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
|
//@ compile-flags: -C no-prepopulate-passes
#![crate_type = "lib"]
#![feature(repr_simd, intrinsics)]
#![allow(non_camel_case_types)]
#[repr(simd)]
#[derive(Copy, Clone, PartialEq, Debug)]
pub struct f32x2(pub [f32; 2]);
#[repr(simd)]
#[derive(Copy, Clone, PartialEq, Debug)]
pub struct f32x4(pub [f32; 4]);
#[repr(simd)]
#[derive(Copy, Clone, PartialEq, Debug)]
pub struct f32x8(pub [f32; 8]);
#[repr(simd)]
#[derive(Copy, Clone, PartialEq, Debug)]
pub struct f32x16(pub [f32; 16]);
extern "rust-intrinsic" {
fn simd_fsqrt<T>(x: T) -> T;
}
// CHECK-LABEL: @fsqrt_32x2
#[no_mangle]
pub unsafe fn fsqrt_32x2(a: f32x2) -> f32x2 {
// CHECK: call <2 x float> @llvm.sqrt.v2f32
simd_fsqrt(a)
}
// CHECK-LABEL: @fsqrt_32x4
#[no_mangle]
pub unsafe fn fsqrt_32x4(a: f32x4) -> f32x4 {
// CHECK: call <4 x float> @llvm.sqrt.v4f32
simd_fsqrt(a)
}
// CHECK-LABEL: @fsqrt_32x8
#[no_mangle]
pub unsafe fn fsqrt_32x8(a: f32x8) -> f32x8 {
// CHECK: call <8 x float> @llvm.sqrt.v8f32
simd_fsqrt(a)
}
// CHECK-LABEL: @fsqrt_32x16
#[no_mangle]
pub unsafe fn fsqrt_32x16(a: f32x16) -> f32x16 {
// CHECK: call <16 x float> @llvm.sqrt.v16f32
simd_fsqrt(a)
}
#[repr(simd)]
#[derive(Copy, Clone, PartialEq, Debug)]
pub struct f64x2(pub [f64; 2]);
#[repr(simd)]
#[derive(Copy, Clone, PartialEq, Debug)]
pub struct f64x4(pub [f64; 4]);
#[repr(simd)]
#[derive(Copy, Clone, PartialEq, Debug)]
pub struct f64x8(pub [f64; 8]);
// CHECK-LABEL: @fsqrt_64x4
#[no_mangle]
pub unsafe fn fsqrt_64x4(a: f64x4) -> f64x4 {
// CHECK: call <4 x double> @llvm.sqrt.v4f64
simd_fsqrt(a)
}
// CHECK-LABEL: @fsqrt_64x2
#[no_mangle]
pub unsafe fn fsqrt_64x2(a: f64x2) -> f64x2 {
// CHECK: call <2 x double> @llvm.sqrt.v2f64
simd_fsqrt(a)
}
// CHECK-LABEL: @fsqrt_64x8
#[no_mangle]
pub unsafe fn fsqrt_64x8(a: f64x8) -> f64x8 {
// CHECK: call <8 x double> @llvm.sqrt.v8f64
simd_fsqrt(a)
}
|
