mirror of
https://chromium.googlesource.com/libyuv/libyuv
synced 2026-06-15 00:16:08 +08:00
Removes non-matrix implementations for RGB24, RAW, RGB565, ARGB1555, and ARGB4444 conversions. Introduces RGBToYMatrixRow, RGBToUVMatrixRow, and equivalent functions for 16-bit and 24-bit formats. These functions utilize a 2-step conversion internally (to ARGB, then to YUV) inside row_common.cc for C, AVX2, and NEON, allowing the high-level convert.cc logic to execute in a single pass using ArgbConstants. Benchmark on Zen4 Test: libyuv_unittest --gtest_filter=*RGB*ToI420* Was BT.601-only ARGBToI420_Opt (115 ms) ARGB4444ToI420_Opt (190 ms) RGB565ToI420_Opt (194 ms) ARGB1555ToI420_Opt (207 ms) RGB24ToI420_Opt (143 ms) RGBAToI420_Opt (167 ms) 28.07% ARGBToUVMatrixRow_AVX512BW 19.65% ARGBToYMatrixRow_AVX512BW 11.32% RGBAToUVRow_SSSE3 10.24% ARGB1555ToARGBRow_SSE2 8.56% ARGB4444ToARGBRow_SSE2 8.47% RGB565ToARGBRow_SSE2 4.17% RGBAToYRow_AVX512BW 4.04% RGB24ToARGBRow_AVX512BW Now Matrix ARGBToI420_Opt (124 ms) ARGB4444ToI420_Opt (287 ms) RGB565ToI420_Opt (292 ms) ARGB1555ToI420_Opt (324 ms) RGB24ToI420_Opt (236 ms) RGBAToI420_Opt (126 ms) 29.74% ARGBToUVMatrixRow_AVX2 14.58% ARGB1555ToARGBRow_SSE2 12.59% RGB565ToARGBRow_SSE2 11.32% ARGB4444ToARGBRow_SSE2 9.35% ARGBToYMatrixRow_AVX2 8.45% RGB24ToARGBRow_SSSE3 5.56% ARGBToYMatrixRow_AVX512BW 1.37% ARGBToUVMatrixRow_Any_AVX2 0.74% ARGBToYMatrixRow_Any_AVX2 0.49% ARGB4444ToARGBRow_Any_SSE2 0.46% RGB565ToARGBRow_Any_SSE2 0.39% ARGB1555ToARGBRow_Any_SSE2 0.28% RGB24ToARGBRow_Any_SSSE3 0.11% ARGB4444ToYMatrixRow_AVX2 0.09% RGB565ToUVMatrixRow_AVX2 0.09% RGB565ToYMatrixRow_AVX2 0.07% RGBToYMatrixRow_AVX2 0.05% ARGB1555ToUVMatrixRow_AVX2 0.04% ARGB1555ToYMatrixRow_AVX2 0.03% RGBToUVMatrixRow_AVX2 0.02% ARGB4444ToUVMatrixRow_AVX2 Bug: libyuv:508639302 Change-Id: I362c0cfe4c86ee1f3ffb569fa4f784b84148f11a Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/7891045 Commit-Queue: Frank Barchard <fbarchard@google.com> Reviewed-by: richard winterton <rrwinterton@gmail.com> |
||
|---|---|---|
| build_overrides | ||
| docs | ||
| include | ||
| infra/config | ||
| riscv_script | ||
| source | ||
| tools_libyuv | ||
| unit_test | ||
| util | ||
| .clang-format | ||
| .gitignore | ||
| .gn | ||
| .vpython3 | ||
| Android.bp | ||
| Android.mk | ||
| AUTHORS | ||
| BUILD.bazel | ||
| BUILD.gn | ||
| CM_linux_packages.cmake | ||
| CMakeLists.txt | ||
| codereview.settings | ||
| DEPS | ||
| DIR_METADATA | ||
| download_vs_toolchain.py | ||
| GEMINI.md | ||
| libyuv.bzl | ||
| libyuv.gni | ||
| libyuv.gyp | ||
| libyuv.gypi | ||
| LICENSE | ||
| linux.mk | ||
| OWNERS | ||
| PATENTS | ||
| PRESUBMIT.py | ||
| public.mk | ||
| pylintrc | ||
| README.chromium | ||
| README.md | ||
| winarm.mk | ||
| WORKSPACE.bazel | ||
libyuv is an open source project that includes YUV scaling and conversion functionality.
- Scale YUV to prepare content for compression, with point, bilinear or box filter.
- Convert to YUV from webcam formats for compression.
- Convert to RGB formats for rendering/effects.
- Rotate by 90/180/270 degrees to adjust for mobile devices in portrait mode.
- Optimized for SSSE3/AVX2 on x86/x64.
- Optimized for Neon/SVE2/SME on Arm.
- Optimized for MSA on Mips.
- Optimized for RVV on RISC-V.
Development
See Getting started for instructions on how to get started developing.
You can also browse the docs directory for more documentation.