1 /* { dg-do compile } */
2 /* { dg-options "-march=rv32gcv -mabi=ilp32 -O3 -fno-tree-vectorize" } */
3
4 #include "riscv_vector.h"
5
6 void fn3 (void);
7
8 void f (int8_t * restrict in, int8_t * restrict out, int n)
9 {
10 for (int i = 0 ; i < n * n * n * n; i++) {
11 vuint8mf8_t v = *(vuint8mf8_t*)(in + 900 + i);
12 *(vuint8mf8_t*)(out + 900 + i) = v;
13 }
14 fn3 ();
15
16 for (int i = 0 ; i < n * n * n * n; i++) {
17 vuint8mf8_t v = *(vuint8mf8_t*)(in + 900 + i);
18 *(vuint8mf8_t*)(out + 900 + i) = v;
19 }
20 fn3 ();
21
22 for (int i = 0 ; i < n * n * n * n; i++) {
23 vuint8mf8_t v = *(vuint8mf8_t*)(in + 900 + i);
24 *(vuint8mf8_t*)(out + 900 + i) = v;
25 }
26 fn3 ();
27 for (int i = 0 ; i < n * n * n * n; i++) {
28 vuint8mf8_t v = *(vuint8mf8_t*)(in + 900 + i);
29 *(vuint8mf8_t*)(out + 900 + i) = v;
30 }
31 fn3 ();
32 for (int i = 0 ; i < n * n * n * n; i++) {
33 vuint8mf8_t v = *(vuint8mf8_t*)(in + 900 + i);
34 *(vuint8mf8_t*)(out + 900 + i) = v;
35 }
36
37 }
38
39 void f2 (int8_t * restrict in, int8_t * restrict out, int n)
40 {
41 for (int i = 0 ; i < n * n * n * n; i++) {
42 vuint8mf4_t v = *(vuint8mf4_t*)(in + 900 + i);
43 *(vuint8mf4_t*)(out + 900 + i) = v;
44 }
45 fn3 ();
46
47 for (int i = 0 ; i < n * n * n * n; i++) {
48 vuint8mf4_t v = *(vuint8mf4_t*)(in + 900 + i);
49 *(vuint8mf4_t*)(out + 900 + i) = v;
50 }
51 fn3 ();
52
53 for (int i = 0 ; i < n * n * n * n; i++) {
54 vuint8mf4_t v = *(vuint8mf4_t*)(in + 900 + i);
55 *(vuint8mf4_t*)(out + 900 + i) = v;
56 }
57 fn3 ();
58 for (int i = 0 ; i < n * n * n * n; i++) {
59 vuint8mf4_t v = *(vuint8mf4_t*)(in + 900 + i);
60 *(vuint8mf4_t*)(out + 900 + i) = v;
61 }
62 fn3 ();
63 for (int i = 0 ; i < n * n * n * n; i++) {
64 vuint8mf4_t v = *(vuint8mf4_t*)(in + 900 + i);
65 *(vuint8mf4_t*)(out + 900 + i) = v;
66 }
67
68 }
69
70 void f3 (int8_t * restrict in, int8_t * restrict out, int n)
71 {
72 for (int i = 0 ; i < n * n * n * n; i++) {
73 vuint8mf2_t v = *(vuint8mf2_t*)(in + 900 + i);
74 *(vuint8mf2_t*)(out + 900 + i) = v;
75 }
76 fn3 ();
77
78 for (int i = 0 ; i < n * n * n * n; i++) {
79 vuint8mf2_t v = *(vuint8mf2_t*)(in + 900 + i);
80 *(vuint8mf2_t*)(out + 900 + i) = v;
81 }
82 fn3 ();
83
84 for (int i = 0 ; i < n * n * n * n; i++) {
85 vuint8mf2_t v = *(vuint8mf2_t*)(in + 900 + i);
86 *(vuint8mf2_t*)(out + 900 + i) = v;
87 }
88 fn3 ();
89 for (int i = 0 ; i < n * n * n * n; i++) {
90 vuint8mf2_t v = *(vuint8mf2_t*)(in + 900 + i);
91 *(vuint8mf2_t*)(out + 900 + i) = v;
92 }
93 fn3 ();
94 for (int i = 0 ; i < n * n * n * n; i++) {
95 vuint8mf2_t v = *(vuint8mf2_t*)(in + 900 + i);
96 *(vuint8mf2_t*)(out + 900 + i) = v;
97 }
98
99 }
100
101 void f4 (int8_t * restrict in, int8_t * restrict out, int n)
102 {
103 for (int i = 0 ; i < n * n * n * n; i++) {
104 vuint16mf4_t v = *(vuint16mf4_t*)(in + 900 + i);
105 *(vuint16mf4_t*)(out + 900 + i) = v;
106 }
107 fn3 ();
108
109 for (int i = 0 ; i < n * n * n * n; i++) {
110 vuint16mf4_t v = *(vuint16mf4_t*)(in + 900 + i);
111 *(vuint16mf4_t*)(out + 900 + i) = v;
112 }
113 fn3 ();
114
115 for (int i = 0 ; i < n * n * n * n; i++) {
116 vuint16mf4_t v = *(vuint16mf4_t*)(in + 900 + i);
117 *(vuint16mf4_t*)(out + 900 + i) = v;
118 }
119 fn3 ();
120 for (int i = 0 ; i < n * n * n * n; i++) {
121 vuint16mf4_t v = *(vuint16mf4_t*)(in + 900 + i);
122 *(vuint16mf4_t*)(out + 900 + i) = v;
123 }
124 fn3 ();
125 for (int i = 0 ; i < n * n * n * n; i++) {
126 vuint16mf4_t v = *(vuint16mf4_t*)(in + 900 + i);
127 *(vuint16mf4_t*)(out + 900 + i) = v;
128 }
129
130 }
131
132 void f5 (int8_t * restrict in, int8_t * restrict out, int n)
133 {
134 for (int i = 0 ; i < n * n * n * n; i++) {
135 vuint16mf2_t v = *(vuint16mf2_t*)(in + 900 + i);
136 *(vuint16mf2_t*)(out + 900 + i) = v;
137 }
138 fn3 ();
139
140 for (int i = 0 ; i < n * n * n * n; i++) {
141 vuint16mf2_t v = *(vuint16mf2_t*)(in + 900 + i);
142 *(vuint16mf2_t*)(out + 900 + i) = v;
143 }
144 fn3 ();
145
146 for (int i = 0 ; i < n * n * n * n; i++) {
147 vuint16mf2_t v = *(vuint16mf2_t*)(in + 900 + i);
148 *(vuint16mf2_t*)(out + 900 + i) = v;
149 }
150 fn3 ();
151 for (int i = 0 ; i < n * n * n * n; i++) {
152 vuint16mf2_t v = *(vuint16mf2_t*)(in + 900 + i);
153 *(vuint16mf2_t*)(out + 900 + i) = v;
154 }
155 fn3 ();
156 for (int i = 0 ; i < n * n * n * n; i++) {
157 vuint16mf2_t v = *(vuint16mf2_t*)(in + 900 + i);
158 *(vuint16mf2_t*)(out + 900 + i) = v;
159 }
160
161 }
162
163 void f6 (int8_t * restrict in, int8_t * restrict out, int n)
164 {
165 for (int i = 0 ; i < n * n * n * n; i++) {
166 vuint32mf2_t v = *(vuint32mf2_t*)(in + 900 + i);
167 *(vuint32mf2_t*)(out + 900 + i) = v;
168 }
169 fn3 ();
170
171 for (int i = 0 ; i < n * n * n * n; i++) {
172 vuint32mf2_t v = *(vuint32mf2_t*)(in + 900 + i);
173 *(vuint32mf2_t*)(out + 900 + i) = v;
174 }
175 fn3 ();
176
177 for (int i = 0 ; i < n * n * n * n; i++) {
178 vuint32mf2_t v = *(vuint32mf2_t*)(in + 900 + i);
179 *(vuint32mf2_t*)(out + 900 + i) = v;
180 }
181 fn3 ();
182 for (int i = 0 ; i < n * n * n * n; i++) {
183 vuint32mf2_t v = *(vuint32mf2_t*)(in + 900 + i);
184 *(vuint32mf2_t*)(out + 900 + i) = v;
185 }
186 fn3 ();
187 for (int i = 0 ; i < n * n * n * n; i++) {
188 vuint32mf2_t v = *(vuint32mf2_t*)(in + 900 + i);
189 *(vuint32mf2_t*)(out + 900 + i) = v;
190 }
191
192 }
193
194 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e8,\s*mf8,\s*t[au],\s*m[au]} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
195 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e8,\s*mf8,\s*t[au],\s*m[au]\s+\.L[0-9]\:\s+vle8\.v\s+(?:v[0-9]|v[1-2][0-9]|v3[0-1]),0\s*\([a-x0-9]+\)} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
196 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e8,\s*mf4,\s*t[au],\s*m[au]} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
197 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e8,\s*mf4,\s*t[au],\s*m[au]\s+\.L[0-9][0-9]\:\s+vle8\.v\s+(?:v[0-9]|v[1-2][0-9]|v3[0-1]),0\s*\([a-x0-9]+\)} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
198 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e8,\s*mf2,\s*t[au],\s*m[au]} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
199 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e8,\s*mf2,\s*t[au],\s*m[au]\s+\.L[0-9][0-9]\:\s+vle8\.v\s+(?:v[0-9]|v[1-2][0-9]|v3[0-1]),0\s*\([a-x0-9]+\)} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
200
201 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e16,\s*mf4,\s*t[au],\s*m[au]} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
202 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e16,\s*mf4,\s*t[au],\s*m[au]\s+\.L[0-9][0-9]\:\s+vle16\.v\s+(?:v[0-9]|v[1-2][0-9]|v3[0-1]),0\s*\([a-x0-9]+\)} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
203 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e16,\s*mf2,\s*t[au],\s*m[au]} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
204 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e16,\s*mf2,\s*t[au],\s*m[au]\s+\.L[0-9][0-9]\:\s+vle16\.v\s+(?:v[0-9]|v[1-2][0-9]|v3[0-1]),0\s*\([a-x0-9]+\)} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
205
206 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e32,\s*mf2,\s*t[au],\s*m[au]} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
207 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e32,\s*mf2,\s*t[au],\s*m[au]\s+\.L[0-9][0-9]\:\s+vle32\.v\s+(?:v[0-9]|v[1-2][0-9]|v3[0-1]),0\s*\([a-x0-9]+\)} 5 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */