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
|
// RUN: %clang_cc1 -verify -fopenmp -x c++ -std=c++11 -ast-print %s | FileCheck %s
// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp -x c++ -std=c++11 -include-pch %t -fsyntax-only -verify %s -ast-print | FileCheck %s
// RUN: %clang_cc1 -verify -fopenmp-simd -x c++ -std=c++11 -ast-print %s | FileCheck %s
// RUN: %clang_cc1 -fopenmp-simd -x c++ -std=c++11 -emit-pch -o %t %s
// RUN: %clang_cc1 -fopenmp-simd -x c++ -std=c++11 -include-pch %t -fsyntax-only -verify %s -ast-print | FileCheck %s
// expected-no-diagnostics
#ifndef HEADER
#define HEADER
#pragma omp declare simd linear(d: 8)
#pragma omp declare simd inbranch simdlen(32)
#pragma omp declare simd notinbranch
void add_1(float *d) __attribute__((cold));
// CHECK: #pragma omp declare simd notinbranch
// CHECK-NEXT: #pragma omp declare simd inbranch simdlen(32)
// CHECK-NEXT: #pragma omp declare simd linear(val(d): 8)
// CHECK-NEXT: void add_1(float *d) __attribute__((cold));
//
#pragma omp declare simd aligned(hp, hp2)
template <class C> void h(C *hp, C *hp2, C *hq, C *lin) {
}
// CHECK: #pragma omp declare simd aligned(hp) aligned(hp2)
// CHECK-NEXT: template <class C> void h(C *hp, C *hp2, C *hq, C *lin) {
// CHECK-NEXT: }
// CHECK: #pragma omp declare simd aligned(hp) aligned(hp2)
// CHECK-NEXT: template<> void h<float>(float *hp, float *hp2, float *hq, float *lin) {
// CHECK-NEXT: }
// CHECK-NEXT: template<> void h<int>(int *hp, int *hp2, int *hq, int *lin) {
// CHECK-NEXT: h((float *)hp, (float *)hp2, (float *)hq, (float *)lin);
// CHECK-NEXT: }
// Explicit specialization with <C=int>.
// Pragmas need to be same, otherwise standard says that's undefined behavior.
#pragma omp declare simd aligned(hp, hp2)
template <>
void h(int *hp, int *hp2, int *hq, int *lin)
{
// Implicit specialization with <C=float>.
// This is special case where the directive is stored by Sema and is
// generated together with the (pending) function instatiation.
h((float*) hp, (float*) hp2, (float*) hq, (float*) lin);
}
class VV {
// CHECK: #pragma omp declare simd uniform(this, a) linear(val(b): a)
// CHECK-NEXT: int add(int a, int b) __attribute__((cold)) {
// CHECK-NEXT: return a + b;
// CHECK-NEXT: }
#pragma omp declare simd uniform(this, a) linear(val(b): a)
int add(int a, int b) __attribute__((cold)) { return a + b; }
// CHECK: #pragma omp declare simd aligned(b: 4) aligned(a) linear(ref(b): 4) linear(val(this)) linear(val(a))
// CHECK-NEXT: float taddpf(float *a, float *&b) {
// CHECK-NEXT: return *a + *b;
// CHECK-NEXT: }
#pragma omp declare simd aligned (b: 4) aligned(a) linear(ref(b): 4) linear(this, a)
float taddpf(float *a, float *&b) { return *a + *b; }
// CHECK: #pragma omp declare simd aligned(b: 8)
// CHECK-NEXT: #pragma omp declare simd linear(uval(c): 8)
// CHECK-NEXT: int tadd(int (&b)[], int &c) {
// CHECK-NEXT: return this->x[b[0]] + b[0];
// CHECK-NEXT: }
#pragma omp declare simd linear(uval(c): 8)
#pragma omp declare simd aligned(b : 8)
int tadd(int (&b)[], int &c) { return x[b[0]] + b[0]; }
private:
int x[10];
};
// CHECK: template <int X, typename T> class TVV {
// CHECK: #pragma omp declare simd simdlen(X)
// CHECK-NEXT: int tadd(int a, int b) {
// CHECK: #pragma omp declare simd aligned(a: X * 2) aligned(b) linear(ref(b): X)
// CHECK-NEXT: float taddpf(float *a, T *&b) {
// CHECK-NEXT: return *a + *b;
// CHECK-NEXT: }
// CHECK: #pragma omp declare simd
// CHECK-NEXT: #pragma omp declare simd
// CHECK-NEXT: int tadd(int b) {
// CHECK-NEXT: return this->x[b] + b;
// CHECK-NEXT: }
// CHECK: }
template <int X, typename T>
class TVV {
public:
// CHECK: template<> class TVV<16, float> {
#pragma omp declare simd simdlen(X)
int tadd(int a, int b) { return a + b; }
// CHECK: #pragma omp declare simd simdlen(16)
// CHECK-NEXT: int tadd(int a, int b);
#pragma omp declare simd aligned(a : X * 2) aligned(b) linear(ref(b): X)
float taddpf(float *a, T *&b) { return *a + *b; }
// CHECK: #pragma omp declare simd aligned(a: 16 * 2) aligned(b) linear(ref(b): 16)
// CHECK-NEXT: float taddpf(float *a, float *&b) {
// CHECK-NEXT: return *a + *b;
// CHECK-NEXT: }
#pragma omp declare simd
#pragma omp declare simd uniform(this, b)
int tadd(int b) { return x[b] + b; }
// CHECK: #pragma omp declare simd uniform(this, b)
// CHECK-NEXT: #pragma omp declare simd
// CHECK-NEXT: int tadd(int b) {
// CHECK-NEXT: return this->x[b] + b;
// CHECK-NEXT: }
private:
int x[X];
};
// CHECK: };
// CHECK: #pragma omp declare simd simdlen(N) aligned(b: N * 2) linear(uval(c): N)
// CHECK: template <int N> void foo(int (&b)[N], float *&c)
// CHECK: #pragma omp declare simd simdlen(64) aligned(b: 64 * 2) linear(uval(c): 64)
// CHECK: template<> void foo<64>(int (&b)[64], float *&c)
#pragma omp declare simd simdlen(N) aligned(b : N * 2) linear(uval(c): N)
template <int N>
void foo(int (&b)[N], float *&c);
// CHECK: TVV<16, float> t16;
TVV<16, float> t16;
void f() {
float a = 1.0f, b = 2.0f;
float *p = &b;
float r = t16.taddpf(&a, p);
int res = t16.tadd(b);
int c[64];
foo(c, p);
}
#endif
|
