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
|
/* { dg-do run { target { powerpc*-*-* && vmx_hw } } } */
/* { dg-do compile { target { powerpc*-*-* && { ! vmx_hw } } } } */
/* { dg-require-effective-target powerpc_altivec_ok } */
/* { dg-options "-maltivec -mabi=altivec -O2" } */
/* Check that "easy" AltiVec constants are correctly synthesized. */
extern void abort (void);
typedef __attribute__ ((vector_size (16))) unsigned char v16qi;
typedef __attribute__ ((vector_size (16))) unsigned short v8hi;
typedef __attribute__ ((vector_size (16))) unsigned int v4si;
char w[16] __attribute__((aligned(16)));
/* Emulate the vspltis? instructions on a 16-byte array of chars. */
void vspltisb (char *v, int val)
{
int i;
for (i = 0; i < 16; i++)
v[i] = val;
}
void vspltish (char *v, int val)
{
int i;
for (i = 0; i < 16; i += 2)
v[i] = val >> 7, v[i + 1] = val;
}
void vspltisw (char *v, int val)
{
int i;
for (i = 0; i < 16; i += 4)
v[i] = v[i + 1] = v[i + 2] = val >> 7, v[i + 3] = val;
}
/* Use three different check functions for each mode-instruction pair.
The callers have no typecasting and no addressable vectors, to make
the test more robust. */
void __attribute__ ((noinline)) check_v16qi (v16qi v1, char *v2)
{
if (memcmp (&v1, v2, 16))
abort ();
}
void __attribute__ ((noinline)) check_v8hi (v8hi v1, char *v2)
{
if (memcmp (&v1, v2, 16))
abort ();
}
void __attribute__ ((noinline)) check_v4si (v4si v1, char *v2)
{
if (memcmp (&v1, v2, 16))
abort ();
}
/* V16QI tests. */
void v16qi_vspltisb ()
{
v16qi v = { 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15 };
vspltisb (w, 15);
check_v16qi (v, w);
}
void v16qi_vspltisb_neg ()
{
v16qi v = { -5, -5, -5, -5, -5, -5, -5, -5, -5, -5, -5, -5, -5, -5, -5, -5 };
vspltisb (w, -5);
check_v16qi (v, w);
}
void v16qi_vspltisb_addself ()
{
v16qi v = { 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30, 30 };
vspltisb (w, 30);
check_v16qi (v, w);
}
void v16qi_vspltisb_neg_addself ()
{
v16qi v = { -24, -24, -24, -24, -24, -24, -24, -24, -24, -24, -24, -24, -24, -24, -24, -24 };
vspltisb (w, -24);
check_v16qi (v, w);
}
void v16qi_vspltish ()
{
v16qi v = { 0, 15, 0, 15, 0, 15, 0, 15, 0, 15, 0, 15, 0, 15, 0, 15 };
vspltish (w, 15);
check_v16qi (v, w);
}
void v16qi_vspltish_addself ()
{
v16qi v = { 0, 30, 0, 30, 0, 30, 0, 30, 0, 30, 0, 30, 0, 30, 0, 30 };
vspltish (w, 30);
check_v16qi (v, w);
}
void v16qi_vspltish_neg ()
{
v16qi v = { -1, -5, -1, -5, -1, -5, -1, -5, -1, -5, -1, -5, -1, -5, -1, -5 };
vspltish (w, -5);
check_v16qi (v, w);
}
void v16qi_vspltisw ()
{
v16qi v = { 0, 0, 0, 15, 0, 0, 0, 15, 0, 0, 0, 15, 0, 0, 0, 15 };
vspltisw (w, 15);
check_v16qi (v, w);
}
void v16qi_vspltisw_addself ()
{
v16qi v = { 0, 0, 0, 30, 0, 0, 0, 30, 0, 0, 0, 30, 0, 0, 0, 30 };
vspltisw (w, 30);
check_v16qi (v, w);
}
void v16qi_vspltisw_neg ()
{
v16qi v = { -1, -1, -1, -5, -1, -1, -1, -5, -1, -1, -1, -5, -1, -1, -1, -5 };
vspltisw (w, -5);
check_v16qi (v, w);
}
/* V8HI tests. */
void v8hi_vspltisb ()
{
v8hi v = { 0x0F0F, 0x0F0F, 0x0F0F, 0x0F0F, 0x0F0F, 0x0F0F, 0x0F0F, 0x0F0F };
vspltisb (w, 15);
check_v8hi (v, w);
}
void v8hi_vspltisb_addself ()
{
v8hi v = { 0x1E1E, 0x1E1E, 0x1E1E, 0x1E1E, 0x1E1E, 0x1E1E, 0x1E1E, 0x1E1E };
vspltisb (w, 30);
check_v8hi (v, w);
}
void v8hi_vspltisb_neg ()
{
v8hi v = { 0xFBFB, 0xFBFB, 0xFBFB, 0xFBFB, 0xFBFB, 0xFBFB, 0xFBFB, 0xFBFB };
vspltisb (w, -5);
check_v8hi (v, w);
}
void v8hi_vspltish ()
{
v8hi v = { 15, 15, 15, 15, 15, 15, 15, 15 };
vspltish (w, 15);
check_v8hi (v, w);
}
void v8hi_vspltish_neg ()
{
v8hi v = { -5, -5, -5, -5, -5, -5, -5, -5 };
vspltish (w, -5);
check_v8hi (v, w);
}
void v8hi_vspltish_addself ()
{
v8hi v = { 30, 30, 30, 30, 30, 30, 30, 30 };
vspltish (w, 30);
check_v8hi (v, w);
}
void v8hi_vspltish_neg_addself ()
{
v8hi v = { -24, -24, -24, -24, -24, -24, -24, -24 };
vspltish (w, -24);
check_v8hi (v, w);
}
void v8hi_vspltisw ()
{
v8hi v = { 0, 15, 0, 15, 0, 15, 0, 15 };
vspltisw (w, 15);
check_v8hi (v, w);
}
void v8hi_vspltisw_addself ()
{
v8hi v = { 0, 30, 0, 30, 0, 30, 0, 30 };
vspltisw (w, 30);
check_v8hi (v, w);
}
void v8hi_vspltisw_neg ()
{
v8hi v = { -1, -5, -1, -5, -1, -5, -1, -5 };
vspltisw (w, -5);
check_v8hi (v, w);
}
/* V4SI tests. */
void v4si_vspltisb ()
{
v4si v = { 0x0F0F0F0F, 0x0F0F0F0F, 0x0F0F0F0F, 0x0F0F0F0F };
vspltisb (w, 15);
check_v4si (v, w);
}
void v4si_vspltisb_addself ()
{
v4si v = { 0x1E1E1E1E, 0x1E1E1E1E, 0x1E1E1E1E, 0x1E1E1E1E };
vspltisb (w, 30);
check_v4si (v, w);
}
void v4si_vspltisb_neg ()
{
v4si v = { 0xFBFBFBFB, 0xFBFBFBFB, 0xFBFBFBFB, 0xFBFBFBFB };
vspltisb (w, -5);
check_v4si (v, w);
}
void v4si_vspltish ()
{
v4si v = { 0x000F000F, 0x000F000F, 0x000F000F, 0x000F000F };
vspltish (w, 15);
check_v4si (v, w);
}
void v4si_vspltish_addself ()
{
v4si v = { 0x001E001E, 0x001E001E, 0x001E001E, 0x001E001E };
vspltish (w, 30);
check_v4si (v, w);
}
void v4si_vspltish_neg ()
{
v4si v = { 0xFFFBFFFB, 0xFFFBFFFB, 0xFFFBFFFB, 0xFFFBFFFB };
vspltish (w, -5);
check_v4si (v, w);
}
void v4si_vspltisw ()
{
v4si v = { 15, 15, 15, 15 };
vspltisw (w, 15);
check_v4si (v, w);
}
void v4si_vspltisw_neg ()
{
v4si v = { -5, -5, -5, -5 };
vspltisw (w, -5);
check_v4si (v, w);
}
void v4si_vspltisw_addself ()
{
v4si v = { 30, 30, 30, 30 };
vspltisw (w, 30);
check_v4si (v, w);
}
void v4si_vspltisw_neg_addself ()
{
v4si v = { -24, -24, -24, -24 };
vspltisw (w, -24);
check_v4si (v, w);
}
int main ()
{
v16qi_vspltisb ();
v16qi_vspltisb_neg ();
v16qi_vspltisb_addself ();
v16qi_vspltisb_neg_addself ();
v16qi_vspltish ();
v16qi_vspltish_addself ();
v16qi_vspltish_neg ();
v16qi_vspltisw ();
v16qi_vspltisw_addself ();
v16qi_vspltisw_neg ();
v8hi_vspltisb ();
v8hi_vspltisb_addself ();
v8hi_vspltisb_neg ();
v8hi_vspltish ();
v8hi_vspltish_neg ();
v8hi_vspltish_addself ();
v8hi_vspltish_neg_addself ();
v8hi_vspltisw ();
v8hi_vspltisw_addself ();
v8hi_vspltisw_neg ();
v4si_vspltisb ();
v4si_vspltisb_addself ();
v4si_vspltisb_neg ();
v4si_vspltish ();
v4si_vspltish_addself ();
v4si_vspltish_neg ();
v4si_vspltisw ();
v4si_vspltisw_neg ();
v4si_vspltisw_addself ();
v4si_vspltisw_neg_addself ();
return 0;
}
|
