pr43166-fold-tail-by-masking.ll
5.26 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
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
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -loop-vectorize -force-vector-width=4 -S | FileCheck %s
; Test cases below are reduced (and slightly modified) reproducers based on a
; problem seen when compiling a C program like this:
;
; #include <stdint.h>
; #include <stdio.h>
;
; int y = 0;
; int b = 1;
; int d = 1;
;
; int main() {
; #pragma clang loop vectorize_width(4)
; for (int i = 0; i < 3; ++i) {
; b = (y == 0) ? d : (d / y);
; }
;
; if (b == 1)
; printf("GOOD!\n");
; else
; printf("BAD!\n");
; }
;
; When compiled+executed using
; build-all/bin/clang -O1 lv-bug.c && ./a.out
; the result is "GOOD!"
;
; When compiled+executed using
; build-all/bin/clang -O1 lv-bug.c -fvectorize && ./a.out
; the result is "BAD!"
; This test case miscompiled with clang 8.0.0 (see PR43166), now we get
; loop not vectorized: Cannot fold tail by masking in the presence of live outs.
; instead.
define i64 @test1(i64 %y) {
; CHECK-LABEL: @test1(
; CHECK-NEXT: entry:
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.body:
; CHECK-NEXT: [[I:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[INC:%.*]], [[COND_END:%.*]] ]
; CHECK-NEXT: [[CMP:%.*]] = icmp eq i64 [[Y:%.*]], 0
; CHECK-NEXT: br i1 [[CMP]], label [[COND_END]], label [[COND_FALSE:%.*]]
; CHECK: cond.false:
; CHECK-NEXT: [[DIV:%.*]] = xor i64 3, [[Y]]
; CHECK-NEXT: br label [[COND_END]]
; CHECK: cond.end:
; CHECK-NEXT: [[COND:%.*]] = phi i64 [ [[DIV]], [[COND_FALSE]] ], [ 77, [[FOR_BODY]] ]
; CHECK-NEXT: [[INC]] = add nuw nsw i32 [[I]], 1
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[INC]], 3
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]]
; CHECK: for.cond.cleanup:
; CHECK-NEXT: [[COND_LCSSA:%.*]] = phi i64 [ [[COND]], [[COND_END]] ]
; CHECK-NEXT: ret i64 [[COND_LCSSA]]
;
entry:
br label %for.body
for.body:
%i = phi i32 [ 0, %entry ], [ %inc, %cond.end ]
%cmp = icmp eq i64 %y, 0
br i1 %cmp, label %cond.end, label %cond.false
cond.false:
%div = xor i64 3, %y
br label %cond.end
cond.end:
%cond = phi i64 [ %div, %cond.false ], [ 77, %for.body ]
%inc = add nuw nsw i32 %i, 1
%exitcond = icmp eq i32 %inc, 3
br i1 %exitcond, label %for.cond.cleanup, label %for.body
for.cond.cleanup:
ret i64 %cond
}
; This test case miscompiled with clang 8.0.0 (see PR43166), now we get
; loop not vectorized: Cannot fold tail by masking in the presence of live outs.
; instead.
define i64 @test2(i64 %y) {
; CHECK-LABEL: @test2(
; CHECK-NEXT: entry:
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.body:
; CHECK-NEXT: [[I:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[INC:%.*]], [[COND_END:%.*]] ]
; CHECK-NEXT: [[CMP:%.*]] = icmp eq i64 [[Y:%.*]], 0
; CHECK-NEXT: br i1 [[CMP]], label [[COND_END]], label [[COND_FALSE:%.*]]
; CHECK: cond.false:
; CHECK-NEXT: br label [[COND_END]]
; CHECK: cond.end:
; CHECK-NEXT: [[COND:%.*]] = phi i64 [ 55, [[COND_FALSE]] ], [ 77, [[FOR_BODY]] ]
; CHECK-NEXT: [[INC]] = add nuw nsw i32 [[I]], 1
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[INC]], 3
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]]
; CHECK: for.cond.cleanup:
; CHECK-NEXT: [[COND_LCSSA:%.*]] = phi i64 [ [[COND]], [[COND_END]] ]
; CHECK-NEXT: ret i64 [[COND_LCSSA]]
;
entry:
br label %for.body
for.body:
%i = phi i32 [ 0, %entry ], [ %inc, %cond.end ]
%cmp = icmp eq i64 %y, 0
br i1 %cmp, label %cond.end, label %cond.false
cond.false:
br label %cond.end
cond.end:
%cond = phi i64 [ 55, %cond.false ], [ 77, %for.body ]
%inc = add nuw nsw i32 %i, 1
%exitcond = icmp eq i32 %inc, 3
br i1 %exitcond, label %for.cond.cleanup, label %for.body
for.cond.cleanup:
ret i64 %cond
}
; This test case miscompiled with clang 8.0.0 (see PR43166), now we get
; loop not vectorized: Cannot fold tail by masking in the presence of live outs.
; instead.
define i32 @test3(i64 %y) {
; CHECK-LABEL: @test3(
; CHECK-NEXT: entry:
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.body:
; CHECK-NEXT: [[I:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[INC:%.*]], [[COND_END:%.*]] ]
; CHECK-NEXT: [[CMP:%.*]] = icmp eq i64 [[Y:%.*]], 0
; CHECK-NEXT: br i1 [[CMP]], label [[COND_END]], label [[COND_FALSE:%.*]]
; CHECK: cond.false:
; CHECK-NEXT: br label [[COND_END]]
; CHECK: cond.end:
; CHECK-NEXT: [[COND:%.*]] = phi i32 [ 55, [[COND_FALSE]] ], [ [[I]], [[FOR_BODY]] ]
; CHECK-NEXT: [[INC]] = add nuw nsw i32 [[I]], 1
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[INC]], 3
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]]
; CHECK: for.cond.cleanup:
; CHECK-NEXT: [[COND_LCSSA:%.*]] = phi i32 [ [[COND]], [[COND_END]] ]
; CHECK-NEXT: ret i32 [[COND_LCSSA]]
;
entry:
br label %for.body
for.body:
%i = phi i32 [ 0, %entry ], [ %inc, %cond.end ]
%cmp = icmp eq i64 %y, 0
br i1 %cmp, label %cond.end, label %cond.false
cond.false:
br label %cond.end
cond.end:
%cond = phi i32 [ 55, %cond.false ], [ %i, %for.body ]
%inc = add nuw nsw i32 %i, 1
%exitcond = icmp eq i32 %inc, 3
br i1 %exitcond, label %for.cond.cleanup, label %for.body
for.cond.cleanup:
ret i32 %cond
}