sdfg::passes::GPUNestedParallelizationPass Class Reference

Phased GPU nested map parallelization pass. More...

#include <sdfg/passes/offloading/gpu_nested_parallelization_pass.h>

Inheritance diagram for sdfg::passes::GPUNestedParallelizationPass:

Public Member Functions
	GPUNestedParallelizationPass (const std::vector< structured_control_flow::StructuredLoop * > &loops, GPUTarget target, size_t block_size)
	~GPUNestedParallelizationPass () override=default
bool	run_pass (builder::StructuredSDFGBuilder &builder, analysis::AnalysisManager &analysis_manager) override
std::string	name () override
Public Member Functions inherited from sdfg::passes::Pass
	Pass ()=default
	Pass (const Options &options)
virtual	~Pass ()=default
virtual std::vector< OptionSpec >	options ()
bool	run (builder::SDFGBuilder &builder, bool create_report=false)
bool	run (builder::StructuredSDFGBuilder &builder, analysis::AnalysisManager &analysis_manager, bool create_report=false)
virtual bool	run_pass (builder::SDFGBuilder &builder)
virtual void	invalidates (analysis::AnalysisManager &analysis_manager, bool applied)

Additional Inherited Members
Protected Member Functions inherited from sdfg::passes::Pass
template<class T>
T	option (const OptionKey< T > &key, T fallback=T{}) const

Detailed Description

Phased GPU nested map parallelization pass.

Given a set of outer loops, finds all nested loops within them and applies nested parallelization (CUDA or ROCM) in two phases:

1. can_be_applied phase: collects all nested loops where parallelization is applicable
2. apply phase: applies parallelization to all collected loops

Constructor & Destructor Documentation

GPUNestedParallelizationPass()

sdfg::passes::GPUNestedParallelizationPass::GPUNestedParallelizationPass	(	const std::vector< structured_control_flow::StructuredLoop * > &	loops,
		GPUTarget	target,
		size_t	block_size )

~GPUNestedParallelizationPass()

sdfg::passes::GPUNestedParallelizationPass::~GPUNestedParallelizationPass ( )

overridedefault

Member Function Documentation

name()

std::string sdfg::passes::GPUNestedParallelizationPass::name ( )

inlineoverridevirtual

Implements sdfg::passes::Pass.

run_pass()

bool sdfg::passes::GPUNestedParallelizationPass::run_pass ( builder::StructuredSDFGBuilder & builder, analysis::AnalysisManager & analysis_manager )

overridevirtual

Reimplemented from sdfg::passes::Pass.

---

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

• opt/include/sdfg/passes/offloading/gpu_nested_parallelization_pass.h