ruPRIM 0.1.8

Parallel primitives, reductions, scans, and indexing for Ruda.
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

use ruda_kernel::dsl as kernel_dsl;
use crate::reduce::{
    ReduceInstruction, ReducePrecision, VectorizationMode,
    components::{
        args::NumericVector,
        global::idle_check,
        instructions::{Accumulator, reduce_inplace},
        readers::{Reader, plane::PlaneReader},
        writers::Writer,
    },
    routines::{PlaneMergeStrategy, PlaneReduceBlueprint},
};

use crate::reduce::components::instructions::ReduceStep;
use ruda_kernel::dsl::prelude::*;
use ruda_kernel::library::tensor::r#virtual::VirtualTensor;

#[derive(RudaType)]
pub struct GlobalFullPlaneReduce;

#[ruda]
impl GlobalFullPlaneReduce {
    pub fn execute<P: ReducePrecision, Out: NumericVector, I: ReduceInstruction<P>>(
        input: &VirtualTensor<P::EI, P::SI>,
        output: &mut VirtualTensor<Out::T, Out::N, ReadWrite>,
        reduce_axis: usize,
        out_vec_axis: usize,
        inst: &I,
        #[comptime] vectorization_mode: VectorizationMode,
        #[comptime] blueprint: PlaneReduceBlueprint,
    ) {
        let acc_format = I::accumulator_format(inst);
        let planes_per_ruda = RUDA_DIM_Y as usize * RUDA_DIM_Z as usize;
        let plane_index = PLANE_POS as usize;
        let assigned = plane_index < planes_per_ruda;
        let leader = plane_exclusive_sum(1u32) == 0;
        let reduction_index = RUDA_POS * planes_per_ruda + plane_index;
        let write_index = reduction_index;

        let mut writer = Writer::<Out>::new::<P>(
            input,
            output,
            reduce_axis,
            out_vec_axis,
            write_index,
            vectorization_mode,
            acc_format,
        );

        let write_count = writer.write_count();
        let reduce_index_start = write_index * write_count;

        let idle = idle_check::<P, Out>(
            input,
            output,
            reduce_axis,
            reduce_index_start,
            vectorization_mode,
            blueprint.plane_idle,
        );
        #[comptime]
        let idle = match idle {
            ComptimeOption::Some(idle) => idle || !assigned,
            ComptimeOption::None => !assigned,
        };
        let idle = ComptimeOption::new_Some(idle);

        for b in 0..write_count {
            let reduce_index = reduce_index_start + b;
            let result = Self::reduce_single::<P, Out, I>(
                input,
                output,
                reduce_axis,
                reduce_index,
                inst,
                idle,
                vectorization_mode,
                blueprint,
            );

            if leader && assigned {
                writer.write::<P, I>(b, result, inst);
            }
        }

        let commit_required = writer.commit_required();

        #[allow(clippy::collapsible_if)]
        if commit_required {
            if leader && assigned {
                writer.commit();
            }
        }
    }

    #[allow(clippy::too_many_arguments)]
    fn reduce_single<P: ReducePrecision, Out: NumericVector, I: ReduceInstruction<P>>(
        input: &VirtualTensor<P::EI, P::SI>,
        output: &mut VirtualTensor<Out::T, Out::N, ReadWrite>,
        reduce_axis: usize,
        reduce_index: usize,
        inst: &I,
        idle: ComptimeOption<bool>,
        #[comptime] vectorization_mode: VectorizationMode,
        #[comptime] blueprint: PlaneReduceBlueprint,
    ) -> Accumulator<P> {
        let reader = Reader::<P>::new::<I, Out>(
            input,
            output,
            inst,
            reduce_axis,
            reduce_index,
            idle,
            blueprint.bound_checks,
            vectorization_mode,
            blueprint.plane_dim_ceil,
        );
        let reader = PlaneReader::<P>::new(reader);

        let mut accumulator = I::null_accumulator(inst);

        let iteration_plane_reduce_mode = match blueprint.plane_merge_strategy {
            PlaneMergeStrategy::Eager => ReduceStep::Plane,
            PlaneMergeStrategy::Lazy => ReduceStep::Identity,
        };
        for i in 0..reader.length() {
            let item = reader.read(i);
            reduce_inplace::<P, I>(inst, &mut accumulator, item, iteration_plane_reduce_mode);
        }

        match blueprint.plane_merge_strategy {
            PlaneMergeStrategy::Lazy => {
                I::plane_reduce_inplace(inst, &mut accumulator);
                accumulator
            }
            PlaneMergeStrategy::Eager => accumulator,
        }
    }
}