1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
|
/* copy pixel --> dst */
#ifdef BUILD_NEON
static void
_op_copy_p_dp_neon(DATA32 *s, DATA8 *m __UNUSED__, DATA32 c __UNUSED__, DATA32 *d, int l) {
//#define USENEON 1
#ifndef USENEON
memcpy(d, s, l * sizeof(DATA32));
return;
#else
DATA32 *e;
e = d + l - 23;
if (e > d)
{
int dl;
asm volatile
(".fpu neon \n\t"
"_op_copy_p_dp_neon_asmloop: \n\t"
"pld [%[s], #192] \n\t" // preload 256 bytes ahead
"pld [%[s], #320] \n\t" // preload 320 bytes ahead
"vld1.32 {d0-d3}, [%[s]]! \n\t" // load 256bits (32 bytes 8 pix), 32bit aligned
"vld1.32 {d4-d7} , [%[s]]! \n\t" // load 256bits (32 bytes 8 pix), 32bit aligned
"vld1.32 {d8-d11}, [%[s]]! \n\t" // load 256bits (32 bytes 8 pix), 32bit aligned
"vst1.32 {d0-d3}, [%[d]]! \n\t" // store 256bits (32 bytes 8 pix), 32bit aligned
"vst1.32 {d4-d7}, [%[d]]! \n\t" // store 256bits (32 bytes 8 pix), 32bit aligned
"vst1.32 {d8-d11}, [%[d]]! \n\t" // store 256bits (32 bytes 8 pix), 32bit aligned
"cmp %[e], %[d] \n\t" // compare current and end ptr
"bgt _op_copy_p_dp_neon_asmloop \n\t"
: /*out*/
: /*in */ [s] "r" (s), [e] "r" (e), [d] "r" (d)
: /*clobber*/
"q0", "q1", "q2","q3", "q4", "q5", "q6",
"d0", "d1", "d2", "d3",
"d4", "d5", "d6", "d7",
"d8", "d9", "d10", "d11",
"memory" // clobbered
);
dl = l % 24; // dl is how many pixels at end that is not a multiple of 24
l = l - dl; // jump to there at the end of the run?
s = s + l;
d = d + l;
}
e += 23;
for (;d < e; d++, s++) *d = *s;
#endif
}
#define _op_copy_pan_dp_neon _op_copy_p_dp_neon
#define _op_copy_pas_dp_neon _op_copy_p_dp_neon
#define _op_copy_p_dpan_neon _op_copy_p_dp_neon
#define _op_copy_pan_dpan_neon _op_copy_pan_dp_neon
#define _op_copy_pas_dpan_neon _op_copy_pas_dp_neon
static void
init_copy_pixel_span_funcs_neon(void)
{
op_copy_span_funcs[SP][SM_N][SC_N][DP][CPU_NEON] = _op_copy_p_dp_neon;
op_copy_span_funcs[SP_AN][SM_N][SC_N][DP][CPU_NEON] = _op_copy_pan_dp_neon;
op_copy_span_funcs[SP_AS][SM_N][SC_N][DP][CPU_NEON] = _op_copy_pas_dp_neon;
op_copy_span_funcs[SP][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_p_dpan_neon;
op_copy_span_funcs[SP_AN][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_pan_dpan_neon;
op_copy_span_funcs[SP_AS][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_pas_dpan_neon;
}
#endif
#ifdef BUILD_NEON
static void
_op_copy_pt_p_dp_neon(DATA32 s, DATA8 m __UNUSED__, DATA32 c __UNUSED__, DATA32 *d) {
*d = s;
}
#define _op_copy_pt_pan_dp_neon _op_copy_pt_p_dp_neon
#define _op_copy_pt_pas_dp_neon _op_copy_pt_p_dp_neon
#define _op_copy_pt_p_dpan_neon _op_copy_pt_p_dp_neon
#define _op_copy_pt_pan_dpan_neon _op_copy_pt_pan_dp_neon
#define _op_copy_pt_pas_dpan_neon _op_copy_pt_pas_dp_neon
static void
init_copy_pixel_pt_funcs_neon(void)
{
op_copy_pt_funcs[SP][SM_N][SC_N][DP][CPU_NEON] = _op_copy_pt_p_dp_neon;
op_copy_pt_funcs[SP_AN][SM_N][SC_N][DP][CPU_NEON] = _op_copy_pt_pan_dp_neon;
op_copy_pt_funcs[SP_AS][SM_N][SC_N][DP][CPU_NEON] = _op_copy_pt_pas_dp_neon;
op_copy_pt_funcs[SP][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_pt_p_dpan_neon;
op_copy_pt_funcs[SP_AN][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_pt_pan_dpan_neon;
op_copy_pt_funcs[SP_AS][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_pt_pas_dpan_neon;
}
#endif
/*-----*/
/* copy_rel pixel --> dst */
#ifdef BUILD_NEON
static void
_op_copy_rel_p_dp_neon(DATA32 *s, DATA8 *m __UNUSED__, DATA32 c __UNUSED__, DATA32 *d, int l) {
// FIXME: neon-it
DATA32 *e;
UNROLL8_PLD_WHILE(d, l, e,
{
*d = MUL_SYM(*d >> 24, c);
d++;
});
}
#define _op_copy_rel_pas_dp_neon _op_copy_rel_p_dp_neon
#define _op_copy_rel_pan_dp_neon _op_copy_rel_p_dp_neon
#define _op_copy_rel_p_dpan_neon _op_copy_p_dpan_neon
#define _op_copy_rel_pan_dpan_neon _op_copy_pan_dpan_neon
#define _op_copy_rel_pas_dpan_neon _op_copy_pas_dpan_neon
static void
init_copy_rel_pixel_span_funcs_neon(void)
{
op_copy_rel_span_funcs[SP][SM_N][SC_N][DP][CPU_NEON] = _op_copy_rel_p_dp_neon;
op_copy_rel_span_funcs[SP_AN][SM_N][SC_N][DP][CPU_NEON] = _op_copy_rel_pan_dp_neon;
op_copy_rel_span_funcs[SP_AS][SM_N][SC_N][DP][CPU_NEON] = _op_copy_rel_pas_dp_neon;
op_copy_rel_span_funcs[SP][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_rel_p_dpan_neon;
op_copy_rel_span_funcs[SP_AN][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_rel_pan_dpan_neon;
op_copy_rel_span_funcs[SP_AS][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_rel_pas_dpan_neon;
}
#endif
#ifdef BUILD_NEON
static void
_op_copy_rel_pt_p_dp_neon(DATA32 s, DATA8 m __UNUSED__, DATA32 c, DATA32 *d) {
s = 1 + (*d >> 24);
*d = MUL_256(s, c);
}
#define _op_copy_rel_pt_pan_dp_neon _op_copy_rel_pt_p_dp_neon
#define _op_copy_rel_pt_pas_dp_neon _op_copy_rel_pt_p_dp_neon
#define _op_copy_rel_pt_p_dpan_neon _op_copy_pt_p_dpan_neon
#define _op_copy_rel_pt_pan_dpan_neon _op_copy_pt_pan_dpan_neon
#define _op_copy_rel_pt_pas_dpan_neon _op_copy_pt_pas_dpan_neon
static void
init_copy_rel_pixel_pt_funcs_neon(void)
{
op_copy_rel_pt_funcs[SP][SM_N][SC_N][DP][CPU_NEON] = _op_copy_rel_pt_p_dp_neon;
op_copy_rel_pt_funcs[SP_AN][SM_N][SC_N][DP][CPU_NEON] = _op_copy_rel_pt_pan_dp_neon;
op_copy_rel_pt_funcs[SP_AS][SM_N][SC_N][DP][CPU_NEON] = _op_copy_rel_pt_pas_dp_neon;
op_copy_rel_pt_funcs[SP][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_rel_pt_p_dpan_neon;
op_copy_rel_pt_funcs[SP_AN][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_rel_pt_pan_dpan_neon;
op_copy_rel_pt_funcs[SP_AS][SM_N][SC_N][DP_AN][CPU_NEON] = _op_copy_rel_pt_pas_dpan_neon;
}
#endif
|
