GPUOcelot
|
#include <KernelExtractor.h>
Public Types | |
typedef std::unordered_map < CUmodule, std::string > | ModuleNameMap |
typedef std::unordered_map < CUfunction, std::pair < std::string, std::string > > | FunctionNameMap |
typedef std::unordered_map < CUtexref, std::pair < std::string, std::string > > | TextureNameMap |
Public Member Functions | |
KernelExtractorDriver () | |
virtual | ~KernelExtractorDriver () |
void | synchronizeFromDevice () |
copies data from device to host-side allocations | |
void | synchronizeToDevice () |
copies data from host-side allocations to device | |
void | loadModule (CUresult result, CUmodule module, const char *ptxImage, const char *name=0) |
binds module handle to PTX image | |
void | bindKernel (CUresult result, CUmodule module, CUfunction function, const char *name) |
binds a function handle to a module and kernel name | |
void | bindTexture (CUresult result, CUmodule module, CUtexref texture, const char *name) |
binds a texture handle to a module and texture name | |
void | bindGlobal (CUresult result, CUmodule module, void *ptr, const char *name) |
binds a global variable to a pointer | |
void | kernelLaunch (CUfunction f, int gridX=1, int gridY=1) |
called when a kernel is launched | |
void | kernelReturn (CUresult result) |
called when a kernel returns | |
void | allocate (CUresult result, void *dptr, size_t bytes) |
allocates device memory | |
void | free (void *dptr) |
deletes an allocation | |
CUresult | cuInit (unsigned int Flags) |
CUresult | cuDriverGetVersion (int *driverVersion) |
CUresult | cuGetExportTable (const void **ppExportTable, const CUuuid *pExportTableId) |
CUresult | cuDeviceGet (CUdevice *device, int ordinal) |
CUresult | cuDeviceGetCount (int *count) |
CUresult | cuDeviceGetName (char *name, int len, CUdevice dev) |
CUresult | cuDeviceComputeCapability (int *major, int *minor, CUdevice dev) |
CUresult | cuDeviceTotalMem (size_t *bytes, CUdevice dev) |
CUresult | cuDeviceGetProperties (CUdevprop *prop, CUdevice dev) |
CUresult | cuDeviceGetAttribute (int *pi, CUdevice_attribute attrib, CUdevice dev) |
CUresult | cuCtxCreate (CUcontext *pctx, unsigned int flags, CUdevice dev) |
CUresult | cuCtxDestroy (CUcontext ctx) |
CUresult | cuCtxAttach (CUcontext *pctx, unsigned int flags) |
CUresult | cuCtxDetach (CUcontext ctx) |
CUresult | cuCtxPushCurrent (CUcontext ctx) |
CUresult | cuCtxPopCurrent (CUcontext *pctx) |
CUresult | cuCtxGetDevice (CUdevice *device) |
CUresult | cuCtxSynchronize (void) |
CUresult | cuModuleLoad (CUmodule *module, const char *fname) |
CUresult | cuModuleLoadData (CUmodule *module, const void *image) |
CUresult | cuModuleLoadDataEx (CUmodule *module, const void *image, unsigned int numOptions, CUjit_option *options, void **optionValues) |
CUresult | cuModuleLoadFatBinary (CUmodule *module, const void *fatCubin) |
CUresult | cuModuleUnload (CUmodule hmod) |
CUresult | cuModuleGetFunction (CUfunction *hfunc, CUmodule hmod, const char *name) |
CUresult | cuModuleGetGlobal (CUdeviceptr *dptr, size_t *bytes, CUmodule hmod, const char *name) |
CUresult | cuModuleGetTexRef (CUtexref *pTexRef, CUmodule hmod, const char *name) |
CUresult | cuMemGetInfo (size_t *free, size_t *total) |
CUresult | cuMemAlloc (CUdeviceptr *dptr, unsigned int bytesize) |
CUresult | cuMemAllocPitch (CUdeviceptr *dptr, size_t *pPitch, unsigned int WidthInBytes, unsigned int Height, unsigned int ElementSizeBytes) |
CUresult | cuMemFree (CUdeviceptr dptr) |
CUresult | cuMemGetAddressRange (CUdeviceptr *pbase, size_t *psize, CUdeviceptr dptr) |
CUresult | cuMemAllocHost (void **pp, unsigned int bytesize) |
CUresult | cuMemFreeHost (void *p) |
CUresult | cuMemHostAlloc (void **pp, unsigned long long bytesize, unsigned int Flags) |
CUresult | cuMemHostGetDevicePointer (CUdeviceptr *pdptr, void *p, unsigned int Flags) |
CUresult | cuMemHostGetFlags (unsigned int *pFlags, void *p) |
CUresult | cuMemcpyHtoD (CUdeviceptr dstDevice, const void *srcHost, unsigned int ByteCount) |
CUresult | cuMemcpyDtoH (void *dstHost, CUdeviceptr srcDevice, unsigned int ByteCount) |
CUresult | cuMemcpyDtoD (CUdeviceptr dstDevice, CUdeviceptr srcDevice, unsigned int ByteCount) |
CUresult | cuMemcpyDtoA (CUarray dstArray, unsigned int dstIndex, CUdeviceptr srcDevice, unsigned int ByteCount) |
CUresult | cuMemcpyAtoD (CUdeviceptr dstDevice, CUarray hSrc, unsigned int SrcIndex, unsigned int ByteCount) |
CUresult | cuMemcpyHtoA (CUarray dstArray, unsigned int dstIndex, const void *pSrc, unsigned int ByteCount) |
CUresult | cuMemcpyAtoH (void *dstHost, CUarray srcArray, unsigned int srcIndex, unsigned int ByteCount) |
CUresult | cuMemcpyAtoA (CUarray dstArray, unsigned int dstIndex, CUarray srcArray, unsigned int srcIndex, unsigned int ByteCount) |
CUresult | cuMemcpy2D (const CUDA_MEMCPY2D *pCopy) |
CUresult | cuMemcpy2DUnaligned (const CUDA_MEMCPY2D *pCopy) |
CUresult | cuMemcpy3D (const CUDA_MEMCPY3D *pCopy) |
CUresult | cuMemcpyHtoDAsync (CUdeviceptr dstDevice, const void *srcHost, unsigned int ByteCount, CUstream hStream) |
CUresult | cuMemcpyDtoHAsync (void *dstHost, CUdeviceptr srcDevice, unsigned int ByteCount, CUstream hStream) |
CUresult | cuMemcpyHtoAAsync (CUarray dstArray, unsigned int dstIndex, const void *pSrc, unsigned int ByteCount, CUstream hStream) |
CUresult | cuMemcpyAtoHAsync (void *dstHost, CUarray srcArray, unsigned int srcIndex, unsigned int ByteCount, CUstream hStream) |
CUresult | cuMemcpy2DAsync (const CUDA_MEMCPY2D *pCopy, CUstream hStream) |
CUresult | cuMemcpy3DAsync (const CUDA_MEMCPY3D *pCopy, CUstream hStream) |
CUresult | cuMemsetD8 (CUdeviceptr dstDevice, unsigned char uc, unsigned int N) |
CUresult | cuMemsetD16 (CUdeviceptr dstDevice, unsigned short us, unsigned int N) |
CUresult | cuMemsetD32 (CUdeviceptr dstDevice, unsigned int ui, unsigned int N) |
CUresult | cuMemsetD2D8 (CUdeviceptr dstDevice, unsigned int dstPitch, unsigned char uc, unsigned int Width, unsigned int Height) |
CUresult | cuMemsetD2D16 (CUdeviceptr dstDevice, unsigned int dstPitch, unsigned short us, unsigned int Width, unsigned int Height) |
CUresult | cuMemsetD2D32 (CUdeviceptr dstDevice, unsigned int dstPitch, unsigned int ui, unsigned int Width, unsigned int Height) |
CUresult | cuFuncSetBlockShape (CUfunction hfunc, int x, int y, int z) |
CUresult | cuFuncSetSharedSize (CUfunction hfunc, unsigned int bytes) |
CUresult | cuFuncGetAttribute (int *pi, CUfunction_attribute attrib, CUfunction hfunc) |
CUresult | cuArrayCreate (CUarray *pHandle, const CUDA_ARRAY_DESCRIPTOR *pAllocateArray) |
CUresult | cuArrayGetDescriptor (CUDA_ARRAY_DESCRIPTOR *pArrayDescriptor, CUarray hArray) |
CUresult | cuArrayDestroy (CUarray hArray) |
CUresult | cuArray3DCreate (CUarray *pHandle, const CUDA_ARRAY3D_DESCRIPTOR *pAllocateArray) |
CUresult | cuArray3DGetDescriptor (CUDA_ARRAY3D_DESCRIPTOR *pArrayDescriptor, CUarray hArray) |
CUresult | cuTexRefCreate (CUtexref *pTexRef) |
CUresult | cuTexRefDestroy (CUtexref hTexRef) |
CUresult | cuTexRefSetArray (CUtexref hTexRef, CUarray hArray, unsigned int Flags) |
CUresult | cuTexRefSetAddress (size_t *ByteOffset, CUtexref hTexRef, CUdeviceptr dptr, size_t bytes) |
CUresult | cuTexRefSetAddress2D (CUtexref hTexRef, const CUDA_ARRAY_DESCRIPTOR *desc, CUdeviceptr dptr, unsigned int Pitch) |
CUresult | cuTexRefSetFormat (CUtexref hTexRef, CUarray_format fmt, int NumPackedComponents) |
CUresult | cuTexRefSetAddressMode (CUtexref hTexRef, int dim, CUaddress_mode am) |
CUresult | cuTexRefSetFilterMode (CUtexref hTexRef, CUfilter_mode fm) |
CUresult | cuTexRefSetFlags (CUtexref hTexRef, unsigned int Flags) |
CUresult | cuTexRefGetAddress (CUdeviceptr *pdptr, CUtexref hTexRef) |
CUresult | cuTexRefGetArray (CUarray *phArray, CUtexref hTexRef) |
CUresult | cuTexRefGetAddressMode (CUaddress_mode *pam, CUtexref hTexRef, int dim) |
CUresult | cuTexRefGetFilterMode (CUfilter_mode *pfm, CUtexref hTexRef) |
CUresult | cuTexRefGetFormat (CUarray_format *pFormat, int *pNumChannels, CUtexref hTexRef) |
CUresult | cuTexRefGetFlags (unsigned int *pFlags, CUtexref hTexRef) |
CUresult | cuParamSetSize (CUfunction hfunc, unsigned int numbytes) |
CUresult | cuParamSeti (CUfunction hfunc, int offset, unsigned int value) |
CUresult | cuParamSetf (CUfunction hfunc, int offset, float value) |
CUresult | cuParamSetv (CUfunction hfunc, int offset, void *ptr, unsigned int numbytes) |
CUresult | cuParamSetTexRef (CUfunction hfunc, int texunit, CUtexref hTexRef) |
CUresult | cuLaunch (CUfunction f) |
CUresult | cuLaunchGrid (CUfunction f, int grid_width, int grid_height) |
CUresult | cuLaunchGridAsync (CUfunction f, int grid_width, int grid_height, CUstream hStream) |
CUresult | cuEventCreate (CUevent *phEvent, unsigned int Flags) |
CUresult | cuEventRecord (CUevent hEvent, CUstream hStream) |
CUresult | cuEventQuery (CUevent hEvent) |
CUresult | cuEventSynchronize (CUevent hEvent) |
CUresult | cuEventDestroy (CUevent hEvent) |
CUresult | cuEventElapsedTime (float *pMilliseconds, CUevent hStart, CUevent hEnd) |
CUresult | cuStreamCreate (CUstream *phStream, unsigned int Flags) |
CUresult | cuStreamQuery (CUstream hStream) |
CUresult | cuStreamSynchronize (CUstream hStream) |
CUresult | cuStreamDestroy (CUstream hStream) |
CUresult | cuGraphicsUnregisterResource (CUgraphicsResource resource) |
CUresult | cuGraphicsSubResourceGetMappedArray (CUarray *pArray, CUgraphicsResource resource, unsigned int arrayIndex, unsigned int mipLevel) |
CUresult | cuGraphicsResourceGetMappedPointer (CUdeviceptr *pDevPtr, size_t *pSize, CUgraphicsResource resource) |
CUresult | cuGraphicsResourceSetMapFlags (CUgraphicsResource resource, unsigned int flags) |
CUresult | cuGraphicsMapResources (unsigned int count, CUgraphicsResource *resources, CUstream hStream) |
CUresult | cuGraphicsUnmapResources (unsigned int count, CUgraphicsResource *resources, CUstream hStream) |
CUresult | cuGLInit () |
CUresult | cuGLCtxCreate (CUcontext *pCtx, unsigned int Flags, CUdevice device) |
CUresult | cuGraphicsGLRegisterBuffer (CUgraphicsResource *pCudaResource, unsigned int buffer, unsigned int Flags) |
CUresult | cuGraphicsGLRegisterImage (CUgraphicsResource *pCudaResource, unsigned int image, int target, unsigned int Flags) |
CUresult | cuGLRegisterBufferObject (GLuint bufferobj) |
CUresult | cuGLSetBufferObjectMapFlags (GLuint buffer, unsigned int flags) |
std::string | toString (CUresult result) |
Public Attributes | |
cuda::CudaDriver::Interface | cudaDriver |
ExtractedDeviceState | state |
object for serializing CUDA kernels and device state | |
bool | enabled |
determines whether kernels are actually extracted | |
ModuleNameMap | moduleNameMap |
maps module handles to module names | |
FunctionNameMap | functionNameMap |
maps function handles to (module name, function name) pairs | |
TextureNameMap | textureNameMap |
maps texture handles to (module name, texture name) pairs | |
Static Public Attributes | |
static KernelExtractorDriver | instance |
singleton instance |
typedef std::unordered_map< CUfunction, std::pair< std::string, std::string > > util::KernelExtractorDriver::FunctionNameMap |
typedef std::unordered_map< CUmodule, std::string > util::KernelExtractorDriver::ModuleNameMap |
typedef std::unordered_map< CUtexref, std::pair< std::string, std::string > > util::KernelExtractorDriver::TextureNameMap |
util::KernelExtractorDriver::KernelExtractorDriver | ( | ) |
util::KernelExtractorDriver::~KernelExtractorDriver | ( | ) | [virtual] |
allocates device memory
void util::KernelExtractorDriver::bindGlobal | ( | CUresult | result, |
CUmodule | module, | ||
void * | ptr, | ||
const char * | name | ||
) |
binds a global variable to a pointer
void util::KernelExtractorDriver::bindKernel | ( | CUresult | result, |
CUmodule | module, | ||
CUfunction | function, | ||
const char * | name | ||
) |
binds a function handle to a module and kernel name
void util::KernelExtractorDriver::bindTexture | ( | CUresult | result, |
CUmodule | module, | ||
CUtexref | texture, | ||
const char * | name | ||
) |
binds a texture handle to a module and texture name
CUresult util::KernelExtractorDriver::cuArray3DCreate | ( | CUarray * | pHandle, |
const CUDA_ARRAY3D_DESCRIPTOR * | pAllocateArray | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuArray3DGetDescriptor | ( | CUDA_ARRAY3D_DESCRIPTOR * | pArrayDescriptor, |
CUarray | hArray | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuArrayCreate | ( | CUarray * | pHandle, |
const CUDA_ARRAY_DESCRIPTOR * | pAllocateArray | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuArrayGetDescriptor | ( | CUDA_ARRAY_DESCRIPTOR * | pArrayDescriptor, |
CUarray | hArray | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuCtxAttach | ( | CUcontext * | pctx, |
unsigned int | flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuCtxCreate | ( | CUcontext * | pctx, |
unsigned int | flags, | ||
CUdevice | dev | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuCtxSynchronize | ( | void | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuDeviceComputeCapability | ( | int * | major, |
int * | minor, | ||
CUdevice | dev | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuDeviceGetAttribute | ( | int * | pi, |
CUdevice_attribute | attrib, | ||
CUdevice | dev | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuDeviceGetCount | ( | int * | count | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuDeviceGetName | ( | char * | name, |
int | len, | ||
CUdevice | dev | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuDeviceGetProperties | ( | CUdevprop * | prop, |
CUdevice | dev | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuDriverGetVersion | ( | int * | driverVersion | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuEventCreate | ( | CUevent * | phEvent, |
unsigned int | Flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuEventElapsedTime | ( | float * | pMilliseconds, |
CUevent | hStart, | ||
CUevent | hEnd | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuFuncGetAttribute | ( | int * | pi, |
CUfunction_attribute | attrib, | ||
CUfunction | hfunc | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuFuncSetBlockShape | ( | CUfunction | hfunc, |
int | x, | ||
int | y, | ||
int | z | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuFuncSetSharedSize | ( | CUfunction | hfunc, |
unsigned int | bytes | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGetExportTable | ( | const void ** | ppExportTable, |
const CUuuid * | pExportTableId | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGLCtxCreate | ( | CUcontext * | pCtx, |
unsigned int | Flags, | ||
CUdevice | device | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGLInit | ( | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGLSetBufferObjectMapFlags | ( | GLuint | buffer, |
unsigned int | flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGraphicsGLRegisterBuffer | ( | CUgraphicsResource * | pCudaResource, |
unsigned int | buffer, | ||
unsigned int | Flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGraphicsGLRegisterImage | ( | CUgraphicsResource * | pCudaResource, |
unsigned int | image, | ||
int | target, | ||
unsigned int | Flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGraphicsMapResources | ( | unsigned int | count, |
CUgraphicsResource * | resources, | ||
CUstream | hStream | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGraphicsResourceGetMappedPointer | ( | CUdeviceptr * | pDevPtr, |
size_t * | pSize, | ||
CUgraphicsResource | resource | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGraphicsResourceSetMapFlags | ( | CUgraphicsResource | resource, |
unsigned int | flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGraphicsSubResourceGetMappedArray | ( | CUarray * | pArray, |
CUgraphicsResource | resource, | ||
unsigned int | arrayIndex, | ||
unsigned int | mipLevel | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGraphicsUnmapResources | ( | unsigned int | count, |
CUgraphicsResource * | resources, | ||
CUstream | hStream | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuGraphicsUnregisterResource | ( | CUgraphicsResource | resource | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuInit | ( | unsigned int | Flags | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuLaunch | ( | CUfunction | f | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuLaunchGrid | ( | CUfunction | f, |
int | grid_width, | ||
int | grid_height | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuLaunchGridAsync | ( | CUfunction | f, |
int | grid_width, | ||
int | grid_height, | ||
CUstream | hStream | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemAlloc | ( | CUdeviceptr * | dptr, |
unsigned int | bytesize | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemAllocHost | ( | void ** | pp, |
unsigned int | bytesize | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemAllocPitch | ( | CUdeviceptr * | dptr, |
size_t * | pPitch, | ||
unsigned int | WidthInBytes, | ||
unsigned int | Height, | ||
unsigned int | ElementSizeBytes | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpy2D | ( | const CUDA_MEMCPY2D * | pCopy | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpy2DAsync | ( | const CUDA_MEMCPY2D * | pCopy, |
CUstream | hStream | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpy2DUnaligned | ( | const CUDA_MEMCPY2D * | pCopy | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpy3D | ( | const CUDA_MEMCPY3D * | pCopy | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpy3DAsync | ( | const CUDA_MEMCPY3D * | pCopy, |
CUstream | hStream | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyAtoA | ( | CUarray | dstArray, |
unsigned int | dstIndex, | ||
CUarray | srcArray, | ||
unsigned int | srcIndex, | ||
unsigned int | ByteCount | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyAtoD | ( | CUdeviceptr | dstDevice, |
CUarray | hSrc, | ||
unsigned int | SrcIndex, | ||
unsigned int | ByteCount | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyAtoH | ( | void * | dstHost, |
CUarray | srcArray, | ||
unsigned int | srcIndex, | ||
unsigned int | ByteCount | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyAtoHAsync | ( | void * | dstHost, |
CUarray | srcArray, | ||
unsigned int | srcIndex, | ||
unsigned int | ByteCount, | ||
CUstream | hStream | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyDtoA | ( | CUarray | dstArray, |
unsigned int | dstIndex, | ||
CUdeviceptr | srcDevice, | ||
unsigned int | ByteCount | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyDtoD | ( | CUdeviceptr | dstDevice, |
CUdeviceptr | srcDevice, | ||
unsigned int | ByteCount | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyDtoH | ( | void * | dstHost, |
CUdeviceptr | srcDevice, | ||
unsigned int | ByteCount | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyDtoHAsync | ( | void * | dstHost, |
CUdeviceptr | srcDevice, | ||
unsigned int | ByteCount, | ||
CUstream | hStream | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyHtoA | ( | CUarray | dstArray, |
unsigned int | dstIndex, | ||
const void * | pSrc, | ||
unsigned int | ByteCount | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyHtoAAsync | ( | CUarray | dstArray, |
unsigned int | dstIndex, | ||
const void * | pSrc, | ||
unsigned int | ByteCount, | ||
CUstream | hStream | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyHtoD | ( | CUdeviceptr | dstDevice, |
const void * | srcHost, | ||
unsigned int | ByteCount | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemcpyHtoDAsync | ( | CUdeviceptr | dstDevice, |
const void * | srcHost, | ||
unsigned int | ByteCount, | ||
CUstream | hStream | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemFree | ( | CUdeviceptr | dptr | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemFreeHost | ( | void * | p | ) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemGetAddressRange | ( | CUdeviceptr * | pbase, |
size_t * | psize, | ||
CUdeviceptr | dptr | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemHostAlloc | ( | void ** | pp, |
unsigned long long | bytesize, | ||
unsigned int | Flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemHostGetDevicePointer | ( | CUdeviceptr * | pdptr, |
void * | p, | ||
unsigned int | Flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemHostGetFlags | ( | unsigned int * | pFlags, |
void * | p | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemsetD16 | ( | CUdeviceptr | dstDevice, |
unsigned short | us, | ||
unsigned int | N | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemsetD2D16 | ( | CUdeviceptr | dstDevice, |
unsigned int | dstPitch, | ||
unsigned short | us, | ||
unsigned int | Width, | ||
unsigned int | Height | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemsetD2D32 | ( | CUdeviceptr | dstDevice, |
unsigned int | dstPitch, | ||
unsigned int | ui, | ||
unsigned int | Width, | ||
unsigned int | Height | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemsetD2D8 | ( | CUdeviceptr | dstDevice, |
unsigned int | dstPitch, | ||
unsigned char | uc, | ||
unsigned int | Width, | ||
unsigned int | Height | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemsetD32 | ( | CUdeviceptr | dstDevice, |
unsigned int | ui, | ||
unsigned int | N | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuMemsetD8 | ( | CUdeviceptr | dstDevice, |
unsigned char | uc, | ||
unsigned int | N | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuModuleGetFunction | ( | CUfunction * | hfunc, |
CUmodule | hmod, | ||
const char * | name | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuModuleGetGlobal | ( | CUdeviceptr * | dptr, |
size_t * | bytes, | ||
CUmodule | hmod, | ||
const char * | name | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuModuleGetTexRef | ( | CUtexref * | pTexRef, |
CUmodule | hmod, | ||
const char * | name | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuModuleLoad | ( | CUmodule * | module, |
const char * | fname | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuModuleLoadData | ( | CUmodule * | module, |
const void * | image | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuModuleLoadDataEx | ( | CUmodule * | module, |
const void * | image, | ||
unsigned int | numOptions, | ||
CUjit_option * | options, | ||
void ** | optionValues | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuModuleLoadFatBinary | ( | CUmodule * | module, |
const void * | fatCubin | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuParamSetf | ( | CUfunction | hfunc, |
int | offset, | ||
float | value | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuParamSeti | ( | CUfunction | hfunc, |
int | offset, | ||
unsigned int | value | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuParamSetSize | ( | CUfunction | hfunc, |
unsigned int | numbytes | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuParamSetTexRef | ( | CUfunction | hfunc, |
int | texunit, | ||
CUtexref | hTexRef | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuParamSetv | ( | CUfunction | hfunc, |
int | offset, | ||
void * | ptr, | ||
unsigned int | numbytes | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuStreamCreate | ( | CUstream * | phStream, |
unsigned int | Flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefGetAddress | ( | CUdeviceptr * | pdptr, |
CUtexref | hTexRef | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefGetAddressMode | ( | CUaddress_mode * | pam, |
CUtexref | hTexRef, | ||
int | dim | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefGetArray | ( | CUarray * | phArray, |
CUtexref | hTexRef | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefGetFilterMode | ( | CUfilter_mode * | pfm, |
CUtexref | hTexRef | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefGetFlags | ( | unsigned int * | pFlags, |
CUtexref | hTexRef | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefGetFormat | ( | CUarray_format * | pFormat, |
int * | pNumChannels, | ||
CUtexref | hTexRef | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefSetAddress | ( | size_t * | ByteOffset, |
CUtexref | hTexRef, | ||
CUdeviceptr | dptr, | ||
size_t | bytes | ||
) |
CUresult util::KernelExtractorDriver::cuTexRefSetAddress2D | ( | CUtexref | hTexRef, |
const CUDA_ARRAY_DESCRIPTOR * | desc, | ||
CUdeviceptr | dptr, | ||
unsigned int | Pitch | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefSetAddressMode | ( | CUtexref | hTexRef, |
int | dim, | ||
CUaddress_mode | am | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefSetArray | ( | CUtexref | hTexRef, |
CUarray | hArray, | ||
unsigned int | Flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefSetFilterMode | ( | CUtexref | hTexRef, |
CUfilter_mode | fm | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefSetFlags | ( | CUtexref | hTexRef, |
unsigned int | Flags | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
CUresult util::KernelExtractorDriver::cuTexRefSetFormat | ( | CUtexref | hTexRef, |
CUarray_format | fmt, | ||
int | NumPackedComponents | ||
) | [virtual] |
Reimplemented from cuda::CudaDriverInterface.
void util::KernelExtractorDriver::free | ( | void * | dptr | ) |
deletes an allocation
void util::KernelExtractorDriver::kernelLaunch | ( | CUfunction | f, |
int | gridX = 1 , |
||
int | gridY = 1 |
||
) |
called when a kernel is launched
void util::KernelExtractorDriver::kernelReturn | ( | CUresult | result | ) |
called when a kernel returns
void util::KernelExtractorDriver::loadModule | ( | CUresult | result, |
CUmodule | module, | ||
const char * | ptxImage, | ||
const char * | name = 0 |
||
) |
binds module handle to PTX image
void util::KernelExtractorDriver::synchronizeFromDevice | ( | ) |
copies data from device to host-side allocations
void util::KernelExtractorDriver::synchronizeToDevice | ( | ) |
copies data from host-side allocations to device
copies data to host-side allocations to device
std::string util::KernelExtractorDriver::toString | ( | CUresult | result | ) |
Reimplemented from cuda::CudaDriverInterface.
determines whether kernels are actually extracted
maps function handles to (module name, function name) pairs
singleton instance
maps module handles to module names
object for serializing CUDA kernels and device state
maps texture handles to (module name, texture name) pairs