1 ; RUN: opt -loop-vectorize -force-vector-width=2 -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefix=VF_2
2 ; RUN: opt -loop-vectorize -force-vector-width=4 -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefix=VF_4
3 ; RUN: opt -loop-vectorize -force-vector-width=8 -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefix=VF_8
4 ; RUN: opt -loop-vectorize -force-vector-width=16 -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefix=VF_16
7 target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
8 target triple = "armv8--linux-gnueabihf"
11 define void @i8_factor_2(%i8.2* %data, i64 %n) {
15 ; VF_8-LABEL: Checking a loop in "i8_factor_2"
16 ; VF_8: Found an estimated cost of 2 for VF 8 For instruction: %tmp2 = load i8, i8* %tmp0, align 1
17 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: %tmp3 = load i8, i8* %tmp1, align 1
18 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: store i8 0, i8* %tmp0, align 1
19 ; VF_8-NEXT: Found an estimated cost of 2 for VF 8 For instruction: store i8 0, i8* %tmp1, align 1
20 ; VF_16-LABEL: Checking a loop in "i8_factor_2"
21 ; VF_16: Found an estimated cost of 2 for VF 16 For instruction: %tmp2 = load i8, i8* %tmp0, align 1
22 ; VF_16-NEXT: Found an estimated cost of 0 for VF 16 For instruction: %tmp3 = load i8, i8* %tmp1, align 1
23 ; VF_16-NEXT: Found an estimated cost of 0 for VF 16 For instruction: store i8 0, i8* %tmp0, align 1
24 ; VF_16-NEXT: Found an estimated cost of 2 for VF 16 For instruction: store i8 0, i8* %tmp1, align 1
26 %i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
27 %tmp0 = getelementptr inbounds %i8.2, %i8.2* %data, i64 %i, i32 0
28 %tmp1 = getelementptr inbounds %i8.2, %i8.2* %data, i64 %i, i32 1
29 %tmp2 = load i8, i8* %tmp0, align 1
30 %tmp3 = load i8, i8* %tmp1, align 1
31 store i8 0, i8* %tmp0, align 1
32 store i8 0, i8* %tmp1, align 1
33 %i.next = add nuw nsw i64 %i, 1
34 %cond = icmp slt i64 %i.next, %n
35 br i1 %cond, label %for.body, label %for.end
41 %i16.2 = type {i16, i16}
42 define void @i16_factor_2(%i16.2* %data, i64 %n) {
46 ; VF_4-LABEL: Checking a loop in "i16_factor_2"
47 ; VF_4: Found an estimated cost of 2 for VF 4 For instruction: %tmp2 = load i16, i16* %tmp0, align 2
48 ; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: %tmp3 = load i16, i16* %tmp1, align 2
49 ; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: store i16 0, i16* %tmp0, align 2
50 ; VF_4-NEXT: Found an estimated cost of 2 for VF 4 For instruction: store i16 0, i16* %tmp1, align 2
51 ; VF_8-LABEL: Checking a loop in "i16_factor_2"
52 ; VF_8: Found an estimated cost of 2 for VF 8 For instruction: %tmp2 = load i16, i16* %tmp0, align 2
53 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: %tmp3 = load i16, i16* %tmp1, align 2
54 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: store i16 0, i16* %tmp0, align 2
55 ; VF_8-NEXT: Found an estimated cost of 2 for VF 8 For instruction: store i16 0, i16* %tmp1, align 2
56 ; VF_16-LABEL: Checking a loop in "i16_factor_2"
57 ; VF_16: Found an estimated cost of 4 for VF 16 For instruction: %tmp2 = load i16, i16* %tmp0, align 2
58 ; VF_16-NEXT: Found an estimated cost of 0 for VF 16 For instruction: %tmp3 = load i16, i16* %tmp1, align 2
59 ; VF_16-NEXT: Found an estimated cost of 0 for VF 16 For instruction: store i16 0, i16* %tmp0, align 2
60 ; VF_16-NEXT: Found an estimated cost of 4 for VF 16 For instruction: store i16 0, i16* %tmp1, align 2
62 %i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
63 %tmp0 = getelementptr inbounds %i16.2, %i16.2* %data, i64 %i, i32 0
64 %tmp1 = getelementptr inbounds %i16.2, %i16.2* %data, i64 %i, i32 1
65 %tmp2 = load i16, i16* %tmp0, align 2
66 %tmp3 = load i16, i16* %tmp1, align 2
67 store i16 0, i16* %tmp0, align 2
68 store i16 0, i16* %tmp1, align 2
69 %i.next = add nuw nsw i64 %i, 1
70 %cond = icmp slt i64 %i.next, %n
71 br i1 %cond, label %for.body, label %for.end
77 %i32.2 = type {i32, i32}
78 define void @i32_factor_2(%i32.2* %data, i64 %n) {
82 ; VF_2-LABEL: Checking a loop in "i32_factor_2"
83 ; VF_2: Found an estimated cost of 2 for VF 2 For instruction: %tmp2 = load i32, i32* %tmp0, align 4
84 ; VF_2-NEXT: Found an estimated cost of 0 for VF 2 For instruction: %tmp3 = load i32, i32* %tmp1, align 4
85 ; VF_2-NEXT: Found an estimated cost of 0 for VF 2 For instruction: store i32 0, i32* %tmp0, align 4
86 ; VF_2-NEXT: Found an estimated cost of 2 for VF 2 For instruction: store i32 0, i32* %tmp1, align 4
87 ; VF_4-LABEL: Checking a loop in "i32_factor_2"
88 ; VF_4: Found an estimated cost of 2 for VF 4 For instruction: %tmp2 = load i32, i32* %tmp0, align 4
89 ; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: %tmp3 = load i32, i32* %tmp1, align 4
90 ; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: store i32 0, i32* %tmp0, align 4
91 ; VF_4-NEXT: Found an estimated cost of 2 for VF 4 For instruction: store i32 0, i32* %tmp1, align 4
92 ; VF_8-LABEL: Checking a loop in "i32_factor_2"
93 ; VF_8: Found an estimated cost of 4 for VF 8 For instruction: %tmp2 = load i32, i32* %tmp0, align 4
94 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: %tmp3 = load i32, i32* %tmp1, align 4
95 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: store i32 0, i32* %tmp0, align 4
96 ; VF_8-NEXT: Found an estimated cost of 4 for VF 8 For instruction: store i32 0, i32* %tmp1, align 4
97 ; VF_16-LABEL: Checking a loop in "i32_factor_2"
98 ; VF_16: Found an estimated cost of 8 for VF 16 For instruction: %tmp2 = load i32, i32* %tmp0, align 4
99 ; VF_16-NEXT: Found an estimated cost of 0 for VF 16 For instruction: %tmp3 = load i32, i32* %tmp1, align 4
100 ; VF_16-NEXT: Found an estimated cost of 0 for VF 16 For instruction: store i32 0, i32* %tmp0, align 4
101 ; VF_16-NEXT: Found an estimated cost of 8 for VF 16 For instruction: store i32 0, i32* %tmp1, align 4
103 %i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
104 %tmp0 = getelementptr inbounds %i32.2, %i32.2* %data, i64 %i, i32 0
105 %tmp1 = getelementptr inbounds %i32.2, %i32.2* %data, i64 %i, i32 1
106 %tmp2 = load i32, i32* %tmp0, align 4
107 %tmp3 = load i32, i32* %tmp1, align 4
108 store i32 0, i32* %tmp0, align 4
109 store i32 0, i32* %tmp1, align 4
110 %i.next = add nuw nsw i64 %i, 1
111 %cond = icmp slt i64 %i.next, %n
112 br i1 %cond, label %for.body, label %for.end
118 %half.2 = type {half, half}
119 define void @half_factor_2(%half.2* %data, i64 %n) {
123 ; VF_4-LABEL: Checking a loop in "half_factor_2"
124 ; VF_4: Found an estimated cost of 40 for VF 4 For instruction: %tmp2 = load half, half* %tmp0, align 2
125 ; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: %tmp3 = load half, half* %tmp1, align 2
126 ; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: store half 0xH0000, half* %tmp0, align 2
127 ; VF_4-NEXT: Found an estimated cost of 32 for VF 4 For instruction: store half 0xH0000, half* %tmp1, align 2
128 ; VF_8-LABEL: Checking a loop in "half_factor_2"
129 ; VF_8: Found an estimated cost of 80 for VF 8 For instruction: %tmp2 = load half, half* %tmp0, align 2
130 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: %tmp3 = load half, half* %tmp1, align 2
131 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: store half 0xH0000, half* %tmp0, align 2
132 ; VF_8-NEXT: Found an estimated cost of 64 for VF 8 For instruction: store half 0xH0000, half* %tmp1, align 2
134 %i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
135 %tmp0 = getelementptr inbounds %half.2, %half.2* %data, i64 %i, i32 0
136 %tmp1 = getelementptr inbounds %half.2, %half.2* %data, i64 %i, i32 1
137 %tmp2 = load half, half* %tmp0, align 2
138 %tmp3 = load half, half* %tmp1, align 2
139 store half 0., half* %tmp0, align 2
140 store half 0., half* %tmp1, align 2
141 %i.next = add nuw nsw i64 %i, 1
142 %cond = icmp slt i64 %i.next, %n
143 br i1 %cond, label %for.body, label %for.end