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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=x86_64-- -mcpu=corei7 | FileCheck %s
define void @foo(ptr %v0, ptr readonly %v1) {
; CHECK-LABEL: @foo(
; CHECK-NEXT: [[T14:%.*]] = getelementptr inbounds i32, ptr [[V1:%.*]], i64 4
; CHECK-NEXT: [[T142:%.*]] = getelementptr inbounds i64, ptr [[V1]], i64 8
; CHECK-NEXT: [[T222:%.*]] = getelementptr inbounds i64, ptr [[V1]], i64 10
; CHECK-NEXT: [[T21:%.*]] = getelementptr inbounds i32, ptr [[V0:%.*]], i64 4
; CHECK-NEXT: [[T212:%.*]] = getelementptr inbounds i64, ptr [[V0]], i64 8
; CHECK-NEXT: [[T292:%.*]] = getelementptr inbounds i64, ptr [[V0]], i64 10
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, ptr [[T14]], align 4
; CHECK-NEXT: [[TMP2:%.*]] = add nsw <4 x i32> [[TMP1]], <i32 4, i32 4, i32 6, i32 7>
; CHECK-NEXT: [[TMP3:%.*]] = load <2 x i64>, ptr [[T142]], align 8
; CHECK-NEXT: [[TMP4:%.*]] = add nsw <2 x i64> [[TMP3]], <i64 4, i64 4>
; CHECK-NEXT: [[TMP5:%.*]] = load <2 x i64>, ptr [[T222]], align 8
; CHECK-NEXT: [[TMP6:%.*]] = add nsw <2 x i64> [[TMP5]], <i64 6, i64 7>
; CHECK-NEXT: store <2 x i64> [[TMP4]], ptr [[T212]], align 8
; CHECK-NEXT: store <2 x i64> [[TMP6]], ptr [[T292]], align 8
; CHECK-NEXT: store <4 x i32> [[TMP2]], ptr [[T21]], align 4
; CHECK-NEXT: ret void
;
%t14 = getelementptr inbounds i32, ptr %v1, i64 4
%t18 = getelementptr inbounds i32, ptr %v1, i64 5
%t22 = getelementptr inbounds i32, ptr %v1, i64 6
%t26 = getelementptr inbounds i32, ptr %v1, i64 7
%t142 = getelementptr inbounds i64, ptr %v1, i64 8
%t182 = getelementptr inbounds i64, ptr %v1, i64 9
%t222 = getelementptr inbounds i64, ptr %v1, i64 10
%t262 = getelementptr inbounds i64, ptr %v1, i64 11
%t21 = getelementptr inbounds i32, ptr %v0, i64 4
%t25 = getelementptr inbounds i32, ptr %v0, i64 5
%t29 = getelementptr inbounds i32, ptr %v0, i64 6
%t32 = getelementptr inbounds i32, ptr %v0, i64 7
%t212 = getelementptr inbounds i64, ptr %v0, i64 8
%t252 = getelementptr inbounds i64, ptr %v0, i64 9
%t292 = getelementptr inbounds i64, ptr %v0, i64 10
%t322 = getelementptr inbounds i64, ptr %v0, i64 11
%t19 = load i32, ptr %t14, align 4
%t23 = load i32, ptr %t18, align 4
%t27 = load i32, ptr %t22, align 4
%t30 = load i32, ptr %t26, align 4
%t192 = load i64, ptr %t142, align 8
%t232 = load i64, ptr %t182, align 8
%t272 = load i64, ptr %t222, align 8
%t302 = load i64, ptr %t262, align 8
%t20 = add nsw i32 %t19, 4
%t24 = add nsw i32 %t23, 4
%t28 = add nsw i32 %t27, 6
%t31 = add nsw i32 %t30, 7
%t202 = add nsw i64 %t192, 4
%t242 = add nsw i64 %t232, 4
%t282 = add nsw i64 %t272, 6
%t312 = add nsw i64 %t302, 7
store i64 %t202, ptr %t212, align 8
store i64 %t242, ptr %t252, align 8
store i64 %t282, ptr %t292, align 8
store i64 %t312, ptr %t322, align 8
store i32 %t20, ptr %t21, align 4
store i32 %t24, ptr %t25, align 4
store i32 %t28, ptr %t29, align 4
store i32 %t31, ptr %t32, align 4
ret void
}
|
