mlir::gpu::GPUToNVVMPipelineOptions Struct Reference

Options for the gpu to nvvm pipeline. More...

#include "mlir/Dialect/GPU/Pipelines/Passes.h"

Inheritance diagram for mlir::gpu::GPUToNVVMPipelineOptions:

Public Attributes
PassOptions::Option< int64_t >	indexBitWidth
PassOptions::Option< std::string >	cubinTriple
PassOptions::Option< std::string >	cubinChip
PassOptions::Option< std::string >	cubinFeatures
PassOptions::Option< std::string >	cubinFormat
PassOptions::Option< int >	optLevel
PassOptions::Option< bool >	kernelUseBarePtrCallConv
PassOptions::Option< bool >	hostUseBarePtrCallConv

Additional Inherited Members
Public Types inherited from mlir::detail::PassOptions
template<typename DataType >
using	OptionParser = std::conditional_t< std::is_base_of< llvm::cl::generic_parser_base, llvm::cl::parser< DataType > >::value, GenericOptionParser< DataType >, llvm::cl::parser< DataType > >
	The specific parser to use depending on llvm::cl parser used. More...
Public Member Functions inherited from mlir::detail::PassOptions
	PassOptions ()=default
	PassOptions (const PassOptions &)=delete
	Delete the copy constructor to avoid copying the internal options map. More...
	PassOptions (PassOptions &&)=delete
void	copyOptionValuesFrom (const PassOptions &other)
	Copy the option values from 'other' into 'this', where 'other' has the same options as 'this'. More...
LogicalResult	parseFromString (StringRef options, raw_ostream &errorStream=llvm::errs())
	Parse options out as key=value pairs that can then be handed off to the llvm::cl command line passing infrastructure. More...
void	print (raw_ostream &os)
	Print the options held by this struct in a form that can be parsed via 'parseFromString'. More...
void	printHelp (size_t indent, size_t descIndent) const
	Print the help string for the options held by this struct. More...
size_t	getOptionWidth () const
	Return the maximum width required when printing the help string. More...
Static Public Member Functions inherited from mlir::PassPipelineOptions< GPUToNVVMPipelineOptions >
static std::unique_ptr< GPUToNVVMPipelineOptions >	createFromString (StringRef options)
	Factory that parses the provided options and returns a unique_ptr to the struct. More...

Detailed Description

Options for the gpu to nvvm pipeline.

Definition at line 18 of file Passes.h.

Member Data Documentation

cubinChip

PassOptions::Option<std::string> mlir::gpu::GPUToNVVMPipelineOptions::cubinChip

Initial value:

{
      *this, "cubin-chip", llvm::cl::desc("Chip to use to serialize to cubin."),
      llvm::cl::init("sm_50")}

Definition at line 29 of file Passes.h.

cubinFeatures

PassOptions::Option<std::string> mlir::gpu::GPUToNVVMPipelineOptions::cubinFeatures

Initial value:

{
      *this, "cubin-features",
      llvm::cl::desc("Features to use to serialize to cubin."),
      llvm::cl::init("+ptx60")}

Definition at line 32 of file Passes.h.

cubinFormat

PassOptions::Option<std::string> mlir::gpu::GPUToNVVMPipelineOptions::cubinFormat

Initial value:

{
      *this, "cubin-format",
      llvm::cl::desc("Compilation format to use to serialize to cubin."),
      llvm::cl::init("fatbin")}

Definition at line 36 of file Passes.h.

cubinTriple

PassOptions::Option<std::string> mlir::gpu::GPUToNVVMPipelineOptions::cubinTriple

Initial value:

{
      *this, "cubin-triple",
      llvm::cl::desc("Triple to use to serialize to cubin."),
      llvm::cl::init("nvptx64-nvidia-cuda")}

Definition at line 25 of file Passes.h.

hostUseBarePtrCallConv

PassOptions::Option<bool> mlir::gpu::GPUToNVVMPipelineOptions::hostUseBarePtrCallConv

Initial value:

{
      *this, "host-bare-ptr-calling-convention",
      llvm::cl::desc(
          "Whether to use the bareptr calling convention on the host (warning "
          "this should be false until the GPU layering is fixed)"),
      llvm::cl::init(false)}

Definition at line 50 of file Passes.h.

indexBitWidth

PassOptions::Option<int64_t> mlir::gpu::GPUToNVVMPipelineOptions::indexBitWidth

Initial value:

{
      *this, "index-bitwidth",
      llvm::cl::desc("Bitwidth of the index type for the host (warning this "
                     "should be 64 until the GPU layering is fixed)"),
      llvm::cl::init(64)}

Definition at line 20 of file Passes.h.

kernelUseBarePtrCallConv

PassOptions::Option<bool> mlir::gpu::GPUToNVVMPipelineOptions::kernelUseBarePtrCallConv

Initial value:

{
      *this, "kernel-bare-ptr-calling-convention",
      llvm::cl::desc(
          "Whether to use the bareptr calling convention on the kernel "
          "(warning this should be false until the GPU layering is fixed)"),
      llvm::cl::init(false)}

Definition at line 44 of file Passes.h.

optLevel

PassOptions::Option<int> mlir::gpu::GPUToNVVMPipelineOptions::optLevel

Initial value:

{
      *this, "opt-level",
      llvm::cl::desc("Optimization level for NVVM compilation"),
      llvm::cl::init(2)}

Definition at line 40 of file Passes.h.

---

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

• include/mlir/Dialect/GPU/Pipelines/Passes.h