2 ; RUN: opt -passes=loop-vectorize -mtriple=riscv64 -mattr=+v -force-vector-width=2 -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefix=VF_2
3 ; RUN: opt -passes=loop-vectorize -mtriple=riscv64 -mattr=+v -force-vector-width=4 -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefix=VF_4
4 ; RUN: opt -passes=loop-vectorize -mtriple=riscv64 -mattr=+v -force-vector-width=8 -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefix=VF_8
5 ; RUN: opt -passes=loop-vectorize -mtriple=riscv64 -mattr=+v -force-vector-width=16 -debug-only=loop-vectorize -disable-output < %s 2>&1 | FileCheck %s --check-prefix=VF_16
8 define void @i8_factor_2(ptr %data, i64 %n) {
11 ; VF_2-LABEL: Checking a loop in 'i8_factor_2'
12 ; VF_2: Found an estimated cost of 2 for VF 2 For instruction: %l0 = load i8, ptr %p0, align 1
13 ; VF_2-NEXT: Found an estimated cost of 0 for VF 2 For instruction: %l1 = load i8, ptr %p1, align 1
14 ; VF_2: Found an estimated cost of 0 for VF 2 For instruction: store i8 %a0, ptr %p0, align 1
15 ; VF_2-NEXT: Found an estimated cost of 2 for VF 2 For instruction: store i8 %a1, ptr %p1, align 1
16 ; VF_4-LABEL: Checking a loop in 'i8_factor_2'
17 ; VF_4: Found an estimated cost of 2 for VF 4 For instruction: %l0 = load i8, ptr %p0, align 1
18 ; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: %l1 = load i8, ptr %p1, align 1
19 ; VF_4: Found an estimated cost of 0 for VF 4 For instruction: store i8 %a0, ptr %p0, align 1
20 ; VF_4-NEXT: Found an estimated cost of 2 for VF 4 For instruction: store i8 %a1, ptr %p1, align 1
21 ; VF_8-LABEL: Checking a loop in 'i8_factor_2'
22 ; VF_8: Found an estimated cost of 2 for VF 8 For instruction: %l0 = load i8, ptr %p0, align 1
23 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: %l1 = load i8, ptr %p1, align 1
24 ; VF_8: Found an estimated cost of 0 for VF 8 For instruction: store i8 %a0, ptr %p0, align 1
25 ; VF_8-NEXT: Found an estimated cost of 2 for VF 8 For instruction: store i8 %a1, ptr %p1, align 1
26 ; VF_16-LABEL: Checking a loop in 'i8_factor_2'
27 ; VF_16: Found an estimated cost of 3 for VF 16 For instruction: %l0 = load i8, ptr %p0, align 1
28 ; VF_16-NEXT: Found an estimated cost of 0 for VF 16 For instruction: %l1 = load i8, ptr %p1, align 1
29 ; VF_16: Found an estimated cost of 0 for VF 16 For instruction: store i8 %a0, ptr %p0, align 1
30 ; VF_16-NEXT: Found an estimated cost of 3 for VF 16 For instruction: store i8 %a1, ptr %p1, align 1
32 %i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
33 %p0 = getelementptr inbounds %i8.2, ptr %data, i64 %i, i32 0
34 %p1 = getelementptr inbounds %i8.2, ptr %data, i64 %i, i32 1
35 %l0 = load i8, ptr %p0, align 1
36 %l1 = load i8, ptr %p1, align 1
39 store i8 %a0, ptr %p0, align 1
40 store i8 %a1, ptr %p1, align 1
41 %i.next = add nuw nsw i64 %i, 1
42 %cond = icmp slt i64 %i.next, %n
43 br i1 %cond, label %for.body, label %for.end
49 %i8.3 = type {i8, i8, i8}
50 define void @i8_factor_3(ptr %data, i64 %n) {
53 ; VF_2-LABEL: Checking a loop in 'i8_factor_3'
54 ; VF_2: Found an estimated cost of 2 for VF 2 For instruction: %l0 = load i8, ptr %p0, align 1
55 ; VF_2-NEXT: Found an estimated cost of 0 for VF 2 For instruction: %l1 = load i8, ptr %p1, align 1
56 ; VF_2-NEXT: Found an estimated cost of 0 for VF 2 For instruction: %l2 = load i8, ptr %p2, align 1
57 ; VF_2: Found an estimated cost of 0 for VF 2 For instruction: store i8 %a0, ptr %p0, align 1
58 ; VF_2: Found an estimated cost of 0 for VF 2 For instruction: store i8 %a1, ptr %p1, align 1
59 ; VF_2-NEXT: Found an estimated cost of 2 for VF 2 For instruction: store i8 %a2, ptr %p2, align 1
60 ; VF_4-LABEL: Checking a loop in 'i8_factor_3'
61 ; VF_4: Found an estimated cost of 2 for VF 4 For instruction: %l0 = load i8, ptr %p0, align 1
62 ; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: %l1 = load i8, ptr %p1, align 1
63 ; VF_4-NEXT: Found an estimated cost of 0 for VF 4 For instruction: %l2 = load i8, ptr %p2, align 1
64 ; VF_4: Found an estimated cost of 0 for VF 4 For instruction: store i8 %a0, ptr %p0, align 1
65 ; VF_4: Found an estimated cost of 0 for VF 4 For instruction: store i8 %a1, ptr %p1, align 1
66 ; VF_4-NEXT: Found an estimated cost of 2 for VF 4 For instruction: store i8 %a2, ptr %p2, align 1
67 ; VF_8-LABEL: Checking a loop in 'i8_factor_3'
68 ; VF_8: Found an estimated cost of 3 for VF 8 For instruction: %l0 = load i8, ptr %p0, align 1
69 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: %l1 = load i8, ptr %p1, align 1
70 ; VF_8-NEXT: Found an estimated cost of 0 for VF 8 For instruction: %l2 = load i8, ptr %p2, align 1
71 ; VF_8: Found an estimated cost of 0 for VF 8 For instruction: store i8 %a0, ptr %p0, align 1
72 ; VF_8: Found an estimated cost of 0 for VF 8 For instruction: store i8 %a1, ptr %p1, align 1
73 ; VF_8-NEXT: Found an estimated cost of 3 for VF 8 For instruction: store i8 %a2, ptr %p2, align 1
74 ; VF_16-LABEL: Checking a loop in 'i8_factor_3'
75 ; VF_16: Found an estimated cost of 48 for VF 16 For instruction: %l0 = load i8, ptr %p0, align 1
76 ; VF_16-NEXT: Found an estimated cost of 0 for VF 16 For instruction: %l1 = load i8, ptr %p1, align 1
77 ; VF_16-NEXT: Found an estimated cost of 0 for VF 16 For instruction: %l2 = load i8, ptr %p2, align 1
78 ; VF_16: Found an estimated cost of 0 for VF 16 For instruction: store i8 %a0, ptr %p0, align 1
79 ; VF_16: Found an estimated cost of 0 for VF 16 For instruction: store i8 %a1, ptr %p1, align 1
80 ; VF_16-NEXT: Found an estimated cost of 48 for VF 16 For instruction: store i8 %a2, ptr %p2, align 1
82 %i = phi i64 [ 0, %entry ], [ %i.next, %for.body ]
83 %p0 = getelementptr inbounds %i8.3, ptr %data, i64 %i, i32 0
84 %p1 = getelementptr inbounds %i8.3, ptr %data, i64 %i, i32 1
85 %p2 = getelementptr inbounds %i8.3, ptr %data, i64 %i, i32 2
86 %l0 = load i8, ptr %p0, align 1
87 %l1 = load i8, ptr %p1, align 1
88 %l2 = load i8, ptr %p2, align 1
92 store i8 %a0, ptr %p0, align 1
93 store i8 %a1, ptr %p1, align 1
94 store i8 %a2, ptr %p2, align 1
95 %i.next = add nuw nsw i64 %i, 1
96 %cond = icmp slt i64 %i.next, %n
97 br i1 %cond, label %for.body, label %for.end