[ARM] MVE shuffle broadcast costs

[dmgreen]

Summary

A VDUP will perform a vector broadcast in a single instruction. Update the cost model for MVE accordingly.

Code originally by David Sherwood.

[SjoerdMeijer]

LGTM

llvm/trunk/lib/Target/ARM/ARMTargetTransformInfo.cpp

if (Kind == TTI::SK_Select) {
           {ISD::VECTOR_SHUFFLE, MVT::v16i8, 32}};
 
       std::pair<int, MVT> LT = TLI->getTypeLegalizationCost(DL, Tp);
       if (const auto *Entry = CostTableLookup(NEONSelShuffleTbl,
                                               ISD::VECTOR_SHUFFLE, LT.second))
         return LT.first * Entry->Cost;
     }
   }
+  if (ST->hasMVEIntegerOps()) {
+    if (Kind == TTI::SK_Broadcast) {
+      static const CostTblEntry MVEDupTbl[] = {
+          // VDUP handles these cases.
+          {ISD::VECTOR_SHUFFLE, MVT::v4i32, 1},
+          {ISD::VECTOR_SHUFFLE, MVT::v8i16, 1},
+          {ISD::VECTOR_SHUFFLE, MVT::v16i8, 1},
+          {ISD::VECTOR_SHUFFLE, MVT::v4f32, 1},
+          {ISD::VECTOR_SHUFFLE, MVT::v8f16, 1}};
+
+      std::pair<int, MVT> LT = TLI->getTypeLegalizationCost(DL, Tp);
+
+      if (const auto *Entry = CostTableLookup(MVEDupTbl, ISD::VECTOR_SHUFFLE,
+                                              LT.second))
+        return LT.first * Entry->Cost;
+    }
+  }
   return BaseT::getShuffleCost(Kind, Tp, Index, SubTp);
 }
 
 int ARMTTIImpl::getArithmeticInstrCost(
     unsigned Opcode, Type *Ty, TTI::OperandValueKind Op1Info,
     TTI::OperandValueKind Op2Info, TTI::OperandValueProperties Opd1PropInfo,
     TTI::OperandValueProperties Opd2PropInfo,
     ArrayRef<const Value *> Args) {

llvm/trunk/test/Analysis/CostModel/ARM/shuffle.ll

 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
 ; RUN: opt < %s  -cost-model -analyze -mtriple=thumbv8.1-m.main-none-eabi -mattr=+mve.fp | FileCheck %s --check-prefix=CHECK-MVE
 ; RUN: opt < %s  -cost-model -analyze -mtriple=thumbv7-apple-ios6.0.0 -mcpu=swift | FileCheck %s --check-prefix=CHECK-NEON
 
 define void @broadcast() {
 ; CHECK-MVE-LABEL: 'broadcast'
 ; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v7 = shufflevector <2 x i8> undef, <2 x i8> undef, <2 x i32> zeroinitializer
-; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %v8 = shufflevector <4 x i8> undef, <4 x i8> undef, <4 x i32> zeroinitializer
-; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v9 = shufflevector <8 x i8> undef, <8 x i8> undef, <8 x i32> zeroinitializer
-; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %v10 = shufflevector <16 x i8> undef, <16 x i8> undef, <16 x i32> zeroinitializer
+; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8 = shufflevector <4 x i8> undef, <4 x i8> undef, <4 x i32> zeroinitializer
+; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v9 = shufflevector <8 x i8> undef, <8 x i8> undef, <8 x i32> zeroinitializer
+; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v10 = shufflevector <16 x i8> undef, <16 x i8> undef, <16 x i32> zeroinitializer
 ; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v11 = shufflevector <2 x i16> undef, <2 x i16> undef, <2 x i32> zeroinitializer
-; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %v12 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> zeroinitializer
-; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %v13 = shufflevector <8 x i16> undef, <8 x i16> undef, <8 x i32> zeroinitializer
+; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v12 = shufflevector <4 x i16> undef, <4 x i16> undef, <4 x i32> zeroinitializer
+; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v13 = shufflevector <8 x i16> undef, <8 x i16> undef, <8 x i32> zeroinitializer
 ; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v14 = shufflevector <2 x i32> undef, <2 x i32> undef, <2 x i32> zeroinitializer
-; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %v15 = shufflevector <4 x i32> undef, <4 x i32> undef, <4 x i32> zeroinitializer
-; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v16 = shufflevector <2 x float> undef, <2 x float> undef, <2 x i32> zeroinitializer
-; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %v17 = shufflevector <4 x float> undef, <4 x float> undef, <4 x i32> zeroinitializer
+; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v15 = shufflevector <4 x i32> undef, <4 x i32> undef, <4 x i32> zeroinitializer
+; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v16 = shufflevector <2 x float> undef, <2 x float> undef, <2 x i32> zeroinitializer
+; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v17 = shufflevector <4 x float> undef, <4 x float> undef, <4 x i32> zeroinitializer
 ; CHECK-MVE-NEXT:  Cost Model: Unknown cost for instruction: %v18 = shufflevector <8 x half> undef, <8 x half> undef, <4 x i32> zeroinitializer
 ; CHECK-MVE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 ; CHECK-NEON-LABEL: 'broadcast'
 ; CHECK-NEON-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v7 = shufflevector <2 x i8> undef, <2 x i8> undef, <2 x i32> zeroinitializer
 ; CHECK-NEON-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8 = shufflevector <4 x i8> undef, <4 x i8> undef, <4 x i32> zeroinitializer
 ; CHECK-NEON-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v9 = shufflevector <8 x i8> undef, <8 x i8> undef, <8 x i32> zeroinitializer
 ; CHECK-NEON-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v10 = shufflevector <16 x i8> undef, <16 x i8> undef, <16 x i32> zeroinitializer