File: aarch64-neon-vcadd.c

package info (click to toggle)
swiftlang 6.0.3-2
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 2,519,992 kB
  • sloc: cpp: 9,107,863; ansic: 2,040,022; asm: 1,135,751; python: 296,500; objc: 82,456; f90: 60,502; lisp: 34,951; pascal: 19,946; sh: 18,133; perl: 7,482; ml: 4,937; javascript: 4,117; makefile: 3,840; awk: 3,535; xml: 914; fortran: 619; cs: 573; ruby: 573
file content (67 lines) | stat: -rw-r--r-- 1,922 bytes parent folder | download | duplicates (9) 
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
 
// RUN: %clang_cc1 -triple aarch64-linux-gnu -target-feature +neon \
// RUN:  -target-feature +v8.3a -target-feature +fullfp16 -S -emit-llvm -o - %s \
// RUN:  | FileCheck %s

// REQUIRES: aarch64-registered-target || arm-registered-target

#include <arm_neon.h>

void foo16x4_rot90(float16x4_t a, float16x4_t b)
{
// CHECK: call <4 x half> @llvm.aarch64.neon.vcadd.rot90.v4f16
  float16x4_t result = vcadd_rot90_f16(a, b);
}

void foo32x2_rot90(float32x2_t a, float32x2_t b)
{
// CHECK: call <2 x float> @llvm.aarch64.neon.vcadd.rot90.v2f32
  float32x2_t result = vcadd_rot90_f32(a, b);
}

void foo16x8_rot90(float16x8_t a, float16x8_t b)
{
// CHECK: call <8 x half> @llvm.aarch64.neon.vcadd.rot90.v8f16
  float16x8_t result = vcaddq_rot90_f16(a, b);
}

void foo32x4_rot90(float32x4_t a, float32x4_t b)
{
// CHECK: call <4 x float> @llvm.aarch64.neon.vcadd.rot90.v4f32
  float32x4_t result = vcaddq_rot90_f32(a, b);
}

void foo64x2_rot90(float64x2_t a, float64x2_t b)
{
// CHECK: call <2 x double> @llvm.aarch64.neon.vcadd.rot90.v2f64
  float64x2_t result = vcaddq_rot90_f64(a, b);
}

void foo16x4_rot270(float16x4_t a, float16x4_t b)
{
// CHECK: call <4 x half> @llvm.aarch64.neon.vcadd.rot270.v4f16
  float16x4_t result = vcadd_rot270_f16(a, b);
}

void foo32x2_rot270(float32x2_t a, float32x2_t b)
{
// CHECK: call <2 x float> @llvm.aarch64.neon.vcadd.rot270.v2f32
  float32x2_t result = vcadd_rot270_f32(a, b);
}

void foo16x8_rot270(float16x8_t a, float16x8_t b)
{
// CHECK: call <8 x half> @llvm.aarch64.neon.vcadd.rot270.v8f16
  float16x8_t result = vcaddq_rot270_f16(a, b);
}

void foo32x4_rot270(float32x4_t a, float32x4_t b)
{
// CHECK: call <4 x float> @llvm.aarch64.neon.vcadd.rot270.v4f32
  float32x4_t result = vcaddq_rot270_f32(a, b);
}

void foo64x2_rot270(float64x2_t a, float64x2_t b)
{
// CHECK: call <2 x double> @llvm.aarch64.neon.vcadd.rot270.v2f64
  float64x2_t result = vcaddq_rot270_f64(a, b);
}