summaryrefslogtreecommitdiff
path: root/vp9/vp9cx.mk
AgeCommit message (Collapse)Author
2015-12-14move vp9_avg to vpx_dspJames Zern
Change-Id: I7bc991abea383db1f86c1bb0f2e849837b54d90f
2015-11-11Add AVX vectorized vp9_diamond_search_sadGeza Lore
This function now has an AVX intrinsics version which is about 80% faster compared to the C implementation. This provides a 2-4% total speed-up for encode, depending on encoding parameters. The function utilizes 3 properties of the cost function lookup table, constructed in 'cal_nmvjointsadcost' and 'cal_nmvsadcosts'. For the joint cost: - mvjointsadcost[1] == mvjointsadcost[2] == mvjointsadcost[3] For the component costs: - For all i: mvsadcost[0][i] == mvsadcost[1][i] (equal per component cost) - For all i: mvsadcost[0][i] == mvsadcost[0][-i] (Cost function is even) These must hold, otherwise the AVX version of the function cannot be used. Change-Id: I6c2791d43022822a9e6ab43cd124a773946d0bdc
2015-11-06Revert "Add AVX vectorized vp9_diamond_search_sad"James Zern
This reverts commit f1342a7b070ef61b9fbdf03e899ac2107cfcb6bd. This breaks 32-bit builds: runtime error: load of misaligned address 0xf72fdd48 for type 'const __m128i' (vector of 2 'long long' values), which requires 16 byte alignment + _mm_set1_epi64x is incompatible with some versions of visual studio Change-Id: I6f6fc3c11403344cef78d1c432cdc9147e5c1673
2015-11-05Merge "Add AVX vectorized vp9_diamond_search_sad"Yunqing Wang
2015-11-05Add AVX vectorized vp9_diamond_search_sadGeza Lore
This function now has an AVX intrinsics version which is about 80% faster compared to the C implementation. This provides a 2-4% total speed-up for encode, depending on encoding parameters. The function utilizes 3 properties of the cost function lookup table, constructed in 'cal_nmvjointsadcost' and 'cal_nmvsadcosts'. For the joint cost: - mvjointsadcost[1] == mvjointsadcost[2] == mvjointsadcost[3] For the component costs: - For all i: mvsadcost[0][i] == mvsadcost[1][i] (equal per component cost) - For all i: mvsadcost[0][i] == mvsadcost[0][-i] (Cost function is even) These must hold, otherwise the AVX version of the function cannot be used. Change-Id: I184055b864c5a2dc37b2d8c5c9012eb801e9daf6
2015-11-02Move noise level estimate outside denoiser.Marco
Source noise level estimate is also useful for setting variance encoder parameters (variance thresholds, qp-delta, mode selection, etc), so allow it to be used also if denoising is not on. Change-Id: I4fe23d47607b4e17a35287057f489c29114beed1
2015-10-21Optimize vp9_highbd_block_error_8bit assembly.Geza Lore
A new version of vp9_highbd_error_8bit is now available which is optimized with AVX assembly. AVX itself does not buy us too much, but the non-destructive 3 operand format encoding of the 128bit SSEn integer instructions helps to eliminate move instructions. The Sandy Bridge micro-architecture cannot eliminate move instructions in the processor front end, so AVX will help on these machines. Further 2 optimizations are applied: 1. The common case of computing block error on 4x4 blocks is optimized as a special case. 2. All arithmetic is speculatively done on 32 bits only. At the end of the loop, the code detects if overflow might have happened and if so, the whole computation is re-executed using higher precision arithmetic. This case however is extremely rare in real use, so we can achieve a large net gain here. The optimizations rely on the fact that the coefficients are in the range [-(2^15-1), 2^15-1], and that the quantized coefficients always have the same sign as the input coefficients (in the worst case they are 0). These are the same assumptions that the old SSE2 assembly code for the non high bitdepth configuration relied on. The unit tests have been updated to take this constraint into consideration when generating test input data. Change-Id: I57d9888a74715e7145a5d9987d67891ef68f39b7
2015-10-08Optimization of 8bit block error for high bitdepthGeza Lore
If high bit depth configuration is enabled, but encoding in profile 0, the code now falls back on optimized SSE2 assembler to compute the block errors, similar to when high bit depth is not enabled. Change-Id: I471d1494e541de61a4008f852dbc0d548856484f
2015-08-06Move VP9 SSIM metrics to vpx_dsp.Alex Converse
Change-Id: I20c7b42631b579fade6cf7ebf6d4c69b2fcb5e5e
2015-07-29Merge "add vp9_block_error_fp_neon"James Zern
2015-07-28Remove vp9_dct.h fileJingning Han
The forward 32x32 2D-DCT functions are aligned in vpx_dsp folder. The vp9_dct.h file is not effectively used now. Change-Id: Ie7946b6fdd784b8e91496242337bc9002c75c281
2015-07-28Move DC only forward 2D-DCT functions to vpx_dspJingning Han
This completes the forward transform functions layout refactoring. Change-Id: I996fb0fb795f41e2040f7b21db985774098aedbd
2015-07-28Factor 32x32 fwd DCT to vpx_dsp folderJingning Han
Move the 32x32 2D-DCT implementations from vp9/ to vpx_dsp/. Change-Id: Id3980696f8b69906ff7a59ff9fb2b9013d60047d
2015-07-27add vp9_block_error_fp_neonJames Zern
~60-70% faster depending on the block size Change-Id: Icdbaa9977a91a63cbcc6ead0cf19d5a2af7f27e1
2015-07-27Move forward dct sse2 header file to vpx_dspJingning Han
Change-Id: Iba03852ce778c956200818e3473cfb2b48cf8d8e
2015-07-22Factor forward 2D-DCT transforms into vpx_dspJingning Han
This commit factors the 4x4, 8x8, and 16x16 2D-DCT forward transform operations into vpx_dsp folder. Change-Id: I084b117b79c0925edcbcabb93f62b9f4bf8dbe7d
2015-07-20Move bit writer files to vpx_dsp/Yaowu Xu
Change-Id: Id27e0007a0feac821ca66bcecbf3a723305da82d
2015-07-17Migrate quantization functions from vp9/ to vpx_dsp/Yunqing Wang
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
2015-07-07Move sub pixel variance to vpx_dspJohann
Change-Id: I66bf6720c396c89aa2d1fd26d5d52bf5d5e3dff1
2015-07-06Merge "mips msa vp9 subpel variance optimization"James Zern
2015-07-06mips msa vp9 subpel variance optimizationParag Salasakar
Change-Id: If88401bf8c5d8ee58200278734d7a5058d1585d0
2015-07-06Move subtract functions from vp9 to vpx_dspJingning Han
Factor out the subtraction operator as common function. Change-Id: I526e703477c6a290e0e3e3c8898f8bb1ca82779b
2015-07-02Revert "mips msa vp9 subpel variance optimization"James Zern
This reverts commit a42df86c035d1bc007dbc816b4b758605135d6d9. this change causes MSA/VP9SubpelVarianceTest.Ref and MSA/VP9SubpelVarianceTest.ExtremeRef failures under mips32r5el-msa-linux-gnu and mips64r6el-msa-linux-gnu Change-Id: I40b71a0b774eaeb31f66f795733f95cf360909f7
2015-07-01Fix --disable-use-x86incJohann
Change-Id: I374fcd8fb45a6893dcdeac6896671be142a99f06
2015-07-01mips msa vp9 subpel variance optimizationParag Salasakar
average improvement ~3x-5x Change-Id: I4cbba2711467b0e205904769ebbb4a1fcbb1a311
2015-06-26mips msa vp9 temporal filter optimizationParag Salasakar
average improvement ~4x-5x Change-Id: Iad9c0a296dbc2ea96d000bd009077999ed58a3c5
2015-06-26mips msa vp9 subtract block optimizationParag Salasakar
average improvement ~3x-4x Change-Id: Idbe4d13a00d05ff8be6559b116f416e42c3b4097
2015-06-26mips msa vp9 block error optimizationParag Salasakar
average improvement ~3x-4x Change-Id: If0fdcc34b17437a7e3e7fb4caaf1067bc175f291
2015-06-23mips msa vp9 avg optimizationParag Salasakar
average improvement ~2x-3x Change-Id: I76f7fc00c0ffdf2b4ba41bf3819f3b6044bcdeff
2015-06-22mips msa vp9 fdct 4x4 optimizationParag Salasakar
average improvement ~2x-3x Change-Id: Idf8be780b8b4228fc91f110a94e4ee1fd9af0163
2015-06-20mips msa vp9 fdct 8x8 optimizationParag Salasakar
average improvement ~4x-5x Change-Id: I37582efc2622bc20b2bf99617a76110ab24e9f6a
2015-06-17mips msa vp9 fdct 32x32 optimizationParag Salasakar
average improvement ~4x-6x Change-Id: Ibcac3ef8ed5e207cf8c121e696570e6b63d3c0f4
2015-06-16mips msa vp9 fdct 16x16 optimizationParag Salasakar
average improvement ~4x-6x Change-Id: Id3b2243e5b3c7844c90c4231a5e75fa69911362c
2015-05-26Move variance functions to vpx_dspJohann
subpel functions will be moved in another patch. Change-Id: Idb2e049bad0b9b32ac42cc7731cd6903de2826ce
2015-05-15rename vp9_dct_impl_sse2.c to vp9_dct_sse2_impl.hJames Zern
this file shouldn't be built directly, it is included in vp9_dct_sse2.c to create a non-high-bitdepth and a high-bitdepth version silences missing prototype warnings for the unused FDCT* functions Change-Id: Ide6ff8c24ab31bdb0f833260505ae33660a1ad5b
2015-05-15rename vp9_dct32x32_sse2.c to vp9_dct32x32_sse2_impl.hJames Zern
this file shouldn't be built directly, it is included in vp9_dct_sse2.c to create a non-high-bitdepth and a high-bitdepth version silences missing prototype warnings for the unused FDCT32x32* functions Change-Id: I0e38f16dae5ea1728de184ee2c89287d48675c51
2015-05-15rename vp9_dct32x32_avx2.c to vp9_dct32x32_avx2_impl.hJames Zern
this file shouldn't be built directly, it is included in vp9_dct_avx2.c to create a non-high-bitdepth and a high-bitdepth version silences missing prototype warnings for the unused FDCT32x32* functions Change-Id: I4c19935c0e035b393be513bde735e9a78064a494
2015-05-06Move shared SAD code to vpx_dspJohann
Create a new component, vpx_dsp, for code that can be shared between codecs. Move the SAD code into the component. This reduces the size of vpxenc/dec by 36k on x86_64 builds. Change-Id: I73f837ddaecac6b350bf757af0cfe19c4ab9327a
2015-04-17Merge "Adds a blockiness metric to internal stats."Jim Bankoski
2015-04-17Merge "adds psnrhvs to internal stats."Jim Bankoski
2015-04-17Merge "Adds a fastssim metric to VPX internal stats."Jim Bankoski
2015-04-17Adds a blockiness metric to internal stats.Jim Bankoski
Change-Id: Iedceeb020492050063acf3fd2326f96c29db9ae5
2015-04-17adds psnrhvs to internal stats.Jim Bankoski
PSNR HVS is a human visual system weighted version of SNR that's gained some popularity from academia and apparently better matches MOS testing. This code is borrowed from the Daala Project but uses our FDCT code. Change-Id: Idd10fbc93129f7f4734946f6009f87d0f44cd2d7
2015-04-17Adds a fastssim metric to VPX internal stats.Jim Bankoski
This code appeared in the Daala project first and was originally committed by Nathan Egge. Change-Id: Iadce416a091929c51b46637ebdec984cddcaf18c
2015-03-31vp9/neon: skip some files in high-bitdepth buildJames Zern
exclude files that only contain functions for non-high-bitdepth builds. this removes some warnings related to missing prototypes Change-Id: Ic6642998c46a7b808c6c53b2f9c34bcd4d037abe
2015-02-11Add skin detection.Marco
Simple skin detection, from vp8; works reasonable on most of the RTC clips, but could miss sometimes. Added debug flag to write out skin map over source input. Change-Id: I2caea7592f1c459047aac46627eeb24a94946464
2015-01-27Add vp9_sad64x64x4d_neon Neon intrinsic function.Frank Galligan
On Nexus 7 speed -6 saw ~30% increase in perf. Tested on Nexus 7, built with ndk r10d, gcc 4.9. BUG=https://code.google.com/p/webm/issues/detail?id=908 Change-Id: Id12af7d1883243c23e6692e898aea82299633d58
2015-01-15Add Neon intrinsics for vp9_avg_8x8_neonFrank Galligan
On Nexus 7 speed -5, -6, -7, and -8 saw about a 1% increase in perf for 480p. Speeds -5, -6, -7, and -8 saw about a 1.5% increase in perf for 720p. Tested on Nexus 7, built with ndk r10d, gcc 4.9. Change-Id: Ibf17ebfd952a6aec941719bd8306df8ec4574bee
2014-12-05Merge "vp9_ethread: the tile-based multi-threaded encoder"Yunqing Wang
2014-12-04vp9_ethread: the tile-based multi-threaded encoderYunqing Wang
Currently, VP9 supports column-tile encoding, which allows a frame to be encoded in multiple column tiles independently. The number of column tiles are set by encoder option "--tile-columns". This provides a way to encode a frame in parallel. Based on previous set of patches, this patch implemented the tile- based multi-threaded encoder. Each thread processes one or more tiles. Usage: For HD clips: --tile-columns=2 --threads=1/2/3/4 While using 4 threads, tests showed that the encoder achieved 2.3X - 2.5X speedup at good-quality speed 3, and 2X speedup at realtime speed 5. Change-Id: Ied987f8f2618b1283a8643ad255e88341733c9d4
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-15 15:32:34 +0000