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
|
; RUN: opt -mtriple=amdgcn-amd-amdhsa -basic-aa -load-store-vectorizer -S -o - %s | FileCheck %s
target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5"
declare i32 @llvm.amdgcn.workitem.id.x() #1
; CHECK-LABEL: @merge_v2p1i8(
; CHECK: load <2 x i64>
; CHECK: inttoptr i64 %{{[^ ]+}} to i8 addrspace(1)*
; CHECK: inttoptr i64 %{{[^ ]+}} to i8 addrspace(1)*
; CHECK: store <2 x i64> zeroinitializer
define amdgpu_kernel void @merge_v2p1i8(i8 addrspace(1)* addrspace(1)* nocapture %a, i8 addrspace(1)* addrspace(1)* nocapture readonly %b) #0 {
entry:
%a.1 = getelementptr inbounds i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %a, i64 1
%b.1 = getelementptr inbounds i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %b, i64 1
%ld.c = load i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %b, align 4
%ld.c.idx.1 = load i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %b.1, align 4
store i8 addrspace(1)* null, i8 addrspace(1)* addrspace(1)* %a, align 4
store i8 addrspace(1)* null, i8 addrspace(1)* addrspace(1)* %a.1, align 4
ret void
}
; CHECK-LABEL: @merge_v2p3i8(
; CHECK: load <2 x i32>
; CHECK: inttoptr i32 %{{[^ ]+}} to i8 addrspace(3)*
; CHECK: inttoptr i32 %{{[^ ]+}} to i8 addrspace(3)*
; CHECK: store <2 x i32> zeroinitializer
define amdgpu_kernel void @merge_v2p3i8(i8 addrspace(3)* addrspace(3)* nocapture %a, i8 addrspace(3)* addrspace(3)* nocapture readonly %b) #0 {
entry:
%a.1 = getelementptr inbounds i8 addrspace(3)*, i8 addrspace(3)* addrspace(3)* %a, i64 1
%b.1 = getelementptr inbounds i8 addrspace(3)*, i8 addrspace(3)* addrspace(3)* %b, i64 1
%ld.c = load i8 addrspace(3)*, i8 addrspace(3)* addrspace(3)* %b, align 4
%ld.c.idx.1 = load i8 addrspace(3)*, i8 addrspace(3)* addrspace(3)* %b.1, align 4
store i8 addrspace(3)* null, i8 addrspace(3)* addrspace(3)* %a, align 4
store i8 addrspace(3)* null, i8 addrspace(3)* addrspace(3)* %a.1, align 4
ret void
}
; CHECK-LABEL: @merge_load_i64_ptr64(
; CHECK: load <2 x i64>
; CHECK: [[ELT1:%[^ ]+]] = extractelement <2 x i64> %{{[^ ]+}}, i32 1
; CHECK: inttoptr i64 [[ELT1]] to i8 addrspace(1)*
define amdgpu_kernel void @merge_load_i64_ptr64(i64 addrspace(1)* nocapture %a) #0 {
entry:
%a.1 = getelementptr inbounds i64, i64 addrspace(1)* %a, i64 1
%a.1.cast = bitcast i64 addrspace(1)* %a.1 to i8 addrspace(1)* addrspace(1)*
%ld.0 = load i64, i64 addrspace(1)* %a
%ld.1 = load i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %a.1.cast
ret void
}
; CHECK-LABEL: @merge_load_ptr64_i64(
; CHECK: load <2 x i64>
; CHECK: [[ELT0:%[^ ]+]] = extractelement <2 x i64> %{{[^ ]+}}, i32 0
; CHECK: inttoptr i64 [[ELT0]] to i8 addrspace(1)*
define amdgpu_kernel void @merge_load_ptr64_i64(i64 addrspace(1)* nocapture %a) #0 {
entry:
%a.cast = bitcast i64 addrspace(1)* %a to i8 addrspace(1)* addrspace(1)*
%a.1 = getelementptr inbounds i64, i64 addrspace(1)* %a, i64 1
%ld.0 = load i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %a.cast
%ld.1 = load i64, i64 addrspace(1)* %a.1
ret void
}
; CHECK-LABEL: @merge_store_ptr64_i64(
; CHECK: [[ELT0:%[^ ]+]] = ptrtoint i8 addrspace(1)* %ptr0 to i64
; CHECK: insertelement <2 x i64> poison, i64 [[ELT0]], i32 0
; CHECK: store <2 x i64>
define amdgpu_kernel void @merge_store_ptr64_i64(i64 addrspace(1)* nocapture %a, i8 addrspace(1)* %ptr0, i64 %val1) #0 {
entry:
%a.cast = bitcast i64 addrspace(1)* %a to i8 addrspace(1)* addrspace(1)*
%a.1 = getelementptr inbounds i64, i64 addrspace(1)* %a, i64 1
store i8 addrspace(1)* %ptr0, i8 addrspace(1)* addrspace(1)* %a.cast
store i64 %val1, i64 addrspace(1)* %a.1
ret void
}
; CHECK-LABEL: @merge_store_i64_ptr64(
; CHECK: [[ELT1:%[^ ]+]] = ptrtoint i8 addrspace(1)* %ptr1 to i64
; CHECK: insertelement <2 x i64> %{{[^ ]+}}, i64 [[ELT1]], i32 1
; CHECK: store <2 x i64>
define amdgpu_kernel void @merge_store_i64_ptr64(i8 addrspace(1)* addrspace(1)* nocapture %a, i64 %val0, i8 addrspace(1)* %ptr1) #0 {
entry:
%a.1 = getelementptr inbounds i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %a, i64 1
%a.cast = bitcast i8 addrspace(1)* addrspace(1)* %a to i64 addrspace(1)*
store i64 %val0, i64 addrspace(1)* %a.cast
store i8 addrspace(1)* %ptr1, i8 addrspace(1)* addrspace(1)* %a.1
ret void
}
; CHECK-LABEL: @merge_load_i32_ptr32(
; CHECK: load <2 x i32>
; CHECK: [[ELT1:%[^ ]+]] = extractelement <2 x i32> %{{[^ ]+}}, i32 1
; CHECK: inttoptr i32 [[ELT1]] to i8 addrspace(3)*
define amdgpu_kernel void @merge_load_i32_ptr32(i32 addrspace(3)* nocapture %a) #0 {
entry:
%a.1 = getelementptr inbounds i32, i32 addrspace(3)* %a, i32 1
%a.1.cast = bitcast i32 addrspace(3)* %a.1 to i8 addrspace(3)* addrspace(3)*
%ld.0 = load i32, i32 addrspace(3)* %a
%ld.1 = load i8 addrspace(3)*, i8 addrspace(3)* addrspace(3)* %a.1.cast
ret void
}
; CHECK-LABEL: @merge_load_ptr32_i32(
; CHECK: load <2 x i32>
; CHECK: [[ELT0:%[^ ]+]] = extractelement <2 x i32> %{{[^ ]+}}, i32 0
; CHECK: inttoptr i32 [[ELT0]] to i8 addrspace(3)*
define amdgpu_kernel void @merge_load_ptr32_i32(i32 addrspace(3)* nocapture %a) #0 {
entry:
%a.cast = bitcast i32 addrspace(3)* %a to i8 addrspace(3)* addrspace(3)*
%a.1 = getelementptr inbounds i32, i32 addrspace(3)* %a, i32 1
%ld.0 = load i8 addrspace(3)*, i8 addrspace(3)* addrspace(3)* %a.cast
%ld.1 = load i32, i32 addrspace(3)* %a.1
ret void
}
; CHECK-LABEL: @merge_store_ptr32_i32(
; CHECK: [[ELT0:%[^ ]+]] = ptrtoint i8 addrspace(3)* %ptr0 to i32
; CHECK: insertelement <2 x i32> poison, i32 [[ELT0]], i32 0
; CHECK: store <2 x i32>
define amdgpu_kernel void @merge_store_ptr32_i32(i32 addrspace(3)* nocapture %a, i8 addrspace(3)* %ptr0, i32 %val1) #0 {
entry:
%a.cast = bitcast i32 addrspace(3)* %a to i8 addrspace(3)* addrspace(3)*
%a.1 = getelementptr inbounds i32, i32 addrspace(3)* %a, i32 1
store i8 addrspace(3)* %ptr0, i8 addrspace(3)* addrspace(3)* %a.cast
store i32 %val1, i32 addrspace(3)* %a.1
ret void
}
; CHECK-LABEL: @merge_store_i32_ptr32(
; CHECK: [[ELT1:%[^ ]+]] = ptrtoint i8 addrspace(3)* %ptr1 to i32
; CHECK: insertelement <2 x i32> %{{[^ ]+}}, i32 [[ELT1]], i32 1
; CHECK: store <2 x i32>
define amdgpu_kernel void @merge_store_i32_ptr32(i8 addrspace(3)* addrspace(3)* nocapture %a, i32 %val0, i8 addrspace(3)* %ptr1) #0 {
entry:
%a.1 = getelementptr inbounds i8 addrspace(3)*, i8 addrspace(3)* addrspace(3)* %a, i32 1
%a.cast = bitcast i8 addrspace(3)* addrspace(3)* %a to i32 addrspace(3)*
store i32 %val0, i32 addrspace(3)* %a.cast
store i8 addrspace(3)* %ptr1, i8 addrspace(3)* addrspace(3)* %a.1
ret void
}
; CHECK-LABEL: @no_merge_store_ptr32_i64(
; CHECK: store i8 addrspace(3)*
; CHECK: store i64
define amdgpu_kernel void @no_merge_store_ptr32_i64(i64 addrspace(1)* nocapture %a, i8 addrspace(3)* %ptr0, i64 %val1) #0 {
entry:
%a.cast = bitcast i64 addrspace(1)* %a to i8 addrspace(3)* addrspace(1)*
%a.1 = getelementptr inbounds i64, i64 addrspace(1)* %a, i64 1
store i8 addrspace(3)* %ptr0, i8 addrspace(3)* addrspace(1)* %a.cast
store i64 %val1, i64 addrspace(1)* %a.1
ret void
}
; CHECK-LABEL: @no_merge_store_i64_ptr32(
; CHECK: store i64
; CHECK: store i8 addrspace(3)*
define amdgpu_kernel void @no_merge_store_i64_ptr32(i8 addrspace(3)* addrspace(1)* nocapture %a, i64 %val0, i8 addrspace(3)* %ptr1) #0 {
entry:
%a.1 = getelementptr inbounds i8 addrspace(3)*, i8 addrspace(3)* addrspace(1)* %a, i64 1
%a.cast = bitcast i8 addrspace(3)* addrspace(1)* %a to i64 addrspace(1)*
store i64 %val0, i64 addrspace(1)* %a.cast
store i8 addrspace(3)* %ptr1, i8 addrspace(3)* addrspace(1)* %a.1
ret void
}
; CHECK-LABEL: @no_merge_load_i64_ptr32(
; CHECK: load i64,
; CHECK: load i8 addrspace(3)*,
define amdgpu_kernel void @no_merge_load_i64_ptr32(i64 addrspace(1)* nocapture %a) #0 {
entry:
%a.1 = getelementptr inbounds i64, i64 addrspace(1)* %a, i64 1
%a.1.cast = bitcast i64 addrspace(1)* %a.1 to i8 addrspace(3)* addrspace(1)*
%ld.0 = load i64, i64 addrspace(1)* %a
%ld.1 = load i8 addrspace(3)*, i8 addrspace(3)* addrspace(1)* %a.1.cast
ret void
}
; CHECK-LABEL: @no_merge_load_ptr32_i64(
; CHECK: load i8 addrspace(3)*,
; CHECK: load i64,
define amdgpu_kernel void @no_merge_load_ptr32_i64(i64 addrspace(1)* nocapture %a) #0 {
entry:
%a.cast = bitcast i64 addrspace(1)* %a to i8 addrspace(3)* addrspace(1)*
%a.1 = getelementptr inbounds i64, i64 addrspace(1)* %a, i64 1
%ld.0 = load i8 addrspace(3)*, i8 addrspace(3)* addrspace(1)* %a.cast
%ld.1 = load i64, i64 addrspace(1)* %a.1
ret void
}
; XXX - This isn't merged for some reason
; CHECK-LABEL: @merge_v2p1i8_v2p1i8(
; CHECK: load <2 x i8 addrspace(1)*>
; CHECK: load <2 x i8 addrspace(1)*>
; CHECK: store <2 x i8 addrspace(1)*>
; CHECK: store <2 x i8 addrspace(1)*>
define amdgpu_kernel void @merge_v2p1i8_v2p1i8(<2 x i8 addrspace(1)*> addrspace(1)* nocapture noalias %a, <2 x i8 addrspace(1)*> addrspace(1)* nocapture readonly noalias %b) #0 {
entry:
%a.1 = getelementptr inbounds <2 x i8 addrspace(1)*>, <2 x i8 addrspace(1)*> addrspace(1)* %a, i64 1
%b.1 = getelementptr inbounds <2 x i8 addrspace(1)*>, <2 x i8 addrspace(1)*> addrspace(1)* %b, i64 1
%ld.c = load <2 x i8 addrspace(1)*>, <2 x i8 addrspace(1)*> addrspace(1)* %b, align 4
%ld.c.idx.1 = load <2 x i8 addrspace(1)*>, <2 x i8 addrspace(1)*> addrspace(1)* %b.1, align 4
store <2 x i8 addrspace(1)*> zeroinitializer, <2 x i8 addrspace(1)*> addrspace(1)* %a, align 4
store <2 x i8 addrspace(1)*> zeroinitializer, <2 x i8 addrspace(1)*> addrspace(1)* %a.1, align 4
ret void
}
; CHECK-LABEL: @merge_load_ptr64_f64(
; CHECK: load <2 x i64>
; CHECK: [[ELT0:%[^ ]+]] = extractelement <2 x i64> %{{[^ ]+}}, i32 0
; CHECK: [[ELT0_INT:%[^ ]+]] = inttoptr i64 [[ELT0]] to i8 addrspace(1)*
; CHECK: [[ELT1_INT:%[^ ]+]] = extractelement <2 x i64> %{{[^ ]+}}, i32 1
; CHECK: bitcast i64 [[ELT1_INT]] to double
define amdgpu_kernel void @merge_load_ptr64_f64(double addrspace(1)* nocapture %a) #0 {
entry:
%a.cast = bitcast double addrspace(1)* %a to i8 addrspace(1)* addrspace(1)*
%a.1 = getelementptr inbounds double, double addrspace(1)* %a, i64 1
%ld.0 = load i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %a.cast
%ld.1 = load double, double addrspace(1)* %a.1
ret void
}
; CHECK-LABEL: @merge_load_f64_ptr64(
; CHECK: load <2 x i64>
; CHECK: [[ELT0:%[^ ]+]] = extractelement <2 x i64> %{{[^ ]+}}, i32 0
; CHECK: bitcast i64 [[ELT0]] to double
; CHECK: [[ELT1:%[^ ]+]] = extractelement <2 x i64> %{{[^ ]+}}, i32 1
; CHECK: inttoptr i64 [[ELT1]] to i8 addrspace(1)*
define amdgpu_kernel void @merge_load_f64_ptr64(double addrspace(1)* nocapture %a) #0 {
entry:
%a.1 = getelementptr inbounds double, double addrspace(1)* %a, i64 1
%a.1.cast = bitcast double addrspace(1)* %a.1 to i8 addrspace(1)* addrspace(1)*
%ld.0 = load double, double addrspace(1)* %a
%ld.1 = load i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %a.1.cast
ret void
}
; CHECK-LABEL: @merge_store_ptr64_f64(
; CHECK: [[ELT0_INT:%[^ ]+]] = ptrtoint i8 addrspace(1)* %ptr0 to i64
; CHECK: insertelement <2 x i64> poison, i64 [[ELT0_INT]], i32 0
; CHECK: [[ELT1_INT:%[^ ]+]] = bitcast double %val1 to i64
; CHECK: insertelement <2 x i64> %{{[^ ]+}}, i64 [[ELT1_INT]], i32 1
; CHECK: store <2 x i64>
define amdgpu_kernel void @merge_store_ptr64_f64(double addrspace(1)* nocapture %a, i8 addrspace(1)* %ptr0, double %val1) #0 {
entry:
%a.cast = bitcast double addrspace(1)* %a to i8 addrspace(1)* addrspace(1)*
%a.1 = getelementptr inbounds double, double addrspace(1)* %a, i64 1
store i8 addrspace(1)* %ptr0, i8 addrspace(1)* addrspace(1)* %a.cast
store double %val1, double addrspace(1)* %a.1
ret void
}
; CHECK-LABEL: @merge_store_f64_ptr64(
; CHECK: [[ELT0_INT:%[^ ]+]] = bitcast double %val0 to i64
; CHECK: insertelement <2 x i64> poison, i64 [[ELT0_INT]], i32 0
; CHECK: [[ELT1_INT:%[^ ]+]] = ptrtoint i8 addrspace(1)* %ptr1 to i64
; CHECK: insertelement <2 x i64> %{{[^ ]+}}, i64 [[ELT1_INT]], i32 1
; CHECK: store <2 x i64>
define amdgpu_kernel void @merge_store_f64_ptr64(i8 addrspace(1)* addrspace(1)* nocapture %a, double %val0, i8 addrspace(1)* %ptr1) #0 {
entry:
%a.1 = getelementptr inbounds i8 addrspace(1)*, i8 addrspace(1)* addrspace(1)* %a, i64 1
%a.cast = bitcast i8 addrspace(1)* addrspace(1)* %a to double addrspace(1)*
store double %val0, double addrspace(1)* %a.cast
store i8 addrspace(1)* %ptr1, i8 addrspace(1)* addrspace(1)* %a.1
ret void
}
attributes #0 = { nounwind }
attributes #1 = { nounwind readnone }
|
