mirror of
https://chromium.googlesource.com/libyuv/libyuv
synced 2026-10-29 02:17:13 +08:00
2404f9730e
Reduce loop unrolling across LoongArch LSX (128-bit) and LASX (256-bit) compare, row, rotate, and scale kernels to match SSE2/SSSE3 and AVX2 (1-2 vectors per iteration), rewrite intrinsics/macro implementations into GCC/Clang inline assembly, update all _Any wrapper masks and IS_ALIGNED dispatch checks, and port all remaining LSX Convert (+19) and Rotate (+2) kernels to LASX for full parity: ISA | Compare | Convert | Rotate | Scale -----------------|---------|---------|---------|-------- MIPS/Loong (Any) | 1 | 68 | 4 | 20 LSX | 1 | 68 | 4 | 20 LASX | 1 | 68 | 4 | 0 Removes 1,935 macro/helper lines from loongson_intrinsics.h (-99.3%) and reduces total LoongArch compiled .text size by 9,799 bytes (-25.1% overall, -37.4% row_lsx.cc, -52.3% scale_lsx.cc, -31.2% rotate_lsx.cc, -4.9% row_lasx.cc while adding +25 LASX functions): File | Before | After | Change -------------------|----------|----------|-------- scale_lsx.cc.o | 6,055 B | 2,888 B | -52.3% row_lsx.cc.o | 17,252 B | 10,792 B | -37.4% rotate_lsx.cc.o | 1,488 B | 1,024 B | -31.2% row_lasx.cc.o | 13,740 B | 13,072 B | -4.9% rotate_lasx.cc.o | 544 B | 1,504 B | +176.5% TOTAL .text | 39,079 B | 29,280 B | -25.1% R=wtc@google.com, yuanhecai@loongson.cn Test: qemu-loongarch64 -cpu la464 libyuv_unittest (LASX and LIBYUV_DISABLE_LASX=1) Change-Id: I11c1143ad2aad0399f10b2c3fc9bc4c02c740c59 Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/8564659 Reviewed-by: richard winterton <rrwinterton@gmail.com>
183 lines
7.6 KiB
C++
183 lines
7.6 KiB
C++
/*
|
|
* Copyright 2015 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.h"
|
|
#include "libyuv/rotate_row.h"
|
|
|
|
#ifdef __cplusplus
|
|
namespace libyuv {
|
|
extern "C" {
|
|
#endif
|
|
|
|
#define TANY(NAMEANY, TPOS_SIMD, TPOS_C, MASK) \
|
|
void NAMEANY(const uint8_t* src, int src_stride, uint8_t* dst, \
|
|
int dst_stride, int width) { \
|
|
int r = width & MASK; \
|
|
int n = width - r; \
|
|
if (n > 0) { \
|
|
TPOS_SIMD(src, src_stride, dst, dst_stride, n); \
|
|
} \
|
|
ptrdiff_t np = n; \
|
|
TPOS_C(src + np, src_stride, dst + np * dst_stride, dst_stride, r); \
|
|
}
|
|
|
|
#ifdef HAS_TRANSPOSEWX8_NEON
|
|
TANY(TransposeWx8_Any_NEON, TransposeWx8_NEON, TransposeWx8_C, 7)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX16_NEON
|
|
TANY(TransposeWx16_Any_NEON, TransposeWx16_NEON, TransposeWx16_C, 15)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX8_SSSE3
|
|
TANY(TransposeWx8_Any_SSSE3, TransposeWx8_SSSE3, TransposeWx8_C, 15)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX16_AVX2
|
|
TANY(TransposeWx16_Any_AVX2, TransposeWx16_AVX2, TransposeWx16_C, 15)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX8_LSX
|
|
TANY(TransposeWx8_Any_LSX, TransposeWx8_LSX, TransposeWx8_C, 15)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX16_LASX
|
|
TANY(TransposeWx16_Any_LASX, TransposeWx16_LASX, TransposeWx16_C, 15)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX8_WASMSIMD
|
|
TANY(TransposeWx8_Any_WASMSIMD, TransposeWx8_WASMSIMD, TransposeWx8_C, 15)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX16_WASMSIMD
|
|
TANY(TransposeWx16_Any_WASMSIMD, TransposeWx16_WASMSIMD, TransposeWx16_C, 15)
|
|
#endif
|
|
#undef TANY
|
|
|
|
#define TUVANY(NAMEANY, TPOS_SIMD, TPOS_C, MASK) \
|
|
void NAMEANY(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 r = width & MASK; \
|
|
int n = width - r; \
|
|
if (n > 0) { \
|
|
TPOS_SIMD(src, src_stride, dst_a, dst_stride_a, dst_b, dst_stride_b, n); \
|
|
} \
|
|
ptrdiff_t np = n; \
|
|
TPOS_C(src + np * 2, src_stride, dst_a + np * dst_stride_a, dst_stride_a, \
|
|
dst_b + np * dst_stride_b, dst_stride_b, r); \
|
|
}
|
|
|
|
#ifdef HAS_TRANSPOSEUVWX8_NEON
|
|
TUVANY(TransposeUVWx8_Any_NEON, TransposeUVWx8_NEON, TransposeUVWx8_C, 7)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEUVWX8_SSE2
|
|
TUVANY(TransposeUVWx8_Any_SSE2, TransposeUVWx8_SSE2, TransposeUVWx8_C, 7)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEUVWX16_AVX2
|
|
TUVANY(TransposeUVWx16_Any_AVX2, TransposeUVWx16_AVX2, TransposeUVWx16_C, 7)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEUVWX8_LSX
|
|
TUVANY(TransposeUVWx8_Any_LSX, TransposeUVWx8_LSX, TransposeUVWx8_C, 7)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEUVWX16_LASX
|
|
TUVANY(TransposeUVWx16_Any_LASX, TransposeUVWx16_LASX, TransposeUVWx16_C, 7)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEUVWX8_WASMSIMD
|
|
TUVANY(TransposeUVWx8_Any_WASMSIMD,
|
|
TransposeUVWx8_WASMSIMD,
|
|
TransposeUVWx8_C,
|
|
7)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEUVWX16_WASMSIMD
|
|
TUVANY(TransposeUVWx16_Any_WASMSIMD,
|
|
TransposeUVWx16_WASMSIMD,
|
|
TransposeUVWx16_C,
|
|
7)
|
|
#endif
|
|
#undef TUVANY
|
|
|
|
#define TANY32(NAMEANY, TPOS_SIMD, TPOS_C, MASK) \
|
|
void NAMEANY(const uint8_t* src, int src_stride, uint8_t* dst, \
|
|
int dst_stride, int width) { \
|
|
int r = width & MASK; \
|
|
int n = width - r; \
|
|
if (n > 0) { \
|
|
TPOS_SIMD(src, src_stride, dst, dst_stride, n); \
|
|
} \
|
|
if (r > 0) { \
|
|
TPOS_C(src + (ptrdiff_t)n * src_stride, src_stride, \
|
|
dst + (ptrdiff_t)n * 4, dst_stride, 4, r); \
|
|
} \
|
|
}
|
|
|
|
#ifdef HAS_TRANSPOSEWX4_32_SSE2
|
|
TANY32(TransposeWx4_32_Any_SSE2, TransposeWx4_32_SSE2, TransposeWxH_32_C, 3)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX4_32_LSX
|
|
TANY32(TransposeWx4_32_Any_LSX, TransposeWx4_32_LSX, TransposeWxH_32_C, 3)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX4_32_LASX
|
|
TANY32(TransposeWx4_32_Any_LASX, TransposeWx4_32_LASX, TransposeWxH_32_C, 7)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX4_32_WASMSIMD
|
|
TANY32(TransposeWx4_32_Any_WASMSIMD,
|
|
TransposeWx4_32_WASMSIMD,
|
|
TransposeWxH_32_C,
|
|
3)
|
|
#endif
|
|
#undef TANY32
|
|
|
|
#define TANY32_SRC(NAMEANY, TPOS_SIMD, MASK, HEIGHT) \
|
|
void NAMEANY(const uint8_t* src, int src_stride, uint8_t* dst, \
|
|
int dst_stride, int width) { \
|
|
int r = width & MASK; \
|
|
int n = width - r; \
|
|
if (n > 0) { \
|
|
TPOS_SIMD(src, src_stride, dst, dst_stride, n); \
|
|
} \
|
|
ptrdiff_t np = n; \
|
|
TransposeWxH_32_C(src + np * 4, src_stride, dst + np * dst_stride, \
|
|
dst_stride, r, HEIGHT); \
|
|
}
|
|
|
|
#ifdef HAS_TRANSPOSEWX8_32_AVX2
|
|
TANY32_SRC(TransposeWx8_32_Any_AVX2, TransposeWx8_32_AVX2, 3, 8)
|
|
#endif
|
|
#undef TANY32_SRC
|
|
|
|
#define TANY16(NAMEANY, TPOS_SIMD, MASK, HEIGHT) \
|
|
void NAMEANY(const uint16_t* src, int src_stride, uint16_t* dst, \
|
|
int dst_stride, int width) { \
|
|
int r = width & MASK; \
|
|
int n = width - r; \
|
|
if (n > 0) { \
|
|
TPOS_SIMD(src, src_stride, dst, dst_stride, n); \
|
|
} \
|
|
ptrdiff_t np = n; \
|
|
TransposeWxH_16_C(src + np, src_stride, dst + np * dst_stride, dst_stride, \
|
|
r, HEIGHT); \
|
|
}
|
|
|
|
#ifdef HAS_TRANSPOSEWX8_16_NEON
|
|
TANY16(TransposeWx8_16_Any_NEON, TransposeWx8_16_NEON, 7, 8)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX16_16_AVX2
|
|
TANY16(TransposeWx16_16_Any_AVX2, TransposeWx16_16_AVX2, 7, 16)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX8_16_LSX
|
|
TANY16(TransposeWx8_16_Any_LSX, TransposeWx8_16_LSX, 7, 8)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX16_16_LASX
|
|
TANY16(TransposeWx16_16_Any_LASX, TransposeWx16_16_LASX, 7, 16)
|
|
#endif
|
|
#ifdef HAS_TRANSPOSEWX8_16_WASMSIMD
|
|
TANY16(TransposeWx8_16_Any_WASMSIMD, TransposeWx8_16_WASMSIMD, 7, 8)
|
|
#endif
|
|
#undef TANY16
|
|
|
|
#ifdef __cplusplus
|
|
} // extern "C"
|
|
} // namespace libyuv
|
|
#endif
|