[AArch64] Fix cost modelling for SVE Min/Max Intrinsics

[chuongg3]

Summary

Add more legal types for SMIN, SMAX, UMIN, UMAX in cost modelling for AArch64

[dmgreen]

Thanks, Looks good. Can you split out the tests? I think this has included the two patches together.

And remember to upload with context. https://llvm.org/docs/DeveloperPolicy.html#making-and-submitting-a-patch

llvm/test/Analysis/CostModel/AArch64/sve-min-max.ll

2

This can remove the --check-prefixes=CHECK,CHECK-NOF16, as it won't be needed for sve tests.

346

And so this can be removed too.

[chuongg3]

Removed arguments for the min/max test

[dmgreen]

Thanks. The SVE types could be predicated on hasSVE, but it doesn't make much sense to cost them without it.

LGTM

llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp

AArch64TTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
                                       TTI::TargetCostKind CostKind) {
   auto *RetTy = ICA.getReturnType();
   switch (ICA.getID()) {
   case Intrinsic::umin:
   case Intrinsic::umax:
   case Intrinsic::smin:
   case Intrinsic::smax: {
     static const auto ValidMinMaxTys = {MVT::v8i8,  MVT::v16i8, MVT::v4i16,
-                                        MVT::v8i16, MVT::v2i32, MVT::v4i32};
+                                        MVT::v8i16, MVT::v2i32, MVT::v4i32,
+                                        MVT::nxv16i8, MVT::nxv8i16, MVT::nxv4i32,
+                                        MVT::nxv2i64};
     auto LT = getTypeLegalizationCost(RetTy);
     // v2i64 types get converted to cmp+bif hence the cost of 2
     if (LT.second == MVT::v2i64)
       return LT.first * 2;
     if (any_of(ValidMinMaxTys, [&LT](MVT M) { return M == LT.second; }))
       return LT.first;
     break;
   }

llvm/test/Analysis/CostModel/AArch64/sve-intrinsics.ll

 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %mul_nxv4i64 = call i64 @llvm.vector.reduce.mul.nxv4i64(<vscale x 4 x i64> %v1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %and_nxv4i32 = call i32 @llvm.vector.reduce.and.nxv4i32(<vscale x 4 x i32> %v0)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %and_nxv4i64 = call i64 @llvm.vector.reduce.and.nxv4i64(<vscale x 4 x i64> %v1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %or_nxv4i32 = call i32 @llvm.vector.reduce.or.nxv4i32(<vscale x 4 x i32> %v0)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %or_nxv4i64 = call i64 @llvm.vector.reduce.or.nxv4i64(<vscale x 4 x i64> %v1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %xor_nxv4i32 = call i32 @llvm.vector.reduce.xor.nxv4i32(<vscale x 4 x i32> %v0)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %xor_nxv4i64 = call i64 @llvm.vector.reduce.xor.nxv4i64(<vscale x 4 x i64> %v1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %umin_nxv4i32 = call i32 @llvm.vector.reduce.umin.nxv4i32(<vscale x 4 x i32> %v0)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %umin_nxv4i64 = call i64 @llvm.vector.reduce.umin.nxv4i64(<vscale x 4 x i64> %v1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %umin_nxv4i64 = call i64 @llvm.vector.reduce.umin.nxv4i64(<vscale x 4 x i64> %v1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %smin_nxv4i32 = call i32 @llvm.vector.reduce.smin.nxv4i32(<vscale x 4 x i32> %v0)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %smin_nxv4i64 = call i64 @llvm.vector.reduce.smin.nxv4i64(<vscale x 4 x i64> %v1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %smin_nxv4i64 = call i64 @llvm.vector.reduce.smin.nxv4i64(<vscale x 4 x i64> %v1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %umax_nxv4i32 = call i32 @llvm.vector.reduce.umax.nxv4i32(<vscale x 4 x i32> %v0)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %umax_nxv4i64 = call i64 @llvm.vector.reduce.umax.nxv4i64(<vscale x 4 x i64> %v1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %umax_nxv4i64 = call i64 @llvm.vector.reduce.umax.nxv4i64(<vscale x 4 x i64> %v1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %smax_nxv4i32 = call i32 @llvm.vector.reduce.smax.nxv4i32(<vscale x 4 x i32> %v0)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %smax_nxv4i64 = call i64 @llvm.vector.reduce.smax.nxv4i64(<vscale x 4 x i64> %v1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %smax_nxv4i64 = call i64 @llvm.vector.reduce.smax.nxv4i64(<vscale x 4 x i64> %v1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %fadd_nxv4f32 = call fast float @llvm.vector.reduce.fadd.nxv4f32(float 0.000000e+00, <vscale x 4 x float> %v2)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %fadd_nxv4f64 = call fast double @llvm.vector.reduce.fadd.nxv4f64(double 0.000000e+00, <vscale x 4 x double> %v3)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %fmin_nxv4f32 = call fast float @llvm.vector.reduce.fmin.nxv4f32(<vscale x 4 x float> %v2)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %fmin_nxv4f64 = call fast double @llvm.vector.reduce.fmin.nxv4f64(<vscale x 4 x double> %v3)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %fmax_nxv4f32 = call fast float @llvm.vector.reduce.fmax.nxv4f32(<vscale x 4 x float> %v2)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %fmax_nxv4f64 = call fast double @llvm.vector.reduce.fmax.nxv4f64(<vscale x 4 x double> %v3)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 ; TYPE_BASED_ONLY-LABEL: 'reductions'
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %add_nxv4i32 = call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> %v0)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %add_nxv4i64 = call i64 @llvm.vector.reduce.add.nxv4i64(<vscale x 4 x i64> %v1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %mul_nxv4i32 = call i32 @llvm.vector.reduce.mul.nxv4i32(<vscale x 4 x i32> %v0)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %mul_nxv4i64 = call i64 @llvm.vector.reduce.mul.nxv4i64(<vscale x 4 x i64> %v1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %and_nxv4i32 = call i32 @llvm.vector.reduce.and.nxv4i32(<vscale x 4 x i32> %v0)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %and_nxv4i64 = call i64 @llvm.vector.reduce.and.nxv4i64(<vscale x 4 x i64> %v1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %or_nxv4i32 = call i32 @llvm.vector.reduce.or.nxv4i32(<vscale x 4 x i32> %v0)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %or_nxv4i64 = call i64 @llvm.vector.reduce.or.nxv4i64(<vscale x 4 x i64> %v1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %xor_nxv4i32 = call i32 @llvm.vector.reduce.xor.nxv4i32(<vscale x 4 x i32> %v0)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %xor_nxv4i64 = call i64 @llvm.vector.reduce.xor.nxv4i64(<vscale x 4 x i64> %v1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %umin_nxv4i32 = call i32 @llvm.vector.reduce.umin.nxv4i32(<vscale x 4 x i32> %v0)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %umin_nxv4i64 = call i64 @llvm.vector.reduce.umin.nxv4i64(<vscale x 4 x i64> %v1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %umin_nxv4i64 = call i64 @llvm.vector.reduce.umin.nxv4i64(<vscale x 4 x i64> %v1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %smin_nxv4i32 = call i32 @llvm.vector.reduce.smin.nxv4i32(<vscale x 4 x i32> %v0)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %smin_nxv4i64 = call i64 @llvm.vector.reduce.smin.nxv4i64(<vscale x 4 x i64> %v1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %smin_nxv4i64 = call i64 @llvm.vector.reduce.smin.nxv4i64(<vscale x 4 x i64> %v1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %umax_nxv4i32 = call i32 @llvm.vector.reduce.umax.nxv4i32(<vscale x 4 x i32> %v0)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %umax_nxv4i64 = call i64 @llvm.vector.reduce.umax.nxv4i64(<vscale x 4 x i64> %v1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %umax_nxv4i64 = call i64 @llvm.vector.reduce.umax.nxv4i64(<vscale x 4 x i64> %v1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %smax_nxv4i32 = call i32 @llvm.vector.reduce.smax.nxv4i32(<vscale x 4 x i32> %v0)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %smax_nxv4i64 = call i64 @llvm.vector.reduce.smax.nxv4i64(<vscale x 4 x i64> %v1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %smax_nxv4i64 = call i64 @llvm.vector.reduce.smax.nxv4i64(<vscale x 4 x i64> %v1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %fadd_nxv4f32 = call fast float @llvm.vector.reduce.fadd.nxv4f32(float 0.000000e+00, <vscale x 4 x float> %v2)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %fadd_nxv4f64 = call fast double @llvm.vector.reduce.fadd.nxv4f64(double 0.000000e+00, <vscale x 4 x double> %v3)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %fmin_nxv4f32 = call fast float @llvm.vector.reduce.fmin.nxv4f32(<vscale x 4 x float> %v2)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %fmin_nxv4f64 = call fast double @llvm.vector.reduce.fmin.nxv4f64(<vscale x 4 x double> %v3)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %fmax_nxv4f32 = call fast float @llvm.vector.reduce.fmax.nxv4f32(<vscale x 4 x float> %v2)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %fmax_nxv4f64 = call fast double @llvm.vector.reduce.fmax.nxv4f64(<vscale x 4 x double> %v3)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;

llvm/test/Analysis/CostModel/AArch64/sve-min-max.ll

 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
 ; RUN: opt < %s -mtriple=aarch64-unknown-linux-gnu -mattr=+sve -passes="print<cost-model>" 2>&1 -disable-output -cost-kind=throughput | FileCheck %s
 
 target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
 
 define void @umin() {
 ; CHECK-LABEL: 'umin'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i8 = call <vscale x 4 x i8> @llvm.umin.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i8 = call <vscale x 8 x i8> @llvm.umin.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16i8 = call <vscale x 16 x i8> @llvm.umin.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32i8 = call <vscale x 32 x i8> @llvm.umin.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64i8 = call <vscale x 64 x i8> @llvm.umin.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i16 = call <vscale x 2 x i16> @llvm.umin.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i16 = call <vscale x 4 x i16> @llvm.umin.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i16 = call <vscale x 8 x i16> @llvm.umin.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16i16 = call <vscale x 16 x i16> @llvm.umin.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i32 = call <vscale x 2 x i32> @llvm.umin.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i32 = call <vscale x 4 x i32> @llvm.umin.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8i32 = call <vscale x 8 x i32> @llvm.umin.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i64 = call <vscale x 2 x i64> @llvm.umin.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4i64 = call <vscale x 4 x i64> @llvm.umin.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i8 = call <vscale x 4 x i8> @llvm.umin.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8i8 = call <vscale x 8 x i8> @llvm.umin.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16i8 = call <vscale x 16 x i8> @llvm.umin.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32i8 = call <vscale x 32 x i8> @llvm.umin.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64i8 = call <vscale x 64 x i8> @llvm.umin.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i16 = call <vscale x 2 x i16> @llvm.umin.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i16 = call <vscale x 4 x i16> @llvm.umin.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8i16 = call <vscale x 8 x i16> @llvm.umin.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16i16 = call <vscale x 16 x i16> @llvm.umin.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i32 = call <vscale x 2 x i32> @llvm.umin.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i32 = call <vscale x 4 x i32> @llvm.umin.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i32 = call <vscale x 8 x i32> @llvm.umin.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i64 = call <vscale x 2 x i64> @llvm.umin.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i64 = call <vscale x 4 x i64> @llvm.umin.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   %V4i8 = call <vscale x 4 x i8> @llvm.umin.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
   %V8i8 = call <vscale x 8 x i8> @llvm.umin.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
   %V16i8 = call <vscale x 16 x i8> @llvm.umin.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
   %V32i8 = call <vscale x 32 x i8> @llvm.umin.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
   %V64i8 = call <vscale x 64 x i8> @llvm.umin.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
   %V2i16 = call <vscale x 2 x i16> @llvm.umin.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
   %V4i16 = call <vscale x 4 x i16> @llvm.umin.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
   %V8i16 = call <vscale x 8 x i16> @llvm.umin.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
   %V16i16 = call <vscale x 16 x i16> @llvm.umin.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
   %V2i32 = call <vscale x 2 x i32> @llvm.umin.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
   %V4i32 = call <vscale x 4 x i32> @llvm.umin.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
   %V8i32 = call <vscale x 8 x i32> @llvm.umin.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
   %V2i64 = call <vscale x 2 x i64> @llvm.umin.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
   %V4i64 = call <vscale x 4 x i64> @llvm.umin.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
   ret void
 }
 
 define void @umax() {
 ; CHECK-LABEL: 'umax'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i8 = call <vscale x 4 x i8> @llvm.umax.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i8 = call <vscale x 8 x i8> @llvm.umax.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16i8 = call <vscale x 16 x i8> @llvm.umax.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32i8 = call <vscale x 32 x i8> @llvm.umax.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64i8 = call <vscale x 64 x i8> @llvm.umax.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i16 = call <vscale x 2 x i16> @llvm.umax.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i16 = call <vscale x 4 x i16> @llvm.umax.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i16 = call <vscale x 8 x i16> @llvm.umax.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16i16 = call <vscale x 16 x i16> @llvm.umax.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i32 = call <vscale x 2 x i32> @llvm.umax.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i32 = call <vscale x 4 x i32> @llvm.umax.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8i32 = call <vscale x 8 x i32> @llvm.umax.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i64 = call <vscale x 2 x i64> @llvm.umax.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4i64 = call <vscale x 4 x i64> @llvm.umax.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i8 = call <vscale x 4 x i8> @llvm.umax.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8i8 = call <vscale x 8 x i8> @llvm.umax.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16i8 = call <vscale x 16 x i8> @llvm.umax.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32i8 = call <vscale x 32 x i8> @llvm.umax.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64i8 = call <vscale x 64 x i8> @llvm.umax.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i16 = call <vscale x 2 x i16> @llvm.umax.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i16 = call <vscale x 4 x i16> @llvm.umax.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8i16 = call <vscale x 8 x i16> @llvm.umax.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16i16 = call <vscale x 16 x i16> @llvm.umax.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i32 = call <vscale x 2 x i32> @llvm.umax.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i32 = call <vscale x 4 x i32> @llvm.umax.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i32 = call <vscale x 8 x i32> @llvm.umax.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i64 = call <vscale x 2 x i64> @llvm.umax.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i64 = call <vscale x 4 x i64> @llvm.umax.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   %V4i8 = call <vscale x 4 x i8> @llvm.umax.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
   %V8i8 = call <vscale x 8 x i8> @llvm.umax.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
   %V16i8 = call <vscale x 16 x i8> @llvm.umax.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
   %V32i8 = call <vscale x 32 x i8> @llvm.umax.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
   %V64i8 = call <vscale x 64 x i8> @llvm.umax.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
   %V2i16 = call <vscale x 2 x i16> @llvm.umax.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
   %V4i16 = call <vscale x 4 x i16> @llvm.umax.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
   %V8i16 = call <vscale x 8 x i16> @llvm.umax.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
   %V16i16 = call <vscale x 16 x i16> @llvm.umax.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
   %V2i32 = call <vscale x 2 x i32> @llvm.umax.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
   %V4i32 = call <vscale x 4 x i32> @llvm.umax.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
   %V8i32 = call <vscale x 8 x i32> @llvm.umax.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
   %V2i64 = call <vscale x 2 x i64> @llvm.umax.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
   %V4i64 = call <vscale x 4 x i64> @llvm.umax.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
   ret void
 }
 
 define void @smin() {
 ; CHECK-LABEL: 'smin'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i8 = call <vscale x 4 x i8> @llvm.smin.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i8 = call <vscale x 8 x i8> @llvm.smin.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16i8 = call <vscale x 16 x i8> @llvm.smin.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32i8 = call <vscale x 32 x i8> @llvm.smin.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64i8 = call <vscale x 64 x i8> @llvm.smin.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i16 = call <vscale x 2 x i16> @llvm.smin.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i16 = call <vscale x 4 x i16> @llvm.smin.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i16 = call <vscale x 8 x i16> @llvm.smin.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16i16 = call <vscale x 16 x i16> @llvm.smin.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i32 = call <vscale x 2 x i32> @llvm.smin.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i32 = call <vscale x 4 x i32> @llvm.smin.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8i32 = call <vscale x 8 x i32> @llvm.smin.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i64 = call <vscale x 2 x i64> @llvm.smin.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4i64 = call <vscale x 4 x i64> @llvm.smin.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i8 = call <vscale x 4 x i8> @llvm.smin.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8i8 = call <vscale x 8 x i8> @llvm.smin.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16i8 = call <vscale x 16 x i8> @llvm.smin.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32i8 = call <vscale x 32 x i8> @llvm.smin.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64i8 = call <vscale x 64 x i8> @llvm.smin.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i16 = call <vscale x 2 x i16> @llvm.smin.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i16 = call <vscale x 4 x i16> @llvm.smin.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8i16 = call <vscale x 8 x i16> @llvm.smin.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16i16 = call <vscale x 16 x i16> @llvm.smin.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i32 = call <vscale x 2 x i32> @llvm.smin.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i32 = call <vscale x 4 x i32> @llvm.smin.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i32 = call <vscale x 8 x i32> @llvm.smin.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i64 = call <vscale x 2 x i64> @llvm.smin.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i64 = call <vscale x 4 x i64> @llvm.smin.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   %V4i8 = call <vscale x 4 x i8> @llvm.smin.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
   %V8i8 = call <vscale x 8 x i8> @llvm.smin.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
   %V16i8 = call <vscale x 16 x i8> @llvm.smin.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
   %V32i8 = call <vscale x 32 x i8> @llvm.smin.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
   %V64i8 = call <vscale x 64 x i8> @llvm.smin.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
   %V2i16 = call <vscale x 2 x i16> @llvm.smin.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
   %V4i16 = call <vscale x 4 x i16> @llvm.smin.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
   %V8i16 = call <vscale x 8 x i16> @llvm.smin.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
   %V16i16 = call <vscale x 16 x i16> @llvm.smin.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
   %V2i32 = call <vscale x 2 x i32> @llvm.smin.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
   %V4i32 = call <vscale x 4 x i32> @llvm.smin.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
   %V8i32 = call <vscale x 8 x i32> @llvm.smin.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
   %V2i64 = call <vscale x 2 x i64> @llvm.smin.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
   %V4i64 = call <vscale x 4 x i64> @llvm.smin.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
   ret void
 }
 
 define void @smax() {
 ; CHECK-LABEL: 'smax'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i8 = call <vscale x 4 x i8> @llvm.smax.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i8 = call <vscale x 8 x i8> @llvm.smax.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16i8 = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32i8 = call <vscale x 32 x i8> @llvm.smax.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64i8 = call <vscale x 64 x i8> @llvm.smax.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i16 = call <vscale x 2 x i16> @llvm.smax.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i16 = call <vscale x 4 x i16> @llvm.smax.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i16 = call <vscale x 8 x i16> @llvm.smax.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16i16 = call <vscale x 16 x i16> @llvm.smax.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i32 = call <vscale x 2 x i32> @llvm.smax.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i32 = call <vscale x 4 x i32> @llvm.smax.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8i32 = call <vscale x 8 x i32> @llvm.smax.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2i64 = call <vscale x 2 x i64> @llvm.smax.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4i64 = call <vscale x 4 x i64> @llvm.smax.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i8 = call <vscale x 4 x i8> @llvm.smax.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8i8 = call <vscale x 8 x i8> @llvm.smax.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16i8 = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32i8 = call <vscale x 32 x i8> @llvm.smax.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64i8 = call <vscale x 64 x i8> @llvm.smax.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i16 = call <vscale x 2 x i16> @llvm.smax.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i16 = call <vscale x 4 x i16> @llvm.smax.nxv4i16(<vscale x 4 x i16> undef, <vscale x 4 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8i16 = call <vscale x 8 x i16> @llvm.smax.nxv8i16(<vscale x 8 x i16> undef, <vscale x 8 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16i16 = call <vscale x 16 x i16> @llvm.smax.nxv16i16(<vscale x 16 x i16> undef, <vscale x 16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i32 = call <vscale x 2 x i32> @llvm.smax.nxv2i32(<vscale x 2 x i32> undef, <vscale x 2 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V4i32 = call <vscale x 4 x i32> @llvm.smax.nxv4i32(<vscale x 4 x i32> undef, <vscale x 4 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8i32 = call <vscale x 8 x i32> @llvm.smax.nxv8i32(<vscale x 8 x i32> undef, <vscale x 8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V2i64 = call <vscale x 2 x i64> @llvm.smax.nxv2i64(<vscale x 2 x i64> undef, <vscale x 2 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4i64 = call <vscale x 4 x i64> @llvm.smax.nxv4i64(<vscale x 4 x i64> undef, <vscale x 4 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   %V4i8 = call <vscale x 4 x i8> @llvm.smax.nxv4i8(<vscale x 4 x i8> undef, <vscale x 4 x i8> undef)
   %V8i8 = call <vscale x 8 x i8> @llvm.smax.nxv8i8(<vscale x 8 x i8> undef, <vscale x 8 x i8> undef)
   %V16i8 = call <vscale x 16 x i8> @llvm.smax.nxv16i8(<vscale x 16 x i8> undef, <vscale x 16 x i8> undef)
   %V32i8 = call <vscale x 32 x i8> @llvm.smax.nxv32i8(<vscale x 32 x i8> undef, <vscale x 32 x i8> undef)
   %V64i8 = call <vscale x 64 x i8> @llvm.smax.nxv64i8(<vscale x 64 x i8> undef, <vscale x 64 x i8> undef)
   %V2i16 = call <vscale x 2 x i16> @llvm.smax.nxv2i16(<vscale x 2 x i16> undef, <vscale x 2 x i16> undef)

 declare <vscale x 2 x half> @llvm.maximum.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>)
 declare <vscale x 4 x half> @llvm.maximum.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>)
 declare <vscale x 8 x half> @llvm.maximum.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>)
 declare <vscale x 16 x half> @llvm.maximum.nxv16f16(<vscale x 16 x half>, <vscale x 16 x half>)
 declare <vscale x 2 x float> @llvm.maximum.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>)
 declare <vscale x 4 x float> @llvm.maximum.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>)
 declare <vscale x 8 x float> @llvm.maximum.nxv8f32(<vscale x 8 x float>, <vscale x 8 x float>)
 declare <vscale x 2 x double> @llvm.maximum.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>)
 declare <vscale x 4 x double> @llvm.maximum.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>)