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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
; RUN: opt -p loop-vectorize -mtriple=x86_64-apple-macosx -mcpu=skylake-avx512 -force-vector-width=4 -force-vector-interleave=1 -S %s | FileCheck %s
; Test case for https://github.com/llvm/llvm-project/issues/102934.
define void @gep_use_in_dead_block(ptr noalias %dst, ptr %src) {
; CHECK-LABEL: define void @gep_use_in_dead_block(
; CHECK-SAME: ptr noalias [[DST:%.*]], ptr [[SRC:%.*]]) #[[ATTR0:[0-9]+]] {
; CHECK-NEXT: [[ENTRY:.*]]:
; CHECK-NEXT: br i1 false, label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP0:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT: [[TMP4:%.*]] = getelementptr i16, ptr [[SRC]], i64 [[TMP0]]
; CHECK-NEXT: [[TMP5:%.*]] = getelementptr i16, ptr [[TMP4]], i32 0
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <4 x i16>, ptr [[TMP5]], align 2
; CHECK-NEXT: [[TMP6:%.*]] = icmp eq <4 x i16> [[WIDE_LOAD]], splat (i16 10)
; CHECK-NEXT: [[TMP7:%.*]] = xor <4 x i1> [[TMP6]], splat (i1 true)
; CHECK-NEXT: [[TMP8:%.*]] = getelementptr i16, ptr [[DST]], i64 [[TMP0]]
; CHECK-NEXT: [[TMP12:%.*]] = getelementptr i16, ptr [[TMP8]], i32 0
; CHECK-NEXT: call void @llvm.masked.store.v4i16.p0(<4 x i16> zeroinitializer, ptr [[TMP12]], i32 2, <4 x i1> [[TMP7]])
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT: [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], 96
; CHECK-NEXT: br i1 [[TMP13]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: br i1 false, label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK: [[SCALAR_PH]]:
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ 96, %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT: br label %[[LOOP_HEADER:.*]]
; CHECK: [[LOOP_HEADER]]:
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP_LATCH:.*]] ]
; CHECK-NEXT: [[GEP_SRC:%.*]] = getelementptr i16, ptr [[SRC]], i64 [[IV]]
; CHECK-NEXT: [[L:%.*]] = load i16, ptr [[GEP_SRC]], align 2
; CHECK-NEXT: [[C:%.*]] = icmp eq i16 [[L]], 10
; CHECK-NEXT: br i1 [[C]], label %[[LOOP_LATCH]], label %[[THEN:.*]]
; CHECK: [[THEN]]:
; CHECK-NEXT: [[GEP_DST:%.*]] = getelementptr i16, ptr [[DST]], i64 [[IV]]
; CHECK-NEXT: store i16 0, ptr [[GEP_DST]], align 2
; CHECK-NEXT: br label %[[LOOP_LATCH]]
; CHECK: [[DEAD:.*]]:
; CHECK-NEXT: store i16 0, ptr [[GEP_DST]], align 2
; CHECK-NEXT: br label %[[DEAD]]
; CHECK: [[LOOP_LATCH]]:
; CHECK-NEXT: [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT: [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 99
; CHECK-NEXT: br i1 [[EC]], label %[[EXIT]], label %[[LOOP_HEADER]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: ret void
;
entry:
br label %loop.header
loop.header:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.latch ]
%gep.src = getelementptr i16, ptr %src, i64 %iv
%l = load i16, ptr %gep.src
%c = icmp eq i16 %l, 10
br i1 %c, label %loop.latch, label %then
then:
%gep.dst = getelementptr i16, ptr %dst, i64 %iv
store i16 0, ptr %gep.dst, align 2
br label %loop.latch
dead:
store i16 0, ptr %gep.dst, align 2
br label %dead
loop.latch:
%iv.next = add i64 %iv, 1
%ec = icmp eq i64 %iv.next, 99
br i1 %ec, label %exit, label %loop.header
exit:
ret void
}
define void @gep_use_outside_loop(ptr noalias %dst, ptr %src) {
; CHECK-LABEL: define void @gep_use_outside_loop(
; CHECK-SAME: ptr noalias [[DST:%.*]], ptr [[SRC:%.*]]) #[[ATTR0]] {
; CHECK-NEXT: [[ENTRY:.*]]:
; CHECK-NEXT: br i1 false, label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 0, i64 1, i64 2, i64 3>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP0:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i16, ptr [[DST]], <4 x i64> [[VEC_IND]]
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr i16, ptr [[SRC]], i64 [[TMP0]]
; CHECK-NEXT: [[TMP3:%.*]] = getelementptr i16, ptr [[TMP2]], i32 0
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <4 x i16>, ptr [[TMP3]], align 2
; CHECK-NEXT: [[TMP4:%.*]] = icmp eq <4 x i16> [[WIDE_LOAD]], splat (i16 10)
; CHECK-NEXT: [[TMP5:%.*]] = xor <4 x i1> [[TMP4]], splat (i1 true)
; CHECK-NEXT: [[TMP6:%.*]] = extractelement <4 x ptr> [[TMP1]], i32 0
; CHECK-NEXT: [[TMP7:%.*]] = getelementptr i16, ptr [[TMP6]], i32 0
; CHECK-NEXT: call void @llvm.masked.store.v4i16.p0(<4 x i16> zeroinitializer, ptr [[TMP7]], i32 2, <4 x i1> [[TMP5]])
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <4 x i64> [[VEC_IND]], splat (i64 4)
; CHECK-NEXT: [[TMP8:%.*]] = icmp eq i64 [[INDEX_NEXT]], 96
; CHECK-NEXT: br i1 [[TMP8]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: [[TMP9:%.*]] = extractelement <4 x ptr> [[TMP1]], i32 3
; CHECK-NEXT: br i1 false, label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK: [[SCALAR_PH]]:
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ 96, %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT: br label %[[LOOP_HEADER:.*]]
; CHECK: [[LOOP_HEADER]]:
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP_LATCH:.*]] ]
; CHECK-NEXT: [[GEP_DST:%.*]] = getelementptr i16, ptr [[DST]], i64 [[IV]]
; CHECK-NEXT: [[GEP_SRC:%.*]] = getelementptr i16, ptr [[SRC]], i64 [[IV]]
; CHECK-NEXT: [[L:%.*]] = load i16, ptr [[GEP_SRC]], align 2
; CHECK-NEXT: [[C:%.*]] = icmp eq i16 [[L]], 10
; CHECK-NEXT: br i1 [[C]], label %[[LOOP_LATCH]], label %[[THEN:.*]]
; CHECK: [[THEN]]:
; CHECK-NEXT: store i16 0, ptr [[GEP_DST]], align 2
; CHECK-NEXT: br label %[[LOOP_LATCH]]
; CHECK: [[LOOP_LATCH]]:
; CHECK-NEXT: [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT: [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 99
; CHECK-NEXT: br i1 [[EC]], label %[[EXIT]], label %[[LOOP_HEADER]], !llvm.loop [[LOOP5:![0-9]+]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: [[GEP_DST_LCSSA:%.*]] = phi ptr [ [[GEP_DST]], %[[LOOP_LATCH]] ], [ [[TMP9]], %[[MIDDLE_BLOCK]] ]
; CHECK-NEXT: store i16 0, ptr [[GEP_DST_LCSSA]], align 2
; CHECK-NEXT: ret void
;
entry:
br label %loop.header
loop.header:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.latch ]
%gep.dst = getelementptr i16, ptr %dst, i64 %iv
%gep.src = getelementptr i16, ptr %src, i64 %iv
%l = load i16, ptr %gep.src
%c = icmp eq i16 %l, 10
br i1 %c, label %loop.latch, label %then
then:
store i16 0, ptr %gep.dst, align 2
br label %loop.latch
loop.latch:
%iv.next = add i64 %iv, 1
%ec = icmp eq i64 %iv.next, 99
br i1 %ec, label %exit, label %loop.header
exit:
store i16 0, ptr %gep.dst, align 2
ret void
}
;.
; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
; CHECK: [[LOOP5]] = distinct !{[[LOOP5]], [[META2]], [[META1]]}
;.
|
