mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-10-23 04:52:54 +02:00
da52f0e643
As we introduced target triple environment amdgiz and amdgizcl, the address space values are no longer enums. We have to decide the value by target triple. The basic idea is to use struct AMDGPUAS to represent address space values. For address space values which are not depend on target triple, use static const members, so that they don't occupy extra memory space and is equivalent to a compile time constant. Since the struct is lightweight and cheap, it can be created on the fly at the point of usage. Or it can be added as member to a pass and created at the beginning of the run* function. Differential Revision: https://reviews.llvm.org/D31284 llvm-svn: 298846
122 lines
4.2 KiB
C++
122 lines
4.2 KiB
C++
//===-- AMDGPUTargetTransformInfo.h - AMDGPU specific TTI -------*- C++ -*-===//
|
|
//
|
|
// The LLVM Compiler Infrastructure
|
|
//
|
|
// This file is distributed under the University of Illinois Open Source
|
|
// License. See LICENSE.TXT for details.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
/// \file
|
|
/// This file a TargetTransformInfo::Concept conforming object 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.
|
|
///
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#ifndef LLVM_LIB_TARGET_AMDGPU_AMDGPUTARGETTRANSFORMINFO_H
|
|
#define LLVM_LIB_TARGET_AMDGPU_AMDGPUTARGETTRANSFORMINFO_H
|
|
|
|
#include "AMDGPU.h"
|
|
#include "AMDGPUTargetMachine.h"
|
|
#include "llvm/Analysis/TargetTransformInfo.h"
|
|
#include "llvm/CodeGen/BasicTTIImpl.h"
|
|
|
|
namespace llvm {
|
|
class AMDGPUTargetLowering;
|
|
|
|
class AMDGPUTTIImpl final : public BasicTTIImplBase<AMDGPUTTIImpl> {
|
|
typedef BasicTTIImplBase<AMDGPUTTIImpl> BaseT;
|
|
typedef TargetTransformInfo TTI;
|
|
friend BaseT;
|
|
|
|
const AMDGPUSubtarget *ST;
|
|
const AMDGPUTargetLowering *TLI;
|
|
bool IsGraphicsShader;
|
|
|
|
const AMDGPUSubtarget *getST() const { return ST; }
|
|
const AMDGPUTargetLowering *getTLI() const { return TLI; }
|
|
|
|
|
|
static inline int getFullRateInstrCost() {
|
|
return TargetTransformInfo::TCC_Basic;
|
|
}
|
|
|
|
static inline int getHalfRateInstrCost() {
|
|
return 2 * TargetTransformInfo::TCC_Basic;
|
|
}
|
|
|
|
// TODO: The size is usually 8 bytes, but takes 4x as many cycles. Maybe
|
|
// should be 2 or 4.
|
|
static inline int getQuarterRateInstrCost() {
|
|
return 3 * TargetTransformInfo::TCC_Basic;
|
|
}
|
|
|
|
// On some parts, normal fp64 operations are half rate, and others
|
|
// quarter. This also applies to some integer operations.
|
|
inline int get64BitInstrCost() const {
|
|
return ST->hasHalfRate64Ops() ?
|
|
getHalfRateInstrCost() : getQuarterRateInstrCost();
|
|
}
|
|
|
|
public:
|
|
explicit AMDGPUTTIImpl(const AMDGPUTargetMachine *TM, const Function &F)
|
|
: BaseT(TM, F.getParent()->getDataLayout()),
|
|
ST(TM->getSubtargetImpl(F)),
|
|
TLI(ST->getTargetLowering()),
|
|
IsGraphicsShader(AMDGPU::isShader(F.getCallingConv())) {}
|
|
|
|
bool hasBranchDivergence() { return true; }
|
|
|
|
void getUnrollingPreferences(Loop *L, TTI::UnrollingPreferences &UP);
|
|
|
|
TTI::PopcntSupportKind getPopcntSupport(unsigned TyWidth) {
|
|
assert(isPowerOf2_32(TyWidth) && "Ty width must be power of 2");
|
|
return TTI::PSK_FastHardware;
|
|
}
|
|
|
|
unsigned getNumberOfRegisters(bool Vector);
|
|
unsigned getRegisterBitWidth(bool Vector);
|
|
unsigned getLoadStoreVecRegBitWidth(unsigned AddrSpace) const;
|
|
|
|
bool isLegalToVectorizeMemChain(unsigned ChainSizeInBytes,
|
|
unsigned Alignment,
|
|
unsigned AddrSpace) const;
|
|
bool isLegalToVectorizeLoadChain(unsigned ChainSizeInBytes,
|
|
unsigned Alignment,
|
|
unsigned AddrSpace) const;
|
|
bool isLegalToVectorizeStoreChain(unsigned ChainSizeInBytes,
|
|
unsigned Alignment,
|
|
unsigned AddrSpace) const;
|
|
|
|
unsigned getMaxInterleaveFactor(unsigned VF);
|
|
|
|
int getArithmeticInstrCost(
|
|
unsigned Opcode, Type *Ty,
|
|
TTI::OperandValueKind Opd1Info = TTI::OK_AnyValue,
|
|
TTI::OperandValueKind Opd2Info = TTI::OK_AnyValue,
|
|
TTI::OperandValueProperties Opd1PropInfo = TTI::OP_None,
|
|
TTI::OperandValueProperties Opd2PropInfo = TTI::OP_None,
|
|
ArrayRef<const Value *> Args = ArrayRef<const Value *>());
|
|
|
|
unsigned getCFInstrCost(unsigned Opcode);
|
|
|
|
int getVectorInstrCost(unsigned Opcode, Type *ValTy, unsigned Index);
|
|
bool isSourceOfDivergence(const Value *V) const;
|
|
|
|
unsigned getFlatAddressSpace() const {
|
|
// Don't bother running InferAddressSpaces pass on graphics shaders which
|
|
// don't use flat addressing.
|
|
if (IsGraphicsShader)
|
|
return -1;
|
|
return ST->hasFlatAddressSpace() ?
|
|
ST->getAMDGPUAS().FLAT_ADDRESS : ST->getAMDGPUAS().UNKNOWN_ADDRESS_SPACE;
|
|
}
|
|
|
|
unsigned getVectorSplitCost() { return 0; }
|
|
};
|
|
|
|
} // end namespace llvm
|
|
|
|
#endif
|