GPUOcelot
Public Member Functions | Protected Member Functions | Protected Attributes

executive::NVIDIAExecutableKernel Class Reference

#include <NVIDIAExecutableKernel.h>

Inheritance diagram for executive::NVIDIAExecutableKernel:
Inheritance graph
[legend]
Collaboration diagram for executive::NVIDIAExecutableKernel:
Collaboration graph
[legend]

List of all members.

Public Member Functions

 NVIDIAExecutableKernel (ir::IRKernel &kernel, const CUfunction &function, executive::Device *d=0)
 NVIDIAExecutableKernel ()
 ~NVIDIAExecutableKernel ()
void launchGrid (int width, int height, int depth)
void setKernelShape (int x, int y, int z)
void setDevice (const Device *device, unsigned int limit)
void setExternSharedMemorySize (unsigned int bytes)
void updateArgumentMemory ()
 Indicate that the kernels arguments have been updated.
void updateMemory ()
 Indicate that other memory has been updated.
TextureVector textureReferences () const
 Get a vector of all textures references by the kernel.
void updateGlobalMemory ()
void updateConstantMemory ()
void addTraceGenerator (trace::TraceGenerator *generator)
void removeTraceGenerator (trace::TraceGenerator *generator)
void setExternalFunctionSet (const ir::ExternalFunctionSet &s)
void clearExternalFunctionSet ()
void setWorkerThreads (unsigned int limit)
 Sets the max number of pthreads this kernel can use.

Protected Member Functions

void configureArguments ()
bool initialize ()

Protected Attributes

CUfunction cuFunction

Constructor & Destructor Documentation

executive::NVIDIAExecutableKernel::NVIDIAExecutableKernel ( ir::IRKernel kernel,
const CUfunction function,
executive::Device d = 0 
)

Construct a NVIDIAExecutableKernel from an existing kernel

executive::NVIDIAExecutableKernel::NVIDIAExecutableKernel ( )
executive::NVIDIAExecutableKernel::~NVIDIAExecutableKernel ( )

Member Function Documentation

void executive::NVIDIAExecutableKernel::addTraceGenerator ( trace::TraceGenerator generator) [virtual]

adds a trace generator to the EmulatedKernel

Reimplemented from executive::ExecutableKernel.

void executive::NVIDIAExecutableKernel::clearExternalFunctionSet ( ) [virtual]

clear the external function table for the emulated kernel

Implements executive::ExecutableKernel.

void executive::NVIDIAExecutableKernel::configureArguments ( ) [protected]

Configures the argument block for the CUDA driver API

bool executive::NVIDIAExecutableKernel::initialize ( ) [protected]

Construct

void executive::NVIDIAExecutableKernel::launchGrid ( int  width,
int  height,
int  depth 
) [virtual]

Launch a kernel on a 2D grid

Implements executive::ExecutableKernel.

void executive::NVIDIAExecutableKernel::removeTraceGenerator ( trace::TraceGenerator generator) [virtual]

removes a trace generator from an EmulatedKernel

Reimplemented from executive::ExecutableKernel.

void executive::NVIDIAExecutableKernel::setDevice ( const Device device,
unsigned int  limit 
)

Sets the device used to execute the kernel

void executive::NVIDIAExecutableKernel::setExternalFunctionSet ( const ir::ExternalFunctionSet s) [virtual]

sets an external function table for the emulated kernel

Implements executive::ExecutableKernel.

void executive::NVIDIAExecutableKernel::setExternSharedMemorySize ( unsigned int  bytes) [virtual]

sets the size of shared memory in bytes

Implements executive::ExecutableKernel.

void executive::NVIDIAExecutableKernel::setKernelShape ( int  x,
int  y,
int  z 
) [virtual]

Sets the shape of a kernel

Implements executive::ExecutableKernel.

void executive::NVIDIAExecutableKernel::setWorkerThreads ( unsigned int  workerThreadLimit) [virtual]

Sets the max number of pthreads this kernel can use.

Implements executive::ExecutableKernel.

executive::ExecutableKernel::TextureVector executive::NVIDIAExecutableKernel::textureReferences ( ) const [virtual]

Get a vector of all textures references by the kernel.

Implements executive::ExecutableKernel.

void executive::NVIDIAExecutableKernel::updateArgumentMemory ( ) [virtual]

Indicate that the kernels arguments have been updated.

Implements executive::ExecutableKernel.

void executive::NVIDIAExecutableKernel::updateConstantMemory ( )
void executive::NVIDIAExecutableKernel::updateGlobalMemory ( )
void executive::NVIDIAExecutableKernel::updateMemory ( ) [virtual]

Indicate that other memory has been updated.

Implements executive::ExecutableKernel.


Member Data Documentation

CUDA function refering to this kernel


The documentation for this class was generated from the following files:
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Defines