mirror of
https://chromium.googlesource.com/libyuv/libyuv
synced 2026-10-28 18:07:12 +08:00
16af162cf9
Implement SIMD 16-bit plane rotation (TransposePlane_16, RotatePlane90_16, RotatePlane180_16, RotatePlane270_16) used by I010Rotate, I210Rotate, and I410Rotate, keeping tile sizes compact with full-register writes and zero cross-lane shuffles: - AVX2: 8x16 tile in ymm0..ymm15 with _Any tail (low-end x86 SIMD) - AVX-512BW: 8x32 tile in zmm0..zmm15 with k1/k2 right/bottom masks - NEON (ARM64/ARM32): 8x16 tile in v0..v23 / q0..q11 with _Any tail - SME: 8xVL tile using ZA16 horizontal/vertical array slices - RVV: 8xVL tile using widening unpacks with hardware vl tail - WASM SIMD: 8x8 tile in v128 with _Any tail Also update 32-bit ARGBRotate x86 transposes to compact vertical in-lane tiles with full-register writes and zero cross-lane shuffles: - TransposeWxH_32_AVX2: 4x8 tile in ymm0..ymm7 with vpmaskmovd tails - TransposeWxH_32_AVX512BW: 4x16 tile in zmm0..zmm7 with k1/k2 tails Add RotatePlane16_SplitMergeUV_Opt and RotatePlane16_SplitMergeUV_Odd unit tests comparing 16-bit rotation against SplitUVPlane + RotatePlane + MergeUVPlane across 0/90/180/270 degrees. Performance on 1275x1650 (1000 iterations, before -> after): CPU 90 degree 180 degree Coral RVV (Verilator) 7306 -> 1415c (5.16x) 2274 -> 719c (3.16x) AMD Rome (Zen 2) 821 -> 215ms (3.82x) 259 -> 181ms (1.44x) AMD Genoa (Zen 4) 483 -> 129ms (3.73x) 149 -> 111ms (1.34x) AMD Turin (Zen 5) 321 -> 101ms (3.18x) 109 -> 78ms (1.40x) AMD Milan (Zen 3) 505 -> 189ms (2.67x) 199 -> 118ms (1.69x) Intel Skylake-X 1076 -> 447ms (2.41x) 353 -> 317ms (1.11x) WASM SIMD (d8) 415 -> 203ms (2.04x) 370 -> 106ms (3.49x) Intel Emerald Rapids 735 -> 422ms (1.74x) 238 -> 233ms (1.02x) Intel Haswell 1120 -> 648ms (1.73x) 438 -> 393ms (1.11x) Intel Ice Lake 774 -> 474ms (1.63x) 279 -> 290ms (0.96x) ARM Neoverse-N1 864 -> 544ms (1.59x) 404 -> 266ms (1.52x) Ampere Siryn 862 -> 576ms (1.50x) 568 -> 329ms (1.72x) Google Axion (Maple) 554 -> 379ms (1.46x) 199 -> 142ms (1.41x) Intel Sapphire Rapids 1126 -> 772ms (1.46x) 301 -> 288ms (1.04x) Intel Granite Rapids 730 -> 565ms (1.29x) 280 -> 280ms (1.00x) NVIDIA Grace 462 -> 365ms (1.26x) 192 -> 126ms (1.52x) ARM Cortex-A53 28290 -> 24386ms (1.16x) 7178 -> 5075ms (1.41x) Google Axion 2 (Tamar) 523 -> 453ms (1.15x) 217 -> 112ms (1.93x) Test: libyuv_unittest --gtest_filter=*Rotate* Bug: b/42280942 Change-Id: I91ba6cac22185b5d294fcb98344fdb8a23cc94ff Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8545294 Reviewed-by: richard winterton <rrwinterton@gmail.com> Commit-Queue: Frank Barchard <fbarchard@google.com>
442 lines
18 KiB
C++
442 lines
18 KiB
C++
/*
|
|
* Copyright 2012 The LibYuv Project Authors. All rights reserved.
|
|
*
|
|
* Use of this source code is governed by a BSD-style license
|
|
* that can be found in the LICENSE file in the root of the source
|
|
* tree. An additional intellectual property rights grant can be found
|
|
* in the file PATENTS. All contributing project authors may
|
|
* be found in the AUTHORS file in the root of the source tree.
|
|
*/
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include "../unit_test/unit_test.h"
|
|
#include "libyuv/cpu_id.h"
|
|
#include "libyuv/planar_functions.h"
|
|
#include "libyuv/rotate_argb.h"
|
|
|
|
namespace libyuv {
|
|
|
|
static void TestRotateBpp(int src_width,
|
|
int src_height,
|
|
int dst_width,
|
|
int dst_height,
|
|
libyuv::RotationMode mode,
|
|
int benchmark_iterations,
|
|
int disable_cpu_flags,
|
|
int benchmark_cpu_info,
|
|
const int kBpp) {
|
|
if (src_width < 1) {
|
|
src_width = 1;
|
|
}
|
|
if (src_height < 1) {
|
|
src_height = 1;
|
|
}
|
|
if (dst_width < 1) {
|
|
dst_width = 1;
|
|
}
|
|
if (dst_height < 1) {
|
|
dst_height = 1;
|
|
}
|
|
int src_stride_argb = src_width * kBpp;
|
|
int src_argb_plane_size = src_stride_argb * abs(src_height);
|
|
align_buffer_page_end(src_argb, src_argb_plane_size);
|
|
for (int i = 0; i < src_argb_plane_size; ++i) {
|
|
src_argb[i] = fastrand() & 0xff;
|
|
}
|
|
|
|
int dst_stride_argb = dst_width * kBpp;
|
|
int dst_argb_plane_size = dst_stride_argb * dst_height;
|
|
align_buffer_page_end(dst_argb_c, dst_argb_plane_size);
|
|
align_buffer_page_end(dst_argb_opt, dst_argb_plane_size);
|
|
memset(dst_argb_c, 2, dst_argb_plane_size);
|
|
memset(dst_argb_opt, 3, dst_argb_plane_size);
|
|
|
|
if (kBpp == 1) {
|
|
MaskCpuFlags(disable_cpu_flags); // Disable all CPU optimization.
|
|
RotatePlane(src_argb, src_stride_argb, dst_argb_c, dst_stride_argb,
|
|
src_width, src_height, mode);
|
|
|
|
MaskCpuFlags(benchmark_cpu_info); // Enable all CPU optimization.
|
|
for (int i = 0; i < benchmark_iterations; ++i) {
|
|
RotatePlane(src_argb, src_stride_argb, dst_argb_opt, dst_stride_argb,
|
|
src_width, src_height, mode);
|
|
}
|
|
} else if (kBpp == 4) {
|
|
MaskCpuFlags(disable_cpu_flags); // Disable all CPU optimization.
|
|
ARGBRotate(src_argb, src_stride_argb, dst_argb_c, dst_stride_argb,
|
|
src_width, src_height, mode);
|
|
|
|
MaskCpuFlags(benchmark_cpu_info); // Enable all CPU optimization.
|
|
for (int i = 0; i < benchmark_iterations; ++i) {
|
|
ARGBRotate(src_argb, src_stride_argb, dst_argb_opt, dst_stride_argb,
|
|
src_width, src_height, mode);
|
|
}
|
|
}
|
|
|
|
// Rotation should be exact.
|
|
for (int i = 0; i < dst_argb_plane_size; ++i) {
|
|
ASSERT_EQ(dst_argb_c[i], dst_argb_opt[i]);
|
|
}
|
|
|
|
free_aligned_buffer_page_end(dst_argb_c);
|
|
free_aligned_buffer_page_end(dst_argb_opt);
|
|
free_aligned_buffer_page_end(src_argb);
|
|
}
|
|
|
|
static void ARGBTestRotate(int src_width,
|
|
int src_height,
|
|
int dst_width,
|
|
int dst_height,
|
|
libyuv::RotationMode mode,
|
|
int benchmark_iterations,
|
|
int disable_cpu_flags,
|
|
int benchmark_cpu_info) {
|
|
TestRotateBpp(src_width, src_height, dst_width, dst_height, mode,
|
|
benchmark_iterations, disable_cpu_flags, benchmark_cpu_info, 4);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, ARGBRotate0_Opt) {
|
|
ARGBTestRotate(benchmark_width_, benchmark_height_, benchmark_width_,
|
|
benchmark_height_, kRotate0, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, ARGBRotate90_Opt) {
|
|
ARGBTestRotate(benchmark_width_, benchmark_height_, benchmark_height_,
|
|
benchmark_width_, kRotate90, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, ARGBRotate180_Opt) {
|
|
ARGBTestRotate(benchmark_width_, benchmark_height_, benchmark_width_,
|
|
benchmark_height_, kRotate180, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, ARGBRotate270_Opt) {
|
|
ARGBTestRotate(benchmark_width_, benchmark_height_, benchmark_height_,
|
|
benchmark_width_, kRotate270, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, ARGBRotate90_Odd) {
|
|
ARGBTestRotate(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_height_ + 1, benchmark_width_ + 1, kRotate90,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, ARGBRotate270_Odd) {
|
|
ARGBTestRotate(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_height_ + 1, benchmark_width_ + 1, kRotate270,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
static void TestRotatePlane(int src_width,
|
|
int src_height,
|
|
int dst_width,
|
|
int dst_height,
|
|
libyuv::RotationMode mode,
|
|
int benchmark_iterations,
|
|
int disable_cpu_flags,
|
|
int benchmark_cpu_info) {
|
|
TestRotateBpp(src_width, src_height, dst_width, dst_height, mode,
|
|
benchmark_iterations, disable_cpu_flags, benchmark_cpu_info, 1);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane0_Opt) {
|
|
TestRotatePlane(benchmark_width_, benchmark_height_, benchmark_width_,
|
|
benchmark_height_, kRotate0, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane90_Opt) {
|
|
TestRotatePlane(benchmark_width_, benchmark_height_, benchmark_height_,
|
|
benchmark_width_, kRotate90, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane180_Opt) {
|
|
TestRotatePlane(benchmark_width_, benchmark_height_, benchmark_width_,
|
|
benchmark_height_, kRotate180, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane270_Opt) {
|
|
TestRotatePlane(benchmark_width_, benchmark_height_, benchmark_height_,
|
|
benchmark_width_, kRotate270, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, DISABLED_RotatePlane0_Odd) {
|
|
TestRotatePlane(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_width_ + 1, benchmark_height_ + 1, kRotate0,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, DISABLED_RotatePlane90_Odd) {
|
|
TestRotatePlane(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_height_ + 1, benchmark_width_ + 1, kRotate90,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, DISABLED_RotatePlane180_Odd) {
|
|
TestRotatePlane(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_width_ + 1, benchmark_height_ + 1, kRotate180,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, DISABLED_RotatePlane270_Odd) {
|
|
TestRotatePlane(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_height_ + 1, benchmark_width_ + 1, kRotate270,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane90_TestStride) {
|
|
int argb_plane_size = benchmark_width_ * 4 * abs(benchmark_height_);
|
|
|
|
align_buffer_page_end(src_argb, argb_plane_size);
|
|
align_buffer_page_end(dst_argb, argb_plane_size);
|
|
|
|
ASSERT_EQ(0, ARGBRotate(src_argb, benchmark_width_ * 4, dst_argb,
|
|
benchmark_width_ * 4, benchmark_width_,
|
|
benchmark_height_, kRotate0));
|
|
|
|
ASSERT_EQ(0, ARGBRotate(src_argb, benchmark_width_ * 4 - 1, dst_argb,
|
|
benchmark_width_ * 4 - 1, benchmark_width_ - 1,
|
|
benchmark_height_, kRotate0));
|
|
|
|
ASSERT_EQ(0, ARGBRotate(src_argb, benchmark_width_ * 4, dst_argb,
|
|
benchmark_width_ * 4, benchmark_width_,
|
|
benchmark_height_, kRotate180));
|
|
|
|
ASSERT_EQ(0, ARGBRotate(src_argb, benchmark_width_ * 4 - 1, dst_argb,
|
|
benchmark_width_ * 4 - 1, benchmark_width_ - 1,
|
|
benchmark_height_, kRotate180));
|
|
|
|
ASSERT_EQ(0, ARGBRotate(src_argb, benchmark_width_ * 4, dst_argb,
|
|
abs(benchmark_height_) * 4, benchmark_width_,
|
|
benchmark_height_, kRotate90));
|
|
|
|
ASSERT_EQ(-1, ARGBRotate(src_argb, benchmark_width_ * 4 - 1, dst_argb,
|
|
abs(benchmark_height_) * 4, benchmark_width_ - 1,
|
|
benchmark_height_, kRotate90));
|
|
|
|
ASSERT_EQ(0, ARGBRotate(src_argb, benchmark_width_ * 4, dst_argb,
|
|
abs(benchmark_height_) * 4, benchmark_width_,
|
|
benchmark_height_, kRotate270));
|
|
|
|
ASSERT_EQ(-1, ARGBRotate(src_argb, benchmark_width_ * 4 - 1, dst_argb,
|
|
abs(benchmark_height_) * 4, benchmark_width_ - 1,
|
|
benchmark_height_, kRotate270));
|
|
|
|
free_aligned_buffer_page_end(dst_argb);
|
|
free_aligned_buffer_page_end(src_argb);
|
|
}
|
|
|
|
static void TestRotatePlane_16(int src_width,
|
|
int src_height,
|
|
int dst_width,
|
|
int dst_height,
|
|
libyuv::RotationMode mode,
|
|
int benchmark_iterations,
|
|
int disable_cpu_flags,
|
|
int benchmark_cpu_info) {
|
|
if (src_width < 1) {
|
|
src_width = 1;
|
|
}
|
|
if (src_height < 1) {
|
|
src_height = 1;
|
|
}
|
|
if (dst_width < 1) {
|
|
dst_width = 1;
|
|
}
|
|
if (dst_height < 1) {
|
|
dst_height = 1;
|
|
}
|
|
int src_stride = src_width;
|
|
int src_plane_size = src_stride * abs(src_height);
|
|
align_buffer_page_end_16(src, src_plane_size);
|
|
for (int i = 0; i < src_plane_size; ++i) {
|
|
src[i] = fastrand() & 0xffff;
|
|
}
|
|
|
|
int dst_stride = dst_width;
|
|
int dst_plane_size = dst_stride * dst_height;
|
|
align_buffer_page_end_16(dst_c, dst_plane_size);
|
|
align_buffer_page_end_16(dst_opt, dst_plane_size);
|
|
memset(dst_c, 2, dst_plane_size * 2);
|
|
memset(dst_opt, 3, dst_plane_size * 2);
|
|
|
|
MaskCpuFlags(disable_cpu_flags); // Disable all CPU optimization.
|
|
RotatePlane_16(src, src_stride, dst_c, dst_stride, src_width, src_height,
|
|
mode);
|
|
|
|
MaskCpuFlags(benchmark_cpu_info); // Enable all CPU optimization.
|
|
for (int i = 0; i < benchmark_iterations; ++i) {
|
|
RotatePlane_16(src, src_stride, dst_opt, dst_stride, src_width, src_height,
|
|
mode);
|
|
}
|
|
|
|
// Rotation should be exact.
|
|
for (int i = 0; i < dst_plane_size; ++i) {
|
|
ASSERT_EQ(dst_c[i], dst_opt[i]);
|
|
}
|
|
|
|
free_aligned_buffer_page_end_16(dst_c);
|
|
free_aligned_buffer_page_end_16(dst_opt);
|
|
free_aligned_buffer_page_end_16(src);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane0_16_Opt) {
|
|
TestRotatePlane_16(benchmark_width_, benchmark_height_, benchmark_width_,
|
|
benchmark_height_, kRotate0, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane90_16_Opt) {
|
|
TestRotatePlane_16(benchmark_width_, benchmark_height_, benchmark_height_,
|
|
benchmark_width_, kRotate90, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane180_16_Opt) {
|
|
TestRotatePlane_16(benchmark_width_, benchmark_height_, benchmark_width_,
|
|
benchmark_height_, kRotate180, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane270_16_Opt) {
|
|
TestRotatePlane_16(benchmark_width_, benchmark_height_, benchmark_height_,
|
|
benchmark_width_, kRotate270, benchmark_iterations_,
|
|
disable_cpu_flags_, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane0_16_Odd) {
|
|
TestRotatePlane_16(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_width_ + 1, benchmark_height_ + 1, kRotate0,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane90_16_Odd) {
|
|
TestRotatePlane_16(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_height_ + 1, benchmark_width_ + 1, kRotate90,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane180_16_Odd) {
|
|
TestRotatePlane_16(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_width_ + 1, benchmark_height_ + 1, kRotate180,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane270_16_Odd) {
|
|
TestRotatePlane_16(benchmark_width_ + 1, benchmark_height_ + 1,
|
|
benchmark_height_ + 1, benchmark_width_ + 1, kRotate270,
|
|
benchmark_iterations_, disable_cpu_flags_,
|
|
benchmark_cpu_info_);
|
|
}
|
|
|
|
static void TestRotatePlane16_SplitMergeUV(int src_width,
|
|
int src_height,
|
|
int dst_width,
|
|
int dst_height,
|
|
libyuv::RotationMode mode,
|
|
int benchmark_cpu_info) {
|
|
if (src_width < 1) {
|
|
src_width = 1;
|
|
}
|
|
if (src_height < 1) {
|
|
src_height = 1;
|
|
}
|
|
if (dst_width < 1) {
|
|
dst_width = 1;
|
|
}
|
|
if (dst_height < 1) {
|
|
dst_height = 1;
|
|
}
|
|
const int src_pixels = src_width * src_height;
|
|
const int dst_pixels = dst_width * dst_height;
|
|
align_buffer_page_end(src_16, src_pixels * 2);
|
|
align_buffer_page_end(dst_16, dst_pixels * 2);
|
|
align_buffer_page_end(ref_16, dst_pixels * 2);
|
|
align_buffer_page_end(src_lo, src_pixels);
|
|
align_buffer_page_end(src_hi, src_pixels);
|
|
align_buffer_page_end(dst_lo, dst_pixels);
|
|
align_buffer_page_end(dst_hi, dst_pixels);
|
|
for (int i = 0; i < src_pixels * 2; ++i) {
|
|
src_16[i] = fastrand() & 0xff;
|
|
}
|
|
memset(dst_16, 2, dst_pixels * 2);
|
|
memset(ref_16, 3, dst_pixels * 2);
|
|
|
|
MaskCpuFlags(benchmark_cpu_info);
|
|
SplitUVPlane(src_16, src_width * 2, src_lo, src_width, src_hi, src_width,
|
|
src_width, src_height);
|
|
RotatePlane(src_lo, src_width, dst_lo, dst_width, src_width, src_height,
|
|
mode);
|
|
RotatePlane(src_hi, src_width, dst_hi, dst_width, src_width, src_height,
|
|
mode);
|
|
MergeUVPlane(dst_lo, dst_width, dst_hi, dst_width, ref_16, dst_width * 2,
|
|
dst_width, dst_height);
|
|
|
|
I410Rotate(reinterpret_cast<const uint16_t*>(src_16), src_width,
|
|
reinterpret_cast<const uint16_t*>(src_16), src_width,
|
|
reinterpret_cast<const uint16_t*>(src_16), src_width,
|
|
reinterpret_cast<uint16_t*>(dst_16), dst_width,
|
|
reinterpret_cast<uint16_t*>(dst_16), dst_width,
|
|
reinterpret_cast<uint16_t*>(dst_16), dst_width, src_width,
|
|
src_height, mode);
|
|
|
|
EXPECT_EQ(0, memcmp(ref_16, dst_16, dst_pixels * 2));
|
|
|
|
free_aligned_buffer_page_end(dst_hi);
|
|
free_aligned_buffer_page_end(dst_lo);
|
|
free_aligned_buffer_page_end(src_hi);
|
|
free_aligned_buffer_page_end(src_lo);
|
|
free_aligned_buffer_page_end(ref_16);
|
|
free_aligned_buffer_page_end(dst_16);
|
|
free_aligned_buffer_page_end(src_16);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane16_SplitMergeUV_Opt) {
|
|
TestRotatePlane16_SplitMergeUV(benchmark_width_, benchmark_height_,
|
|
benchmark_width_, benchmark_height_, kRotate0,
|
|
benchmark_cpu_info_);
|
|
TestRotatePlane16_SplitMergeUV(benchmark_width_, benchmark_height_,
|
|
benchmark_height_, benchmark_width_, kRotate90,
|
|
benchmark_cpu_info_);
|
|
TestRotatePlane16_SplitMergeUV(benchmark_width_, benchmark_height_,
|
|
benchmark_width_, benchmark_height_,
|
|
kRotate180, benchmark_cpu_info_);
|
|
TestRotatePlane16_SplitMergeUV(benchmark_width_, benchmark_height_,
|
|
benchmark_height_, benchmark_width_,
|
|
kRotate270, benchmark_cpu_info_);
|
|
}
|
|
|
|
TEST_F(LibYUVRotateTest, RotatePlane16_SplitMergeUV_Odd) {
|
|
TestRotatePlane16_SplitMergeUV(benchmark_width_ + 3, benchmark_height_ + 5,
|
|
benchmark_width_ + 3, benchmark_height_ + 5,
|
|
kRotate0, benchmark_cpu_info_);
|
|
TestRotatePlane16_SplitMergeUV(benchmark_width_ + 3, benchmark_height_ + 5,
|
|
benchmark_height_ + 5, benchmark_width_ + 3,
|
|
kRotate90, benchmark_cpu_info_);
|
|
TestRotatePlane16_SplitMergeUV(benchmark_width_ + 3, benchmark_height_ + 5,
|
|
benchmark_width_ + 3, benchmark_height_ + 5,
|
|
kRotate180, benchmark_cpu_info_);
|
|
TestRotatePlane16_SplitMergeUV(benchmark_width_ + 3, benchmark_height_ + 5,
|
|
benchmark_height_ + 5, benchmark_width_ + 3,
|
|
kRotate270, benchmark_cpu_info_);
|
|
}
|
|
|
|
} // namespace libyuv
|