![]() Generated automatically by Doxygen from the source code. Offset - Offset in argument stack to push new argĬudaLaunch (C API), cudaSetDoubleForDevice, cudaSetDoubleForHost, cudaSetupArgument (C API), Parameters: arg - Argument to push for a kernel launch cudaSetupArgument() must be preceded by a call to Parameter passing area, which starts at offset 0. Pushes size bytes of the argument pointed to by arg at offset bytes from the start of the ![]() Use of a string naming a variable as the symbol parameter was removed in CUDA 5.0.ĬudaLaunchKernel (C API), cudaFuncSetCacheConfig (C API), cudaFuncGetAttributes (C API),ĬudaLaunch (C API), cudaSetDoubleForDevice, cudaSetDoubleForHost, cudaSetupArgument (C API), cudaThreadGetCacheConfig, cudaThreadSetCacheConfig cudaError_t cudaSetupArgument (const void * arg, size_t size, size_t offset) Deprecated Returns: cudaSuccess, cudaErrorInvalidDeviceFunction, cudaErrorInvalidConfiguration,ĬudaErrorLaunchFailure, cudaErrorLaunchTimeout, cudaErrorLaunchOutOfResources,ĬudaErrorSharedObjectInitFailed, cudaErrorInvalidPtx, cudaErrorNoKernelImageForDevice, Parameters: func - Device function symbol ![]() Templated functions, pass the function symbol as follows:įunc_name cudaLaunch() must be preceded by a call toĬudaConfigureCall() since it pops the data that was pushed by cudaConfigureCall() from the The parameter specified by func must be declared as a _global_ function. The parameter func must be a device function See also: cudaLaunchKernel (C API), cudaFuncSetCacheConfig (C API), cudaFuncGetAttributes (C API), cudaLaunch (C API), cudaSetDoubleForDevice, cudaSetDoubleForHost,ĬudaError_t cudaLaunch (const void * func) Deprecated gpgpu-simdistribution/libcuda/ Go to file Cannot retrieve contributors at this time 7012 lines (6426 sloc) 232 KB Raw Blame // This file created from cudaruntimeapi.h distributed with CUDA 1.1 // Changes Copyright 2009, Tor M. Note that this function may also return error codes from previous, asynchronous Returns: cudaSuccess, cudaErrorInvalidConfiguration Note: This data contains the dimension for the grid and threadīlocks, together with any arguments for the call. Specifies the grid and block dimensions for the device call to be executed similar to theĮxecution configuration syntax. This function is deprecated as of CUDA 7.0 Some functions have overloaded C API template versions documented separately in the C API Routines module.įunction Documentation cudaError_t cudaConfigureCall (dim3 gridDim, dim3 blockDim, size_t sharedMem = 0, cudaStream_t stream = 0) Deprecated This section describes the deprecated execution control functions of the CUDA runtime \brief deprecated execution control functions of the CUDA runtime API (cuda_runtime_api.h) cudaError_t cudaSetupArgument (const void *arg, size_t size, size_t offset)Ĭonfigure a device launch. cudaError_t cudaLaunch (const void *func) Provided by: nvidia-cuda-dev_9.1.85-3ubuntu1_amd64įunctions cudaError_t cudaConfigureCall (dim3 gridDim, dim3 blockDim, size_t sharedMem=0,Ĭonfigure a device-launch.
0 Comments
Leave a Reply. |