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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -S -passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu -mcpu=cascadelake < %s | FileCheck %s
define void @test() {
; CHECK-LABEL: @test(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[ARRAYIDX10_I_I86:%.*]] = getelementptr inbounds float, ptr undef, i64 2
; CHECK-NEXT: [[ARRAYIDX21_I:%.*]] = getelementptr inbounds [4 x float], ptr undef, i64 2
; CHECK-NEXT: br label [[BB1:%.*]]
; CHECK: bb1:
; CHECK-NEXT: [[TMP0:%.*]] = load <2 x float>, ptr undef, align 4
; CHECK-NEXT: [[TMP1:%.*]] = fsub <2 x float> zeroinitializer, [[TMP0]]
; CHECK-NEXT: [[TMP2:%.*]] = load float, ptr [[ARRAYIDX10_I_I86]], align 4
; CHECK-NEXT: [[TMP3:%.*]] = load float, ptr undef, align 4
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <2 x float> [[TMP0]], <2 x float> <float poison, float 0.000000e+00>, <2 x i32> <i32 1, i32 3>
; CHECK-NEXT: [[TMP5:%.*]] = insertelement <2 x float> poison, float [[TMP2]], i32 0
; CHECK-NEXT: [[TMP6:%.*]] = insertelement <2 x float> [[TMP5]], float [[TMP3]], i32 1
; CHECK-NEXT: [[TMP7:%.*]] = shufflevector <2 x float> [[TMP6]], <2 x float> <float 0.000000e+00, float poison>, <2 x i32> <i32 2, i32 0>
; CHECK-NEXT: [[TMP8:%.*]] = call <2 x float> @llvm.fmuladd.v2f32(<2 x float> [[TMP4]], <2 x float> [[TMP6]], <2 x float> [[TMP7]])
; CHECK-NEXT: br i1 false, label [[BB2:%.*]], label [[BB3:%.*]]
; CHECK: bb2:
; CHECK-NEXT: [[TMP9:%.*]] = fmul <2 x float> [[TMP8]], zeroinitializer
; CHECK-NEXT: br label [[BB3]]
; CHECK: bb3:
; CHECK-NEXT: [[TMP10:%.*]] = phi <2 x float> [ [[TMP9]], [[BB2]] ], [ zeroinitializer, [[BB1]] ]
; CHECK-NEXT: [[TMP11:%.*]] = fadd <2 x float> [[TMP1]], [[TMP10]]
; CHECK-NEXT: [[TMP12:%.*]] = fadd <2 x float> [[TMP11]], zeroinitializer
; CHECK-NEXT: [[TMP13:%.*]] = fsub <2 x float> [[TMP12]], zeroinitializer
; CHECK-NEXT: [[TMP14:%.*]] = fsub <2 x float> [[TMP13]], zeroinitializer
; CHECK-NEXT: store <2 x float> [[TMP14]], ptr [[ARRAYIDX21_I]], align 16
; CHECK-NEXT: ret void
;
entry:
%arrayidx10.i.i86 = getelementptr inbounds float, ptr undef, i64 2
%arrayidx6.i66.i = getelementptr inbounds float, ptr undef, i64 1
%arrayidx21.i = getelementptr inbounds [4 x float], ptr undef, i64 2
%arrayidx6.i109.i = getelementptr inbounds [4 x float], ptr undef, i64 2, i64 1
br label %bb1
bb1:
%0 = load float, ptr undef, align 4
%sub.i71.i = fsub float 0.000000e+00, %0
%1 = load float, ptr %arrayidx6.i66.i, align 4
%sub5.i74.i = fsub float 0.000000e+00, %1
%2 = load float, ptr %arrayidx10.i.i86, align 4
%3 = call float @llvm.fmuladd.f32(float %1, float %2, float 0.000000e+00)
%4 = load float, ptr undef, align 4
%5 = call float @llvm.fmuladd.f32(float 0.000000e+00, float %4, float %2)
br i1 false, label %bb2, label %bb3
bb2:
%mul.i95 = fmul float %3, 0.000000e+00
%mul3.i96 = fmul float %5, 0.000000e+00
br label %bb3
bb3:
%vddir.sroa.8.0.i = phi float [ %mul3.i96, %bb2 ], [ 0.000000e+00, %bb1 ]
%vddir.sroa.0.0.i = phi float [ %mul.i95, %bb2 ], [ 0.000000e+00, %bb1 ]
%add.i.i = fadd float %sub.i71.i, %vddir.sroa.0.0.i
%add5.i.i = fadd float %sub5.i74.i, %vddir.sroa.8.0.i
%add.i105.i = fadd float %add.i.i, 0.000000e+00
%add5.i108.i = fadd float %add5.i.i, 0.000000e+00
%sub.i114.i = fsub float %add.i105.i, 0.000000e+00
%sub4.i.i = fsub float %add5.i108.i, 0.000000e+00
%sub.i118.i = fsub float %sub.i114.i, 0.000000e+00
store float %sub.i118.i, ptr %arrayidx21.i, align 16
%sub4.i121.i = fsub float %sub4.i.i, 0.000000e+00
store float %sub4.i121.i, ptr %arrayidx6.i109.i, align 4
ret void
}
declare float @llvm.fmuladd.f32(float, float, float)
|