summaryrefslogtreecommitdiff
path: root/vpx_dsp
AgeCommit message (Collapse)Author
2017-08-24quantize ssse3: copy implementation to intrinsicsJohann
Still does not pass tests. Does match the previous assembly, although saving the sign before multiplying is dubious. Change-Id: Ia163f18c755aba542d6e93f7bf7343184660df5a
2017-08-24Merge "vpx_dsp:loongson optimize vpx_mseWxH_c(case 16x16,16X8,8X16,8X8) with ↵Shiyou Yin
mmi."
2017-08-23Merge "quantize avx: copy implementation to intrinsics"Johann Koenig
2017-08-23Merge "vpx_dsp: get32x32var_avx2() cleanup"Scott LaVarnway
2017-08-23Merge "quantize neon: round dqcoeff towards zero"Johann Koenig
2017-08-23quantize avx: copy implementation to intrinsicsJohann
Adds an early exit based on ptest. Slightly slower than ssse3 in the full case because of the extra check, but potentially faster if lots of rows can be skipped. Very close in speed to the assembly. Can run in 32 bit, unlike the assembly. Allows reworking the function prototype to use structs. Change-Id: If80e2b9ba059370a4cad3c973196e82a97b4330e
2017-08-23quantize neon: round dqcoeff towards zeroJohann
Add 1 if negative to get dqcoeff to round towards zero. 10-15% faster than converting to positive before shifting. Change-Id: I01a62fd0c9bca786b6885b318bd447bb9229903d
2017-08-23vpx_dsp:loongson optimize vpx_mseWxH_c(case 16x16,16X8,8X16,8X8) with mmi.Shiyou Yin
Change-Id: I2c782d18d9004414ba61b77238e0caf3e022d8f2
2017-08-22quantize ssse3: copy style from sse2Johann
Change-Id: I53f8a160e640c674ea035fc112e207b6dca42598
2017-08-22quantize sse2: copy opts from ssse3Johann
Simplify eob calculations based on ssse3 implementation. General clean up and re-scoping. Change-Id: I48f282bf9bd28ee9bc2c7a6779be9d45b5a3a3ee
2017-08-22Merge changes Icfb70687,I9a963e99,Ie8ac00ef,I1272917cJohann Koenig
* changes: quantize: ignore skip_block in arm quantize: ignore skip_block in x86 quantize fp: ignore skip_block in arm quantize fp: ignore skip_block in x86
2017-08-22Merge "ppc: Add vpx_idct16x16_256_add_vsx"James Zern
2017-08-22Merge "vpx_dsp:loongson optimize vpx_subtract_block_c (case 4x4,8x8,16x16) ↵Shiyou Yin
with mmi."
2017-08-21quantize: ignore skip_block in armJohann
Change-Id: Icfb70687476b2edb25d255793ba325b261d40584
2017-08-21quantize: ignore skip_block in x86Johann
Change-Id: I9a963e99f08761f0c8d6a305619270b2f1c4edf8
2017-08-21Remove skip_block from quantizeJohann
This condition is handled before this code is reached. The ssse3 version of the function has always crashed when attempting to handle the skip_block condition. Add assert() and comments regarding the usage of skip_block. Removing the parameter is a fairly involved process so leave it be for the moment. Change-Id: Ib299f6fc6589d7ee102262cc74a7aeb60110bc5a
2017-08-18vpx_dsp: get32x32var_avx2() cleanupScott LaVarnway
renamed to get32x16var_avx2() BUG=webm:1404 Change-Id: Icb8f3986c9c9c646e13a69430db7235fc7e1a036
2017-08-18Merge "vpx_dsp: vpx_get16x16var_avx2() cleanup"Scott LaVarnway
2017-08-18vpx_dsp: vpx_get16x16var_avx2() cleanupScott LaVarnway
BUG=webm:1404 Change-Id: I88aceb07f4db4870a06eee21d87296974ce3221a
2017-08-18Merge "quantize: normalize intermediate types"Johann Koenig
2017-08-18vpx_dsp:loongson optimize vpx_subtract_block_c (case 4x4,8x8,16x16) with mmi.Shiyou Yin
Change-Id: Ia120ad1064d0b6106d9685cf075bdab373eef19e
2017-08-17highbd_idct32x32*,idct32_34_4x32_quarter_1_2: fix typoJames Zern
135 -> 34 fixes unused function warnings for highbd_idct32_34_4x32_quarter_[12] Change-Id: I4f50ff6ea514200af93dd59ff94c7f9717409682
2017-08-17quantize: normalize intermediate typesJohann
Despite abs_coeff being a positive value, all the other implementations treat it as signed which simplifies restoring the sign. HBD builds cast qcoeff to avoid a visual studio warning. Match vp9_quantize.c style of casting the entire expression. Change-Id: I62b539b8df05364df3d7644311e325288da7c5b5
2017-08-16inv_txfm_sse2.h: correct idct*/iadst* prototypesJames Zern
fixes mismatch between prototypes and definitions Change-Id: Ib5e7dfcce244dbb8401815be2cdd183d96792652
2017-08-16Merge changes I08b562b6,Ia275940a,I51106e90Linfeng Zhang
* changes: Add vpx_highbd_idct32x32_{34, 135, 1024}_add_{sse2, sse4_1} Update highbd idct x86 optimizations. Update 32x32 idct sse2 and ssse3 optimizations.
2017-08-15Merge "Clean up writing YUV files for debug purpose."Jerome Jiang
2017-08-15Clean up writing YUV files for debug purpose.Jerome Jiang
Change legacy vp8/9_write_yuv_frame to vpx_write_yuv_files. Delete some flags that can be enabled during build. To enable writing denoised YUV, use the following command line: CFLAGS='-DOUTPUT_YUV_DENOISED' ./configure --enable-vp9-temporal-denoising For skinmap, use CFLAGS='-DOUTPUT_YUV_SKINMAP' Change-Id: I236974ac8b3cf279d20c4dc7f6162d8b480b6528
2017-08-15quantize: silence unsigned overflow warningJohann
The result of the xor operation is unsigned. If coeff was negative, this results in an unsigned value - INT_MIN. Change-Id: I1f1edeaa6de1f4c68b848e8a82a666d390b749f0
2017-08-14Add vpx_highbd_idct32x32_{34, 135, 1024}_add_{sse2, sse4_1}Linfeng Zhang
BUG=webm:1412 Change-Id: I08b562b60fa85fbc2fec1c15c323a3444b44618f
2017-08-14Update highbd idct x86 optimizations.Linfeng Zhang
BUG=webm:1412 Change-Id: Ia275940af7d7d8637e9a851a9e39d655bfbe4069
2017-08-14Update 32x32 idct sse2 and ssse3 optimizations.Linfeng Zhang
Change-Id: I51106e90344035452621c49a6e1be7d5276b6c70
2017-08-10Merge "Clean highbd idct x86 code with inline functions"Linfeng Zhang
2017-08-10Merge "neon: vpx_quantize_b_32x32"Johann Koenig
2017-08-10Merge "quantize: copy ssse3 optimizations to intrinsics"Johann Koenig
2017-08-08Clean highbd idct x86 code with inline functionsLinfeng Zhang
Created inline functions highbd_butterfly_cospi16_sse2() and highbd_butterfly_cospi16_sse4_1() BUG=webm:1412 Change-Id: Icbc53a73712b6207379872a5e88d0a4d09e2322a
2017-08-08neon: vpx_quantize_b_32x32Johann
With skip block the neon is about twice as fast as C. The neon has no shortcut for coeff < zbin so it always takes the same amount of time. Even if the C can take the shortcut, it is over twice as fast in neon. If it can't, that gap increases to over 10x. BUG=webm:1426 Change-Id: I400722146c1b5a5f6289f67d85fd642463d2bfc6
2017-08-08quantize: copy ssse3 optimizations to intrinsicsJohann
Fairly minor differences from sse2. pabsw and psignw are the big gains. Also re-uses some values in eob calculation to avoid an extra pcmp. Fixes test failures in HBD and OS X builds. Allows using it in 32bit builds, where it is about 40% faster than sse2. Substantially faster than the assembly for skip_block. 10-20% faster the rest of the time. Change-Id: If783bb3567e561e47667e10133b9c84414a334e2
2017-08-07Update 32x32 idct sse2 funcs, add partial case 135Linfeng Zhang
Change-Id: I2b9add83f6fd8f9138fed3bec04a59877a237a6a
2017-08-04Rename highbd_multiplication_and_add_xx() to highbd_butterfly_xx()Linfeng Zhang
in idct x86 code Change-Id: I5159499a73a5c1b680516f6ca9c3d84f00c35083
2017-08-04Replace multiplication_and_add() with butterfly() in idct x86 codeLinfeng Zhang
Change-Id: I266e45a3d75a5357c7d6e6f20ab5c6fdbfe4982e
2017-08-04Update butterfly() in idct x86 optimizations.Linfeng Zhang
Change-Id: Ic73e03bab9fdc085146f52094014db4af36ad701
2017-08-04Add vpx_highbd_idct16x16_{10, 38, 256}_add_sse4_1Linfeng Zhang
BUG=webm:1412 Change-Id: I8877c986b4042f7b8e33f5674c86700675a0e4ca
2017-08-04Update for loop increment of idct x86 functionsLinfeng Zhang
Change-Id: Ided7895eaf41d5bc9d64fe536a17f5a078da68d4
2017-08-04Update high bitdepth 16x16 idct x86 codeLinfeng Zhang
Prepare for high bitdepth 16x16 idct sse4.1 code. Just functions moving and renaming. BUG=webm:1412 Change-Id: Ie056fe4494b1f299491968beadcef990e2ab714a
2017-08-04vpx_dsp: merge avx2 variance filesScott LaVarnway
BUG=webm:1404 Change-Id: Ieb8f85c3811b05df78722cb41eeb1166966ceec4
2017-08-04Merge "Rewrite vpx_idct16x16_{10,256}_add_sse2() and add case 38 function"Linfeng Zhang
2017-08-03Merge "vpx_dsp: Use correct check for halfpel in"Scott LaVarnway
2017-08-03Rewrite vpx_idct16x16_{10,256}_add_sse2() and add case 38 functionLinfeng Zhang
BUG=webm:1412 Change-Id: I945f0fb6807b8948747243794dc7352b959221f7
2017-08-03Merge changes I76727df0,I66297d78,I1d000c6bLinfeng Zhang
* changes: Extract inlined 16x16 idct sse2 code into header file Add transpose_32bit_8x4() sse2 optimization Update x86 idct optimization
2017-08-03vpx_dsp: Use correct check for halfpel inScott LaVarnway
vpx_sub_pixel_variance32xh_avx2() and vpx_sub_pixel_avg_variance32xh_avx2 see: 17fae3a Change to use correct check for halfpel Change-Id: Ib0741c5c2fd011e9650ca62b76009f1b59fdbe4c
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-11 07:27:33 +0000