sse_reload_fold.ll
5.15 KB
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 < %s -mtriple=x86_64-linux -mattr=+64bit,+sse3 -print-failed-fuse-candidates -regalloc=basic 2>&1 | FileCheck %s
; CHECK: fail
; CHECK-NOT: fail
declare float @test_f(float %f)
declare double @test_d(double %f)
declare <4 x float> @test_vf(<4 x float> %f)
declare <2 x double> @test_vd(<2 x double> %f)
declare float @llvm.sqrt.f32(float)
declare double @llvm.sqrt.f64(double)
declare <4 x float> @llvm.x86.sse.rsqrt.ps(<4 x float>)
declare <4 x float> @llvm.x86.sse.sqrt.ps(<4 x float>)
declare <4 x float> @llvm.x86.sse.rcp.ps(<4 x float>)
declare <4 x float> @llvm.x86.sse.min.ps(<4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.sse.max.ps(<4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.sse.cmp.ps(<4 x float>, <4 x float>, i8)
declare <4 x float> @llvm.x86.sse3.addsub.ps(<4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.sse3.hadd.ps(<4 x float>, <4 x float>)
declare <4 x float> @llvm.x86.sse3.hsub.ps(<4 x float>, <4 x float>)
declare <2 x double> @llvm.x86.sse2.sqrt.pd(<2 x double>)
declare <2 x double> @llvm.x86.sse2.min.pd(<2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.sse2.max.pd(<2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double>, <2 x double>, i8)
declare <2 x double> @llvm.x86.sse3.addsub.pd(<2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.sse3.hadd.pd(<2 x double>, <2 x double>)
declare <2 x double> @llvm.x86.sse3.hsub.pd(<2 x double>, <2 x double>)
define float @foo(float %f) {
%a = call float @test_f(float %f)
%t = call float @llvm.sqrt.f32(float %f)
ret float %t
}
define double @doo(double %f) {
%a = call double @test_d(double %f)
%t = call double @llvm.sqrt.f64(double %f)
ret double %t
}
define <4 x float> @a0(<4 x float> %f) {
%a = call <4 x float> @test_vf(<4 x float> %f)
%t = call <4 x float> @llvm.x86.sse.rsqrt.ps(<4 x float> %f)
ret <4 x float> %t
}
define <4 x float> @a1(<4 x float> %f) {
%a = call <4 x float> @test_vf(<4 x float> %f)
%t = call <4 x float> @llvm.x86.sse.sqrt.ps(<4 x float> %f)
ret <4 x float> %t
}
define <4 x float> @a2(<4 x float> %f) {
%a = call <4 x float> @test_vf(<4 x float> %f)
%t = call <4 x float> @llvm.x86.sse.rcp.ps(<4 x float> %f)
ret <4 x float> %t
}
define <4 x float> @b3(<4 x float> %f) {
%y = call <4 x float> @test_vf(<4 x float> %f)
%t = call <4 x float> @llvm.x86.sse.min.ps(<4 x float> %y, <4 x float> %f)
ret <4 x float> %t
}
define <4 x float> @b4(<4 x float> %f) {
%y = call <4 x float> @test_vf(<4 x float> %f)
%t = call <4 x float> @llvm.x86.sse.max.ps(<4 x float> %y, <4 x float> %f)
ret <4 x float> %t
}
define <4 x float> @b5(<4 x float> %f) {
%y = call <4 x float> @test_vf(<4 x float> %f)
%t = call <4 x float> @llvm.x86.sse.cmp.ps(<4 x float> %y, <4 x float> %f, i8 7)
ret <4 x float> %t
}
define <4 x float> @b6(<4 x float> %f) {
%y = call <4 x float> @test_vf(<4 x float> %f)
%t = call <4 x float> @llvm.x86.sse3.addsub.ps(<4 x float> %y, <4 x float> %f)
ret <4 x float> %t
}
define <4 x float> @b7(<4 x float> %f) {
%y = call <4 x float> @test_vf(<4 x float> %f)
%t = call <4 x float> @llvm.x86.sse3.hadd.ps(<4 x float> %y, <4 x float> %f)
ret <4 x float> %t
}
define <4 x float> @b8(<4 x float> %f) {
%y = call <4 x float> @test_vf(<4 x float> %f)
%t = call <4 x float> @llvm.x86.sse3.hsub.ps(<4 x float> %y, <4 x float> %f)
ret <4 x float> %t
}
define <2 x double> @c1(<2 x double> %f) {
%a = call <2 x double> @test_vd(<2 x double> %f)
%t = call <2 x double> @llvm.x86.sse2.sqrt.pd(<2 x double> %f)
ret <2 x double> %t
}
define <2 x double> @d3(<2 x double> %f) {
%y = call <2 x double> @test_vd(<2 x double> %f)
%t = call <2 x double> @llvm.x86.sse2.min.pd(<2 x double> %y, <2 x double> %f)
ret <2 x double> %t
}
define <2 x double> @d4(<2 x double> %f) {
%y = call <2 x double> @test_vd(<2 x double> %f)
%t = call <2 x double> @llvm.x86.sse2.max.pd(<2 x double> %y, <2 x double> %f)
ret <2 x double> %t
}
define <2 x double> @d5(<2 x double> %f) {
%y = call <2 x double> @test_vd(<2 x double> %f)
%t = call <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double> %y, <2 x double> %f, i8 7)
ret <2 x double> %t
}
define <2 x double> @d6(<2 x double> %f) {
%y = call <2 x double> @test_vd(<2 x double> %f)
%t = call <2 x double> @llvm.x86.sse3.addsub.pd(<2 x double> %y, <2 x double> %f)
ret <2 x double> %t
}
define <2 x double> @d7(<2 x double> %f) {
%y = call <2 x double> @test_vd(<2 x double> %f)
%t = call <2 x double> @llvm.x86.sse3.hadd.pd(<2 x double> %y, <2 x double> %f)
ret <2 x double> %t
}
define <2 x double> @d8(<2 x double> %f) {
%y = call <2 x double> @test_vd(<2 x double> %f)
%t = call <2 x double> @llvm.x86.sse3.hsub.pd(<2 x double> %y, <2 x double> %f)
ret <2 x double> %t
}
; This one should fail to fuse, but -regalloc=greedy isn't even trying. Instead
; it produces:
; callq test_vd
; movapd (%rsp), %xmm1 # 16-byte Reload
; hsubpd %xmm0, %xmm1
; movapd %xmm1, %xmm0
; addq $24, %rsp
; ret
; RABasic still tries to fold this one.
define <2 x double> @z0(<2 x double> %f) {
%y = call <2 x double> @test_vd(<2 x double> %f)
%t = call <2 x double> @llvm.x86.sse3.hsub.pd(<2 x double> %f, <2 x double> %y)
ret <2 x double> %t
}