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
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -O3 -S | FileCheck --check-prefix=OLDPM %s
; RUN: opt < %s -passes='default<O3>' -S | FileCheck --check-prefix=NEWPM %s
target datalayout = "e-m:e-p:32:32-Fi8-i64:64-v128:64:128-a:0:32-n32-S64"
target triple = "thumbv6m-none-none-eabi"
; Not only should we be able to to form memsets here, the original loops
; should be deleted, too.
; Function Attrs: nounwind
define dso_local void @arm_fill_q7(i8 signext %value, ptr %pDst, i32 %blockSize) #0 {
; OLDPM-LABEL: @arm_fill_q7(
; OLDPM-NEXT: entry:
; OLDPM-NEXT: [[CMP_NOT20:%.*]] = icmp ult i32 [[BLOCKSIZE:%.*]], 4
; OLDPM-NEXT: br i1 [[CMP_NOT20]], label [[WHILE_END:%.*]], label [[WHILE_BODY_PREHEADER:%.*]]
; OLDPM: while.body.preheader:
; OLDPM-NEXT: [[TMP0:%.*]] = and i32 [[BLOCKSIZE]], -4
; OLDPM-NEXT: call void @llvm.memset.p0.i32(ptr align 1 [[PDST:%.*]], i8 [[VALUE:%.*]], i32 [[TMP0]], i1 false)
; OLDPM-NEXT: [[SCEVGEP:%.*]] = getelementptr i8, ptr [[PDST]], i32 [[TMP0]]
; OLDPM-NEXT: br label [[WHILE_END]]
; OLDPM: while.end:
; OLDPM-NEXT: [[PDST_ADDR_0_LCSSA:%.*]] = phi ptr [ [[PDST]], [[ENTRY:%.*]] ], [ [[SCEVGEP]], [[WHILE_BODY_PREHEADER]] ]
; OLDPM-NEXT: [[REM:%.*]] = and i32 [[BLOCKSIZE]], 3
; OLDPM-NEXT: [[CMP14_NOT17:%.*]] = icmp eq i32 [[REM]], 0
; OLDPM-NEXT: br i1 [[CMP14_NOT17]], label [[WHILE_END18:%.*]], label [[WHILE_BODY16_PREHEADER:%.*]]
; OLDPM: while.body16.preheader:
; OLDPM-NEXT: call void @llvm.memset.p0.i32(ptr align 1 [[PDST_ADDR_0_LCSSA]], i8 [[VALUE]], i32 [[REM]], i1 false)
; OLDPM-NEXT: br label [[WHILE_END18]]
; OLDPM: while.end18:
; OLDPM-NEXT: ret void
;
; NEWPM-LABEL: @arm_fill_q7(
; NEWPM-NEXT: entry:
; NEWPM-NEXT: [[CMP_NOT17:%.*]] = icmp ult i32 [[BLOCKSIZE:%.*]], 4
; NEWPM-NEXT: br i1 [[CMP_NOT17]], label [[WHILE_END:%.*]], label [[WHILE_BODY_PREHEADER:%.*]]
; NEWPM: while.body.preheader:
; NEWPM-NEXT: [[TMP0:%.*]] = and i32 [[BLOCKSIZE]], -4
; NEWPM-NEXT: call void @llvm.memset.p0.i32(ptr align 1 [[PDST:%.*]], i8 [[VALUE:%.*]], i32 [[TMP0]], i1 false)
; NEWPM-NEXT: [[SCEVGEP:%.*]] = getelementptr i8, ptr [[PDST]], i32 [[TMP0]]
; NEWPM-NEXT: br label [[WHILE_END]]
; NEWPM: while.end:
; NEWPM-NEXT: [[PDST_ADDR_0_LCSSA:%.*]] = phi ptr [ [[PDST]], [[ENTRY:%.*]] ], [ [[SCEVGEP]], [[WHILE_BODY_PREHEADER]] ]
; NEWPM-NEXT: [[REM:%.*]] = and i32 [[BLOCKSIZE]], 3
; NEWPM-NEXT: [[CMP14_NOT20:%.*]] = icmp eq i32 [[REM]], 0
; NEWPM-NEXT: br i1 [[CMP14_NOT20]], label [[WHILE_END18:%.*]], label [[WHILE_BODY16_PREHEADER:%.*]]
; NEWPM: while.body16.preheader:
; NEWPM-NEXT: call void @llvm.memset.p0.i32(ptr align 1 [[PDST_ADDR_0_LCSSA]], i8 [[VALUE]], i32 [[REM]], i1 false)
; NEWPM-NEXT: br label [[WHILE_END18]]
; NEWPM: while.end18:
; NEWPM-NEXT: ret void
;
entry:
%value.addr = alloca i8, align 1
%pDst.addr = alloca ptr, align 4
%blockSize.addr = alloca i32, align 4
%blkCnt = alloca i32, align 4
%packedValue = alloca i32, align 4
store i8 %value, ptr %value.addr, align 1, !tbaa !3
store ptr %pDst, ptr %pDst.addr, align 4, !tbaa !6
store i32 %blockSize, ptr %blockSize.addr, align 4, !tbaa !8
call void @llvm.lifetime.start.p0(i64 4, ptr %blkCnt) #3
call void @llvm.lifetime.start.p0(i64 4, ptr %packedValue) #3
%0 = load i8, ptr %value.addr, align 1, !tbaa !3
%conv = sext i8 %0 to i32
%shl = shl i32 %conv, 0
%and = and i32 %shl, 255
%1 = load i8, ptr %value.addr, align 1, !tbaa !3
%conv1 = sext i8 %1 to i32
%shl2 = shl i32 %conv1, 8
%and3 = and i32 %shl2, 65280
%or = or i32 %and, %and3
%2 = load i8, ptr %value.addr, align 1, !tbaa !3
%conv4 = sext i8 %2 to i32
%shl5 = shl i32 %conv4, 16
%and6 = and i32 %shl5, 16711680
%or7 = or i32 %or, %and6
%3 = load i8, ptr %value.addr, align 1, !tbaa !3
%conv8 = sext i8 %3 to i32
%shl9 = shl i32 %conv8, 24
%and10 = and i32 %shl9, -16777216
%or11 = or i32 %or7, %and10
store i32 %or11, ptr %packedValue, align 4, !tbaa !8
%4 = load i32, ptr %blockSize.addr, align 4, !tbaa !8
%shr = lshr i32 %4, 2
store i32 %shr, ptr %blkCnt, align 4, !tbaa !8
br label %while.cond
while.cond: ; preds = %while.body, %entry
%5 = load i32, ptr %blkCnt, align 4, !tbaa !8
%cmp = icmp ugt i32 %5, 0
br i1 %cmp, label %while.body, label %while.end
while.body: ; preds = %while.cond
%6 = load i32, ptr %packedValue, align 4, !tbaa !8
call void @write_q7x4_ia(ptr %pDst.addr, i32 %6)
%7 = load i32, ptr %blkCnt, align 4, !tbaa !8
%dec = add i32 %7, -1
store i32 %dec, ptr %blkCnt, align 4, !tbaa !8
br label %while.cond, !llvm.loop !10
while.end: ; preds = %while.cond
%8 = load i32, ptr %blockSize.addr, align 4, !tbaa !8
%rem = urem i32 %8, 4
store i32 %rem, ptr %blkCnt, align 4, !tbaa !8
br label %while.cond13
while.cond13: ; preds = %while.body16, %while.end
%9 = load i32, ptr %blkCnt, align 4, !tbaa !8
%cmp14 = icmp ugt i32 %9, 0
br i1 %cmp14, label %while.body16, label %while.end18
while.body16: ; preds = %while.cond13
%10 = load i8, ptr %value.addr, align 1, !tbaa !3
%11 = load ptr, ptr %pDst.addr, align 4, !tbaa !6
%incdec.ptr = getelementptr inbounds i8, ptr %11, i32 1
store ptr %incdec.ptr, ptr %pDst.addr, align 4, !tbaa !6
store i8 %10, ptr %11, align 1, !tbaa !3
%12 = load i32, ptr %blkCnt, align 4, !tbaa !8
%dec17 = add i32 %12, -1
store i32 %dec17, ptr %blkCnt, align 4, !tbaa !8
br label %while.cond13, !llvm.loop !12
while.end18: ; preds = %while.cond13
call void @llvm.lifetime.end.p0(i64 4, ptr %packedValue) #3
call void @llvm.lifetime.end.p0(i64 4, ptr %blkCnt) #3
ret void
}
; Function Attrs: argmemonly nofree nosync nounwind willreturn
declare void @llvm.lifetime.start.p0(i64 immarg, ptr nocapture) #1
; Function Attrs: alwaysinline nounwind
define internal void @write_q7x4_ia(ptr %pQ7, i32 %value) #2 {
entry:
%pQ7.addr = alloca ptr, align 4
%value.addr = alloca i32, align 4
%val = alloca i32, align 4
store ptr %pQ7, ptr %pQ7.addr, align 4, !tbaa !6
store i32 %value, ptr %value.addr, align 4, !tbaa !8
call void @llvm.lifetime.start.p0(i64 4, ptr %val) #3
%0 = load i32, ptr %value.addr, align 4, !tbaa !8
store i32 %0, ptr %val, align 4, !tbaa !8
%1 = load i32, ptr %val, align 4, !tbaa !8
%and = and i32 %1, 255
%conv = trunc i32 %and to i8
%2 = load ptr, ptr %pQ7.addr, align 4, !tbaa !6
%3 = load ptr, ptr %2, align 4, !tbaa !6
store i8 %conv, ptr %3, align 1, !tbaa !3
%4 = load i32, ptr %val, align 4, !tbaa !8
%shr = ashr i32 %4, 8
%and1 = and i32 %shr, 255
%conv2 = trunc i32 %and1 to i8
%5 = load ptr, ptr %pQ7.addr, align 4, !tbaa !6
%6 = load ptr, ptr %5, align 4, !tbaa !6
%arrayidx3 = getelementptr inbounds i8, ptr %6, i32 1
store i8 %conv2, ptr %arrayidx3, align 1, !tbaa !3
%7 = load i32, ptr %val, align 4, !tbaa !8
%shr4 = ashr i32 %7, 16
%and5 = and i32 %shr4, 255
%conv6 = trunc i32 %and5 to i8
%8 = load ptr, ptr %pQ7.addr, align 4, !tbaa !6
%9 = load ptr, ptr %8, align 4, !tbaa !6
%arrayidx7 = getelementptr inbounds i8, ptr %9, i32 2
store i8 %conv6, ptr %arrayidx7, align 1, !tbaa !3
%10 = load i32, ptr %val, align 4, !tbaa !8
%shr8 = ashr i32 %10, 24
%and9 = and i32 %shr8, 255
%conv10 = trunc i32 %and9 to i8
%11 = load ptr, ptr %pQ7.addr, align 4, !tbaa !6
%12 = load ptr, ptr %11, align 4, !tbaa !6
%arrayidx11 = getelementptr inbounds i8, ptr %12, i32 3
store i8 %conv10, ptr %arrayidx11, align 1, !tbaa !3
%13 = load ptr, ptr %pQ7.addr, align 4, !tbaa !6
%14 = load ptr, ptr %13, align 4, !tbaa !6
%add.ptr = getelementptr inbounds i8, ptr %14, i32 4
store ptr %add.ptr, ptr %13, align 4, !tbaa !6
call void @llvm.lifetime.end.p0(i64 4, ptr %val) #3
ret void
}
; Function Attrs: argmemonly nofree nosync nounwind willreturn
declare void @llvm.lifetime.end.p0(i64 immarg, ptr nocapture) #1
attributes #0 = { nounwind "disable-tail-calls"="false" "frame-pointer"="all" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="cortex-m0plus" "target-features"="+armv6-m,+strict-align,+thumb-mode,-aes,-bf16,-cdecp0,-cdecp1,-cdecp2,-cdecp3,-cdecp4,-cdecp5,-cdecp6,-cdecp7,-crc,-crypto,-dotprod,-dsp,-fp16fml,-fullfp16,-hwdiv,-hwdiv-arm,-i8mm,-lob,-mve,-mve.fp,-ras,-sb,-sha2" "unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #1 = { argmemonly nofree nosync nounwind willreturn }
attributes #2 = { alwaysinline nounwind "disable-tail-calls"="false" "frame-pointer"="all" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="cortex-m0plus" "target-features"="+armv6-m,+strict-align,+thumb-mode,-aes,-bf16,-cdecp0,-cdecp1,-cdecp2,-cdecp3,-cdecp4,-cdecp5,-cdecp6,-cdecp7,-crc,-crypto,-dotprod,-dsp,-fp16fml,-fullfp16,-hwdiv,-hwdiv-arm,-i8mm,-lob,-mve,-mve.fp,-ras,-sb,-sha2" "unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #3 = { nounwind }
!llvm.module.flags = !{!0, !1}
!llvm.ident = !{!2}
!0 = !{i32 1, !"wchar_size", i32 4}
!1 = !{i32 1, !"min_enum_size", i32 4}
!2 = !{!"clang version 12.0.0 (https://github.com/llvm/llvm-project.git 07f234be1ccbce131704f580aa3f117083a887f7)"}
!3 = !{!4, !4, i64 0}
!4 = !{!"omnipotent char", !5, i64 0}
!5 = !{!"Simple C/C++ TBAA"}
!6 = !{!7, !7, i64 0}
!7 = !{!"any pointer", !4, i64 0}
!8 = !{!9, !9, i64 0}
!9 = !{!"int", !4, i64 0}
!10 = distinct !{!10, !11}
!11 = !{!"llvm.loop.mustprogress"}
!12 = distinct !{!12, !11}
|
