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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=loop-reroll -S | FileCheck %s
target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:32:64-v128:32:128-a0:0:32-n32-S32"
target triple = "thumbv7-none-linux"
;void foo(int *A, int *B, int m, int n) {
; for (int i = m; i < n; i+=4) {
; A[i+0] = B[i+0] * 4;
; A[i+1] = B[i+1] * 4;
; A[i+2] = B[i+2] * 4;
; A[i+3] = B[i+3] * 4;
; }
;}
define void @foo(ptr nocapture %A, ptr nocapture readonly %B, i32 %m, i32 %n) {
; CHECK-LABEL: @foo(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[CMP34:%.*]] = icmp slt i32 [[M:%.*]], [[N:%.*]]
; CHECK-NEXT: br i1 [[CMP34]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END:%.*]]
; CHECK: for.body.preheader:
; CHECK-NEXT: [[TMP0:%.*]] = add i32 [[M]], 4
; CHECK-NEXT: [[SMAX:%.*]] = call i32 @llvm.smax.i32(i32 [[N]], i32 [[TMP0]])
; CHECK-NEXT: [[TMP1:%.*]] = add i32 [[SMAX]], -1
; CHECK-NEXT: [[TMP2:%.*]] = sub i32 [[TMP1]], [[M]]
; CHECK-NEXT: [[TMP3:%.*]] = lshr i32 [[TMP2]], 2
; CHECK-NEXT: [[TMP4:%.*]] = shl nuw i32 [[TMP3]], 2
; CHECK-NEXT: [[TMP5:%.*]] = add nuw nsw i32 [[TMP4]], 3
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.body:
; CHECK-NEXT: [[INDVAR:%.*]] = phi i32 [ 0, [[FOR_BODY_PREHEADER]] ], [ [[INDVAR_NEXT:%.*]], [[FOR_BODY]] ]
; CHECK-NEXT: [[TMP6:%.*]] = add i32 [[M]], [[INDVAR]]
; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i32 [[TMP6]]
; CHECK-NEXT: [[TMP7:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
; CHECK-NEXT: [[MUL:%.*]] = shl nsw i32 [[TMP7]], 2
; CHECK-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i32 [[TMP6]]
; CHECK-NEXT: store i32 [[MUL]], ptr [[ARRAYIDX2]], align 4
; CHECK-NEXT: [[INDVAR_NEXT]] = add i32 [[INDVAR]], 1
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[INDVAR]], [[TMP5]]
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_END_LOOPEXIT:%.*]], label [[FOR_BODY]]
; CHECK: for.end.loopexit:
; CHECK-NEXT: br label [[FOR_END]]
; CHECK: for.end:
; CHECK-NEXT: ret void
;
entry:
%cmp34 = icmp slt i32 %m, %n
br i1 %cmp34, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%i.035 = phi i32 [ %add18, %for.body ], [ %m, %entry ]
%arrayidx = getelementptr inbounds i32, ptr %B, i32 %i.035
%0 = load i32, ptr %arrayidx, align 4
%mul = shl nsw i32 %0, 2
%arrayidx2 = getelementptr inbounds i32, ptr %A, i32 %i.035
store i32 %mul, ptr %arrayidx2, align 4
%add3 = add nsw i32 %i.035, 1
%arrayidx4 = getelementptr inbounds i32, ptr %B, i32 %add3
%1 = load i32, ptr %arrayidx4, align 4
%mul5 = shl nsw i32 %1, 2
%arrayidx7 = getelementptr inbounds i32, ptr %A, i32 %add3
store i32 %mul5, ptr %arrayidx7, align 4
%add8 = add nsw i32 %i.035, 2
%arrayidx9 = getelementptr inbounds i32, ptr %B, i32 %add8
%2 = load i32, ptr %arrayidx9, align 4
%mul10 = shl nsw i32 %2, 2
%arrayidx12 = getelementptr inbounds i32, ptr %A, i32 %add8
store i32 %mul10, ptr %arrayidx12, align 4
%add13 = add nsw i32 %i.035, 3
%arrayidx14 = getelementptr inbounds i32, ptr %B, i32 %add13
%3 = load i32, ptr %arrayidx14, align 4
%mul15 = shl nsw i32 %3, 2
%arrayidx17 = getelementptr inbounds i32, ptr %A, i32 %add13
store i32 %mul15, ptr %arrayidx17, align 4
%add18 = add nsw i32 %i.035, 4
%cmp = icmp slt i32 %add18, %n
br i1 %cmp, label %for.body, label %for.end
for.end: ; preds = %for.body, %entry
ret void
}
;void daxpy_ur(int n,float da,ptr dx,ptr dy)
; {
; int m = n % 4;
; for (int i = m; i < n; i = i + 4)
; {
; dy[i] = dy[i] + da*dx[i];
; dy[i+1] = dy[i+1] + da*dx[i+1];
; dy[i+2] = dy[i+2] + da*dx[i+2];
; dy[i+3] = dy[i+3] + da*dx[i+3];
; }
; }
define void @daxpy_ur(i32 %n, float %da, ptr nocapture readonly %dx, ptr nocapture %dy) {
; CHECK-LABEL: @daxpy_ur(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[REM:%.*]] = srem i32 [[N:%.*]], 4
; CHECK-NEXT: [[CMP55:%.*]] = icmp slt i32 [[REM]], [[N]]
; CHECK-NEXT: br i1 [[CMP55]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END:%.*]]
; CHECK: for.body.preheader:
; CHECK-NEXT: [[TMP0:%.*]] = add i32 [[N]], -1
; CHECK-NEXT: [[TMP1:%.*]] = sub i32 [[TMP0]], [[REM]]
; CHECK-NEXT: [[TMP2:%.*]] = lshr i32 [[TMP1]], 2
; CHECK-NEXT: [[TMP3:%.*]] = shl nuw i32 [[TMP2]], 2
; CHECK-NEXT: [[TMP4:%.*]] = add nuw nsw i32 [[TMP3]], 3
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.body:
; CHECK-NEXT: [[INDVAR:%.*]] = phi i32 [ 0, [[FOR_BODY_PREHEADER]] ], [ [[INDVAR_NEXT:%.*]], [[FOR_BODY]] ]
; CHECK-NEXT: [[TMP5:%.*]] = add i32 [[REM]], [[INDVAR]]
; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds float, ptr [[DY:%.*]], i32 [[TMP5]]
; CHECK-NEXT: [[TMP6:%.*]] = load float, ptr [[ARRAYIDX]], align 4
; CHECK-NEXT: [[ARRAYIDX1:%.*]] = getelementptr inbounds float, ptr [[DX:%.*]], i32 [[TMP5]]
; CHECK-NEXT: [[TMP7:%.*]] = load float, ptr [[ARRAYIDX1]], align 4
; CHECK-NEXT: [[MUL:%.*]] = fmul float [[TMP7]], [[DA:%.*]]
; CHECK-NEXT: [[ADD:%.*]] = fadd float [[TMP6]], [[MUL]]
; CHECK-NEXT: store float [[ADD]], ptr [[ARRAYIDX]], align 4
; CHECK-NEXT: [[INDVAR_NEXT]] = add i32 [[INDVAR]], 1
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[INDVAR]], [[TMP4]]
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_END_LOOPEXIT:%.*]], label [[FOR_BODY]]
; CHECK: for.end.loopexit:
; CHECK-NEXT: br label [[FOR_END]]
; CHECK: for.end:
; CHECK-NEXT: ret void
;
entry:
%rem = srem i32 %n, 4
%cmp55 = icmp slt i32 %rem, %n
br i1 %cmp55, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%i.056 = phi i32 [ %add27, %for.body ], [ %rem, %entry ]
%arrayidx = getelementptr inbounds float, ptr %dy, i32 %i.056
%0 = load float, ptr %arrayidx, align 4
%arrayidx1 = getelementptr inbounds float, ptr %dx, i32 %i.056
%1 = load float, ptr %arrayidx1, align 4
%mul = fmul float %1, %da
%add = fadd float %0, %mul
store float %add, ptr %arrayidx, align 4
%add3 = add nsw i32 %i.056, 1
%arrayidx4 = getelementptr inbounds float, ptr %dy, i32 %add3
%2 = load float, ptr %arrayidx4, align 4
%arrayidx6 = getelementptr inbounds float, ptr %dx, i32 %add3
%3 = load float, ptr %arrayidx6, align 4
%mul7 = fmul float %3, %da
%add8 = fadd float %2, %mul7
store float %add8, ptr %arrayidx4, align 4
%add11 = add nsw i32 %i.056, 2
%arrayidx12 = getelementptr inbounds float, ptr %dy, i32 %add11
%4 = load float, ptr %arrayidx12, align 4
%arrayidx14 = getelementptr inbounds float, ptr %dx, i32 %add11
%5 = load float, ptr %arrayidx14, align 4
%mul15 = fmul float %5, %da
%add16 = fadd float %4, %mul15
store float %add16, ptr %arrayidx12, align 4
%add19 = add nsw i32 %i.056, 3
%arrayidx20 = getelementptr inbounds float, ptr %dy, i32 %add19
%6 = load float, ptr %arrayidx20, align 4
%arrayidx22 = getelementptr inbounds float, ptr %dx, i32 %add19
%7 = load float, ptr %arrayidx22, align 4
%mul23 = fmul float %7, %da
%add24 = fadd float %6, %mul23
store float %add24, ptr %arrayidx20, align 4
%add27 = add nsw i32 %i.056, 4
%cmp = icmp slt i32 %add27, %n
br i1 %cmp, label %for.body, label %for.end
for.end: ; preds = %for.body, %entry
ret void
}
|
