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
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
|
; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 5
; RUN: opt -passes='print<access-info>' -disable-output %s 2>&1 | FileCheck %s
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
define void @loads_of_same_pointer_with_different_sizes1(ptr %A, ptr %B, i64 %N) {
; CHECK-LABEL: 'loads_of_same_pointer_with_different_sizes1'
; CHECK-NEXT: loop:
; CHECK-NEXT: Memory dependences are safe with run-time checks
; CHECK-NEXT: Dependences:
; CHECK-NEXT: Run-time memory checks:
; CHECK-NEXT: Check 0:
; CHECK-NEXT: Comparing group ([[GRP1:0x[0-9a-f]+]]):
; CHECK-NEXT: %gep.B = getelementptr inbounds i32, ptr %B, i64 %iv
; CHECK-NEXT: Against group ([[GRP2:0x[0-9a-f]+]]):
; CHECK-NEXT: %gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
; CHECK-NEXT: %gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
; CHECK-NEXT: Grouped accesses:
; CHECK-NEXT: Group [[GRP1]]:
; CHECK-NEXT: (Low: %B High: ((4 * %N) + %B))
; CHECK-NEXT: Member: {%B,+,4}<nuw><%loop>
; CHECK-NEXT: Group [[GRP2]]:
; CHECK-NEXT: (Low: %A High: (3 + %N + %A))
; CHECK-NEXT: Member: {%A,+,1}<nuw><%loop>
; CHECK-NEXT: Member: {%A,+,1}<nuw><%loop>
; CHECK-EMPTY:
; CHECK-NEXT: Non vectorizable stores to invariant address were not found in loop.
; CHECK-NEXT: SCEV assumptions:
; CHECK-EMPTY:
; CHECK-NEXT: Expressions re-written:
;
entry:
br label %loop
loop:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
%gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
%l0 = load i8, ptr %gep.A, align 1
%l1 = load i32, ptr %gep.A, align 1
%l0.ext = sext i8 %l0 to i32
%iv.trunc = trunc nuw i64 %iv to i32
%sub.0 = sub i32 %l0.ext, %iv.trunc
%sub.1 = sub i32 %l1, %sub.0
%gep.B = getelementptr inbounds i32, ptr %B, i64 %iv
store i32 %sub.1, ptr %gep.B, align 4
%iv.next = add nuw nsw i64 %iv, 1
%exitcond.not = icmp eq i64 %iv.next, %N
br i1 %exitcond.not, label %exit, label %loop
exit:
ret void
}
define void @loads_of_same_pointer_with_different_sizes2(ptr %A, ptr %B, i64 %N) {
; CHECK-LABEL: 'loads_of_same_pointer_with_different_sizes2'
; CHECK-NEXT: loop:
; CHECK-NEXT: Memory dependences are safe with run-time checks
; CHECK-NEXT: Dependences:
; CHECK-NEXT: Run-time memory checks:
; CHECK-NEXT: Check 0:
; CHECK-NEXT: Comparing group ([[GRP3:0x[0-9a-f]+]]):
; CHECK-NEXT: %gep.B = getelementptr inbounds i32, ptr %B, i64 %iv
; CHECK-NEXT: Against group ([[GRP4:0x[0-9a-f]+]]):
; CHECK-NEXT: %gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
; CHECK-NEXT: %gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
; CHECK-NEXT: Grouped accesses:
; CHECK-NEXT: Group [[GRP3]]:
; CHECK-NEXT: (Low: %B High: ((4 * %N) + %B))
; CHECK-NEXT: Member: {%B,+,4}<nuw><%loop>
; CHECK-NEXT: Group [[GRP4]]:
; CHECK-NEXT: (Low: %A High: (3 + %N + %A))
; CHECK-NEXT: Member: {%A,+,1}<nuw><%loop>
; CHECK-NEXT: Member: {%A,+,1}<nuw><%loop>
; CHECK-EMPTY:
; CHECK-NEXT: Non vectorizable stores to invariant address were not found in loop.
; CHECK-NEXT: SCEV assumptions:
; CHECK-EMPTY:
; CHECK-NEXT: Expressions re-written:
;
entry:
br label %loop
loop:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
%gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
%l1 = load i32, ptr %gep.A, align 1
%l0 = load i8, ptr %gep.A, align 1
%l0.ext = sext i8 %l0 to i32
%iv.trunc = trunc nuw i64 %iv to i32
%sub.0 = sub i32 %l0.ext, %iv.trunc
%sub.1 = sub i32 %l1, %sub.0
%gep.B = getelementptr inbounds i32, ptr %B, i64 %iv
store i32 %sub.1, ptr %gep.B, align 4
%iv.next = add nuw nsw i64 %iv, 1
%exitcond.not = icmp eq i64 %iv.next, %N
br i1 %exitcond.not, label %exit, label %loop
exit:
ret void
}
define void @loads_of_same_pointer_with_different_sizes_retry_with_runtime_checks(ptr %A, ptr %B, i64 %N, i64 %off) {
; CHECK-LABEL: 'loads_of_same_pointer_with_different_sizes_retry_with_runtime_checks'
; CHECK-NEXT: loop:
; CHECK-NEXT: Memory dependences are safe with run-time checks
; CHECK-NEXT: Dependences:
; CHECK-NEXT: Run-time memory checks:
; CHECK-NEXT: Check 0:
; CHECK-NEXT: Comparing group ([[GRP5:0x[0-9a-f]+]]):
; CHECK-NEXT: %gep.B.iv = getelementptr inbounds i32, ptr %B, i64 %iv
; CHECK-NEXT: Against group ([[GRP6:0x[0-9a-f]+]]):
; CHECK-NEXT: %gep.B.inc = getelementptr inbounds i32, ptr %B, i64 %inc
; CHECK-NEXT: Check 1:
; CHECK-NEXT: Comparing group ([[GRP5]]):
; CHECK-NEXT: %gep.B.iv = getelementptr inbounds i32, ptr %B, i64 %iv
; CHECK-NEXT: Against group ([[GRP7:0x[0-9a-f]+]]):
; CHECK-NEXT: %gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
; CHECK-NEXT: Check 2:
; CHECK-NEXT: Comparing group ([[GRP5]]):
; CHECK-NEXT: %gep.B.iv = getelementptr inbounds i32, ptr %B, i64 %iv
; CHECK-NEXT: Against group ([[GRP8:0x[0-9a-f]+]]):
; CHECK-NEXT: %gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
; CHECK-NEXT: Check 3:
; CHECK-NEXT: Comparing group ([[GRP6]]):
; CHECK-NEXT: %gep.B.inc = getelementptr inbounds i32, ptr %B, i64 %inc
; CHECK-NEXT: Against group ([[GRP7]]):
; CHECK-NEXT: %gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
; CHECK-NEXT: Check 4:
; CHECK-NEXT: Comparing group ([[GRP6]]):
; CHECK-NEXT: %gep.B.inc = getelementptr inbounds i32, ptr %B, i64 %inc
; CHECK-NEXT: Against group ([[GRP8]]):
; CHECK-NEXT: %gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
; CHECK-NEXT: Grouped accesses:
; CHECK-NEXT: Group [[GRP5]]:
; CHECK-NEXT: (Low: %B High: ((4 * %N) + %B))
; CHECK-NEXT: Member: {%B,+,4}<nuw><%loop>
; CHECK-NEXT: Group [[GRP6]]:
; CHECK-NEXT: (Low: ((4 * %off) + %B) High: ((4 * %N) + (4 * %off) + %B))
; CHECK-NEXT: Member: {((4 * %off) + %B),+,4}<%loop>
; CHECK-NEXT: Group [[GRP7]]:
; CHECK-NEXT: (Low: %A High: (%N + %A))
; CHECK-NEXT: Member: {%A,+,1}<nuw><%loop>
; CHECK-NEXT: Group [[GRP8]]:
; CHECK-NEXT: (Low: %A High: (3 + %N + %A))
; CHECK-NEXT: Member: {%A,+,1}<nuw><%loop>
; CHECK-EMPTY:
; CHECK-NEXT: Non vectorizable stores to invariant address were not found in loop.
; CHECK-NEXT: SCEV assumptions:
; CHECK-EMPTY:
; CHECK-NEXT: Expressions re-written:
;
entry:
br label %loop
loop:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
%gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
%l0 = load i8, ptr %gep.A, align 1
%l1 = load i32, ptr %gep.A, align 1
%l0.ext = sext i8 %l0 to i32
%iv.trunc = trunc nuw i64 %iv to i32
%sub.0 = sub i32 %l0.ext, %iv.trunc
%sub.1 = sub i32 %l1, %iv.trunc
%gep.B.iv = getelementptr inbounds i32, ptr %B, i64 %iv
store i32 %sub.0, ptr %gep.B.iv, align 4
%inc = add nuw nsw i64 %iv, %off
%gep.B.inc = getelementptr inbounds i32, ptr %B, i64 %inc
store i32 %sub.1, ptr %gep.B.inc , align 4
%iv.next = add nuw nsw i64 %iv, 1
%exitcond.not = icmp eq i64 %iv.next, %N
br i1 %exitcond.not, label %exit, label %loop
exit:
ret void
}
|
