mirror of
https://chromium.googlesource.com/libyuv/libyuv
synced 2026-10-28 18:07:12 +08:00
63b56324de
Replace stack-buffer memcpy _Any wrappers with native AVX-512 k-mask
tail handling across 8 AVX-512 row kernels (unlocking composite
RGBToUVMatrixRow_AVX512BW as well) and remove IS_ALIGNED dispatch
constraints:
- MirrorSplitUVRow_AVX512BW: Masked 16-bit load (vmovdqu16 %{k1%}%{z%}),
dynamic reverse index (vpsubw + vpermw via kMirrorUVPermute_AVX512),
and vpmovwb %{k1%} U/V stores; unlocks unaligned SplitRotateUV180.
- ARGBToUVMatrixRow_AVX512BW: Fix MASK=63 vs 32-pixel loop bug;
zero-load tail dwords per row, duplicate odd final pixel via knotw +
vpshufd, and store masked U/V bytes.
- I422ToARGBRow_AVX512BW & I422ToAR30Row_AVX512BW: Add masked Y/U/V
loads (READYUV422_MASK_AVX512BW) and masked ARGB/AR30 stores.
- CopyRow_AVX512BW: Add 64-byte full vector step + bzhi masked tail.
- BlendPlaneRow_AVX512BW, ARGBShuffleRow_AVX512BW, RGBToARGBRow_AVX512BW
(RAWToARGBRow_AVX512BW / RGB24ToARGBRow_AVX512BW): Masked loads and
stores in both row_gcc.cc inline assembly and row_win.cc intrinsics.
Reduces x86-64 binary .text size by 2,592 bytes with 0 bytes added to
.rodata.
Performance on Intel Sapphire Rapids (1275x1650, 1000 iterations):
Function (Test) | Before | After | Speedup
-------------------------|---------|---------|--------
NV12Rotate180_Opt | 429 ms | 260 ms | 1.65x
RGB24ToI420_Opt | 615 ms | 531 ms | 1.16x
I420ToAR30_Opt | 425 ms | 383 ms | 1.11x
ARGBToI420_Opt | 427 ms | 411 ms | 1.04x
I420ToARGB_Opt | 423 ms | 411 ms | 1.03x
RGB24ToARGB_Opt | 493 ms | 490 ms | 1.01x
ARGBToBGRA_Opt | 544 ms | 542 ms | 1.00x
CopyPlane_Opt | 246 ms | 245 ms | 1.00x
BlendPlane_Opt | 259 ms | 258 ms | 1.00x
RAWToARGB_Opt | 486 ms | 486 ms | 1.00x
R=rrwinterton@gmail.com
Test: LIBYUV_WIDTH=1275 LIBYUV_HEIGHT=1650 LIBYUV_REPEAT=1000 out/Release/libyuv_unittest
Change-Id: Ib5e5ae52076352649108b20f8df2b12d81860fb1
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8563977
Reviewed-by: richard winterton <rrwinterton@gmail.com>
|
||
|---|---|---|
| .. | ||
| compare_common.cc | ||
| compare_gcc.cc | ||
| compare_neon64.cc | ||
| compare_neon.cc | ||
| compare_wasm.cc | ||
| compare_win.cc | ||
| compare.cc | ||
| convert_argb.cc | ||
| convert_from_argb.cc | ||
| convert_from.cc | ||
| convert_jpeg.cc | ||
| convert_to_argb.cc | ||
| convert_to_i420.cc | ||
| convert.cc | ||
| cpu_id.cc | ||
| mjpeg_decoder.cc | ||
| mjpeg_validate.cc | ||
| planar_functions.cc | ||
| rotate_any.cc | ||
| rotate_argb.cc | ||
| rotate_common.cc | ||
| rotate_gcc.cc | ||
| rotate_lasx.cc | ||
| rotate_lsx.cc | ||
| rotate_neon64.cc | ||
| rotate_neon.cc | ||
| rotate_rvv.cc | ||
| rotate_sme.cc | ||
| rotate_sve.cc | ||
| rotate_wasm.cc | ||
| rotate_win.cc | ||
| rotate.cc | ||
| row_any.cc | ||
| row_common.cc | ||
| row_gcc.cc | ||
| row_lasx.cc | ||
| row_lsx.cc | ||
| row_neon64.cc | ||
| row_neon.cc | ||
| row_rvv.cc | ||
| row_sme.cc | ||
| row_sve.cc | ||
| row_wasm.cc | ||
| row_win.cc | ||
| scale_any.cc | ||
| scale_argb.cc | ||
| scale_common.cc | ||
| scale_gcc.cc | ||
| scale_lsx.cc | ||
| scale_neon64.cc | ||
| scale_neon.cc | ||
| scale_rgb.cc | ||
| scale_rvv.cc | ||
| scale_sme.cc | ||
| scale_uv.cc | ||
| scale_wasm.cc | ||
| scale_win.cc | ||
| scale.cc | ||
| test.sh | ||
| video_common.cc | ||