aboutsummaryrefslogtreecommitdiffstats
path: root/test/Transforms/IndVarSimplify/widen-loop-comp.ll
blob: 0930a0c413904a9378caf27c4d78ef4ecfd505cc (plain)
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
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
; RUN: opt < %s -indvars -S | FileCheck %s
target triple = "aarch64--linux-gnu"

; Check the loop exit i32 compare instruction and operand are widened to i64
; instead of truncating IV before its use in the i32 compare instruction.

@idx = common global i32 0, align 4
@e = common global i32 0, align 4
@ptr = common global i32* null, align 8

; CHECK-LABEL: @test1
; CHECK: for.body.lr.ph:
; CHECK: sext i32
; CHECK: for.cond:
; CHECK: icmp slt i64
; CHECK: for.body:
; CHECK: phi i64

define i32 @test1() {
entry:
  store i32 -1, i32* @idx, align 4
  %0 = load i32* @e, align 4
  %cmp4 = icmp slt i32 %0, 0
  br i1 %cmp4, label %for.end.loopexit, label %for.body.lr.ph

for.body.lr.ph:
  %1 = load i32** @ptr, align 8
  %2 = load i32* @e, align 4
  br label %for.body

for.cond:
  %inc = add nsw i32 %i.05, 1
  %cmp = icmp slt i32 %i.05, %2
  br i1 %cmp, label %for.body, label %for.cond.for.end.loopexit_crit_edge

for.body:
  %i.05 = phi i32 [ 0, %for.body.lr.ph ], [ %inc, %for.cond ]
  %idxprom = sext i32 %i.05 to i64
  %arrayidx = getelementptr inbounds i32* %1, i64 %idxprom
  %3 = load i32* %arrayidx, align 4
  %tobool = icmp eq i32 %3, 0
  br i1 %tobool, label %if.then, label %for.cond

if.then:
  %i.05.lcssa = phi i32 [ %i.05, %for.body ]
  store i32 %i.05.lcssa, i32* @idx, align 4
  br label %for.end

for.cond.for.end.loopexit_crit_edge:
  br label %for.end.loopexit

for.end.loopexit:
  br label %for.end

for.end:
  %4 = load i32* @idx, align 4
  ret i32 %4
}

; CHECK-LABEL: @test2
; CHECK: for.body4.us
; CHECK: %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
; CHECK: %cmp2.us = icmp slt i64
; CHECK-NOT: %2 = trunc i64 %indvars.iv.next to i32
; CHECK-NOT: %cmp2.us = icmp slt i32

define void @test2([8 x i8]* %a, i8* %b, i8 %limit) {
entry:
  %conv = zext i8 %limit to i32
  br i1 undef, label %for.cond1.preheader, label %for.cond1.preheader.us

for.cond1.preheader.us:
  %storemerge5.us = phi i32 [ 0, %entry ], [ %inc14.us, %for.inc13.us ]
  br i1 true, label %for.body4.lr.ph.us, label %for.inc13.us

for.inc13.us:
  %inc14.us = add nsw i32 %storemerge5.us, 1
  %cmp.us = icmp slt i32 %inc14.us, 4
  br i1 %cmp.us, label %for.cond1.preheader.us, label %for.end

for.body4.us:
  %storemerge14.us = phi i32 [ 0, %for.body4.lr.ph.us ], [ %inc.us, %for.body4.us ]
  %idxprom.us = sext i32 %storemerge14.us to i64
  %arrayidx6.us = getelementptr inbounds [8 x i8]* %a, i64 %idxprom5.us, i64 %idxprom.us
  %0 = load i8* %arrayidx6.us, align 1
  %idxprom7.us = zext i8 %0 to i64
  %arrayidx8.us = getelementptr inbounds i8* %b, i64 %idxprom7.us
  %1 = load i8* %arrayidx8.us, align 1
  store i8 %1, i8* %arrayidx6.us, align 1
  %inc.us = add nsw i32 %storemerge14.us, 1
  %cmp2.us = icmp slt i32 %inc.us, %conv
  br i1 %cmp2.us, label %for.body4.us, label %for.inc13.us

for.body4.lr.ph.us:
  %idxprom5.us = sext i32 %storemerge5.us to i64
  br label %for.body4.us

for.cond1.preheader:
  %storemerge5 = phi i32 [ 0, %entry ], [ %inc14, %for.inc13 ]
  br i1 false, label %for.inc13, label %for.inc13

for.inc13:
  %inc14 = add nsw i32 %storemerge5, 1
  %cmp = icmp slt i32 %inc14, 4
  br i1 %cmp, label %for.cond1.preheader, label %for.end

for.end:
  ret void
}

; CHECK-LABEL: @test3
; CHECK: sext i32 %b
; CHECK: for.cond:
; CHECK: phi i64
; CHECK: icmp slt i64

define i32 @test3(i32* %a, i32 %b) {
entry:
  br label %for.cond

for.cond:
  %sum.0 = phi i32 [ 0, %entry ], [ %add, %for.body ]
  %i.0 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
  %cmp = icmp slt i32 %i.0, %b
  br i1 %cmp, label %for.body, label %for.end

for.body:
  %idxprom = sext i32 %i.0 to i64
  %arrayidx = getelementptr inbounds i32* %a, i64 %idxprom
  %0 = load i32* %arrayidx, align 4
  %add = add nsw i32 %sum.0, %0
  %inc = add nsw i32 %i.0, 1
  br label %for.cond

for.end:
  ret i32 %sum.0
}

declare i32 @fn1(i8 signext)

; PR21030
; CHECK-LABEL: @test4
; CHECK: for.body:
; CHECK: phi i32
; CHECK: icmp sgt i8

define i32 @test4(i32 %a) {
entry:
  br label %for.body

for.body:
  %c.07 = phi i8 [ -3, %entry ], [ %dec, %for.body ]
  %conv6 = zext i8 %c.07 to i32
  %or = or i32 %a, %conv6
  %conv3 = trunc i32 %or to i8
  %call = call i32 @fn1(i8 signext %conv3)
  %dec = add i8 %c.07, -1
  %cmp = icmp sgt i8 %dec, -14
  br i1 %cmp, label %for.body, label %for.end

for.end:
  ret i32 0
}

; CHECK-LABEL: @test5
; CHECK: zext i32 %b
; CHECK: for.cond:
; CHECK: phi i64
; CHECK: icmp ule i64

define i32 @test5(i32* %a, i32 %b) {
entry:
  br label %for.cond

for.cond:
  %sum.0 = phi i32 [ 0, %entry ], [ %add, %for.body ]
  %i.0 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
  %cmp = icmp ule i32 %i.0, %b
  br i1 %cmp, label %for.body, label %for.end

for.body:
  %idxprom = zext i32 %i.0 to i64
  %arrayidx = getelementptr inbounds i32* %a, i64 %idxprom
  %0 = load i32* %arrayidx, align 4
  %add = add nsw i32 %sum.0, %0
  %inc = add nsw i32 %i.0, 1
  br label %for.cond

for.end:
  ret i32 %sum.0
}