class cv::ocl::Kernel
Overview
#include <ocl.hpp> class Kernel { public: // construction Kernel(); Kernel( const char* kname, const Program& prog ); Kernel( const char* kname, const ProgramSource& prog, const String& buildopts = String(), String* errmsg = 0 ); Kernel(const Kernel& k); // methods template <typename _Tp0> Kernel& args(const _Tp0& a0); template < typename _Tp0, typename _Tp1 > Kernel& args( const _Tp0& a0, const _Tp1& a1 ); template < typename _Tp0, typename _Tp1, typename _Tp2 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6, typename _Tp7 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6, const _Tp7& a7 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6, typename _Tp7, typename _Tp8 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6, const _Tp7& a7, const _Tp8& a8 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6, typename _Tp7, typename _Tp8, typename _Tp9 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6, const _Tp7& a7, const _Tp8& a8, const _Tp9& a9 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6, typename _Tp7, typename _Tp8, typename _Tp9, typename _Tp10 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6, const _Tp7& a7, const _Tp8& a8, const _Tp9& a9, const _Tp10& a10 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6, typename _Tp7, typename _Tp8, typename _Tp9, typename _Tp10, typename _Tp11 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6, const _Tp7& a7, const _Tp8& a8, const _Tp9& a9, const _Tp10& a10, const _Tp11& a11 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6, typename _Tp7, typename _Tp8, typename _Tp9, typename _Tp10, typename _Tp11, typename _Tp12 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6, const _Tp7& a7, const _Tp8& a8, const _Tp9& a9, const _Tp10& a10, const _Tp11& a11, const _Tp12& a12 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6, typename _Tp7, typename _Tp8, typename _Tp9, typename _Tp10, typename _Tp11, typename _Tp12, typename _Tp13 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6, const _Tp7& a7, const _Tp8& a8, const _Tp9& a9, const _Tp10& a10, const _Tp11& a11, const _Tp12& a12, const _Tp13& a13 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6, typename _Tp7, typename _Tp8, typename _Tp9, typename _Tp10, typename _Tp11, typename _Tp12, typename _Tp13, typename _Tp14 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6, const _Tp7& a7, const _Tp8& a8, const _Tp9& a9, const _Tp10& a10, const _Tp11& a11, const _Tp12& a12, const _Tp13& a13, const _Tp14& a14 ); template < typename _Tp0, typename _Tp1, typename _Tp2, typename _Tp3, typename _Tp4, typename _Tp5, typename _Tp6, typename _Tp7, typename _Tp8, typename _Tp9, typename _Tp10, typename _Tp11, typename _Tp12, typename _Tp13, typename _Tp14, typename _Tp15 > Kernel& args( const _Tp0& a0, const _Tp1& a1, const _Tp2& a2, const _Tp3& a3, const _Tp4& a4, const _Tp5& a5, const _Tp6& a6, const _Tp7& a7, const _Tp8& a8, const _Tp9& a9, const _Tp10& a10, const _Tp11& a11, const _Tp12& a12, const _Tp13& a13, const _Tp14& a14, const _Tp15& a15 ); bool compileWorkGroupSize(size_t wsz []) const; bool create( const char* kname, const Program& prog ); bool create( const char* kname, const ProgramSource& prog, const String& buildopts, String* errmsg = 0 ); bool empty() const; size_t localMemSize() const; Kernel& operator=(const Kernel& k); size_t preferedWorkGroupSizeMultiple() const; void* ptr() const; bool run( int dims, size_t globalsize [], size_t localsize [], bool sync, const Queue& q = Queue() ); bool runTask( bool sync, const Queue& q = Queue() ); int set( int i, const void* value, size_t sz ); int set( int i, const Image2D& image2D ); int set( int i, const UMat& m ); int set( int i, const KernelArg& arg ); template <typename _Tp> int set( int i, const _Tp& value ); size_t workGroupSize() const; protected: // fields Impl* p; };
Detailed Documentation
Methods
bool run( int dims, size_t globalsize [], size_t localsize [], bool sync, const Queue& q = Queue() )
Run the OpenCL kernel.
Parameters:
| dims | the work problem dimensions. It is the length of globalsize and localsize. It can be either 1, 2 or 3. |
| globalsize | work items for each dimension. It is not the final globalsize passed to OpenCL. Each dimension will be adjusted to the nearest integer divisible by the corresponding value in localsize. If localsize is NULL, it will still be adjusted depending on dims. The adjusted values are greater than or equal to the original values. |
| localsize | work-group size for each dimension. |
| sync | specify whether to wait for OpenCL computation to finish before return. |
| q | command queue |