#include <rle.h>

Inheritance diagram for fz::RLEStage< T >:

Public Member Functions
void	setInverse (bool inverse) override

void	execute (cudaStream_t stream, MemoryPool pool, const std::vector< void > &inputs, const std::vector< void * > &outputs, const std::vector< size_t > &sizes) override

void	postStreamSync (cudaStream_t stream) override

std::string	getName () const override

size_t	estimateScratchBytes (const std::vector< size_t > &input_sizes) const override

std::vector< size_t >	estimateOutputSizes (const std::vector< size_t > &input_sizes) const override

std::unordered_map< std::string, size_t >	getActualOutputSizesByName () const override

size_t	getActualOutputSize (int index) const override

uint16_t	getStageTypeId () const override

uint8_t	getOutputDataType (size_t output_index) const override

uint8_t	getInputDataType (size_t) const override

size_t	serializeHeader (size_t output_index, uint8_t *header_buffer, size_t max_size) const override

void	deserializeHeader (const uint8_t *header_buffer, size_t size) override

size_t	getMaxHeaderSize (size_t output_index) const override

Public Member Functions inherited from fz::Stage
virtual size_t	getRequiredInputAlignment () const

virtual std::vector< std::string >	getOutputNames () const

int	getOutputIndex (const std::string &name) const

virtual void	saveState ()

virtual void	setDims (const std::array< size_t, 3 > &dims)

virtual void	onFinalize (size_t, MemoryPool *)

virtual size_t	estimateDeviceFootprintBytes (size_t) const

virtual size_t	estimatePinnedFootprintBytes (size_t) const

virtual bool	isGraphCompatible () const

Detailed Description

template<typename T = uint16_t>
class fz::RLEStage< T >

Run-Length Encoding stage. Lossless; effective when data has long runs of identical values (e.g. quantized codes).

Forward wire format: [num_runs:u32][values:T×n (4B-aligned)][lengths:u32×n]

Worst-case output is 2× input + 4 bytes (no repeated values), so RLE should follow a predictor/quantizer stage that creates repetition.

Template Parameters

T	Element type (`uint8_t`, `uint16_t`, `uint32_t`, …). Run counts are always `uint32_t`.

Member Function Documentation

◆ setInverse()

template<typename T = uint16_t>

void fz::RLEStage< T >::setInverse ( bool inverse )

inlineoverridevirtual

Switch between forward (compression) and inverse (decompression) mode. Affects getNumInputs()/getNumOutputs() for stages with asymmetric port counts.

Reimplemented from fz::Stage.

◆ execute()

template<typename T = uint16_t>

void fz::RLEStage< T >::execute	(	cudaStream_t	stream,
		MemoryPool *	pool,
		const std::vector< void * > &	inputs,
		const std::vector< void * > &	outputs,
		const std::vector< size_t > &	sizes
	)

overridevirtual

Execute the stage. Inputs, outputs, and sizes are device pointers/bytes.

Stages may call cudaStreamSynchronize(stream) or issue blocking D2H copies when the algorithm requires it (e.g. Huffman histogram readback for codebook construction, ANS renormalization tables). Such stages must return false from isGraphCompatible() and must document the sync points.

Note: the DAG dispatches sibling nodes (same topological level) via a sequential CPU loop, each enqueuing to its own stream. A sync inside execute() blocks the CPU from dispatching subsequent siblings until the synced stream is idle — this delays parallel branches in wide DAGs. In a linear pipeline there are no siblings and no extra cost.

Implements fz::Stage.

◆ postStreamSync()

template<typename T = uint16_t>

void fz::RLEStage< T >::postStreamSync ( cudaStream_t stream )

overridevirtual

Completes the async D2H readback of num_runs started during forward execute() and sets actual_output_sizes_. Must be called after the stream passed to execute() has been synchronized.

Reimplemented from fz::Stage.

◆ getName()

template<typename T = uint16_t>

std::string fz::RLEStage< T >::getName ( ) const

inlineoverridevirtual

Human-readable name used in error messages and debug output.

Implements fz::Stage.

◆ estimateScratchBytes()

template<typename T = uint16_t>

size_t fz::RLEStage< T >::estimateScratchBytes ( const std::vector< size_t > & input_sizes ) const

inlineoverridevirtual

Persistent forward-path scratch: d_is_boundary_ : n bytes d_boundary_scan_ : n × u32 d_boundary_positions_: n × u32 (worst-case, avoids D2H for num_runs) d_values_scratch_ : n × T d_lengths_scratch_ : n × u32 All five arrays are sized to the largest n seen so far and reused across calls, eliminating per-call cudaMallocAsync overhead.

Reimplemented from fz::Stage.

◆ estimateOutputSizes()

template<typename T = uint16_t>

std::vector< size_t > fz::RLEStage< T >::estimateOutputSizes ( const std::vector< size_t > & input_sizes ) const

inlineoverridevirtual

Estimate output buffer sizes given input sizes. Used for buffer allocation planning in PREALLOCATE mode — must be a safe upper bound; under-estimation causes buffer overruns.

Implements fz::Stage.

◆ getActualOutputSizesByName()

template<typename T = uint16_t>

std::unordered_map< std::string, size_t > fz::RLEStage< T >::getActualOutputSizesByName ( ) const

inlineoverridevirtual

Actual output sizes after execute(), keyed by output port name.

Implements fz::Stage.

◆ getActualOutputSize()

template<typename T = uint16_t>

size_t fz::RLEStage< T >::getActualOutputSize ( int index ) const

inlineoverridevirtual

Actual size of a single output by index after execute(). Avoids constructing the map for the common single-output case. Default delegates to getActualOutputSizesByName(); override to return directly from an internal field.

Reimplemented from fz::Stage.

◆ getStageTypeId()

template<typename T = uint16_t>

uint16_t fz::RLEStage< T >::getStageTypeId ( ) const

inlineoverridevirtual

Stage type identifier written into the FZM file header.

Implements fz::Stage.

◆ getOutputDataType()

template<typename T = uint16_t>

uint8_t fz::RLEStage< T >::getOutputDataType ( size_t output_index ) const

inlineoverridevirtual

DataType enum of the given output port.

Implements fz::Stage.

◆ getInputDataType()

template<typename T = uint16_t>

uint8_t fz::RLEStage< T >::getInputDataType ( size_t ) const

inlineoverridevirtual

Expected DataType of the given input port.

Used by Pipeline::finalize() to detect type mismatches between connected stages before any execution. Return DataType::UNKNOWN to opt out of checking — byte-transparent stages (Bitshuffle, RZE) and mock stages must return UNKNOWN; finalize() skips any connection where either side is UNKNOWN.

Reimplemented from fz::Stage.

◆ serializeHeader()

template<typename T = uint16_t>

size_t fz::RLEStage< T >::serializeHeader	(	size_t	output_index,
		uint8_t *	header_buffer,
		size_t	max_size
	)		const

inlineoverridevirtual

Serialize stage config into header_buffer (max 128 bytes) for the FZM file. Return the number of bytes written, or 0 if the stage has no config.

Reimplemented from fz::Stage.

◆ deserializeHeader()

template<typename T = uint16_t>

void fz::RLEStage< T >::deserializeHeader	(	const uint8_t *	header_buffer,
		size_t	size
	)

inlineoverridevirtual

Restore stage config from header_buffer during decompression.

Reimplemented from fz::Stage.

◆ getMaxHeaderSize()

template<typename T = uint16_t>

size_t fz::RLEStage< T >::getMaxHeaderSize ( size_t output_index ) const

inlineoverridevirtual

Maximum bytes this stage writes into its per-output FZM header slot.

Reimplemented from fz::Stage.

Public Member Functions

Detailed Description

Member Function Documentation

◆ setInverse()

◆ execute()

◆ postStreamSync()

◆ getName()

◆ estimateScratchBytes()

◆ estimateOutputSizes()

◆ getActualOutputSizesByName()

◆ getActualOutputSize()

◆ getStageTypeId()

◆ getOutputDataType()

◆ getInputDataType()

◆ serializeHeader()

◆ deserializeHeader()

◆ getMaxHeaderSize()