ordered_doacross_codegen.cpp
11.8 KB
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
// RUN: %clang_cc1 -verify -fopenmp -x c++ -triple x86_64-unknown-unknown -emit-llvm %s -fexceptions -fcxx-exceptions -o - | FileCheck %s
// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp -x c++ -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s
// RUN: %clang_cc1 -verify -fopenmp-simd -x c++ -triple x86_64-unknown-unknown -emit-llvm %s -fexceptions -fcxx-exceptions -o - | FileCheck --check-prefix SIMD-ONLY0 %s
// RUN: %clang_cc1 -fopenmp-simd -x c++ -std=c++11 -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp-simd -x c++ -triple x86_64-unknown-unknown -fexceptions -fcxx-exceptions -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck --check-prefix SIMD-ONLY0 %s
// SIMD-ONLY0-NOT: {{__kmpc|__tgt}}
// expected-no-diagnostics
#ifndef HEADER
#define HEADER
// CHECK: [[KMP_DIM:%.+]] = type { i64, i64, i64 }
extern int n;
int a[10], b[10], c[10], d[10];
void foo();
// CHECK-LABEL:bar
void bar() {
int i,j;
// CHECK: call void @__kmpc_doacross_init(
// CHECK: call void @__kmpc_doacross_fini(
#pragma omp parallel for ordered(2)
for (i = 0; i < n; ++i)
for (j = 0; j < n; ++j)
a[i] = b[i] + 1;
// CHECK: call void @__kmpc_doacross_init(
// CHECK: call void @__kmpc_doacross_fini(
#pragma omp for collapse(2) ordered(2)
for (int i = 0; i < n; i++)
for (int j = 0; j < n; j++)
;
}
// CHECK-LABEL: @main()
int main() {
int i;
// CHECK: [[DIMS:%.+]] = alloca [1 x [[KMP_DIM]]],
// CHECK: [[GTID:%.+]] = call i32 @__kmpc_global_thread_num([[IDENT:%.+]])
// CHECK: icmp
// CHECK-NEXT: br i1 %
// CHECK: [[CAST:%.+]] = bitcast [1 x [[KMP_DIM]]]* [[DIMS]] to i8*
// CHECK: call void @llvm.memset.p0i8.i64(i8* align 8 [[CAST]], i8 0, i64 24, i1 false)
// CHECK: [[DIM:%.+]] = getelementptr inbounds [1 x [[KMP_DIM]]], [1 x [[KMP_DIM]]]* [[DIMS]], i64 0, i64 0
// CHECK: getelementptr inbounds [[KMP_DIM]], [[KMP_DIM]]* [[DIM]], i32 0, i32 1
// CHECK: store i64 %{{.+}}, i64* %
// CHECK: getelementptr inbounds [[KMP_DIM]], [[KMP_DIM]]* [[DIM]], i32 0, i32 2
// CHECK: store i64 1, i64* %
// CHECK: [[DIM:%.+]] = getelementptr inbounds [1 x [[KMP_DIM]]], [1 x [[KMP_DIM]]]* [[DIMS]], i64 0, i64 0
// CHECK: [[CAST:%.+]] = bitcast [[KMP_DIM]]* [[DIM]] to i8*
// CHECK: call void @__kmpc_doacross_init([[IDENT]], i32 [[GTID]], i32 1, i8* [[CAST]])
// CHECK: call void @__kmpc_for_static_init_4(%struct.ident_t* @{{.+}}, i32 [[GTID]], i32 33, i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
#pragma omp for ordered(1)
for (int i = 0; i < n; ++i) {
a[i] = b[i] + 1;
foo();
// CHECK: invoke void [[FOO:.+]](
// CHECK: load i32, i32* [[I:%.+]],
// CHECK-NEXT: sub nsw i32 %{{.+}}, 0
// CHECK-NEXT: sdiv i32 %{{.+}}, 1
// CHECK-NEXT: sext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [1 x i64], [1 x i64]* [[CNT:%.+]], i64 0, i64 0
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP]],
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [1 x i64], [1 x i64]* [[CNT]], i64 0, i64 0
// CHECK-NEXT: call void @__kmpc_doacross_post([[IDENT]], i32 [[GTID]], i64* [[TMP]])
#pragma omp ordered depend(source)
c[i] = c[i] + 1;
foo();
// CHECK: invoke void [[FOO]]
// CHECK: load i32, i32* [[I]],
// CHECK-NEXT: sub nsw i32 %{{.+}}, 2
// CHECK-NEXT: sub nsw i32 %{{.+}}, 0
// CHECK-NEXT: sdiv i32 %{{.+}}, 1
// CHECK-NEXT: sext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [1 x i64], [1 x i64]* [[CNT:%.+]], i64 0, i64 0
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP]],
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [1 x i64], [1 x i64]* [[CNT]], i64 0, i64 0
// CHECK-NEXT: call void @__kmpc_doacross_wait([[IDENT]], i32 [[GTID]], i64* [[TMP]])
#pragma omp ordered depend(sink : i - 2)
d[i] = a[i - 2];
}
// CHECK: landingpad
// CHECK: call void @__kmpc_doacross_fini([[IDENT]], i32 [[GTID]])
// CHECK: br label %
// CHECK: call void @__kmpc_for_static_fini(
// CHECK: call void @__kmpc_doacross_fini([[IDENT]], i32 [[GTID]])
// CHECK: ret i32 0
return 0;
}
// CHECK-LABEL: main1
int main1() {
// CHECK: [[DIMS:%.+]] = alloca [1 x [[KMP_DIM]]],
// CHECK: [[GTID:%.+]] = call i32 @__kmpc_global_thread_num([[IDENT:%.+]])
// CHECK: icmp
// CHECK-NEXT: br i1 %
// CHECK: [[CAST:%.+]] = bitcast [1 x [[KMP_DIM]]]* [[DIMS]] to i8*
// CHECK: call void @llvm.memset.p0i8.i64(i8* align 8 [[CAST]], i8 0, i64 24, i1 false)
// CHECK: [[DIM:%.+]] = getelementptr inbounds [1 x [[KMP_DIM]]], [1 x [[KMP_DIM]]]* [[DIMS]], i64 0, i64 0
// CHECK: getelementptr inbounds [[KMP_DIM]], [[KMP_DIM]]* [[DIM]], i32 0, i32 1
// CHECK: store i64 %{{.+}}, i64* %
// CHECK: getelementptr inbounds [[KMP_DIM]], [[KMP_DIM]]* [[DIM]], i32 0, i32 2
// CHECK: store i64 1, i64* %
// CHECK: [[DIM:%.+]] = getelementptr inbounds [1 x [[KMP_DIM]]], [1 x [[KMP_DIM]]]* [[DIMS]], i64 0, i64 0
// CHECK: [[CAST:%.+]] = bitcast [[KMP_DIM]]* [[DIM]] to i8*
// CHECK: call void @__kmpc_doacross_init([[IDENT]], i32 [[GTID]], i32 1, i8* [[CAST]])
// CHECK: call void @__kmpc_for_static_init_4(%struct.ident_t* @{{.+}}, i32 [[GTID]], i32 33, i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
#pragma omp for ordered(1)
for (int i = n; i > 0; --i) {
a[i] = b[i] + 1;
foo();
// CHECK: invoke void [[FOO:.+]](
// CHECK: [[UB_VAL:%.+]] = load i32, i32* [[UB:%.+]],
// CHECK-NEXT: [[I_VAL:%.+]] = load i32, i32* [[I:%.+]],
// CHECK-NEXT: sub i32 [[UB_VAL]], [[I_VAL]]
// CHECK-NEXT: udiv i32 %{{.+}}, 1
// CHECK-NEXT: zext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [1 x i64], [1 x i64]* [[CNT:%.+]], i64 0, i64 0
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP]],
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [1 x i64], [1 x i64]* [[CNT]], i64 0, i64 0
// CHECK-NEXT: call void @__kmpc_doacross_post([[IDENT]], i32 [[GTID]], i64* [[TMP]])
#pragma omp ordered depend(source)
c[i] = c[i] + 1;
foo();
// CHECK: invoke void [[FOO]]
// CHECK: [[UB_VAL:%.+]] = load i32, i32* [[UB]],
// CHECK-NEXT: [[I_VAL:%.+]] = load i32, i32* [[I]],
// CHECK-NEXT: [[SUB:%.+]] = sub nsw i32 [[I_VAL]], 2
// CHECK-NEXT: sub i32 [[UB_VAL]], [[SUB]]
// CHECK-NEXT: udiv i32 %{{.+}}, 1
// CHECK-NEXT: zext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [1 x i64], [1 x i64]* [[CNT:%.+]], i64 0, i64 0
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP]],
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [1 x i64], [1 x i64]* [[CNT]], i64 0, i64 0
// CHECK-NEXT: call void @__kmpc_doacross_wait([[IDENT]], i32 [[GTID]], i64* [[TMP]])
#pragma omp ordered depend(sink : i - 2)
d[i] = a[i - 2];
}
// CHECK: landingpad
// CHECK: call void @__kmpc_doacross_fini([[IDENT]], i32 [[GTID]])
// CHECK: br label %
// CHECK: call void @__kmpc_for_static_fini(
// CHECK: call void @__kmpc_doacross_fini([[IDENT]], i32 [[GTID]])
// CHECK: ret i32 0
return 0;
}
// CHECK: define {{.+}}TestStruct
template <typename T>
struct TestStruct {
static const int M = 10;
static const int N = 20;
T i;
T a[N][M];
T b[N][M];
T foo(T, T);
T bar(T, T, T);
void baz(T, T);
TestStruct() {
// CHECK: [[DIMS:%.+]] = alloca [2 x [[KMP_DIM]]],
// CHECK: [[GTID:%.+]] = call i32 @__kmpc_global_thread_num([[IDENT:%.+]])
// CHECK: [[CAST:%.+]] = bitcast [2 x [[KMP_DIM]]]* [[DIMS]] to i8*
// CHECK: call void @llvm.memset.p0i8.i64(i8* align 8 [[CAST]], i8 0, i64 48, i1 false)
// CHECK: [[DIM:%.+]] = getelementptr inbounds [2 x [[KMP_DIM]]], [2 x [[KMP_DIM]]]* [[DIMS]], i64 0, i64 0
// CHECK: getelementptr inbounds [[KMP_DIM]], [[KMP_DIM]]* [[DIM]], i32 0, i32 1
// CHECK: store i64 10, i64* %
// CHECK: getelementptr inbounds [[KMP_DIM]], [[KMP_DIM]]* [[DIM]], i32 0, i32 2
// CHECK: store i64 1, i64* %
// CHECK: [[DIM:%.+]] = getelementptr inbounds [2 x [[KMP_DIM]]], [2 x [[KMP_DIM]]]* [[DIMS]], i64 0, i64 1
// CHECK: getelementptr inbounds [[KMP_DIM]], [[KMP_DIM]]* [[DIM]], i32 0, i32 1
// CHECK: store i64 %{{.+}}, i64* %
// CHECK: getelementptr inbounds [[KMP_DIM]], [[KMP_DIM]]* [[DIM]], i32 0, i32 2
// CHECK: store i64 1, i64* %
// CHECK: [[DIM:%.+]] = getelementptr inbounds [2 x [[KMP_DIM]]], [2 x [[KMP_DIM]]]* [[DIMS]], i64 0, i64 0
// CHECK: [[CAST:%.+]] = bitcast [[KMP_DIM]]* [[DIM]] to i8*
// CHECK: call void @__kmpc_doacross_init([[IDENT]], i32 [[GTID]], i32 2, i8* [[CAST]])
// CHECK: call void @__kmpc_for_static_init_4(%struct.ident_t* @{{.+}}, i32 [[GTID]], i32 33, i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
#pragma omp for ordered(2)
for (T j = 0; j < M; j++)
for (i = 0; i < n; i += 2) {
a[i][j] = foo(i, j);
// CHECK: invoke {{.+TestStruct.+foo}}
// CHECK: load i32*, i32** %
// CHECK: load i32, i32* %
// CHECK: load i32, i32* %
// CHECK: load i32, i32* [[J:%.+]],
// CHECK-NEXT: sub nsw i32 %{{.+}}, 0
// CHECK-NEXT: sdiv i32 %{{.+}}, 1
// CHECK-NEXT: sext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [2 x i64], [2 x i64]* [[CNT:%.+]], i64 0, i64 0
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP:%.+]],
// CHECK-NEXT: [[I:%.+]] = load i32*, i32** [[I_REF:%.+]],
// CHECK-NEXT: load i32, i32* [[I]],
// CHECK-NEXT: sub nsw i32 %{{.+}}, 2
// CHECK-NEXT: sub nsw i32 %{{.+}}, 0
// CHECK-NEXT: sdiv i32 %{{.+}}, 2
// CHECK-NEXT: sext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [2 x i64], [2 x i64]* [[CNT]], i64 0, i64 1
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP]],
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [2 x i64], [2 x i64]* [[CNT]], i64 0, i64 0
// CHECK-NEXT: call void @__kmpc_doacross_wait([[IDENT]], i32 [[GTID]], i64* [[TMP]])
// CHECK-NEXT: load i32, i32* [[J:%.+]],
// CHECK-NEXT: sub nsw i32 %{{.+}}, 1
// CHECK-NEXT: sub nsw i32 %{{.+}}, 0
// CHECK-NEXT: sdiv i32 %{{.+}}, 1
// CHECK-NEXT: sext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [2 x i64], [2 x i64]* [[CNT:%.+]], i64 0, i64 0
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP:%.+]],
// CHECK-NEXT: [[I:%.+]] = load i32*, i32** [[I_REF]],
// CHECK-NEXT: load i32, i32* [[I]],
// CHECK-NEXT: sub nsw i32 %{{.+}}, 0
// CHECK-NEXT: sdiv i32 %{{.+}}, 2
// CHECK-NEXT: sext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [2 x i64], [2 x i64]* [[CNT]], i64 0, i64 1
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP]],
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [2 x i64], [2 x i64]* [[CNT]], i64 0, i64 0
// CHECK-NEXT: call void @__kmpc_doacross_wait([[IDENT]], i32 [[GTID]], i64* [[TMP]])
#pragma omp ordered depend(sink : j, i - 2) depend(sink : j - 1, i)
b[i][j] = bar(a[i][j], b[i - 1][j], b[i][j - 1]);
// CHECK: invoke {{.+TestStruct.+bar}}
// CHECK: load i32*, i32** %
// CHECK: load i32, i32* %
// CHECK: load i32, i32* %
// CHECK: load i32, i32* [[J]],
// CHECK-NEXT: sub nsw i32 %{{.+}}, 0
// CHECK-NEXT: sdiv i32 %{{.+}}, 1
// CHECK-NEXT: sext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [2 x i64], [2 x i64]* [[CNT:%.+]], i64 0, i64 0
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP:%.+]],
// CHECK-NEXT: [[I:%.+]] = load i32*, i32** [[I_REF]],
// CHECK-NEXT: load i32, i32* [[I]],
// CHECK-NEXT: sub nsw i32 %{{.+}}, 0
// CHECK-NEXT: sdiv i32 %{{.+}}, 2
// CHECK-NEXT: sext i32 %{{.+}} to i64
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [2 x i64], [2 x i64]* [[CNT]], i64 0, i64 1
// CHECK-NEXT: store i64 %{{.+}}, i64* [[TMP]],
// CHECK-NEXT: [[TMP:%.+]] = getelementptr inbounds [2 x i64], [2 x i64]* [[CNT]], i64 0, i64 0
// CHECK-NEXT: call void @__kmpc_doacross_post([[IDENT]], i32 [[GTID]], i64* [[TMP]])
#pragma omp ordered depend(source)
baz(a[i][j], b[i][j]);
}
}
// CHECK: landingpad
// CHECK: call void @__kmpc_doacross_fini([[IDENT]], i32 [[GTID]])
// CHECK: br label %
// CHECK: call void @__kmpc_for_static_fini(
// CHECK: call void @__kmpc_doacross_fini([[IDENT]], i32 [[GTID]])
// CHECK: ret
};
TestStruct<int> s;
#endif // HEADER