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
| ; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
; RUN: llc -march=amdgcn -mcpu=kaveri -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,SICIVI,FUNC %s
; RUN: llc -march=amdgcn -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,SICIVI,FUNC %s
; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,SICIVI,FUNC %s
; RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,GFX9,FUNC %s
; FUNC-LABEL: sin_f32
; EG: MULADD_IEEE *
; EG: FRACT *
; EG: ADD *
; EG: SIN * T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
; EG-NOT: SIN
; GCN: v_mul_f32
; SICIVI: v_fract_f32
; GFX9-NOT: v_fract_f32
; GCN: v_sin_f32
; GCN-NOT: v_sin_f32
define amdgpu_kernel void @sin_f32(float addrspace(1)* %out, float %x) #1 {
%sin = call float @llvm.sin.f32(float %x)
store float %sin, float addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}safe_sin_3x_f32:
; GCN: v_mul_f32
; GCN: v_mul_f32
; SICIVI: v_fract_f32
; GFX9-NOT: v_fract_f32
; GCN: v_sin_f32
; GCN-NOT: v_sin_f32
define amdgpu_kernel void @safe_sin_3x_f32(float addrspace(1)* %out, float %x) #1 {
%y = fmul float 3.0, %x
%sin = call float @llvm.sin.f32(float %y)
store float %sin, float addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}unsafe_sin_3x_f32:
; GCN-NOT: v_add_f32
; GCN: 0x3ef47644
; GCN: v_mul_f32
; SICIVI: v_fract_f32
; GFX9-NOT: v_fract_f32
; GCN: v_sin_f32
; GCN-NOT: v_sin_f32
define amdgpu_kernel void @unsafe_sin_3x_f32(float addrspace(1)* %out, float %x) #2 {
%y = fmul float 3.0, %x
%sin = call float @llvm.sin.f32(float %y)
store float %sin, float addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}safe_sin_2x_f32:
; GCN: v_add_f32
; GCN: v_mul_f32
; SICIVI: v_fract_f32
; GFX9-NOT: v_fract_f32
; GCN: v_sin_f32
; GCN-NOT: v_sin_f32
define amdgpu_kernel void @safe_sin_2x_f32(float addrspace(1)* %out, float %x) #1 {
%y = fmul float 2.0, %x
%sin = call float @llvm.sin.f32(float %y)
store float %sin, float addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}unsafe_sin_2x_f32:
; GCN-NOT: v_add_f32
; GCN: 0x3ea2f983
; GCN: v_mul_f32
; SICIVI: v_fract_f32
; GFX9-NOT: v_fract_f32
; GCN: v_sin_f32
; GCN-NOT: v_sin_f32
define amdgpu_kernel void @unsafe_sin_2x_f32(float addrspace(1)* %out, float %x) #2 {
%y = fmul float 2.0, %x
%sin = call float @llvm.sin.f32(float %y)
store float %sin, float addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}test_safe_2sin_f32:
; GCN: v_add_f32
; GCN: v_mul_f32
; SICIVI: v_fract_f32
; GFX9-NOT: v_fract_f32
; GCN: v_sin_f32
; GCN-NOT: v_sin_f32
define amdgpu_kernel void @test_safe_2sin_f32(float addrspace(1)* %out, float %x) #1 {
%y = fmul float 2.0, %x
%sin = call float @llvm.sin.f32(float %y)
store float %sin, float addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}test_unsafe_2sin_f32:
; GCN: 0x3ea2f983
; GCN: v_mul_f32
; SICIVI: v_fract_f32
; GFX9-NOT: v_fract_f32
; GCN: v_sin_f32
; GCN-NOT: v_sin_f32
define amdgpu_kernel void @test_unsafe_2sin_f32(float addrspace(1)* %out, float %x) #2 {
%y = fmul float 2.0, %x
%sin = call float @llvm.sin.f32(float %y)
store float %sin, float addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}sin_v4f32:
; EG: SIN * T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
; EG: SIN * T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
; EG: SIN * T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
; EG: SIN * T{{[0-9]+\.[XYZW], PV\.[XYZW]}}
; EG-NOT: SIN
; GCN: v_sin_f32
; GCN: v_sin_f32
; GCN: v_sin_f32
; GCN: v_sin_f32
; GCN-NOT: v_sin_f32
define amdgpu_kernel void @sin_v4f32(<4 x float> addrspace(1)* %out, <4 x float> %vx) #1 {
%sin = call <4 x float> @llvm.sin.v4f32( <4 x float> %vx)
store <4 x float> %sin, <4 x float> addrspace(1)* %out
ret void
}
declare float @llvm.sin.f32(float) #0
declare <4 x float> @llvm.sin.v4f32(<4 x float>) #0
attributes #0 = { nounwind readnone }
attributes #1 = { nounwind "unsafe-fp-math"="false" }
attributes #2 = { nounwind "unsafe-fp-math"="true" }
|