1 /* { dg-do compile } */
2 /* { dg-options "-march=rv32gcv -mabi=ilp32 -fno-tree-vectorize" } */
3
4 #include "riscv_vector.h"
5
6 void f (int8_t * restrict in, int8_t * restrict out, int n)
7 {
8 switch (n)
9 {
10 case 0:{
11 vint8mf8_t v = *(vint8mf8_t*)(in + 500);
12 *(vint8mf8_t*)(out + 500) = v;
13 break;
14 }
15 case 1:{
16 vint8mf8_t v = *(vint8mf8_t*)(in + 500);
17 *(vint8mf8_t*)(out + 500) = v;
18 break;
19 }
20 default:{
21 break;
22 }
23 }
24
25 for (int i = 0 ; i < n * n; i++)
26 out[i] = out[i] + out[i];
27
28 for (int i = 0 ; i < n * n * n; i++)
29 out[i] = out[i] * out[i];
30
31 for (int i = 0 ; i < n * n * n * n; i++)
32 out[i] = out[i] * out[i];
33
34 for (int i = 0 ; i < n * n * n * n; i++) {
35 vint8mf8_t v = *(vint8mf8_t*)(in + 900 + i);
36 *(vint8mf8_t*)(out + 900 + i) = v;
37 }
38 }
39
40 void f2 (int8_t * restrict in, int8_t * restrict out, int n)
41 {
42 switch (n)
43 {
44 case 0:{
45 vint8mf4_t v = *(vint8mf4_t*)(in + 500);
46 *(vint8mf4_t*)(out + 500) = v;
47 break;
48 }
49 case 1:{
50 vint8mf4_t v = *(vint8mf4_t*)(in + 500);
51 *(vint8mf4_t*)(out + 500) = v;
52 break;
53 }
54 default:{
55 break;
56 }
57 }
58
59 for (int i = 0 ; i < n * n; i++)
60 out[i] = out[i] + out[i];
61
62 for (int i = 0 ; i < n * n * n; i++)
63 out[i] = out[i] * out[i];
64
65 for (int i = 0 ; i < n * n * n * n; i++)
66 out[i] = out[i] * out[i];
67
68 for (int i = 0 ; i < n * n * n * n; i++) {
69 vint8mf4_t v = *(vint8mf4_t*)(in + 900 + i);
70 *(vint8mf4_t*)(out + 900 + i) = v;
71 }
72 }
73
74 void f3 (int8_t * restrict in, int8_t * restrict out, int n)
75 {
76 switch (n)
77 {
78 case 0:{
79 vint8mf2_t v = *(vint8mf2_t*)(in + 500);
80 *(vint8mf2_t*)(out + 500) = v;
81 break;
82 }
83 case 1:{
84 vint8mf2_t v = *(vint8mf2_t*)(in + 500);
85 *(vint8mf2_t*)(out + 500) = v;
86 break;
87 }
88 default:{
89 break;
90 }
91 }
92
93 for (int i = 0 ; i < n * n; i++)
94 out[i] = out[i] + out[i];
95
96 for (int i = 0 ; i < n * n * n; i++)
97 out[i] = out[i] * out[i];
98
99 for (int i = 0 ; i < n * n * n * n; i++)
100 out[i] = out[i] * out[i];
101
102 for (int i = 0 ; i < n * n * n * n; i++) {
103 vint8mf2_t v = *(vint8mf2_t*)(in + 900 + i);
104 *(vint8mf2_t*)(out + 900 + i) = v;
105 }
106 }
107
108 void f4 (int8_t * restrict in, int8_t * restrict out, int n)
109 {
110 switch (n)
111 {
112 case 0:{
113 vint16mf4_t v = *(vint16mf4_t*)(in + 500);
114 *(vint16mf4_t*)(out + 500) = v;
115 break;
116 }
117 case 1:{
118 vint16mf4_t v = *(vint16mf4_t*)(in + 500);
119 *(vint16mf4_t*)(out + 500) = v;
120 break;
121 }
122 default:{
123 break;
124 }
125 }
126
127 for (int i = 0 ; i < n * n; i++)
128 out[i] = out[i] + out[i];
129
130 for (int i = 0 ; i < n * n * n; i++)
131 out[i] = out[i] * out[i];
132
133 for (int i = 0 ; i < n * n * n * n; i++)
134 out[i] = out[i] * out[i];
135
136 for (int i = 0 ; i < n * n * n * n; i++) {
137 vint16mf4_t v = *(vint16mf4_t*)(in + 900 + i);
138 *(vint16mf4_t*)(out + 900 + i) = v;
139 }
140 }
141
142 void f5 (int8_t * restrict in, int8_t * restrict out, int n)
143 {
144 switch (n)
145 {
146 case 0:{
147 vint16mf2_t v = *(vint16mf2_t*)(in + 500);
148 *(vint16mf2_t*)(out + 500) = v;
149 break;
150 }
151 case 1:{
152 vint16mf2_t v = *(vint16mf2_t*)(in + 500);
153 *(vint16mf2_t*)(out + 500) = v;
154 break;
155 }
156 default:{
157 break;
158 }
159 }
160
161 for (int i = 0 ; i < n * n; i++)
162 out[i] = out[i] + out[i];
163
164 for (int i = 0 ; i < n * n * n; i++)
165 out[i] = out[i] * out[i];
166
167 for (int i = 0 ; i < n * n * n * n; i++)
168 out[i] = out[i] * out[i];
169
170 for (int i = 0 ; i < n * n * n * n; i++) {
171 vint16mf2_t v = *(vint16mf2_t*)(in + 900 + i);
172 *(vint16mf2_t*)(out + 900 + i) = v;
173 }
174 }
175
176 void f6 (int8_t * restrict in, int8_t * restrict out, int n)
177 {
178 switch (n)
179 {
180 case 0:{
181 vint32mf2_t v = *(vint32mf2_t*)(in + 500);
182 *(vint32mf2_t*)(out + 500) = v;
183 break;
184 }
185 case 1:{
186 vint32mf2_t v = *(vint32mf2_t*)(in + 500);
187 *(vint32mf2_t*)(out + 500) = v;
188 break;
189 }
190 default:{
191 break;
192 }
193 }
194
195 for (int i = 0 ; i < n * n; i++)
196 out[i] = out[i] + out[i];
197
198 for (int i = 0 ; i < n * n * n; i++)
199 out[i] = out[i] * out[i];
200
201 for (int i = 0 ; i < n * n * n * n; i++)
202 out[i] = out[i] * out[i];
203
204 for (int i = 0 ; i < n * n * n * n; i++) {
205 vint32mf2_t v = *(vint32mf2_t*)(in + 900 + i);
206 *(vint32mf2_t*)(out + 900 + i) = v;
207 }
208 }
209
210 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e8,\s*mf8,\s*t[au],\s*m[au]} 3 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
211 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e8,\s*mf4,\s*t[au],\s*m[au]} 3 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
212 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e8,\s*mf2,\s*t[au],\s*m[au]} 3 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
213 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e16,\s*mf4,\s*t[au],\s*m[au]} 3 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
214 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e16,\s*mf2,\s*t[au],\s*m[au]} 3 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */
215 /* { dg-final { scan-assembler-times {vsetvli\s+[a-x0-9]+,\s*zero,\s*e32,\s*mf2,\s*t[au],\s*m[au]} 3 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" } } } } */