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
|
// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -ast-print %s | FileCheck %s
// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -std=c++11 -include-pch %t -fsyntax-only -verify %s -ast-print | FileCheck %s
// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=45 -ast-print %s | FileCheck %s
// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -x c++ -std=c++11 -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -std=c++11 -include-pch %t -fsyntax-only -verify %s -ast-print | FileCheck %s
// expected-no-diagnostics
#ifndef HEADER
#define HEADER
void foo() {}
struct S1 {
S1() : a(0) {}
S1(int v) : a(v) {}
int a;
typedef int type;
};
template <typename T>
class S7 : public T {
protected:
T a;
S7() : a(0) {}
public:
S7(typename T::type v) : a(v) {
#pragma omp parallel for simd private(a) private(this->a) private(T::a)
for (int k = 0; k < a.a; ++k)
++this->a.a;
}
S7 &operator=(S7 &s) {
#pragma omp parallel for simd private(a) private(this->a)
for (int k = 0; k < s.a.a; ++k)
++s.a.a;
return *this;
}
};
// CHECK: #pragma omp parallel for simd private(this->a) private(this->a) private(T::a){{$}}
// CHECK: #pragma omp parallel for simd private(this->a) private(this->a)
// CHECK: #pragma omp parallel for simd private(this->a) private(this->a) private(this->S1::a)
class S8 : public S7<S1> {
S8() {}
public:
S8(int v) : S7<S1>(v){
#pragma omp parallel for simd private(a) private(this->a) private(S7 <S1>::a)
for (int k = 0; k < a.a; ++k)
++this->a.a;
}
S8 &operator=(S8 &s) {
#pragma omp parallel for simd private(a) private(this->a)
for (int k = 0; k < s.a.a; ++k)
++s.a.a;
return *this;
}
};
// CHECK: #pragma omp parallel for simd private(this->a) private(this->a) private(this->S7<S1>::a)
// CHECK: #pragma omp parallel for simd private(this->a) private(this->a)
int g_ind = 1;
template<class T, class N> T reduct(T* arr, N num) {
N i;
N ind;
N myind;
T sum = (T)0;
// CHECK: T sum = (T)0;
#pragma omp parallel for simd private(myind, g_ind), linear(ind), aligned(arr) if (parallel :num)
// CHECK-NEXT: #pragma omp parallel for simd private(myind,g_ind) linear(ind) aligned(arr) if(parallel: num)
for (i = 0; i < num; ++i) {
myind = ind;
T cur = arr[myind];
ind += g_ind;
sum += cur;
}
}
template<class T> struct S {
S(const T &a)
:m_a(a)
{}
T result(T *v) const {
T res;
T val;
T lin = 0;
// CHECK: T res;
// CHECK: T val;
// CHECK: T lin = 0;
#pragma omp parallel for simd private(val) safelen(7) linear(lin : -5) lastprivate(res) simdlen(5) if(7)
// CHECK-NEXT: #pragma omp parallel for simd private(val) safelen(7) linear(lin: -5) lastprivate(res) simdlen(5) if(7)
for (T i = 7; i < m_a; ++i) {
val = v[i-7] + m_a;
res = val;
lin -= 5;
}
const T clen = 3;
// CHECK: T clen = 3;
#pragma omp parallel for simd safelen(clen-1) simdlen(clen-1) ordered
// CHECK-NEXT: #pragma omp parallel for simd safelen(clen - 1) simdlen(clen - 1) ordered
for(T i = clen+2; i < 20; ++i) {
// CHECK-NEXT: for (T i = clen + 2; i < 20; ++i) {
v[i] = v[v-clen] + 1;
// CHECK-NEXT: v[i] = v[v - clen] + 1;
}
// CHECK-NEXT: }
return res;
}
~S()
{}
T m_a;
};
template<int LEN> struct S2 {
static void func(int n, float *a, float *b, float *c) {
int k1 = 0, k2 = 0;
#pragma omp parallel for simd safelen(LEN) linear(k1,k2:LEN) aligned(a:LEN) simdlen(LEN)
for(int i = 0; i < n; i++) {
c[i] = a[i] + b[i];
c[k1] = a[k1] + b[k1];
c[k2] = a[k2] + b[k2];
k1 = k1 + LEN;
k2 = k2 + LEN;
}
}
};
// S2<4>::func is called below in main.
// CHECK: template<> struct S2<4> {
// CHECK-NEXT: static void func(int n, float *a, float *b, float *c) {
// CHECK-NEXT: int k1 = 0, k2 = 0;
// CHECK-NEXT: #pragma omp parallel for simd safelen(4) linear(k1,k2: 4) aligned(a: 4) simdlen(4)
// CHECK-NEXT: for (int i = 0; i < n; i++) {
// CHECK-NEXT: c[i] = a[i] + b[i];
// CHECK-NEXT: c[k1] = a[k1] + b[k1];
// CHECK-NEXT: c[k2] = a[k2] + b[k2];
// CHECK-NEXT: k1 = k1 + 4;
// CHECK-NEXT: k2 = k2 + 4;
// CHECK-NEXT: }
// CHECK-NEXT: }
int main (int argc, char **argv) {
int b = argc, c, d, e, f, g;
int k1=0,k2=0;
static int *a;
// CHECK: static int *a;
#pragma omp parallel for simd if(parallel :b) ordered
// CHECK-NEXT: #pragma omp parallel for simd if(parallel: b) ordered
for (int i=0; i < 2; ++i)*a=2;
// CHECK-NEXT: for (int i = 0; i < 2; ++i)
// CHECK-NEXT: *a = 2;
#pragma omp parallel
#pragma omp parallel for simd private(argc, b),lastprivate(d,f) collapse(2) aligned(a : 4) ,firstprivate( g ) if(g)
for (int i = 0; i < 10; ++i)
for (int j = 0; j < 10; ++j) {foo(); k1 += 8; k2 += 8;}
// CHECK-NEXT: #pragma omp parallel
// CHECK-NEXT: #pragma omp parallel for simd private(argc,b) lastprivate(d,f) collapse(2) aligned(a: 4) firstprivate(g) if(g)
// CHECK-NEXT: for (int i = 0; i < 10; ++i)
// CHECK-NEXT: for (int j = 0; j < 10; ++j) {
// CHECK-NEXT: foo();
// CHECK-NEXT: k1 += 8;
// CHECK-NEXT: k2 += 8;
// CHECK-NEXT: }
for (int i = 0; i < 10; ++i)foo();
// CHECK-NEXT: for (int i = 0; i < 10; ++i)
// CHECK-NEXT: foo();
const int CLEN = 4;
// CHECK-NEXT: const int CLEN = 4;
#pragma omp parallel for simd aligned(a:CLEN) linear(a:CLEN) safelen(CLEN) collapse( 1 ) simdlen(CLEN)
// CHECK-NEXT: #pragma omp parallel for simd aligned(a: CLEN) linear(a: CLEN) safelen(CLEN) collapse(1) simdlen(CLEN)
for (int i = 0; i < 10; ++i)foo();
// CHECK-NEXT: for (int i = 0; i < 10; ++i)
// CHECK-NEXT: foo();
float arr[16];
S2<4>::func(0,arr,arr,arr);
return (0);
}
#endif
|
