gpu-tRNA-mapper/smallkernels.cuh at main · fkallen/gpu-tRNA-mapper · GitHub

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
#ifndef SMALLKERNELS_CUH
#define SMALLKERNELS_CUH

#include "smallkernels.cuh"

#include <rmm/device_uvector.hpp>


void callVerticalMaxReduceWithIndexKernel(
    const int* d_input,
    int numRows,
    int numColumns,
    int* d_maxOutput,
    int* d_rowIndexOfMax,
    cudaStream_t stream
);

void callVerticalMaxReduceFindNumBestScoresKernel(
    const int* d_input,
    int numRows,
    int numColumns,
    int* d_maxOutput,
    int* d_numMax,
    int minScore,
    cudaStream_t stream
);

void callVerticalMaxReduceFindIndicesOfBestScoresKernel(
    const int* d_input,
    int numRows,
    int numColumns,
    const int* d_maxOutput,
    const int* d_numMax,
    const int* d_numMaxPrefixSum,
    int minScore,
    int* d_rowIndicesOfMax,
    cudaStream_t stream
);


void getSegmentIdsPerElement(
    int* d_output,
    const int* d_segmentSizes,
    const int* d_segmentSizesPrefixSum,
    int numSegments,
    int numElements,
    cudaStream_t stream
);

rmm::device_uvector<int> getSegmentIdsPerElement(
    const int* d_segmentSizes,
    const int* d_segmentSizesPrefixSum,
    int numSegments,
    int numElements,
    cudaStream_t stream
);


#endif