This project is archived. Its data is read-only.

Commit fe182333 authored Jan 12, 2021 by Tomáš Oberhuber

Writting documentation - general concepts.

parent 75a8f3e3

Documentation/Tutorials/CMakeLists.txt

+1 −0

Original line number	Diff line number	Diff line
		add_subdirectory( GeneralConcepts )
		add_subdirectory( Arrays )
		add_subdirectory( Vectors )
		add_subdirectory( ReductionAndScan )

Documentation/Tutorials/CoreConcepts/tutorial_CoreConcepts.md

deleted100644 → 0

+0 −15

Original line number	Diff line number	Diff line
		\page tutorial_CoreConcepts Core concepts

		## Introduction




		## Table of Contents
		1. [Devices and allocators](#devices_and_allocators)
		2. [Algorithms and lambda functions](#algorithms_and_lambda_functions)
		3. [Smart pointers and views](#static_for)


		## Devices and allocators<a name="devices_and_allocators"></a>

Documentation/Tutorials/GeneralConcepts/snippet_algorithms_and_lambda_functions_parallel_for.cpp

0 → 100644

+8 −0

Original line number	Diff line number	Diff line
		template< typename Device >
		void vectorAddition( double* v1, double* v2, double* sum, const int size )
		{
		auto sum_lambda = [=] __cuda_callable__ ( int i ) mutable {
		sum[ i ] = v1[ i ] + v2[ i ];
		}
		TNL::Algorithms::ParalellFor< Device >::exec( 0, size, sum_lambda );
		}
		No newline at end of file

Documentation/Tutorials/GeneralConcepts/snippet_algorithms_and_lambda_functions_reduction.cpp

0 → 100644

+10 −0

Original line number	Diff line number	Diff line
		template< typename Device >
		void scalarProduct( double* v1, double* v2, double* product, const int size )
		{
		auto fetch = [=] __cuda_callable__ ( int i ) -> double {
		return = v1[ i ] * v2[ i ];
		}
		auto reduce = [] __cuda_callable__ ( const double& a, const double& b ) {
		return a + b; };
		TNL::Algorithms::Reduction< Device >::reduce( 0, size, reduce, fetch, 0.0 );
		}
		No newline at end of file

Documentation/Tutorials/GeneralConcepts/snippet_algorithms_and_lambda_functions_reduction_2.cpp

0 → 100644

+12 −0

Original line number	Diff line number	Diff line
		template< typename Device >
		void scalarProduct( double* u1, double* u2,
		double* v1, double* v2,
		double* product, const int size )
		{
		auto fetch = [=] __cuda_callable__ ( int i ) -> double {
		return = ( u1[ i ] + u2[ i ] ) * ( v1[ i ] + v2[ i ] );
		}
		auto reduce = [] __cuda_callable__ ( const double& a, const double& b ) {
		return a + b; };
		TNL::Algorithms::Reduction< Device >::reduce( 0, size, reduce, fetch, 0.0 );
		}
		No newline at end of file