From 84f24c2daffc40fc10b4ea2ae69016ebdabfc0ed Mon Sep 17 00:00:00 2001
From: Shih-Po Hung <shihpo.hung@sifive.com>
Date: Tue, 2 Apr 2024 09:26:27 +0800
Subject: [PATCH] [RISCV][TTI] Scale the cost of intrinsic umin/umax/smin/smax
 with LMUL (#87245)

Use the return type to measure the LMUL size for throughput/latency cost
---
 .../Target/RISCV/RISCVTargetTransformInfo.cpp |  22 +++-
 .../Analysis/CostModel/RISCV/int-min-max.ll   | 120 +++++++++---------
 2 files changed, 80 insertions(+), 62 deletions(-)

diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index efcaa65605e0..ed4b0ca8c941 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -810,9 +810,27 @@ RISCVTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
   case Intrinsic::smin:
   case Intrinsic::smax: {
     auto LT = getTypeLegalizationCost(RetTy);
-    if ((ST->hasVInstructions() && LT.second.isVector()) ||
-        (LT.second.isScalarInteger() && ST->hasStdExtZbb()))
+    if (LT.second.isScalarInteger() && ST->hasStdExtZbb())
       return LT.first;
+
+    if (ST->hasVInstructions() && LT.second.isVector()) {
+      unsigned Op;
+      switch (ICA.getID()) {
+      case Intrinsic::umin:
+        Op = RISCV::VMINU_VV;
+        break;
+      case Intrinsic::umax:
+        Op = RISCV::VMAXU_VV;
+        break;
+      case Intrinsic::smin:
+        Op = RISCV::VMIN_VV;
+        break;
+      case Intrinsic::smax:
+        Op = RISCV::VMAX_VV;
+        break;
+      }
+      return LT.first * getRISCVInstructionCost(Op, LT.second, CostKind);
+    }
     break;
   }
   case Intrinsic::sadd_sat:
diff --git a/llvm/test/Analysis/CostModel/RISCV/int-min-max.ll b/llvm/test/Analysis/CostModel/RISCV/int-min-max.ll
index ec669c986c15..79cf1c84ed49 100644
--- a/llvm/test/Analysis/CostModel/RISCV/int-min-max.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/int-min-max.ll
@@ -12,36 +12,36 @@ define void @smax() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x i8> @llvm.smax.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = 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: %9 = 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: %10 = 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: %10 = 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: %11 = call i16 @llvm.smax.i16(i16 undef, i16 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i16> @llvm.smax.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i16> @llvm.smax.v4i16(<4 x i16> undef, <4 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i16> @llvm.smax.v8i16(<8 x i16> undef, <8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i16> @llvm.smax.v16i16(<16 x i16> undef, <16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %15 = call <16 x i16> @llvm.smax.v16i16(<16 x i16> undef, <16 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i16> @llvm.smax.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %17 = 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: %18 = 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: %19 = 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 1 for instruction: %20 = 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: %19 = 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: %20 = 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: %21 = call i32 @llvm.smax.i32(i32 undef, i32 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %22 = call <2 x i32> @llvm.smax.v2i32(<2 x i32> undef, <2 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %23 = call <4 x i32> @llvm.smax.v4i32(<4 x i32> undef, <4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %24 = call <8 x i32> @llvm.smax.v8i32(<8 x i32> undef, <8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %25 = call <16 x i32> @llvm.smax.v16i32(<16 x i32> undef, <16 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %24 = call <8 x i32> @llvm.smax.v8i32(<8 x i32> undef, <8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %25 = call <16 x i32> @llvm.smax.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %26 = call <vscale x 1 x i32> @llvm.smax.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %27 = 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: %28 = 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 1 for instruction: %29 = 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: %30 = call <vscale x 16 x i32> @llvm.smax.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %28 = 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: %29 = 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 8 for instruction: %30 = call <vscale x 16 x i32> @llvm.smax.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %31 = call i64 @llvm.smax.i64(i64 undef, i64 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %32 = call <2 x i64> @llvm.smax.v2i64(<2 x i64> undef, <2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %33 = call <4 x i64> @llvm.smax.v4i64(<4 x i64> undef, <4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %34 = call <8 x i64> @llvm.smax.v8i64(<8 x i64> undef, <8 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %35 = call <16 x i64> @llvm.smax.v16i64(<16 x i64> undef, <16 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %33 = call <4 x i64> @llvm.smax.v4i64(<4 x i64> undef, <4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %34 = call <8 x i64> @llvm.smax.v8i64(<8 x i64> undef, <8 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %35 = call <16 x i64> @llvm.smax.v16i64(<16 x i64> undef, <16 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %36 = call <vscale x 1 x i64> @llvm.smax.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %37 = 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 1 for instruction: %38 = 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: %39 = call <vscale x 8 x i64> @llvm.smax.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %37 = 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: %38 = 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 8 for instruction: %39 = call <vscale x 8 x i64> @llvm.smax.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   call i8 @llvm.smax.i8(i8 undef, i8 undef)
@@ -97,36 +97,36 @@ define void @smin() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x i8> @llvm.smin.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = 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: %9 = 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: %10 = 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: %10 = 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: %11 = call i16 @llvm.smin.i16(i16 undef, i16 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i16> @llvm.smin.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i16> @llvm.smin.v4i16(<4 x i16> undef, <4 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i16> @llvm.smin.v8i16(<8 x i16> undef, <8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i16> @llvm.smin.v16i16(<16 x i16> undef, <16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %15 = call <16 x i16> @llvm.smin.v16i16(<16 x i16> undef, <16 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i16> @llvm.smin.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %17 = 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: %18 = 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: %19 = 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 1 for instruction: %20 = 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: %19 = 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: %20 = 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: %21 = call i32 @llvm.smin.i32(i32 undef, i32 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %22 = call <2 x i32> @llvm.smin.v2i32(<2 x i32> undef, <2 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %23 = call <4 x i32> @llvm.smin.v4i32(<4 x i32> undef, <4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %24 = call <8 x i32> @llvm.smin.v8i32(<8 x i32> undef, <8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %25 = call <16 x i32> @llvm.smin.v16i32(<16 x i32> undef, <16 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %24 = call <8 x i32> @llvm.smin.v8i32(<8 x i32> undef, <8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %25 = call <16 x i32> @llvm.smin.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %26 = call <vscale x 1 x i32> @llvm.smin.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %27 = 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: %28 = 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 1 for instruction: %29 = 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: %30 = call <vscale x 16 x i32> @llvm.smin.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %28 = 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: %29 = 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 8 for instruction: %30 = call <vscale x 16 x i32> @llvm.smin.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %31 = call i64 @llvm.smin.i64(i64 undef, i64 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %32 = call <2 x i64> @llvm.smin.v2i64(<2 x i64> undef, <2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %33 = call <4 x i64> @llvm.smin.v4i64(<4 x i64> undef, <4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %34 = call <8 x i64> @llvm.smin.v8i64(<8 x i64> undef, <8 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %35 = call <16 x i64> @llvm.smin.v16i64(<16 x i64> undef, <16 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %33 = call <4 x i64> @llvm.smin.v4i64(<4 x i64> undef, <4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %34 = call <8 x i64> @llvm.smin.v8i64(<8 x i64> undef, <8 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %35 = call <16 x i64> @llvm.smin.v16i64(<16 x i64> undef, <16 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %36 = call <vscale x 1 x i64> @llvm.smin.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %37 = 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 1 for instruction: %38 = 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: %39 = call <vscale x 8 x i64> @llvm.smin.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %37 = 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: %38 = 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 8 for instruction: %39 = call <vscale x 8 x i64> @llvm.smin.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   call i8 @llvm.smin.i8(i8 undef, i8 undef)
@@ -182,36 +182,36 @@ define void @umax() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x i8> @llvm.umax.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = 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: %9 = 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: %10 = 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: %10 = 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: %11 = call i16 @llvm.umax.i16(i16 undef, i16 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i16> @llvm.umax.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i16> @llvm.umax.v4i16(<4 x i16> undef, <4 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i16> @llvm.umax.v8i16(<8 x i16> undef, <8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i16> @llvm.umax.v16i16(<16 x i16> undef, <16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %15 = call <16 x i16> @llvm.umax.v16i16(<16 x i16> undef, <16 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i16> @llvm.umax.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %17 = 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: %18 = 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: %19 = 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 1 for instruction: %20 = 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: %19 = 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: %20 = 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: %21 = call i32 @llvm.umax.i32(i32 undef, i32 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %22 = call <2 x i32> @llvm.umax.v2i32(<2 x i32> undef, <2 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %23 = call <4 x i32> @llvm.umax.v4i32(<4 x i32> undef, <4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %24 = call <8 x i32> @llvm.umax.v8i32(<8 x i32> undef, <8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %25 = call <16 x i32> @llvm.umax.v16i32(<16 x i32> undef, <16 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %24 = call <8 x i32> @llvm.umax.v8i32(<8 x i32> undef, <8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %25 = call <16 x i32> @llvm.umax.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %26 = call <vscale x 1 x i32> @llvm.umax.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %27 = 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: %28 = 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 1 for instruction: %29 = 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: %30 = call <vscale x 16 x i32> @llvm.umax.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %28 = 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: %29 = 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 8 for instruction: %30 = call <vscale x 16 x i32> @llvm.umax.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %31 = call i64 @llvm.umax.i64(i64 undef, i64 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %32 = call <2 x i64> @llvm.umax.v2i64(<2 x i64> undef, <2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %33 = call <4 x i64> @llvm.umax.v4i64(<4 x i64> undef, <4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %34 = call <8 x i64> @llvm.umax.v8i64(<8 x i64> undef, <8 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %35 = call <16 x i64> @llvm.umax.v16i64(<16 x i64> undef, <16 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %33 = call <4 x i64> @llvm.umax.v4i64(<4 x i64> undef, <4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %34 = call <8 x i64> @llvm.umax.v8i64(<8 x i64> undef, <8 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %35 = call <16 x i64> @llvm.umax.v16i64(<16 x i64> undef, <16 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %36 = call <vscale x 1 x i64> @llvm.umax.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %37 = 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 1 for instruction: %38 = 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: %39 = call <vscale x 8 x i64> @llvm.umax.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %37 = 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: %38 = 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 8 for instruction: %39 = call <vscale x 8 x i64> @llvm.umax.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   call i8 @llvm.umax.i8(i8 undef, i8 undef)
@@ -267,36 +267,36 @@ define void @umin() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x i8> @llvm.umin.nxv2i8(<vscale x 2 x i8> undef, <vscale x 2 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = 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: %9 = 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: %10 = 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: %10 = 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: %11 = call i16 @llvm.umin.i16(i16 undef, i16 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i16> @llvm.umin.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i16> @llvm.umin.v4i16(<4 x i16> undef, <4 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i16> @llvm.umin.v8i16(<8 x i16> undef, <8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i16> @llvm.umin.v16i16(<16 x i16> undef, <16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %15 = call <16 x i16> @llvm.umin.v16i16(<16 x i16> undef, <16 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i16> @llvm.umin.nxv1i16(<vscale x 1 x i16> undef, <vscale x 1 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %17 = 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: %18 = 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: %19 = 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 1 for instruction: %20 = 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: %19 = 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: %20 = 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: %21 = call i32 @llvm.umin.i32(i32 undef, i32 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %22 = call <2 x i32> @llvm.umin.v2i32(<2 x i32> undef, <2 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %23 = call <4 x i32> @llvm.umin.v4i32(<4 x i32> undef, <4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %24 = call <8 x i32> @llvm.umin.v8i32(<8 x i32> undef, <8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %25 = call <16 x i32> @llvm.umin.v16i32(<16 x i32> undef, <16 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %24 = call <8 x i32> @llvm.umin.v8i32(<8 x i32> undef, <8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %25 = call <16 x i32> @llvm.umin.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %26 = call <vscale x 1 x i32> @llvm.umin.nxv1i32(<vscale x 1 x i32> undef, <vscale x 1 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %27 = 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: %28 = 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 1 for instruction: %29 = 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: %30 = call <vscale x 16 x i32> @llvm.umin.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %28 = 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: %29 = 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 8 for instruction: %30 = call <vscale x 16 x i32> @llvm.umin.nxv16i32(<vscale x 16 x i32> undef, <vscale x 16 x i32> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %31 = call i64 @llvm.umin.i64(i64 undef, i64 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %32 = call <2 x i64> @llvm.umin.v2i64(<2 x i64> undef, <2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %33 = call <4 x i64> @llvm.umin.v4i64(<4 x i64> undef, <4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %34 = call <8 x i64> @llvm.umin.v8i64(<8 x i64> undef, <8 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %35 = call <16 x i64> @llvm.umin.v16i64(<16 x i64> undef, <16 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %33 = call <4 x i64> @llvm.umin.v4i64(<4 x i64> undef, <4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %34 = call <8 x i64> @llvm.umin.v8i64(<8 x i64> undef, <8 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %35 = call <16 x i64> @llvm.umin.v16i64(<16 x i64> undef, <16 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %36 = call <vscale x 1 x i64> @llvm.umin.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %37 = 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 1 for instruction: %38 = 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: %39 = call <vscale x 8 x i64> @llvm.umin.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %37 = 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: %38 = 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 8 for instruction: %39 = call <vscale x 8 x i64> @llvm.umin.nxv8i64(<vscale x 8 x i64> undef, <vscale x 8 x i64> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   call i8 @llvm.umin.i8(i8 undef, i8 undef)