This restores commitc7fdd8c11e. Previously reverted inf010b1bef4. LLVM function calls carry convergence control tokens as operand bundles, where the tokens themselves are produced by convergence control intrinsics. This patch implements convergence control tokens in MIR as follows: 1. Introduce target-independent ISD opcodes and MIR opcodes for convergence control intrinsics. 2. Model token values as untyped virtual registers in MIR. The change also introduces an additional ISD opcode CONVERGENCECTRL_GLUE and a corresponding machine opcode with the same spelling. This glues the convergence control token to SDNodes that represent calls to intrinsics. The glued token is later translated to an implicit argument in the MIR. The lowering of calls to user-defined functions is target-specific. On AMDGPU, the convergence control operand bundle at a non-intrinsic call is translated to an explicit argument to the SI_CALL_ISEL instruction. Post-selection adjustment converts this explicit argument to an implicit argument on the SI_CALL instruction.
154 lines
5.4 KiB
C++
154 lines
5.4 KiB
C++
//===-- AMDGPUTargetMachine.h - AMDGPU TargetMachine Interface --*- C++ -*-===//
|
|
//
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
//
|
|
/// \file
|
|
/// The AMDGPU TargetMachine interface definition for hw codegen targets.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#ifndef LLVM_LIB_TARGET_AMDGPU_AMDGPUTARGETMACHINE_H
|
|
#define LLVM_LIB_TARGET_AMDGPU_AMDGPUTARGETMACHINE_H
|
|
|
|
#include "GCNSubtarget.h"
|
|
#include "llvm/CodeGen/TargetPassConfig.h"
|
|
#include "llvm/Target/TargetMachine.h"
|
|
#include <optional>
|
|
#include <utility>
|
|
|
|
namespace llvm {
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// AMDGPU Target Machine (R600+)
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
class AMDGPUTargetMachine : public LLVMTargetMachine {
|
|
protected:
|
|
std::unique_ptr<TargetLoweringObjectFile> TLOF;
|
|
|
|
StringRef getGPUName(const Function &F) const;
|
|
StringRef getFeatureString(const Function &F) const;
|
|
|
|
public:
|
|
static bool EnableLateStructurizeCFG;
|
|
static bool EnableFunctionCalls;
|
|
static bool EnableLowerModuleLDS;
|
|
static bool DisableStructurizer;
|
|
|
|
AMDGPUTargetMachine(const Target &T, const Triple &TT, StringRef CPU,
|
|
StringRef FS, const TargetOptions &Options,
|
|
std::optional<Reloc::Model> RM,
|
|
std::optional<CodeModel::Model> CM, CodeGenOptLevel OL);
|
|
~AMDGPUTargetMachine() override;
|
|
|
|
const TargetSubtargetInfo *getSubtargetImpl() const;
|
|
const TargetSubtargetInfo *getSubtargetImpl(const Function &) const override = 0;
|
|
|
|
TargetLoweringObjectFile *getObjFileLowering() const override {
|
|
return TLOF.get();
|
|
}
|
|
|
|
void registerPassBuilderCallbacks(PassBuilder &PB,
|
|
bool PopulateClassToPassNames) override;
|
|
void registerDefaultAliasAnalyses(AAManager &) override;
|
|
|
|
/// Get the integer value of a null pointer in the given address space.
|
|
static int64_t getNullPointerValue(unsigned AddrSpace);
|
|
|
|
bool isNoopAddrSpaceCast(unsigned SrcAS, unsigned DestAS) const override;
|
|
|
|
unsigned getAssumedAddrSpace(const Value *V) const override;
|
|
|
|
std::pair<const Value *, unsigned>
|
|
getPredicatedAddrSpace(const Value *V) const override;
|
|
|
|
unsigned getAddressSpaceForPseudoSourceKind(unsigned Kind) const override;
|
|
};
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// GCN Target Machine (SI+)
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
class GCNTargetMachine final : public AMDGPUTargetMachine {
|
|
private:
|
|
mutable StringMap<std::unique_ptr<GCNSubtarget>> SubtargetMap;
|
|
|
|
public:
|
|
GCNTargetMachine(const Target &T, const Triple &TT, StringRef CPU,
|
|
StringRef FS, const TargetOptions &Options,
|
|
std::optional<Reloc::Model> RM,
|
|
std::optional<CodeModel::Model> CM, CodeGenOptLevel OL,
|
|
bool JIT);
|
|
|
|
TargetPassConfig *createPassConfig(PassManagerBase &PM) override;
|
|
|
|
const TargetSubtargetInfo *getSubtargetImpl(const Function &) const override;
|
|
|
|
TargetTransformInfo getTargetTransformInfo(const Function &F) const override;
|
|
|
|
bool useIPRA() const override {
|
|
return true;
|
|
}
|
|
|
|
void registerMachineRegisterInfoCallback(MachineFunction &MF) const override;
|
|
|
|
MachineFunctionInfo *
|
|
createMachineFunctionInfo(BumpPtrAllocator &Allocator, const Function &F,
|
|
const TargetSubtargetInfo *STI) const override;
|
|
|
|
yaml::MachineFunctionInfo *createDefaultFuncInfoYAML() const override;
|
|
yaml::MachineFunctionInfo *
|
|
convertFuncInfoToYAML(const MachineFunction &MF) const override;
|
|
bool parseMachineFunctionInfo(const yaml::MachineFunctionInfo &,
|
|
PerFunctionMIParsingState &PFS,
|
|
SMDiagnostic &Error,
|
|
SMRange &SourceRange) const override;
|
|
};
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
// AMDGPU Pass Setup
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
class AMDGPUPassConfig : public TargetPassConfig {
|
|
public:
|
|
AMDGPUPassConfig(LLVMTargetMachine &TM, PassManagerBase &PM);
|
|
|
|
AMDGPUTargetMachine &getAMDGPUTargetMachine() const {
|
|
return getTM<AMDGPUTargetMachine>();
|
|
}
|
|
|
|
ScheduleDAGInstrs *
|
|
createMachineScheduler(MachineSchedContext *C) const override;
|
|
|
|
void addEarlyCSEOrGVNPass();
|
|
void addStraightLineScalarOptimizationPasses();
|
|
void addIRPasses() override;
|
|
void addCodeGenPrepare() override;
|
|
bool addPreISel() override;
|
|
bool addInstSelector() override;
|
|
bool addGCPasses() override;
|
|
|
|
std::unique_ptr<CSEConfigBase> getCSEConfig() const override;
|
|
|
|
/// Check if a pass is enabled given \p Opt option. The option always
|
|
/// overrides defaults if explicitly used. Otherwise its default will
|
|
/// be used given that a pass shall work at an optimization \p Level
|
|
/// minimum.
|
|
bool isPassEnabled(const cl::opt<bool> &Opt,
|
|
CodeGenOptLevel Level = CodeGenOptLevel::Default) const {
|
|
if (Opt.getNumOccurrences())
|
|
return Opt;
|
|
if (TM->getOptLevel() < Level)
|
|
return false;
|
|
return Opt;
|
|
}
|
|
};
|
|
|
|
} // end namespace llvm
|
|
|
|
#endif // LLVM_LIB_TARGET_AMDGPU_AMDGPUTARGETMACHINE_H
|