megenginelite-sys 1.8.2

A safe megenginelite wrapper in Rust
Documentation 
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
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89

/**
 * \file dnn/src/naive/tile/opr_impl.cpp
 * MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
 *
 * Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
 *
 * Unless required by applicable law or agreed to in writing,
 * software distributed under the License is distributed on an
 * "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 */
#include "src/naive/tile/opr_impl.h"

#include <cstring>
#include "src/common/utils.h"
#include "src/naive/handle.h"

namespace megdnn {
namespace naive {

template <typename T>
void exec_tile_forward(
        _megdnn_tensor_in src, _megdnn_tensor_out dst,
        _megdnn_workspace /* workspace */) {
    auto ndim = src.layout.ndim;
    auto sptr = src.ptr<T>(), dptr = dst.ptr<T>();
    auto sshape = src.layout.shape, dshape = dst.layout.shape;
    size_t didx[TensorShape::MAX_NDIM];
    std::memset(didx, 0, sizeof(didx));
    do {
        size_t sidx[TensorShape::MAX_NDIM];
        rep(i, ndim) sidx[i] = didx[i] % sshape[i];
        auto si = get_linear_addr(sidx, sshape, ndim);
        auto di = get_linear_addr(didx, dshape, ndim);
        std::memcpy(dptr + di, sptr + si, sizeof(T) * sshape[ndim - 1]);
    } while (get_next_addr(didx, dshape, ndim, sshape[ndim - 1]));
}

void TileForwardImpl::exec(
        _megdnn_tensor_in src, _megdnn_tensor_out dst, _megdnn_workspace workspace) {
    check_exec(src.layout, dst.layout, workspace.size);
#define cb(DType)                                                                    \
    if (src.layout.dtype == DType()) {                                               \
        using ctype = typename DTypeTrait<DType>::ctype;                             \
        MEGDNN_DISPATCH_CPU_KERN_OPR(exec_tile_forward<ctype>(src, dst, workspace)); \
        return;                                                                      \
    }
    MEGDNN_FOREACH_COMPUTING_DTYPE(cb)
#undef cb
    megdnn_assert_internal(0);
}

template <typename T>
void exec_tile_backward(
        _megdnn_tensor_in diff, _megdnn_tensor_out grad,
        _megdnn_workspace /* workspace */) {
    auto ndim = diff.layout.ndim;
    auto hptr = diff.ptr<T>(), gptr = grad.ptr<T>();
    auto dshape = diff.layout.shape, sshape = grad.layout.shape;
    size_t didx[TensorShape::MAX_NDIM], sidx[TensorShape::MAX_NDIM];
    std::memset(didx, 0, sizeof(didx));
    std::memset(sidx, 0, sizeof(sidx));
    std::memset(gptr, 0, sizeof(T) * grad.layout.total_nr_elems());
    do {
        size_t sidx[TensorShape::MAX_NDIM];
        rep(i, ndim) sidx[i] = didx[i] % sshape[i];
        auto si = get_linear_addr(sidx, sshape, ndim);
        auto di = get_linear_addr(didx, dshape, ndim);
        gptr[si] += hptr[di];
    } while (get_next_addr(didx, dshape, ndim));
}

void TileBackwardImpl::exec(
        _megdnn_tensor_in diff, _megdnn_tensor_out grad, _megdnn_workspace workspace) {
    check_exec(diff.layout, grad.layout, workspace.size);
#define cb(DType)                                                  \
    if (diff.layout.dtype == DType()) {                            \
        using ctype = typename DTypeTrait<DType>::ctype;           \
        MEGDNN_DISPATCH_CPU_KERN_OPR(                              \
                exec_tile_backward<ctype>(diff, grad, workspace)); \
        return;                                                    \
    }
    MEGDNN_FOREACH_COMPUTING_DTYPE(cb)
#undef cb
    megdnn_assert_internal(0);
}

}  // namespace naive
}  // namespace megdnn
// vim: syntax=cpp.doxygen