cubek-convolution 0.2.0-pre.5

CubeK: Convolution Kernels
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
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
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147

use std::ops::Deref;

use cubecl::CubeDim;
use cubek_matmul::{
    components::global::{GlobalConfig, memory::GlobalMemoryConfig},
    definition::{MatmulSetupError, MatmulVectorSizes},
};
use std::{fmt::Debug, hash::Hash};

use super::*;

/// Convolution specific config, extends regular matmul `Config`.
pub trait ConvGemmConfig:
    Deref<Target: GlobalConfig> + Copy + Clone + Eq + PartialEq + Hash + Debug + Send + Sync + 'static
{
    type GlobalMatmulConfig: GlobalConfig;

    fn matmul_config(&self) -> Self::GlobalMatmulConfig;

    /// The size of the convolution kernel at `dim`
    fn params(&self) -> ConvolutionParams;
    fn operation(&self) -> ConvolutionOperation;
    fn vector_sizes(&self) -> MatmulVectorSizes;
    fn check_spatial_bounds(&self) -> bool;
    fn cube_dim(&self) -> CubeDim;
    fn lhs_global_memory_config(&self) -> GlobalMemoryConfig;
    fn rhs_global_memory_config(&self) -> GlobalMemoryConfig;
    fn out_global_memory_config(&self) -> GlobalMemoryConfig;
}

#[derive(Copy, Clone, Debug, Hash, PartialEq, Eq)]
pub struct ConvolutionConfig<M: GlobalConfig> {
    pub matmul: M,
    pub params: ConvolutionParams,
}

#[derive(Copy, Clone, Debug, Hash, PartialEq, Eq)]
pub struct ConvolutionParams {
    pub kernel_size: [u32; 3],
    pub stride: [u32; 3],
    pub dilation: [u32; 3],
    pub padding: [i32; 3],
    pub dimensionality: Dimensionality,
    pub operation: ConvolutionOperation,
}

impl ConvolutionParams {
    pub fn from_problem(problem: &ConvolutionProblem) -> Self {
        let dims = problem.dimensionality.num_dims();

        let mut params = ConvolutionParams {
            kernel_size: [0; 3],
            stride: [0; 3],
            dilation: [0; 3],
            padding: [0; 3],
            dimensionality: problem.dimensionality,
            operation: problem.operation,
        };
        params.kernel_size[0..dims].copy_from_slice(&problem.kernel_size);
        params.stride[0..dims].copy_from_slice(&problem.stride);
        params.dilation[0..dims].copy_from_slice(&problem.dilation);
        params.padding[0..dims].copy_from_slice(&problem.padding);
        params
    }
}

impl<M: GlobalConfig> Deref for ConvolutionConfig<M> {
    type Target = M;

    fn deref(&self) -> &Self::Target {
        &self.matmul
    }
}

impl<M: GlobalConfig> ConvGemmConfig for ConvolutionConfig<M> {
    type GlobalMatmulConfig = M;

    fn matmul_config(&self) -> Self::GlobalMatmulConfig {
        self.matmul
    }

    fn vector_sizes(&self) -> MatmulVectorSizes {
        self.matmul.global_vector_sizes()
    }

    fn cube_dim(&self) -> CubeDim {
        self.matmul.cube_dim()
    }

    fn check_spatial_bounds(&self) -> bool {
        let spatial_dims = self.params.dimensionality.num_dims();
        let mut has_padding = false;
        for i in 0..spatial_dims {
            has_padding |= self.params.padding[i] != 0;
        }
        has_padding
    }

    fn params(&self) -> ConvolutionParams {
        self.params
    }

    fn operation(&self) -> ConvolutionOperation {
        self.params.operation
    }

    fn lhs_global_memory_config(&self) -> GlobalMemoryConfig {
        self.matmul.lhs_reader_config().gmem_config
    }

    fn rhs_global_memory_config(&self) -> GlobalMemoryConfig {
        self.matmul.rhs_reader_config().gmem_config
    }

    fn out_global_memory_config(&self) -> GlobalMemoryConfig {
        self.matmul.writer_config().gmem_config
    }
}

impl<M: GlobalConfig> ConvolutionConfig<M> {
    #[allow(clippy::too_many_arguments)]
    pub fn new(
        matmul: M,
        kernel_size: &[u32],
        stride: &[u32],
        dilation: &[u32],
        padding: &[i32],
        dim: Dimensionality,
        operation: ConvolutionOperation,
    ) -> Result<Self, MatmulSetupError> {
        let dims = kernel_size.len();

        let mut params = ConvolutionParams {
            kernel_size: [0; 3],
            stride: [0; 3],
            dilation: [0; 3],
            padding: [0; 3],
            dimensionality: dim,
            operation,
        };
        params.kernel_size[0..dims].copy_from_slice(kernel_size);
        params.stride[0..dims].copy_from_slice(stride);
        params.dilation[0..dims].copy_from_slice(dilation);
        params.padding[0..dims].copy_from_slice(padding);
        Ok(Self { matmul, params })
    }
}