Skip to content

GitLab

Switch branch/tag
Find file Normal viewHistoryPermalink
demo_ir.ll 19.9 KB
Newer Older
dugupeiwen's avatar
init  
dugupeiwen committed
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 
; ModuleID = 'copy_kernel_1d'
target datalayout = "e-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-n32"
target triple = "amdgcn--amdhsa"

define internal spir_func i32 @hsapy_devfn__5F__5F_main_5F__5F__2E_copy_5F_kernel_5F_1d_24_1_2E_array_28_float32_2C__20_1d_2C__20_C_29__2E_array_28_float32_2C__20_1d_2C__20_C_29_(i8** %.ret, i8* %arg.out.0, i8* %arg.out.1, i64 %arg.out.2, i64 %arg.out.3, float addrspace(4)* %arg.out.4, i64 %arg.out.5.0, i64 %arg.out.6.0, i8* %arg.inp.0, i8* %arg.inp.1, i64 %arg.inp.2, i64 %arg.inp.3, float addrspace(4)* %arg.inp.4, i64 %arg.inp.5.0, i64 %arg.inp.6.0) {
entry:
  %inserted.meminfo = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } undef, i8* %arg.out.0, 0
  %inserted.parent = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.meminfo, i8* %arg.out.1, 1
  %inserted.nitems = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.parent, i64 %arg.out.2, 2
  %inserted.itemsize = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.nitems, i64 %arg.out.3, 3
  %inserted.data = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.itemsize, float addrspace(4)* %arg.out.4, 4
  %.17 = insertvalue [1 x i64] undef, i64 %arg.out.5.0, 0
  %inserted.shape = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.data, [1 x i64] %.17, 5
  %.18 = insertvalue [1 x i64] undef, i64 %arg.out.6.0, 0
  %inserted.strides = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.shape, [1 x i64] %.18, 6
  %inserted.meminfo.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } undef, i8* %arg.inp.0, 0
  %inserted.parent.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.meminfo.1, i8* %arg.inp.1, 1
  %inserted.nitems.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.parent.1, i64 %arg.inp.2, 2
  %inserted.itemsize.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.nitems.1, i64 %arg.inp.3, 3
  %inserted.data.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.itemsize.1, float addrspace(4)* %arg.inp.4, 4
  %.19 = insertvalue [1 x i64] undef, i64 %arg.inp.5.0, 0
  %inserted.shape.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.data.1, [1 x i64] %.19, 5
  %.20 = insertvalue [1 x i64] undef, i64 %arg.inp.6.0, 0
  %inserted.strides.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.shape.1, [1 x i64] %.20, 6
  %out = alloca { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } zeroinitializer, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %out
  %inp = alloca { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } zeroinitializer, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %inp
  %"$0.1" = alloca i8*
  store i8* null, i8** %"$0.1"
  %"$0.2" = alloca i8*
  store i8* null, i8** %"$0.2"
  %"$const0.3" = alloca i64
  store i64 0, i64* %"$const0.3"
  %"$0.4" = alloca i64
  store i64 0, i64* %"$0.4"
  %i = alloca i64
  store i64 0, i64* %i
  %.56 = alloca { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } zeroinitializer, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.56
  %"$0.7" = alloca i64
  store i64 0, i64* %"$0.7"
  %"$0.8" = alloca i1
  store i1 false, i1* %"$0.8"
  %.78 = alloca { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } zeroinitializer, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.78
  %"$30.3" = alloca float
  store float 0.000000e+00, float* %"$30.3"
  %.114 = alloca { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } zeroinitializer, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.114
  %"$const47.1" = alloca i8*
  store i8* null, i8** %"$const47.1"
  %"$47.2" = alloca i8*
  store i8* null, i8** %"$47.2"
  br label %B0

B0:                                               ; preds = %entry
  %.22 = load { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %out
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %out
  %.25 = load { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %inp
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides.1, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %inp
  %.28 = load i8*, i8** %"$0.1"
  store i8* null, i8** %"$0.1"
  %.30 = load i8*, i8** %"$0.1"
  %.32 = load i8*, i8** %"$0.2"
  store i8* null, i8** %"$0.2"
  %.34 = load i8*, i8** %"$0.1"
  store i8* null, i8** %"$0.1"
  %.37 = load i64, i64* %"$const0.3"
  store i64 0, i64* %"$const0.3"
  %.39 = load i64, i64* %"$const0.3"
  %.40 = trunc i64 %.39 to i32
  %.41 = call spir_func i64 @_Z13get_global_idj(i32 %.40)
  %.43 = load i64, i64* %"$0.4"
  store i64 %.41, i64* %"$0.4"
  %.45 = load i64, i64* %"$const0.3"
  store i64 0, i64* %"$const0.3"
  %.47 = load i8*, i8** %"$0.2"
  store i8* null, i8** %"$0.2"
  %.49 = load i64, i64* %"$0.4"
  %.51 = load i64, i64* %i
  store i64 %.49, i64* %i
  %.53 = load i64, i64* %"$0.4"
  store i64 0, i64* %"$0.4"
  %.55 = load { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %out
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %.55, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.56
  %.59 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.56, i32 0, i32 2
  %.60 = load i64, i64* %.59
  %.62 = load i64, i64* %"$0.7"
  store i64 %.60, i64* %"$0.7"
  %.64 = load i64, i64* %i
  %.65 = load i64, i64* %"$0.7"
  %.66 = icmp slt i64 %.64, %.65
  %.68 = load i1, i1* %"$0.8"
  store i1 %.66, i1* %"$0.8"
  %.70 = load i64, i64* %"$0.7"
  store i64 0, i64* %"$0.7"
  %.72 = load i1, i1* %"$0.8"
  br i1 %.72, label %B30, label %B47

B30:                                              ; preds = %B0
  %.74 = load i1, i1* %"$0.8"
  store i1 false, i1* %"$0.8"
  %.76 = load { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %inp
  %.77 = load i64, i64* %i
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %.76, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.78
  %.81 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.78, i32 0, i32 5
  %.82 = getelementptr inbounds [1 x i64], [1 x i64]* %.81, i32 0, i32 0
  %.83 = load i64, i64* %.82, !range !7
  %.84 = insertvalue [1 x i64] undef, i64 %.83, 0
  %.85 = extractvalue [1 x i64] %.84, 0
  %.86 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.78, i32 0, i32 6
  %.87 = load [1 x i64], [1 x i64]* %.86
  %.88 = extractvalue [1 x i64] %.87, 0
  %.89 = icmp slt i64 %.77, 0
  %.90 = add i64 %.77, %.85
  %.91 = select i1 %.89, i64 %.90, i64 %.77
  %.92 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.78, i32 0, i32 5
  %.93 = getelementptr inbounds [1 x i64], [1 x i64]* %.92, i32 0, i32 0
  %.94 = load i64, i64* %.93, !range !7
  %.95 = insertvalue [1 x i64] undef, i64 %.94, 0
  %.96 = extractvalue [1 x i64] %.95, 0
  %.97 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.78, i32 0, i32 6
  %.98 = load [1 x i64], [1 x i64]* %.97
  %.99 = extractvalue [1 x i64] %.98, 0
  %.100 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.78, i32 0, i32 4
  %.101 = load float addrspace(4)*, float addrspace(4)** %.100
  %.102 = mul i64 %.91, 1
  %.103 = add i64 0, %.102
  %.104 = getelementptr float, float addrspace(4)* %.101, i64 %.103
  %.105 = load float, float addrspace(4)* %.104
  %.107 = load float, float* %"$30.3"
  store float %.105, float* %"$30.3"
  %.109 = load { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %inp
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } zeroinitializer, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %inp
  %.111 = load { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %out
  %.112 = load float, float* %"$30.3"
  %.113 = load i64, i64* %i
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %.111, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.114
  %.117 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.114, i32 0, i32 5
  %.118 = getelementptr inbounds [1 x i64], [1 x i64]* %.117, i32 0, i32 0
  %.119 = load i64, i64* %.118, !range !7
  %.120 = insertvalue [1 x i64] undef, i64 %.119, 0
  %.121 = extractvalue [1 x i64] %.120, 0
  %.122 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.114, i32 0, i32 6
  %.123 = load [1 x i64], [1 x i64]* %.122
  %.124 = extractvalue [1 x i64] %.123, 0
  %.125 = icmp slt i64 %.113, 0
  %.126 = add i64 %.113, %.121
  %.127 = select i1 %.125, i64 %.126, i64 %.113
  %.128 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.114, i32 0, i32 5
  %.129 = getelementptr inbounds [1 x i64], [1 x i64]* %.128, i32 0, i32 0
  %.130 = load i64, i64* %.129, !range !7
  %.131 = insertvalue [1 x i64] undef, i64 %.130, 0
  %.132 = extractvalue [1 x i64] %.131, 0
  %.133 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.114, i32 0, i32 6
  %.134 = load [1 x i64], [1 x i64]* %.133
  %.135 = extractvalue [1 x i64] %.134, 0
  %.136 = getelementptr inbounds { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %.114, i32 0, i32 4
  %.137 = load float addrspace(4)*, float addrspace(4)** %.136
  %.138 = mul i64 %.127, 1
  %.139 = add i64 0, %.138
  %.140 = getelementptr float, float addrspace(4)* %.137, i64 %.139
  store float %.112, float addrspace(4)* %.140
  %.142 = load { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %out
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } zeroinitializer, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %out
  %.144 = load i64, i64* %i
  store i64 0, i64* %i
  %.146 = load float, float* %"$30.3"
  store float 0.000000e+00, float* %"$30.3"
  br label %B47

B47:                                              ; preds = %B30, %B0
  %.149 = load { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %out
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } zeroinitializer, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %out
  %.151 = load { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %inp
  store { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } zeroinitializer, { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] }* %inp
  %.153 = load i64, i64* %i
  store i64 0, i64* %i
  %.155 = load i1, i1* %"$0.8"
  store i1 false, i1* %"$0.8"
  %.158 = load i8*, i8** %"$const47.1"
  store i8* null, i8** %"$const47.1"
  %.160 = load i8*, i8** %"$const47.1"
  %.162 = load i8*, i8** %"$47.2"
  store i8* %.160, i8** %"$47.2"
  %.164 = load i8*, i8** %"$const47.1"
  store i8* null, i8** %"$const47.1"
  %.166 = load i8*, i8** %"$47.2"
  store i8* %.166, i8** %.ret
  ret i32 0
}

declare spir_func i64 @_Z13get_global_idj(i32)

define spir_kernel void @hsaPy_hsapy_devfn__5F__5F_main_5F__5F__2E_copy_5F_kernel_5F_1d_24_1_2E_array_28_float32_2C__20_1d_2C__20_C_29__2E_array_28_float32_2C__20_1d_2C__20_C_29_(i8 addrspace(1)* %.1, i8 addrspace(1)* %.2, i64 %.3, i64 %.4, float addrspace(1)* %.5, i64 %.6, i64 %.7, i8 addrspace(1)* %.8, i8 addrspace(1)* %.9, i64 %.10, i64 %.11, float addrspace(1)* %.12, i64 %.13, i64 %.14) {
.16:
  %.17 = addrspacecast i8 addrspace(1)* %.1 to i8*
  %.18 = addrspacecast i8 addrspace(1)* %.2 to i8*
  %.19 = addrspacecast float addrspace(1)* %.5 to float addrspace(4)*
  %.20 = addrspacecast i8 addrspace(1)* %.8 to i8*
  %.21 = addrspacecast i8 addrspace(1)* %.9 to i8*
  %.22 = addrspacecast float addrspace(1)* %.12 to float addrspace(4)*
  %inserted.meminfo = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } undef, i8* %.17, 0
  %inserted.parent = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.meminfo, i8* %.18, 1
  %inserted.nitems = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.parent, i64 %.3, 2
  %inserted.itemsize = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.nitems, i64 %.4, 3
  %inserted.data = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.itemsize, float addrspace(4)* %.19, 4
  %.23 = insertvalue [1 x i64] undef, i64 %.6, 0
  %inserted.shape = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.data, [1 x i64] %.23, 5
  %.24 = insertvalue [1 x i64] undef, i64 %.7, 0
  %inserted.strides = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.shape, [1 x i64] %.24, 6
  %inserted.meminfo.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } undef, i8* %.20, 0
  %inserted.parent.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.meminfo.1, i8* %.21, 1
  %inserted.nitems.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.parent.1, i64 %.10, 2
  %inserted.itemsize.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.nitems.1, i64 %.11, 3
  %inserted.data.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.itemsize.1, float addrspace(4)* %.22, 4
  %.25 = insertvalue [1 x i64] undef, i64 %.13, 0
  %inserted.shape.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.data.1, [1 x i64] %.25, 5
  %.26 = insertvalue [1 x i64] undef, i64 %.14, 0
  %inserted.strides.1 = insertvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.shape.1, [1 x i64] %.26, 6
  %.27 = alloca i8*
  store i8* null, i8** %.27
  %extracted.meminfo = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides, 0
  %extracted.parent = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides, 1
  %extracted.nitems = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides, 2
  %extracted.itemsize = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides, 3
  %extracted.data = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides, 4
  %extracted.shape = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides, 5
  %.29 = extractvalue [1 x i64] %extracted.shape, 0
  %extracted.strides = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides, 6
  %.30 = extractvalue [1 x i64] %extracted.strides, 0
  %extracted.meminfo.1 = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides.1, 0
  %extracted.parent.1 = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides.1, 1
  %extracted.nitems.1 = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides.1, 2
  %extracted.itemsize.1 = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides.1, 3
  %extracted.data.1 = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides.1, 4
  %extracted.shape.1 = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides.1, 5
  %.31 = extractvalue [1 x i64] %extracted.shape.1, 0
  %extracted.strides.1 = extractvalue { i8*, i8*, i64, i64, float addrspace(4)*, [1 x i64], [1 x i64] } %inserted.strides.1, 6
  %.32 = extractvalue [1 x i64] %extracted.strides.1, 0
  %.33 = call spir_func i32 @hsapy_devfn__5F__5F_main_5F__5F__2E_copy_5F_kernel_5F_1d_24_1_2E_array_28_float32_2C__20_1d_2C__20_C_29__2E_array_28_float32_2C__20_1d_2C__20_C_29_(i8** %.27, i8* %extracted.meminfo, i8* %extracted.parent, i64 %extracted.nitems, i64 %extracted.itemsize, float addrspace(4)* %extracted.data, i64 %.29, i64 %.30, i8* %extracted.meminfo.1, i8* %extracted.parent.1, i64 %extracted.nitems.1, i64 %extracted.itemsize.1, float addrspace(4)* %extracted.data.1, i64 %.31, i64 %.32)
  %.34 = icmp eq i32 %.33, 0
  %.35 = icmp eq i32 %.33, -2
  %.36 = or i1 %.34, %.35
  %.37 = xor i1 %.36, true
  %.38 = icmp eq i32 %.33, -1
  %.39 = icmp eq i32 %.33, -3
  %.40 = icmp sge i32 %.33, 1
  %.41 = load i8*, i8** %.27
  ret void
}

!opencl.kernels = !{!0}
!opencl.ocl.version = !{!6}
!opencl.spir.version = !{!6}

!0 = !{void (i8 addrspace(1)*, i8 addrspace(1)*, i64, i64, float addrspace(1)*, i64, i64, i8 addrspace(1)*, i8 addrspace(1)*, i64, i64, float addrspace(1)*, i64, i64)* @hsaPy_hsapy_devfn__5F__5F_main_5F__5F__2E_copy_5F_kernel_5F_1d_24_1_2E_array_28_float32_2C__20_1d_2C__20_C_29__2E_array_28_float32_2C__20_1d_2C__20_C_29_, !1, !2, !3, !4, !5}
!1 = !{!"kernel_arg_addr_space", i32 1, i32 1, i32 0, i32 0, i32 1, i32 0, i32 0, i32 1, i32 1, i32 0, i32 0, i32 1, i32 0, i32 0}
!2 = !{!"kernel_arg_access_qual", !"none", !"none", !"none", !"none", !"none", !"none", !"none", !"none", !"none", !"none", !"none", !"none", !"none", !"none"}
!3 = !{!"kernel_arg_type", !"i8 addrspace(1)* ", !"i8 addrspace(1)* ", !"i64", !"i64", !"float addrspace(1)* ", !"i64", !"i64", !"i8 addrspace(1)* ", !"i8 addrspace(1)* ", !"i64", !"i64", !"float addrspace(1)* ", !"i64", !"i64"}
!4 = !{!"kernel_arg_type_qual", !"", !"", !"", !"", !"", !"", !"", !"", !"", !"", !"", !"", !"", !""}
!5 = !{!"kernel_arg_base_type", !"i8 addrspace(1)* ", !"i8 addrspace(1)* ", !"i64", !"i64", !"float addrspace(1)* ", !"i64", !"i64", !"i8 addrspace(1)* ", !"i8 addrspace(1)* ", !"i64", !"i64", !"float addrspace(1)* ", !"i64", !"i64"}
!6 = !{i32 2, i32 0}
!7 = !{i64 0, i64 9223372036854775807}