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
| ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-unknown -mcpu=core2 -mattr=+sse2 | FileCheck %s
; FIXME: Ideally we should be able to fold the entire body of @test1 into a
; single paddd instruction. At the moment we produce the sequence
; pshufd+paddq+pshufd. This is fixed with the widening legalization.
define double @test1(double %A) {
; CHECK-LABEL: test1:
; CHECK: # %bb.0:
; CHECK-NEXT: paddd {{.*}}(%rip), %xmm0
; CHECK-NEXT: retq
%1 = bitcast double %A to <2 x i32>
%add = add <2 x i32> %1, <i32 3, i32 5>
%2 = bitcast <2 x i32> %add to double
ret double %2
}
define double @test2(double %A, double %B) {
; CHECK-LABEL: test2:
; CHECK: # %bb.0:
; CHECK-NEXT: paddd %xmm1, %xmm0
; CHECK-NEXT: retq
%1 = bitcast double %A to <2 x i32>
%2 = bitcast double %B to <2 x i32>
%add = add <2 x i32> %1, %2
%3 = bitcast <2 x i32> %add to double
ret double %3
}
define i64 @test3(i64 %A) {
; CHECK-LABEL: test3:
; CHECK: # %bb.0:
; CHECK-NEXT: movq %rdi, %xmm0
; CHECK-NEXT: addps {{.*}}(%rip), %xmm0
; CHECK-NEXT: movq %xmm0, %rax
; CHECK-NEXT: retq
%1 = bitcast i64 %A to <2 x float>
%add = fadd <2 x float> %1, <float 3.0, float 5.0>
%2 = bitcast <2 x float> %add to i64
ret i64 %2
}
; FIXME: Ideally we should be able to fold the entire body of @test4 into a
; single paddd instruction. This is fixed with the widening legalization.
define i64 @test4(i64 %A) {
; CHECK-LABEL: test4:
; CHECK: # %bb.0:
; CHECK-NEXT: movq %rdi, %xmm0
; CHECK-NEXT: paddd {{.*}}(%rip), %xmm0
; CHECK-NEXT: movq %xmm0, %rax
; CHECK-NEXT: retq
%1 = bitcast i64 %A to <2 x i32>
%add = add <2 x i32> %1, <i32 3, i32 5>
%2 = bitcast <2 x i32> %add to i64
ret i64 %2
}
define double @test5(double %A) {
; CHECK-LABEL: test5:
; CHECK: # %bb.0:
; CHECK-NEXT: addps {{.*}}(%rip), %xmm0
; CHECK-NEXT: retq
%1 = bitcast double %A to <2 x float>
%add = fadd <2 x float> %1, <float 3.0, float 5.0>
%2 = bitcast <2 x float> %add to double
ret double %2
}
; FIXME: Ideally we should be able to fold the entire body of @test6 into a
; single paddw instruction. This is fixed with the widening legalization.
define double @test6(double %A) {
; CHECK-LABEL: test6:
; CHECK: # %bb.0:
; CHECK-NEXT: paddw {{.*}}(%rip), %xmm0
; CHECK-NEXT: retq
%1 = bitcast double %A to <4 x i16>
%add = add <4 x i16> %1, <i16 3, i16 4, i16 5, i16 6>
%2 = bitcast <4 x i16> %add to double
ret double %2
}
define double @test7(double %A, double %B) {
; CHECK-LABEL: test7:
; CHECK: # %bb.0:
; CHECK-NEXT: paddw %xmm1, %xmm0
; CHECK-NEXT: retq
%1 = bitcast double %A to <4 x i16>
%2 = bitcast double %B to <4 x i16>
%add = add <4 x i16> %1, %2
%3 = bitcast <4 x i16> %add to double
ret double %3
}
; FIXME: Ideally we should be able to fold the entire body of @test8 into a
; single paddb instruction. At the moment we produce the sequence
; pshufd+paddw+pshufd. This is fixed with the widening legalization.
define double @test8(double %A) {
; CHECK-LABEL: test8:
; CHECK: # %bb.0:
; CHECK-NEXT: paddb {{.*}}(%rip), %xmm0
; CHECK-NEXT: retq
%1 = bitcast double %A to <8 x i8>
%add = add <8 x i8> %1, <i8 3, i8 4, i8 5, i8 6, i8 7, i8 8, i8 9, i8 10>
%2 = bitcast <8 x i8> %add to double
ret double %2
}
define double @test9(double %A, double %B) {
; CHECK-LABEL: test9:
; CHECK: # %bb.0:
; CHECK-NEXT: paddb %xmm1, %xmm0
; CHECK-NEXT: retq
%1 = bitcast double %A to <8 x i8>
%2 = bitcast double %B to <8 x i8>
%add = add <8 x i8> %1, %2
%3 = bitcast <8 x i8> %add to double
ret double %3
}
|