Commit Graph

57 Commits

Author SHA1 Message Date
Scott LaVarnway
195883023b VPX: subpixel_8t_ssse3 asm using x86inc
This is based on the original patch optimized for 32bit
platforms by Tamar/Ilya and now uses the x86inc style asm.
The assembly was also modified to support 64bit platforms.

Change-Id: Ice12f249bbbc162a7427e3d23fbf0cbe4135aff2
2015-09-03 20:35:51 -07:00
Johann Koenig
5c245a46d8 Merge changes I53b5bdc5,Ib81168a7,Ie0113945
* changes:
  Only build ssse3 filter functions on 64 bit
  Clean up unused function warnings in vp8 encoder
  Clean up unused function warnings in vp8 onyx_if.c
2015-08-27 20:58:53 +00:00
Johann
a28b2c6ff0 Add sse2 versions of halfpix variance
These were lost in the great sub pixel variance move of
6a82f0d7fb

Not having these functions caused a ~10% performance regression in
some realtime vp8 encodes.

Change-Id: I50658483d9198391806b27899f2c0d309233c4b5
2015-08-27 11:58:38 -07:00
Johann
f5507b514c Only build ssse3 filter functions on 64 bit
Avoid an unused function warning by only building the functions when
they will be used.

Change-Id: I53b5bdc5a180c79d63b34e4c8921d679bbc54009
2015-08-26 10:32:18 -07:00
Scott LaVarnway
6c0f6dd817 Merge "VPX: scaled convolve : fix windows build errors" 2015-08-21 12:06:34 +00:00
Scott LaVarnway
acf24cc1b8 VPX: scaled convolve : fix windows build errors
Change-Id: Ic81d435ea928183197040cdf64b6afd7dbaf57e4
2015-08-20 13:09:27 -07:00
Scott LaVarnway
6a21ca20cc Merge "VPX ssse3 scaled convolve" 2015-08-19 22:12:21 +00:00
Jingning Han
49f6ff1103 Rename inv_txfm_sse2.asm to inv_wht_sse2.asm
Change-Id: I43bcc70680503e4c18d8f021097307778cf9ea70
2015-08-19 10:29:53 -07:00
Scott LaVarnway
2030c49cf8 VPX ssse3 scaled convolve
Change-Id: I71d5994e21813554a927d35ebcc26bf7a68984fd
2015-08-18 15:13:02 -07:00
Scott LaVarnway
6cf95bd1e7 Merge "VPX: remove step == 16 and filter[3] != 128 checks" 2015-08-12 20:13:33 +00:00
Scott LaVarnway
b04dad328c Merge "VPX: remove scaled calls from FUN_CONV_1D" 2015-08-11 21:46:50 +00:00
Scott LaVarnway
4ef08dcec8 Merge "VPX: Add rtcd support for scaling." 2015-08-11 13:19:00 +00:00
James Zern
9265bad906 Merge changes from topic 'x86inc'
* changes:
  Only use .text sections for aout
  Use newer x86inc.asm
  Use .text instead of .rodata on macho
  Copy PIC handling code from x86_abi_support
  Set 'private_extern' visibility for macho targets
  Avoid 'amdnop' when building with nasm
  Catch all elf formats
  Expand PIC default to macho64 and respect CONFIG_PIC from libvpx
  Use libvpx defines to set name mangling rules
  Customize x86inc.asm for libvpx
2015-08-10 21:20:38 +00:00
Scott LaVarnway
a229dbc1f0 VPX: remove step == 16 and filter[3] != 128 checks
from FUN_CONV_1D and FUN_CONV_2D macros.  The functions
will not be called with these inputs.

Change-Id: I67ec75e4edafc0acee70190521a80ea85dfa521b
2015-08-10 13:44:32 -07:00
Johann
41a0a0cb35 Use newer x86inc.asm
Rename updated version of x86inc.asm

Use "private_prefix" instead of "program_name" and make vpx the default
prefix.

Change-Id: I4883a99b2aee8e5dc9f2c16a2e6f4b5d6e4de458
2015-08-07 16:44:44 -07:00
Alex Converse
c65e79d2e5 ssim: Replace unsigned long with uint32_t.
The assembly only writes the low 4 bytes, and the HBD version only uses
uint32_t bytes.

Change-Id: Ie3694ecda511c231e55870df814cbae30e588073
2015-08-07 11:48:31 -07:00
Alex Converse
c7b7011b9b Move VP9 SSIM metrics to vpx_dsp.
Change-Id: I20c7b42631b579fade6cf7ebf6d4c69b2fcb5e5e
2015-08-06 18:25:25 -07:00
Aℓex Converse
7ac505c726 Merge "Narrow a load in iwht4x4_16_add." 2015-08-06 22:21:16 +00:00
Alex Converse
0572052725 Narrow a load in iwht4x4_16_add.
The top half is unused.

Change-Id: I29b2f6a93e20ea43aff4ad0bd2d52257e1e752b6
2015-08-05 12:16:12 -07:00
Scott LaVarnway
4e6b5079c6 VPX: remove scaled calls from FUN_CONV_1D
and FUN_CONV_2D macros.  The predict lut now handles
this case.  The encoder now calls vpx_scaled_2d() instead
of vpx_convolve8() for scaling.

Change-Id: Ia1c8af8a31e4cb4887a587143108cb45835f7df7
2015-08-05 10:47:06 -07:00
James Zern
afd2f68dae Revert "VP9_COPY_CONVOLVE_SSE2 optimization"
This reverts commit a5e97d874b.

Additionally:
Revert "vpx_convolve_copy_sse2: fix win64"

This reverts commit 22a8474fe7.

This change performs poorly on various x86_64 devices affecting
performance by 1-3% at 1080P. Performance on chromebook like devices was
mixed neutral to slightly negative, so there should be minimal change
there.

Change-Id: I95831233b4b84ee96369baa192a2d4cc7639658c
2015-08-04 17:57:01 -07:00
Jingning Han
d621de7e8d Change vp9_quantize to vpx_quantize
This commit clears all the vp9_ prefix use case in vpx_dsp. It gets
the vp9 folder ready to branch out vp10.

Change-Id: I2906eec179ee792b4af8c9b4161313653050e931
2015-08-04 15:31:49 -07:00
Jingning Han
08a453b9de Replace vp9_ prefix with vpx_ prefix in vpx_dsp function names
This commit clears the function naming convention in vpx_dsp. It
replaces vp9_ prefix of global functions with vpx_ prefix. It also
removes the vp9_ prefix from static functions.

Change-Id: I6394359a63b71a51dda01342eec6a3cc08dfeedf
2015-08-04 13:46:11 -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
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
b4c7d0523a Merge "Factor inverse transform functions into vpx_dsp" 2015-08-01 16:20:24 +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
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
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
Hui Su
4cbf36b105 Merge "Replace prefix vp9_ with vpx_ for intra prediction functions" 2015-07-29 00:38:48 +00:00
Jingning Han
d12a4a825c Merge "Replace vp9_ prefix in 2D-DCT functions with vpx_" 2015-07-29 00:07:31 +00:00
Jingning Han
fc18cf7a11 Merge "Move DC only forward 2D-DCT functions to vpx_dsp" 2015-07-29 00:06:37 +00:00
Jingning Han
4b5109cd73 Replace vp9_ prefix in 2D-DCT functions with vpx_
Clean up the forward 2D-DCT function names in vpx_dsp.

Change-Id: I3117978596d198b690036e7eb05fe429caf3bc25
2015-07-28 16:06:44 -07:00
Jingning Han
d19033fa4e Move DC only forward 2D-DCT functions to vpx_dsp
This completes the forward transform functions layout refactoring.

Change-Id: I996fb0fb795f41e2040f7b21db985774098aedbd
2015-07-28 14:52:30 -07:00
Johann
124ada514b Don't use 'h' for functions using x86inc.asm
In newer version of x86inc.asm 'h' is used as a modifier for register
names.

Change-Id: Ie5b9dd2f91ecdc8f6f18b2701b6dc23042b604e4
2015-07-28 14:00:32 -07:00
Hui Su
fe7cabe8b6 Merge "Move intra prediction functions from vp9/common/ to vpx_dsp/" 2015-07-28 20:41:01 +00:00
Jingning Han
a6a4659bea Factor 32x32 fwd DCT to vpx_dsp folder
Move the 32x32 2D-DCT implementations from vp9/ to vpx_dsp/.

Change-Id: Id3980696f8b69906ff7a59ff9fb2b9013d60047d
2015-07-28 11:13:41 -07:00
Jingning Han
8eefb36ca9 Move forward dct sse2 header file to vpx_dsp
Change-Id: Iba03852ce778c956200818e3473cfb2b48cf8d8e
2015-07-27 14:59:57 -07:00
hui su
4013645353 Replace prefix vp9_ with vpx_ for intra prediction functions
Change-Id: I8ae6fb586f8d5d018ace228df11714f82b085076
2015-07-27 13:42:06 -07:00
hui su
7971846a5e Move intra prediction functions from vp9/common/ to vpx_dsp/
Change-Id: I64edc26cf4aab050c83f2d393df6250628ad43b8
2015-07-27 13:38:16 -07:00
Jingning Han
5ebc8febdc Refactor vp9_idct.h file
Separate the common coefficient constant into vpx_dsp/txfm_common.h.
Move the SSE2 macro definitions to vpx_dsp/x86/txfm_common_sse2.h.
This clears the use case of vp9_idct.h in vpx_dsp folder.

Change-Id: I319735a2abf42888e5080ac14cfbcde34be7b121
2015-07-26 08:26:32 -07:00
Jingning Han
252ec59821 Remove vp9_dct.h from fwd_txfm_impl_sse2 header file
Change-Id: Ib3a4814fdb9d69cf6cc23bdd208f9bc9e7972edc
2015-07-24 21:11:44 +00:00
Jingning Han
5ddfa101c9 Add x86_64 flag to guard fwd_txfm_ssse3.asm in make file
This fixes a VS build error. Fix by @johannkoenig.

Change-Id: I6e71435d70ae56079db7328e4c7915416ece8fda
2015-07-23 12:55:50 -07:00
Jingning Han
b67821f37b Factor forward 2D-DCT transforms into vpx_dsp
This commit factors the 4x4, 8x8, and 16x16 2D-DCT forward
transform operations into vpx_dsp folder.

Change-Id: I084b117b79c0925edcbcabb93f62b9f4bf8dbe7d
2015-07-22 15:48:17 -07:00
Yunqing Wang
f65473c036 Merge "Migrate quantization functions from vp9/ to vpx_dsp/" 2015-07-20 16:20:07 +00:00
Yunqing Wang
38f1fbbb75 Migrate quantization functions from vp9/ to vpx_dsp/
The following quantization functions were moved:
vp9_quantize_b
vp9_quantize_b_32x32
vp9_highbd_quantize_b
vp9_highbd_quantize_b_32x32

vp9_quantize_dc
vp9_quantize_dc_32x32
vp9_highbd_quantize_dc
vp9_highbd_quantize_dc_32x32

The purpose of doing that was to allow these functions to be shared
by multiple codecs.

Change-Id: Id8ab939f283353cdd07bd930d47db3d932a5d87f
2015-07-17 16:38:14 -07:00
Jingning Han
2992739b5d Rename loop filter function from vp9_ to vpx_
Change-Id: I6f424bb8daec26bf8482b5d75dd9b0e45c11a665
2015-07-17 15:55:02 -07:00
Jingning Han
6e5ab70a93 Take out unnecessary header file from highbd_loopfilter_sse2
The dependency on vp9_loopfilter.h is not needed.

Change-Id: Ic0583c43d3d63f19cef06cf9d8e5c8031601be6a
2015-07-16 18:15:02 -07:00