1; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
2; RUN: opt < %s -mtriple=riscv32 -mattr=+v -riscv-v-vector-bits-min=256 -passes='print<cost-model>' -cost-kind=throughput 2>&1 -disable-output | FileCheck %s
3; RUN: opt < %s -mtriple=riscv64 -mattr=+v -riscv-v-vector-bits-min=256 -passes='print<cost-model>' -cost-kind=throughput 2>&1 -disable-output | FileCheck %s
4
5define i32 @reduce_i1(i32 %arg) {
6; CHECK-LABEL: 'reduce_i1'
7; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i1 @llvm.vector.reduce.add.v1i1(<1 x i1> undef)
8; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i1 @llvm.vector.reduce.add.v2i1(<2 x i1> undef)
9; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i1 @llvm.vector.reduce.add.v4i1(<4 x i1> undef)
10; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i1 @llvm.vector.reduce.add.v8i1(<8 x i1> undef)
11; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16 = call i1 @llvm.vector.reduce.add.v16i1(<16 x i1> undef)
12; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32 = call i1 @llvm.vector.reduce.add.v32i1(<32 x i1> undef)
13; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64 = call i1 @llvm.vector.reduce.add.v64i1(<64 x i1> undef)
14; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V128 = call i1 @llvm.vector.reduce.add.v128i1(<128 x i1> undef)
15; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
16;
17  %V1   = call i1 @llvm.vector.reduce.add.v1i1(<1 x i1> undef)
18  %V2   = call i1 @llvm.vector.reduce.add.v2i1(<2 x i1> undef)
19  %V4   = call i1 @llvm.vector.reduce.add.v4i1(<4 x i1> undef)
20  %V8   = call i1 @llvm.vector.reduce.add.v8i1(<8 x i1> undef)
21  %V16  = call i1 @llvm.vector.reduce.add.v16i1(<16 x i1> undef)
22  %V32  = call i1 @llvm.vector.reduce.add.v32i1(<32 x i1> undef)
23  %V64  = call i1 @llvm.vector.reduce.add.v64i1(<64 x i1> undef)
24  %V128 = call i1 @llvm.vector.reduce.add.v128i1(<128 x i1> undef)
25  ret i32 undef
26}
27
28define i32 @reduce_i8(i32 %arg) {
29; CHECK-LABEL: 'reduce_i8'
30; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i8 @llvm.vector.reduce.add.v1i8(<1 x i8> undef)
31; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
32; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
33; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
34; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
35; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i8 @llvm.vector.reduce.add.v32i8(<32 x i8> undef)
36; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.vector.reduce.add.v64i8(<64 x i8> undef)
37; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i8 @llvm.vector.reduce.add.v128i8(<128 x i8> undef)
38; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
39;
40  %V1   = call i8 @llvm.vector.reduce.add.v1i8(<1 x i8> undef)
41  %V2   = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
42  %V4   = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
43  %V8   = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
44  %V16  = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
45  %V32  = call i8 @llvm.vector.reduce.add.v32i8(<32 x i8> undef)
46  %V64  = call i8 @llvm.vector.reduce.add.v64i8(<64 x i8> undef)
47  %V128 = call i8 @llvm.vector.reduce.add.v128i8(<128 x i8> undef)
48  ret i32 undef
49}
50
51define i32 @reduce_i16(i32 %arg) {
52; CHECK-LABEL: 'reduce_i16'
53; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i16 @llvm.vector.reduce.add.v1i16(<1 x i16> undef)
54; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i16 @llvm.vector.reduce.add.v2i16(<2 x i16> undef)
55; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i16 @llvm.vector.reduce.add.v4i16(<4 x i16> undef)
56; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i16 @llvm.vector.reduce.add.v8i16(<8 x i16> undef)
57; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i16 @llvm.vector.reduce.add.v16i16(<16 x i16> undef)
58; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i16 @llvm.vector.reduce.add.v32i16(<32 x i16> undef)
59; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i16 @llvm.vector.reduce.add.v64i16(<64 x i16> undef)
60; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V128 = call i16 @llvm.vector.reduce.add.v128i16(<128 x i16> undef)
61; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
62;
63  %V1   = call i16 @llvm.vector.reduce.add.v1i16(<1 x i16> undef)
64  %V2   = call i16 @llvm.vector.reduce.add.v2i16(<2 x i16> undef)
65  %V4   = call i16 @llvm.vector.reduce.add.v4i16(<4 x i16> undef)
66  %V8   = call i16 @llvm.vector.reduce.add.v8i16(<8 x i16> undef)
67  %V16  = call i16 @llvm.vector.reduce.add.v16i16(<16 x i16> undef)
68  %V32  = call i16 @llvm.vector.reduce.add.v32i16(<32 x i16> undef)
69  %V64  = call i16 @llvm.vector.reduce.add.v64i16(<64 x i16> undef)
70  %V128 = call i16 @llvm.vector.reduce.add.v128i16(<128 x i16> undef)
71  ret i32 undef
72}
73
74define i32 @reduce_i32(i32 %arg) {
75; CHECK-LABEL: 'reduce_i32'
76; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i32 @llvm.vector.reduce.add.v1i32(<1 x i32> undef)
77; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i32 @llvm.vector.reduce.add.v2i32(<2 x i32> undef)
78; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> undef)
79; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> undef)
80; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> undef)
81; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i32 @llvm.vector.reduce.add.v32i32(<32 x i32> undef)
82; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i32 @llvm.vector.reduce.add.v64i32(<64 x i32> undef)
83; CHECK-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V128 = call i32 @llvm.vector.reduce.add.v128i32(<128 x i32> undef)
84; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
85;
86  %V1   = call i32 @llvm.vector.reduce.add.v1i32(<1 x i32> undef)
87  %V2   = call i32 @llvm.vector.reduce.add.v2i32(<2 x i32> undef)
88  %V4   = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> undef)
89  %V8   = call i32 @llvm.vector.reduce.add.v8i32(<8 x i32> undef)
90  %V16  = call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> undef)
91  %V32  = call i32 @llvm.vector.reduce.add.v32i32(<32 x i32> undef)
92  %V64  = call i32 @llvm.vector.reduce.add.v64i32(<64 x i32> undef)
93  %V128 = call i32 @llvm.vector.reduce.add.v128i32(<128 x i32> undef)
94  ret i32 undef
95}
96
97define i32 @reduce_i64(i32 %arg) {
98; CHECK-LABEL: 'reduce_i64'
99; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1 = call i64 @llvm.vector.reduce.add.v1i64(<1 x i64> undef)
100; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2 = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
101; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
102; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8 = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
103; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16 = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
104; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V32 = call i64 @llvm.vector.reduce.add.v32i64(<32 x i64> undef)
105; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V64 = call i64 @llvm.vector.reduce.add.v64i64(<64 x i64> undef)
106; CHECK-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V128 = call i64 @llvm.vector.reduce.add.v128i64(<128 x i64> undef)
107; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
108;
109  %V1   = call i64 @llvm.vector.reduce.add.v1i64(<1 x i64> undef)
110  %V2   = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
111  %V4   = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
112  %V8   = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
113  %V16  = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
114  %V32  = call i64 @llvm.vector.reduce.add.v32i64(<32 x i64> undef)
115  %V64  = call i64 @llvm.vector.reduce.add.v64i64(<64 x i64> undef)
116  %V128 = call i64 @llvm.vector.reduce.add.v128i64(<128 x i64> undef)
117  ret i32 undef
118}
119
120declare i1 @llvm.vector.reduce.add.v1i1(<1 x i1>)
121declare i1 @llvm.vector.reduce.add.v2i1(<2 x i1>)
122declare i1 @llvm.vector.reduce.add.v4i1(<4 x i1>)
123declare i1 @llvm.vector.reduce.add.v8i1(<8 x i1>)
124declare i1 @llvm.vector.reduce.add.v16i1(<16 x i1>)
125declare i1 @llvm.vector.reduce.add.v32i1(<32 x i1>)
126declare i1 @llvm.vector.reduce.add.v64i1(<64 x i1>)
127declare i1 @llvm.vector.reduce.add.v128i1(<128 x i1>)
128declare i8 @llvm.vector.reduce.add.v1i8(<1 x i8>)
129declare i8 @llvm.vector.reduce.add.v2i8(<2 x i8>)
130declare i8 @llvm.vector.reduce.add.v4i8(<4 x i8>)
131declare i8 @llvm.vector.reduce.add.v8i8(<8 x i8>)
132declare i8 @llvm.vector.reduce.add.v16i8(<16 x i8>)
133declare i8 @llvm.vector.reduce.add.v32i8(<32 x i8>)
134declare i8 @llvm.vector.reduce.add.v64i8(<64 x i8>)
135declare i8 @llvm.vector.reduce.add.v128i8(<128 x i8>)
136declare i16 @llvm.vector.reduce.add.v1i16(<1 x i16>)
137declare i16 @llvm.vector.reduce.add.v2i16(<2 x i16>)
138declare i16 @llvm.vector.reduce.add.v4i16(<4 x i16>)
139declare i16 @llvm.vector.reduce.add.v8i16(<8 x i16>)
140declare i16 @llvm.vector.reduce.add.v16i16(<16 x i16>)
141declare i16 @llvm.vector.reduce.add.v32i16(<32 x i16>)
142declare i16 @llvm.vector.reduce.add.v64i16(<64 x i16>)
143declare i16 @llvm.vector.reduce.add.v128i16(<128 x i16>)
144declare i32 @llvm.vector.reduce.add.v1i32(<1 x i32>)
145declare i32 @llvm.vector.reduce.add.v2i32(<2 x i32>)
146declare i32 @llvm.vector.reduce.add.v4i32(<4 x i32>)
147declare i32 @llvm.vector.reduce.add.v8i32(<8 x i32>)
148declare i32 @llvm.vector.reduce.add.v16i32(<16 x i32>)
149declare i32 @llvm.vector.reduce.add.v32i32(<32 x i32>)
150declare i32 @llvm.vector.reduce.add.v64i32(<64 x i32>)
151declare i32 @llvm.vector.reduce.add.v128i32(<128 x i32>)
152declare i64 @llvm.vector.reduce.add.v1i64(<1 x i64>)
153declare i64 @llvm.vector.reduce.add.v2i64(<2 x i64>)
154declare i64 @llvm.vector.reduce.add.v4i64(<4 x i64>)
155declare i64 @llvm.vector.reduce.add.v8i64(<8 x i64>)
156declare i64 @llvm.vector.reduce.add.v16i64(<16 x i64>)
157declare i64 @llvm.vector.reduce.add.v32i64(<32 x i64>)
158declare i64 @llvm.vector.reduce.add.v64i64(<64 x i64>)
159declare i64 @llvm.vector.reduce.add.v128i64(<128 x i64>)
160