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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --check-globals none --filter-out-after "scalar.ph:" --version 5
; RUN: opt -passes='loop-vectorize' -force-vector-width=2 -S < %s | FileCheck %s
;
; Forcing VF=2 to trigger vector code gen
;
; This is a test case that let's vectorizer's code gen to generate
; more than one BasicBlocks in the loop body (emulated masked scatter)
; for those targets that do not support masked scatter. Broadcast
; code generation was previously dependent on loop body being
; a single basic block and this test case exposed incorrect code gen
; resulting in an assert in IL verification.
;
@a = external global [2 x i16], align 1
define void @f1(ptr noalias %b, i1 %c, i32 %start) {
; CHECK-LABEL: define void @f1(
; CHECK-SAME: ptr noalias [[B:%.*]], i1 [[C:%.*]], i32 [[START:%.*]]) {
; CHECK-NEXT: [[ENTRY:.*:]]
; CHECK-NEXT: [[TMP0:%.*]] = add i32 [[START]], 1
; CHECK-NEXT: [[SMIN1:%.*]] = call i32 @llvm.smin.i32(i32 [[START]], i32 1)
; CHECK-NEXT: [[TMP1:%.*]] = sub i32 [[TMP0]], [[SMIN1]]
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[TMP1]], 2
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_SCEVCHECK:.*]]
; CHECK: [[VECTOR_SCEVCHECK]]:
; CHECK-NEXT: [[SMIN:%.*]] = call i32 @llvm.smin.i32(i32 [[START]], i32 1)
; CHECK-NEXT: [[TMP2:%.*]] = sub i32 [[START]], [[SMIN]]
; CHECK-NEXT: [[TMP3:%.*]] = trunc i32 [[START]] to i16
; CHECK-NEXT: [[TMP4:%.*]] = trunc i32 [[TMP2]] to i16
; CHECK-NEXT: [[TMP5:%.*]] = sub i16 [[TMP3]], [[TMP4]]
; CHECK-NEXT: [[TMP6:%.*]] = icmp sgt i16 [[TMP5]], [[TMP3]]
; CHECK-NEXT: [[TMP8:%.*]] = icmp ugt i32 [[TMP2]], 65535
; CHECK-NEXT: [[TMP9:%.*]] = or i1 [[TMP6]], [[TMP8]]
; CHECK-NEXT: br i1 [[TMP9]], label %[[SCALAR_PH]], label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[TMP1]], 2
; CHECK-NEXT: [[N_VEC:%.*]] = sub i32 [[TMP1]], [[N_MOD_VF]]
; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i1> poison, i1 [[C]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i1> [[BROADCAST_SPLATINSERT]], <2 x i1> poison, <2 x i32> zeroinitializer
; CHECK-NEXT: [[TMP10:%.*]] = sub i32 [[START]], [[N_VEC]]
; CHECK-NEXT: [[TMP11:%.*]] = xor <2 x i1> [[BROADCAST_SPLAT]], splat (i1 true)
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[PRED_STORE_CONTINUE3:.*]] ]
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = sub i32 [[START]], [[INDEX]]
; CHECK-NEXT: [[TMP12:%.*]] = trunc i32 [[OFFSET_IDX]] to i16
; CHECK-NEXT: [[TMP13:%.*]] = extractelement <2 x i1> [[TMP11]], i32 0
; CHECK-NEXT: br i1 [[TMP13]], label %[[PRED_STORE_IF:.*]], label %[[PRED_STORE_CONTINUE:.*]]
; CHECK: [[PRED_STORE_IF]]:
; CHECK-NEXT: store i32 10, ptr [[B]], align 1
; CHECK-NEXT: br label %[[PRED_STORE_CONTINUE]]
; CHECK: [[PRED_STORE_CONTINUE]]:
; CHECK-NEXT: [[TMP14:%.*]] = extractelement <2 x i1> [[TMP11]], i32 1
; CHECK-NEXT: br i1 [[TMP14]], label %[[PRED_STORE_IF2:.*]], label %[[PRED_STORE_CONTINUE3]]
; CHECK: [[PRED_STORE_IF2]]:
; CHECK-NEXT: store i32 10, ptr [[B]], align 1
; CHECK-NEXT: br label %[[PRED_STORE_CONTINUE3]]
; CHECK: [[PRED_STORE_CONTINUE3]]:
; CHECK-NEXT: [[TMP15:%.*]] = getelementptr inbounds [2 x i16], ptr @a, i16 0, i16 [[TMP12]]
; CHECK-NEXT: [[TMP16:%.*]] = getelementptr inbounds i16, ptr [[TMP15]], i32 0
; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds i16, ptr [[TMP16]], i32 -1
; CHECK-NEXT: store <2 x i16> zeroinitializer, ptr [[TMP17]], align 1
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 2
; CHECK-NEXT: [[TMP18:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP18]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP1]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], [[EXIT:label %.*]], label %[[SCALAR_PH]]
; CHECK: [[SCALAR_PH]]:
;
entry:
br label %loop.header
loop.header:
%iv = phi i32 [ %start, %entry ], [ %iv.next, %loop.latch ]
br i1 %c, label %loop.latch, label %else
else:
store i32 10, ptr %b, align 1
br label %loop.latch
loop.latch:
%t = trunc i32 %iv to i16
%arrayidx = getelementptr inbounds [2 x i16], ptr @a, i16 0, i16 %t
store i16 0, ptr %arrayidx, align 1
%iv.next = add nsw i32 %iv, -1
%cmp = icmp sgt i32 %iv, 1
br i1 %cmp, label %loop.header, label %exit
exit:
ret void
}
define void @f2(ptr noalias %b, i1 %c, i32 %start) {
; CHECK-LABEL: define void @f2(
; CHECK-SAME: ptr noalias [[B:%.*]], i1 [[C:%.*]], i32 [[START:%.*]]) {
; CHECK-NEXT: [[ENTRY:.*:]]
; CHECK-NEXT: [[TMP0:%.*]] = add i32 [[START]], 1
; CHECK-NEXT: [[SMIN1:%.*]] = call i32 @llvm.smin.i32(i32 [[START]], i32 1)
; CHECK-NEXT: [[TMP1:%.*]] = sub i32 [[TMP0]], [[SMIN1]]
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[TMP1]], 2
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_SCEVCHECK:.*]]
; CHECK: [[VECTOR_SCEVCHECK]]:
; CHECK-NEXT: [[SMIN:%.*]] = call i32 @llvm.smin.i32(i32 [[START]], i32 1)
; CHECK-NEXT: [[TMP2:%.*]] = sub i32 [[START]], [[SMIN]]
; CHECK-NEXT: [[TMP3:%.*]] = trunc i32 [[START]] to i16
; CHECK-NEXT: [[TMP4:%.*]] = trunc i32 [[TMP2]] to i16
; CHECK-NEXT: [[TMP5:%.*]] = sub i16 [[TMP3]], [[TMP4]]
; CHECK-NEXT: [[TMP6:%.*]] = icmp sgt i16 [[TMP5]], [[TMP3]]
; CHECK-NEXT: [[TMP8:%.*]] = icmp ugt i32 [[TMP2]], 65535
; CHECK-NEXT: [[TMP9:%.*]] = or i1 [[TMP6]], [[TMP8]]
; CHECK-NEXT: br i1 [[TMP9]], label %[[SCALAR_PH]], label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i32 [[TMP1]], 2
; CHECK-NEXT: [[N_VEC:%.*]] = sub i32 [[TMP1]], [[N_MOD_VF]]
; CHECK-NEXT: [[TMP10:%.*]] = sub i32 [[START]], [[N_VEC]]
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i32 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = sub i32 [[START]], [[INDEX]]
; CHECK-NEXT: [[TMP11:%.*]] = trunc i32 [[OFFSET_IDX]] to i16
; CHECK-NEXT: [[TMP12:%.*]] = getelementptr inbounds [2 x i16], ptr @a, i16 0, i16 [[TMP11]]
; CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds i16, ptr [[TMP12]], i32 0
; CHECK-NEXT: [[TMP14:%.*]] = getelementptr inbounds i16, ptr [[TMP13]], i32 -1
; CHECK-NEXT: store <2 x i16> zeroinitializer, ptr [[TMP14]], align 1
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 2
; CHECK-NEXT: [[TMP15:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP15]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP1]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], [[EXIT:label %.*]], label %[[SCALAR_PH]]
; CHECK: [[SCALAR_PH]]:
;
entry:
br label %loop.header
loop.header:
%iv = phi i32 [ %start, %entry ], [ %iv.next, %loop.latch ]
br i1 %c, label %loop.latch, label %else
else:
%1 = load i32, ptr %b, align 1
br label %loop.latch
loop.latch:
%2 = trunc i32 %iv to i16
%arrayidx = getelementptr inbounds [2 x i16], ptr @a, i16 0, i16 %2
store i16 0, ptr %arrayidx, align 1
%iv.next = add nsw i32 %iv, -1
%cmp = icmp sgt i32 %iv, 1
br i1 %cmp, label %loop.header, label %exit
exit:
ret void
}
|
