1 /*
2 * Copyright (c) 2015 The WebM project authors. All Rights Reserved.
3 *
4 * Use of this source code is governed by a BSD-style license
5 * that can be found in the LICENSE file in the root of the source
6 * tree. An additional intellectual property rights grant can be found
7 * in the file PATENTS. All contributing project authors may
8 * be found in the AUTHORS file in the root of the source tree.
9 */
10
11 #include "./vp8_rtcd.h"
12 #include "vp8/common/mips/msa/vp8_macros_msa.h"
13
copy_8x4_msa(uint8_t * src,int32_t src_stride,uint8_t * dst,int32_t dst_stride)14 static void copy_8x4_msa(uint8_t *src, int32_t src_stride, uint8_t *dst,
15 int32_t dst_stride) {
16 uint64_t src0, src1, src2, src3;
17
18 LD4(src, src_stride, src0, src1, src2, src3);
19 SD4(src0, src1, src2, src3, dst, dst_stride);
20 }
21
copy_8x8_msa(uint8_t * src,int32_t src_stride,uint8_t * dst,int32_t dst_stride)22 static void copy_8x8_msa(uint8_t *src, int32_t src_stride, uint8_t *dst,
23 int32_t dst_stride) {
24 uint64_t src0, src1, src2, src3;
25
26 LD4(src, src_stride, src0, src1, src2, src3);
27 src += (4 * src_stride);
28 SD4(src0, src1, src2, src3, dst, dst_stride);
29 dst += (4 * dst_stride);
30
31 LD4(src, src_stride, src0, src1, src2, src3);
32 SD4(src0, src1, src2, src3, dst, dst_stride);
33 }
34
copy_16x16_msa(uint8_t * src,int32_t src_stride,uint8_t * dst,int32_t dst_stride)35 static void copy_16x16_msa(uint8_t *src, int32_t src_stride, uint8_t *dst,
36 int32_t dst_stride) {
37 v16u8 src0, src1, src2, src3, src4, src5, src6, src7;
38 v16u8 src8, src9, src10, src11, src12, src13, src14, src15;
39
40 LD_UB8(src, src_stride, src0, src1, src2, src3, src4, src5, src6, src7);
41 src += (8 * src_stride);
42 LD_UB8(src, src_stride, src8, src9, src10, src11, src12, src13, src14, src15);
43
44 ST_UB8(src0, src1, src2, src3, src4, src5, src6, src7, dst, dst_stride);
45 dst += (8 * dst_stride);
46 ST_UB8(src8, src9, src10, src11, src12, src13, src14, src15, dst, dst_stride);
47 }
48
vp8_copy_mem16x16_msa(uint8_t * src,int32_t src_stride,uint8_t * dst,int32_t dst_stride)49 void vp8_copy_mem16x16_msa(uint8_t *src, int32_t src_stride, uint8_t *dst,
50 int32_t dst_stride) {
51 copy_16x16_msa(src, src_stride, dst, dst_stride);
52 }
53
vp8_copy_mem8x8_msa(uint8_t * src,int32_t src_stride,uint8_t * dst,int32_t dst_stride)54 void vp8_copy_mem8x8_msa(uint8_t *src, int32_t src_stride, uint8_t *dst,
55 int32_t dst_stride) {
56 copy_8x8_msa(src, src_stride, dst, dst_stride);
57 }
58
vp8_copy_mem8x4_msa(uint8_t * src,int32_t src_stride,uint8_t * dst,int32_t dst_stride)59 void vp8_copy_mem8x4_msa(uint8_t *src, int32_t src_stride, uint8_t *dst,
60 int32_t dst_stride) {
61 copy_8x4_msa(src, src_stride, dst, dst_stride);
62 }
63