llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp

//===-- AMDGPUTargetTransformInfo.cpp - AMDGPU specific TTI pass ---------===//
//
//                     The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// \file
// This file implements a TargetTransformInfo analysis pass specific to the
// AMDGPU target machine. It uses the target's detailed information to provide
// more precise answers to certain TTI queries, while letting the target
// independent and default TTI implementations handle the rest.
//
//===----------------------------------------------------------------------===//

#include "AMDGPUTargetTransformInfo.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/TargetTransformInfo.h"
#include "llvm/Analysis/ValueTracking.h"
#include "llvm/CodeGen/BasicTTIImpl.h"
#include "llvm/IR/Module.h"
#include "llvm/IR/Intrinsics.h"
#include "llvm/Support/Debug.h"
#include "llvm/Target/CostTable.h"
#include "llvm/Target/TargetLowering.h"
using namespace llvm;

#define DEBUG_TYPE "AMDGPUtti"

static cl::opt<unsigned> UnrollThresholdPrivate(
  "amdgpu-unroll-threshold-private",
  cl::desc("Unroll threshold for AMDGPU if private memory used in a loop"),
  cl::init(2500), cl::Hidden);

static cl::opt<unsigned> UnrollThresholdLocal(
  "amdgpu-unroll-threshold-local",
  cl::desc("Unroll threshold for AMDGPU if local memory used in a loop"),
  cl::init(1000), cl::Hidden);

static cl::opt<unsigned> UnrollThresholdIf(
  "amdgpu-unroll-threshold-if",
  cl::desc("Unroll threshold increment for AMDGPU for each if statement inside loop"),
  cl::init(150), cl::Hidden);

static bool dependsOnLocalPhi(const Loop *L, const Value *Cond,
                              unsigned Depth = 0) {
  const Instruction *I = dyn_cast<Instruction>(Cond);
  if (!I)
    return false;

  for (const Value *V : I->operand_values()) {
    if (!L->contains(I))
      continue;
    if (const PHINode *PHI = dyn_cast<PHINode>(V)) {
      if (none_of(L->getSubLoops(), [PHI](const Loop* SubLoop) {
                  return SubLoop->contains(PHI); }))
        return true;
    } else if (Depth < 10 && dependsOnLocalPhi(L, V, Depth+1))
      return true;
  }
  return false;
}

void AMDGPUTTIImpl::getUnrollingPreferences(Loop *L,
                                            TTI::UnrollingPreferences &UP) {
  UP.Threshold = 300; // Twice the default.
  UP.MaxCount = UINT_MAX;
  UP.Partial = true;

  // TODO: Do we want runtime unrolling?

  // Maximum alloca size than can fit registers. Reserve 16 registers.
  const unsigned MaxAlloca = (256 - 16) * 4;
  unsigned ThresholdPrivate = UnrollThresholdPrivate;
  unsigned ThresholdLocal = UnrollThresholdLocal;
  unsigned MaxBoost = std::max(ThresholdPrivate, ThresholdLocal);
  AMDGPUAS ASST = ST->getAMDGPUAS();
  for (const BasicBlock *BB : L->getBlocks()) {
    const DataLayout &DL = BB->getModule()->getDataLayout();
    unsigned LocalGEPsSeen = 0;

    if (any_of(L->getSubLoops(), [BB](const Loop* SubLoop) {
               return SubLoop->contains(BB); }))
        continue; // Block belongs to an inner loop.

    for (const Instruction &I : *BB) {

      // Unroll a loop which contains an "if" statement whose condition
      // defined by a PHI belonging to the loop. This may help to eliminate
      // if region and potentially even PHI itself, saving on both divergence
      // and registers used for the PHI.
      // Add a small bonus for each of such "if" statements.
      if (const BranchInst *Br = dyn_cast<BranchInst>(&I)) {
        if (UP.Threshold < MaxBoost && Br->isConditional()) {
          if (L->isLoopExiting(Br->getSuccessor(0)) ||
              L->isLoopExiting(Br->getSuccessor(1)))
            continue;
          if (dependsOnLocalPhi(L, Br->getCondition())) {
            UP.Threshold += UnrollThresholdIf;
            DEBUG(dbgs() << "Set unroll threshold " << UP.Threshold
                         << " for loop:\n" << *L << " due to " << *Br << '\n');
            if (UP.Threshold >= MaxBoost)
              return;
          }
        }
        continue;
      }

      const GetElementPtrInst *GEP = dyn_cast<GetElementPtrInst>(&I);
      if (!GEP)
        continue;

      unsigned AS = GEP->getAddressSpace();
      unsigned Threshold = 0;
      if (AS == ASST.PRIVATE_ADDRESS)
        Threshold = ThresholdPrivate;
      else if (AS == ASST.LOCAL_ADDRESS)
        Threshold = ThresholdLocal;
      else
        continue;

      if (UP.Threshold >= Threshold)
        continue;

      if (AS == ASST.PRIVATE_ADDRESS) {
        const Value *Ptr = GEP->getPointerOperand();
        const AllocaInst *Alloca =
            dyn_cast<AllocaInst>(GetUnderlyingObject(Ptr, DL));
        if (!Alloca || !Alloca->isStaticAlloca())
          continue;
        Type *Ty = Alloca->getAllocatedType();
        unsigned AllocaSize = Ty->isSized() ? DL.getTypeAllocSize(Ty) : 0;
        if (AllocaSize > MaxAlloca)
          continue;
      } else if (AS == ASST.LOCAL_ADDRESS) {
        LocalGEPsSeen++;
        // Inhibit unroll for local memory if we have seen addressing not to
        // a variable, most likely we will be unable to combine it.
        // Do not unroll too deep inner loops for local memory to give a chance
        // to unroll an outer loop for a more important reason.
        if (LocalGEPsSeen > 1 || L->getLoopDepth() > 2 ||
            (!isa<GlobalVariable>(GEP->getPointerOperand()) &&
             !isa<Argument>(GEP->getPointerOperand())))
          continue;
      }

      // Check if GEP depends on a value defined by this loop itself.
      bool HasLoopDef = false;
      for (const Value *Op : GEP->operands()) {
        const Instruction *Inst = dyn_cast<Instruction>(Op);
        if (!Inst || L->isLoopInvariant(Op))
          continue;

        if (any_of(L->getSubLoops(), [Inst](const Loop* SubLoop) {
             return SubLoop->contains(Inst); }))
          continue;
        HasLoopDef = true;
        break;
      }
      if (!HasLoopDef)
        continue;

      // We want to do whatever we can to limit the number of alloca
      // instructions that make it through to the code generator.  allocas
      // require us to use indirect addressing, which is slow and prone to
      // compiler bugs.  If this loop does an address calculation on an
      // alloca ptr, then we want to use a higher than normal loop unroll
      // threshold. This will give SROA a better chance to eliminate these
      // allocas.
      //
      // We also want to have more unrolling for local memory to let ds
      // instructions with different offsets combine.
      //
      // Don't use the maximum allowed value here as it will make some
      // programs way too big.
      UP.Threshold = Threshold;
      DEBUG(dbgs() << "Set unroll threshold " << Threshold << " for loop:\n"
                   << *L << " due to " << *GEP << '\n');
      if (UP.Threshold >= MaxBoost)
        return;
    }
  }
}

unsigned AMDGPUTTIImpl::getNumberOfRegisters(bool Vec) {
  if (Vec)
    return 0;

  // Number of VGPRs on SI.
  if (ST->getGeneration() >= AMDGPUSubtarget::SOUTHERN_ISLANDS)
    return 256;

  return 4 * 128; // XXX - 4 channels. Should these count as vector instead?
}

unsigned AMDGPUTTIImpl::getRegisterBitWidth(bool Vector) {
  return Vector ? 0 : 32;
}

unsigned AMDGPUTTIImpl::getLoadStoreVecRegBitWidth(unsigned AddrSpace) const {
  AMDGPUAS AS = ST->getAMDGPUAS();
  if (AddrSpace == AS.GLOBAL_ADDRESS ||
      AddrSpace == AS.CONSTANT_ADDRESS ||
      AddrSpace == AS.FLAT_ADDRESS)
    return 128;
  if (AddrSpace == AS.LOCAL_ADDRESS ||
      AddrSpace == AS.REGION_ADDRESS)
    return 64;
  if (AddrSpace == AS.PRIVATE_ADDRESS)
    return 8 * ST->getMaxPrivateElementSize();

  if (ST->getGeneration() <= AMDGPUSubtarget::NORTHERN_ISLANDS &&
      (AddrSpace == AS.PARAM_D_ADDRESS ||
      AddrSpace == AS.PARAM_I_ADDRESS ||
      (AddrSpace >= AS.CONSTANT_BUFFER_0 &&
      AddrSpace <= AS.CONSTANT_BUFFER_15)))
    return 128;
  llvm_unreachable("unhandled address space");
}

bool AMDGPUTTIImpl::isLegalToVectorizeMemChain(unsigned ChainSizeInBytes,
                                               unsigned Alignment,
                                               unsigned AddrSpace) const {
  // We allow vectorization of flat stores, even though we may need to decompose
  // them later if they may access private memory. We don't have enough context
  // here, and legalization can handle it.
  if (AddrSpace == ST->getAMDGPUAS().PRIVATE_ADDRESS) {
    return (Alignment >= 4 || ST->hasUnalignedScratchAccess()) &&
      ChainSizeInBytes <= ST->getMaxPrivateElementSize();
  }
  return true;
}

bool AMDGPUTTIImpl::isLegalToVectorizeLoadChain(unsigned ChainSizeInBytes,
                                                unsigned Alignment,
                                                unsigned AddrSpace) const {
  return isLegalToVectorizeMemChain(ChainSizeInBytes, Alignment, AddrSpace);
}

bool AMDGPUTTIImpl::isLegalToVectorizeStoreChain(unsigned ChainSizeInBytes,
                                                 unsigned Alignment,
                                                 unsigned AddrSpace) const {
  return isLegalToVectorizeMemChain(ChainSizeInBytes, Alignment, AddrSpace);
}

unsigned AMDGPUTTIImpl::getMaxInterleaveFactor(unsigned VF) {
  // Disable unrolling if the loop is not vectorized.
  if (VF == 1)
    return 1;

  // Semi-arbitrary large amount.
  return 64;
}

int AMDGPUTTIImpl::getArithmeticInstrCost(
    unsigned Opcode, Type *Ty, TTI::OperandValueKind Opd1Info,
    TTI::OperandValueKind Opd2Info, TTI::OperandValueProperties Opd1PropInfo,
    TTI::OperandValueProperties Opd2PropInfo, ArrayRef<const Value *> Args ) {

  EVT OrigTy = TLI->getValueType(DL, Ty);
  if (!OrigTy.isSimple()) {
    return BaseT::getArithmeticInstrCost(Opcode, Ty, Opd1Info, Opd2Info,
                                         Opd1PropInfo, Opd2PropInfo);
  }

  // Legalize the type.
  std::pair<int, MVT> LT = TLI->getTypeLegalizationCost(DL, Ty);
  int ISD = TLI->InstructionOpcodeToISD(Opcode);

  // Because we don't have any legal vector operations, but the legal types, we
  // need to account for split vectors.
  unsigned NElts = LT.second.isVector() ?
    LT.second.getVectorNumElements() : 1;

  MVT::SimpleValueType SLT = LT.second.getScalarType().SimpleTy;

  switch (ISD) {
  case ISD::SHL:
  case ISD::SRL:
  case ISD::SRA: {
    if (SLT == MVT::i64)
      return get64BitInstrCost() * LT.first * NElts;

    // i32
    return getFullRateInstrCost() * LT.first * NElts;
  }
  case ISD::ADD:
  case ISD::SUB:
  case ISD::AND:
  case ISD::OR:
  case ISD::XOR: {
    if (SLT == MVT::i64){
      // and, or and xor are typically split into 2 VALU instructions.
      return 2 * getFullRateInstrCost() * LT.first * NElts;
    }

    return LT.first * NElts * getFullRateInstrCost();
  }
  case ISD::MUL: {
    const int QuarterRateCost = getQuarterRateInstrCost();
    if (SLT == MVT::i64) {
      const int FullRateCost = getFullRateInstrCost();
      return (4 * QuarterRateCost + (2 * 2) * FullRateCost) * LT.first * NElts;
    }

    // i32
    return QuarterRateCost * NElts * LT.first;
  }
  case ISD::FADD:
  case ISD::FSUB:
  case ISD::FMUL:
    if (SLT == MVT::f64)
      return LT.first * NElts * get64BitInstrCost();

    if (SLT == MVT::f32 || SLT == MVT::f16)
      return LT.first * NElts * getFullRateInstrCost();
    break;

  case ISD::FDIV:
  case ISD::FREM:
    // FIXME: frem should be handled separately. The fdiv in it is most of it,
    // but the current lowering is also not entirely correct.
    if (SLT == MVT::f64) {
      int Cost = 4 * get64BitInstrCost() + 7 * getQuarterRateInstrCost();

      // Add cost of workaround.
      if (ST->getGeneration() == AMDGPUSubtarget::SOUTHERN_ISLANDS)
        Cost += 3 * getFullRateInstrCost();

      return LT.first * Cost * NElts;
    }

    // Assuming no fp32 denormals lowering.
    if (SLT == MVT::f32 || SLT == MVT::f16) {
      assert(!ST->hasFP32Denormals() && "will change when supported");
      int Cost = 7 * getFullRateInstrCost() + 1 * getQuarterRateInstrCost();
      return LT.first * NElts * Cost;
    }

    break;
  default:
    break;
  }

  return BaseT::getArithmeticInstrCost(Opcode, Ty, Opd1Info, Opd2Info,
                                       Opd1PropInfo, Opd2PropInfo);
}

unsigned AMDGPUTTIImpl::getCFInstrCost(unsigned Opcode) {
  // XXX - For some reason this isn't called for switch.
  switch (Opcode) {
  case Instruction::Br:
  case Instruction::Ret:
    return 10;
  default:
    return BaseT::getCFInstrCost(Opcode);
  }
}

int AMDGPUTTIImpl::getVectorInstrCost(unsigned Opcode, Type *ValTy,
                                      unsigned Index) {
  switch (Opcode) {
  case Instruction::ExtractElement:
  case Instruction::InsertElement:
    // Extracts are just reads of a subregister, so are free. Inserts are
    // considered free because we don't want to have any cost for scalarizing
    // operations, and we don't have to copy into a different register class.

    // Dynamic indexing isn't free and is best avoided.
    return Index == ~0u ? 2 : 0;
  default:
    return BaseT::getVectorInstrCost(Opcode, ValTy, Index);
  }
}

static bool isIntrinsicSourceOfDivergence(const IntrinsicInst *I) {
  switch (I->getIntrinsicID()) {
  case Intrinsic::amdgcn_workitem_id_x:
  case Intrinsic::amdgcn_workitem_id_y:
  case Intrinsic::amdgcn_workitem_id_z:
  case Intrinsic::amdgcn_interp_mov:
  case Intrinsic::amdgcn_interp_p1:
  case Intrinsic::amdgcn_interp_p2:
  case Intrinsic::amdgcn_mbcnt_hi:
  case Intrinsic::amdgcn_mbcnt_lo:
  case Intrinsic::r600_read_tidig_x:
  case Intrinsic::r600_read_tidig_y:
  case Intrinsic::r600_read_tidig_z:
  case Intrinsic::amdgcn_atomic_inc:
  case Intrinsic::amdgcn_atomic_dec:
  case Intrinsic::amdgcn_image_atomic_swap:
  case Intrinsic::amdgcn_image_atomic_add:
  case Intrinsic::amdgcn_image_atomic_sub:
  case Intrinsic::amdgcn_image_atomic_smin:
  case Intrinsic::amdgcn_image_atomic_umin:
  case Intrinsic::amdgcn_image_atomic_smax:
  case Intrinsic::amdgcn_image_atomic_umax:
  case Intrinsic::amdgcn_image_atomic_and:
  case Intrinsic::amdgcn_image_atomic_or:
  case Intrinsic::amdgcn_image_atomic_xor:
  case Intrinsic::amdgcn_image_atomic_inc:
  case Intrinsic::amdgcn_image_atomic_dec:
  case Intrinsic::amdgcn_image_atomic_cmpswap:
  case Intrinsic::amdgcn_buffer_atomic_swap:
  case Intrinsic::amdgcn_buffer_atomic_add:
  case Intrinsic::amdgcn_buffer_atomic_sub:
  case Intrinsic::amdgcn_buffer_atomic_smin:
  case Intrinsic::amdgcn_buffer_atomic_umin:
  case Intrinsic::amdgcn_buffer_atomic_smax:
  case Intrinsic::amdgcn_buffer_atomic_umax:
  case Intrinsic::amdgcn_buffer_atomic_and:
  case Intrinsic::amdgcn_buffer_atomic_or:
  case Intrinsic::amdgcn_buffer_atomic_xor:
  case Intrinsic::amdgcn_buffer_atomic_cmpswap:
  case Intrinsic::amdgcn_ps_live:
  case Intrinsic::amdgcn_ds_swizzle:
    return true;
  default:
    return false;
  }
}

static bool isArgPassedInSGPR(const Argument *A) {
  const Function *F = A->getParent();

  // Arguments to compute shaders are never a source of divergence.
  CallingConv::ID CC = F->getCallingConv();
  switch (CC) {
  case CallingConv::AMDGPU_KERNEL:
  case CallingConv::SPIR_KERNEL:
    return true;
  case CallingConv::AMDGPU_VS:
  case CallingConv::AMDGPU_GS:
  case CallingConv::AMDGPU_PS:
  case CallingConv::AMDGPU_CS:
    // For non-compute shaders, SGPR inputs are marked with either inreg or byval.
    // Everything else is in VGPRs.
    return F->getAttributes().hasParamAttribute(A->getArgNo(), Attribute::InReg) ||
           F->getAttributes().hasParamAttribute(A->getArgNo(), Attribute::ByVal);
  default:
    // TODO: Should calls support inreg for SGPR inputs?
    return false;
  }
}

///
/// \returns true if the result of the value could potentially be
/// different across workitems in a wavefront.
bool AMDGPUTTIImpl::isSourceOfDivergence(const Value *V) const {

  if (const Argument *A = dyn_cast<Argument>(V))
    return !isArgPassedInSGPR(A);

  // Loads from the private address space are divergent, because threads
  // can execute the load instruction with the same inputs and get different
  // results.
  //
  // All other loads are not divergent, because if threads issue loads with the
  // same arguments, they will always get the same result.
  if (const LoadInst *Load = dyn_cast<LoadInst>(V))
    return Load->getPointerAddressSpace() == ST->getAMDGPUAS().PRIVATE_ADDRESS;

  // Atomics are divergent because they are executed sequentially: when an
  // atomic operation refers to the same address in each thread, then each
  // thread after the first sees the value written by the previous thread as
  // original value.
  if (isa<AtomicRMWInst>(V) || isa<AtomicCmpXchgInst>(V))
    return true;

  if (const IntrinsicInst *Intrinsic = dyn_cast<IntrinsicInst>(V))
    return isIntrinsicSourceOfDivergence(Intrinsic);

  // Assume all function calls are a source of divergence.
  if (isa<CallInst>(V) || isa<InvokeInst>(V))
    return true;

  return false;
}