ndrs 0.3.0

A tensor library with GPU support
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
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131

use std::ffi::c_void;

extern "C" {
    pub fn cpu_strided_add_f32(
        a: *const f32,
        a_strides: *const usize,
        b: *const f32,
        b_strides: *const usize,
        c: *mut f32,
        c_strides: *const usize,
        shape: *const usize,
        ndim: i32,
        total_elements: usize,
    );
    pub fn gpu_strided_add_f32(
        a: *const f32,
        a_strides: *const usize,
        b: *const f32,
        b_strides: *const usize,
        c: *mut f32,
        c_strides: *const usize,
        shape: *const usize,
        ndim: i32,
        total_elements: usize,
        stream: *mut c_void,
    ) -> i32;

    pub fn cpu_strided_add_i32(
        a: *const i32,
        a_strides: *const usize,
        b: *const i32,
        b_strides: *const usize,
        c: *mut i32,
        c_strides: *const usize,
        shape: *const usize,
        ndim: i32,
        total_elements: usize,
    );
    pub fn gpu_strided_add_i32(
        a: *const i32,
        a_strides: *const usize,
        b: *const i32,
        b_strides: *const usize,
        c: *mut i32,
        c_strides: *const usize,
        shape: *const usize,
        ndim: i32,
        total_elements: usize,
        stream: *mut c_void,
    ) -> i32;

    pub fn cpu_strided_copy(
        src: *const u8,
        src_offset: usize,
        src_strides: *const usize,
        ndim: i32,
        shape: *const usize,
        dst: *mut u8,
        dst_offset: usize,
        dst_strides: *const usize,
        elem_size: usize,
        total_elements: usize,
    );

    pub fn gpu_strided_copy(
        src: *const u8,
        src_offset: usize,
        src_strides: *const usize,
        ndim: i32,
        shape: *const usize,
        dst: *mut u8,
        dst_offset: usize,
        dst_strides: *const usize,
        elem_size: usize,
        total_elements: usize,
        stream: *mut c_void,
    ) -> i32;

    pub fn cpu_contiguous(
        src: *const u8,
        src_offset: usize,
        src_strides: *const usize,
        ndim: i32,
        shape: *const usize,
        dst: *mut u8,
        elem_size: usize,
        total_elements: usize,
    );

    pub fn gpu_contiguous(
        src: *const u8,
        src_offset: usize,
        src_strides: *const usize,
        ndim: i32,
        shape: *const usize,
        dst: *mut u8,
        elem_size: usize,
        total_elements: usize,
        stream: *mut c_void,
    ) -> i32;

    pub fn cpu_matmul_strided_f32(
        a: *const f32,
        a_stride_row: usize,
        a_stride_col: usize,
        b: *const f32,
        b_stride_row: usize,
        b_stride_col: usize,
        c: *mut f32,
        c_stride_row: usize,
        c_stride_col: usize,
        m: i32,
        n: i32,
        k: i32,
    );
    pub fn gpu_matmul_strided_f32(
        a: *const f32,
        a_stride_row: usize,
        a_stride_col: usize,
        b: *const f32,
        b_stride_row: usize,
        b_stride_col: usize,
        c: *mut f32,
        c_stride_row: usize,
        c_stride_col: usize,
        m: i32,
        n: i32,
        k: i32,
        stream: *mut std::ffi::c_void,
    ) -> i32;
}