summaryrefslogtreecommitdiff
path: root/vpx_dsp
AgeCommit message (Collapse)Author
2018-01-23Add vp9_highbd_iht16x16_256_add_sse4_1()Linfeng Zhang
BUG=webm:1413 Change-Id: I8d7eeae1bd219eb848c1a86071046a477f7a91af
2018-01-23Add "vpx_" prefix to 2 idct x86 functionsLinfeng Zhang
Change-Id: I4f3052d8748e16b06e9155f8daf22f867dfaa7a3
2018-01-23Merge "Add vp9_highbd_iht8x8_64_add_sse4_1()"Linfeng Zhang
2018-01-18Add vp9_highbd_iht8x8_64_add_sse4_1()Linfeng Zhang
BUG=webm:1413 Change-Id: Id9038226902b2d793fc6c17ac81bb104c1a18988
2018-01-18clang-format v5.0.0 vpx_dsp/Johann
Remove comments above #define statements because they get indented unnecessarily. https://bugs.llvm.org/show_bug.cgi?id=35930 Add blank lines to prevent comments from being treated as blocks. Change-Id: I04dce21b2a10e13b8dc07411a0019c098f6dd705
2018-01-11adopt some clang 5.0.0 formattingJohann
At least the changes that don't conflict with 4.0.1 Change-Id: I9b6a7c14dadc0738cd0f628a10ece90fc7ee89fd
2018-01-08Add vp9_highbd_iht4x4_16_add_sse4_1()Linfeng Zhang
BUG=webm:1413 Change-Id: I14930d0af24370a44ab359de5bba5512eef4e29f
2018-01-08Update dct_test.ccLinfeng Zhang
Make 8-bit functions testing available in high bitdepth. Change-Id: Ic030c75aa4c6b649c52426abb4bb2122882de0fe
2017-12-28Update iadst4_sse2()Linfeng Zhang
Change-Id: I21ff81df0d6898170a3b80b3b5220f9f3ac7f4e8
2017-12-14add copyright to rtcd filesJohann
Allows them to pass the license check in chromium. BUG=chromium:98319 Change-Id: Iefc1706152a549d8c4ae774c917596bf1c9492d8
2017-12-04Merge "vpx_dsp: [loongson] optimize variance v2."Shiyou Yin
2017-12-01explicitly label .text sectionsJohann
nasm should infer .text but does not for windows: https://bugzilla.nasm.us/show_bug.cgi?id=3392451 Change-Id: Ib195465e5f33405f5ff61c4cf88aa2a72640cacb
2017-12-01vpx_dsp: [loongson] optimize variance v2.Shiyou Yin
1. Delete unnecessary zero setting process. 2. Optimize the method of calculating SSE in vpx_varianceWxH. Change-Id: I58890c6a2ed1543379acb48e03e620c144f6515f
2017-12-01Merge "mips msa optimize vpx_scaled_2d function"Kaustubh Raste
2017-11-30Merge "vpx: [loongson] fix bug in var_filter_block2d_bil_16x"Shiyou Yin
2017-11-29Merge "Remove unnecessary includes of emmintrin_compat.h"Kyle Siefring
2017-11-29Remove unnecessary includes of emmintrin_compat.hKyle Siefring
Change-Id: Ie60381a0c6ee01f828cd364a43f01517f4cb03e9
2017-11-29mips msa optimize vpx_scaled_2d functionKaustubh Raste
Change-Id: I638507b360c71489ab0e87bd558d2719ad995333
2017-11-29vpx: [loongson] fix bug in var_filter_block2d_bil_16xShiyou Yin
Which cause failed case: 1. MMI/VpxSubpelVarianceTest.Ref/6 2. MMI/VpxSubpelVarianceTest.Ref/7 3. MMI/VpxSubpelVarianceTest.ExtremeRef/6 4. MMI/VpxSubpelVarianceTest.ExtremeRef/7 Change-Id: I122ca20089e14ac324edd61295cf8f506e06afc8
2017-11-27quantize x86: dedup some partsJohann
Change-Id: I9f95f47bc7ecbb7980f21cbc3a91f699624141af
2017-11-20Merge "Optimize AVX2 get16x16var and get32x16var functions"Kyle Siefring
2017-11-17Optimize AVX2 get16x16var and get32x16var functionsKyle Siefring
Change-Id: If8b91aaa883c01107f0ea3468139fa24cfb301d2
2017-11-15fwd txfm ssse3: use GLOBAL() for loading constantsJohann
Fixes a build issue when relocation is not allowed: relocation R_X86_64_32 against '.rodata' can not be used when making a shared object Change-Id: Ica3e90c926847bc384e818d7854f0030f4d69aa0
2017-11-10vpx: [x86] add vpx_satd_avx2()Scott LaVarnway
SSE2 instrinsic vs AVX2 intrinsic speed gains: blocksize 16: ~1.33 blocksize 64: ~1.51 blocksize 256: ~3.03 blocksize 1024: ~3.71 Change-Id: I79b28cba82d21f9dd765e79881aa16d24fd0cb58
2017-11-09runtime error fix: bitdepth_conversion_avx2.hScott LaVarnway
Change-Id: I7364a157de39eb7137b599808474b8d46d19d376
2017-11-03Support building AVX-512 and implement sadx4 for AVX-512Kyle Siefring
The added AVX-512 support requires the subset of AVX-512 added in Skylake-X. Change-Id: I39666b00d10bf96d06c709823663eb09b89265b7
2017-10-26vpx: hadamard: use ptrdiff_t instead of int for strideScott LaVarnway
Eliminates the following instruction for the x86 (64 bit) intrinsic code: movslq %esi,%rax Change-Id: I8f5ebd40726f998708a668b0f52ea7a0576befae
2017-10-24Merge "Optimize convolve8 SSSE3 and AVX2 intrinsics"Kyle Siefring
2017-10-24Optimize convolve8 SSSE3 and AVX2 intrinsicsKyle Siefring
Changed the intrinsics to perform summation similiar to the way the assembly does. The new code diverges from the assembly by preferring unsaturated additions. Results for haswell SSSE3 Horiz/Vert Size Speedup Horiz x4 ~32% Horiz x8 ~6% Vert x8 ~4% AVX2 Horiz/Vert Size Speedup Horiz x16 ~16% Vert x16 ~14% BUG=webm:1471 Change-Id: I7ad98ea688c904b1ba324adf8eb977873c8b8668
2017-10-23vpx: [x86] vpx_hadamard_16x16_avx2() highbitdepth fixScott LaVarnway
Use an intermediate buffer before storing to coeffs when highbitdepth is enabled. Change-Id: I101981a1995f1108ad107c55c37d6e09eadb404b
2017-10-20vpx: [x86] vpx_hadamard_16x16_avx2() improvementsScott LaVarnway
~10% performance gain. Fixed the cosmetics noted in the previous commit. Change-Id: Iddf475f34d0d0a3e356b2143682aeabac459ed13
2017-10-19Merge "vpx: [x86] add vpx_hadamard_16x16_avx2()"Scott LaVarnway
2017-10-18vpx: [x86] add vpx_hadamard_16x16_avx2()Scott LaVarnway
This version is ~1.91x faster than the sse2 version. When highbitdepth is enabled, it is ~1.74x. Change-Id: I2b0e92ede9f55c6259ca07bf1f8c8a5d0d0955bd
2017-10-18Merge "Refactor x86/vpx_subpixel_8t_intrin_avx2.c"Kyle Siefring
2017-10-17Merge changes I17fff122,Ic149e3cbLinfeng Zhang
* changes: Add 4 to 3 scaling SSSE3 optimization Test extreme inputs in frame scale functions
2017-10-17Refactor x86/vpx_subpixel_8t_intrin_avx2.cKyle Siefring
Change-Id: I6539111dfb35a43028e9755785b2e9ea31854305
2017-10-16Add 4 to 3 scaling SSSE3 optimizationLinfeng Zhang
Note this change will trigger the different C version on SSSE3 and generate different scaled output. Its speed is 2x compared with the version calling vpx_scaled_2d_ssse3(). Change-Id: I17fff122cd0a5ac8aa451d84daa606582da8e194
2017-10-12Merge changes I38783d97,If5160c0cKyle Siefring
* changes: Extend 16 wide AVX2 convolve8 code to support averaging. Add AVX2 version of vpx_convolve8_avg.
2017-10-10Add 4 to 1 scaling x86 optimizationLinfeng Zhang
Change-Id: I51c190f0a88685867df36912522e67bdae58a673
2017-10-10Merge changes I9d4c1af5,I882da3a0Linfeng Zhang
* changes: Rename some inline functions in NEON scaling Generalize 2:1 vp9_scale_and_extend_frame_ssse3()
2017-10-09Extend 16 wide AVX2 convolve8 code to support averaging.Kyle Siefring
Also adds vpx_convolve8_avg_horiz_avx2. Change-Id: I38783d972ac26bec77610e9e15a0a058ed498cbf
2017-10-07Add AVX2 version of vpx_convolve8_avg.Kyle Siefring
vpx_convolve8_avg works by first running a normal horizontal filter then a vertical filter averages at the end. The added vpx_convolve8_avg_avx2 calls pre-existing AVX2 code for the horizontal step. vpx_convolve8_avg_vert_avx2 is also added, but only uses ssse3 code. Change-Id: If5160c0c8e778e10de61ee9bf42ee4be5975c983
2017-10-07Merge "ppc: Add vpx_idct32x32_1024_add_vsx"James Zern
2017-10-04Generalize 2:1 vp9_scale_and_extend_frame_ssse3()Linfeng Zhang
Change-Id: I882da3a04884d5fabd4cd591c28682cbb2d76aa5
2017-10-04Merge changes Id6a8c549,Ib1e0650b,Ic369dd86Linfeng Zhang
* changes: Refactor x86/vpx_subpixel_8t_intrin_ssse3.c Add vpx_dsp/x86/mem_sse2.h Add transpose_8bit_{4x4,8x8}() x86 optimization
2017-10-03Merge "vpx: fix nasm build errors"James Zern
2017-10-03vpx: fix nasm build errorsScott LaVarnway
BUG=webm:1462,766721 Change-Id: Icfa536a8e38623636b96c396e3c94889bfde7a98
2017-10-03Refactor x86/vpx_subpixel_8t_intrin_ssse3.cLinfeng Zhang
Change-Id: Id6a8c549709a3c516ed5d7b719b05117c5ef8bac
2017-10-03Add vpx_dsp/x86/mem_sse2.hLinfeng Zhang
Add some load and store sse2 inline functions. Change-Id: Ib1e0650b5a3d8e2b3736ab7c7642d6e384354222
2017-10-03Add transpose_8bit_{4x4,8x8}() x86 optimizationLinfeng Zhang
Change-Id: Ic369dd86b3b81686f68fbc13ad34ab8ea8846878
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-09 19:30:14 +0000