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
|
// RUN: %clang_cc1 %s -triple=amdgcn-amd-amdhsa -emit-llvm -o - | FileCheck %s
template<typename T, unsigned int n> struct my_vector_base;
template<typename T>
struct my_vector_base<T, 1> {
typedef T Native_vec_ __attribute__((ext_vector_type(1)));
union {
Native_vec_ data;
struct {
T x;
};
};
};
template<typename T, unsigned int rank>
struct my_vector_type : public my_vector_base<T, rank> {
using my_vector_base<T, rank>::data;
using typename my_vector_base<T, rank>::Native_vec_;
template< typename U>
my_vector_type(U x) noexcept
{
for (auto i = 0u; i != rank; ++i) data[i] = x;
}
my_vector_type& operator+=(const my_vector_type& x) noexcept
{
data += x.data;
return *this;
}
};
template<typename T, unsigned int n>
inline
my_vector_type<T, n> operator+(
const my_vector_type<T, n>& x, const my_vector_type<T, n>& y) noexcept
{
return my_vector_type<T, n>{x} += y;
}
using char1 = my_vector_type<char, 1>;
int mane() {
char1 f1{1};
char1 f2{1};
// CHECK: [[TMP:%.+]] = alloca i16
// CHECK: [[COERCE:%.+]] = addrspacecast ptr addrspace(5) [[TMP]] to ptr
// CHECK: call void @llvm.memcpy.p0.p0.i64(ptr align 1 %{{.+}}, ptr align 2 [[COERCE]], i64 1, i1 false)
char1 f3 = f1 + f2;
}
|
