Wrapper class for a CUDA code module. More...

#include <module.hpp>

Public Member Functions
module::handle_t	handle () const

context::handle_t	context_handle () const

device::id_t	device_id () const

context_t	context () const

device_t	device () const

cuda::kernel_t	get_kernel (const char *name) const
	Obtains an already-compiled kernel previously associated with this module. More...

cuda::kernel_t	get_kernel (const ::std::string &name) const

memory::region_t	get_global_region (const char *name) const

CUsurfref *	get_surface (const char *name) const

CUtexref *	get_texture_reference (const char *name) const

	module_t (const module_t &)=delete

	module_t (module_t &&other) noexcept

module_t &	operator= (const module_t &)=delete

module_t &	operator= (module_t &&other) noexcept

Detailed Description

Wrapper class for a CUDA code module.

Note: This class is a "reference type", not a "value type". Therefore, making changes to the module is a const-respecting operation on this class.

Member Function Documentation

cuda::kernel_t cuda::module_t::get_kernel ( const char * name ) const

inline

Obtains an already-compiled kernel previously associated with this module.

Parameters

name	The function name, in case of a C-style function, or the mangled function signature, in case of a C++-style function.

The documentation for this class was generated from the following file: