torsh-backend 0.1.2

Backend abstraction layer for ToRSh
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

//! Rust bindings for reduction operation kernels

#![allow(unused_imports)]
#[allow(unused_imports)]
use crate::cuda::error::{CudaError, CudaResult};

/// Launch sum reduction kernel
pub fn launch_sum_f32(
    input: *mut f32,
    output: *mut f32,
    size: usize,
    axis: i32,
    stream: crate::cuda::CUstream,
) {
    unsafe {
        super::cuda_kernels::launch_sum_f32(input, output, size as i32, axis, stream);
    }
}

/// Launch mean reduction kernel
pub fn launch_mean_f32(
    input: *mut f32,
    output: *mut f32,
    size: usize,
    axis: i32,
    stream: crate::cuda::CUstream,
) {
    unsafe {
        super::cuda_kernels::launch_mean_f32(input, output, size as i32, axis, stream);
    }
}

/// Launch max reduction kernel
pub fn launch_max_f32(
    input: *mut f32,
    output: *mut f32,
    size: usize,
    axis: i32,
    stream: crate::cuda::CUstream,
) {
    unsafe {
        super::cuda_kernels::launch_max_f32(input, output, size as i32, axis, stream);
    }
}

/// Launch min reduction kernel
pub fn launch_min_f32(
    input: *mut f32,
    output: *mut f32,
    size: usize,
    axis: i32,
    stream: crate::cuda::CUstream,
) {
    unsafe {
        super::cuda_kernels::launch_min_f32(input, output, size as i32, axis, stream);
    }
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn test_reduction_kernel_bindings_exist() {
        // Verify the functions are properly linked
    }
}