summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2014-09-07Merge "Replacing vp9_get_mb_ss_sse2 asm implementation with intrinsics."Dmitry Kovalev
2014-09-06libyuv: cherry-pick MSVC arm build fixJames Zern
Issue 24479004: Fix building with MSVC for arm https://webrtc-codereview.appspot.com/24479004/ Change-Id: I758b33a21f46e46af6e58d83b7c045262ac9c7d9
2014-09-06Replacing vp9_get_mb_ss_sse2 asm implementation with intrinsics.Dmitry Kovalev
Change-Id: Ib4f5dd733eb2939b108070a01e83da5d9990bac0
2014-09-05Merge "Removing postproc mmx code."Dmitry Kovalev
2014-09-05Merge "Correct the mode decisions in special cases"Yunqing Wang
2014-09-05Merge "Initializing intra modes without vpx_once()."Dmitry Kovalev
2014-09-05Merge changes I7b9f40dc,I76e74f2eJames Zern
* changes: vp9: correct context buffer resize check vp9: fail decode if block/frame refs are corrupt
2014-09-05Correct the mode decisions in special casesYunqing Wang
The rate costs calculated for inter modes are not precise in some cases, which causes NEWMV is chosen instead of NEARESTMV, NEARMV, and ZEROMV. This patch added checks for these cases, and corrected the mode decisions. Borg tests at speed 3 showed: 1. stdhd set: 0.102% PSNR gain and 0.088% SSIM gain. 2. derf set: 0.147% PSNR gain and 0.132% SSIM gain. No speed change. Change-Id: I35d17684b89ad4734fb610942d707899146426db
2014-09-05Merge "fix x86-darwin* build"James Zern
2014-09-05Merge "vp9: skip loopfilter when the frame is corrupt"James Zern
2014-09-05Removing postproc mmx code.Dmitry Kovalev
Removed functions: * vp9_post_proc_down_and_across_mmx * vp9_mbpost_proc_down_mmx * vp9_plane_add_noise_mmx They all have sse2 equivalent. Change-Id: I59c1fac12b7c96ca4538d455e4400c2b7875feff
2014-09-05Merge "Adding temp cpi var."Dmitry Kovalev
2014-09-05vp8 common: change 'HAVE_NEON_ASM' to 'HAVE_NEON' for compiling functions of ↵Jia Jia
NEON intrinsics. Change-Id: I975e5eac16f8b623ff589f0ec072cdaff2183b04
2014-09-04fix x86-darwin* buildJames Zern
vp9_variance_sse2.c contains a mix of intrinsics and references to assembly which uses x86inc.asm; it's conditionally included as a result. Change-Id: I254451483a65881c0b8e18e27bf0c3ddef60c4ec
2014-09-04bilinearpredict_neon: fix type conversion warningsJames Zern
make bifilter4_coeff[][] uint8_t, no values exceed this range and they're loaded with vdup_n_u8(). Change-Id: I921983e9edd828d29820e40ac30a7801dbe0fb4f
2014-09-04vp9: correct context buffer resize checkJames Zern
allocations within vp9_alloc_context_buffers() rely on mi_rows/mi_cols individually, use those to determine whether to realloc rather than stride and stride * rows. this fixes a crash with some fuzzed files for invalid accesses into last_frame_seg_map and above_context. Change-Id: I7b9f40dcf170d443890f3bd2acd285507943c7d4
2014-09-04vp9: fail decode if block/frame refs are corruptJames Zern
proceeding using a corrupt (incompletely decoded) frame reference may lead to incorrect assumptions about allocation sizes leading to a crash. Change-Id: I76e74f2e1be127c2e2c7e1174bb3307497dfd23d
2014-09-04Merge "Removing sz member from vpx_codec_priv. "Dmitry Kovalev
2014-09-04Merge "Removing unused function prototypes."Dmitry Kovalev
2014-09-04Merge "Map motion magnitude in VP9 denoiser."JackyChen
2014-09-04Merge "Speed up compound inter prediction mode check"Jingning Han
2014-09-04Merge "Update the condition when COPY_BLOCK is chosen."JackyChen
2014-09-04Merge "Removing local set_speed_features() function."Dmitry Kovalev
2014-09-04Map motion magnitude in VP9 denoiser.JackyChen
This is to keep the same with VP8 denoiser. If motion magnitude is small, make denoiser more aggressive. Change-Id: I942a6e2f2ed9aec6f0c4c1f9e5fa47066cadcc0c
2014-09-04Adding temp cpi var.Dmitry Kovalev
Change-Id: Ifa3c1cc2317c1bc21d1042b9662b35056d1e9ed0
2014-09-04Removing sz member from vpx_codec_priv. Dmitry Kovalev
Change-Id: I811526a9ee9f237604f72abe7fc677e39e0f457f
2014-09-04Update the condition when COPY_BLOCK is chosen.JackyChen
The change is just to keep the condition the same with VP8. Change-Id: I9662b40996126605945dd853c0cbe8916c1ce578
2014-09-04Merge "Actually resetting random generator for all variance test cases."Dmitry Kovalev
2014-09-04Removing unused function prototypes.Dmitry Kovalev
Change-Id: Ia5e383e2cf18052f6f1eacf8b9495ab8e4d58878
2014-09-04Merge "Fix a bug in VP9 denoiser."JackyChen
2014-09-04Merge "Adding sse2 variant for vp9_mse{8x8, 8x16, 16x8}."Dmitry Kovalev
2014-09-04Actually resetting random generator for all variance test cases.Dmitry Kovalev
Calling Reset(int) method instead of overloaded operator()(int). Adding underscore at the end of class member name. Change-Id: I01934e7bc056d4b594e5d05d693328febd34ac3c
2014-09-04Fix a bug in VP9 denoiser.JackyChen
When the first try of denoising turns out to be too much, we will use a softer filter by adopting an adjustment to make the result closer to original pixel (as in VP8 denoiser). The old code made the adjustment in the wrong direction. Change-Id: I84e28fa9e01eef47c5a37d5a2e6d3d378a06786b
2014-09-04Merge "arm: Fix building vp8_subpixelvariance_neon.c with MSVC"James Zern
2014-09-04Revert "Revert "VP8 for ARMv8 by using NEON intrinsics 10""Scott LaVarnway
This reverts commit 677fb5123e0ece1d6c30be9d0282e1e1f594a56d Compiles with 4.6. Change-Id: I7f87048911b6bc28a61741d95501fa45ee97b819
2014-09-04vp8 common: change 'HAVE_NEON_ASM' to 'HAVE_NEON' for compiling idct_blk_neon.c.Jia Jia
Change-Id: Ib89107fb824b5fe58afef6841104d5a27b2e0f2d
2014-09-04arm: Fix building vp8_subpixelvariance_neon.c with MSVCMartin Storsjo
Use the right return values - vget_low_s64 returns int64x1_t, not a normal int64_t. Also make __builtin_prefetch a no-op on MSVC for this file. Change-Id: I4d2fce01d0ba106b98d3d53b137803119c2c2c08
2014-09-03Merge "Consistent allocation of vpx_codec_alg_priv_t."Dmitry Kovalev
2014-09-03Adding sse2 variant for vp9_mse{8x8, 8x16, 16x8}.Dmitry Kovalev
Change-Id: I6786d25ce4f32b8d8912f2d239a45ca15b310c4b
2014-09-03Merge "Replacing asm 16x16 variance calculation with intrinsics."Dmitry Kovalev
2014-09-03Merge "Small cleanup: reusing existing code."Dmitry Kovalev
2014-09-03Merge "Add a more complex SEG_LVL_SKIP test vector."Yaowu Xu
2014-09-03Speed up compound inter prediction mode checkJingning Han
This commit allows the encoder to store outcomes of single reference frame modes and compares them to decide if the inter prediction filter, forward transform, and quantization can be skipped. The compression performance of speed 3 is down derf -0.364% stdhd -0.198% For test sequences, the speed 3 runtime is reduced highway CIF 100 kbps, 51976 ms -> 45033 ms, 13% speed-up stockholm 720p 1000 kbps, 71826 ms -> 67838 ms, 5.5% speed-up pedestrian 1080p 2000 kbps, 154924 ms -> 150702 ms, 2.6% speed-up Change-Id: I5aa26f918d2b4b5197a2c0afa2779319f1c88e44
2014-09-03Add a more complex SEG_LVL_SKIP test vector.Alex Converse
This file changes the segmentation map, deactivates the map, and reactivates the map. Change-Id: I1c8b8949887257b9f4c49a1a9ba9a7c74666ef2e
2014-09-03Merge "Change last_partition_redo_frequency for speed 3"Yaowu Xu
2014-09-03Merge "Remove redundant code"Yaowu Xu
2014-09-03Merge "Make process_denoise_mode_change function cond-compiled."Marco
2014-09-03Neon version of vp8_build_intra_predictors_mby_s() andScott LaVarnway
vp8_build_intra_predictors_mbuv_s(). This patch replaces the assembly version with an intrinsic version. On a Nexus 7, vpxenc (in realtime mode, speed -12) reported a performance improvement of ~2.6%. Change-Id: I9ef65bad929450c0215253fdae1c16c8b4a8f26f
2014-09-03VP8 for ARMv8 by using NEON intrinsics 17Scott LaVarnway
Add vp8_subpixelvariance_neon.c - vp8_sub_pixel_variance16x16_neon_func - vp8_variance_halfpixvar16x16_h_neon - vp8_variance_halfpixvar16x16_v_neon - vp8_variance_halfpixvar16x16_hv_neon - vp8_sub_pixel_variance8x8_neon Change-Id: I3e5d85b2eafc26be0eef6a777789b80e4579257b Signed-off-by: James Yu <james.yu@linaro.org>
2014-09-03Merge "Revert "Revert "VP8 for ARMv8 by using NEON intrinsics 06" This ↵Johann
reverts commit 81ad047ee57ecb0e2c1ee4dcebda54a44ea54ae9. Revert "VP8 for ARMv8 by using NEON intrinsics 15" This reverts commit 727af7cebe3698b8493ba6c1360b0a6606c310fb.""
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-15 17:13:34 +0000