ViennaCL - The Vienna Computing Library  1.7.0
Free open-source GPU-accelerated linear algebra and solver library.
vandermonde_matrix_operations.hpp
Go to the documentation of this file.
1 #ifndef VIENNACL_LINALG_OPENCL_VANDERMONDE_MATRIX_OPERATIONS_HPP_
2 #define VIENNACL_LINALG_OPENCL_VANDERMONDE_MATRIX_OPERATIONS_HPP_
3 
4 /* =========================================================================
5  Copyright (c) 2010-2015, Institute for Microelectronics,
6  Institute for Analysis and Scientific Computing,
7  TU Wien.
8  Portions of this software are copyright by UChicago Argonne, LLC.
9 
10  -----------------
11  ViennaCL - The Vienna Computing Library
12  -----------------
13 
14  Project Head: Karl Rupp rupp@iue.tuwien.ac.at
15 
16  (A list of authors and contributors can be found in the manual)
17 
18  License: MIT (X11), see file LICENSE in the base directory
19 ============================================================================= */
20 
25 #include "viennacl/forwards.h"
26 #include "viennacl/ocl/backend.hpp"
27 #include "viennacl/scalar.hpp"
28 #include "viennacl/vector.hpp"
29 #include "viennacl/tools/tools.hpp"
30 #include "viennacl/fft.hpp"
31 #include "viennacl/linalg/opencl/kernels/fft.hpp"
32 
33 namespace viennacl
34 {
35 namespace linalg
36 {
37 namespace opencl
38 {
39 
48 template<typename NumericT, unsigned int AlignmentV>
49 void prod_impl(viennacl::vandermonde_matrix<NumericT, AlignmentV> const & A,
50  viennacl::vector_base<NumericT> const & x,
51  viennacl::vector_base<NumericT> & y)
52 {
53  viennacl::ocl::context & ctx = const_cast<viennacl::ocl::context &>(viennacl::traits::opencl_handle(A).context());
54  viennacl::linalg::opencl::kernels::fft<NumericT>::init(ctx);
55 
56  viennacl::ocl::kernel & kernel = ctx.get_kernel(viennacl::linalg::opencl::kernels::fft<NumericT>::program_name(), "vandermonde_prod");
57  viennacl::ocl::enqueue(kernel(viennacl::traits::opencl_handle(A),
58  viennacl::traits::opencl_handle(x),
59  viennacl::traits::opencl_handle(y),
60  static_cast<cl_uint>(A.size1())));
61 }
62 
63 } //namespace opencl
64 } //namespace linalg
65 } //namespace viennacl
66 
67 
68 #endif
fft.hpp
OpenCL kernel file for FFT operations.
viennacl::ocl::kernel
Represents an OpenCL kernel within ViennaCL.
Definition: kernel.hpp:58
tools.hpp
Various little tools used here and there in ViennaCL.
viennacl::linalg::opencl::kernels::fft
Main kernel class for generating OpenCL kernels for the fast Fourier transform.
Definition: fft.hpp:260
viennacl::ocl::context
Manages an OpenCL context and provides the respective convenience functions for creating buffers...
Definition: context.hpp:55
forwards.h
This file provides the forward declarations for the main types used within ViennaCL.
viennacl::vandermonde_matrix::size1
vcl_size_t size1() const
Returns the number of rows of the matrix.
Definition: vandermonde_matrix.hpp:95
viennacl::vandermonde_matrix
A Vandermonde matrix class.
Definition: forwards.h:418
viennacl
Main namespace in ViennaCL. Holds all the basic types such as vector, matrix, etc. and defines operations upon them.
Definition: cpu_ram.hpp:34
viennacl::ocl::context::get_kernel
viennacl::ocl::kernel & get_kernel(std::string const &program_name, std::string const &kernel_name)
Convenience function for retrieving the kernel of a program directly from the context.
Definition: context.hpp:605
viennacl::vector_base< NumericT >
backend.hpp
Implementations of the OpenCL backend, where all contexts are stored in.
viennacl::ocl::enqueue
void enqueue(KernelType &k, viennacl::ocl::command_queue const &queue)
Enqueues a kernel in the provided queue.
Definition: enqueue.hpp:50
vector.hpp
The vector type with operator-overloads and proxy classes is defined here. Linear algebra operations ...
fft.hpp
All routines related to the Fast Fourier Transform. Experimental.
viennacl::linalg::opencl::kernels::fft::init
static void init(viennacl::ocl::context &ctx)
Definition: fft.hpp:267
scalar.hpp
Implementation of the ViennaCL scalar class.
viennacl::linalg::opencl::prod_impl
void prod_impl(const matrix_base< NumericT > &A, bool trans_A, const vector_base< NumericT > &vec, vector_base< NumericT > &result)
Carries out matrix-vector multiplication.
Definition: matrix_operations.hpp:305