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
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
|
/* { dg-do compile } */
/* { dg-options "-march=rv32gcv -mabi=ilp32d -O3" } */
#include "riscv_vector.h"
void f0 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16mf4_t src = __riscv_vle16_v_u16mf4 (base, vl);
vuint8mf8_t v = __riscv_vnclipu_wx_u8mf8(src,31,vl);
__riscv_vse8_v_u8mf8 (out,v,vl);
}
void f1 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16mf4_t src = __riscv_vle16_v_u16mf4 (base, vl);
vuint8mf8_t src2 = __riscv_vle8_v_u8mf8 ((int8_t *)(base + 100), vl);
vuint8mf8_t v = __riscv_vnclipu_wx_u8mf8_tu(src2,src,31,vl);
__riscv_vse8_v_u8mf8 (out,v,vl);
}
void f2 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16mf4_t src = __riscv_vle16_v_u16mf4 (base, vl);
vuint8mf8_t v = __riscv_vnclipu_wx_u8mf8(src,31,vl);
vuint16mf4_t v2 = __riscv_vadd_vv_u16mf4 (src, src,vl);
__riscv_vse8_v_u8mf8 (out,v,vl);
__riscv_vse16_v_u16mf4 ((int16_t *)out,v2,vl);
}
void f3 (int16_t *base,int8_t *out,size_t vl, int n, size_t shift)
{
for (int i = 0; i < n; i++){
vuint16mf4_t src = __riscv_vle16_v_u16mf4 (base + 100*i, vl);
vuint8mf8_t v = __riscv_vnclipu_wx_u8mf8(src,31,vl);
vuint16mf4_t v2 = __riscv_vadd_vv_u16mf4 (src, src,vl);
__riscv_vse8_v_u8mf8 (out + 100*i,v,vl);
__riscv_vse16_v_u16mf4 ((int16_t *)(out + 200*i),v2,vl);
}
}
void f4 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16mf4_t src = __riscv_vle16_v_u16mf4 (base, vl);
vuint8mf8_t v = __riscv_vnclipu_wx_u8mf8(src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
vuint16mf4_t v2 = __riscv_vadd_vv_u16mf4 (src, src,vl);
__riscv_vse8_v_u8mf8 (out,v,vl);
__riscv_vse16_v_u16mf4 ((int16_t *)out,v2,vl);
}
void f5 (void *base,void *base2,void *out,size_t vl, int n, size_t shift)
{
vuint16mf4_t src = __riscv_vle16_v_u16mf4 (base + 100, vl);
for (int i = 0; i < n; i++){
vbool64_t m = __riscv_vlm_v_b64 (base + i, vl);
vuint8mf8_t v = __riscv_vnclipu_wx_u8mf8_m(m,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vle8_v_u8mf8_tu (v, base2, vl);
__riscv_vse8_v_u8mf8 (out + 100*i,v,vl);
}
}
void f6 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16m2_t src = __riscv_vle16_v_u16m2 (base, vl);
vuint8m1_t v = __riscv_vnclipu_wx_u8m1(src,31,vl);
__riscv_vse8_v_u8m1 (out,v,vl);
}
void f7 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16m2_t src = __riscv_vle16_v_u16m2 (base, vl);
vuint8m1_t src2 = __riscv_vle8_v_u8m1 ((int8_t *)(base + 100), vl);
vuint8m1_t v = __riscv_vnclipu_wx_u8m1_tu(src2,src,31,vl);
__riscv_vse8_v_u8m1 (out,v,vl);
}
void f8 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16m2_t src = __riscv_vle16_v_u16m2 (base, vl);
vuint8m1_t v = __riscv_vnclipu_wx_u8m1(src,31,vl);
vuint16m2_t v2 = __riscv_vadd_vv_u16m2 (src, src,vl);
__riscv_vse8_v_u8m1 (out,v,vl);
__riscv_vse16_v_u16m2 ((int16_t *)out,v2,vl);
}
void f9 (int16_t *base,int8_t *out,size_t vl, int n, size_t shift)
{
for (int i = 0; i < n; i++){
vuint16m2_t src = __riscv_vle16_v_u16m2 (base + 100*i, vl);
vuint8m1_t v = __riscv_vnclipu_wx_u8m1(src,31,vl);
vuint16m2_t v2 = __riscv_vadd_vv_u16m2 (src, src,vl);
__riscv_vse8_v_u8m1 (out + 100*i,v,vl);
__riscv_vse16_v_u16m2 ((int16_t *)(out + 200*i),v2,vl);
}
}
void f10 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16m2_t src = __riscv_vle16_v_u16m2 (base, vl);
vuint8m1_t v = __riscv_vnclipu_wx_u8m1(src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
vuint16m2_t v2 = __riscv_vadd_vv_u16m2 (src, src,vl);
__riscv_vse8_v_u8m1 (out,v,vl);
__riscv_vse16_v_u16m2 ((int16_t *)out,v2,vl);
}
void f11 (void *base,void *base2,void *out,size_t vl, int n, size_t shift)
{
vuint16m2_t src = __riscv_vle16_v_u16m2 (base + 100, vl);
for (int i = 0; i < n; i++){
vbool8_t m = __riscv_vlm_v_b8 (base + i, vl);
vuint8m1_t v = __riscv_vnclipu_wx_u8m1_m(m,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vle8_v_u8m1_tu (v, base2, vl);
__riscv_vse8_v_u8m1 (out + 100*i,v,vl);
}
}
void f12 (int16_t *base,int8_t *out,size_t vl, int n, size_t shift)
{
vuint8mf8_t v = __riscv_vle8_v_u8mf8 ((int8_t *)(base + 1000), vl);
for (int i = 0; i < n; i++){
vuint16mf4_t src = __riscv_vle16_v_u16mf4 (base + 100*i, vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
__riscv_vse8_v_u8mf8 (out + 100*i,v,vl);
}
}
void f13 (int16_t *base,int8_t *out,size_t vl, int n, size_t shift)
{
vuint8m1_t v = __riscv_vle8_v_u8m1 ((int8_t *)(base + 1000), vl);
for (int i = 0; i < n; i++){
vuint16m2_t src = __riscv_vle16_v_u16m2 (base + 100*i, vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
__riscv_vse8_v_u8m1 (out + 100*i,v,vl);
}
}
void f14 (int16_t *base,int8_t *out,size_t vl, int n, size_t shift)
{
for (int i = 0; i < n; i++){
vuint8mf8_t v = __riscv_vle8_v_u8mf8 ((int8_t *)(base + 1000 * i), vl);
vuint16mf4_t src = __riscv_vle16_v_u16mf4 (base + 100*i, vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src,31,vl);
__riscv_vse8_v_u8mf8 (out + 100*i,v,vl);
}
}
void f15 (int16_t *base,int8_t *out,size_t vl, int n, size_t shift)
{
for (int i = 0; i < n; i++){
vuint8m1_t v = __riscv_vle8_v_u8m1 ((int8_t *)(base + 1000 * i), vl);
vuint16m2_t src = __riscv_vle16_v_u16m2 (base + 100*i, vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src,31,vl);
__riscv_vse8_v_u8m1 (out + 100*i,v,vl);
}
}
void f16 (int16_t *base,int8_t *out,size_t vl, int n, size_t shift)
{
for (int i = 0; i < n; i++){
vuint8mf8_t v = __riscv_vle8_v_u8mf8 ((int8_t *)(base + 1000 * i), vl);
vuint16mf4_t src1 = __riscv_vle16_v_u16mf4 (base + 100*i, vl);
vuint16mf4_t src2 = __riscv_vle16_v_u16mf4 (base + 200*i, vl);
vuint16mf4_t src3 = __riscv_vle16_v_u16mf4 (base + 300*i, vl);
vuint16mf4_t src4 = __riscv_vle16_v_u16mf4 (base + 400*i, vl);
vuint16mf4_t src5 = __riscv_vle16_v_u16mf4 (base + 500*i, vl);
vuint16mf4_t src6 = __riscv_vle16_v_u16mf4 (base + 600*i, vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src1,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src2,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src3,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src4,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src5,31,vl);
v = __riscv_vnclipu_wx_u8mf8_tu(v,src6,31,vl);
__riscv_vse8_v_u8mf8 (out + 100*i,v,vl);
}
}
void f17 (int16_t *base,int8_t *out,size_t vl, int n, size_t shift)
{
for (int i = 0; i < n; i++){
vuint8m1_t v = __riscv_vle8_v_u8m1 ((int8_t *)(base + 1000 * i), vl);
vuint16m2_t src1 = __riscv_vle16_v_u16m2 (base + 100*i, vl);
vuint16m2_t src2 = __riscv_vle16_v_u16m2 (base + 200*i, vl);
vuint16m2_t src3 = __riscv_vle16_v_u16m2 (base + 300*i, vl);
vuint16m2_t src4 = __riscv_vle16_v_u16m2 (base + 400*i, vl);
vuint16m2_t src5 = __riscv_vle16_v_u16m2 (base + 500*i, vl);
vuint16m2_t src6 = __riscv_vle16_v_u16m2 (base + 600*i, vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src1,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src2,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src3,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src4,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src5,31,vl);
v = __riscv_vnclipu_wx_u8m1_tu(v,src6,31,vl);
__riscv_vse8_v_u8m1 (out + 100*i,v,vl);
}
}
void f18 (void *base,void *base2,void *out,size_t vl, int n, size_t shift)
{
vuint32mf2_t src = __riscv_vle32_v_u32mf2 (base + 100, vl);
for (int i = 0; i < n; i++){
vbool64_t m = __riscv_vlm_v_b64 (base + i, vl);
vuint16mf4_t v = __riscv_vnclipu_wx_u16mf4_m(m,src,31,vl);
vuint16mf4_t v2 = __riscv_vle16_v_u16mf4_tu (v, base2 + i, vl);
vuint8mf8_t v3 = __riscv_vnclipu_wx_u8mf8_m(m,v2,31,vl);
__riscv_vse8_v_u8mf8 (out + 100*i,v3,vl);
}
}
void f19 (void *base,void *base2,void *out,size_t vl, int n, size_t shift)
{
vuint32m4_t src = __riscv_vle32_v_u32m4 (base + 100, vl);
for (int i = 0; i < n; i++){
vbool8_t m = __riscv_vlm_v_b8 (base + i, vl);
vuint16m2_t v = __riscv_vnclipu_wx_u16m2_m(m,src,31,vl);
vuint16m2_t v2 = __riscv_vle16_v_u16m2_tu (v, base2 + i, vl);
vuint8m1_t v3 = __riscv_vnclipu_wx_u8m1_m(m,v2,31,vl);
vuint8m1_t v4 = __riscv_vnclipu_wx_u8m1_tumu(m,v3,v2,31,vl);
__riscv_vse8_v_u8m1 (out + 100*i,v3,vl);
__riscv_vse8_v_u8m1 (out + 222*i,v4,vl);
}
}
void f20 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16m2_t src = __riscv_vle16_v_u16m2 (base, vl);
/* Only allow load v30,v31. */
asm volatile("#" ::
: "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9",
"v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17",
"v18", "v19", "v20", "v21", "v22", "v23", "v24", "v25",
"v26", "v27", "v28", "v29");
vuint8m1_t v = __riscv_vnclipu_wx_u8m1(src,31,vl);
/* Only allow vncvt SRC == DEST v30. */
asm volatile("#" ::
: "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9",
"v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17",
"v18", "v19", "v20", "v21", "v22", "v23", "v24", "v25",
"v26", "v27", "v28", "v29", "v31");
__riscv_vse8_v_u8m1 (out,v,vl);
}
void f21 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16m1_t src = __riscv_vle16_v_u16m1 (base, vl);
/* Only allow load v31. */
asm volatile("#" ::
: "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9",
"v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17",
"v18", "v19", "v20", "v21", "v22", "v23", "v24", "v25",
"v26", "v27", "v28", "v29", "v30");
vuint8mf2_t v = __riscv_vnclipu_wx_u8mf2(src,31,vl);
/* Only allow vncvt SRC == DEST v31. */
asm volatile("#" ::
: "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9",
"v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17",
"v18", "v19", "v20", "v21", "v22", "v23", "v24", "v25",
"v26", "v27", "v28", "v29", "v30");
__riscv_vse8_v_u8mf2 (out,v,vl);
}
void f22 (int16_t *base,int8_t *out,size_t vl, size_t shift)
{
vuint16m2_t src = __riscv_vle16_v_u16m2 (base, vl);
/* Only allow load v30,v31. */
asm volatile("#" ::
: "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9",
"v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17",
"v18", "v19", "v20", "v21", "v22", "v23", "v24", "v25",
"v26", "v27", "v28", "v29");
vuint8m1_t v = __riscv_vnclipu_wx_u8m1(src,31,vl);
/* Only allow v29. */
asm volatile("#" ::
: "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9",
"v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17",
"v18", "v19", "v20", "v21", "v22", "v23", "v24", "v25",
"v26", "v27", "v28", "v30", "v31");
v = __riscv_vadd_vv_u8m1 (v,v,vl);
/* Only allow v29. */
asm volatile("#" ::
: "v0", "v1", "v2", "v3", "v4", "v5", "v6", "v7", "v8", "v9",
"v10", "v11", "v12", "v13", "v14", "v15", "v16", "v17",
"v18", "v19", "v20", "v21", "v22", "v23", "v24", "v25",
"v26", "v27", "v28", "v30", "v31");
__riscv_vse8_v_u8m1 (out,v,vl);
}
/* { dg-final { scan-assembler-not {vmv} } } */
/* { dg-final { scan-assembler-not {csrr} } } */
|
