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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
; RUN: opt -mtriple=x86_64-apple-macosx -passes=loop-vectorize,simplifycfg,dce -force-vector-width=2 -force-vector-interleave=1 -S %s | FileCheck %s
define void @test_pr55096(i64 %c, ptr %p) {
; CHECK-LABEL: define void @test_pr55096(
; CHECK-SAME: i64 [[C:%.*]], ptr [[P:%.*]]) {
; CHECK-NEXT: entry:
; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i64> poison, i64 [[C]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i64> [[BROADCAST_SPLATINSERT]], <2 x i64> poison, <2 x i32> zeroinitializer
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH:%.*]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE3:%.*]] ]
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 122, i64 123>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[PRED_STORE_CONTINUE3]] ]
; CHECK-NEXT: [[DOTCAST:%.*]] = trunc i64 [[INDEX]] to i16
; CHECK-NEXT: [[TMP0:%.*]] = mul i16 [[DOTCAST]], 2008
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = add i16 6229, [[TMP0]]
; CHECK-NEXT: [[TMP1:%.*]] = icmp ult <2 x i64> [[VEC_IND]], [[BROADCAST_SPLAT]]
; CHECK-NEXT: [[TMP2:%.*]] = xor <2 x i1> [[TMP1]], <i1 true, i1 true>
; CHECK-NEXT: [[TMP3:%.*]] = extractelement <2 x i1> [[TMP2]], i32 0
; CHECK-NEXT: br i1 [[TMP3]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
; CHECK: pred.store.if:
; CHECK-NEXT: [[TMP4:%.*]] = add i16 [[OFFSET_IDX]], 0
; CHECK-NEXT: [[TMP5:%.*]] = add i16 [[TMP4]], 2008
; CHECK-NEXT: [[TMP6:%.*]] = udiv i16 4943, [[TMP5]]
; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds i16, ptr [[P]], i16 [[TMP6]]
; CHECK-NEXT: store i16 0, ptr [[TMP7]], align 2
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE]]
; CHECK: pred.store.continue:
; CHECK-NEXT: [[TMP9:%.*]] = extractelement <2 x i1> [[TMP2]], i32 1
; CHECK-NEXT: br i1 [[TMP9]], label [[PRED_STORE_IF2:%.*]], label [[PRED_STORE_CONTINUE3]]
; CHECK: pred.store.if2:
; CHECK-NEXT: [[TMP10:%.*]] = add i16 [[OFFSET_IDX]], 2008
; CHECK-NEXT: [[TMP11:%.*]] = add i16 [[TMP10]], 2008
; CHECK-NEXT: [[TMP12:%.*]] = udiv i16 4943, [[TMP11]]
; CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds i16, ptr [[P]], i16 [[TMP12]]
; CHECK-NEXT: store i16 0, ptr [[TMP13]], align 2
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE3]]
; CHECK: pred.store.continue3:
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i64> [[VEC_IND]], <i64 2, i64 2>
; CHECK-NEXT: [[TMP15:%.*]] = icmp eq i64 [[INDEX_NEXT]], 340
; CHECK-NEXT: br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: exit:
; CHECK-NEXT: ret void
;
entry:
br label %loop.header
loop.header:
%iv.1 = phi i64 [ 122, %entry ], [ %iv.1.next, %loop.latch ]
%iv.2 = phi i16 [ 6229, %entry ], [ %iv.2.next, %loop.latch ]
%iv.2.next = add i16 %iv.2, 2008
%cmp = icmp ult i64 %iv.1, %c
br i1 %cmp, label %loop.latch, label %loop.then
loop.then:
%div = udiv i16 4943, %iv.2.next
%gep = getelementptr inbounds i16, ptr %p, i16 %div
store i16 0, ptr %gep, align 2
br label %loop.latch
loop.latch:
%iv.1.next = add nuw nsw i64 %iv.1, 1
%exitcond.not = icmp eq i64 %iv.1.next, 462
br i1 %exitcond.not, label %exit, label %loop.header
exit:
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"}
;.
|
