xref: /llvm-project/llvm/test/Analysis/CostModel/RISCV/reduce-fmaximum.ll (revision 400b725c2740ee29560bac9ad870b4ddb56bf3f2) 
1; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 4
2; RUN: opt < %s -mtriple=riscv32 -mattr=+v -passes="print<cost-model>" -cost-kind=throughput 2>&1 -disable-output | FileCheck %s
3; RUN: opt < %s -mtriple=riscv64 -mattr=+v -passes="print<cost-model>" -cost-kind=throughput 2>&1 -disable-output | FileCheck %s
4; RUN: opt < %s -mtriple=riscv32 -mattr=+v -passes="print<cost-model>" -cost-kind=code-size 2>&1 -disable-output | FileCheck %s --check-prefix=SIZE
5; RUN: opt < %s -mtriple=riscv64 -mattr=+v -passes="print<cost-model>" -cost-kind=code-size 2>&1 -disable-output | FileCheck %s --check-prefix=SIZE
6
7define float @reduce_fmaximum_f32(float %arg) {
8; CHECK-LABEL: 'reduce_fmaximum_f32'
9; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
10; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
11; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8 = call float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
12; CHECK-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V16 = call float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
13; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V32 = call float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
14; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V64 = call float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
15; CHECK-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V128 = call float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
16; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call fast float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
17; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %2 = call fast float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
18; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %3 = call fast float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
19; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %4 = call fast float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
20; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %5 = call fast float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
21; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %6 = call fast float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
22; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %7 = call fast float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
23; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret float undef
24;
25; SIZE-LABEL: 'reduce_fmaximum_f32'
26; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
27; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
28; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
29; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16 = call float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
30; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V32 = call float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
31; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V64 = call float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
32; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V128 = call float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
33; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call fast float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
34; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call fast float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
35; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call fast float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
36; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call fast float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
37; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call fast float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
38; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call fast float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
39; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call fast float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
40; SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret float undef
41;
42%V2   = call float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
43%V4   = call float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
44%V8   = call float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
45%V16   = call float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
46%V32   = call float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
47%V64   = call float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
48%V128   = call float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
49call fast float @llvm.vector.reduce.fmaximum.v2f32(<2 x float> undef)
50call fast float @llvm.vector.reduce.fmaximum.v4f32(<4 x float> undef)
51call fast float @llvm.vector.reduce.fmaximum.v8f32(<8 x float> undef)
52call fast float @llvm.vector.reduce.fmaximum.v16f32(<16 x float> undef)
53call fast float @llvm.vector.reduce.fmaximum.v32f32(<32 x float> undef)
54call fast float @llvm.vector.reduce.fmaximum.v64f32(<64 x float> undef)
55call fast float @llvm.vector.reduce.fmaximum.v128f32(<128 x float> undef)
56ret float undef
57}
58declare float @llvm.vector.reduce.fmaximum.v2f32(<2 x float>)
59declare float @llvm.vector.reduce.fmaximum.v4f32(<4 x float>)
60declare float @llvm.vector.reduce.fmaximum.v8f32(<8 x float>)
61declare float @llvm.vector.reduce.fmaximum.v16f32(<16 x float>)
62declare float @llvm.vector.reduce.fmaximum.v32f32(<32 x float>)
63declare float @llvm.vector.reduce.fmaximum.v64f32(<64 x float>)
64declare float @llvm.vector.reduce.fmaximum.v128f32(<128 x float>)
65
66define double @reduce_fmaximum_f64(double %arg) {
67; CHECK-LABEL: 'reduce_fmaximum_f64'
68; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2 = call double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
69; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4 = call double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
70; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V8 = call double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
71; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16 = call double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
72; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32 = call double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
73; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V64 = call double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
74; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call fast double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
75; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %2 = call fast double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
76; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %3 = call fast double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
77; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %4 = call fast double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
78; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %5 = call fast double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
79; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %6 = call fast double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
80; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret double undef
81;
82; SIZE-LABEL: 'reduce_fmaximum_f64'
83; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2 = call double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
84; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4 = call double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
85; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8 = call double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
86; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16 = call double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
87; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V32 = call double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
88; SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V64 = call double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
89; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call fast double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
90; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %2 = call fast double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
91; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call fast double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
92; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call fast double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
93; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call fast double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
94; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call fast double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
95; SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret double undef
96;
97%V2   = call double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
98%V4   = call double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
99%V8   = call double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
100%V16   = call double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
101%V32   = call double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
102%V64   = call double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
103call fast double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)
104call fast double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)
105call fast double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)
106call fast double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)
107call fast double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)
108call fast double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)
109ret double undef
110}
111declare double @llvm.vector.reduce.fmaximum.v2f64(<2 x double>)
112declare double @llvm.vector.reduce.fmaximum.v4f64(<4 x double>)
113declare double @llvm.vector.reduce.fmaximum.v8f64(<8 x double>)
114declare double @llvm.vector.reduce.fmaximum.v16f64(<16 x double>)
115declare double @llvm.vector.reduce.fmaximum.v32f64(<32 x double>)
116declare double @llvm.vector.reduce.fmaximum.v64f64(<64 x double>)
117