GPUOcelot
Public Member Functions

cuda::TraceGeneratingCudaRuntime Class Reference

The main high performance implementation of the CUDA API. More...

#include <TraceGeneratingCudaRuntime.h>

Inheritance diagram for cuda::TraceGeneratingCudaRuntime:
Inheritance graph
[legend]
Collaboration diagram for cuda::TraceGeneratingCudaRuntime:
Collaboration graph
[legend]

List of all members.

Public Member Functions

 TraceGeneratingCudaRuntime ()
 Create the trace file.
 ~TraceGeneratingCudaRuntime ()
 Destroy the trace file.
cudaError_t cudaMalloc3D (cudaPitchedPtr *pitchedDevPtr, cudaExtent extent)
cudaError_t cudaMalloc3DArray (cudaArray **arrayPtr, const cudaChannelFormatDesc *desc, cudaExtent extent)
cudaError_t cudaMemset3D (cudaPitchedPtr pitchedDevPtr, int value, cudaExtent extent)
cudaError_t cudaMemcpy3D (const cudaMemcpy3DParms *p)
cudaError_t cudaMemcpy3DAsync (const cudaMemcpy3DParms *p, cudaStream_t stream)
cudaError_t cudaMalloc (void **devPtr, size_t size)
cudaError_t cudaMallocHost (void **ptr, size_t size)
cudaError_t cudaMallocPitch (void **devPtr, size_t *pitch, size_t width, size_t height)
cudaError_t cudaMallocArray (cudaArray **array, const cudaChannelFormatDesc *desc, size_t width, size_t height=1)
cudaError_t cudaFree (void *devPtr)
cudaError_t cudaFreeHost (void *ptr)
cudaError_t cudaFreeArray (cudaArray *array)
cudaError_t cudaHostAlloc (void **pHost, size_t bytes, unsigned int flags)
cudaError_t cudaHostGetDevicePointer (void **pDevice, void *pHost, unsigned int flags)
cudaError_t cudaMemcpy (void *dst, const void *src, size_t count, cudaMemcpyKind kind)
cudaError_t cudaMemcpyToArray (cudaArray *dst, size_t wOffset, size_t hOffset, const void *src, size_t count, cudaMemcpyKind kind)
cudaError_t cudaMemcpyFromArray (void *dst, const cudaArray *src, size_t wOffset, size_t hOffset, size_t count, cudaMemcpyKind kind)
cudaError_t cudaMemcpyArrayToArray (cudaArray *dst, size_t wOffsetDst, size_t hOffsetDst, const cudaArray *src, size_t wOffsetSrc, size_t hOffsetSrc, size_t count, cudaMemcpyKind kind=cudaMemcpyDeviceToDevice)
cudaError_t cudaMemcpy2D (void *dst, size_t dpitch, const void *src, size_t spitch, size_t width, size_t height, cudaMemcpyKind kind)
cudaError_t cudaMemcpy2DToArray (cudaArray *dst, size_t wOffset, size_t hOffset, const void *src, size_t spitch, size_t width, size_t height, cudaMemcpyKind kind)
cudaError_t cudaMemcpy2DFromArray (void *dst, size_t dpitch, const cudaArray *src, size_t wOffset, size_t hOffset, size_t width, size_t height, cudaMemcpyKind kind)
cudaError_t cudaMemcpy2DArrayToArray (cudaArray *dst, size_t wOffsetDst, size_t hOffsetDst, const cudaArray *src, size_t wOffsetSrc, size_t hOffsetSrc, size_t width, size_t height, cudaMemcpyKind kind=cudaMemcpyDeviceToDevice)
cudaError_t cudaMemcpyToSymbol (const char *symbol, const void *src, size_t count, size_t offset, cudaMemcpyKind kind=cudaMemcpyHostToDevice)
cudaError_t cudaMemcpyFromSymbol (void *dst, const char *symbol, size_t count, size_t offset, cudaMemcpyKind kind=cudaMemcpyDeviceToHost)
cudaError_t cudaMemcpyAsync (void *dst, const void *src, size_t count, cudaMemcpyKind kind, cudaStream_t stream)
cudaError_t cudaMemcpyToArrayAsync (cudaArray *dst, size_t wOffset, size_t hOffset, const void *src, size_t count, cudaMemcpyKind kind, cudaStream_t stream)
cudaError_t cudaMemcpyFromArrayAsync (void *dst, const cudaArray *src, size_t wOffset, size_t hOffset, size_t count, cudaMemcpyKind kind, cudaStream_t stream)
cudaError_t cudaMemcpy2DAsync (void *dst, size_t dpitch, const void *src, size_t spitch, size_t width, size_t height, cudaMemcpyKind kind, cudaStream_t stream)
cudaError_t cudaMemcpy2DToArrayAsync (cudaArray *dst, size_t wOffset, size_t hOffset, const void *src, size_t spitch, size_t width, size_t height, cudaMemcpyKind kind, cudaStream_t stream)
cudaError_t cudaMemcpy2DFromArrayAsync (void *dst, size_t dpitch, const cudaArray *src, size_t wOffset, size_t hOffset, size_t width, size_t height, cudaMemcpyKind kind, cudaStream_t stream)
cudaError_t cudaMemcpyToSymbolAsync (const char *symbol, const void *src, size_t count, size_t offset, cudaMemcpyKind kind, cudaStream_t stream)
cudaError_t cudaMemcpyFromSymbolAsync (void *dst, const char *symbol, size_t count, size_t offset, cudaMemcpyKind kind, cudaStream_t stream)
cudaError_t cudaMemset (void *devPtr, int value, size_t count)
cudaError_t cudaMemset2D (void *devPtr, size_t pitch, int value, size_t width, size_t height)
cudaError_t cudaGetSymbolAddress (void **devPtr, const char *symbol)
cudaError_t cudaGetSymbolSize (size_t *size, const char *symbol)
cudaError_t cudaGetDeviceCount (int *count)
cudaError_t cudaGetDeviceProperties (cudaDeviceProp *prop, int device)
cudaError_t cudaChooseDevice (int *device, const cudaDeviceProp *prop)
cudaError_t cudaSetDevice (int device)
cudaError_t cudaGetDevice (int *device)
cudaError_t cudaSetValidDevices (int *device_arr, int len)
cudaError_t cudaSetDeviceFlags (int flags)
cudaError_t cudaBindTexture (size_t *offset, const textureReference *texref, const void *devPtr, const cudaChannelFormatDesc *desc, size_t size=UINT_MAX)
cudaError_t cudaBindTexture2D (size_t *offset, const textureReference *texref, const void *devPtr, const cudaChannelFormatDesc *desc, size_t width, size_t height, size_t pitch)
cudaError_t cudaBindTextureToArray (const textureReference *texref, const cudaArray *array, const cudaChannelFormatDesc *desc)
cudaError_t cudaUnbindTexture (const textureReference *texref)
cudaError_t cudaGetTextureAlignmentOffset (size_t *offset, const textureReference *texref)
cudaError_t cudaGetTextureReference (const textureReference **texref, const char *symbol)
cudaError_t cudaGetChannelDesc (cudaChannelFormatDesc *desc, const cudaArray *array)
cudaChannelFormatDesc cudaCreateChannelDesc (int x, int y, int z, int w, cudaChannelFormatKind f)
cudaError_t cudaGetLastError (void)
const char * cudaGetErrorString (cudaError_t error)
cudaError_t cudaConfigureCall (dim3 gridDim, dim3 blockDim, size_t sharedMem, cudaStream_t stream)
cudaError_t cudaSetupArgument (const void *arg, size_t size, size_t offset)
cudaError_t cudaLaunch (const char *entry)
cudaError_t cudaFuncGetAttributes (cudaFuncAttributes *attr, const char *func)
cudaError_t cudaStreamCreate (cudaStream_t *pStream)
cudaError_t cudaStreamDestroy (cudaStream_t stream)
cudaError_t cudaStreamSynchronize (cudaStream_t stream)
cudaError_t cudaStreamQuery (cudaStream_t stream)
cudaError_t cudaEventCreate (cudaEvent_t *event)
cudaError_t cudaEventCreateWithFlags (cudaEvent_t *event, int flags)
cudaError_t cudaEventRecord (cudaEvent_t event, cudaStream_t stream)
cudaError_t cudaEventQuery (cudaEvent_t event)
cudaError_t cudaEventSynchronize (cudaEvent_t event)
cudaError_t cudaEventDestroy (cudaEvent_t event)
cudaError_t cudaEventElapsedTime (float *ms, cudaEvent_t start, cudaEvent_t end)
cudaError_t cudaSetDoubleForDevice (double *d)
cudaError_t cudaSetDoubleForHost (double *d)
cudaError_t cudaThreadExit (void)
cudaError_t cudaThreadSynchronize (void)
cudaError_t cudaDriverGetVersion (int *driverVersion)
cudaError_t cudaRuntimeGetVersion (int *runtimeVersion)
cudaError_t cudaGLSetGLDevice (int device)
cudaError_t cudaGLRegisterBufferObject (GLuint bufObj)
cudaError_t cudaGLMapBufferObject (void **devPtr, GLuint bufObj)
cudaError_t cudaGLUnmapBufferObject (GLuint bufObj)
cudaError_t cudaGLUnregisterBufferObject (GLuint bufObj)
void cudaTextureFetch (const void *tex, void *index, int integer, void *val)
void ** cudaRegisterFatBinary (void *fatCubin)
void cudaUnregisterFatBinary (void **fatCubinHandle)
void cudaRegisterVar (void **fatCubinHandle, char *hostVar, char *deviceAddress, const char *deviceName, int ext, int size, int constant, int global)
void cudaRegisterTexture (void **fatCubinHandle, const struct textureReference *hostVar, const void **deviceAddress, const char *deviceName, int dim, int norm, int ext)
void cudaRegisterShared (void **fatCubinHandle, void **devicePtr)
void cudaRegisterSharedVar (void **fatCubinHandle, void **devicePtr, size_t size, size_t alignment, int storage)
void cudaRegisterFunction (void **fatCubinHandle, const char *hostFun, char *deviceFun, const char *deviceName, int thread_limit, uint3 *tid, uint3 *bid, dim3 *bDim, dim3 *gDim, int *wSize)
void cudaMutexOperation (int lock)
int cudaSynchronizeThreads (void **, void *)
void addTraceGenerator (trace::TraceGenerator &gen, bool persistent, bool safe)
void clearTraceGenerators (bool safe)
void limitWorkerThreads (unsigned int limit)
 Sets a limit on the number of host worker threads to launch when executing a CUDA kernel on a Multi-Core CPU.
void registerPTXModule (std::istream &stream, const std::string &name)
 Register an istream containing a PTX module.
const char * getKernelPointer (const std::string &name, const std::string &module)
void ** getFatBinaryHandle (const std::string &name)
void clearErrors ()
 Clear all errors in the Cuda Runtime.
void configure (const Configuration &c)

Detailed Description

The main high performance implementation of the CUDA API.


Constructor & Destructor Documentation

cuda::TraceGeneratingCudaRuntime::TraceGeneratingCudaRuntime ( )

Create the trace file.

cuda::TraceGeneratingCudaRuntime::~TraceGeneratingCudaRuntime ( )

Destroy the trace file.


Member Function Documentation

void cuda::TraceGeneratingCudaRuntime::addTraceGenerator ( trace::TraceGenerator gen,
bool  persistent,
bool  safe 
)
void cuda::TraceGeneratingCudaRuntime::clearErrors ( ) [virtual]

Clear all errors in the Cuda Runtime.

Reimplemented from cuda::CudaRuntimeInterface.

void cuda::TraceGeneratingCudaRuntime::clearTraceGenerators ( bool  safe)
void cuda::TraceGeneratingCudaRuntime::configure ( const Configuration &  c)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaBindTexture ( size_t offset,
const textureReference texref,
const void *  devPtr,
const cudaChannelFormatDesc desc,
size_t  size = UINT_MAX 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaBindTexture2D ( size_t offset,
const textureReference texref,
const void *  devPtr,
const cudaChannelFormatDesc desc,
size_t  width,
size_t  height,
size_t  pitch 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaBindTextureToArray ( const textureReference texref,
const cudaArray *  array,
const cudaChannelFormatDesc desc 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaChooseDevice ( int *  device,
const cudaDeviceProp prop 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaConfigureCall ( dim3  gridDim,
dim3  blockDim,
size_t  sharedMem,
cudaStream_t  stream 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaChannelFormatDesc cuda::TraceGeneratingCudaRuntime::cudaCreateChannelDesc ( int  x,
int  y,
int  z,
int  w,
cudaChannelFormatKind  f 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaDriverGetVersion ( int *  driverVersion) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaEventCreate ( cudaEvent_t event) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaEventCreateWithFlags ( cudaEvent_t event,
int  flags 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaEventDestroy ( cudaEvent_t  event) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaEventElapsedTime ( float *  ms,
cudaEvent_t  start,
cudaEvent_t  end 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaEventQuery ( cudaEvent_t  event) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaEventRecord ( cudaEvent_t  event,
cudaStream_t  stream 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaEventSynchronize ( cudaEvent_t  event) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaFree ( void *  devPtr) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaFreeArray ( cudaArray *  array) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaFreeHost ( void *  ptr) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaFuncGetAttributes ( cudaFuncAttributes attr,
const char *  func 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGetChannelDesc ( cudaChannelFormatDesc desc,
const cudaArray *  array 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGetDevice ( int *  device) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGetDeviceCount ( int *  count) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGetDeviceProperties ( cudaDeviceProp prop,
int  device 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

const char* cuda::TraceGeneratingCudaRuntime::cudaGetErrorString ( cudaError_t  error)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGetLastError ( void  ) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGetSymbolAddress ( void **  devPtr,
const char *  symbol 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGetSymbolSize ( size_t size,
const char *  symbol 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGetTextureAlignmentOffset ( size_t offset,
const textureReference texref 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGetTextureReference ( const textureReference **  texref,
const char *  symbol 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGLMapBufferObject ( void **  devPtr,
GLuint  bufObj 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGLRegisterBufferObject ( GLuint  bufObj) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGLSetGLDevice ( int  device) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGLUnmapBufferObject ( GLuint  bufObj) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaGLUnregisterBufferObject ( GLuint  bufObj) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaHostAlloc ( void **  pHost,
size_t  bytes,
unsigned int  flags 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaHostGetDevicePointer ( void **  pDevice,
void *  pHost,
unsigned int  flags 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaLaunch ( const char *  entry) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMalloc ( void **  devPtr,
size_t  size 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMalloc3D ( cudaPitchedPtr pitchedDevPtr,
cudaExtent  extent 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMalloc3DArray ( cudaArray **  arrayPtr,
const cudaChannelFormatDesc desc,
cudaExtent  extent 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMallocArray ( cudaArray **  array,
const cudaChannelFormatDesc desc,
size_t  width,
size_t  height = 1 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMallocHost ( void **  ptr,
size_t  size 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMallocPitch ( void **  devPtr,
size_t pitch,
size_t  width,
size_t  height 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy ( void *  dst,
const void *  src,
size_t  count,
cudaMemcpyKind  kind 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy2D ( void *  dst,
size_t  dpitch,
const void *  src,
size_t  spitch,
size_t  width,
size_t  height,
cudaMemcpyKind  kind 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy2DArrayToArray ( cudaArray *  dst,
size_t  wOffsetDst,
size_t  hOffsetDst,
const cudaArray *  src,
size_t  wOffsetSrc,
size_t  hOffsetSrc,
size_t  width,
size_t  height,
cudaMemcpyKind  kind = cudaMemcpyDeviceToDevice 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy2DAsync ( void *  dst,
size_t  dpitch,
const void *  src,
size_t  spitch,
size_t  width,
size_t  height,
cudaMemcpyKind  kind,
cudaStream_t  stream 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy2DFromArray ( void *  dst,
size_t  dpitch,
const cudaArray *  src,
size_t  wOffset,
size_t  hOffset,
size_t  width,
size_t  height,
cudaMemcpyKind  kind 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy2DFromArrayAsync ( void *  dst,
size_t  dpitch,
const cudaArray *  src,
size_t  wOffset,
size_t  hOffset,
size_t  width,
size_t  height,
cudaMemcpyKind  kind,
cudaStream_t  stream 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy2DToArray ( cudaArray *  dst,
size_t  wOffset,
size_t  hOffset,
const void *  src,
size_t  spitch,
size_t  width,
size_t  height,
cudaMemcpyKind  kind 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy2DToArrayAsync ( cudaArray *  dst,
size_t  wOffset,
size_t  hOffset,
const void *  src,
size_t  spitch,
size_t  width,
size_t  height,
cudaMemcpyKind  kind,
cudaStream_t  stream 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy3D ( const cudaMemcpy3DParms p)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpy3DAsync ( const cudaMemcpy3DParms p,
cudaStream_t  stream 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyArrayToArray ( cudaArray *  dst,
size_t  wOffsetDst,
size_t  hOffsetDst,
const cudaArray *  src,
size_t  wOffsetSrc,
size_t  hOffsetSrc,
size_t  count,
cudaMemcpyKind  kind = cudaMemcpyDeviceToDevice 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyAsync ( void *  dst,
const void *  src,
size_t  count,
cudaMemcpyKind  kind,
cudaStream_t  stream 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyFromArray ( void *  dst,
const cudaArray *  src,
size_t  wOffset,
size_t  hOffset,
size_t  count,
cudaMemcpyKind  kind 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyFromArrayAsync ( void *  dst,
const cudaArray *  src,
size_t  wOffset,
size_t  hOffset,
size_t  count,
cudaMemcpyKind  kind,
cudaStream_t  stream 
)
cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyFromSymbol ( void *  dst,
const char *  symbol,
size_t  count,
size_t  offset,
cudaMemcpyKind  kind = cudaMemcpyDeviceToHost 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyFromSymbolAsync ( void *  dst,
const char *  symbol,
size_t  count,
size_t  offset,
cudaMemcpyKind  kind,
cudaStream_t  stream 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyToArray ( cudaArray *  dst,
size_t  wOffset,
size_t  hOffset,
const void *  src,
size_t  count,
cudaMemcpyKind  kind 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyToArrayAsync ( cudaArray *  dst,
size_t  wOffset,
size_t  hOffset,
const void *  src,
size_t  count,
cudaMemcpyKind  kind,
cudaStream_t  stream 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyToSymbol ( const char *  symbol,
const void *  src,
size_t  count,
size_t  offset,
cudaMemcpyKind  kind = cudaMemcpyHostToDevice 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemcpyToSymbolAsync ( const char *  symbol,
const void *  src,
size_t  count,
size_t  offset,
cudaMemcpyKind  kind,
cudaStream_t  stream 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemset ( void *  devPtr,
int  value,
size_t  count 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemset2D ( void *  devPtr,
size_t  pitch,
int  value,
size_t  width,
size_t  height 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaMemset3D ( cudaPitchedPtr  pitchedDevPtr,
int  value,
cudaExtent  extent 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

void cuda::TraceGeneratingCudaRuntime::cudaMutexOperation ( int  lock) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

void** cuda::TraceGeneratingCudaRuntime::cudaRegisterFatBinary ( void *  fatCubin) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

void cuda::TraceGeneratingCudaRuntime::cudaRegisterFunction ( void **  fatCubinHandle,
const char *  hostFun,
char *  deviceFun,
const char *  deviceName,
int  thread_limit,
uint3 tid,
uint3 bid,
dim3 bDim,
dim3 gDim,
int *  wSize 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

void cuda::TraceGeneratingCudaRuntime::cudaRegisterShared ( void **  fatCubinHandle,
void **  devicePtr 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

void cuda::TraceGeneratingCudaRuntime::cudaRegisterSharedVar ( void **  fatCubinHandle,
void **  devicePtr,
size_t  size,
size_t  alignment,
int  storage 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

void cuda::TraceGeneratingCudaRuntime::cudaRegisterTexture ( void **  fatCubinHandle,
const struct textureReference hostVar,
const void **  deviceAddress,
const char *  deviceName,
int  dim,
int  norm,
int  ext 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

void cuda::TraceGeneratingCudaRuntime::cudaRegisterVar ( void **  fatCubinHandle,
char *  hostVar,
char *  deviceAddress,
const char *  deviceName,
int  ext,
int  size,
int  constant,
int  global 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaRuntimeGetVersion ( int *  runtimeVersion) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaSetDevice ( int  device) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaSetDeviceFlags ( int  flags) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaSetDoubleForDevice ( double *  d) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaSetDoubleForHost ( double *  d) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaSetupArgument ( const void *  arg,
size_t  size,
size_t  offset 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaSetValidDevices ( int *  device_arr,
int  len 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaStreamCreate ( cudaStream_t pStream) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaStreamDestroy ( cudaStream_t  stream) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaStreamQuery ( cudaStream_t  stream) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaStreamSynchronize ( cudaStream_t  stream) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

int cuda::TraceGeneratingCudaRuntime::cudaSynchronizeThreads ( void **  ,
void *   
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

void cuda::TraceGeneratingCudaRuntime::cudaTextureFetch ( const void *  tex,
void *  index,
int  integer,
void *  val 
) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaThreadExit ( void  ) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaThreadSynchronize ( void  ) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

cudaError_t cuda::TraceGeneratingCudaRuntime::cudaUnbindTexture ( const textureReference texref)
void cuda::TraceGeneratingCudaRuntime::cudaUnregisterFatBinary ( void **  fatCubinHandle) [virtual]

Reimplemented from cuda::CudaRuntimeInterface.

void** cuda::TraceGeneratingCudaRuntime::getFatBinaryHandle ( const std::string &  name)
const char* cuda::TraceGeneratingCudaRuntime::getKernelPointer ( const std::string &  name,
const std::string &  module 
)
void cuda::TraceGeneratingCudaRuntime::limitWorkerThreads ( unsigned int  limit) [virtual]

Sets a limit on the number of host worker threads to launch when executing a CUDA kernel on a Multi-Core CPU.

Parameters:
limitThe max number of worker threads to launch per kernel.

Reimplemented from cuda::CudaRuntimeInterface.

void cuda::TraceGeneratingCudaRuntime::registerPTXModule ( std::istream &  stream,
const std::string &  name 
) [virtual]

Register an istream containing a PTX module.

Parameters:
streamAn input stream containing a PTX module
Thename of the module being registered. Must be Unique.

Reimplemented from cuda::CudaRuntimeInterface.


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