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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -S -mtriple=x86_64-unknown -mattr=+avx -passes=slp-vectorizer | FileCheck %s
define i32 @jumbled-load(ptr noalias nocapture %in, ptr noalias nocapture %inn, ptr noalias nocapture %out) {
; CHECK-LABEL: @jumbled-load(
; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, ptr [[IN:%.*]], align 4
; CHECK-NEXT: [[TMP4:%.*]] = load <4 x i32>, ptr [[INN:%.*]], align 4
; CHECK-NEXT: [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP4]], <4 x i32> poison, <4 x i32> <i32 2, i32 0, i32 3, i32 1>
; CHECK-NEXT: [[TMP5:%.*]] = mul <4 x i32> [[TMP2]], [[SHUFFLE]]
; CHECK-NEXT: [[SHUFFLE1:%.*]] = shufflevector <4 x i32> [[TMP5]], <4 x i32> poison, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
; CHECK-NEXT: store <4 x i32> [[SHUFFLE1]], ptr [[OUT:%.*]], align 4
; CHECK-NEXT: ret i32 undef
;
%load.1 = load i32, ptr %in, align 4
%gep.1 = getelementptr inbounds i32, ptr %in, i64 3
%load.2 = load i32, ptr %gep.1, align 4
%gep.2 = getelementptr inbounds i32, ptr %in, i64 1
%load.3 = load i32, ptr %gep.2, align 4
%gep.3 = getelementptr inbounds i32, ptr %in, i64 2
%load.4 = load i32, ptr %gep.3, align 4
%load.5 = load i32, ptr %inn, align 4
%gep.4 = getelementptr inbounds i32, ptr %inn, i64 2
%load.6 = load i32, ptr %gep.4, align 4
%gep.5 = getelementptr inbounds i32, ptr %inn, i64 3
%load.7 = load i32, ptr %gep.5, align 4
%gep.6 = getelementptr inbounds i32, ptr %inn, i64 1
%load.8 = load i32, ptr %gep.6, align 4
%mul.1 = mul i32 %load.3, %load.5
%mul.2 = mul i32 %load.2, %load.8
%mul.3 = mul i32 %load.4, %load.7
%mul.4 = mul i32 %load.1, %load.6
store i32 %mul.1, ptr %out, align 4
%gep.8 = getelementptr inbounds i32, ptr %out, i64 1
store i32 %mul.2, ptr %gep.8, align 4
%gep.9 = getelementptr inbounds i32, ptr %out, i64 2
store i32 %mul.3, ptr %gep.9, align 4
%gep.10 = getelementptr inbounds i32, ptr %out, i64 3
store i32 %mul.4, ptr %gep.10, align 4
ret i32 undef
}
define i32 @jumbled-load-multiuses(ptr noalias nocapture %in, ptr noalias nocapture %out) {
; CHECK-LABEL: @jumbled-load-multiuses(
; CHECK-NEXT: [[TMP2:%.*]] = load <4 x i32>, ptr [[IN:%.*]], align 4
; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <4 x i32> [[TMP2]], <4 x i32> poison, <4 x i32> <i32 1, i32 2, i32 0, i32 3>
; CHECK-NEXT: [[TMP4:%.*]] = mul <4 x i32> [[TMP2]], [[TMP3]]
; CHECK-NEXT: [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[TMP4]], <4 x i32> poison, <4 x i32> <i32 1, i32 3, i32 2, i32 0>
; CHECK-NEXT: store <4 x i32> [[SHUFFLE]], ptr [[OUT:%.*]], align 4
; CHECK-NEXT: ret i32 undef
;
%load.1 = load i32, ptr %in, align 4
%gep.1 = getelementptr inbounds i32, ptr %in, i64 3
%load.2 = load i32, ptr %gep.1, align 4
%gep.2 = getelementptr inbounds i32, ptr %in, i64 1
%load.3 = load i32, ptr %gep.2, align 4
%gep.3 = getelementptr inbounds i32, ptr %in, i64 2
%load.4 = load i32, ptr %gep.3, align 4
%mul.1 = mul i32 %load.3, %load.4
%mul.2 = mul i32 %load.2, %load.2
%mul.3 = mul i32 %load.4, %load.1
%mul.4 = mul i32 %load.1, %load.3
store i32 %mul.1, ptr %out, align 4
%gep.8 = getelementptr inbounds i32, ptr %out, i64 1
store i32 %mul.2, ptr %gep.8, align 4
%gep.9 = getelementptr inbounds i32, ptr %out, i64 2
store i32 %mul.3, ptr %gep.9, align 4
%gep.10 = getelementptr inbounds i32, ptr %out, i64 3
store i32 %mul.4, ptr %gep.10, align 4
ret i32 undef
}
|
