1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
|
// =================================================================================================
// This file is part of the CLBlast project. The project is licensed under Apache Version 2.0. This
// project loosely follows the Google C++ styleguide and uses a tab-size of two spaces and a max-
// width of 100 characters per line.
//
// Author(s):
// Cedric Nugteren <www.cedricnugteren.nl>
//
// This file implements the Xgemv routine. The precision is implemented using a template argument.
//
// =================================================================================================
#ifndef CLBLAST_ROUTINES_XGEMV_H_
#define CLBLAST_ROUTINES_XGEMV_H_
#include "internal/routine.h"
namespace clblast {
// =================================================================================================
// See comment at top of file for a description of the class
template <typename T>
class Xgemv: public Routine<T> {
public:
// Members and methods from the base class
using Routine<T>::db_;
using Routine<T>::source_string_;
using Routine<T>::queue_;
using Routine<T>::GetProgramFromCache;
using Routine<T>::TestVectorX;
using Routine<T>::TestVectorY;
using Routine<T>::TestMatrixA;
using Routine<T>::TestMatrixAP;
using Routine<T>::RunKernel;
using Routine<T>::ErrorIn;
// Constructor
Xgemv(Queue &queue, Event &event, const std::string &name = "GEMV");
// Templated-precision implementation of the routine
StatusCode DoGemv(const Layout layout, const Transpose a_transpose,
const size_t m, const size_t n,
const T alpha,
const Buffer<T> &a_buffer, const size_t a_offset, const size_t a_ld,
const Buffer<T> &x_buffer, const size_t x_offset, const size_t x_inc,
const T beta,
const Buffer<T> &y_buffer, const size_t y_offset, const size_t y_inc);
// Generic version used also for other matrix-vector multiplications
StatusCode MatVec(const Layout layout, const Transpose a_transpose,
const size_t m, const size_t n,
const T alpha,
const Buffer<T> &a_buffer, const size_t a_offset, const size_t a_ld,
const Buffer<T> &x_buffer, const size_t x_offset, const size_t x_inc,
const T beta,
const Buffer<T> &y_buffer, const size_t y_offset, const size_t y_inc,
bool fast_kernel, bool fast_kernel_rot,
const size_t parameter, const bool packed,
const size_t kl, const size_t ku);
private:
// Static variable to get the precision
const static Precision precision_;
};
// =================================================================================================
} // namespace clblast
// CLBLAST_ROUTINES_XGEMV_H_
#endif
|
