1 /*
2  * Copyright (c) 2016 Alexandra Hájková
3  *
4  * This file is part of FFmpeg.
5  *
6  * FFmpeg is free software; you can redistribute it and/or modify
7  * it under the terms of the GNU General Public License as published by
8  * the Free Software Foundation; either version 2 of the License, or
9  * (at your option) any later version.
10  *
11  * FFmpeg is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
14  * GNU General Public License for more details.
15  *
16  * You should have received a copy of the GNU General Public License along
17  * with FFmpeg; if not, write to the Free Software Foundation, Inc.,
18  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
19  */
20 
21 #include <string.h>
22 
23 #include "libavutil/intreadwrite.h"
24 
25 #include "libavcodec/hevcdsp.h"
26 
27 #include "checkasm.h"
28 
29 #define randomize_buffers(buf, size)            \
30     do {                                        \
31         int j;                                  \
32         for (j = 0; j < size; j++) {            \
33             int16_t r = rnd();                  \
34             AV_WN16A(buf + j, r >> 3);          \
35         }                                       \
36     } while (0)
37 
38 #define randomize_buffers2(buf, size)             \
39     do {                                          \
40         int j;                                    \
41         for (j = 0; j < size; j++)                \
42             AV_WN16A(buf + j * 2, rnd() & 0x3FF); \
43     } while (0)
44 
compare_add_res(int size,ptrdiff_t stride,int overflow_test)45 static void compare_add_res(int size, ptrdiff_t stride, int overflow_test)
46 {
47     LOCAL_ALIGNED_32(int16_t, res0, [32 * 32]);
48     LOCAL_ALIGNED_32(int16_t, res1, [32 * 32]);
49     LOCAL_ALIGNED_32(uint8_t, dst0, [32 * 32 * 2]);
50     LOCAL_ALIGNED_32(uint8_t, dst1, [32 * 32 * 2]);
51 
52     declare_func_emms(AV_CPU_FLAG_MMX, void, uint8_t *dst, int16_t *res, ptrdiff_t stride);
53 
54     randomize_buffers(res0, size);
55     randomize_buffers2(dst0, size);
56     if (overflow_test)
57         res0[0] = 0x8000;
58     memcpy(res1, res0, sizeof(*res0) * size);
59     memcpy(dst1, dst0, sizeof(int16_t) * size);
60 
61     call_ref(dst0, res0, stride);
62     call_new(dst1, res1, stride);
63     if (memcmp(dst0, dst1, size))
64         fail();
65     bench_new(dst1, res1, stride);
66 }
67 
check_add_res(HEVCDSPContext h,int bit_depth)68 static void check_add_res(HEVCDSPContext h, int bit_depth)
69 {
70     int i;
71 
72     for (i = 2; i <= 5; i++) {
73         int block_size = 1 << i;
74         int size = block_size * block_size;
75         ptrdiff_t stride = block_size << (bit_depth > 8);
76 
77         if (check_func(h.add_residual[i - 2], "hevc_add_res_%dx%d_%d", block_size, block_size, bit_depth)) {
78             compare_add_res(size, stride, 0);
79             // overflow test for res = -32768
80             compare_add_res(size, stride, 1);
81         }
82     }
83 }
84 
checkasm_check_hevc_add_res(void)85 void checkasm_check_hevc_add_res(void)
86 {
87     int bit_depth;
88 
89     for (bit_depth = 8; bit_depth <= 10; bit_depth++) {
90         HEVCDSPContext h;
91 
92         ff_hevc_dsp_init(&h, bit_depth);
93         check_add_res(h, bit_depth);
94     }
95     report("add_residual");
96 }
97