generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
14,553 Commits 55 Branches 16 Tags
Commit Graph

14553 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
James Zern
a0fd7a9831 Merge "add vp9_vector_var_neon" 2015-08-04 02:30:41 +00:00
James Zern
d011808777 gen_msvs_proj.sh: avoid asm object name collisions 
fixes link under vs9; this is the same change as:
dbf6e3f gen_msvs_vcxproj.sh: Avoid object name collisions.

Change-Id: I2a188c9024d0605e60e5e03ddcef1a25e7e53585
 2015-08-03 16:22:21 -07:00
Johann
749c393c8d Rename vp8 loopfilter_mmx.asm 
Chromium puts all the yasm output in the same directory. Looking at ways
to improve this but in the meantime get rid of collisions.

Change-Id: I923c5231d14e895ab96521eb89807ede868a0753
 2015-08-03 14:27:03 -07:00
Jingning Han
bfad9d2fe6 Move inverse transfrom dspr2 functions from vp9 to vpx_dsp 
Change-Id: Ia9cf7c31cab4ba3dd6b9bb668c4b3e84bd55cf69
 2015-08-03 11:59:50 -07:00
Jingning Han
7723b8df6a Merge "Add common_dspr2.c file to vpx_dsp/mips" 2015-08-03 18:58:37 +00:00
Yaowu Xu
9d9273726b Merge "Correct the allocation size for ssim_vars" 2015-08-03 18:43:55 +00:00
Jingning Han
92b08f516a Add common_dspr2.c file to vpx_dsp/mips 
Move the declaration of commonly referenced variable to
vpx_dsp/mips/common_dspr2.c.

Change-Id: Ia51287b02e2ac5cfae0fba98c721f0810618f28e
 2015-08-03 10:53:47 -07:00
Yaowu Xu
8f089cbd2e Correct the allocation size for ssim_vars 
Ssim_vars is used to accumulate stats based 4x4 pixel blocks, this
commit changes the allocations size to be based on mi_rows and mi_cols
to avoid out-of-bound memory access for larger size videos. The hard
coded 720x480 can only work for image size up to 2880x1920.

Change-Id: Id9d07f3f777385b448ac88a6034b7472e4cf3c79
 2015-08-03 10:46:12 -07:00
Jingning Han
a68356202d Remove vpx_ prefix from the dspr2 file name in vpx_dsp/mips 
Make it consistent with other formats.

Change-Id: I28f0d05ff7c5bf2b815989b3f1bd6c6b25608677
 2015-08-03 09:59:14 -07:00
Scott LaVarnway
8f6b943100 VPX: Add rtcd support for scaling. 
Change-Id: If34bfb0d918967445aea7dc30cd7b55ebfedb1f2
 2015-08-03 09:43:34 -07:00
Jingning Han
d10fc5af8f Merge "Add vpx_dsp_rtcd.h to inv_txfm_sse2.c" 2015-08-03 16:03:09 +00:00
Jingning Han
b096db5ad4 Merge "Remove vp9_common.h from idct16x16_neon.c" 2015-08-03 16:03:02 +00:00
Parag Salasakar
1579bb88c5 mips msa vpx convolve optimzation 
Removed redundant clip/saturate code from 2tap filter functions
average improvement 10%-40%

Change-Id: I1dafb5f7d2ce7a021d883d8af30fb93cd9ace173
 2015-08-03 14:03:40 +05:30
Parag Salasakar
9b375871db mips msa vpx subpel variance optimization 
Removed redundant clip/saturate code from 2tap filter functions
average improvement 20%-40%

Change-Id: I362540b0c7d5d3d69932c39d61b7d2a44da533d2
 2015-08-03 13:00:55 +05:30
Parag Salasakar
4bfe3bdf27 mips msa vpx subtract test added 
Change-Id: I0f0827a665c4d3039d3e5f09fa8c75c8f2bb2bab
 2015-08-03 09:42:11 +05:30
Jingning Han
0b0eba728d Add _dspr2 to local function names 
It avoids symbol conflicts between function names of various
implementation versions.

Change-Id: Iad79ebcb8e289457801812a7745c8380b5b06a46
 2015-08-02 20:21:59 -07:00
Jingning Han
da7dc59837 Merge "Factor out mips/msa inverse transform implementations" 2015-08-03 03:18:39 +00:00
Jingning Han
0fcfc613c6 Merge "Add x86inc flag guard to inv_txfm_sse2.asm" 2015-08-02 21:56:09 +00:00
Jingning Han
6eabf229e2 Remove vp9_common.h from idct16x16_neon.c 
Change-Id: I3df35a99900ef8ce549d315866849a10db1a4c7b
 2015-08-02 09:57:25 -07:00
Jingning Han
4f7a7d29fa Add x86inc flag guard to inv_txfm_sse2.asm 
Fix the VS build failure.

Change-Id: I4fb9d1c83980c4b52d5a848a9cb02ec72493dccb
 2015-08-02 08:43:51 -07:00
Jingning Han
80ae856c8b Add vpx_dsp_rtcd.h to inv_txfm_sse2.c 
Change-Id: Ibab434fb4bd6da02dba087582ed74811f555c3ed
 2015-08-02 08:25:13 -07:00
James Zern
22a8474fe7 vpx_convolve_copy_sse2: fix win64 
xmm6-7 need to be stored

Change-Id: I6c51559598d335946ec91be6246b49589c63b724
 2015-08-01 11:45:49 -07:00
Jingning Han
44849516d4 Factor out mips/msa inverse transform implementations 
Move mips/msa inverse transform implementations from vp9 folder to
vpx_dsp.

Change-Id: Ic4cf3f05247c3c63db7b532a0e5000017a962391
 2015-08-01 09:25:12 -07:00
Jingning Han
b37494cfb5 Merge "Use precise header files in inverse transform msa implementations" 2015-08-01 16:20:43 +00:00
Jingning Han
b4c7d0523a Merge "Factor inverse transform functions into vpx_dsp" 2015-08-01 16:20:24 +00:00
Parag Salasakar
d35f992599 mips msa vp8 denoising filter optimization 
average improvement ~2x-3x

Change-Id: I6c17012c731fa4d56e0343f8de0df47b2dde289b
 2015-08-01 08:05:25 +05:30
Parag Salasakar
c1b233dd43 Merge "mips msa vp8 temporal filter optimization" 2015-08-01 02:12:20 +00:00
James Zern
fcb4253c9c update libyuv to r1456 
picks up build warning fixes for visual studio 2015

Change-Id: Idea85fa70d1aeb2a46ea355b87fe41ec5b2b9520
 2015-07-31 18:34:20 -07:00
Jingning Han
4dc390b15d Merge "Add dynamic range notes to vp9_vector_var_c" 2015-08-01 01:01:37 +00:00
James Zern
7dc5a689b4 add vp9_vector_var_neon 
~50-60% faster depending on the width

Change-Id: I9d007cfa10b9aaa2169c8c009d95522df6123a92
 2015-07-31 17:31:58 -07:00
Aℓex Converse
fd22c492f7 Merge "Turn off simple_model_rd_from_var at speed 4." 2015-07-31 23:51:01 +00:00
Jingning Han
36a9a33b90 Add dynamic range notes to vp9_vector_var_c 
Change-Id: If536ad31046ecd9e2ecd9c21f52f8192c8153ad7
 2015-07-31 16:42:09 -07:00
Jingning Han
56c2cb7553 Use precise header files in inverse transform msa implementations 
Change-Id: Ie8a79d9e2837842c3f60776b661cd42782b108d5
 2015-07-31 23:24:54 +00:00
James Zern
d8642d831f Merge "VP9_COPY_CONVOLVE_SSE2 optimization" 2015-07-31 23:22:34 +00:00
Jingning Han
e8b133c79c Factor inverse transform functions into vpx_dsp 
This commit moves the module inverse transform functions from vp9
to vpx_dsp folder. The hybrid transform wrapper functions stay in
the vp9 folder, since it involves codec-specific data structures.

Change-Id: Ib066367c953d3d024c73ba65157bbd70a95c9ef8
 2015-07-31 16:21:00 -07:00
Alex Converse
af6d2c7d42 Turn off simple_model_rd_from_var at speed 4. 
This got erroneously changed during the refactor. This fixes
SvcTest.TwoPassEncode2TemporalLayersWithMultipleFrameContextsAndTiles.

Change-Id: Ifa5ab0e098396c5e2d10478db87df256eadfa4c7
 2015-07-31 15:50:17 -07:00
James Zern
e184b613b9 Merge changes Iecdbbc34,I8b4db93f 
* changes:
  Android.mk: fix *_rtcd.h deps for armeabi-v7a
  Android.mk: add a dep on vpx_config.asm for x86_64
 2015-07-31 22:22:48 +00:00
Scott LaVarnway
a5e97d874b VP9_COPY_CONVOLVE_SSE2 optimization 
This function suffers from a couple problems in small core(tablets):
-The load of the next iteration is blocked by the store of previous iteration
-4k aliasing (between future store and older loads)
-current small core machine are in-order machine and because of it the store will spin the rehabQ until the load is finished
fixed by:
- prefetching 2 lines ahead
- unroll copy of 2 rows of block
- pre-load all xmm regiters before the loop, final stores after the loop
The function is optimized by:
copy_convolve_sse2 64x64 - 16%
copy_convolve_sse2 32x32 - 52%
copy_convolve_sse2 16x16 - 6%
copy_convolve_sse2 8x8 - 2.5%
copy_convolve_sse2 4x4 - 2.7%
credit goes to Tom Craver(tom.r.craver@intel.com) and Ilya Albrekht(ilya.albrekht@intel.com)

Change-Id: I63d3428799c50b2bf7b5677c8268bacb9fc29671
 2015-07-31 14:51:51 -07:00
Jingning Han
6025c6d65b Merge "Fix compiler warning in mips/dspr2" 2015-07-31 21:29:50 +00:00
Aℓex Converse
dd4b416412 Merge "Compute skippable inside the block_rd_txfm loop." 2015-07-31 21:19:11 +00:00
Jingning Han
135b43ccf3 Fix compiler warning in mips/dspr2 
This commit fixes the mix declaration and definition warning when
mips/dspr2 is turned on.

Change-Id: I633d6fe42368b9ac35b106786ebac6969ad53552
 2015-07-31 12:34:34 -07:00
Aℓex Converse
90e563d91f Merge changes Ic1ce346a,Ic0b4e92c 
* changes:
  Simplify model_rd_for_sb HBD ifdefs
  Simplify dist_block HBD ifdefs
 2015-07-31 19:05:54 +00:00
Alex Converse
ab20c98e84 Compute skippable inside the block_rd_txfm loop. 
Change-Id: Iaa43aeeb7a2074495e00cdb83bb551c3f13d3ed2
 2015-07-31 11:45:59 -07:00
Zoe Liu
7f8dd35329 Merge "Refactor mips/dspr2 on convolution." 2015-07-31 18:23:19 +00:00
Zoe Liu
873a158f14 Merge "Code refactor on InterpKernel" 2015-07-31 18:20:14 +00:00
Alex Converse
c62228f273 Simplify model_rd_for_sb HBD ifdefs 
Change-Id: Ic1ce346a053800ae3b2d77178f46e6a388357f6d
 2015-07-31 11:16:59 -07:00
Alex Converse
da9c73c293 Simplify dist_block HBD ifdefs 
Change-Id: Ic0b4e92cbaf813bcca8a8e9052c936c2e025e114
 2015-07-31 11:04:01 -07:00
Aℓex Converse
8abd0c2a12 Merge "Short circuit rate_block in block_rd_txfm." 2015-07-31 17:59:22 +00:00
Zoe Liu
7cfdc00337 Refactor mips/dspr2 on convolution. 
Change-Id: If59a39d5a92c261537342726f94bb7f7f26dfff3
 2015-07-31 10:27:42 -07:00
Zoe Liu
7186a2dd86 Code refactor on InterpKernel 
It in essence refactors the code for both the interpolation
filtering and the convolution. This change includes the moving
of all the files as well as the changing of the code from vp9_
prefix to vpx_ prefix accordingly, for underneath architectures:
(1) x86;
(2) arm/neon; and
(3) mips/msa.
The work on mips/drsp2 will be done in a separate change list.

Change-Id: Ic3ce7fb7f81210db7628b373c73553db68793c46
 2015-07-31 10:27:33 -07:00