generateKernels.h File Reference

Contains functions that generate code for CUDA kernels. Part of the code generation section. More...

#include "modelSpec.h"
#include <string>
#include <fstream>

Go to the source code of this file.

Functions
void	genNeuronKernel (const NNmodel &model, const string &path)
	Function for generating the CUDA kernel that simulates all neurons in the model. More...
void	generate_process_presynaptic_events_code (ostream &os, const NNmodel &model, unsigned int src, unsigned int trg, int i, const string &localID, unsigned int inSynNo, const string &postfix)
	Function for generating the CUDA synapse kernel code that handles presynaptic spikes or spike type events. More...
void	genSynapseKernel (const NNmodel &model, const string &path)
	Function for generating a CUDA kernel for simulating all synapses. More...

Detailed Description

Contains functions that generate code for CUDA kernels. Part of the code generation section.

Function Documentation

void generate_process_presynaptic_events_code	(	ostream &	os,
		const NNmodel &	model,
		unsigned int	src,
		unsigned int	trg,
		int	i,
		const string &	localID,
		unsigned int	inSynNo,
		const string &	postfix
	)

Function for generating the CUDA synapse kernel code that handles presynaptic spikes or spike type events.

Parameters

os	output stream for code
model	the neuronal network model to generate code for
src	the number of the src neuron population
trg	the number of the target neuron population
i	the index of the synapse group being processed
localID	the variable name of the local ID of the thread within the synapse group
inSynNo	the ID number of the current synapse population as the incoming population to the target neuron population
postfix	whether to generate code for true spikes or spike type events

void genNeuronKernel	(	const NNmodel &	model,
		const string &	path
	)

Function for generating the CUDA kernel that simulates all neurons in the model.

The code generated upon execution of this function is for defining GPU side global variables that will hold model state in the GPU global memory and for the actual kernel function for simulating the neurons for one time step.

Binary flag for the sparse synapses to use atomic operations when the number of connections is bigger than the block size, and shared variables otherwise

Binary flag for the sparse synapses to use atomic operations when the number of connections is bigger than the block size, and shared variables otherwise

Parameters

model	Model description
path	Path for code generation

void genSynapseKernel	(	const NNmodel &	model,
		const string &	path
	)

Function for generating a CUDA kernel for simulating all synapses.

This functions generates code for global variables on the GPU side that are synapse-related and the actual CUDA kernel for simulating one time step of the synapses.

< "id" if first synapse group, else "lid". lid =(thread index- last thread of the last synapse group)

Parameters

model	Model description
path	Path for code generation