MLIRToLLVMPassPipelineConfig Struct Reference

Configuriation for the MLIR to LLVM pass pipeline. More...

#include <flang/Tools/CrossToolHelpers.h>

Inheritance diagram for MLIRToLLVMPassPipelineConfig:

Public Member Functions
	MLIRToLLVMPassPipelineConfig (llvm::OptimizationLevel level)
	MLIRToLLVMPassPipelineConfig (llvm::OptimizationLevel level, const Fortran::frontend::CodeGenOptions &opts, const Fortran::common::MathOptionsBase &mathOpts)
Public Member Functions inherited from FlangEPCallBacks
void	registerFIROptEarlyEPCallbacks (const std::function< void(mlir::PassManager &, llvm::OptimizationLevel)> &C)
void	registerFIRInlinerCallback (const std::function< void(mlir::PassManager &, llvm::OptimizationLevel)> &C)
void	registerFIROptLastEPCallbacks (const std::function< void(mlir::PassManager &, llvm::OptimizationLevel)> &C)
void	invokeFIROptEarlyEPCallbacks (mlir::PassManager &pm, llvm::OptimizationLevel optLevel)
void	invokeFIRInlinerCallback (mlir::PassManager &pm, llvm::OptimizationLevel optLevel)
void	invokeFIROptLastEPCallbacks (mlir::PassManager &pm, llvm::OptimizationLevel optLevel)

Public Attributes
llvm::OptimizationLevel	OptLevel
	optimisation level
bool	StackArrays = false
	convert memory allocations to alloca.
bool	Underscoring = true
	add underscores to function names.
bool	LoopVersioning = false
	Run the version loop pass.
bool	AliasAnalysis = false
	Add TBAA tags to generated LLVMIR.
llvm::codegenoptions::DebugInfoKind	DebugInfo
	Debug info generation.
llvm::FramePointerKind	FramePointerKind
	Add frame pointer to functions.
unsigned	VScaleMin = 0
	SVE vector range minimum.
unsigned	VScaleMax = 0
	SVE vector range maximum.
bool	NoInfsFPMath = false
	Set no-infs-fp-math attribute for functions.
bool	NoNaNsFPMath = false
	Set no-nans-fp-math attribute for functions.
bool	ApproxFuncFPMath = false
	Set afn flag for instructions.
bool	NoSignedZerosFPMath
	Set no-signed-zeros-fp-math attribute for functions.
bool	UnsafeFPMath = false
	Set unsafe-fp-math attribute for functions.
std::string	Reciprocals = ""
std::string	PreferVectorWidth = ""
bool	NSWOnLoopVarInc = true
	Add nsw flag to loop variable increments.
bool	EnableOpenMP = false
	Enable OpenMP lowering.
bool	EnableOpenMPSimd = false
	Enable OpenMP simd-only mode.
std::string	InstrumentFunctionEntry
std::string	InstrumentFunctionExit
Fortran::frontend::CodeGenOptions::ComplexRangeKind	ComplexRange
	Method for calculating complex number division.

Detailed Description

Configuriation for the MLIR to LLVM pass pipeline.

Member Data Documentation

ComplexRange

Fortran::frontend::CodeGenOptions::ComplexRangeKind MLIRToLLVMPassPipelineConfig::ComplexRange

Initial value:

=
      Fortran::frontend::CodeGenOptions::ComplexRangeKind::
          CX_Full

Method for calculating complex number division.

DebugInfo

llvm::codegenoptions::DebugInfoKind MLIRToLLVMPassPipelineConfig::DebugInfo

Initial value:

=
      llvm::codegenoptions::NoDebugInfo

Debug info generation.

FramePointerKind

llvm::FramePointerKind MLIRToLLVMPassPipelineConfig::FramePointerKind

Initial value:

=
      llvm::FramePointerKind::None

Add frame pointer to functions.

InstrumentFunctionEntry

std::string MLIRToLLVMPassPipelineConfig::InstrumentFunctionEntry

Initial value:

=
      ""

Name of the instrument-function that is called on each function-entry

InstrumentFunctionExit

std::string MLIRToLLVMPassPipelineConfig::InstrumentFunctionExit

Initial value:

=
      ""

Name of the instrument-function that is called on each function-exit

NoSignedZerosFPMath

bool MLIRToLLVMPassPipelineConfig::NoSignedZerosFPMath

Initial value:

=
      false

Set no-signed-zeros-fp-math attribute for functions.

PreferVectorWidth

std::string MLIRToLLVMPassPipelineConfig::PreferVectorWidth = ""

Set prefer-vector-width attribute for functions.

Reciprocals

std::string MLIRToLLVMPassPipelineConfig::Reciprocals = ""

Set reciprocal-estimate attribute for functions.

---

The documentation for this struct was generated from the following file:

• include/flang/Tools/CrossToolHelpers.h