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
|
// RUN: %clang_cc1 -x c -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +f16c -emit-llvm -o - -Wall -Werror | FileCheck %s
// RUN: %clang_cc1 -x c -ffreestanding %s -triple=i386-apple-darwin -target-feature +f16c -emit-llvm -o - -Wall -Werror | FileCheck %s
// RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +f16c -emit-llvm -o - -Wall -Werror | FileCheck %s
// RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=i386-apple-darwin -target-feature +f16c -emit-llvm -o - -Wall -Werror | FileCheck %s
// RUN: %clang_cc1 -x c -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +f16c -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s
// RUN: %clang_cc1 -x c -ffreestanding %s -triple=i386-apple-darwin -target-feature +f16c -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s
// RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=x86_64-apple-darwin -target-feature +f16c -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s
// RUN: %clang_cc1 -x c++ -ffreestanding %s -triple=i386-apple-darwin -target-feature +f16c -emit-llvm -o - -Wall -Werror -fexperimental-new-constant-interpreter | FileCheck %s
#include <immintrin.h>
#include "builtin_test_helpers.h"
float test_cvtsh_ss(unsigned short a) {
// CHECK-LABEL: test_cvtsh_ss
// CHECK: [[CONV:%.*]] = fpext half %{{.*}} to float
// CHECK: ret float [[CONV]]
return _cvtsh_ss(a);
}
TEST_CONSTEXPR(_cvtsh_ss(0x0000) == 0.0f);
TEST_CONSTEXPR(_cvtsh_ss(0x4500) == 5.0f);
TEST_CONSTEXPR(_cvtsh_ss(0xC000) == -2.0f);
unsigned short test_cvtss_sh(float a) {
// CHECK-LABEL: test_cvtss_sh
// CHECK: insertelement <4 x float> poison, float %{{.*}}, i32 0
// CHECK: insertelement <4 x float> %{{.*}}, float 0.000000e+00, i32 1
// CHECK: insertelement <4 x float> %{{.*}}, float 0.000000e+00, i32 2
// CHECK: insertelement <4 x float> %{{.*}}, float 0.000000e+00, i32 3
// CHECK: call <8 x i16> @llvm.x86.vcvtps2ph.128(<4 x float> %{{.*}}, i32 0)
// CHECK: extractelement <8 x i16> %{{.*}}, i32 0
return _cvtss_sh(a, 0);
}
TEST_CONSTEXPR(match_m128(
_mm_cvtph_ps(_mm_setr_epi16(0x3C00, 0x4000, 0x4200, 0x4400, 0, 0, 0, 0)),
1.0f, 2.0f, 3.0f, 4.0f
));
__m128 test_mm_cvtph_ps(__m128i a) {
// CHECK-LABEL: test_mm_cvtph_ps
// CHECK: shufflevector <8 x i16> %{{.*}}, <8 x i16> %{{.*}}, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
// CHECK: fpext <4 x half> %{{.*}} to <4 x float>
return _mm_cvtph_ps(a);
}
__m256 test_mm256_cvtph_ps(__m128i a) {
// CHECK-LABEL: test_mm256_cvtph_ps
// CHECK: fpext <8 x half> %{{.*}} to <8 x float>
return _mm256_cvtph_ps(a);
}
TEST_CONSTEXPR(match_m256(
_mm256_cvtph_ps(_mm_setr_epi16(0x3C00, 0x4000, 0x4200, 0x4400, 0x4500, 0x3800, 0xC000, 0x0000)),
1.0f, 2.0f, 3.0f, 4.0f, 5.0f, 0.5f, -2.0f, 0.0f
));
__m128i test_mm_cvtps_ph(__m128 a) {
// CHECK-LABEL: test_mm_cvtps_ph
// CHECK: call <8 x i16> @llvm.x86.vcvtps2ph.128(<4 x float> %{{.*}}, i32 0)
return _mm_cvtps_ph(a, 0);
}
__m128i test_mm256_cvtps_ph(__m256 a) {
// CHECK-LABEL: test_mm256_cvtps_ph
// CHECK: call <8 x i16> @llvm.x86.vcvtps2ph.256(<8 x float> %{{.*}}, i32 0)
return _mm256_cvtps_ph(a, 0);
}
|
