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
|
/*
* Header file for wrappers around MSA instructions assembler invocations
*
* Copyright (C) 2018 Wave Computing, Inc.
* Copyright (C) 2018 Aleksandar Markovic <amarkovic@wavecomp.com>
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <https://www.gnu.org/licenses/>.
*
*/
#ifndef WRAPPERS_MSA_H
#define WRAPPERS_MSA_H
#define DO_MSA__WD__WS(suffix, mnemonic) \
static inline void do_msa_##suffix(void *input, void *output) \
{ \
__asm__ volatile ( \
"move $t0, %0\n\t" \
"ld.d $w11, 0($t0)\n\t" \
#mnemonic " $w10, $w11\n\t" \
"move $t0, %1\n\t" \
"st.d $w10, 0($t0)\n\t" \
: \
: "r" (input), "r" (output) \
: "t0", "memory" \
); \
}
DO_MSA__WD__WS(NLOC_B, nloc.b)
DO_MSA__WD__WS(NLOC_H, nloc.h)
DO_MSA__WD__WS(NLOC_W, nloc.w)
DO_MSA__WD__WS(NLOC_D, nloc.d)
DO_MSA__WD__WS(NLZC_B, nlzc.b)
DO_MSA__WD__WS(NLZC_H, nlzc.h)
DO_MSA__WD__WS(NLZC_W, nlzc.w)
DO_MSA__WD__WS(NLZC_D, nlzc.d)
DO_MSA__WD__WS(PCNT_B, pcnt.b)
DO_MSA__WD__WS(PCNT_H, pcnt.h)
DO_MSA__WD__WS(PCNT_W, pcnt.w)
DO_MSA__WD__WS(PCNT_D, pcnt.d)
#define DO_MSA__WD__WS_WT(suffix, mnemonic) \
static inline void do_msa_##suffix(void *input1, void *input2, \
void *output) \
{ \
__asm__ volatile ( \
"move $t0, %0\n\t" \
"ld.d $w11, 0($t0)\n\t" \
"move $t0, %1\n\t" \
"ld.d $w12, 0($t0)\n\t" \
#mnemonic " $w10, $w11, $w12\n\t" \
"move $t0, %2\n\t" \
"st.d $w10, 0($t0)\n\t" \
: \
: "r" (input1), "r" (input2), "r" (output) \
: "t0", "memory" \
); \
}
DO_MSA__WD__WS_WT(ILVEV_B, ilvev.b)
DO_MSA__WD__WS_WT(ILVEV_H, ilvev.h)
DO_MSA__WD__WS_WT(ILVEV_W, ilvev.w)
DO_MSA__WD__WS_WT(ILVEV_D, ilvev.d)
DO_MSA__WD__WS_WT(ILVOD_B, ilvod.b)
DO_MSA__WD__WS_WT(ILVOD_H, ilvod.h)
DO_MSA__WD__WS_WT(ILVOD_W, ilvod.w)
DO_MSA__WD__WS_WT(ILVOD_D, ilvod.d)
DO_MSA__WD__WS_WT(ILVL_B, ilvl.b)
DO_MSA__WD__WS_WT(ILVL_H, ilvl.h)
DO_MSA__WD__WS_WT(ILVL_W, ilvl.w)
DO_MSA__WD__WS_WT(ILVL_D, ilvl.d)
DO_MSA__WD__WS_WT(ILVR_B, ilvr.b)
DO_MSA__WD__WS_WT(ILVR_H, ilvr.h)
DO_MSA__WD__WS_WT(ILVR_W, ilvr.w)
DO_MSA__WD__WS_WT(ILVR_D, ilvr.d)
DO_MSA__WD__WS_WT(AND_V, and.v)
DO_MSA__WD__WS_WT(NOR_V, nor.v)
DO_MSA__WD__WS_WT(OR_V, or.v)
DO_MSA__WD__WS_WT(XOR_V, xor.v)
#endif
|
