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
| ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -indvars -S | FileCheck %s
; Indvars should be able to compute the exit value of this loop
; without any additional arithmetic. The only add needed should
; be the canonical IV increment.
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128"
define void @foo(double* %p, i32 %n) nounwind {
; CHECK-LABEL: @foo(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = icmp sgt i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[TMP0]], label [[BB_NPH:%.*]], label [[RETURN:%.*]]
; CHECK: bb.nph:
; CHECK-NEXT: br label [[BB2:%.*]]
; CHECK: bb2:
; CHECK-NEXT: [[I_01:%.*]] = phi i32 [ [[TMP7:%.*]], [[BB3:%.*]] ], [ 0, [[BB_NPH]] ]
; CHECK-NEXT: [[TMP1:%.*]] = sext i32 [[I_01]] to i64
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr double, double* [[P:%.*]], i64 [[TMP1]]
; CHECK-NEXT: [[TMP3:%.*]] = load double, double* [[TMP2]], align 8
; CHECK-NEXT: [[TMP4:%.*]] = fmul double [[TMP3]], 1.100000e+00
; CHECK-NEXT: [[TMP5:%.*]] = sext i32 [[I_01]] to i64
; CHECK-NEXT: [[TMP6:%.*]] = getelementptr double, double* [[P]], i64 [[TMP5]]
; CHECK-NEXT: store double [[TMP4]], double* [[TMP6]], align 8
; CHECK-NEXT: [[TMP7]] = add nuw nsw i32 [[I_01]], 1
; CHECK-NEXT: br label [[BB3]]
; CHECK: bb3:
; CHECK-NEXT: [[TMP8:%.*]] = icmp slt i32 [[TMP7]], [[N]]
; CHECK-NEXT: br i1 [[TMP8]], label [[BB2]], label [[BB3_RETURN_CRIT_EDGE:%.*]]
; CHECK: bb3.return_crit_edge:
; CHECK-NEXT: br label [[RETURN]]
; CHECK: return:
; CHECK-NEXT: ret void
;
entry:
%0 = icmp sgt i32 %n, 0
br i1 %0, label %bb.nph, label %return
bb.nph: ; preds = %entry
br label %bb2
bb2: ; preds = %bb3, %bb.nph
%i.01 = phi i32 [ %7, %bb3 ], [ 0, %bb.nph ]
%1 = sext i32 %i.01 to i64
%2 = getelementptr double, double* %p, i64 %1
%3 = load double, double* %2, align 8
%4 = fmul double %3, 1.100000e+00
%5 = sext i32 %i.01 to i64
%6 = getelementptr double, double* %p, i64 %5
store double %4, double* %6, align 8
%7 = add i32 %i.01, 1
br label %bb3
bb3: ; preds = %bb2
%8 = icmp slt i32 %7, %n
br i1 %8, label %bb2, label %bb3.return_crit_edge
bb3.return_crit_edge: ; preds = %bb3
br label %return
return: ; preds = %bb3.return_crit_edge, %entry
ret void
}
|