from common/ to encoder/.
This seems to only be used in the encoder. Also remove an empty wrapper
file that contained forward declarations for this function, but didn't
actually define any actual functions.

Change-Id: Ifc561eef7ebe374a7d03698055e51e105f6d614b
2013-06-17 16:54:09 -07:00
Dmitry Kovalev
18c83b3714 Compressed/uncompressed frame header changes.
Adding API to read/write uncompressed frame header bits (it is not final
yet). Separate functions to read/write uncompressed header. Moving
clr_type, error_resilient_mode, refresh_frame_context,
frame_parallel_decoding_mode, frame_context_idx from compressed partition
to uncompressed frame header.

Change-Id: Id3ed8a387980c652ae147549412f4ec24a0a5bd0
2013-05-28 18:07:54 -07:00