• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 /*
2  * Copyright (c) 2015 Henrik Gramner
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 #include "checkasm.h"
23 #include "libavcodec/h264qpel.h"
24 #include "libavutil/common.h"
25 #include "libavutil/internal.h"
26 #include "libavutil/intreadwrite.h"
27 #include "libavutil/mem_internal.h"
28 
29 static const uint32_t pixel_mask[3] = { 0xffffffff, 0x01ff01ff, 0x03ff03ff };
30 
31 #define SIZEOF_PIXEL ((bit_depth + 7) / 8)
32 #define BUF_SIZE (2 * 16 * (16 + 3 + 4))
33 
34 #define randomize_buffers()                        \
35     do {                                           \
36         uint32_t mask = pixel_mask[bit_depth - 8]; \
37         int k;                                     \
38         for (k = 0; k < BUF_SIZE; k += 4) {        \
39             uint32_t r = rnd() & mask;             \
40             AV_WN32A(buf0 + k, r);                 \
41             AV_WN32A(buf1 + k, r);                 \
42             r = rnd();                             \
43             AV_WN32A(dst0 + k, r);                 \
44             AV_WN32A(dst1 + k, r);                 \
45         }                                          \
46     } while (0)
47 
48 #define src0 (buf0 + 3 * 2 * 16) /* h264qpel functions read data from negative src pointer offsets */
49 #define src1 (buf1 + 3 * 2 * 16)
50 
checkasm_check_h264qpel(void)51 void checkasm_check_h264qpel(void)
52 {
53     LOCAL_ALIGNED_16(uint8_t, buf0, [BUF_SIZE]);
54     LOCAL_ALIGNED_16(uint8_t, buf1, [BUF_SIZE]);
55     LOCAL_ALIGNED_16(uint8_t, dst0, [BUF_SIZE]);
56     LOCAL_ALIGNED_16(uint8_t, dst1, [BUF_SIZE]);
57     H264QpelContext h;
58     int op, bit_depth, i, j;
59     declare_func_emms(AV_CPU_FLAG_MMX | AV_CPU_FLAG_MMXEXT, void, uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
60 
61     for (op = 0; op < 2; op++) {
62         qpel_mc_func (*tab)[16] = op ? h.avg_h264_qpel_pixels_tab : h.put_h264_qpel_pixels_tab;
63         const char *op_name = op ? "avg" : "put";
64 
65         for (bit_depth = 8; bit_depth <= 10; bit_depth++) {
66             ff_h264qpel_init(&h, bit_depth);
67             for (i = 0; i < (op ? 3 : 4); i++) {
68                 int size = 16 >> i;
69                 for (j = 0; j < 16; j++)
70                     if (check_func(tab[i][j], "%s_h264_qpel_%d_mc%d%d_%d", op_name, size, j & 3, j >> 2, bit_depth)) {
71                         randomize_buffers();
72                         call_ref(dst0, src0, size * SIZEOF_PIXEL);
73                         call_new(dst1, src1, size * SIZEOF_PIXEL);
74                         if (memcmp(buf0, buf1, BUF_SIZE) || memcmp(dst0, dst1, BUF_SIZE))
75                             fail();
76                         bench_new(dst1, src1, size * SIZEOF_PIXEL);
77                     }
78             }
79         }
80         report("%s", op_name);
81     }
82 }
83