CodeGenerator::CUDA Namespace Reference

Namespaces
	Optimiser
	Utils

Classes
class	Backend
struct	Preferences
	Preferences for CUDA backend. More...

Typedefs
using	KernelBlockSize = std::array< size_t, KernelMax >
	Array of block sizes for each kernel. More...

Enumerations
enum	DeviceSelect { DeviceSelect::OPTIMAL, DeviceSelect::MOST_MEMORY, DeviceSelect::MANUAL }
	Methods for selecting CUDA device. More...
enum	BlockSizeSelect { BlockSizeSelect::OCCUPANCY, BlockSizeSelect::MANUAL }
	Methods for selecting CUDA kernel block size. More...
enum	Kernel {   KernelNeuronUpdate, KernelPresynapticUpdate, KernelPostsynapticUpdate, KernelSynapseDynamicsUpdate,   KernelInitialize, KernelInitializeSparse, KernelPreNeuronReset, KernelPreSynapseReset,   KernelMax }
	Kernels generated by CUDA backend. More...

Typedef Documentation

KernelBlockSize

using CodeGenerator::CUDA::KernelBlockSize = typedef std::array<size_t, KernelMax>

Array of block sizes for each kernel.

Enumeration Type Documentation

BlockSizeSelect

enum CodeGenerator::CUDA::BlockSizeSelect

strong

Methods for selecting CUDA kernel block size.

Enumerator
OCCUPANCY	Pick optimal blocksize for each kernel based on occupancy.
MANUAL	Use block sizes specified by user.

DeviceSelect

enum CodeGenerator::CUDA::DeviceSelect

strong

Methods for selecting CUDA device.

Enumerator
OPTIMAL	Pick optimal device based on how well kernels can be simultaneously simulated and occupancy.
MOST_MEMORY	Pick device with most global memory.
MANUAL	Use device specified by user.

Kernel

enum CodeGenerator::CUDA::Kernel

Kernels generated by CUDA backend.

Enumerator
KernelNeuronUpdate
KernelPresynapticUpdate
KernelPostsynapticUpdate
KernelSynapseDynamicsUpdate
KernelInitialize
KernelInitializeSparse
KernelPreNeuronReset
KernelPreSynapseReset
KernelMax