-
Notifications
You must be signed in to change notification settings - Fork 0
/
kernel_cholesky.ll
241 lines (212 loc) · 9.26 KB
/
kernel_cholesky.ll
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
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
; ModuleID = 'linear-algebra/solvers/cholesky/cholesky.c'
source_filename = "linear-algebra/solvers/cholesky/cholesky.c"
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"
@.str = private unnamed_addr constant [1 x i8] zeroinitializer, align 1
@stderr = external global ptr, align 8
@.str.1 = private unnamed_addr constant [23 x i8] c"==BEGIN DUMP_ARRAYS==\0A\00", align 1
@.str.2 = private unnamed_addr constant [15 x i8] c"begin dump: %s\00", align 1
@.str.3 = private unnamed_addr constant [2 x i8] c"A\00", align 1
@.str.4 = private unnamed_addr constant [2 x i8] c"\0A\00", align 1
@.str.5 = private unnamed_addr constant [8 x i8] c"%0.2lf \00", align 1
@.str.6 = private unnamed_addr constant [17 x i8] c"\0Aend dump: %s\0A\00", align 1
@.str.7 = private unnamed_addr constant [23 x i8] c"==END DUMP_ARRAYS==\0A\00", align 1
; Function Attrs: noinline nounwind optnone uwtable
define internal void @kernel_cholesky(i32 noundef %0, ptr noundef %1) #0 {
%3 = alloca i32, align 4
%4 = alloca ptr, align 8
%5 = alloca i32, align 4
%6 = alloca i32, align 4
%7 = alloca i32, align 4
store i32 %0, ptr %3, align 4
store ptr %1, ptr %4, align 8
store i32 0, ptr %5, align 4
br label %8
8: ; preds = %125, %2
%9 = load i32, ptr %5, align 4
%10 = load i32, ptr %3, align 4
%11 = icmp slt i32 %9, %10
br i1 %11, label %12, label %128
12: ; preds = %8
store i32 0, ptr %6, align 4
br label %13
13: ; preds = %70, %12
%14 = load i32, ptr %6, align 4
%15 = load i32, ptr %5, align 4
%16 = icmp slt i32 %14, %15
br i1 %16, label %17, label %73
17: ; preds = %13
store i32 0, ptr %7, align 4
br label %18
18: ; preds = %49, %17
%19 = load i32, ptr %7, align 4
%20 = load i32, ptr %6, align 4
%21 = icmp slt i32 %19, %20
br i1 %21, label %22, label %52
22: ; preds = %18
%23 = load ptr, ptr %4, align 8
%24 = load i32, ptr %5, align 4
%25 = sext i32 %24 to i64
%26 = getelementptr inbounds [2000 x double], ptr %23, i64 %25
%27 = load i32, ptr %7, align 4
%28 = sext i32 %27 to i64
%29 = getelementptr inbounds [2000 x double], ptr %26, i64 0, i64 %28
%30 = load double, ptr %29, align 8
%31 = load ptr, ptr %4, align 8
%32 = load i32, ptr %6, align 4
%33 = sext i32 %32 to i64
%34 = getelementptr inbounds [2000 x double], ptr %31, i64 %33
%35 = load i32, ptr %7, align 4
%36 = sext i32 %35 to i64
%37 = getelementptr inbounds [2000 x double], ptr %34, i64 0, i64 %36
%38 = load double, ptr %37, align 8
%39 = load ptr, ptr %4, align 8
%40 = load i32, ptr %5, align 4
%41 = sext i32 %40 to i64
%42 = getelementptr inbounds [2000 x double], ptr %39, i64 %41
%43 = load i32, ptr %6, align 4
%44 = sext i32 %43 to i64
%45 = getelementptr inbounds [2000 x double], ptr %42, i64 0, i64 %44
%46 = load double, ptr %45, align 8
%47 = fneg double %30
%48 = call double @llvm.fmuladd.f64(double %47, double %38, double %46)
store double %48, ptr %45, align 8
br label %49
49: ; preds = %22
%50 = load i32, ptr %7, align 4
%51 = add nsw i32 %50, 1
store i32 %51, ptr %7, align 4
br label %18, !llvm.loop !17
52: ; preds = %18
%53 = load ptr, ptr %4, align 8
%54 = load i32, ptr %6, align 4
%55 = sext i32 %54 to i64
%56 = getelementptr inbounds [2000 x double], ptr %53, i64 %55
%57 = load i32, ptr %6, align 4
%58 = sext i32 %57 to i64
%59 = getelementptr inbounds [2000 x double], ptr %56, i64 0, i64 %58
%60 = load double, ptr %59, align 8
%61 = load ptr, ptr %4, align 8
%62 = load i32, ptr %5, align 4
%63 = sext i32 %62 to i64
%64 = getelementptr inbounds [2000 x double], ptr %61, i64 %63
%65 = load i32, ptr %6, align 4
%66 = sext i32 %65 to i64
%67 = getelementptr inbounds [2000 x double], ptr %64, i64 0, i64 %66
%68 = load double, ptr %67, align 8
%69 = fdiv double %68, %60
store double %69, ptr %67, align 8
br label %70
70: ; preds = %52
%71 = load i32, ptr %6, align 4
%72 = add nsw i32 %71, 1
store i32 %72, ptr %6, align 4
br label %13, !llvm.loop !18
73: ; preds = %13
store i32 0, ptr %7, align 4
br label %74
74: ; preds = %105, %73
%75 = load i32, ptr %7, align 4
%76 = load i32, ptr %5, align 4
%77 = icmp slt i32 %75, %76
br i1 %77, label %78, label %108
78: ; preds = %74
%79 = load ptr, ptr %4, align 8
%80 = load i32, ptr %5, align 4
%81 = sext i32 %80 to i64
%82 = getelementptr inbounds [2000 x double], ptr %79, i64 %81
%83 = load i32, ptr %7, align 4
%84 = sext i32 %83 to i64
%85 = getelementptr inbounds [2000 x double], ptr %82, i64 0, i64 %84
%86 = load double, ptr %85, align 8
%87 = load ptr, ptr %4, align 8
%88 = load i32, ptr %5, align 4
%89 = sext i32 %88 to i64
%90 = getelementptr inbounds [2000 x double], ptr %87, i64 %89
%91 = load i32, ptr %7, align 4
%92 = sext i32 %91 to i64
%93 = getelementptr inbounds [2000 x double], ptr %90, i64 0, i64 %92
%94 = load double, ptr %93, align 8
%95 = load ptr, ptr %4, align 8
%96 = load i32, ptr %5, align 4
%97 = sext i32 %96 to i64
%98 = getelementptr inbounds [2000 x double], ptr %95, i64 %97
%99 = load i32, ptr %5, align 4
%100 = sext i32 %99 to i64
%101 = getelementptr inbounds [2000 x double], ptr %98, i64 0, i64 %100
%102 = load double, ptr %101, align 8
%103 = fneg double %86
%104 = call double @llvm.fmuladd.f64(double %103, double %94, double %102)
store double %104, ptr %101, align 8
br label %105
105: ; preds = %78
%106 = load i32, ptr %7, align 4
%107 = add nsw i32 %106, 1
store i32 %107, ptr %7, align 4
br label %74, !llvm.loop !19
108: ; preds = %74
%109 = load ptr, ptr %4, align 8
%110 = load i32, ptr %5, align 4
%111 = sext i32 %110 to i64
%112 = getelementptr inbounds [2000 x double], ptr %109, i64 %111
%113 = load i32, ptr %5, align 4
%114 = sext i32 %113 to i64
%115 = getelementptr inbounds [2000 x double], ptr %112, i64 0, i64 %114
%116 = load double, ptr %115, align 8
%117 = call double @sqrt(double noundef %116) #6
%118 = load ptr, ptr %4, align 8
%119 = load i32, ptr %5, align 4
%120 = sext i32 %119 to i64
%121 = getelementptr inbounds [2000 x double], ptr %118, i64 %120
%122 = load i32, ptr %5, align 4
%123 = sext i32 %122 to i64
%124 = getelementptr inbounds [2000 x double], ptr %121, i64 0, i64 %123
store double %117, ptr %124, align 8
br label %125
125: ; preds = %108
%126 = load i32, ptr %5, align 4
%127 = add nsw i32 %126, 1
store i32 %127, ptr %5, align 4
br label %8, !llvm.loop !20
128: ; preds = %8
ret void
}
; Function Attrs: nounwind
declare void @free(ptr noundef) #3
; Function Attrs: nocallback nofree nosync nounwind speculatable willreturn memory(none)
declare double @llvm.fmuladd.f64(double, double, double) #4
; Function Attrs: nounwind
declare double @sqrt(double noundef) #3
declare i32 @fprintf(ptr noundef, ptr noundef, ...) #1
attributes #0 = { noinline nounwind optnone uwtable "frame-pointer"="all" "min-legal-vector-width"="0" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+cx8,+fxsr,+mmx,+sse,+sse2,+x87" "tune-cpu"="generic" }
attributes #1 = { "frame-pointer"="all" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+cx8,+fxsr,+mmx,+sse,+sse2,+x87" "tune-cpu"="generic" }
attributes #2 = { nounwind willreturn memory(read) "frame-pointer"="all" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+cx8,+fxsr,+mmx,+sse,+sse2,+x87" "tune-cpu"="generic" }
attributes #3 = { nounwind "frame-pointer"="all" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="x86-64" "target-features"="+cx8,+fxsr,+mmx,+sse,+sse2,+x87" "tune-cpu"="generic" }
attributes #4 = { nocallback nofree nosync nounwind speculatable willreturn memory(none) }
attributes #5 = { nounwind willreturn memory(read) }
attributes #6 = { nounwind }
!llvm.module.flags = !{!0, !1, !2, !3, !4}
!llvm.ident = !{!5}
!0 = !{i32 1, !"wchar_size", i32 4}
!1 = !{i32 8, !"PIC Level", i32 2}
!2 = !{i32 7, !"PIE Level", i32 2}
!3 = !{i32 7, !"uwtable", i32 2}
!4 = !{i32 7, !"frame-pointer", i32 2}
!5 = !{!"clang version 16.0.0"}
!6 = distinct !{!6, !7}
!7 = !{!"llvm.loop.mustprogress"}
!8 = distinct !{!8, !7}
!9 = distinct !{!9, !7}
!10 = distinct !{!10, !7}
!11 = distinct !{!11, !7}
!12 = distinct !{!12, !7}
!13 = distinct !{!13, !7}
!14 = distinct !{!14, !7}
!15 = distinct !{!15, !7}
!16 = distinct !{!16, !7}
!17 = distinct !{!17, !7}
!18 = distinct !{!18, !7}
!19 = distinct !{!19, !7}
!20 = distinct !{!20, !7}
!21 = distinct !{!21, !7}
!22 = distinct !{!22, !7}