[llvm-branch-commits] [llvm] TTI: Check legalization cost of abs nodes (PR #100523)

Matt Arsenault via llvm-branch-commits llvm-branch-commits at lists.llvm.org
Thu Aug 8 12:56:58 PDT 2024


https://github.com/arsenm updated https://github.com/llvm/llvm-project/pull/100523

>From 9c21872a5ab0b5e91b90701326a672f4b73b463d Mon Sep 17 00:00:00 2001
From: Matt Arsenault <Matthew.Arsenault at amd.com>
Date: Thu, 25 Jul 2024 10:38:11 +0400
Subject: [PATCH] TTI: Check legalization cost of abs nodes

Also adjust the AMDGPU cost.
---
 llvm/include/llvm/CodeGen/BasicTTIImpl.h      |  32 +-
 .../AMDGPU/AMDGPUTargetTransformInfo.cpp      |   9 +-
 llvm/test/Analysis/CostModel/AMDGPU/abs.ll    | 368 +++++++++---------
 .../Analysis/CostModel/AMDGPU/arith-ssat.ll   |  32 +-
 .../Analysis/CostModel/AMDGPU/arith-usat.ll   |  32 +-
 5 files changed, 242 insertions(+), 231 deletions(-)

diff --git a/llvm/include/llvm/CodeGen/BasicTTIImpl.h b/llvm/include/llvm/CodeGen/BasicTTIImpl.h
index 8a14c8a37577ad..c2bc1353ee8838 100644
--- a/llvm/include/llvm/CodeGen/BasicTTIImpl.h
+++ b/llvm/include/llvm/CodeGen/BasicTTIImpl.h
@@ -2120,20 +2120,9 @@ class BasicTTIImplBase : public TargetTransformInfoImplCRTPBase<T> {
     case Intrinsic::vector_reduce_fminimum:
       return thisT()->getMinMaxReductionCost(getMinMaxReductionIntrinsicOp(IID),
                                              VecOpTy, ICA.getFlags(), CostKind);
-    case Intrinsic::abs: {
-      // abs(X) = select(icmp(X,0),X,sub(0,X))
-      Type *CondTy = RetTy->getWithNewBitWidth(1);
-      CmpInst::Predicate Pred = CmpInst::ICMP_SGT;
-      InstructionCost Cost = 0;
-      Cost += thisT()->getCmpSelInstrCost(BinaryOperator::ICmp, RetTy, CondTy,
-                                          Pred, CostKind);
-      Cost += thisT()->getCmpSelInstrCost(BinaryOperator::Select, RetTy, CondTy,
-                                          Pred, CostKind);
-      // TODO: Should we add an OperandValueProperties::OP_Zero property?
-      Cost += thisT()->getArithmeticInstrCost(
-         BinaryOperator::Sub, RetTy, CostKind, {TTI::OK_UniformConstantValue, TTI::OP_None});
-      return Cost;
-    }
+    case Intrinsic::abs:
+      ISD = ISD::ABS;
+      break;
     case Intrinsic::smax:
       ISD = ISD::SMAX;
       break;
@@ -2402,6 +2391,21 @@ class BasicTTIImplBase : public TargetTransformInfoImplCRTPBase<T> {
       Cost += thisT()->getArithmeticInstrCost(Instruction::Or, RetTy, CostKind);
       return Cost;
     }
+    case Intrinsic::abs: {
+      // abs(X) = select(icmp(X,0),X,sub(0,X))
+      Type *CondTy = RetTy->getWithNewBitWidth(1);
+      CmpInst::Predicate Pred = CmpInst::ICMP_SGT;
+      InstructionCost Cost = 0;
+      Cost += thisT()->getCmpSelInstrCost(BinaryOperator::ICmp, RetTy, CondTy,
+                                          Pred, CostKind);
+      Cost += thisT()->getCmpSelInstrCost(BinaryOperator::Select, RetTy, CondTy,
+                                          Pred, CostKind);
+      // TODO: Should we add an OperandValueProperties::OP_Zero property?
+      Cost += thisT()->getArithmeticInstrCost(
+          BinaryOperator::Sub, RetTy, CostKind,
+          {TTI::OK_UniformConstantValue, TTI::OP_None});
+      return Cost;
+    }
     case Intrinsic::fptosi_sat:
     case Intrinsic::fptoui_sat: {
       if (Tys.empty())
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
index 8d4ff64ac5adcf..c6aadf0b503012 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
@@ -696,6 +696,7 @@ static bool intrinsicHasPackedVectorBenefit(Intrinsic::ID ID) {
   case Intrinsic::usub_sat:
   case Intrinsic::sadd_sat:
   case Intrinsic::ssub_sat:
+  case Intrinsic::abs:
     return true;
   default:
     return false;
@@ -724,7 +725,7 @@ GCNTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
   if (SLT == MVT::f64)
     return LT.first * NElts * get64BitInstrCost(CostKind);
 
-  if ((ST->has16BitInsts() && SLT == MVT::f16) ||
+  if ((ST->has16BitInsts() && (SLT == MVT::f16 || SLT == MVT::i16)) ||
       (ST->hasPackedFP32Ops() && SLT == MVT::f32))
     NElts = (NElts + 1) / 2;
 
@@ -752,11 +753,17 @@ GCNTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
   case Intrinsic::usub_sat:
   case Intrinsic::sadd_sat:
   case Intrinsic::ssub_sat: {
+    // TODO: Full rate for i32/i16
     static const auto ValidSatTys = {MVT::v2i16, MVT::v4i16};
     if (any_of(ValidSatTys, [&LT](MVT M) { return M == LT.second; }))
       NElts = 1;
     break;
   }
+  case Intrinsic::abs:
+    // Expansion takes 2 instructions for VALU
+    if (SLT == MVT::i16 || SLT == MVT::i32)
+      InstRate = 2 * getFullRateInstrCost();
+    break;
   default:
     break;
   }
diff --git a/llvm/test/Analysis/CostModel/AMDGPU/abs.ll b/llvm/test/Analysis/CostModel/AMDGPU/abs.ll
index f65615b07abc0f..b86e99558377bb 100644
--- a/llvm/test/Analysis/CostModel/AMDGPU/abs.ll
+++ b/llvm/test/Analysis/CostModel/AMDGPU/abs.ll
@@ -14,116 +14,116 @@ define void @abs_nonpoison() {
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 false)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 false)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 58 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 114 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 174 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 160 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 198 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 false)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 322 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 false)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 false)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: ret void
 ;
 ; SLOW-LABEL: 'abs_nonpoison'
 ; SLOW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 false)
 ; SLOW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 false)
 ; SLOW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 58 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 162 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 160 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 198 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 false)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 322 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 false)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 false)
 ; SLOW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: ret void
 ;
 ; FAST-SIZE-LABEL: 'abs_nonpoison'
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 false)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 false)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 false)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 false)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 false)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 false)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
 ; SLOW-SIZE-LABEL: 'abs_nonpoison'
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 false)
 ; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 false)
 ; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 false)
 ; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 false)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 false)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 false)
 ; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
   %I64 = call i64 @llvm.abs.i64(i64 undef, i1 false)
@@ -164,116 +164,116 @@ define void @abs_poison() {
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 true)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 true)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 58 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 114 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 174 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 160 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 198 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 true)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 322 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 true)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 true)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: ret void
 ;
 ; SLOW-LABEL: 'abs_poison'
 ; SLOW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 true)
 ; SLOW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 true)
 ; SLOW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 58 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 45 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 102 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 162 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 160 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 198 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 true)
-; SLOW-NEXT:  Cost Model: Found an estimated cost of 322 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 true)
+; SLOW-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 true)
 ; SLOW-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: ret void
 ;
 ; FAST-SIZE-LABEL: 'abs_poison'
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 true)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 true)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 true)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 true)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 true)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 true)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
 ; SLOW-SIZE-LABEL: 'abs_poison'
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I64 = call i64 @llvm.abs.i64(i64 undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.abs.v2i64(<2 x i64> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.abs.v4i64(<4 x i64> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V5I64 = call <5 x i64> @llvm.abs.v5i64(<5 x i64> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V8I64 = call <8 x i64> @llvm.abs.v8i64(<8 x i64> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call i32 @llvm.abs.i32(i32 undef, i1 true)
 ; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I32 = call <2 x i32> @llvm.abs.v2i32(<2 x i32> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 50 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.abs.v4i32(<4 x i32> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.abs.v8i32(<8 x i32> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 54 for instruction: %V9I32 = call <9 x i32> @llvm.abs.v9i32(<9 x i32> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I32 = call <16 x i32> @llvm.abs.v16i32(<16 x i32> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.abs.i16(i16 undef, i1 true)
 ; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I16 = call <2 x i16> @llvm.abs.v2i16(<2 x i16> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I16 = call <4 x i16> @llvm.abs.v4i16(<4 x i16> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.abs.v8i16(<8 x i16> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.abs.v16i16(<16 x i16> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V17I16 = call <17 x i16> @llvm.abs.v17i16(<17 x i16> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V32I16 = call <32 x i16> @llvm.abs.v32i16(<32 x i16> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.abs.i8(i8 undef, i1 true)
 ; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.abs.v2i8(<2 x i8> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 34 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 true)
-; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.abs.v4i8(<4 x i8> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I8 = call <8 x i8> @llvm.abs.v8i8(<8 x i8> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.abs.v16i8(<16 x i8> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.abs.v32i8(<32 x i8> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V33I8 = call <33 x i8> @llvm.abs.v33i8(<33 x i8> undef, i1 true)
+; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 132 for instruction: %V64I8 = call <64 x i8> @llvm.abs.v64i8(<64 x i8> undef, i1 true)
 ; SLOW-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
   %I64 = call i64 @llvm.abs.i64(i64 undef, i1 true)
diff --git a/llvm/test/Analysis/CostModel/AMDGPU/arith-ssat.ll b/llvm/test/Analysis/CostModel/AMDGPU/arith-ssat.ll
index 564bc4912af7d3..b3175067ac5263 100644
--- a/llvm/test/Analysis/CostModel/AMDGPU/arith-ssat.ll
+++ b/llvm/test/Analysis/CostModel/AMDGPU/arith-ssat.ll
@@ -55,10 +55,10 @@ define i32 @add(i32 %arg) {
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I16 = call <2 x i16> @llvm.sadd.sat.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I16 = call <4 x i16> @llvm.sadd.sat.v4i16(<4 x i16> undef, <4 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V17I16 = call <17 x i16> @llvm.sadd.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V17I16 = call <17 x i16> @llvm.sadd.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2I8 = call <2 x i8> @llvm.sadd.sat.v2i8(<2 x i8> undef, <2 x i8> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I8 = call <4 x i8> @llvm.sadd.sat.v4i8(<4 x i8> undef, <4 x i8> undef)
@@ -113,10 +113,10 @@ define i32 @add(i32 %arg) {
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = call <2 x i16> @llvm.sadd.sat.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I16 = call <4 x i16> @llvm.sadd.sat.v4i16(<4 x i16> undef, <4 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V17I16 = call <17 x i16> @llvm.sadd.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V17I16 = call <17 x i16> @llvm.sadd.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.sadd.sat.v2i8(<2 x i8> undef, <2 x i8> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.sadd.sat.v4i8(<4 x i8> undef, <4 x i8> undef)
@@ -235,10 +235,10 @@ define i32 @sub(i32 %arg) {
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I16 = call <2 x i16> @llvm.ssub.sat.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I16 = call <4 x i16> @llvm.ssub.sat.v4i16(<4 x i16> undef, <4 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V17I16 = call <17 x i16> @llvm.ssub.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V17I16 = call <17 x i16> @llvm.ssub.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2I8 = call <2 x i8> @llvm.ssub.sat.v2i8(<2 x i8> undef, <2 x i8> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I8 = call <4 x i8> @llvm.ssub.sat.v4i8(<4 x i8> undef, <4 x i8> undef)
@@ -293,10 +293,10 @@ define i32 @sub(i32 %arg) {
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = call <2 x i16> @llvm.ssub.sat.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I16 = call <4 x i16> @llvm.ssub.sat.v4i16(<4 x i16> undef, <4 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V17I16 = call <17 x i16> @llvm.ssub.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V17I16 = call <17 x i16> @llvm.ssub.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.ssub.sat.v2i8(<2 x i8> undef, <2 x i8> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.ssub.sat.v4i8(<4 x i8> undef, <4 x i8> undef)
diff --git a/llvm/test/Analysis/CostModel/AMDGPU/arith-usat.ll b/llvm/test/Analysis/CostModel/AMDGPU/arith-usat.ll
index d6481caef916df..eeab3cfa9a2d6b 100644
--- a/llvm/test/Analysis/CostModel/AMDGPU/arith-usat.ll
+++ b/llvm/test/Analysis/CostModel/AMDGPU/arith-usat.ll
@@ -55,10 +55,10 @@ define i32 @add(i32 %arg) {
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I16 = call <2 x i16> @llvm.uadd.sat.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I16 = call <4 x i16> @llvm.uadd.sat.v4i16(<4 x i16> undef, <4 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V17I16 = call <17 x i16> @llvm.uadd.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V17I16 = call <17 x i16> @llvm.uadd.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2I8 = call <2 x i8> @llvm.uadd.sat.v2i8(<2 x i8> undef, <2 x i8> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I8 = call <4 x i8> @llvm.uadd.sat.v4i8(<4 x i8> undef, <4 x i8> undef)
@@ -113,10 +113,10 @@ define i32 @add(i32 %arg) {
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = call <2 x i16> @llvm.uadd.sat.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I16 = call <4 x i16> @llvm.uadd.sat.v4i16(<4 x i16> undef, <4 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V17I16 = call <17 x i16> @llvm.uadd.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V17I16 = call <17 x i16> @llvm.uadd.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.uadd.sat.v2i8(<2 x i8> undef, <2 x i8> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.uadd.sat.v4i8(<4 x i8> undef, <4 x i8> undef)
@@ -235,10 +235,10 @@ define i32 @sub(i32 %arg) {
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I16 = call <2 x i16> @llvm.usub.sat.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I16 = call <4 x i16> @llvm.usub.sat.v4i16(<4 x i16> undef, <4 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V17I16 = call <17 x i16> @llvm.usub.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
-; FAST-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V17I16 = call <17 x i16> @llvm.usub.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
+; FAST-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V2I8 = call <2 x i8> @llvm.usub.sat.v2i8(<2 x i8> undef, <2 x i8> undef)
 ; FAST-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I8 = call <4 x i8> @llvm.usub.sat.v4i8(<4 x i8> undef, <4 x i8> undef)
@@ -293,10 +293,10 @@ define i32 @sub(i32 %arg) {
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = call <2 x i16> @llvm.usub.sat.v2i16(<2 x i16> undef, <2 x i16> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I16 = call <4 x i16> @llvm.usub.sat.v4i16(<4 x i16> undef, <4 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V17I16 = call <17 x i16> @llvm.usub.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
-; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V17I16 = call <17 x i16> @llvm.usub.sat.v17i16(<17 x i16> undef, <17 x i16> undef)
+; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I8 = call <2 x i8> @llvm.usub.sat.v2i8(<2 x i8> undef, <2 x i8> undef)
 ; FAST-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I8 = call <4 x i8> @llvm.usub.sat.v4i8(<4 x i8> undef, <4 x i8> undef)



More information about the llvm-branch-commits mailing list