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>
229 lines
6.6 KiB
C++
229 lines
6.6 KiB
C++
/*
|
|
* Copyright 2011 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 "libyuv/rotate_row.h"
|
|
#include "libyuv/row.h"
|
|
|
|
#ifdef __cplusplus
|
|
namespace libyuv {
|
|
extern "C" {
|
|
#endif
|
|
|
|
void TransposeWx8_C(const uint8_t* src,
|
|
int src_stride,
|
|
uint8_t* dst,
|
|
int dst_stride,
|
|
int width) {
|
|
int i;
|
|
for (i = 0; i < width; ++i) {
|
|
dst[0] = src[0 * src_stride];
|
|
dst[1] = src[1 * src_stride];
|
|
dst[2] = src[2 * src_stride];
|
|
dst[3] = src[3 * src_stride];
|
|
dst[4] = src[4 * src_stride];
|
|
dst[5] = src[5 * src_stride];
|
|
dst[6] = src[6 * src_stride];
|
|
dst[7] = src[7 * src_stride];
|
|
++src;
|
|
dst += dst_stride;
|
|
}
|
|
}
|
|
|
|
void TransposeWx16_C(const uint8_t* src,
|
|
int src_stride,
|
|
uint8_t* dst,
|
|
int dst_stride,
|
|
int width) {
|
|
TransposeWx8_C(src, src_stride, dst, dst_stride, width);
|
|
TransposeWx8_C((src + 8 * src_stride), src_stride, (dst + 8), dst_stride,
|
|
width);
|
|
}
|
|
|
|
void TransposeUVWx8_C(const uint8_t* src,
|
|
int src_stride,
|
|
uint8_t* dst_a,
|
|
int dst_stride_a,
|
|
uint8_t* dst_b,
|
|
int dst_stride_b,
|
|
int width) {
|
|
int i;
|
|
for (i = 0; i < width; ++i) {
|
|
dst_a[0] = src[0 * src_stride + 0];
|
|
dst_b[0] = src[0 * src_stride + 1];
|
|
dst_a[1] = src[1 * src_stride + 0];
|
|
dst_b[1] = src[1 * src_stride + 1];
|
|
dst_a[2] = src[2 * src_stride + 0];
|
|
dst_b[2] = src[2 * src_stride + 1];
|
|
dst_a[3] = src[3 * src_stride + 0];
|
|
dst_b[3] = src[3 * src_stride + 1];
|
|
dst_a[4] = src[4 * src_stride + 0];
|
|
dst_b[4] = src[4 * src_stride + 1];
|
|
dst_a[5] = src[5 * src_stride + 0];
|
|
dst_b[5] = src[5 * src_stride + 1];
|
|
dst_a[6] = src[6 * src_stride + 0];
|
|
dst_b[6] = src[6 * src_stride + 1];
|
|
dst_a[7] = src[7 * src_stride + 0];
|
|
dst_b[7] = src[7 * src_stride + 1];
|
|
src += 2;
|
|
dst_a += dst_stride_a;
|
|
dst_b += dst_stride_b;
|
|
}
|
|
}
|
|
|
|
void TransposeUVWx16_C(const uint8_t* src,
|
|
int src_stride,
|
|
uint8_t* dst_a,
|
|
int dst_stride_a,
|
|
uint8_t* dst_b,
|
|
int dst_stride_b,
|
|
int width) {
|
|
TransposeUVWx8_C(src, src_stride, dst_a, dst_stride_a, dst_b, dst_stride_b,
|
|
width);
|
|
TransposeUVWx8_C((src + 8 * src_stride), src_stride, (dst_a + 8),
|
|
dst_stride_a, (dst_b + 8), dst_stride_b, width);
|
|
}
|
|
|
|
void TransposeWxH_C(const uint8_t* src,
|
|
int src_stride,
|
|
uint8_t* dst,
|
|
int dst_stride,
|
|
int width,
|
|
int height) {
|
|
int i;
|
|
for (i = 0; i < width; ++i) {
|
|
int j;
|
|
for (j = 0; j < height; ++j) {
|
|
dst[i * dst_stride + j] = src[j * src_stride + i];
|
|
}
|
|
}
|
|
}
|
|
|
|
void TransposeUVWxH_C(const uint8_t* src,
|
|
int src_stride,
|
|
uint8_t* dst_a,
|
|
int dst_stride_a,
|
|
uint8_t* dst_b,
|
|
int dst_stride_b,
|
|
int width,
|
|
int height) {
|
|
int i;
|
|
for (i = 0; i < width * 2; i += 2) {
|
|
int j;
|
|
for (j = 0; j < height; ++j) {
|
|
dst_a[((i >> 1) * dst_stride_a) + j] = src[i + (j * src_stride)];
|
|
dst_b[((i >> 1) * dst_stride_b) + j] = src[i + (j * src_stride) + 1];
|
|
}
|
|
}
|
|
}
|
|
|
|
void TransposeWx8_16_C(const uint16_t* src,
|
|
int src_stride,
|
|
uint16_t* dst,
|
|
int dst_stride,
|
|
int width) {
|
|
int i;
|
|
for (i = 0; i < width; ++i) {
|
|
dst[0] = src[0 * src_stride];
|
|
dst[1] = src[1 * src_stride];
|
|
dst[2] = src[2 * src_stride];
|
|
dst[3] = src[3 * src_stride];
|
|
dst[4] = src[4 * src_stride];
|
|
dst[5] = src[5 * src_stride];
|
|
dst[6] = src[6 * src_stride];
|
|
dst[7] = src[7 * src_stride];
|
|
++src;
|
|
dst += dst_stride;
|
|
}
|
|
}
|
|
|
|
void TransposeWxH_16_C(const uint16_t* src,
|
|
int src_stride,
|
|
uint16_t* dst,
|
|
int dst_stride,
|
|
int width,
|
|
int height) {
|
|
int i;
|
|
for (i = 0; i < width; ++i) {
|
|
int j;
|
|
for (j = 0; j < height; ++j) {
|
|
dst[i * dst_stride + j] = src[j * src_stride + i];
|
|
}
|
|
}
|
|
}
|
|
|
|
static inline void Copy4(uint8_t* dst, const uint8_t* src) {
|
|
dst[0] = src[0];
|
|
dst[1] = src[1];
|
|
dst[2] = src[2];
|
|
dst[3] = src[3];
|
|
}
|
|
|
|
// Transpose 32 bit values (ARGB)
|
|
void TransposeWx4_32_C(const uint8_t* src,
|
|
int src_stride,
|
|
uint8_t* dst,
|
|
int dst_stride,
|
|
int width) {
|
|
const uint8_t* src1 = src + src_stride;
|
|
const uint8_t* src2 = src1 + src_stride;
|
|
const uint8_t* src3 = src2 + src_stride;
|
|
uint8_t* dst1 = dst + dst_stride;
|
|
uint8_t* dst2 = dst1 + dst_stride;
|
|
uint8_t* dst3 = dst2 + dst_stride;
|
|
int i;
|
|
for (i = 0; i < width; i += 4) {
|
|
Copy4(dst + 0, src + 0);
|
|
Copy4(dst + 4, src1 + 0);
|
|
Copy4(dst + 8, src2 + 0);
|
|
Copy4(dst + 12, src3 + 0);
|
|
Copy4(dst1 + 0, src + 4);
|
|
Copy4(dst1 + 4, src1 + 4);
|
|
Copy4(dst1 + 8, src2 + 4);
|
|
Copy4(dst1 + 12, src3 + 4);
|
|
Copy4(dst2 + 0, src + 8);
|
|
Copy4(dst2 + 4, src1 + 8);
|
|
Copy4(dst2 + 8, src2 + 8);
|
|
Copy4(dst2 + 12, src3 + 8);
|
|
Copy4(dst3 + 0, src + 12);
|
|
Copy4(dst3 + 4, src1 + 12);
|
|
Copy4(dst3 + 8, src2 + 12);
|
|
Copy4(dst3 + 12, src3 + 12);
|
|
src += (ptrdiff_t)src_stride * 4; // advance 4 rows
|
|
src1 += (ptrdiff_t)src_stride * 4;
|
|
src2 += (ptrdiff_t)src_stride * 4;
|
|
src3 += (ptrdiff_t)src_stride * 4;
|
|
dst += 4 * 4; // advance 4 columns
|
|
dst1 += 4 * 4;
|
|
dst2 += 4 * 4;
|
|
dst3 += 4 * 4;
|
|
}
|
|
}
|
|
|
|
void TransposeWxH_32_C(const uint8_t* src,
|
|
int src_stride,
|
|
uint8_t* dst,
|
|
int dst_stride,
|
|
int width,
|
|
int height) {
|
|
int i;
|
|
for (i = 0; i < width; ++i) {
|
|
int j;
|
|
for (j = 0; j < height; ++j) {
|
|
Copy4(dst + (ptrdiff_t)i * dst_stride + j * 4,
|
|
src + (ptrdiff_t)j * src_stride + i * 4);
|
|
}
|
|
}
|
|
}
|
|
|
|
#ifdef __cplusplus
|
|
} // extern "C"
|
|
} // namespace libyuv
|
|
#endif
|