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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=thumbv8.1m.main-none-none-eabi -mattr=+mve.fp -float-abi=hard -verify-machineinstrs %s -o - | FileCheck %s
define float @add_f32(<8 x float> %a, <4 x float> %b) {
; CHECK-LABEL: add_f32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vadd.f32 q0, q0, q1
; CHECK-NEXT: vadd.f32 q0, q0, q2
; CHECK-NEXT: vadd.f32 s2, s2, s3
; CHECK-NEXT: vadd.f32 s0, s0, s1
; CHECK-NEXT: vadd.f32 s0, s0, s2
; CHECK-NEXT: bx lr
%r1 = call fast float @llvm.vector.reduce.fadd.f32.v8f32(float -0.0, <8 x float> %a)
%r2 = call fast float @llvm.vector.reduce.fadd.f32.v4f32(float -0.0, <4 x float> %b)
%r = fadd fast float %r1, %r2
ret float %r
}
define float @fmul_f32(<8 x float> %a, <4 x float> %b) {
; CHECK-LABEL: fmul_f32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vmul.f32 q0, q0, q1
; CHECK-NEXT: vmul.f32 q0, q0, q2
; CHECK-NEXT: vmul.f32 s2, s2, s3
; CHECK-NEXT: vmul.f32 s0, s0, s1
; CHECK-NEXT: vmul.f32 s0, s0, s2
; CHECK-NEXT: bx lr
%r1 = call fast float @llvm.vector.reduce.fmul.f32.v8f32(float 1.0, <8 x float> %a)
%r2 = call fast float @llvm.vector.reduce.fmul.f32.v4f32(float 1.0, <4 x float> %b)
%r = fmul fast float %r1, %r2
ret float %r
}
define float @fmin_f32(<8 x float> %a, <4 x float> %b) {
; CHECK-LABEL: fmin_f32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vminnm.f32 q0, q0, q1
; CHECK-NEXT: vminnm.f32 q0, q0, q2
; CHECK-NEXT: vminnm.f32 s2, s2, s3
; CHECK-NEXT: vminnm.f32 s0, s0, s1
; CHECK-NEXT: vminnm.f32 s0, s0, s2
; CHECK-NEXT: bx lr
%r1 = call fast float @llvm.vector.reduce.fmin.v8f32(<8 x float> %a)
%r2 = call fast float @llvm.vector.reduce.fmin.v4f32(<4 x float> %b)
%r = call float @llvm.minnum.f32(float %r1, float %r2)
ret float %r
}
define float @fmax_f32(<8 x float> %a, <4 x float> %b) {
; CHECK-LABEL: fmax_f32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vmaxnm.f32 q0, q0, q1
; CHECK-NEXT: vmaxnm.f32 q0, q0, q2
; CHECK-NEXT: vmaxnm.f32 s2, s2, s3
; CHECK-NEXT: vmaxnm.f32 s0, s0, s1
; CHECK-NEXT: vmaxnm.f32 s0, s0, s2
; CHECK-NEXT: bx lr
%r1 = call fast float @llvm.vector.reduce.fmax.v8f32(<8 x float> %a)
%r2 = call fast float @llvm.vector.reduce.fmax.v4f32(<4 x float> %b)
%r = call float @llvm.maxnum.f32(float %r1, float %r2)
ret float %r
}
define i32 @add_i32(<8 x i32> %a, <4 x i32> %b) {
; CHECK-LABEL: add_i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vaddv.u32 r0, q1
; CHECK-NEXT: vaddva.u32 r0, q0
; CHECK-NEXT: vaddva.u32 r0, q2
; CHECK-NEXT: bx lr
%r1 = call i32 @llvm.vector.reduce.add.i32.v8i32(<8 x i32> %a)
%r2 = call i32 @llvm.vector.reduce.add.i32.v4i32(<4 x i32> %b)
%r = add i32 %r1, %r2
ret i32 %r
}
define i16 @add_ext_i16(<16 x i8> %a, <16 x i8> %b) {
; CHECK-LABEL: add_ext_i16:
; CHECK: @ %bb.0:
; CHECK-NEXT: vaddv.u8 r0, q1
; CHECK-NEXT: vaddva.u8 r0, q0
; CHECK-NEXT: bx lr
%ae = zext <16 x i8> %a to <16 x i16>
%be = zext <16 x i8> %b to <16 x i16>
%r1 = call i16 @llvm.vector.reduce.add.i16.v16i16(<16 x i16> %ae)
%r2 = call i16 @llvm.vector.reduce.add.i16.v16i16(<16 x i16> %be)
%r = add i16 %r1, %r2
ret i16 %r
}
define i16 @add_ext_v32i16(<32 x i8> %a, <16 x i8> %b) {
; CHECK-LABEL: add_ext_v32i16:
; CHECK: @ %bb.0:
; CHECK-NEXT: .pad #32
; CHECK-NEXT: sub sp, #32
; CHECK-NEXT: mov r1, sp
; CHECK-NEXT: add r2, sp, #16
; CHECK-NEXT: vstrw.32 q0, [r1]
; CHECK-NEXT: vstrw.32 q1, [r2]
; CHECK-NEXT: vldrb.u16 q1, [r2]
; CHECK-NEXT: vldrb.u16 q0, [r1]
; CHECK-NEXT: vaddv.u16 r0, q1
; CHECK-NEXT: vaddva.u16 r0, q0
; CHECK-NEXT: vldrb.u16 q0, [r1, #8]
; CHECK-NEXT: vaddva.u16 r0, q0
; CHECK-NEXT: vldrb.u16 q0, [r2, #8]
; CHECK-NEXT: vaddva.u16 r0, q0
; CHECK-NEXT: vaddva.u8 r0, q2
; CHECK-NEXT: add sp, #32
; CHECK-NEXT: bx lr
%ae = zext <32 x i8> %a to <32 x i16>
%be = zext <16 x i8> %b to <16 x i16>
%r1 = call i16 @llvm.vector.reduce.add.i16.v32i16(<32 x i16> %ae)
%r2 = call i16 @llvm.vector.reduce.add.i16.v16i16(<16 x i16> %be)
%r = add i16 %r1, %r2
ret i16 %r
}
define i32 @mul_i32(<8 x i32> %a, <4 x i32> %b) {
; CHECK-LABEL: mul_i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vmul.i32 q0, q0, q1
; CHECK-NEXT: vmul.i32 q0, q0, q2
; CHECK-NEXT: vmov r0, r1, d1
; CHECK-NEXT: vmov r2, r3, d0
; CHECK-NEXT: muls r0, r1, r0
; CHECK-NEXT: mul r1, r2, r3
; CHECK-NEXT: muls r0, r1, r0
; CHECK-NEXT: bx lr
%r1 = call i32 @llvm.vector.reduce.mul.i32.v8i32(<8 x i32> %a)
%r2 = call i32 @llvm.vector.reduce.mul.i32.v4i32(<4 x i32> %b)
%r = mul i32 %r1, %r2
ret i32 %r
}
define i32 @and_i32(<8 x i32> %a, <4 x i32> %b) {
; CHECK-LABEL: and_i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vand q0, q0, q1
; CHECK-NEXT: vand q0, q0, q2
; CHECK-NEXT: vmov r0, r1, d1
; CHECK-NEXT: vmov r2, r3, d0
; CHECK-NEXT: ands r0, r1
; CHECK-NEXT: and.w r1, r2, r3
; CHECK-NEXT: ands r0, r1
; CHECK-NEXT: bx lr
%r1 = call i32 @llvm.vector.reduce.and.i32.v8i32(<8 x i32> %a)
%r2 = call i32 @llvm.vector.reduce.and.i32.v4i32(<4 x i32> %b)
%r = and i32 %r1, %r2
ret i32 %r
}
define i32 @or_i32(<8 x i32> %a, <4 x i32> %b) {
; CHECK-LABEL: or_i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vorr q0, q0, q1
; CHECK-NEXT: vorr q0, q0, q2
; CHECK-NEXT: vmov r0, r1, d1
; CHECK-NEXT: vmov r2, r3, d0
; CHECK-NEXT: orrs r0, r1
; CHECK-NEXT: orr.w r1, r2, r3
; CHECK-NEXT: orrs r0, r1
; CHECK-NEXT: bx lr
%r1 = call i32 @llvm.vector.reduce.or.i32.v8i32(<8 x i32> %a)
%r2 = call i32 @llvm.vector.reduce.or.i32.v4i32(<4 x i32> %b)
%r = or i32 %r1, %r2
ret i32 %r
}
define i32 @xor_i32(<8 x i32> %a, <4 x i32> %b) {
; CHECK-LABEL: xor_i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: veor q0, q0, q1
; CHECK-NEXT: veor q0, q0, q2
; CHECK-NEXT: vmov r0, r1, d1
; CHECK-NEXT: vmov r2, r3, d0
; CHECK-NEXT: eors r0, r1
; CHECK-NEXT: eor.w r1, r2, r3
; CHECK-NEXT: eors r0, r1
; CHECK-NEXT: bx lr
%r1 = call i32 @llvm.vector.reduce.xor.i32.v8i32(<8 x i32> %a)
%r2 = call i32 @llvm.vector.reduce.xor.i32.v4i32(<4 x i32> %b)
%r = xor i32 %r1, %r2
ret i32 %r
}
define i32 @umin_i32(<8 x i32> %a, <4 x i32> %b) {
; CHECK-LABEL: umin_i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vmin.u32 q0, q0, q1
; CHECK-NEXT: mov.w r0, #-1
; CHECK-NEXT: vmin.u32 q0, q0, q2
; CHECK-NEXT: vminv.u32 r0, q0
; CHECK-NEXT: bx lr
%r1 = call i32 @llvm.vector.reduce.umin.i32.v8i32(<8 x i32> %a)
%r2 = call i32 @llvm.vector.reduce.umin.i32.v4i32(<4 x i32> %b)
%r = call i32 @llvm.umin.i32(i32 %r1, i32 %r2)
ret i32 %r
}
define i32 @umax_i32(<8 x i32> %a, <4 x i32> %b) {
; CHECK-LABEL: umax_i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vmax.u32 q0, q0, q1
; CHECK-NEXT: movs r0, #0
; CHECK-NEXT: vmax.u32 q0, q0, q2
; CHECK-NEXT: vmaxv.u32 r0, q0
; CHECK-NEXT: bx lr
%r1 = call i32 @llvm.vector.reduce.umax.i32.v8i32(<8 x i32> %a)
%r2 = call i32 @llvm.vector.reduce.umax.i32.v4i32(<4 x i32> %b)
%r = call i32 @llvm.umax.i32(i32 %r1, i32 %r2)
ret i32 %r
}
define i32 @smin_i32(<8 x i32> %a, <4 x i32> %b) {
; CHECK-LABEL: smin_i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vmin.s32 q0, q0, q1
; CHECK-NEXT: mvn r0, #-2147483648
; CHECK-NEXT: vmin.s32 q0, q0, q2
; CHECK-NEXT: vminv.s32 r0, q0
; CHECK-NEXT: bx lr
%r1 = call i32 @llvm.vector.reduce.smin.i32.v8i32(<8 x i32> %a)
%r2 = call i32 @llvm.vector.reduce.smin.i32.v4i32(<4 x i32> %b)
%r = call i32 @llvm.smin.i32(i32 %r1, i32 %r2)
ret i32 %r
}
define i32 @smax_i32(<8 x i32> %a, <4 x i32> %b) {
; CHECK-LABEL: smax_i32:
; CHECK: @ %bb.0:
; CHECK-NEXT: vmax.s32 q0, q0, q1
; CHECK-NEXT: mov.w r0, #-2147483648
; CHECK-NEXT: vmax.s32 q0, q0, q2
; CHECK-NEXT: vmaxv.s32 r0, q0
; CHECK-NEXT: bx lr
%r1 = call i32 @llvm.vector.reduce.smax.i32.v8i32(<8 x i32> %a)
%r2 = call i32 @llvm.vector.reduce.smax.i32.v4i32(<4 x i32> %b)
%r = call i32 @llvm.smax.i32(i32 %r1, i32 %r2)
ret i32 %r
}
declare float @llvm.vector.reduce.fadd.f32.v8f32(float, <8 x float>)
declare float @llvm.vector.reduce.fadd.f32.v4f32(float, <4 x float>)
declare float @llvm.vector.reduce.fmul.f32.v8f32(float, <8 x float>)
declare float @llvm.vector.reduce.fmul.f32.v4f32(float, <4 x float>)
declare float @llvm.vector.reduce.fmin.v8f32(<8 x float>)
declare float @llvm.vector.reduce.fmin.v4f32(<4 x float>)
declare float @llvm.vector.reduce.fmax.v8f32(<8 x float>)
declare float @llvm.vector.reduce.fmax.v4f32(<4 x float>)
declare i32 @llvm.vector.reduce.add.i32.v8i32(<8 x i32>)
declare i32 @llvm.vector.reduce.add.i32.v4i32(<4 x i32>)
declare i16 @llvm.vector.reduce.add.i16.v32i16(<32 x i16>)
declare i16 @llvm.vector.reduce.add.i16.v16i16(<16 x i16>)
declare i32 @llvm.vector.reduce.mul.i32.v8i32(<8 x i32>)
declare i32 @llvm.vector.reduce.mul.i32.v4i32(<4 x i32>)
declare i32 @llvm.vector.reduce.and.i32.v8i32(<8 x i32>)
declare i32 @llvm.vector.reduce.and.i32.v4i32(<4 x i32>)
declare i32 @llvm.vector.reduce.or.i32.v8i32(<8 x i32>)
declare i32 @llvm.vector.reduce.or.i32.v4i32(<4 x i32>)
declare i32 @llvm.vector.reduce.xor.i32.v8i32(<8 x i32>)
declare i32 @llvm.vector.reduce.xor.i32.v4i32(<4 x i32>)
declare i32 @llvm.vector.reduce.umin.i32.v8i32(<8 x i32>)
declare i32 @llvm.vector.reduce.umin.i32.v4i32(<4 x i32>)
declare i32 @llvm.vector.reduce.umax.i32.v8i32(<8 x i32>)
declare i32 @llvm.vector.reduce.umax.i32.v4i32(<4 x i32>)
declare i32 @llvm.vector.reduce.smin.i32.v8i32(<8 x i32>)
declare i32 @llvm.vector.reduce.smin.i32.v4i32(<4 x i32>)
declare i32 @llvm.vector.reduce.smax.i32.v8i32(<8 x i32>)
declare i32 @llvm.vector.reduce.smax.i32.v4i32(<4 x i32>)
declare float @llvm.minnum.f32(float, float)
declare float @llvm.maxnum.f32(float, float)
declare i32 @llvm.umin.i32(i32, i32)
declare i32 @llvm.umax.i32(i32, i32)
declare i32 @llvm.smin.i32(i32, i32)
declare i32 @llvm.smax.i32(i32, i32)
|
