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 vint8mf8_t v = *(vint8mf8_t*)(in + 700);
22 *(vint8mf8_t*)(out + 700) = v;
23 break;
24 }
25 }
26
27 for (int i = 0 ; i < n * n; i++)
28 out[i] = out[i] + out[i];
29
30 for (int i = 0 ; i < n * n * n; i++)
31 out[i] = out[i] * out[i];
32
33 for (int i = 0 ; i < n * n * n * n; i++)
34 out[i] = out[i] * out[i];
35
36 for (int i = 0 ; i < n * n * n * n; i++) {
37 vint8mf8_t v = *(vint8mf8_t*)(in + 900 + i);
38 *(vint8mf8_t*)(out + 900 + i) = v;
39 }
40 }
41
42 void f2 (int8_t * restrict in, int8_t * restrict out, int n)
43 {
44 switch (n)
45 {
46 case 0:{
47 vint8mf4_t v = *(vint8mf4_t*)(in + 500);
48 *(vint8mf4_t*)(out + 500) = v;
49 break;
50 }
51 case 1:{
52 vint8mf4_t v = *(vint8mf4_t*)(in + 500);
53 *(vint8mf4_t*)(out + 500) = v;
54 break;
55 }
56 default:{
57 vint8mf4_t v = *(vint8mf4_t*)(in + 700);
58 *(vint8mf4_t*)(out + 700) = v;
59 break;
60 }
61 }
62
63 for (int i = 0 ; i < n * n; i++)
64 out[i] = out[i] + out[i];
65
66 for (int i = 0 ; i < n * n * n; i++)
67 out[i] = out[i] * out[i];
68
69 for (int i = 0 ; i < n * n * n * n; i++)
70 out[i] = out[i] * out[i];
71
72 for (int i = 0 ; i < n * n * n * n; i++) {
73 vint8mf4_t v = *(vint8mf4_t*)(in + 900 + i);
74 *(vint8mf4_t*)(out + 900 + i) = v;
75 }
76 }
77
78 void f3 (int8_t * restrict in, int8_t * restrict out, int n)
79 {
80 switch (n)
81 {
82 case 0:{
83 vint8mf2_t v = *(vint8mf2_t*)(in + 500);
84 *(vint8mf2_t*)(out + 500) = v;
85 break;
86 }
87 case 1:{
88 vint8mf2_t v = *(vint8mf2_t*)(in + 500);
89 *(vint8mf2_t*)(out + 500) = v;
90 break;
91 }
92 default:{
93 vint8mf2_t v = *(vint8mf2_t*)(in + 700);
94 *(vint8mf2_t*)(out + 700) = v;
95 break;
96 }
97 }
98
99 for (int i = 0 ; i < n * n; i++)
100 out[i] = out[i] + out[i];
101
102 for (int i = 0 ; i < n * n * n; i++)
103 out[i] = out[i] * out[i];
104
105 for (int i = 0 ; i < n * n * n * n; i++)
106 out[i] = out[i] * out[i];
107
108 for (int i = 0 ; i < n * n * n * n; i++) {
109 vint8mf2_t v = *(vint8mf2_t*)(in + 900 + i);
110 *(vint8mf2_t*)(out + 900 + i) = v;
111 }
112 }
113
114 void f4 (int8_t * restrict in, int8_t * restrict out, int n)
115 {
116 switch (n)
117 {
118 case 0:{
119 vint16mf4_t v = *(vint16mf4_t*)(in + 500);
120 *(vint16mf4_t*)(out + 500) = v;
121 break;
122 }
123 case 1:{
124 vint16mf4_t v = *(vint16mf4_t*)(in + 500);
125 *(vint16mf4_t*)(out + 500) = v;
126 break;
127 }
128 default:{
129 vint16mf4_t v = *(vint16mf4_t*)(in + 700);
130 *(vint16mf4_t*)(out + 700) = v;
131 break;
132 }
133 }
134
135 for (int i = 0 ; i < n * n; i++)
136 out[i] = out[i] + out[i];
137
138 for (int i = 0 ; i < n * n * n; i++)
139 out[i] = out[i] * out[i];
140
141 for (int i = 0 ; i < n * n * n * n; i++)
142 out[i] = out[i] * out[i];
143
144 for (int i = 0 ; i < n * n * n * n; i++) {
145 vint16mf4_t v = *(vint16mf4_t*)(in + 900 + i);
146 *(vint16mf4_t*)(out + 900 + i) = v;
147 }
148 }
149
150 void f5 (int8_t * restrict in, int8_t * restrict out, int n)
151 {
152 switch (n)
153 {
154 case 0:{
155 vint16mf2_t v = *(vint16mf2_t*)(in + 500);
156 *(vint16mf2_t*)(out + 500) = v;
157 break;
158 }
159 case 1:{
160 vint16mf2_t v = *(vint16mf2_t*)(in + 500);
161 *(vint16mf2_t*)(out + 500) = v;
162 break;
163 }
164 default:{
165 vint16mf2_t v = *(vint16mf2_t*)(in + 700);
166 *(vint16mf2_t*)(out + 700) = v;
167 break;
168 }
169 }
170
171 for (int i = 0 ; i < n * n; i++)
172 out[i] = out[i] + out[i];
173
174 for (int i = 0 ; i < n * n * n; i++)
175 out[i] = out[i] * out[i];
176
177 for (int i = 0 ; i < n * n * n * n; i++)
178 out[i] = out[i] * out[i];
179
180 for (int i = 0 ; i < n * n * n * n; i++) {
181 vint16mf2_t v = *(vint16mf2_t*)(in + 900 + i);
182 *(vint16mf2_t*)(out + 900 + i) = v;
183 }
184 }
185
186 void f6 (int8_t * restrict in, int8_t * restrict out, int n)
187 {
188 switch (n)
189 {
190 case 0:{
191 vint32mf2_t v = *(vint32mf2_t*)(in + 500);
192 *(vint32mf2_t*)(out + 500) = v;
193 break;
194 }
195 case 1:{
196 vint32mf2_t v = *(vint32mf2_t*)(in + 500);
197 *(vint32mf2_t*)(out + 500) = v;
198 break;
199 }
200 default:{
201 vint32mf2_t v = *(vint32mf2_t*)(in + 700);
202 *(vint32mf2_t*)(out + 700) = v;
203 break;
204 }
205 }
206
207 for (int i = 0 ; i < n * n; i++)
208 out[i] = out[i] + out[i];
209
210 for (int i = 0 ; i < n * n * n; i++)
211 out[i] = out[i] * out[i];
212
213 for (int i = 0 ; i < n * n * n * n; i++)
214 out[i] = out[i] * out[i];
215
216 for (int i = 0 ; i < n * n * n * n; i++) {
217 vint32mf2_t v = *(vint32mf2_t*)(in + 900 + i);
218 *(vint32mf2_t*)(out + 900 + i) = v;
219 }
220 }
221
222 /* { 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" no-opts "-O1" } } } } */
223 /* { 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" no-opts "-O1" } } } } */
224 /* { 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" no-opts "-O1" } } } } */
225 /* { dg-final { scan-assembler-times {add\ta[0-7],a[0-7],a[0-7]\s+\.L[0-9][0-9]\:\s+vle8\.v\s+(?:v[0-9]|v[1-2][0-9]|v3[0-1]),0\s*\([a-x0-9]+\)} 3 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" no-opts "-O1" } } } } */
226 /* { 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" no-opts "-O1" } } } } */
227 /* { 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" no-opts "-O1" } } } } */
228 /* { dg-final { scan-assembler-times {add\ta[0-7],a[0-7],a[0-7]\s+\.L[0-9][0-9]\:\s+vle16\.v\s+(?:v[0-9]|v[1-2][0-9]|v3[0-1]),0\s*\([a-x0-9]+\)} 2 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" no-opts "-O1" } } } } */
229 /* { 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" no-opts "-O1" } } } } */
230 /* { dg-final { scan-assembler-times {add\ta[0-7],a[0-7],a[0-7]\s+\.L[0-9][0-9][0-9]\:\s+vle32\.v\s+(?:v[0-9]|v[1-2][0-9]|v3[0-1]),0\s*\([a-x0-9]+\)} 1 { target { no-opts "-O0" no-opts "-Os" no-opts "-Oz" no-opts "-funroll-loops" no-opts "-g" no-opts "-O1" } } } } */