104 lines · plain
1; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 42; RUN: opt < %s -mtriple=riscv32 -mattr=+v -passes="print<cost-model>" -cost-kind=throughput 2>&1 -disable-output | FileCheck %s3; RUN: opt < %s -mtriple=riscv64 -mattr=+v -passes="print<cost-model>" -cost-kind=throughput 2>&1 -disable-output | FileCheck %s4; RUN: opt < %s -mtriple=riscv32 -mattr=+v -passes="print<cost-model>" -cost-kind=code-size 2>&1 -disable-output | FileCheck %s --check-prefix=SIZE5; RUN: opt < %s -mtriple=riscv64 -mattr=+v -passes="print<cost-model>" -cost-kind=code-size 2>&1 -disable-output | FileCheck %s --check-prefix=SIZE6 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 undef24;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 undef41;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 undef57}58 59define double @reduce_fmaximum_f64(double %arg) {60; CHECK-LABEL: 'reduce_fmaximum_f64'61; CHECK-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V2 = call double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)62; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4 = call double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)63; CHECK-NEXT: Cost Model: Found an estimated cost of 11 for instruction: %V8 = call double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)64; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V16 = call double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)65; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V32 = call double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)66; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V64 = call double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)67; 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)68; 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)69; 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)70; 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)71; 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)72; 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)73; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret double undef74;75; SIZE-LABEL: 'reduce_fmaximum_f64'76; SIZE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V2 = call double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)77; SIZE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V4 = call double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)78; SIZE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V8 = call double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)79; SIZE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V16 = call double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)80; SIZE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V32 = call double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)81; SIZE-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V64 = call double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)82; 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)83; 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)84; 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)85; 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)86; 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)87; 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)88; SIZE-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret double undef89;90%V2 = call double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)91%V4 = call double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)92%V8 = call double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)93%V16 = call double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)94%V32 = call double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)95%V64 = call double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)96call fast double @llvm.vector.reduce.fmaximum.v2f64(<2 x double> undef)97call fast double @llvm.vector.reduce.fmaximum.v4f64(<4 x double> undef)98call fast double @llvm.vector.reduce.fmaximum.v8f64(<8 x double> undef)99call fast double @llvm.vector.reduce.fmaximum.v16f64(<16 x double> undef)100call fast double @llvm.vector.reduce.fmaximum.v32f64(<32 x double> undef)101call fast double @llvm.vector.reduce.fmaximum.v64f64(<64 x double> undef)102ret double undef103}104