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
135
136
137
|
; RUN: opt < %s -scalarrepl -S | FileCheck %s
target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64"
target triple = "x86_64-apple-darwin10.0.0"
define void @test1(<4 x float>* %F, float %f) {
entry:
%G = alloca <4 x float>, align 16 ; <<4 x float>*> [#uses=3]
%tmp = load <4 x float>* %F ; <<4 x float>> [#uses=2]
%tmp3 = fadd <4 x float> %tmp, %tmp ; <<4 x float>> [#uses=1]
store <4 x float> %tmp3, <4 x float>* %G
%G.upgrd.1 = getelementptr <4 x float>* %G, i32 0, i32 0 ; <float*> [#uses=1]
store float %f, float* %G.upgrd.1
%tmp4 = load <4 x float>* %G ; <<4 x float>> [#uses=2]
%tmp6 = fadd <4 x float> %tmp4, %tmp4 ; <<4 x float>> [#uses=1]
store <4 x float> %tmp6, <4 x float>* %F
ret void
; CHECK-LABEL: @test1(
; CHECK-NOT: alloca
; CHECK: %tmp = load <4 x float>* %F
; CHECK: fadd <4 x float> %tmp, %tmp
; CHECK-NEXT: insertelement <4 x float> %tmp3, float %f, i32 0
}
define void @test2(<4 x float>* %F, float %f) {
entry:
%G = alloca <4 x float>, align 16 ; <<4 x float>*> [#uses=3]
%tmp = load <4 x float>* %F ; <<4 x float>> [#uses=2]
%tmp3 = fadd <4 x float> %tmp, %tmp ; <<4 x float>> [#uses=1]
store <4 x float> %tmp3, <4 x float>* %G
%tmp.upgrd.2 = getelementptr <4 x float>* %G, i32 0, i32 2 ; <float*> [#uses=1]
store float %f, float* %tmp.upgrd.2
%tmp4 = load <4 x float>* %G ; <<4 x float>> [#uses=2]
%tmp6 = fadd <4 x float> %tmp4, %tmp4 ; <<4 x float>> [#uses=1]
store <4 x float> %tmp6, <4 x float>* %F
ret void
; CHECK-LABEL: @test2(
; CHECK-NOT: alloca
; CHECK: %tmp = load <4 x float>* %F
; CHECK: fadd <4 x float> %tmp, %tmp
; CHECK-NEXT: insertelement <4 x float> %tmp3, float %f, i32 2
}
define void @test3(<4 x float>* %F, float* %f) {
entry:
%G = alloca <4 x float>, align 16 ; <<4 x float>*> [#uses=2]
%tmp = load <4 x float>* %F ; <<4 x float>> [#uses=2]
%tmp3 = fadd <4 x float> %tmp, %tmp ; <<4 x float>> [#uses=1]
store <4 x float> %tmp3, <4 x float>* %G
%tmp.upgrd.3 = getelementptr <4 x float>* %G, i32 0, i32 2 ; <float*> [#uses=1]
%tmp.upgrd.4 = load float* %tmp.upgrd.3 ; <float> [#uses=1]
store float %tmp.upgrd.4, float* %f
ret void
; CHECK-LABEL: @test3(
; CHECK-NOT: alloca
; CHECK: %tmp = load <4 x float>* %F
; CHECK: fadd <4 x float> %tmp, %tmp
; CHECK-NEXT: extractelement <4 x float> %tmp3, i32 2
}
define void @test4(<4 x float>* %F, float* %f) {
entry:
%G = alloca <4 x float>, align 16 ; <<4 x float>*> [#uses=2]
%tmp = load <4 x float>* %F ; <<4 x float>> [#uses=2]
%tmp3 = fadd <4 x float> %tmp, %tmp ; <<4 x float>> [#uses=1]
store <4 x float> %tmp3, <4 x float>* %G
%G.upgrd.5 = getelementptr <4 x float>* %G, i32 0, i32 0 ; <float*> [#uses=1]
%tmp.upgrd.6 = load float* %G.upgrd.5 ; <float> [#uses=1]
store float %tmp.upgrd.6, float* %f
ret void
; CHECK-LABEL: @test4(
; CHECK-NOT: alloca
; CHECK: %tmp = load <4 x float>* %F
; CHECK: fadd <4 x float> %tmp, %tmp
; CHECK-NEXT: extractelement <4 x float> %tmp3, i32 0
}
define i32 @test5(float %X) { ;; should turn into bitcast.
%X_addr = alloca [4 x float]
%X1 = getelementptr [4 x float]* %X_addr, i32 0, i32 2
store float %X, float* %X1
%a = bitcast float* %X1 to i32*
%tmp = load i32* %a
ret i32 %tmp
; CHECK-LABEL: @test5(
; CHECK-NEXT: bitcast float %X to i32
; CHECK-NEXT: ret i32
}
define i64 @test6(<2 x float> %X) {
%X_addr = alloca <2 x float>
store <2 x float> %X, <2 x float>* %X_addr
%P = bitcast <2 x float>* %X_addr to i64*
%tmp = load i64* %P
ret i64 %tmp
; CHECK-LABEL: @test6(
; CHECK: bitcast <2 x float> %X to i64
; CHECK: ret i64
}
%struct.test7 = type { [6 x i32] }
define void @test7() {
entry:
%memtmp = alloca %struct.test7, align 16
%0 = bitcast %struct.test7* %memtmp to <4 x i32>*
store <4 x i32> zeroinitializer, <4 x i32>* %0, align 16
%1 = getelementptr inbounds %struct.test7* %memtmp, i64 0, i32 0, i64 5
store i32 0, i32* %1, align 4
ret void
; CHECK-LABEL: @test7(
; CHECK-NOT: alloca
; CHECK: and i192
}
; When promoting an alloca to a 1-element vector type, instructions that
; produce that same vector type should not be changed to insert one element
; into a new vector. <rdar://problem/14249078>
define <1 x i64> @test8(<1 x i64> %a) {
entry:
%a.addr = alloca <1 x i64>, align 8
%__a = alloca <1 x i64>, align 8
%tmp = alloca <1 x i64>, align 8
store <1 x i64> %a, <1 x i64>* %a.addr, align 8
%0 = load <1 x i64>* %a.addr, align 8
store <1 x i64> %0, <1 x i64>* %__a, align 8
%1 = load <1 x i64>* %__a, align 8
%2 = bitcast <1 x i64> %1 to <8 x i8>
%3 = bitcast <8 x i8> %2 to <1 x i64>
%vshl_n = shl <1 x i64> %3, <i64 4>
store <1 x i64> %vshl_n, <1 x i64>* %tmp
%4 = load <1 x i64>* %tmp
ret <1 x i64> %4
; CHECK-LABEL: @test8(
; CHECK-NOT: alloca
; CHECK-NOT: insertelement
; CHECK: ret <1 x i64>
}
|
