FFmpeg
h264qpel_init_aarch64.c
Go to the documentation of this file.
1 /*
2  * ARM NEON optimised DSP functions
3  * Copyright (c) 2008 Mans Rullgard <mans@mansr.com>
4  *
5  * This file is part of FFmpeg.
6  *
7  * FFmpeg is free software; you can redistribute it and/or
8  * modify it under the terms of the GNU Lesser General Public
9  * License as published by the Free Software Foundation; either
10  * version 2.1 of the License, or (at your option) any later version.
11  *
12  * FFmpeg is distributed in the hope that it will be useful,
13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15  * Lesser General Public License for more details.
16  *
17  * You should have received a copy of the GNU Lesser General Public
18  * License along with FFmpeg; if not, write to the Free Software
19  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20  */
21 
22 #include <stdint.h>
23 
24 #include "config.h"
25 #include "libavutil/attributes.h"
26 #include "libavutil/cpu.h"
27 #include "libavutil/aarch64/cpu.h"
28 #include "libavcodec/h264qpel.h"
29 
30 void ff_put_h264_qpel16_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
31 void ff_put_h264_qpel16_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
32 void ff_put_h264_qpel16_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
33 void ff_put_h264_qpel16_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
34 void ff_put_h264_qpel16_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
35 void ff_put_h264_qpel16_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
36 void ff_put_h264_qpel16_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
37 void ff_put_h264_qpel16_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
38 void ff_put_h264_qpel16_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
39 void ff_put_h264_qpel16_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
40 void ff_put_h264_qpel16_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
41 void ff_put_h264_qpel16_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
42 void ff_put_h264_qpel16_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
43 void ff_put_h264_qpel16_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
44 void ff_put_h264_qpel16_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
45 void ff_put_h264_qpel16_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
46 
47 void ff_put_h264_qpel8_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
48 void ff_put_h264_qpel8_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
49 void ff_put_h264_qpel8_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
50 void ff_put_h264_qpel8_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
51 void ff_put_h264_qpel8_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
52 void ff_put_h264_qpel8_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
53 void ff_put_h264_qpel8_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
54 void ff_put_h264_qpel8_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
55 void ff_put_h264_qpel8_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
56 void ff_put_h264_qpel8_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
57 void ff_put_h264_qpel8_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
58 void ff_put_h264_qpel8_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
59 void ff_put_h264_qpel8_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
60 void ff_put_h264_qpel8_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
61 void ff_put_h264_qpel8_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
62 void ff_put_h264_qpel8_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
63 
64 void ff_avg_h264_qpel16_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
65 void ff_avg_h264_qpel16_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
66 void ff_avg_h264_qpel16_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
67 void ff_avg_h264_qpel16_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
68 void ff_avg_h264_qpel16_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
69 void ff_avg_h264_qpel16_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
70 void ff_avg_h264_qpel16_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
71 void ff_avg_h264_qpel16_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
72 void ff_avg_h264_qpel16_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
73 void ff_avg_h264_qpel16_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
74 void ff_avg_h264_qpel16_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
75 void ff_avg_h264_qpel16_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
76 void ff_avg_h264_qpel16_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
77 void ff_avg_h264_qpel16_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
78 void ff_avg_h264_qpel16_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
79 void ff_avg_h264_qpel16_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
80 
81 void ff_avg_h264_qpel8_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
82 void ff_avg_h264_qpel8_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
83 void ff_avg_h264_qpel8_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
84 void ff_avg_h264_qpel8_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
85 void ff_avg_h264_qpel8_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
86 void ff_avg_h264_qpel8_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
87 void ff_avg_h264_qpel8_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
88 void ff_avg_h264_qpel8_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
89 void ff_avg_h264_qpel8_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
90 void ff_avg_h264_qpel8_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
91 void ff_avg_h264_qpel8_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
92 void ff_avg_h264_qpel8_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
93 void ff_avg_h264_qpel8_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
94 void ff_avg_h264_qpel8_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
95 void ff_avg_h264_qpel8_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
96 void ff_avg_h264_qpel8_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride);
97 
99 {
100  const int high_bit_depth = bit_depth > 8;
101  int cpu_flags = av_get_cpu_flags();
102 
103  if (have_neon(cpu_flags) && !high_bit_depth) {
120 
137 
154 
171  }
172 }
void ff_put_h264_qpel16_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
av_cold void ff_h264qpel_init_aarch64(H264QpelContext *c, int bit_depth)
static atomic_int cpu_flags
Definition: cpu.c:50
#define src
Definition: vp8dsp.c:254
void ff_avg_h264_qpel16_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
Macro definitions for various function/variable attributes.
void ff_avg_h264_qpel16_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
uint8_t
#define av_cold
Definition: attributes.h:82
void ff_avg_h264_qpel16_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
Definition: undefined.txt:32
void ff_put_h264_qpel8_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
qpel_mc_func avg_h264_qpel_pixels_tab[4][16]
Definition: h264qpel.h:29
void ff_put_h264_qpel16_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
qpel_mc_func put_h264_qpel_pixels_tab[4][16]
Definition: h264qpel.h:28
void ff_avg_h264_qpel16_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
#define have_neon(flags)
Definition: cpu.h:26
void ff_avg_h264_qpel8_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
static void bit_depth(AudioStatsContext *s, uint64_t mask, uint64_t imask, AVRational *depth)
Definition: af_astats.c:155
void ff_put_h264_qpel16_mc00_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc20_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc31_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc32_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
int av_get_cpu_flags(void)
Return the flags which specify extensions supported by the CPU.
Definition: cpu.c:93
void ff_avg_h264_qpel8_mc12_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc11_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc13_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc02_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
GLint GLenum GLboolean GLsizei stride
Definition: opengl_enc.c:105
void ff_put_h264_qpel16_mc22_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc01_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel8_mc21_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel16_mc30_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc03_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc33_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_avg_h264_qpel16_mc10_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)
void ff_put_h264_qpel8_mc23_neon(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)