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
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
|
# RUN: llvm-mc -triple=hexagon -disassemble < %s | FileCheck %s
# Hexagon Programmer's Reference Manual 11.10.5 XTYPE/MPY
# Multiply and use lower result
0xb1 0xdf 0x35 0xd7
# CHECK: r17 = add(#21, mpyi(r21, r31))
0xbf 0xd1 0x35 0xd8
# CHECK: r17 = add(#21, mpyi(r21, #31))
0xb5 0xd1 0x3f 0xdf
# CHECK: r17 = add(r21, mpyi(#84, r31))
0xf5 0xf1 0xb5 0xdf
# CHECK: r17 = add(r21, mpyi(r21, #31))
0x15 0xd1 0x1f 0xe3
# CHECK: r17 = add(r21, mpyi(r17, r31))
0xf1 0xc3 0x15 0xe0
# CHECK: r17 =+ mpyi(r21, #31)
0xf1 0xc3 0x95 0xe0
# CHECK: r17 =- mpyi(r21, #31)
0xf1 0xc3 0x15 0xe1
# CHECK: r17 += mpyi(r21, #31)
0xf1 0xc3 0x95 0xe1
# CHECK: r17 -= mpyi(r21, #31)
0x11 0xdf 0x15 0xed
# CHECK: r17 = mpyi(r21, r31)
0x11 0xdf 0x15 0xef
# CHECK: r17 += mpyi(r21, r31)
# Vector multiply word by signed half (32x16)
0xb0 0xde 0x14 0xe8
# CHECK: r17:16 = vmpyweh(r21:20, r31:30):sat
0xb0 0xde 0x94 0xe8
# CHECK: r17:16 = vmpyweh(r21:20, r31:30):<<1:sat
0xf0 0xde 0x14 0xe8
# CHECK: r17:16 = vmpywoh(r21:20, r31:30):sat
0xf0 0xde 0x94 0xe8
# CHECK: r17:16 = vmpywoh(r21:20, r31:30):<<1:sat
0xb0 0xde 0x34 0xe8
# CHECK: r17:16 = vmpyweh(r21:20, r31:30):rnd:sat
0xb0 0xde 0xb4 0xe8
# CHECK: r17:16 = vmpyweh(r21:20, r31:30):<<1:rnd:sat
0xf0 0xde 0x34 0xe8
# CHECK: r17:16 = vmpywoh(r21:20, r31:30):rnd:sat
0xf0 0xde 0xb4 0xe8
# CHECK: r17:16 = vmpywoh(r21:20, r31:30):<<1:rnd:sat
0xb0 0xde 0x14 0xea
# CHECK: r17:16 += vmpyweh(r21:20, r31:30):sat
0xb0 0xde 0x94 0xea
# CHECK: r17:16 += vmpyweh(r21:20, r31:30):<<1:sat
0xf0 0xde 0x14 0xea
# CHECK: r17:16 += vmpywoh(r21:20, r31:30):sat
0xf0 0xde 0x94 0xea
# CHECK: r17:16 += vmpywoh(r21:20, r31:30):<<1:sat
0xb0 0xde 0x34 0xea
# CHECK: r17:16 += vmpyweh(r21:20, r31:30):rnd:sat
0xb0 0xde 0xb4 0xea
# CHECK: r17:16 += vmpyweh(r21:20, r31:30):<<1:rnd:sat
0xf0 0xde 0x34 0xea
# CHECK: r17:16 += vmpywoh(r21:20, r31:30):rnd:sat
0xf0 0xde 0xb4 0xea
# CHECK: r17:16 += vmpywoh(r21:20, r31:30):<<1:rnd:sat
# Vector multiply word by unsigned half (32x16)
0xb0 0xde 0x54 0xe8
# CHECK: r17:16 = vmpyweuh(r21:20, r31:30):sat
0xb0 0xde 0xd4 0xe8
# CHECK: r17:16 = vmpyweuh(r21:20, r31:30):<<1:sat
0xf0 0xde 0x54 0xe8
# CHECK: r17:16 = vmpywouh(r21:20, r31:30):sat
0xf0 0xde 0xd4 0xe8
# CHECK: r17:16 = vmpywouh(r21:20, r31:30):<<1:sat
0xb0 0xde 0x74 0xe8
# CHECK: r17:16 = vmpyweuh(r21:20, r31:30):rnd:sat
0xb0 0xde 0xf4 0xe8
# CHECK: r17:16 = vmpyweuh(r21:20, r31:30):<<1:rnd:sat
0xf0 0xde 0x74 0xe8
# CHECK: r17:16 = vmpywouh(r21:20, r31:30):rnd:sat
0xf0 0xde 0xf4 0xe8
# CHECK: r17:16 = vmpywouh(r21:20, r31:30):<<1:rnd:sat
0xb0 0xde 0x54 0xea
# CHECK: r17:16 += vmpyweuh(r21:20, r31:30):sat
0xb0 0xde 0xd4 0xea
# CHECK: r17:16 += vmpyweuh(r21:20, r31:30):<<1:sat
0xf0 0xde 0x54 0xea
# CHECK: r17:16 += vmpywouh(r21:20, r31:30):sat
0xf0 0xde 0xd4 0xea
# CHECK: r17:16 += vmpywouh(r21:20, r31:30):<<1:sat
0xb0 0xde 0x74 0xea
# CHECK: r17:16 += vmpyweuh(r21:20, r31:30):rnd:sat
0xb0 0xde 0xf4 0xea
# CHECK: r17:16 += vmpyweuh(r21:20, r31:30):<<1:rnd:sat
0xf0 0xde 0x74 0xea
# CHECK: r17:16 += vmpywouh(r21:20, r31:30):rnd:sat
0xf0 0xde 0xf4 0xea
# CHECK: r17:16 += vmpywouh(r21:20, r31:30):<<1:rnd:sat
# Multiply signed halfwords
0x10 0xdf 0x95 0xe4
# CHECK: r17:16 = mpy(r21.l, r31.l):<<1
0x30 0xdf 0x95 0xe4
# CHECK: r17:16 = mpy(r21.l, r31.h):<<1
0x50 0xdf 0x95 0xe4
# CHECK: r17:16 = mpy(r21.h, r31.l):<<1
0x70 0xdf 0x95 0xe4
# CHECK: r17:16 = mpy(r21.h, r31.h):<<1
0x10 0xdf 0xb5 0xe4
# CHECK: r17:16 = mpy(r21.l, r31.l):<<1:rnd
0x30 0xdf 0xb5 0xe4
# CHECK: r17:16 = mpy(r21.l, r31.h):<<1:rnd
0x50 0xdf 0xb5 0xe4
# CHECK: r17:16 = mpy(r21.h, r31.l):<<1:rnd
0x70 0xdf 0xb5 0xe4
# CHECK: r17:16 = mpy(r21.h, r31.h):<<1:rnd
0x10 0xdf 0x95 0xe6
# CHECK: r17:16 += mpy(r21.l, r31.l):<<1
0x30 0xdf 0x95 0xe6
# CHECK: r17:16 += mpy(r21.l, r31.h):<<1
0x50 0xdf 0x95 0xe6
# CHECK: r17:16 += mpy(r21.h, r31.l):<<1
0x70 0xdf 0x95 0xe6
# CHECK: r17:16 += mpy(r21.h, r31.h):<<1
0x10 0xdf 0xb5 0xe6
# CHECK: r17:16 -= mpy(r21.l, r31.l):<<1
0x30 0xdf 0xb5 0xe6
# CHECK: r17:16 -= mpy(r21.l, r31.h):<<1
0x50 0xdf 0xb5 0xe6
# CHECK: r17:16 -= mpy(r21.h, r31.l):<<1
0x70 0xdf 0xb5 0xe6
# CHECK: r17:16 -= mpy(r21.h, r31.h):<<1
0x11 0xdf 0x95 0xec
# CHECK: r17 = mpy(r21.l, r31.l):<<1
0x31 0xdf 0x95 0xec
# CHECK: r17 = mpy(r21.l, r31.h):<<1
0x51 0xdf 0x95 0xec
# CHECK: r17 = mpy(r21.h, r31.l):<<1
0x71 0xdf 0x95 0xec
# CHECK: r17 = mpy(r21.h, r31.h):<<1
0x91 0xdf 0x95 0xec
# CHECK: r17 = mpy(r21.l, r31.l):<<1:sat
0xb1 0xdf 0x95 0xec
# CHECK: r17 = mpy(r21.l, r31.h):<<1:sat
0xd1 0xdf 0x95 0xec
# CHECK: r17 = mpy(r21.h, r31.l):<<1:sat
0xf1 0xdf 0x95 0xec
# CHECK: r17 = mpy(r21.h, r31.h):<<1:sat
0x11 0xdf 0xb5 0xec
# CHECK: r17 = mpy(r21.l, r31.l):<<1:rnd
0x31 0xdf 0xb5 0xec
# CHECK: r17 = mpy(r21.l, r31.h):<<1:rnd
0x51 0xdf 0xb5 0xec
# CHECK: r17 = mpy(r21.h, r31.l):<<1:rnd
0x71 0xdf 0xb5 0xec
# CHECK: r17 = mpy(r21.h, r31.h):<<1:rnd
0x91 0xdf 0xb5 0xec
# CHECK: r17 = mpy(r21.l, r31.l):<<1:rnd:sat
0xb1 0xdf 0xb5 0xec
# CHECK: r17 = mpy(r21.l, r31.h):<<1:rnd:sat
0xd1 0xdf 0xb5 0xec
# CHECK: r17 = mpy(r21.h, r31.l):<<1:rnd:sat
0xf1 0xdf 0xb5 0xec
# CHECK: r17 = mpy(r21.h, r31.h):<<1:rnd:sat
0x11 0xdf 0x95 0xee
# CHECK: r17 += mpy(r21.l, r31.l):<<1
0x31 0xdf 0x95 0xee
# CHECK: r17 += mpy(r21.l, r31.h):<<1
0x51 0xdf 0x95 0xee
# CHECK: r17 += mpy(r21.h, r31.l):<<1
0x71 0xdf 0x95 0xee
# CHECK: r17 += mpy(r21.h, r31.h):<<1
0x91 0xdf 0x95 0xee
# CHECK: r17 += mpy(r21.l, r31.l):<<1:sat
0xb1 0xdf 0x95 0xee
# CHECK: r17 += mpy(r21.l, r31.h):<<1:sat
0xd1 0xdf 0x95 0xee
# CHECK: r17 += mpy(r21.h, r31.l):<<1:sat
0xf1 0xdf 0x95 0xee
# CHECK: r17 += mpy(r21.h, r31.h):<<1:sat
0x11 0xdf 0xb5 0xee
# CHECK: r17 -= mpy(r21.l, r31.l):<<1
0x31 0xdf 0xb5 0xee
# CHECK: r17 -= mpy(r21.l, r31.h):<<1
0x51 0xdf 0xb5 0xee
# CHECK: r17 -= mpy(r21.h, r31.l):<<1
0x71 0xdf 0xb5 0xee
# CHECK: r17 -= mpy(r21.h, r31.h):<<1
0x91 0xdf 0xb5 0xee
# CHECK: r17 -= mpy(r21.l, r31.l):<<1:sat
0xb1 0xdf 0xb5 0xee
# CHECK: r17 -= mpy(r21.l, r31.h):<<1:sat
0xd1 0xdf 0xb5 0xee
# CHECK: r17 -= mpy(r21.h, r31.l):<<1:sat
0xf1 0xdf 0xb5 0xee
# CHECK: r17 -= mpy(r21.h, r31.h):<<1:sat
# Multiply unsigned halfwords
0x10 0xdf 0xd5 0xe4
# CHECK: r17:16 = mpyu(r21.l, r31.l):<<1
0x30 0xdf 0xd5 0xe4
# CHECK: r17:16 = mpyu(r21.l, r31.h):<<1
0x50 0xdf 0xd5 0xe4
# CHECK: r17:16 = mpyu(r21.h, r31.l):<<1
0x70 0xdf 0xd5 0xe4
# CHECK: r17:16 = mpyu(r21.h, r31.h):<<1
0x10 0xdf 0xd5 0xe6
# CHECK: r17:16 += mpyu(r21.l, r31.l):<<1
0x30 0xdf 0xd5 0xe6
# CHECK: r17:16 += mpyu(r21.l, r31.h):<<1
0x50 0xdf 0xd5 0xe6
# CHECK: r17:16 += mpyu(r21.h, r31.l):<<1
0x70 0xdf 0xd5 0xe6
# CHECK: r17:16 += mpyu(r21.h, r31.h):<<1
0x10 0xdf 0xf5 0xe6
# CHECK: r17:16 -= mpyu(r21.l, r31.l):<<1
0x30 0xdf 0xf5 0xe6
# CHECK: r17:16 -= mpyu(r21.l, r31.h):<<1
0x50 0xdf 0xf5 0xe6
# CHECK: r17:16 -= mpyu(r21.h, r31.l):<<1
0x70 0xdf 0xf5 0xe6
# CHECK: r17:16 -= mpyu(r21.h, r31.h):<<1
0x11 0xdf 0xd5 0xec
# CHECK: r17 = mpyu(r21.l, r31.l):<<1
0x31 0xdf 0xd5 0xec
# CHECK: r17 = mpyu(r21.l, r31.h):<<1
0x51 0xdf 0xd5 0xec
# CHECK: r17 = mpyu(r21.h, r31.l):<<1
0x71 0xdf 0xd5 0xec
# CHECK: r17 = mpyu(r21.h, r31.h):<<1
0x11 0xdf 0xd5 0xee
# CHECK: r17 += mpyu(r21.l, r31.l):<<1
0x31 0xdf 0xd5 0xee
# CHECK: r17 += mpyu(r21.l, r31.h):<<1
0x51 0xdf 0xd5 0xee
# CHECK: r17 += mpyu(r21.h, r31.l):<<1
0x71 0xdf 0xd5 0xee
# CHECK: r17 += mpyu(r21.h, r31.h):<<1
0x11 0xdf 0xf5 0xee
# CHECK: r17 -= mpyu(r21.l, r31.l):<<1
0x31 0xdf 0xf5 0xee
# CHECK: r17 -= mpyu(r21.l, r31.h):<<1
0x51 0xdf 0xf5 0xee
# CHECK: r17 -= mpyu(r21.h, r31.l):<<1
0x71 0xdf 0xf5 0xee
# CHECK: r17 -= mpyu(r21.h, r31.h):<<1
# Polynomial multiply words
0xf0 0xdf 0x55 0xe5
# CHECK: r17:16 = pmpyw(r21, r31)
0xf0 0xdf 0x35 0xe7
# CHECK: r17:16 ^= pmpyw(r21, r31)
# Vector reduce multiply word by signed half (32x16)
0x50 0xde 0x34 0xe8
# CHECK: r17:16 = vrmpywoh(r21:20, r31:30)
0x50 0xde 0xb4 0xe8
# CHECK: r17:16 = vrmpywoh(r21:20, r31:30):<<1
0x90 0xde 0x54 0xe8
# CHECK: r17:16 = vrmpyweh(r21:20, r31:30)
0x90 0xde 0xd4 0xe8
# CHECK: r17:16 = vrmpyweh(r21:20, r31:30):<<1
0xd0 0xde 0x74 0xea
# CHECK: r17:16 += vrmpywoh(r21:20, r31:30)
0xd0 0xde 0xf4 0xea
# CHECK: r17:16 += vrmpywoh(r21:20, r31:30):<<1
0xd0 0xde 0x34 0xea
# CHECK: r17:16 += vrmpyweh(r21:20, r31:30)
0xd0 0xde 0xb4 0xea
# CHECK: r17:16 += vrmpyweh(r21:20, r31:30):<<1
# Multiply and use upper result
0x31 0xdf 0x15 0xed
# CHECK: r17 = mpy(r21, r31)
0x31 0xdf 0x35 0xed
# CHECK: r17 = mpy(r21, r31):rnd
0x31 0xdf 0x55 0xed
# CHECK: r17 = mpyu(r21, r31)
0x31 0xdf 0x75 0xed
# CHECK: r17 = mpysu(r21, r31)
0x11 0xdf 0xb5 0xed
# CHECK: r17 = mpy(r21, r31.h):<<1:sat
0x31 0xdf 0xb5 0xed
# CHECK: r17 = mpy(r21, r31.l):<<1:sat
0x91 0xdf 0xb5 0xed
# CHECK: r17 = mpy(r21, r31.h):<<1:rnd:sat
0x11 0xdf 0xf5 0xed
# CHECK: r17 = mpy(r21, r31):<<1:sat
0x91 0xdf 0xf5 0xed
# CHECK: r17 = mpy(r21, r31.l):<<1:rnd:sat
0x51 0xdf 0xb5 0xed
# CHECK: r17 = mpy(r21, r31):<<1
0x11 0xdf 0x75 0xef
# CHECK: r17 += mpy(r21, r31):<<1:sat
0x31 0xdf 0x75 0xef
# CHECK: r17 -= mpy(r21, r31):<<1:sat
# Multiply and use full result
0x10 0xdf 0x15 0xe5
# CHECK: r17:16 = mpy(r21, r31)
0x10 0xdf 0x55 0xe5
# CHECK: r17:16 = mpyu(r21, r31)
0x10 0xdf 0x15 0xe7
# CHECK: r17:16 += mpy(r21, r31)
0x10 0xdf 0x35 0xe7
# CHECK: r17:16 -= mpy(r21, r31)
0x10 0xdf 0x55 0xe7
# CHECK: r17:16 += mpyu(r21, r31)
0x10 0xdf 0x75 0xe7
# CHECK: r17:16 -= mpyu(r21, r31)
# Vector dual multiply
0x90 0xde 0x14 0xe8
# CHECK: r17:16 = vdmpy(r21:20, r31:30):sat
0x90 0xde 0x94 0xe8
# CHECK: r17:16 = vdmpy(r21:20, r31:30):<<1:sat
0x90 0xde 0x14 0xea
# CHECK: r17:16 += vdmpy(r21:20, r31:30):sat
0x90 0xde 0x94 0xea
# CHECK: r17:16 += vdmpy(r21:20, r31:30):<<1:sat
# Vector dual multiply with round and pack
0x11 0xde 0x14 0xe9
# CHECK: r17 = vdmpy(r21:20, r31:30):rnd:sat
0x11 0xde 0x94 0xe9
# CHECK: r17 = vdmpy(r21:20, r31:30):<<1:rnd:sat
# Vector reduce multiply bytes
0x30 0xde 0x94 0xe8
# CHECK: r17:16 = vrmpybu(r21:20, r31:30)
0x30 0xde 0xd4 0xe8
# CHECK: r17:16 = vrmpybsu(r21:20, r31:30)
0x30 0xde 0x94 0xea
# CHECK: r17:16 += vrmpybu(r21:20, r31:30)
0x30 0xde 0xd4 0xea
# CHECK: r17:16 += vrmpybsu(r21:20, r31:30)
# Vector dual multiply signed by unsigned bytes
0x30 0xde 0xb4 0xe8
# CHECK: r17:16 = vdmpybsu(r21:20, r31:30):sat
0x30 0xde 0x34 0xea
# CHECK: r17:16 += vdmpybsu(r21:20, r31:30):sat
# Vector multiply even haldwords
0xd0 0xde 0x14 0xe8
# CHECK: r17:16 = vmpyeh(r21:20, r31:30):sat
0xd0 0xde 0x94 0xe8
# CHECK: r17:16 = vmpyeh(r21:20, r31:30):<<1:sat
0x50 0xde 0x34 0xea
# CHECK: r17:16 += vmpyeh(r21:20, r31:30)
0xd0 0xde 0x14 0xea
# CHECK: r17:16 += vmpyeh(r21:20, r31:30):sat
0xd0 0xde 0x94 0xea
# CHECK: r17:16 += vmpyeh(r21:20, r31:30):<<1:sat
# Vector multiply halfwords
0xb0 0xdf 0x15 0xe5
# CHECK: r17:16 = vmpyh(r21, r31):sat
0xb0 0xdf 0x95 0xe5
# CHECK: r17:16 = vmpyh(r21, r31):<<1:sat
0x30 0xdf 0x35 0xe7
# CHECK: r17:16 += vmpyh(r21, r31)
0xb0 0xdf 0x15 0xe7
# CHECK: r17:16 += vmpyh(r21, r31):sat
0xb0 0xdf 0x95 0xe7
# CHECK: r17:16 += vmpyh(r21, r31):<<1:sat
# Vector multiply halfwords with round and pack
0xf1 0xdf 0x35 0xed
# CHECK: r17 = vmpyh(r21, r31):rnd:sat
0xf1 0xdf 0xb5 0xed
# CHECK: r17 = vmpyh(r21, r31):<<1:rnd:sat
# Vector multiply halfwords signed by unsigned
0xf0 0xdf 0x15 0xe5
# CHECK: r17:16 = vmpyhsu(r21, r31):sat
0xf0 0xdf 0x95 0xe5
# CHECK: r17:16 = vmpyhsu(r21, r31):<<1:sat
0xb0 0xdf 0x75 0xe7
# CHECK: r17:16 += vmpyhsu(r21, r31):sat
0xb0 0xdf 0xf5 0xe7
# CHECK: r17:16 += vmpyhsu(r21, r31):<<1:sat
# Vector reduce multiply halfwords
0x50 0xde 0x14 0xe8
# CHECK: r17:16 = vrmpyh(r21:20, r31:30)
0x50 0xde 0x14 0xea
# CHECK: r17:16 += vrmpyh(r21:20, r31:30)
# Vector multiply bytes
0x30 0xdf 0x55 0xe5
# CHECK: r17:16 = vmpybsu(r21, r31)
0x30 0xdf 0x95 0xe5
# CHECK: r17:16 = vmpybu(r21, r31)
0x30 0xdf 0x95 0xe7
# CHECK: r17:16 += vmpybu(r21, r31)
0x30 0xdf 0xd5 0xe7
# CHECK: r17:16 += vmpybsu(r21, r31)
# Vector polynomial multiply halfwords
0xf0 0xdf 0xd5 0xe5
# CHECK: r17:16 = vpmpyh(r21, r31)
0xf0 0xdf 0xb5 0xe7
# CHECK: r17:16 ^= vpmpyh(r21, r31)
|
