| 1 |
/* { dg-do compile } */ |
| 2 |
/* { dg-options "-mrvv-vector-bits=scalable -march=rv32gcv -mabi=ilp32 -fno-schedule-insns -fno-schedule-insns2" } */ |
| 3 |
|
| 4 |
#include "riscv_vector.h" |
| 5 |
|
| 6 |
void f1 (void * restrict in, void * restrict out, int l, int n, int m) |
| 7 |
{ |
| 8 |
for (int i = 0; i < l; i++){ |
| 9 |
for (int j = 0; j < m; j++){ |
| 10 |
for (int k = 0; k < n; k++) |
| 11 |
{ |
| 12 |
vint8mf8_t v = __riscv_vle8_v_i8mf8 (in + i + j, 17); |
| 13 |
__riscv_vse8_v_i8mf8 (out + i + j, v, 17); |
| 14 |
} |
| 15 |
} |
| 16 |
} |
| 17 |
} |
| 18 |
|
| 19 |
void f2 (void * restrict in, void * restrict out, int l, int n, int m) |
| 20 |
{ |
| 21 |
for (int i = 0; i < l; i++){ |
| 22 |
for (int j = 0; j < m; j++){ |
| 23 |
for (int k = 0; k < n; k++) |
| 24 |
{ |
| 25 |
vuint8mf8_t v = __riscv_vle8_v_u8mf8 (in + i + j, 17); |
| 26 |
__riscv_vse8_v_u8mf8 (out + i + j, v, 17); |
| 27 |
} |
| 28 |
} |
| 29 |
} |
| 30 |
} |
| 31 |
|
| 32 |
void f3 (void * restrict in, void * restrict out, int l, int n, int m) |
| 33 |
{ |
| 34 |
for (int i = 0; i < l; i++){ |
| 35 |
for (int j = 0; j < m; j++){ |
| 36 |
for (int k = 0; k < n; k++) |
| 37 |
{ |
| 38 |
vint8mf4_t v = __riscv_vle8_v_i8mf4 (in + i + j, 17); |
| 39 |
__riscv_vse8_v_i8mf4 (out + i + j, v, 17); |
| 40 |
} |
| 41 |
} |
| 42 |
} |
| 43 |
} |
| 44 |
|
| 45 |
void f4 (void * restrict in, void * restrict out, int l, int n, int m) |
| 46 |
{ |
| 47 |
for (int i = 0; i < l; i++){ |
| 48 |
for (int j = 0; j < m; j++){ |
| 49 |
for (int k = 0; k < n; k++) |
| 50 |
{ |
| 51 |
vuint8mf4_t v = __riscv_vle8_v_u8mf4 (in + i + j, 17); |
| 52 |
__riscv_vse8_v_u8mf4 (out + i + j, v, 17); |
| 53 |
} |
| 54 |
} |
| 55 |
} |
| 56 |
} |
| 57 |
|
| 58 |
void f5 (void * restrict in, void * restrict out, int l, int n, int m) |
| 59 |
{ |
| 60 |
for (int i = 0; i < l; i++){ |
| 61 |
for (int j = 0; j < m; j++){ |
| 62 |
for (int k = 0; k < n; k++) |
| 63 |
{ |
| 64 |
vint8mf2_t v = __riscv_vle8_v_i8mf2 (in + i + j, 17); |
| 65 |
__riscv_vse8_v_i8mf2 (out + i + j, v, 17); |
| 66 |
} |
| 67 |
} |
| 68 |
} |
| 69 |
} |
| 70 |
|
| 71 |
void f6 (void * restrict in, void * restrict out, int l, int n, int m) |
| 72 |
{ |
| 73 |
for (int i = 0; i < l; i++){ |
| 74 |
for (int j = 0; j < m; j++){ |
| 75 |
for (int k = 0; k < n; k++) |
| 76 |
{ |
| 77 |
vuint8mf2_t v = __riscv_vle8_v_u8mf2 (in + i + j, 17); |
| 78 |
__riscv_vse8_v_u8mf2 (out + i + j, v, 17); |
| 79 |
} |
| 80 |
} |
| 81 |
} |
| 82 |
} |
| 83 |
|
| 84 |
void f7 (void * restrict in, void * restrict out, int l, int n, int m) |
| 85 |
{ |
| 86 |
for (int i = 0; i < l; i++){ |
| 87 |
for (int j = 0; j < m; j++){ |
| 88 |
for (int k = 0; k < n; k++) |
| 89 |
{ |
| 90 |
vint8m1_t v = __riscv_vle8_v_i8m1 (in + i + j, 17); |
| 91 |
__riscv_vse8_v_i8m1 (out + i + j, v, 17); |
| 92 |
} |
| 93 |
} |
| 94 |
} |
| 95 |
} |
| 96 |
|
| 97 |
void f8 (void * restrict in, void * restrict out, int l, int n, int m) |
| 98 |
{ |
| 99 |
for (int i = 0; i < l; i++){ |
| 100 |
for (int j = 0; j < m; j++){ |
| 101 |
for (int k = 0; k < n; k++) |
| 102 |
{ |
| 103 |
vuint8m1_t v = __riscv_vle8_v_u8m1 (in + i + j, 17); |
| 104 |
__riscv_vse8_v_u8m1 (out + i + j, v, 17); |
| 105 |
} |
| 106 |
} |
| 107 |
} |
| 108 |
} |
| 109 |
|
| 110 |
void f9 (void * restrict in, void * restrict out, int l, int n, int m) |
| 111 |
{ |
| 112 |
for (int i = 0; i < l; i++){ |
| 113 |
for (int j = 0; j < m; j++){ |
| 114 |
for (int k = 0; k < n; k++) |
| 115 |
{ |
| 116 |
vint8m2_t v = __riscv_vle8_v_i8m2 (in + i + j, 17); |
| 117 |
__riscv_vse8_v_i8m2 (out + i + j, v, 17); |
| 118 |
} |
| 119 |
} |
| 120 |
} |
| 121 |
} |
| 122 |
|
| 123 |
void f10 (void * restrict in, void * restrict out, int l, int n, int m) |
| 124 |
{ |
| 125 |
for (int i = 0; i < l; i++){ |
| 126 |
for (int j = 0; j < m; j++){ |
| 127 |
for (int k = 0; k < n; k++) |
| 128 |
{ |
| 129 |
vuint8m2_t v = __riscv_vle8_v_u8m2 (in + i + j, 17); |
| 130 |
__riscv_vse8_v_u8m2 (out + i + j, v, 17); |
| 131 |
} |
| 132 |
} |
| 133 |
} |
| 134 |
} |
| 135 |
|
| 136 |
void f11 (void * restrict in, void * restrict out, int l, int n, int m) |
| 137 |
{ |
| 138 |
for (int i = 0; i < l; i++){ |
| 139 |
for (int j = 0; j < m; j++){ |
| 140 |
for (int k = 0; k < n; k++) |
| 141 |
{ |
| 142 |
vint8m4_t v = __riscv_vle8_v_i8m4 (in + i + j, 17); |
| 143 |
__riscv_vse8_v_i8m4 (out + i + j, v, 17); |
| 144 |
} |
| 145 |
} |
| 146 |
} |
| 147 |
} |
| 148 |
|
| 149 |
void f12 (void * restrict in, void * restrict out, int l, int n, int m) |
| 150 |
{ |
| 151 |
for (int i = 0; i < l; i++){ |
| 152 |
for (int j = 0; j < m; j++){ |
| 153 |
for (int k = 0; k < n; k++) |
| 154 |
{ |
| 155 |
vuint8m4_t v = __riscv_vle8_v_u8m4 (in + i + j, 17); |
| 156 |
__riscv_vse8_v_u8m4 (out + i + j, v, 17); |
| 157 |
} |
| 158 |
} |
| 159 |
} |
| 160 |
} |
| 161 |
|
| 162 |
void f13 (void * restrict in, void * restrict out, int l, int n, int m) |
| 163 |
{ |
| 164 |
for (int i = 0; i < l; i++){ |
| 165 |
for (int j = 0; j < m; j++){ |
| 166 |
for (int k = 0; k < n; k++) |
| 167 |
{ |
| 168 |
vint8m8_t v = __riscv_vle8_v_i8m8 (in + i + j, 17); |
| 169 |
__riscv_vse8_v_i8m8 (out + i + j, v, 17); |
| 170 |
} |
| 171 |
} |
| 172 |
} |
| 173 |
} |
| 174 |
|
| 175 |
void f14 (void * restrict in, void * restrict out, int l, int n, int m) |
| 176 |
{ |
| 177 |
for (int i = 0; i < l; i++){ |
| 178 |
for (int j = 0; j < m; j++){ |
| 179 |
for (int k = 0; k < n; k++) |
| 180 |
{ |
| 181 |
vuint8m8_t v = __riscv_vle8_v_u8m8 (in + i + j, 17); |
| 182 |
__riscv_vse8_v_u8m8 (out + i + j, v, 17); |
| 183 |
} |
| 184 |
} |
| 185 |
} |
| 186 |
} |
| 187 |
|
| 188 |
/* { dg-final { scan-assembler-times {\.L[0-9]+\:\s+vle8\.v\s+v[0-9]+,\s*0\s*\([a-x0-9]+\)} 14 { target { no-opts "-O0" no-opts "-g" no-opts "-funroll-loops" } } } } */ |
| 189 |
/* { dg-final { scan-assembler-times {vsetivli\s+zero,\s*17,\s*e8,\s*mf8,\s*t[au],\s*m[au]} 2 { target { no-opts "-O0" no-opts "-g" no-opts "-funroll-loops" } } } } */ |
| 190 |
/* { dg-final { scan-assembler-times {vsetivli\s+zero,\s*17,\s*e8,\s*mf4,\s*t[au],\s*m[au]} 2 { target { no-opts "-O0" no-opts "-g" no-opts "-funroll-loops" } } } } */ |
| 191 |
/* { dg-final { scan-assembler-times {vsetivli\s+zero,\s*17,\s*e8,\s*mf2,\s*t[au],\s*m[au]} 2 { target { no-opts "-O0" no-opts "-g" no-opts "-funroll-loops" } } } } */ |
| 192 |
/* { dg-final { scan-assembler-times {vsetivli\s+zero,\s*17,\s*e8,\s*m1,\s*t[au],\s*m[au]} 2 { target { no-opts "-O0" no-opts "-g" no-opts "-funroll-loops" } } } } */ |
| 193 |
/* { dg-final { scan-assembler-times {vsetivli\s+zero,\s*17,\s*e8,\s*m2,\s*t[au],\s*m[au]} 2 { target { no-opts "-O0" no-opts "-g" no-opts "-funroll-loops" } } } } */ |
| 194 |
/* { dg-final { scan-assembler-times {vsetivli\s+zero,\s*17,\s*e8,\s*m4,\s*t[au],\s*m[au]} 2 { target { no-opts "-O0" no-opts "-g" no-opts "-funroll-loops" } } } } */ |
| 195 |
/* { dg-final { scan-assembler-times {vsetivli\s+zero,\s*17,\s*e8,\s*m8,\s*t[au],\s*m[au]} 2 { target { no-opts "-O0" no-opts "-g" no-opts "-funroll-loops" } } } } */ |