Commit Graph

2659 Commits

Author SHA1 Message Date
Dmitry Kovalev
76d166e413 Removing foreach_predicted_block_uv function.
Adding function build_inter_predictors_for_planes to build inter
predictors for specified planes. This function allows to remove
condition "#if CONFIG_ALPHA" and use MAX_MB_PLANE for general case.
Renaming 'which_mv' local var to 'ref', and 'weight' argument to 'ref'.

Change-Id: I1a97160c9263006929d38953f266bc68e9c56c7d
2013-08-12 13:54:13 -07:00
Dmitry Kovalev
a72e269318 Making scaling code more clear.
Reusing existing functions, using constants instead of magic numbers.

Change-Id: Idc689ffba52c9a8b203fcf26bd67110ecb5635f9
2013-08-12 13:30:26 -07:00
Jingning Han
3984b41c87 Fix a compile failure in vp9_get_compressed_data
The lf struct is now with VP9_COMMON, instead of MACROBLOCKD.

Change-Id: Idfdd4f91f78f486078a138322d58bb61e93e1bc9
2013-08-12 11:42:17 -07:00
Dmitry Kovalev
8b0e6035a2 Entropy context related cleanups.
Adding set_skip_context() function used from both encoder and decoder.

Change-Id: Ia22cfad3211a00a63eb294f64f857b78f4aa9b85
2013-08-12 11:24:24 -07:00
Mans Rullgard
ad7021dd6c vp9: neon: optimise convolve8_vert functions
Invert loops to operate vertically in the inner loop.  This allows
removing redundant loads.

Also add preloading of data.

Change-Id: I4fa85c0ab1735bcb1dd6ea58937efac949172bdc
2013-08-12 15:37:48 +01:00
Dmitry Kovalev
097046ae28 Merge "Removing redundant code and function arguments." 2013-08-11 12:20:58 -07:00
Mans Rullgard
b84dc949c8 vp9: neon: optimise convolve8_horiz functions
Each iteration of the horizontal loop reuses 7 of the 11 source
values.  Loading only the 4 new values saves some time.

Also add preload for source data.

Overall 4% faster on Chromebook.

Change-Id: I8f69e749f2b7f79e9734620dcee51dbfcd716b44
2013-08-11 16:21:55 +01:00
Dmitry Kovalev
3c43ec206c Renaming BLOCK_SIZE_TYPES constant to BLOCK_SIZES.
There will be another change set to rename BLOCK_SIZE_TYPE enum to
BLOCK_SIZE.

Change-Id: I8d1dfc873d6186fa5e554262f5169e929978085e
2013-08-09 17:47:32 -07:00
Guillaume Martres
58b07a6f9d Honor min_partition_size properly
It represents the minimum partition size, so don't split if
bsize == min_partition_size .

Change-Id: Id77c32d6afef7d2ddec0368eaae18fb13227d30e
2013-08-09 17:28:33 -07:00
Dmitry Kovalev
67fe9d17cb Removing redundant code and function arguments.
Change-Id: Ia5cdda0f755befcd1e64397452c42cb7031ca574
2013-08-09 17:24:40 -07:00
Dmitry Kovalev
e7c5ca8983 Merge "Inlining 16 as a stride for BLOCK_OFFSET macro." 2013-08-09 17:22:46 -07:00
James Zern
ef101af8ae Merge "vp9_rd_pick_inter_mode_sb: fix uninitialized value" 2013-08-09 17:13:32 -07:00
Dmitry Kovalev
f1559bdeaf Inlining 16 as a stride for BLOCK_OFFSET macro.
Change-Id: I7f23d174eb089e5500f268a10db09648634c1b82
2013-08-09 16:40:05 -07:00
James Zern
f295774d43 vp9_rd_pick_inter_mode_sb: fix uninitialized value
'skippable' can remain unset and negatively affect later decisions

address one aspect of issue #599

Change-Id: Iffdf0ac2e49ac481c27dc27c87fa546d4167bb28
2013-08-09 16:26:22 -07:00
Dmitry Kovalev
125146034e Merge "Using MV struct instead of int[2] array." 2013-08-09 15:33:08 -07:00
Dmitry Kovalev
cd0629fe68 Merge "Removing plane_block_{width, height}_log2by4 functions." 2013-08-09 15:26:51 -07:00
Dmitry Kovalev
ff7df102d9 Merge "Moving loopfilter struct to VP9_COMMON." 2013-08-09 15:23:00 -07:00
Dmitry Kovalev
816d6c989c Moving loopfilter struct to VP9_COMMON.
Loop filter configuration doesn't belong to macroblock, so moving it from
MACROBLOCKD to VP9_COMMON. Also moving the declaration of loopfilter struct
from vp9_blockd.h to vp9_loopfilter.h.

Change-Id: I4b3e34be9623b47cda35f9b1f9951f8c5b1d5d28
2013-08-09 14:41:51 -07:00
Dmitry Kovalev
8ffe85ad00 Moving scale_factors and related code to separate files.
Change-Id: I531829e5aee2a4a7a112d528ecccbddf052d0e74
2013-08-09 14:07:09 -07:00
Scott LaVarnway
ace93a175d Merge "Bug fix: call set_offsets before rd_auto_partition_range" 2013-08-09 12:30:52 -07:00
Dmitry Kovalev
fa0cd61087 Merge "Using buf_2d struct instead of separate buffer and stride vars." 2013-08-09 11:50:58 -07:00
Scott LaVarnway
41251ae558 Bug fix: call set_offsets before rd_auto_partition_range
The set_offsets call is necessary inorder to set the
mode_info_context ptr correctly.

Change-Id: I644910cc5bacc50ee9cd78458843274ad8ee636d
2013-08-09 14:09:49 -04:00
Adrian Grange
0eef1acbef Merge "Correct bug in loopfilter initialization" 2013-08-09 09:51:58 -07:00
Adrian Grange
12eb2d0267 Correct bug in loopfilter initialization
The memset sets 16 bytes rather than the correct size of the
final array dimension (MAX_MODE_LF_DELTAS).

(In response to bug posted by Manjit Hota to codec-devel
and webm-discuss lists)

Change-Id: I8980f5aa71ddc9d7ef57c5b4700bc28ddf8651c7
2013-08-09 09:21:15 -07:00
Yaowu Xu
6ec2b85bad Added lpf level picking using partial frame
Change-Id: I599ab1bd22b5f3f10d5962c609952abdef8ff67a
2013-08-09 07:37:08 -07:00
Yaowu Xu
6a7a4ba753 renamed vp8_yv12_copy_y to vpx_yv12_copy_y
Becuase the routine is used by both vp8 and vp9

Change-Id: I2d35b287b5bc2394865d931a27da61f4ce7edeeb
2013-08-09 07:37:08 -07:00
Yaowu Xu
c7c9901845 added a speed feature on lpf level picking
Change-Id: Id578f8afdeab3702fc8386969f2d832d8f1b5420
2013-08-09 07:36:32 -07:00
Dmitry Kovalev
6fd2407035 Using buf_2d struct instead of separate buffer and stride vars.
Change-Id: Id5cc3566cc16d1e3030ddb4d1c58459320321dca
2013-08-08 21:25:48 -07:00
Dmitry Kovalev
6a8ec3eac2 General code cleanup.
Removing redundant parenthesis and curly braces. Combining declarations
with initializations. Adding useful intermediate variables instead of
recalculating expressions every time.

Change-Id: I00106f404afd60bfc189905b0fded881684f941a
2013-08-08 21:12:34 -07:00
Dmitry Kovalev
ee40e1a637 Merge "Cleanup inside vp9_reconinter.c." 2013-08-08 14:59:38 -07:00
Deb Mukherjee
2158909fc3 Merge "Adds a new subpel motion function" 2013-08-08 12:26:55 -07:00
Dmitry Kovalev
9e3bcdd135 Merge "Removing unneeded intermediate entropy_nodes_adapt var." 2013-08-08 12:16:57 -07:00
Dmitry Kovalev
47fad4c2d7 Using MV struct instead of int[2] array.
Change-Id: Iab951c555037e36b154f319f351c5e67f9abb931
2013-08-08 12:01:56 -07:00
Dmitry Kovalev
ac008f0030 Removing unneeded intermediate entropy_nodes_adapt var.
Change-Id: I541a178d997b4541e0e2d4d5b854e2ed6b113c3a
2013-08-08 11:52:02 -07:00
Deb Mukherjee
1ba91a84ad Adds a new subpel motion function
Adds a new subpel motion estimation function that uses a 2-level
tree-structured decision tree to eliminate redundant computations.
It searches fewer points than iterative search (which can search
the same point multiple times) but has the same quality roughly.

This is made the default setting at speeds 0 and 1, while at
speed 2 and above only a 1-level search is used.

Also includes various cleanups for consistency and redundancy removal.

Results:
derf: +0.012% psnr
stdhd: +0.09% psnr
Speedup of about 2-3%

Change-Id: Iedde4866f5475586dea0f0ba4cb7428fba24eee9
2013-08-08 11:41:49 -07:00
Adrian Grange
83ee80c045 Moved fast motion search level decision to function
Moving this block of code into a function makes the
code easier to read and change.

Change-Id: If4ede570cce1eab1982b188c4d3e4fd3d4db236e
2013-08-08 11:01:44 -07:00
Adrian Grange
aae6a4c895 Simplify & fix potential bug in rd_pick_partition
Different partitionings were not being evaluated against
best_rd and there were unnecessary calls to RDCOST. This
could have resulted in a non-optimal partioning being
selected.

I simplified the variables used to track the rate,
distortion and RD values throughout the function.

Change-Id: Ifa7085ee80d824e86791432a5bc6d8fea5a3e313
2013-08-08 09:55:45 -07:00
Jingning Han
6bfcce8c7a Merge "Use low precision 32x32fdct for encodemb in speed1" 2013-08-07 19:05:14 -07:00
Dmitry Kovalev
61c33d0ad5 Removing plane_block_{width, height}_log2by4 functions.
Change-Id: I040b82b8e32aee272d10cbb021c7ba1c76343d7a
2013-08-07 17:06:33 -07:00
Dmitry Kovalev
a766d8918e Cleanup inside vp9_reconinter.c.
Using block width and block height instead of their logarithms. Using
SUBPEL_BITS and SUBPEL_SHIFTS constants instead of magic numbers.

Change-Id: I4e10e93c907c8a5e1cb27dfe74d1fcdcc4995448
2013-08-07 17:02:28 -07:00
Dmitry Kovalev
82d7c6fb3c Merge "Using only one scale function in scale_factors struct." 2013-08-07 16:32:09 -07:00
Dmitry Kovalev
1492698ed3 Merge "Adding ss_size_lookup table." 2013-08-07 16:08:24 -07:00
Jingning Han
debb9c68c8 Use low precision 32x32fdct for encodemb in speed1
The low precision 32x32 fdct has all the intermediate steps within
16-bit depth, hence allowing faster SSE2 implementation, at the
expense of larger round-trip error. It was used in the rate-distortion
optimization search loop only.

Using the low precision version, in replace of the high precision one,
affects the compression performance by about 0.7% (derf, stdhd) at
speed 0. For speed 1, it makes derf set down by only 0.017%.

Change-Id: I4e7d18fac5bea5317b91c8e7dabae143bc6b5c8b
2013-08-07 15:34:12 -07:00
Dmitry Kovalev
8db2675b97 Adding ss_size_lookup table.
Removing the old one bsize_from_dim_lookup. Now we have a way to determine
block size for plane using its subsampling values (ss_size_lookup). And
then we can find the number of pixels in the block (num_pels_log2_lookup).

Change-Id: I6fc981da2ae093de81741d3d78eaefed11015db9
2013-08-07 15:33:17 -07:00
Dmitry Kovalev
ea2348ca29 Merge "Removing NMS_STATS defines." 2013-08-07 15:28:30 -07:00
Christian Duvivier
78182538d6 Neon version of vp9_short_idct4x4_add.
Change-Id: Idec4cae0cb9b3a29835fd2750d354c1393d47aa4
2013-08-06 18:41:27 -07:00
Deb Mukherjee
296931c817 Merge "Clean ups of the subpel search functions" 2013-08-06 17:28:48 -07:00
Deb Mukherjee
71b43b0ff0 Clean ups of the subpel search functions
Removes some unused code and speed features, and organizes the
interfaces for fractional mv step functions for use in new speed
features to come.

In the process a new speed feature - number of iterations per
step during the subpel search - is exposed.

No change when this parameter is set as the original value of 3.

Results:
subpel_iters_per_step = 3: baseline
subpel_iters_per_step = 2: psnr -0.067%, 1% speedup
subpel_iters_per_step = 1: psnr -0.331%, 3-4% speedup

Change-Id: I2eba8a21f6461be8caf56af04a5337257a5693a8
2013-08-06 17:23:50 -07:00
Dmitry Kovalev
63ec0587c1 Merge "Motion vector code cleanup." 2013-08-06 16:00:01 -07:00
Dmitry Kovalev
1c552e79bd Using only one scale function in scale_factors struct.
Functions scale_mv_q4 and scale_mv_q3_to_q4 were almost identical except
q3->q4 conversion in scale_mv_q3_to_q4. Now q3->q4 conversion happens
directly in vp9_build_inter_predictor.

Also adding useful constants: SUBPEL_BITS and SUBPEL_MASK.

Change-Id: Ia0a6ad2ac07c45fdf95a5139ece6286c035e9639
2013-08-06 15:43:56 -07:00
Jingning Han
2c091f9768 Merge "Place holder for high-precision 32x32 fdct" 2013-08-06 14:47:30 -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
6eb1254b88 sse3 intrapred x86inc protected
Change-Id: I4a3c83119cdf8a205920034c8019d855d5504605
2013-08-06 14:17:13 -07:00
Deb Mukherjee
fac7c8c9f9 Merge "Flexible support for various pattern searches" 2013-08-06 14:03:27 -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
Dmitry Kovalev
8725ca2ed2 Merge "Inlining vp9_get_pred_probs_switchable_interp function." 2013-08-06 11:57:45 -07:00
Deb Mukherjee
15b5a6a2c7 Flexible support for various pattern searches
Adds a few pattern searches to achieve various tradeoffs
between motion estimation complexity and performance.
The search framework is unified across these searches so that a
common pattern search function is used for all. Besides it will
be easier to experiment with various patterns or combinations
thereof at different scales in the future.

The new pattern search is multi-scale and is capable of using
different patterns at different scales.

The new hex search uses 8 points at the smallest scale
and 6 points at other scales.
Two other pattern searches - big-diamond and square are
also added. Big diamond uses 4 points at the smallest scale and
8 points in diamond shape at the larger scales.
Square is very similar conceptually to the default n-step search
but is somewhat faster since it keeps only one survivor across
all scales.

Psnr/speed-up results on derf300:

hex: -1.6% psnr%, 6-8% speed-up
big-diamond: -0.96% psnr, 4-5% speedup
square: -0.93% psnr, 4-5% speedup

Change-Id: I02a7ef5193f762601e0994e2c99399a3535a43d2
2013-08-06 11:56:39 -07:00
Jingning Han
28566a6cd5 Place holder for high-precision 32x32 fdct
Resolve compile warnings on re-define FDCT32x32_2D template.

Change-Id: Idb3a54ef8d2710ce7245b726379a0e5c875f5cad
2013-08-06 11:44:08 -07:00
Dmitry Kovalev
0c80065694 Inlining vp9_get_pred_probs_switchable_interp function.
There was no benefit having this function. For example, inside
read_switchable_filter_type switchable filter context was calculated twice.

Change-Id: I79cd5bf95cbc0f6d8bf91a2e32289e01b18dcff1
2013-08-06 11:04:31 -07:00
Jingning Han
7d61f8fe53 Merge "Move fdct32x32 SSE2 implementation in separate file." 2013-08-06 10:46:41 -07:00
Jim Bankoski
efc94102f0 Merge "intrapred x86inc guards" 2013-08-06 10:39:19 -07:00
Dmitry Kovalev
a39abe2627 Motion vector code cleanup.
Converting arguments of two functions (clamp_mv_ref, lower_mv_precision)
from int_mv* to MV*. Rewriting is_inside function to make it much shorter.

Change-Id: Ie4c4cf3eccd46707c7df099ec21fb1b61c72fc7a
2013-08-06 10:31:11 -07:00
Dmitry Kovalev
3e51acafec Merge "Finally removing all old block size constants." 2013-08-06 10:30:37 -07:00
Dmitry Kovalev
4a692e4168 Merge "Changing the order switchable filter enum constants." 2013-08-06 10:30:26 -07:00
Dmitry Kovalev
25b7dc08cd Merge "Removing unused functions." 2013-08-06 10:29:57 -07:00
Deb Mukherjee
33afddadb9 Merge "Add variance based mode/skipping" 2013-08-06 10:19:15 -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
25ec1375c9 intrapred x86inc guards
Change-Id: If0399d8e11f4ebe75a5c91abb8d6a52a7709065b
2013-08-06 09:39:30 -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
James Zern
d115cd8b12 Merge changes I082959ab,Ib6932640
* changes:
  vp9/decoder: threaded row-based loop filter
  vp9/decoder: add thread worker
2013-08-05 16:07:09 -07:00
Dmitry Kovalev
b9c7d04e95 Finally removing all old block size constants.
Change-Id: I3aae21e88b876d53ecc955260479980ffe04ad8d
2013-08-05 15:23:49 -07:00
Jim Bankoski
f4837579d1 fixed script problem with config_force_x86_inc
Change-Id: I226e5094d216b09dc47fa5511a66e2d314608000
2013-08-05 14:48:20 -07:00
Jim Bankoski
a5a7322459 Merge "Begin to restrict x86inc.asm usage" 2013-08-05 14:17:49 -07:00
Deb Mukherjee
8b3faccb9e Add variance based mode/skipping
Adds a speed feature to skip all intra modes other than
DC_PRED if the source variance is small. This feature is
made part of speed 1 and up.

Results on derf300: psnr -0.07%, speedup about 1-2%

Also uses the source variance to fine-tune the early
termination criteria when FLAG_EARLY_TERMINATE is on.
This feature is made part of speed 2 and up.

Results on derf300: psnr -0.52%, speedup about 5-7%

Change-Id: I59e38aa836557cfa5405ae706fc64815cbfe4232
2013-08-05 14:14:01 -07:00
Jim Bankoski
9f988a2edf Merge "cleanups after bw bh code" 2013-08-05 14:02:02 -07:00
James Zern
a0ffa2794b vp9/decoder: threaded row-based loop filter
Currently the only threaded option for vp9 decode. Enabled when the
decoder config thread count is > 1.

Change-Id: I082959abac9e31aa4a38ed9fd68b94680e57f4df
2013-08-05 13:22:04 -07:00
James Zern
183b77d5ab vp9/decoder: add thread worker
vp9/decoder/vp9_thread.[hc]
Original source:
 http://git.chromium.org/webm/libwebp.git
 100644 blob b1615d0fb8d311666b2fa4561076c62d72c2e3ff  src/utils/thread.c
 100644 blob 13a61a4c84194c3374080cbf03d881d3cd6af40d  src/utils/thread.h

Local modifications:
 - s/WebP/VP9/g
 - camelcase functions -> lower with _'s

Change-Id: Ib6932640ee34f8b4782c6fbd15864a59d5d4c5fe
2013-08-05 13:21:13 -07:00
Dmitry Kovalev
3f611555d7 Changing the order switchable filter enum constants.
This changeset allows to remove vp9_switchable_interp and
vp9_switchable_interp_map arrays and make code much clear. Actually we
still have to use these mapping but only inside read_interp_filter_type and
write_interp_filter_type functions.

Change-Id: I4026c6f8c4acefba6c81421b7bacbaa52cc45f50
2013-08-05 12:26:15 -07:00
Jim Bankoski
5d2cb7ead0 cleanups after bw bh code
Cons bw/bh parms that should have been const. Additional formatting.

Change-Id: Icd36a5c9dc17dadd7284315ac0d6fef1a565ca16
2013-08-05 12:15:52 -07:00
Jim Bankoski
c3809f3de5 Begin to restrict x86inc.asm usage
Chromium does not support 32bit builds for Mac which use x86inc.asm.
Make the files which include it work if 64bit or not PIC enabled
starting with vp9_copy_sse2.asm

Consolidate these targets in vp9_rtcd_defs.sh

Change-Id: If18f0b957a611efd085a3ee7d245cf1eb91e8248
2013-08-05 12:07:30 -07:00
Dmitry Kovalev
d007446b3f Replacing long block size enum values with shorter ones (2).
Change-Id: I428c4d42212b757112e3acfe5b81314cfbb5fd6b
2013-08-05 10:51:02 -07:00
Dmitry Kovalev
319867d71c Merge "Cleaning up vp9_build_inter_predictor function." 2013-08-05 01:52:11 -07:00
Dmitry Kovalev
78671e2eff Merge "Replacing "txfm" with "tx" in identifiers." 2013-08-04 02:52:22 -07:00
Jim Bankoski
f703f98757 reworked find_mv_ref
This is an attempt at rewriting vp9_find_mv_refs_idx.   I believe that it gains
about 1-2% decode speed

Change-Id: Ia5359c94ce9bb43b32652890e605e9a385485c1b
2013-08-03 20:25:55 -07:00
Dmitry Kovalev
fe2a201eb1 Replacing "txfm" with "tx" in identifiers.
Consistent names with TX_SIZE, TX_MODE, and TX_MODE.

Change-Id: I79592218bf5a40ace89197a34a06ee7de581ed8d
2013-08-02 17:28:23 -07:00
Dmitry Kovalev
5edc65d00d Removing NMS_STATS defines.
Change-Id: Iabab0e59042a33456df1d449c0d0f01debc00c7c
2013-08-02 17:10:15 -07:00
Dmitry Kovalev
7b50333e8f Merge "Adding is_inter_block function." 2013-08-02 16:54:32 -07:00
Dmitry Kovalev
5f0a52faaf Cleaning up vp9_build_inter_predictor function.
Change-Id: I94f6b4272b95ac101de6d10f048116ba065788b0
2013-08-02 16:53:18 -07:00
Dmitry Kovalev
fec4ec4edd Removing unused functions.
Removed functions:
  model_rd_for_sb_y,
  block_error_sby,
  get_sb_variance

Change-Id: Iec458df180caf6f8eac3605773841a4121dd3a8f
2013-08-02 16:41:09 -07:00
Dmitry Kovalev
603931e291 Merge "Changing function arg type from int_mv* to MV*." 2013-08-02 16:30:06 -07:00
Dmitry Kovalev
a6adc82e78 Merge "Cleanups around allow_high_precision_mv flag." 2013-08-02 16:27:05 -07:00
Dmitry Kovalev
680ec32d18 Adding is_inter_block function.
Using it instead of long unclear verbose check
"mbmi->ref_frame[0] != INTRA_FRAME".

Change-Id: I9c7b4b3797942fa962bf3ba7460fff3084beabe9
2013-08-02 16:25:33 -07:00
Dmitry Kovalev
d4e020c4b1 Merge "Cleaning up set_contexts_on_border function." 2013-08-02 16:22:50 -07:00
Yunqing Wang
d340c114fb Merge "Add more checking to using_small_partition_info" 2013-08-02 15:55:09 -07:00
Dmitry Kovalev
769bcab3f5 Cleaning up set_contexts_on_border function.
Change-Id: I8f21c18b29f54b277fb1c167f278f109d9f3b996
2013-08-02 15:52:26 -07:00
Dmitry Kovalev
25b77e2569 Changing function arg type from int_mv* to MV*.
Change-Id: Ic878d31df2ce783a2c9a8c4bc9ed301ec8ffe25e
2013-08-02 15:26:32 -07:00
Dmitry Kovalev
5d86f3886d Moving struct loop_filter_info from *.h to *.c file.
Change-Id: I3fe90eb40088a5b07bdc7d66d93ffe6ef99943d5
2013-08-02 11:53:49 -07:00
Adrian Grange
60ff123536 Merge "Fixed typos and added a few explanatory comments" 2013-08-02 11:37:47 -07:00
Adrian Grange
075b11f004 Merge "Changed name of rd_pick_intra4x4mby_modes" 2013-08-02 11:36:46 -07:00
Johann
8ff58093f0 Merge "vp9: neon: convolve: replace some insns with simpler equivalents" 2013-08-02 11:28:31 -07:00
Johann
8bebfbf7c5 Merge "vp9: neon: convolve: simplify branching to C fallbacks" 2013-08-02 11:28:25 -07:00
Johann
7d14ce8ba5 Merge "vp9: neon: optimise loads in horiz convolve functions" 2013-08-02 11:28:04 -07:00
Johann
319b7dc283 Merge "vp9: neon: add vp9_mb_lpf_* functions" 2013-08-02 11:27:52 -07:00
Dmitry Kovalev
86053d3ae2 Cleanups around allow_high_precision_mv flag.
Change-Id: Ic07f5f8ffeaedd5b7513b464871f83afc82dcd5c
2013-08-02 11:21:16 -07:00
Dmitry Kovalev
b47153deed Replacing long block size enum values with shorter ones.
Change-Id: I0e9329490828684a4fd46f540d89114cc68e8407
2013-08-02 10:48:27 -07:00
Yunqing Wang
0d68080445 Merge "Comment out 2 unused speed features" 2013-08-02 09:58:46 -07:00
Mans Rullgard
355cb14dc7 vp9: neon: convolve: replace some insns with simpler equivalents
Change-Id: I5d6906772e6e6adf68d7f0fd5b8b5207a64a3a37
2013-08-02 08:11:28 -07:00
Mans Rullgard
2003468df8 vp9: neon: convolve: simplify branching to C fallbacks
Change-Id: Ic7cacd02d6dc9243ad8fc85082c5618a9d1e66dc
2013-08-02 08:11:25 -07:00
Mans Rullgard
5e2e78d024 vp9: neon: optimise loads in horiz convolve functions
Loading to single lanes in multiple registers is expensive since
it requires a read and write of each register which saturates
the register file access.  Loading to single registers followed
by a separate transpose reduces this pressure.

Change-Id: I4cc35887ddbca80e5e635b50d2b1d158de9668ee
2013-08-02 08:11:08 -07:00
Mans Rullgard
d85ae87183 vp9: neon: add vp9_mb_lpf_* functions
Change-Id: I13e0880df234f15abc4cc7c57fe84488d5d46a75
2013-08-02 08:10:50 -07:00
Dmitry Kovalev
d91e9f4e36 Merge "Cleanup: replacing xd->seg with seg, and xd->lf with lf." 2013-08-01 23:17:17 -07:00
Dmitry Kovalev
4144fee9e9 Merge "Cleanup: reusing clamp_mv function." 2013-08-01 23:16:56 -07:00
Jingning Han
555bbd68c7 Merge "Remove unused vp9_short_idct10_32x32_add" 2013-08-01 15:41:35 -07:00
Dmitry Kovalev
741537f3ce Cleanup: replacing xd->seg with seg, and xd->lf with lf.
Change-Id: I73b59d7699a8e7e7acd3bf8041cb6c98ce9ba4bf
2013-08-01 15:38:16 -07:00
Dmitry Kovalev
9f4f001ba5 Merge "Cleanup: removing unused function arguments." 2013-08-01 15:07:12 -07:00
Dmitry Kovalev
422d38bca1 Cleanup: reusing clamp_mv function.
Change-Id: I8715f08a3554bdb557c5f935f1dfbd671f18e766
2013-08-01 15:06:34 -07:00
Dmitry Kovalev
ddf02e323a Merge "Nice looking motion vector clamping functions." 2013-08-01 14:50:14 -07:00
Deb Mukherjee
19d42de3ca Merge "Adds a source variance computation function" 2013-08-01 14:18:43 -07:00
Dmitry Kovalev
0497b8d7cd Merge "vp9_get_pred_context_intra_inter cleanup." 2013-08-01 14:15:53 -07:00
Dmitry Kovalev
ce8dedc353 Cleanup: removing unused function arguments.
Change-Id: I27471768980fc631916069f24bc7c482a5c9ca17
2013-08-01 13:41:38 -07:00
Dmitry Kovalev
b621e2d72e Nice looking motion vector clamping functions.
Removing assign_and_clamp_mv function, making implementation of clamp_mv
and clamp_mv2 more clear and consistent.

Change-Id: Iecd08e1c1bf0379f8314ebe01811f8253f4ade58
2013-08-01 13:40:26 -07:00
Deb Mukherjee
dbea726daf Adds a source variance computation function
Adds a function to compute source variance for various
sb_types to be used for pruning mode and partition searches.
[The existing activity measure function is currently specialized
for only 16x16 MBs and needs to be updated].

Change-Id: I22a41e6f1430184201487326fdbebb9b47e6fc24
2013-08-01 13:01:54 -07:00
Jingning Han
67719abde1 Remove unused vp9_short_idct10_32x32_add
The inverse 32x32 transform detects all zero entries and skips the
computations accordingly per 8 rows in the first 1-D operation. The
function vp9_short_idct10_32x32_add performs differently and is not
used anywhere, hence removed.

Change-Id: Ic4fad422debbde7b6b6ffed47c69fbd4268a906c
2013-08-01 12:45:16 -07:00
Jingning Han
56df76bf1b Merge "Optimize 32x32 2D inverse DCT for speed-up" 2013-08-01 11:53:39 -07:00
Yunqing Wang
215b010f4b Add more checking to using_small_partition_info
If the partition is out of partition size range, we don't
need to process small partition information.

Change-Id: Ice9bfbbdebe1f2ef79271a3aee17de0ed4608376
2013-08-01 11:37:41 -07:00
Yunqing Wang
7965a6ea34 Comment out 2 unused speed features
use_min_partition_size and use_max_partition_size are not used
currently, and could be added back if needed later.

Change-Id: Ib22a9c06b064567a7c1d6d5445567ed77e0d3acc
2013-08-01 11:03:34 -07:00
Dmitry Kovalev
ff4bfa726b Merge "Adding missing const to vp9_extra_bits array." 2013-08-01 10:19:51 -07:00
Adrian Grange
89e73c63c0 Fixed typos and added a few explanatory comments
Change-Id: Ib4e4b41094b54874ee34343dd77c0c131ceed9d2
2013-08-01 09:23:49 -07:00
Adrian Grange
5271d47892 Changed name of rd_pick_intra4x4mby_modes
The function name rd_pick_intra4x4mby_modes is confusing, so
I changed it to rd_pick_intra_sub_8x8_y_modes to better
reflect what the function does. Also added const qualifiers
to some of the input parameters and removed camel-case.

Change-Id: I23d53d4c7af5d79ed8a471acd59a09bbb47add39
2013-08-01 09:23:49 -07:00
Dmitry Kovalev
5b65246a71 Adding missing const to vp9_extra_bits array.
Change-Id: Icd128ab58719e0b9066bdfa66a5d0d427a84d6df
2013-07-31 18:51:18 -07:00
Dmitry Kovalev
fb3e78a73a vp9_get_pred_context_intra_inter cleanup.
Change-Id: I8beeee4c020425175f7d5ec83be86afa7b95da1a
2013-07-31 18:33:04 -07:00
Jingning Han
9d67495f72 Optimize 32x32 2D inverse DCT for speed-up
This commit exploits the sparsity of quantized coefficient matrix.
It detects each 32x8 array and skip the corresponding inverse
transformation if all entries are zero.

For ped1080p at 8000 kbps, this on average reduces the runtime of
32x32 inverse 2D-DCT SSE2 function from 6256 cycles -> 5200
cycles. It makes the overall encoding process about 2% faster at
speed 0. The speed-up is more pronounceable for the decoding process.

Change-Id: If20056c3566bd117642a76f8884c83e8bc8efbcf
2013-07-31 17:13:31 -07:00
Jingning Han
12f5762756 Remove unnecessary arguments in rd_pick_ref_frame
This commit removes redundant arguments passing in the function of
rd_pick_reference_frame. This resolves the clang warnings about
potential use of uninitialized values.

Change-Id: Ic68f949a9f8fcd0a583786b0c75321104ea44739
2013-07-31 17:04:13 -07:00
Dmitry Kovalev
8259cdf298 vp9_decodemv.c cleanup.
Inlining VP9_NMV_UPDATE_PROB constant, consistent local variable names.

Change-Id: I01692501982568fa535882d6b320e3c692f88abb
2013-07-31 15:03:36 -07:00
Dmitry Kovalev
9239e96536 Removing get_mi_{row, col} functions.
Passing mi_row and mi_col parameters to functions explicitly. Removing
unused xd argument from scale_mv function.

Change-Id: Icb4c495ec72d26fb066c14470d3ae0b741fbf18a
2013-07-31 14:06:55 -07:00
Dmitry Kovalev
3be9fd9120 Merge "Removing unused "ishp" arguments." 2013-07-31 12:03:04 -07:00
Dmitry Kovalev
0e0a6f840b Merge "Consistent update for inter_mode probabilities." 2013-07-31 12:02:35 -07:00
Dmitry Kovalev
500ade243a Removing unused "ishp" arguments.
Using different variable names "allow_hp" and "use_hp" instead of "usehp".

Change-Id: I0cd5996ddeb46bd754473b680a993c0aaf8eb879
2013-07-31 11:27:53 -07:00
Jingning Han
ac7bab7575 Merge "Make the use of ref_frame index consistent" 2013-07-31 09:11:37 -07:00
Jingning Han
86c384d398 Make the use of ref_frame index consistent
Refactor the frame buffer referencing in choose_partition and make
it consistent with other places. This means to prevent potential
issues when we extend reference frame buffer.

Change-Id: I5ff33ed5f671e1f4cc7049622212769a9b4578d9
2013-07-30 19:49:36 -07:00
Dmitry Kovalev
8701bc11df Consistent update for inter_mode probabilities.
Using inter-mode counts instead of inter-mode-tree branch counts inside
FRAME_COUNTS structure.

Change-Id: I60dde13af37d06146d7d15543311c1b5044e9e04
2013-07-30 18:06:34 -07:00
Adrian Grange
fbd73648dd Merge "Cleanup typos, remove unnecessary lines, replace switch" 2013-07-30 12:59:46 -07:00
Adrian Grange
b30a06b930 Cleanup typos, remove unnecessary lines, replace switch
Removed unnecessary code lines, replaced switch with an if,
fixed spelling errors and formatting.

Change-Id: Ie48aa4604aa0ed48362ca359d792fb21b2ec1dc6
2013-07-30 12:10:32 -07:00
Yaowu Xu
88e48444da Merge "removed duplication" 2013-07-30 09:38:02 -07:00
Yaowu Xu
a15d1f3134 removed duplication
Change-Id: Ica23b66f6664e5a5b168499584f0afffbc54794f
2013-07-30 09:09:14 -07:00
Jingning Han
525745b17a Remove a redundant branching in tokenize_b
The tokenize_b function is only called when output flag is on. Hence
removing the conditional branch on it therein.

Change-Id: Ib709f47f23f39ca05a695faf86fa3377f11f2dd0
2013-07-29 17:08:13 -07:00
Jingning Han
455f2de20b Tune tokenization/detokenization flow for speed-up
This commit optimizes the tokenization and detokenization operational
flow for speed-up. It makes the coding process about 0.3% faster at
speed 0.

Change-Id: I28008df7482874e4b5f237f2d418ff82a249dd56
2013-07-29 16:15:30 -07:00
Jingning Han
b5323ed89a Skip redundant tokenization in rd loop
This commit makes the encoder skip the redundant tokenization process
in the rate-distortion optimization search loop, while updating the
entropy contexts accordingly. It makes the speed 0 encoding process
about 0.5% faster at no performance change.

Change-Id: I34a4155a0b5332afeb45c93a51c7f35a294d685c
2013-07-29 16:09:16 -07:00
Jingning Han
5875d7a4a4 Merge "16x16 inverse 2D-DCT with DC only" 2013-07-29 15:29:25 -07:00
John Koleszar
9c6fafb25b Merge "Remove unnecessary 64 byte alignment" 2013-07-29 15:09:15 -07:00
Jingning Han
a7c4de22e1 16x16 inverse 2D-DCT with DC only
This commit provides special handle on 16x16 inverse 2D-DCT, where
only DC coefficient is quantized to be non-zero value.

Change-Id: I7bf71be7fa13384fab453dc8742b5b50e77a277c
2013-07-29 14:45:53 -07:00
Dmitry Kovalev
828119d6ab Renaming txfm to tx for consistency in some places.
Change-Id: I2a6a646570e2af66315e7c658d00d99f80c4b127
2013-07-29 14:35:55 -07:00
John Koleszar
a31effca75 Remove unnecessary 64 byte alignment
Fixes a warning on MSVS 2012 where the alignment of vp9_default_iscan_8x8
didn't match between its declaration and definition.

Change-Id: I1466a15635f4b22594d705d570b7e399bfb6cf21
2013-07-29 14:02:02 -07:00
Dmitry Kovalev
730a34416f Renaming NB_TXFM_MODES constant to TX_MODES.
Change-Id: I10bf06e3a3d5271221ae6a42a36074d01d493039
2013-07-29 13:38:40 -07:00
Dmitry Kovalev
23391ea835 Renaming TX_SIZE_MAX_SB to TX_SIZES.
Change-Id: I6aa4191935aa93461a07c41b59fdae1eb5f5f107
2013-07-29 12:25:34 -07:00
Jingning Han
decb1b94de Merge "Shortcut 8x8/16x16 inverse 2D-DCT" 2013-07-29 11:04:07 -07:00
Dmitry Kovalev
cc0ff7ecfa Cleanup: replacing xd->mode_info_context with temp variable.
Change-Id: I5a3e83102784cabb918a5404405fcab99c5bb9b6
2013-07-26 19:05:37 -07:00
Ronald S. Bultje
118ccdcd30 Inverse dimension order in token_cost array.
This allows us to increment the position at the band-level only as
we go from one band to the next; more importantly, that allows us to
use an add instead of multiply instruction, and omit the instruction
altogether if the band doesn't change from one coef to the next, thus
being slightly faster (probably more noticeable on systems where a
multiply is expensive, like arm).

Change-Id: I4343fe35b9f9a47fa00b217bdcbf5f91ff96c381
2013-07-26 17:30:04 -07:00
Dmitry Kovalev
35e7e7b614 Merge "vp9_decodemv.c cleanup." 2013-07-26 17:24:34 -07:00
Ronald S. Bultje
6f3054b65d Merge "d45 intra prediction SSSE3 optimizations." 2013-07-26 17:21:09 -07:00
Ronald S. Bultje
dcacce6dd9 Merge "Save pixels instead of coefficients in intra4x4 RD loop." 2013-07-26 17:20:58 -07:00
Ronald S. Bultje
d30c8f41ef Merge "Add best_rd breakout in intra4x4 RD loop." 2013-07-26 17:20:51 -07:00
Jingning Han
38fa487164 Shortcut 8x8/16x16 inverse 2D-DCT
This commit brought back the shortcut implementation of 8x8/16x16
inverse 2D-DCT. When the eob <= 10, it skips the inverse transform
operations on row 4:7/4:15 in the first round. For bus_cif at 1000
kbps, this provides about 2% speed-up at speed 0.

Change-Id: I453e2d72956467d75be4ad8c04b4482ab889d572
2013-07-26 17:19:14 -07:00
Dmitry Kovalev
d42e60d2d8 vp9_decodemv.c cleanup.
Renaming:
  read_intra_mode_info  -> read_intra_frame_mode_info
  read_inter_mode_info  -> read_inter_frame_mode_info
  read_intra_block_part -> read_intra_block_mode_info
  read_inter_block_part -> read_inter_block_mode_info
  read_ref_frame        -> read_ref_frames
  read_reference_frame  -> read_is_inter_block

Using num_4x4_blocks_{wide, high}_lookup instead of bit shifts.

Change-Id: I83c81573b4ef6f53f2f8d24683895014bebfba61
2013-07-26 16:49:49 -07:00
Jingning Han
b9c3dd481a Merge "Special handle on DC only inverse 8x8 2D-DCT" 2013-07-26 16:04:14 -07:00
Dmitry Kovalev
620861dedc Merge "Making read_inter_mode_info function more clear." 2013-07-26 15:47:40 -07:00
hkuang
aaa9755746 Merge "Fix some format error and code error in neon code." 2013-07-26 15:24:28 -07:00
Jingning Han
325e0aa650 Special handle on DC only inverse 8x8 2D-DCT
This commit enables a special handle for the 8x8 inverse 2D-DCT,
where only DC coefficient is quantized to be non-zero. For bus_cif
at 2000 kbps, it provides about 1% speed-up at speed 0.

Change-Id: I2523222359eec26b144cf8fd4c63a4ad63b1b011
2013-07-26 14:16:51 -07:00
hkuang
588b4daf54 Fix some format error and code error in neon code.
Change-Id: I748dee8938dfb19f417f24eed005f3d216f83a82
2013-07-26 14:14:57 -07:00
Dmitry Kovalev
c09b81719f Merge "General cleanups." 2013-07-26 13:59:39 -07:00
Ronald S. Bultje
94b0c6791d d45 intra prediction SSSE3 optimizations.
Change-Id: Ie48035ff4f93c41f8a9b3023e6444fd10432d8fb
2013-07-26 13:30:02 -07:00
Yaowu Xu
4f75a1f4ed Merge "Auto min and max partition size experiment." 2013-07-26 12:10:27 -07:00
Paul Wilkins
fe5e2a91bb Auto min and max partition size experiment.
Speed feature experiment to set an upper and lower
partition size limit based on what has been seen
in spatial neighbors.

This seems to gives quite reasonable speed gains in local
(10-15%) and when used with speed 0 the losses are small
(0.25% derf, 0.35% stdhd). However, for now I am only
enabling it on speed 1 as there may be clashes with the existing
temporal partition selection in speed 2.

Using a tighter min / max around the range derived from the
neighbors increases speed further but at the cost of a
bigger quality loss. However,  I think this spatial method could
be combined with data from either the last frame or a variance
method (or both) to refine the range of minimum and maximum
partition size. I.e. consider the min and max from spatial and
temporal neighbors and the variance recommendation.

Change-Id: I1b96bf8b84368d6aad0c7aa600fe141b4f07435f
2013-07-26 18:30:49 +01:00
Yunqing Wang
52256cdbca Modify static threshold calculation
Used 3 * standard_deviation in internal threshold calculation
instead of fit curve. This actually approached the algorithm
better.
For comparison, similar tests were done:
The overall psnr loss is less than before.
1. derf set:
when static-thresh = 1, psnr loss is 0.329%;
when static-thresh = 500, psnr loss is 0.970%;
2. stdhd set:
when static-thresh = 1, psnr loss is 0.922%;
when static-thresh = 500, psnr loss is 1.307%;

Similar speedup is achieved. For example,
clip            bitrate  static-thresh psnr    time
akiyo(cif)       500        0          48.952  5.077s(50f)
akiyo            500        500        48.866  4.169s(50f)

parkjoy(1080p)   4000       0          30.388  78.20s(30f)
parkjoy          4000       500        30.367  70.85s(30f)

sunflower(1080p) 4000       0          44.402  74.55s(30f)
sunflower        4000       500        44.414  68.69s(30f)

Change-Id: Ic78833642ce1911dbbd1cb6c899a2d7e2dfcc1f3
2013-07-25 19:59:33 -07:00
Dmitry Kovalev
048e9c0991 Making read_inter_mode_info function more clear.
Now read_inter_mode_info calls read_intra_block_part (renamed from
read_intra_block_modes) or read_inter_block_part (just added).

Change-Id: I541badea6b663e0ae692ec158665efb90ed20c03
2013-07-25 15:30:18 -07:00
Johann
67b07c520d Merge "Add const to vp9_accum_mv_refs parameter" 2013-07-25 15:10:52 -07:00
Yunqing Wang
845fd5011c Merge "Add encoding option --static-thresh" 2013-07-25 14:58:00 -07:00
Yunqing Wang
d36852b702 Add encoding option --static-thresh
This option exists in VP8, and it was rewritten in VP9 to support
skipping on different partition levels. After prediction is done,
we can check if the residuals in the partition block will be all
quantized to 0. If this is true, the skip flag is set, and only
prediction data are needed in reconstruction. Based on DCT's energy
conservation property, the skipping check can be estimated in
spatial domain.

The prediction error is calculated and compared to a threshold.
The threshold is determined by the dequant values, and also
adjusted by partition sizes. To be precise, the DC and AC parts
for Y, U, and V planes are checked to decide skipping or not.

Test showed that
1. derf set:
when static-thresh = 1, psnr loss is 0.666%;
when static-thresh = 500, psnr loss is 1.162%;
2. stdhd set:
when static-thresh = 1, psnr loss is 1.249%;
when static-thresh = 500, psnr loss is 1.668%;

For different clips, encoding speedup range is between several
percentage and 20+% when static-thresh <= 500. For example,
clip            bitrate  static-thresh psnr    time
akiyo(cif)       500        0          48.923  5.635s(50f)
akiyo            500        500        48.863  4.402s(50f)

parkjoy(1080p)   4000       0          30.380  77.54s(30f)
parkjoy          4000       500        30.384  69.59s(30f)

sunflower(1080p) 4000       0          44.461  85.2s(30f)
sunflower        4000       500        44.418  78.1s(30f)

Higher static-thresh values give larger speedup with larger
quality loss.

Change-Id: I857031ceb466ff314ab580ac5ec5d18542203c53
2013-07-25 14:28:05 -07:00
Johann
6c8ef8d957 Add const to vp9_accum_mv_refs parameter
Change-Id: I0625d8ffddf590dfecd1bb8b8d6f57ef64b8bf18
2013-07-25 14:25:33 -07:00
Dmitry Kovalev
7131cb0e3d General cleanups.
Removing unused constants, macros, and function declarations. Using
ROUND_POWER_OF_TWO macro, vp9_zero, vp9_copy where possible. Moving
#include from *.h to *.c. Merging for loops for motion vectors.

Change-Id: Ic3bf841764a2bb177128bb3a6d7aa8f68229cd13
2013-07-25 14:13:48 -07:00
Dmitry Kovalev
d53fc9ee4e Merge "Adding lookup table for size group." 2013-07-25 13:57:28 -07:00
Dmitry Kovalev
08fd41ccd7 Adding lookup table for size group.
Change-Id: Ia6144d77ebed66e0739b62e4d673e26a95aa9550
2013-07-25 12:58:54 -07:00
Adrian Grange
e862c6f9eb Merge "Simplify handling of sub-partition motion vectors" 2013-07-25 12:58:38 -07:00
Adrian Grange
6f0f0e4907 Merge "Use local variables rather than structure members" 2013-07-25 12:57:52 -07:00
Dmitry Kovalev
be00d3970d Merge "Removing duplicated code for merging two probabilities." 2013-07-25 12:52:26 -07:00
Dmitry Kovalev
d604914f09 Merge "Removing vp9_adapt_mode_context function." 2013-07-25 12:46:31 -07:00
Jingning Han
d571af76d3 Merge "Make coeff_optimize initialized per-plane" 2013-07-25 12:46:14 -07:00
Dmitry Kovalev
f7ece83141 Merge "Inlining inc_mv_component_count function." 2013-07-25 12:45:23 -07:00
Dmitry Kovalev
9f8335d091 Merge "Removing duplicated PREDICTION_PROBS constant." 2013-07-25 12:45:03 -07:00
Yaowu Xu
51a8458822 Merge "fix a bug where flags are not reset" 2013-07-25 12:18:51 -07:00
Adrian Grange
be700e140a Simplify handling of sub-partition motion vectors
Simplified the code that extracts and uses the motion
vectors for the 4 sub-partitions in rd_pick_partition.

Change-Id: Iaf698ef7ee3aef9edd59015e1ae065dd359b17d9
2013-07-25 11:51:51 -07:00
Jingning Han
2f58faffa4 Make coeff_optimize initialized per-plane
This commit makes the initialization of trellis coeff optimization
a per-plane operation, thereby eliminating the redundant steps in
encode_sby and encode_sbuv. It makes the encoder at speed 0 slightly
faster.

Change-Id: Iffe9faca6a109dafc0dd69dc7273cbdec19b17cd
2013-07-25 11:44:29 -07:00
Dmitry Kovalev
778989a097 Removing duplicated PREDICTION_PROBS constant.
Already defined in vp9_seg_common.h.

Change-Id: I5a0e3fa15966b1ebeb77ccd506b55fc231c22342
2013-07-25 11:08:21 -07:00
Dmitry Kovalev
47d61f008f Removing vp9_adapt_mode_context function.
Moving code from vp9_adapt_mode_context to vp9_adapt_mode_probs.

Change-Id: I60829c30b28968cd813551ef3a206dfb98d323c9
2013-07-25 10:48:45 -07:00
Yaowu Xu
3e386aefc2 fix a bug where flags are not reset
The feature that uses small partition results as a measure to skip
mode evaluation at larger partition requires the flags to be reset.
The reset was missing in the code path that calls rd_use_partition().

Change-Id: Ia0a3a0aee1a862b6e2333d596808db7c48033d50
2013-07-25 10:28:38 -07:00
Jingning Han
242157c756 Merge "SSE2 inverse 4x4 2D-DCT with DC only" 2013-07-25 08:49:37 -07:00
Scott LaVarnway
a0e8b45fee Merge "pack_inter_mode_mvs cleanup" 2013-07-25 04:47:56 -07:00
Jingning Han
384e37e32b SSE2 inverse 4x4 2D-DCT with DC only
Add SSE2 implementation to handle the special case of inverse 2D-DCT
where only DC coefficient is non-zero.

Change-Id: I2c6a59e21e5e77b8cf39a4af5eecf4d5ade32e2f
2013-07-24 23:19:56 -07:00
Jingning Han
91fa12429c Merge "Merge vp9_dc_only_idct_add and vp9_short_idct4x4_1" 2013-07-24 23:18:24 -07:00
Dmitry Kovalev
40358dc406 Removing duplicated code for merging two probabilities.
Adding common merge_probs and merge_probs2 functions. Changing ints to
usigned ints in some places.

Change-Id: Icf088ffdea7cf5b95284a128916409bdd53506b0
2013-07-24 17:44:04 -07:00
Dmitry Kovalev
4450fa4cd9 Inlining vp9_init_mode_contexts function.
Change-Id: I21ee76bcae101cc9f6ef1d867622e50b7ae565fc
2013-07-24 17:03:03 -07:00
Jingning Han
d2de1ca37b Merge vp9_dc_only_idct_add and vp9_short_idct4x4_1
They share the same functionality, so merging together.

Change-Id: I98a0386fcee052cb854f9ff90c283c1b844bcb79
2013-07-24 16:51:15 -07:00
Dmitry Kovalev
fcc34796d2 Removing CONFIG_BALANCED_COEFTREE experiment.
Change-Id: I61a8b0101eac3ee2e0621d56151b90c269fd4db4
2013-07-24 15:53:42 -07:00
Dmitry Kovalev
1787b00214 Merge "Adding condition inside get_tx_type_{4x4, 8x8, 16x16}." 2013-07-24 15:23:22 -07:00
Dmitry Kovalev
0064958c71 Inlining inc_mv_component_count function.
Change-Id: Ic99d07a56b1752ec49fc5074b1dd6804b17609a0
2013-07-24 15:03:00 -07:00
Dmitry Kovalev
9139ee0908 Adding condition inside get_tx_type_{4x4, 8x8, 16x16}.
Adding plane type check condition because it was always used outside of
get_tx_type_{4x4, 8x8, 16x16}.

Change-Id: I02f0bbfee8063474865bd903eb25b54d26e07230
2013-07-24 12:55:45 -07:00
James Zern
9e29b4cd54 Merge "vp9_find_mv_refs_idx: remove unused split_count" 2013-07-24 12:49:15 -07:00
James Zern
e6c0387edd vp9_find_mv_refs_idx: remove unused split_count
variable was write only

Change-Id: I04b002178f66961836ee08fb60a05b91b54e91d8
2013-07-24 11:51:37 -07:00
Adrian Grange
4cfd36d8fd Use local variables rather than structure members
Although local copies of the mode member variables
(mode, ref_frame) were made, they were not used in
all places. Also, made a local copy of the
second_ref_frame member.

Change-Id: I84d8c822e5cb3d8a02fc3de8a4037ca3fea8bfad
2013-07-24 11:17:44 -07:00
Adrian Grange
a183f17d33 Merge "Correct spelling mistakes" 2013-07-24 09:48:57 -07:00
Ronald S. Bultje
7817d3221f Save pixels instead of coefficients in intra4x4 RD loop.
Prevents doing duplicate IDCTs; encoding of first 50 frames of bus
(speed 0) @ 1500kbps goes from 1min4.0 to 1min3.5, i.e. 0.87% faster
overall.

Change-Id: I2df39e29ed9d5ea5e7d2704a34940ba622832ddd
2013-07-24 09:03:20 -07:00
Ronald S. Bultje
b72ecbb1b9 Add best_rd breakout in intra4x4 RD loop.
Encoding time of first 50 frames of bus (speed 0) @ 1500kbps goes from
1min5.4 to 1min4.0, i.e. 2.2% faster overall.

Change-Id: I8c32f2aff9a649ce7dd49d910dc5ba16b99c3bc6
2013-07-24 09:02:05 -07:00
Adrian Grange
bc8b0529db Correct spelling mistakes
Change-Id: Id4138293efeac4503b2e01ce7a6c150a5abeef77
2013-07-24 07:58:26 -07:00
Ronald S. Bultje
47336afd8d Merge "More optimizations for cost_coeffs()." 2013-07-23 21:36:12 -07:00
Jingning Han
666c266623 Merge "Unify the use of encode_b_args/optimize_block_args" 2013-07-23 18:08:50 -07:00
Dmitry Kovalev
1099a436d3 Moving counts from FRAME_CONTEXT to new struct FRAME_COUNTS.
Counts are separate from frame context. We have several frame contexts but
need only one copy of all counts.

Change-Id: I5279b0321cb450bbea7049adaa9275306a7cef7d
2013-07-23 17:02:08 -07:00
Jingning Han
ab77828b36 Unify the use of encode_b_args/optimize_block_args
The struct optimize_block_args is defined same as encode_b_args.
Remove this redundant definition, and use encode_b_args consistently.

Change-Id: I1703aeeb3bacf92e98a34f4355202712110173d9
2013-07-23 16:04:02 -07:00
Dmitry Kovalev
8d13b0d1df Removing LOW_PRECISION_MV_UPDATE define.
Change-Id: I78d16ee758e1fae0200b746f00031f6d9c6d6ce7
2013-07-23 15:41:45 -07:00
Dmitry Kovalev
a9bbabd94b Merge "Removing vp9_is_interpolating_filter array." 2013-07-23 15:01:19 -07:00
Adrian Grange
719cd35f3a Merge "Rolled-up several for loops into one" 2013-07-23 15:00:06 -07:00
Adrian Grange
646edbc1b2 Rolled-up several for loops into one
Several consecutive for loops executed over the same
index range, so I rolled them into one.

Change-Id: I5cfcc8c38c738478965768409cca9d09adf224e1
2013-07-23 14:32:21 -07:00
Dmitry Kovalev
db7f5d28b9 Removing vp9_is_interpolating_filter array.
All filters are interpolating now, so we don't need this array, all
values from this array are evaluated to true.

Change-Id: I9af6d8219ae0eb984063cd15e4e2296374ae4961
2013-07-23 14:24:39 -07:00
Dmitry Kovalev
2855d8aea1 Merge "Adding update_tx_counts function." 2013-07-23 13:57:59 -07:00
Dmitry Kovalev
0d59d6efcd Merge "Removing MODE_COUNT_TESTING from vp9_entropymode.c." 2013-07-23 13:57:05 -07:00
Jingning Han
825f676ceb Merge "Make xform_quant operations tx_type independent" 2013-07-23 13:40:27 -07:00
Dmitry Kovalev
9c2c17dec7 Merge "Cleanup inside vp9_get_pred_context_tx_size." 2013-07-23 12:45:49 -07:00
Dmitry Kovalev
a97d4ab123 Removing MODE_COUNT_TESTING from vp9_entropymode.c.
Change-Id: I5367bc1d9e660d86879d285a6f146d8a47e62464
2013-07-23 12:37:41 -07:00
Jingning Han
e9e2fe8ec3 Make xform_quant operations tx_type independent
The xform_quant() module is only used by inter modes, hence removing
the redundant switches therein conditioned on tx_type.

Change-Id: Ib87ce5b2f2e4cbf3ceb133a1108afa173c933a3f
2013-07-23 12:37:25 -07:00
James Zern
8dede954c7 Merge "vp9: make some static tables const" 2013-07-23 11:37:01 -07:00
Jingning Han
4ef1d35abf Merge "Skip inverse transform when eob is zero" 2013-07-23 10:31:19 -07:00
James Zern
c3871f8f70 Merge "VP9_COMMON: remove unused temp_scale_frame" 2013-07-23 10:30:55 -07:00
Deb Mukherjee
9360fd3dcf Merge "Diamond search change to accelerate movement" 2013-07-23 10:14:10 -07:00
Jingning Han
0359ad7f9a Skip inverse transform when eob is zero
When all the transform coefficients were quantized to zero, skip
the inverse transform operation. For bus_cif at 1000 kbps, the
runtime goes from 154967ms -> 149842ms, i.e., about 3% speed-up,
at speed 0.

Change-Id: Ic0a813fff5e28972d4888ee42d8747846a6c3cc6
2013-07-23 10:06:41 -07:00
Paul Wilkins
cedd24ec61 Merge "Renaming of segment constants." 2013-07-23 08:16:12 -07:00
Scott LaVarnway
7bc294a3fe pack_inter_mode_mvs cleanup
xd->mode_info_context is set to m prior to this call.

Change-Id: Ibc442529961750c29ccf0c6cae08cb2b0431415f
2013-07-23 10:08:28 -04:00
Jim Bankoski
256ee00093 Merge "clean up bw, bh" 2013-07-23 06:58:28 -07:00
Jim Bankoski
86a9dec73c clean up bw, bh
many structures use bw and bh and they have different meanings.   This cl attempts
to start this clean up and remove unneccessary 2 step look up log and then
shift operations...

also removed partition type multiple operation code in bitstream.c.

Change-Id: I7e03e552bdfc0939738e430862e3073d30fdd5db
2013-07-23 06:51:44 -07:00
Scott LaVarnway
2fd20eb37d Merge "Eliminated prev_mip memsets/memcpys in encoder" 2013-07-23 06:43:52 -07:00
Paul Wilkins
7c134bc0cd Merge "Reworked the auto_mv_step_size speed feature" 2013-07-23 04:49:55 -07:00
Paul Wilkins
32042af14b Renaming of segment constants.
Renamed:
  MAX_MB_SEGMENTS to MAX_SEGMENTS
  MB_SEG_TREE_PROBS to SEG_TREE_PROBS

The minimum unit for segmentation in the segment map
is now 8x8 so it is misleading to use MB_ as macro-block
traditionally refers to a 16x16 region.

Change-Id: I0b55a6f0426bb46dd13435fcfa5bae0a30a7fa22
2013-07-23 12:09:04 +01:00
James Zern
3c8cce353f vp9: make some static tables const
Change-Id: I8bcae51271673da8755c66a51aea005dfe6a3739
2013-07-22 19:19:13 -07:00
Frank Galligan
e88db77892 Merge "Speedup loopfilter neon code." 2013-07-22 17:39:42 -07:00
Dmitry Kovalev
0ad079e583 Cleanup inside vp9_get_pred_context_tx_size.
Using max_txsize_lookup to get max transform size.

Change-Id: If4b39beba3c06a581effd8cab698ea90727dc2c9
2013-07-22 17:18:11 -07:00
James Zern
ab139094ed Merge "VP9_COMMON: drop cur_tile_{row,col}_idx" 2013-07-22 17:12:39 -07:00
Frank Galligan
5af6bf6c43 Speedup loopfilter neon code.
Try and cut down the cycle count by rearranging the instructions
so there are less stalls.

Change-Id: Ic1383335ee0f05e656477d9ee9c179ec231285d5
2013-07-22 17:00:01 -07:00
Ronald S. Bultje
e20fcd9585 More optimizations for cost_coeffs().
4x4:    163 ->  123 cycles (33% faster)
8x8:    491 ->  399 cycles (23% faster)
16x16: 1889 -> 1763 cycles (7% faster)
32x32: 8311 -> 8180 cycles (1.6% faster)

Overall encoding time of first 50 frames of bus (speed 0) @ 1500kbps
goes from 1min4.33 to 1min3.00, i.e. 2.11% faster.

Change-Id: Ib52d1dbb5649b14de769d3e7a74af67440b5284f
2013-07-22 16:09:09 -07:00
James Zern
38a4412e1b vp9: apply loopfilter inline if possible
excludes tiled content currently

Change-Id: I44155253e8d6771e5e039d663be5f21cc9d0355d
2013-07-22 15:52:10 -07:00
Dmitry Kovalev
b2fc6fa969 Adding update_tx_counts function.
Moving common encoder/decoder code to update_tx_counts. Also renaming
vp9_get_pred_probs_tx_size to get_tx_probs2 and adding get_tx_probs to
call vp9_get_pred_context_tx_size inside read_selected_tx_size only once
(twice before).

Change-Id: Ia50247f3893de88ef8e9041b0d44be44a40aaa4d
2013-07-22 14:57:43 -07:00
James Zern
746154d905 Merge "filter_block_plane: remove MACROBLOCKD param" 2013-07-22 13:43:34 -07:00
James Zern
0a58f462b8 VP9_COMMON: remove unused temp_scale_frame
Change-Id: I696a0dca1d02d365e283029d1d077710bd5680e0
2013-07-22 13:42:11 -07:00
Dmitry Kovalev
0c5a383b2a Merge "Using update_ct and update_ct2 functions for probability update." 2013-07-22 13:34:30 -07:00
James Zern
ccf6710dc2 VP9_COMMON: drop cur_tile_{row,col}_idx
these were only being written in one location and never read.

Change-Id: If59f3c09aa1485cf89bac0099a8a79e99688b5d1
2013-07-22 13:23:33 -07:00
Yaowu Xu
6261d79206 Merge "fix a build error" 2013-07-22 13:02:15 -07:00
James Zern
76db4d599a Merge "VP[89]_COMMON: remove golden/altref frame counts" 2013-07-22 12:55:07 -07:00
Yaowu Xu
fc186dcad6 fix a build error
Change-Id: I3b05687f439ff6a7c426d2c97a6c58c831fa51ac
2013-07-22 12:37:30 -07:00
Jingning Han
416f315e82 Merge "Skip buffer update in sub8x8 rd loop" 2013-07-22 12:08:22 -07:00
Jingning Han
a5a9f5f7f3 Merge "Optimize operation flow in sub8x8 rd loop" 2013-07-22 12:08:15 -07:00
Dmitry Kovalev
8c5ca9ff14 Using update_ct and update_ct2 functions for probability update.
Update logic for both mode and mvref was the same, so using MODE_COUNT_SAT,
MODE_MAX_UPDATE_FACTOR, update_ct, update_ct2 for both cases. Removing
function update_tx_ct because it was identical to update_mode_ct2.

Change-Id: Iff566be27dbd6cde4c2ec04e8d988f207046b8f0
2013-07-22 12:06:43 -07:00
Deb Mukherjee
a1e2d50be9 Diamond search change to accelerate movement
Optional change in diamond search to continue in the best move
direction until that move turns worse.

This is still WIP since the exact way the new method is to be used is
under investigation. One option is to make it an option in diamond
search and use it only when motion is large.

Overall slightly positive on derfraw300 +0.02%, stdhdraw +0.13%,
but works a lot better for high motion sequences (ex. football : +1%).

Change-Id: If88e01a6021daa0cda934680cdc70be1ee04f798
2013-07-22 11:19:15 -07:00
Paul Wilkins
3798d7a641 Merge "Re-order mode search in rd." 2013-07-22 10:46:04 -07:00
Jingning Han
409e77f2d4 Optimize operation flow in sub8x8 rd loop
Stack the rate-distortion statistics in the sub8x8 rd loop. This allows
the encoder to skip the forward transform, quantization, and coeff cost
estimation, in the sub8x8 rd optimization search, if the motion
vector(s) are of integer pixel value, and have been tested in the
previous prediction filter type rd loops of the same block.

This gives about 2% speed-up for bus_cif at 2000 kpbs, for speed 0.
Its efficacy depends how frequently the motion search will select an
integer motion vector.

Change-Id: Iee15d4283ad4adea05522c1d40b198b127e6dd97
2013-07-22 10:40:33 -07:00
Paul Wilkins
1d189d6464 Re-order mode search in rd.
Mode search order in rd loop changed to better reflect
observed hit counts.

Also some adjustment of the baseline mode rd thresholds
to reflect the order change and observed frequencies.

Change-Id: I47a131cc83e11551df8add6d6d8d413d78d3a63c
2013-07-22 17:21:12 +01:00
Jim Bankoski
9ad604c6fb Merge "fix left over overflow" 2013-07-22 08:51:26 -07:00
Jim Bankoski
2ac8b50cd8 fix left over overflow
This cl fixes issues rbultje brought up. that I somehow neglected when I
submitted yaowu's patch.

Change-Id: I07ad18796317822510b96e951c88d29f194a3c2e
2013-07-22 06:39:39 -07:00
Paul Wilkins
888375d243 Fix build error.
When CONFIG_POSTPROC is set there was a now
invalid reference to cm->filter_level.

Changed to cpi->mb.e_mbd.lf.filter_level in line with
change Iaf5fb71c33719cdfa1b991f671caf071be9ea035

Change-Id: If746e60044903f7ba8d0d346225b3d015226c7d0
2013-07-22 14:01:43 +01:00
Dmitry Kovalev
ee1fe2f750 Merge "Removing pre probabilities from FRAME_CONTEXT." 2013-07-20 22:50:32 -07:00
Dmitry Kovalev
8962d975b2 Merge "Moving all loop filter related variables into new struct." 2013-07-20 22:45:24 -07:00
Dmitry Kovalev
39342db138 Merge "Consistent names for inter mode probabilities and encodings." 2013-07-20 22:40:51 -07:00
Dmitry Kovalev
f66821afbb Merge "Removing frame_type field from MACROBLOCKD struct." 2013-07-20 22:40:06 -07:00
Dmitry Kovalev
2b089f149a Merge "Removing unused static arrays from vp9_reatectrl.c." 2013-07-20 22:39:33 -07:00
Dmitry Kovalev
ad46753378 Merge "Moving vp9_reader into decode_tiles function." 2013-07-20 22:39:22 -07:00
Jingning Han
c725502bf3 Skip buffer update in sub8x8 rd loop
This commit allows the encoder to skip a few buffer update steps in
rd_pick_best_mbsegmentation, when early breakout has been triggered
in the rd_check_segment_txsize. It provides about 1% speed-up for
bus_cif at 2000 kbps, in the settings of speed 0.

Change-Id: Ica034f10a24dec572b397d8389a2b81020ebc0b9
2013-07-20 21:38:12 -07:00
Yaowu Xu
ea284d6281 added checks to prevent rate/distortion overflow
At speed 2, due to the threshold scheme used, it is possible the rate
and distortion assigned with INT_MAX value. The patch added checking
to prevent the INT_MAX value is used in further calculation of RD
scores. The patch also changed the assertion in rd_use_partition() to
be mirror similar assertion in rd_pick_partition().

Change-Id: Idb52c543cc1e10abdf6e6a5d6e9cb535a42214dc
2013-07-19 17:52:50 -07:00
Dmitry Kovalev
7e703de729 Removing pre probabilities from FRAME_CONTEXT.
Using cm->frame_contexts[cm->frame_context_idx] as source of previous
probabilities.

Change-Id: Ie03778acf0e7bebdc3a1f6a51854d4a0712f24a1
2013-07-19 17:33:10 -07:00
Dmitry Kovalev
ee1771ebaa Moving all loop filter related variables into new struct.
Adding loopfilter struct with fields from MACROBLOCKD and VP9Common.
Eventually it will be moved to vp9_loopfilter.h for better code structure.

Change-Id: Iaf5fb71c33719cdfa1b991f671caf071be9ea035
2013-07-19 16:19:10 -07:00
Dmitry Kovalev
f00a237a43 Merge "Fixing problem introduced in one of my previous commits." 2013-07-19 16:14:21 -07:00
Dmitry Kovalev
29f0f79317 Removing unused static arrays from vp9_reatectrl.c.
Removed arrays: kf_boost_seperation_adjustment,
                gf_adjust_table,
                gf_intra_usage_adjustment,
                gf_interval_table.

Change-Id: I62e400cb6e4d039787615169a3779e31ebf95893
2013-07-19 15:55:09 -07:00
Dmitry Kovalev
c3a56ee583 Merge "Moving Scale2Ration function from vp9_onyx.h to vp9_onyx_if.c." 2013-07-19 15:27:24 -07:00
Dmitry Kovalev
2fc927c66a Fixing problem introduced in one of my previous commits.
Changing fc->tx_probs back to fc->pre_tx_probs. This change actually
affects the bitstream but current test vectors work. Chrome branch is not
affected at all. Broken since:

cc662dd Adding struct tx_probs and struct tx_counts to cleanup the code.

Change-Id: I36dd4b3678e902e10aba8dd49b0012eb558c209d
2013-07-19 15:18:43 -07:00
Deb Mukherjee
302698fb12 Reworked the auto_mv_step_size speed feature
This patch modifies the auto_mv_step_size speed feature to
use a combination of the maximum magnitude mv from the last
inter frame, and the maximum magnitude mv for the two reference
mvs with the same reference. For arf frames, the max mav step
for the resolution is used.
The bounds therefore are slightly tighter. The feature is made
a speed 1 feature.

Rebased.

Results (when this feature is turned on over speed 0):
derfraw300: -0.046% psnr, about 5+% speedup
(tested on football: goes from 4m30.760s to 4m17.410s).

Change-Id: If492797a61b0b4b3e58c0b8f86afb880165fc9f6
2013-07-19 15:12:56 -07:00
James Zern
de012cec4f filter_block_plane: remove MACROBLOCKD param
replace with direct use of the plane and MODE_INFO

Change-Id: Icce57bc398a6e3607aedde0573d977e192040696
2013-07-19 14:19:55 -07:00
Dmitry Kovalev
d6e74e0d59 Moving vp9_reader into decode_tiles function.
Change-Id: Ic741054836d6c1b89c4f1c75cc6bd938a7d56723
2013-07-19 12:27:56 -07:00
Dmitry Kovalev
e71a4a77bb Merge "Renaming TXFM_MODE to TX_MODE (like TX_SIZE, TX_TYPE)." 2013-07-19 12:14:32 -07:00
Dmitry Kovalev
97e96bc4e9 Removing frame_type field from MACROBLOCKD struct.
Change-Id: Ia4e83913251c1cdc7aa2abd64bf01ecb1a962119
2013-07-19 11:55:36 -07:00
Dmitry Kovalev
c0eb57406c Renaming TXFM_MODE to TX_MODE (like TX_SIZE, TX_TYPE).
Moving TX_MODE enum to vp9_enums.h. Renaming txfm_mode variables to
tx_mode.

Change-Id: I459d1af6dd928ce7fccdf8ce30b6f1ca057bef92
2013-07-19 11:37:13 -07:00
Dmitry Kovalev
afe43d4089 Removing redundant VP9_COMMON* from function signatures.
Functions: vp9_get_pred_context_switchable_interp,
           vp9_get_pred_context_intra_inter,
           vp9_get_pred_context_single_ref_p1,
           vp9_get_pred_context_single_ref_p2.

Change-Id: I3d6fb8aee23c9062270768e1e6da416dd9bb8f96
2013-07-19 11:20:49 -07:00
Dmitry Kovalev
bc7acb134b Consistent names for inter mode probabilities and encodings.
Renaming vp9_sb_mv_ref_tree to vp9_inter_mode_tree, and
vp9_sb_mv_ref_encoding_array to vp9_inter_mode_encodings.

Change-Id: I0e91fbf81350d3ec5a2599064c74089b5d06133a
2013-07-19 10:40:04 -07:00
Paul Wilkins
2b8ae5a0f7 Merge "Alignment of THR_MODES to vp9_mode_order[]" 2013-07-19 10:15:26 -07:00
Paul Wilkins
b2b5836a16 Merge "Block index variables in MACROBLOCKD reduced to chars." 2013-07-19 10:14:52 -07:00
hkuang
97dbee00dd Merge "Add neon optimize vp9_short_idct8x8_add." 2013-07-19 08:28:39 -07:00
Paul Wilkins
f3ed9f5523 Alignment of THR_MODES to vp9_mode_order[]
Change-Id: I4032dd0442043543954dcb3724df974b7cc7e515
2013-07-19 11:33:39 +01:00
Paul Wilkins
710d10c521 Block index variables in MACROBLOCKD reduced to chars.
Change-Id: I9a4df095732d561807de01a41dcb1a1960726a3c
2013-07-19 11:32:51 +01:00
Dmitry Kovalev
13253d6121 Merge "Removing kf_{y, uv}_mode_prob arrays from VP9Common." 2013-07-19 01:00:46 -07:00
Yaowu Xu
73ff56b68e Merge "Fix slightly quality drop caused at speed 1." 2013-07-18 18:27:49 -07:00
Dmitry Kovalev
b829a9d63c Merge "Removing unused int_mv32 union." 2013-07-18 17:56:11 -07:00
Ronald S. Bultje
e4686c589e Fix slightly quality drop caused at speed 1.
We would skip the rectangular blocks for sub8x8 partitions because
we would conclude that PARTITION_NONE was better than PARTITION_SPLIT,
however, that conclusion was made before we actually really tested
PARTITION_SPLIT.

Change-Id: I8fa91e59894badc1d8cee3ba8a49e40ae4c4a489
2013-07-18 17:52:08 -07:00
Yaowu Xu
37d901a47a Merge "Add best_rd breakout to keyframe partition selection also." 2013-07-18 17:50:39 -07:00
Yaowu Xu
67fb0679ee Merge "Merge scale_factors and scale_factors_uv." 2013-07-18 17:50:34 -07:00
Yaowu Xu
55b52e32da Merge "Do in-place UV intra mode selection." 2013-07-18 17:50:07 -07:00
Yaowu Xu
51972d1279 Merge "Change break statement in a 2d loop to a return statement." 2013-07-18 17:49:58 -07:00
Dmitry Kovalev
92f4198d52 Merge "Using VP9_REF_NO_SCALE instead of (1 << VP9_REF_SCALE_SHIFT)." 2013-07-18 17:29:05 -07:00
hkuang
d757de744c Add neon optimize vp9_short_idct8x8_add.
Change-Id: Ic32acf3e2939c6d12d9c2bf192a5f5da59705fda
2013-07-18 16:40:41 -07:00
Dmitry Kovalev
0b562b2d3d Using VP9_REF_NO_SCALE instead of (1 << VP9_REF_SCALE_SHIFT).
Change-Id: Ide58a74d31ff948319445a6337d2c05e98720e34
2013-07-18 15:12:46 -07:00
Ronald S. Bultje
cf8988bda8 Merge "Remove motion vectors from PARTITION_INFO." 2013-07-18 15:10:41 -07:00
Dmitry Kovalev
39cea29bb7 Merge "Removing unused mv_bias and check_mv_bounds functions." 2013-07-18 14:41:52 -07:00
Dmitry Kovalev
9f193ac75f Merge "Removing unused members of VP9Decompressor: mbc, prob_skip_false." 2013-07-18 14:41:43 -07:00
Ronald S. Bultje
96e4db2660 Add best_rd breakout to keyframe partition selection also.
Change-Id: I96b8058f6dfecf8aa3e152cdcbfd7e10071fbbc9
2013-07-18 14:10:56 -07:00
Ronald S. Bultje
5ebe503f04 Merge scale_factors and scale_factors_uv.
This prevents a duplicate memcpy of a 128-byte struct every time
set_scale_factors() is called (which is a lot), thus leading to a
decrease from 3.7 MB to 1.85 MB of struct copying per 64x64 block
RD/partition loop.

Overall, this decreases encoding time of the first 50 frames of bus
@ 1500kbps (speed 0) from 1min5.9 to 1min4.9, i.e. about a 1.5%
overall speedup. We can likely get more gains by removing the copy
of the other struct (and replacing it with an indexing) as well.

Change-Id: I3dceb7e79f71e6fe911b11cc994cf89a869dde7a
2013-07-18 14:10:56 -07:00
Ronald S. Bultje
df4b4fab26 Do in-place UV intra mode selection.
This means we only do UV intra mode selection if we find any intra
mode to actually be useful at all; in addition, we only do UV intra
mode selection for the transform sizes that were selected, rather
than all sizes available in this partition.

First 50 frames of bus @ 1500kbps (speed 0) gains about 5% with this
change.

Change-Id: I7b461eb8b803247f57896c5a9505f745b55502b3
2013-07-18 14:10:56 -07:00
Ronald S. Bultje
e54a5782b9 Change break statement in a 2d loop to a return statement.
The break statement only breaks out of the nested loop, not the
top-level loop, so it doesn't always work as intended. Changing it
to a return statement does what's intended.

Change-Id: I585419823b39a04ec8826b1c8a216099b1728ba7
2013-07-18 14:10:56 -07:00
Ronald S. Bultje
2d4929e340 Remove motion vectors from PARTITION_INFO.
The same information already exists in union b_mode_info.

Change-Id: Iac5086b99a3c3cc270380138062bb693e58f9e6d
2013-07-18 14:10:52 -07:00
James Zern
5f30a0c687 VP[89]_COMMON: remove golden/altref frame counts
these are only used in the encoder.
frames_since_golden / frames_till_alt_ref_frame -> VP[89]_COMP

Change-Id: Ie14a6f46987bced685ddb449b85dc261caba6dfe
2013-07-18 14:09:21 -07:00
Dmitry Kovalev
9f3c0e34a9 Moving Scale2Ration function from vp9_onyx.h to vp9_onyx_if.c.
Change-Id: Idfe2a850f72b38f519aea1aac1266d8c3aa813ee
2013-07-18 14:05:06 -07:00
Ronald S. Bultje
9da67da04a Merge "Fix bug where we don't choose any mode in RD selection." 2013-07-18 12:47:50 -07:00
Ronald S. Bultje
247197d57b Fix bug where we don't choose any mode in RD selection.
This could happen during golden overlay frame coding from a previous
alt-ref frame if the special overlay code was triggered.

Change-Id: I3056d0c547cd26903b260ef93c94026e96bd9868
2013-07-18 12:13:15 -07:00
Dmitry Kovalev
3de20f3ea9 Removing unused members of VP9Decompressor: mbc, prob_skip_false.
Change-Id: Id5480a4fd56c184ad046c2192b30d190debb3de0
2013-07-18 11:29:34 -07:00
Dmitry Kovalev
b3c0a5fddb Removing unused mv_bias and check_mv_bounds functions.
Change-Id: I1558fd969d9ad112bf6480bdd16ef87edd396ab5
2013-07-18 11:20:48 -07:00
Frank Galligan
5769713608 Merge "Fix horz loopfilter loops" 2013-07-18 10:21:21 -07:00
Ronald S. Bultje
4f5815290c Merge "Fix bug which skips zeromv even if near/nearest is not 0,0." 2013-07-18 10:06:51 -07:00
Frank Galligan
7fd5d8e6a4 Fix horz loopfilter loops
If count was greater than 1 the src pointer would be off on
the second loop.

Change-Id: I8e09037e68dc4ae92076a8067f7b6dacbbef8263
2013-07-18 09:44:15 -07:00
Ronald S. Bultje
deb7456058 Fix bug which skips zeromv even if near/nearest is not 0,0.
Change-Id: Id4f454831f3f11099f39c30246adeaa52857d08d
2013-07-18 09:35:19 -07:00
Jingning Han
ced3c20165 Use mv_check_bounds in sub8x8 rd loop
Make the use of mv_check_bounds consistent for mvs of both ref_frame[0]
and ref_frame[1].

Change-Id: I1ca24865cc7232ca9cbe5db566c53abad1592211
2013-07-17 17:13:51 -07:00
Dmitry Kovalev
7363e668d5 Removing unused int_mv32 union.
Change-Id: Ie692ed6e5fa1d2122e3a03573914d0fcce842f9e
2013-07-17 17:02:44 -07:00
Dmitry Kovalev
f9f453ec8d Removing kf_{y, uv}_mode_prob arrays from VP9Common.
These arrays have constant values (no any updates). Removing two
corresponding memcpy calls. Making a little cleanup in vp9_entropymode.h
as well: removing redundant 'extern' keyword and moving all function
declarations at the end.

Change-Id: Ia16b38b46aec2e2500f5df29c40a297ae241dede
2013-07-17 16:50:52 -07:00
Ronald S. Bultje
facecd80da Merge "Add a best_yrd shortcut in splitmv mode search." 2013-07-17 16:11:13 -07:00
Ronald S. Bultje
056111c822 Merge "Skip redundant nearest/near/zero encodes in splitmv." 2013-07-17 16:10:51 -07:00
Ronald S. Bultje
0b1eba25b2 Merge "Skip nearest/near/zero redundant encodes." 2013-07-17 16:10:41 -07:00
Ronald S. Bultje
607424449c Merge "Best_rd breakout in rd partition search." 2013-07-17 16:10:22 -07:00
Yunqing Wang
b409c1d91e Merge "Remove unnecessary calling of vp9_init_quantizer()" 2013-07-17 15:47:17 -07:00
Yunqing Wang
3798db88e1 Remove unnecessary calling of vp9_init_quantizer()
vp9_init_quantizer() is called in vp9_create_compressor(), and
should not be called in vp9_set_speed_features().

Change-Id: Ic2f1f4b0531b9d46bb841d7e1d8da9812207dad6
2013-07-17 14:59:00 -07:00
hkuang
7b9a652813 Merge "Remove unnecessary buffer copy in idct4x4." 2013-07-17 14:51:53 -07:00
Yaowu Xu
6ac5b7db2c Merge "changed mode checking order" 2013-07-17 14:44:40 -07:00
Dmitry Kovalev
a7a1e96136 Merge changes Ieffea49e,Idf610746
* changes:
  Removing two unused arguments from vp9_inc_mv signature.
  Changing signature of vp9_get_pred_probs_tx_size.
2013-07-17 14:44:20 -07:00
Dmitry Kovalev
b775081283 Merge "Removing experimental code from vp9_entropymv.c." 2013-07-17 14:43:45 -07:00
Dmitry Kovalev
102632aab1 Merge "Adding read_comp_pred function." 2013-07-17 14:26:34 -07:00
Ronald S. Bultje
c6917528a5 Add a best_yrd shortcut in splitmv mode search.
Encoding of first 50 frames of bus (speed 0) @ 1500kbps goes from
1min6.2 to 1min5.9, i.e. 0.5% faster overall.

Change-Id: I59d8a3b2f0a75010fa041d5e2646c8caac5bd683
2013-07-17 14:21:57 -07:00
hkuang
bd6ce7128c Remove unnecessary buffer copy in idct4x4.
Change-Id: I386066b9bcfb4bffb582e6827af36ca0181f6a83
2013-07-17 14:20:56 -07:00
Ronald S. Bultje
161c995658 Skip redundant nearest/near/zero encodes in splitmv.
Encode of first 50 frames of bus @ 1500kbps (speed 0) goes from
1min7.3 to 1min6.2, i.e. 1.7% faster overall.

Change-Id: I19d2deacfbffadd61d32551cee9586757ab4a987
2013-07-17 13:53:48 -07:00
Yaowu Xu
42facc292d changed mode checking order
Change-Id: Ic4c4b363ed840935e42f495f13ea5e601a56f1b2
2013-07-17 13:43:50 -07:00
Ronald S. Bultje
8fea880b6f Skip nearest/near/zero redundant encodes.
Encode of first 50 frames of bus @ 1500kbps (speed 0) goes from 1min12.8
to 1min7.3, i.e. 8% faster.

Change-Id: Ia22d1c7b687316c553cc60eacae988b24e175b62
2013-07-17 11:33:15 -07:00
Yunqing Wang
10e83b0717 Enable disable_splitmv feature for other speeds
Added disable_splitmv feature at other speed levels. For speed 3 or
above, always turn it on.

Change-Id: Ibb36f0a7ef12a34b4f8d0f9cb6193eab43b34360
2013-07-17 10:25:49 -07:00
Dmitry Kovalev
8452c34551 Removing experimental code from vp9_entropymv.c.
Change-Id: I340d06e3bc32c78358654496503cccd4196cbe2e
2013-07-17 10:25:09 -07:00
Johann
9ca66ec050 Merge "vp9_convolve8_neon placeholder" 2013-07-17 10:09:00 -07:00
Ronald S. Bultje
9f427bfe98 Best_rd breakout in rd partition search.
About 15% faster for bus (speed 0) first 50 frames @ 1500kbps, which
goes from 1min36 to 1min24. Results become slightly better (+0.2% on
derf/yt, +0.4% on hd), probably because of a bugfix for skipmode in
super_block_yrd(). Overall speed change (on derfraw300) is roughly
-13%. This can probably be improved further by caching best_yrd
between partition searches. Also, we might be able to get more
speedups by always doing PARTITION_NONE before PARTITIONS_SPLIT, not
just at the sb8x8 level.

Change-Id: I83736949ebd5b4a3b400ee688d7661913fefc98b
2013-07-17 09:56:46 -07:00
Ronald S. Bultje
83c7e13a6b Do a skip-block check for sub8x8 partitions also.
+0.2% SSIM and glbPSNR on derfraw300.

Change-Id: I9cba0bca55e606a22f557c7732b064f738efe84d
2013-07-17 09:46:47 -07:00
Yunqing Wang
df90d58f4f Speed up motion estimation using small partitions' result(experiment)
Current partition checking starts from small sizes, and then goes up
to large sizes. This experiment uses the small partitions' motion
estimation result, which is already available, to speed up the
large partition's motion estimation. We can decide to skip some
patition checkings if they are unlikely choices. We could use the
motion vector(MV) result as current partition's prediction MV, limit
the search range and reference frame.

Current result at speed 1:
psnr loss: 1.19% for stdhd, 0.287% for derf.
speed gain: 14% for sunflower(hd), 11% for akiyo.

Further improvement will be done later.

Change-Id: I5abfd070e9cace2e91e2a0247d1325df313887ab
2013-07-17 09:11:47 -07:00
Johann
59dc4e9cdd vp9_convolve8_neon placeholder
Call the individually optimized horizontal and vertical functions. This
implementation abuses the temp buffer.

This will be replaced with a custom optimized function.

Over 2x speedup.

Change-Id: I5b908d2a73d264e9810d6022bbff73207a3055dd
2013-07-17 08:39:27 -07:00
Paul Wilkins
d66eab15dd Merge "Move uv intra mode selection in rd loop." 2013-07-17 05:19:26 -07:00
Paul Wilkins
154c34a3ee Merge "Limit transform sizes searched for uv intra." 2013-07-17 03:40:11 -07:00