Package | Description |
---|---|
org.bytedeco.cuda.cudart | |
org.bytedeco.cuda.global |
Modifier and Type | Method and Description |
---|---|
CUfunc_st |
CUDA_KERNEL_NODE_PARAMS_v2.func()
Kernel to launch
|
CUfunc_st |
CUDA_KERNEL_NODE_PARAMS_v1.func()
Kernel to launch
|
CUfunc_st |
CUDA_KERNEL_NODE_PARAMS_v3.func()
Kernel to launch
|
CUfunc_st |
CUDA_LAUNCH_PARAMS_v1.function()
Kernel to launch
|
Modifier and Type | Method and Description |
---|---|
CUDA_KERNEL_NODE_PARAMS_v2 |
CUDA_KERNEL_NODE_PARAMS_v2.func(CUfunc_st setter) |
CUDA_KERNEL_NODE_PARAMS_v1 |
CUDA_KERNEL_NODE_PARAMS_v1.func(CUfunc_st setter) |
CUDA_KERNEL_NODE_PARAMS_v3 |
CUDA_KERNEL_NODE_PARAMS_v3.func(CUfunc_st setter) |
CUDA_LAUNCH_PARAMS_v1 |
CUDA_LAUNCH_PARAMS_v1.function(CUfunc_st setter) |
Modifier and Type | Method and Description |
---|---|
static int |
cudart.cudaGetFuncBySymbol(CUfunc_st functionPtr,
Pointer symbolPtr)
\brief Get pointer to device entry function that matches entry function \p symbolPtr
Returns in \p functionPtr the device entry function corresponding to the symbol \p symbolPtr.
|
static int |
cudart.cuFuncGetAttribute(int[] pi,
int attrib,
CUfunc_st hfunc) |
static int |
cudart.cuFuncGetAttribute(IntBuffer pi,
int attrib,
CUfunc_st hfunc) |
static int |
cudart.cuFuncGetAttribute(IntPointer pi,
int attrib,
CUfunc_st hfunc)
\brief Returns information about a function
Returns in \p *pi the integer value of the attribute \p attrib on the kernel
given by \p hfunc.
|
static int |
cudart.cuFuncGetModule(CUmod_st hmod,
CUfunc_st hfunc)
\brief Returns a module handle
Returns in \p *hmod the handle of the module that function \p hfunc
is located in.
|
static int |
cudart.cuFuncGetName(byte[] name,
CUfunc_st hfunc) |
static int |
cudart.cuFuncGetName(ByteBuffer name,
CUfunc_st hfunc) |
static int |
cudart.cuFuncGetName(BytePointer name,
CUfunc_st hfunc) |
static int |
cudart.cuFuncGetName(PointerPointer name,
CUfunc_st hfunc)
\brief Returns the function name for a ::CUfunction handle
Returns in \p **name the function name associated with the function handle \p hfunc .
|
static int |
cudart.cuFuncSetAttribute(CUfunc_st hfunc,
int attrib,
int value)
\brief Sets information about a function
This call sets the value of a specified attribute \p attrib on the kernel given
by \p hfunc to an integer value specified by \p val
This function returns CUDA_SUCCESS if the new value of the attribute could be
successfully set.
|
static int |
cudart.cuFuncSetBlockShape(CUfunc_st hfunc,
int x,
int y,
int z)
Deprecated.
Specifies the \p x, \p y, and \p z dimensions of the thread blocks that are
created when the kernel given by \p hfunc is launched.
|
static int |
cudart.cuFuncSetCacheConfig(CUfunc_st hfunc,
int config)
\brief Sets the preferred cache configuration for a device function
On devices where the L1 cache and shared memory use the same hardware
resources, this sets through \p config the preferred cache configuration for
the device function \p hfunc.
|
static int |
cudart.cuFuncSetSharedMemConfig(CUfunc_st hfunc,
int config)
\brief Sets the shared memory configuration for a device function.
|
static int |
cudart.cuFuncSetSharedSize(CUfunc_st hfunc,
int bytes)
Deprecated.
Sets through \p bytes the amount of dynamic shared memory that will be
available to each thread block when the kernel given by \p hfunc is launched.
|
static int |
cudart.cuKernelGetFunction(CUfunc_st pFunc,
CUkern_st kernel)
\brief Returns a function handle
Returns in \p pFunc the handle of the function for the requested kernel \p kernel and
the current context.
|
static int |
cudart.cuLaunch(CUfunc_st f)
Deprecated.
Invokes the kernel \p f on a 1 x 1 x 1 grid of blocks. The block
contains the number of threads specified by a previous call to
::cuFuncSetBlockShape().
The block shape, dynamic shared memory size, and parameter information
must be set using
::cuFuncSetBlockShape(),
::cuFuncSetSharedSize(),
::cuParamSetSize(),
::cuParamSeti(),
::cuParamSetf(), and
::cuParamSetv()
prior to calling this function.
Launching a function via ::cuLaunchKernel() invalidates the function's
block shape, dynamic shared memory size, and parameter information. After
launching via cuLaunchKernel, this state must be re-initialized prior to
calling this function. Failure to do so results in undefined behavior.
|
static int |
cudart.cuLaunchCooperativeKernel(CUfunc_st f,
int gridDimX,
int gridDimY,
int gridDimZ,
int blockDimX,
int blockDimY,
int blockDimZ,
int sharedMemBytes,
CUstream_st hStream,
Pointer kernelParams) |
static int |
cudart.cuLaunchCooperativeKernel(CUfunc_st f,
int gridDimX,
int gridDimY,
int gridDimZ,
int blockDimX,
int blockDimY,
int blockDimZ,
int sharedMemBytes,
CUstream_st hStream,
PointerPointer kernelParams)
\brief Launches a CUDA function ::CUfunction or a CUDA kernel ::CUkernel where thread blocks
can cooperate and synchronize as they execute
Invokes the function ::CUfunction or the kernel ::CUkernel \p f on a \p gridDimX x \p gridDimY x \p gridDimZ
grid of blocks.
|
static int |
cudart.cuLaunchGrid(CUfunc_st f,
int grid_width,
int grid_height)
Deprecated.
Invokes the kernel \p f on a \p grid_width x \p grid_height grid of
blocks. Each block contains the number of threads specified by a previous
call to ::cuFuncSetBlockShape().
The block shape, dynamic shared memory size, and parameter information
must be set using
::cuFuncSetBlockShape(),
::cuFuncSetSharedSize(),
::cuParamSetSize(),
::cuParamSeti(),
::cuParamSetf(), and
::cuParamSetv()
prior to calling this function.
Launching a function via ::cuLaunchKernel() invalidates the function's
block shape, dynamic shared memory size, and parameter information. After
launching via cuLaunchKernel, this state must be re-initialized prior to
calling this function. Failure to do so results in undefined behavior.
|
static int |
cudart.cuLaunchGridAsync(CUfunc_st f,
int grid_width,
int grid_height,
CUstream_st hStream)
Deprecated.
Invokes the kernel \p f on a \p grid_width x \p grid_height grid of
blocks. Each block contains the number of threads specified by a previous
call to ::cuFuncSetBlockShape().
The block shape, dynamic shared memory size, and parameter information
must be set using
::cuFuncSetBlockShape(),
::cuFuncSetSharedSize(),
::cuParamSetSize(),
::cuParamSeti(),
::cuParamSetf(), and
::cuParamSetv()
prior to calling this function.
Launching a function via ::cuLaunchKernel() invalidates the function's
block shape, dynamic shared memory size, and parameter information. After
launching via cuLaunchKernel, this state must be re-initialized prior to
calling this function. Failure to do so results in undefined behavior.
|
static int |
cudart.cuLaunchKernel(CUfunc_st f,
int gridDimX,
int gridDimY,
int gridDimZ,
int blockDimX,
int blockDimY,
int blockDimZ,
int sharedMemBytes,
CUstream_st hStream,
Pointer kernelParams,
Pointer extra) |
static int |
cudart.cuLaunchKernel(CUfunc_st f,
int gridDimX,
int gridDimY,
int gridDimZ,
int blockDimX,
int blockDimY,
int blockDimZ,
int sharedMemBytes,
CUstream_st hStream,
PointerPointer kernelParams,
PointerPointer extra)
\brief Launches a CUDA function ::CUfunction or a CUDA kernel ::CUkernel
Invokes the function ::CUfunction or the kernel ::CUkernel \p f
on a \p gridDimX x \p gridDimY x \p gridDimZ grid of blocks.
|
static int |
cudart.cuLaunchKernelEx(CUlaunchConfig config,
CUfunc_st f,
Pointer kernelParams,
Pointer extra) |
static int |
cudart.cuLaunchKernelEx(CUlaunchConfig config,
CUfunc_st f,
PointerPointer kernelParams,
PointerPointer extra)
\brief Launches a CUDA function ::CUfunction or a CUDA kernel ::CUkernel with launch-time configuration
Invokes the function ::CUfunction or the kernel ::CUkernel \p f with the specified launch-time configuration
\p config.
|
static int |
cudart.cuModuleGetFunction(CUfunc_st hfunc,
CUmod_st hmod,
BytePointer name)
\brief Returns a function handle
Returns in \p *hfunc the handle of the function of name \p name located in
module \p hmod.
|
static int |
cudart.cuModuleGetFunction(CUfunc_st hfunc,
CUmod_st hmod,
String name) |
static int |
cudart.cuOccupancyAvailableDynamicSMemPerBlock(SizeTPointer dynamicSmemSize,
CUfunc_st func,
int numBlocks,
int blockSize)
\brief Returns dynamic shared memory available per block when launching \p numBlocks blocks on SM
Returns in \p *dynamicSmemSize the maximum size of dynamic shared memory to allow \p numBlocks blocks per SM.
|
static int |
cudart.cuOccupancyMaxActiveBlocksPerMultiprocessor(int[] numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize) |
static int |
cudart.cuOccupancyMaxActiveBlocksPerMultiprocessor(IntBuffer numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize) |
static int |
cudart.cuOccupancyMaxActiveBlocksPerMultiprocessor(IntPointer numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize)
\brief Returns occupancy of a function
Returns in \p *numBlocks the number of the maximum active blocks per
streaming multiprocessor.
|
static int |
cudart.cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(int[] numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize,
int flags) |
static int |
cudart.cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntBuffer numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize,
int flags) |
static int |
cudart.cuOccupancyMaxActiveBlocksPerMultiprocessorWithFlags(IntPointer numBlocks,
CUfunc_st func,
int blockSize,
long dynamicSMemSize,
int flags)
\brief Returns occupancy of a function
Returns in \p *numBlocks the number of the maximum active blocks per
streaming multiprocessor.
|
static int |
cudart.cuOccupancyMaxActiveClusters(int[] numClusters,
CUfunc_st func,
CUlaunchConfig config) |
static int |
cudart.cuOccupancyMaxActiveClusters(IntBuffer numClusters,
CUfunc_st func,
CUlaunchConfig config) |
static int |
cudart.cuOccupancyMaxActiveClusters(IntPointer numClusters,
CUfunc_st func,
CUlaunchConfig config)
\brief Given the kernel function (\p func) and launch configuration
(\p config), return the maximum number of clusters that could co-exist
on the target device in \p *numClusters.
|
static int |
cudart.cuOccupancyMaxPotentialBlockSize(int[] minGridSize,
int[] blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit) |
static int |
cudart.cuOccupancyMaxPotentialBlockSize(IntBuffer minGridSize,
IntBuffer blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit) |
static int |
cudart.cuOccupancyMaxPotentialBlockSize(IntPointer minGridSize,
IntPointer blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit)
\brief Suggest a launch configuration with reasonable occupancy
Returns in \p *blockSize a reasonable block size that can achieve
the maximum occupancy (or, the maximum number of active warps with
the fewest blocks per multiprocessor), and in \p *minGridSize the
minimum grid size to achieve the maximum occupancy.
|
static int |
cudart.cuOccupancyMaxPotentialBlockSizeWithFlags(int[] minGridSize,
int[] blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit,
int flags) |
static int |
cudart.cuOccupancyMaxPotentialBlockSizeWithFlags(IntBuffer minGridSize,
IntBuffer blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit,
int flags) |
static int |
cudart.cuOccupancyMaxPotentialBlockSizeWithFlags(IntPointer minGridSize,
IntPointer blockSize,
CUfunc_st func,
CUoccupancyB2DSize blockSizeToDynamicSMemSize,
long dynamicSMemSize,
int blockSizeLimit,
int flags)
\brief Suggest a launch configuration with reasonable occupancy
An extended version of ::cuOccupancyMaxPotentialBlockSize.
|
static int |
cudart.cuOccupancyMaxPotentialClusterSize(int[] clusterSize,
CUfunc_st func,
CUlaunchConfig config) |
static int |
cudart.cuOccupancyMaxPotentialClusterSize(IntBuffer clusterSize,
CUfunc_st func,
CUlaunchConfig config) |
static int |
cudart.cuOccupancyMaxPotentialClusterSize(IntPointer clusterSize,
CUfunc_st func,
CUlaunchConfig config)
\brief Given the kernel function (\p func) and launch configuration
(\p config), return the maximum cluster size in \p *clusterSize.
|
static int |
cudart.cuParamSetf(CUfunc_st hfunc,
int offset,
float value)
Deprecated.
Sets a floating-point parameter that will be specified the next time the
kernel corresponding to \p hfunc will be invoked. \p offset is a byte offset.
|
static int |
cudart.cuParamSeti(CUfunc_st hfunc,
int offset,
int value)
Deprecated.
Sets an integer parameter that will be specified the next time the
kernel corresponding to \p hfunc will be invoked. \p offset is a byte offset.
|
static int |
cudart.cuParamSetSize(CUfunc_st hfunc,
int numbytes)
Deprecated.
Sets through \p numbytes the total size in bytes needed by the function
parameters of the kernel corresponding to \p hfunc.
|
static int |
cudart.cuParamSetTexRef(CUfunc_st hfunc,
int texunit,
CUtexref_st hTexRef)
Deprecated.
Makes the CUDA array or linear memory bound to the texture reference
\p hTexRef available to a device program as a texture. In this version of
CUDA, the texture-reference must be obtained via ::cuModuleGetTexRef() and
the \p texunit parameter must be set to ::CU_PARAM_TR_DEFAULT.
|
static int |
cudart.cuParamSetv(CUfunc_st hfunc,
int offset,
Pointer ptr,
int numbytes)
Deprecated.
Copies an arbitrary amount of data (specified in \p numbytes) from \p ptr
into the parameter space of the kernel corresponding to \p hfunc. \p offset
is a byte offset.
|
Copyright © 2024. All rights reserved.