forked from AnonymousRepo123/AlphaSparse
-
Notifications
You must be signed in to change notification settings - Fork 0
/
shared_memory_op_warp_compress.hpp
executable file
·145 lines (106 loc) · 6.82 KB
/
shared_memory_op_warp_compress.hpp
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
// 将warp级别的遍历去掉，前提条件的是block内所有线程粒度的块的大小相等
#ifndef SHARED_MEMORY_TEMPLATE_WARP_COMPRESS_H
#define SHARED_MEMORY_TEMPLATE_WARP_COMPRESS_H
#include "struct.hpp"
#include "config.hpp"
#include "arr_optimization.hpp"
#include "code_builder.hpp"
#include "shared_memory_op.hpp"
typedef struct shared_memory_template_warp_compress
{
// 模板对应的稠密矩阵号
unsigned long dense_block_index;
// 对应的密集矩阵
sparse_struct_t *matrix = NULL;
// 当前密集子块的首行行号
unsigned long kernal_first_row_index = 0;
unsigned long kernal_first_col_index = 0;
// 用一个变量来存储有效行数量
unsigned long effective_row_num;
// 用一个变量存是否要用原子加来归约
bool is_atom_add = false;
// 重新构建一个数组，存储每一行的在thread计算结果中的偏移量
void *row_offset_in_thread_tmp_result = NULL;
data_type data_type_of_row_offset_in_thread_tmp_result;
unsigned long size_of_row_offset_in_thread_tmp_result;
// 每个block的首行行号，用来进行归约
void *block_first_row_index = NULL;
data_type data_type_of_block_first_row_index;
unsigned long size_of_block_first_row_index;
// 每个block的第一个线程粒度的块的索引
void *block_begin_thread_index_offset = NULL;
data_type data_type_of_block_begin_thread_index_offset;
unsigned long size_of_block_begin_thread_index_offset;
// 每个block内线程粒度的块的大小
void *thread_block_size_in_block = NULL;
data_type data_type_of_thread_block_size_in_block;
unsigned long size_of_thread_block_size_in_block;
// 排序相关
// 用一个可能存在的数组存储排序之后的输出，可能有全局的和局部的两种情况
bool global_sort_index = false;
bool local_sort_index = false;
void *row_index_before_sort = NULL;
data_type data_type_of_row_index_before_sort;
unsigned long size_of_row_index_before_sort;
// block和warp的偏移量，以及遍历的范围可以尝试使用共享内存的方式来解决。其中block和warp的偏移量的偏移量可以使用
// 大小都和不同层次的block size相同
void *block_nz_begin_offset = NULL;
data_type data_type_of_block_nz_begin_offset;
unsigned long size_of_block_nz_begin_offset;
// 当前稠密视图子块的所有值
void *val_arr = NULL;
data_type data_type_of_val_arr;
unsigned long size_of_val_arr;
// 当前稠密视图子块的所有列号
void *col_index_arr = NULL;
data_type data_type_of_col_index_arr;
unsigned long size_of_col_index_arr;
// 线程的计算结果在结果中的行偏移量
arr_compress_type row_offset_in_thread_tmp_result_compress = NONE_COMPRESS;
void *row_offset_in_thread_tmp_result_compress_meta = NULL;
// 每个block的行起始位置的压缩
arr_compress_type block_first_row_index_compress = NONE_COMPRESS;
void *block_first_row_index_compress_meta = NULL;
// 线程块粒度的块的第一个warp粒度块号，可以使用线性压缩
arr_compress_type block_begin_thread_index_offset_compress = NONE_COMPRESS;
void *block_begin_thread_index_offset_compress_meta = NULL;
// warp中线程粒度的块的大小，可以使用常值压缩
arr_compress_type thread_block_size_in_block_compress = NONE_COMPRESS;
void *thread_block_size_in_block_compress_meta = NULL;
// 排序产生的数组不太可能压缩
arr_compress_type row_index_before_sort_compress = NONE_COMPRESS;
void *row_index_before_sort_compress_meta = NULL;
// 块非零元偏移和warp非零元偏移的的压缩
arr_compress_type block_nz_begin_offset_compress = NONE_COMPRESS;
void *block_nz_begin_offset_compress_meta = NULL;
// 当前内核使用的线程块数量和线程块内的线程数量
unsigned long tblock_num = get_config()["DEFAULT_THREAD_BLOCK_NUM"].as_integer();
unsigned long thread_num_in_block = get_config()["DEFAULT_THREAD_NUM_IN_BLOCK"].as_integer();
// 当前模板每一行的树状规约并行度
unsigned long thread_num_of_row_reduce = 1;
// 用一个数存储一个模板的id的哈希
unsigned long hash_of_this_template;
} shared_memory_template_warp_compress_t;
shared_memory_template_warp_compress_t *init_shared_memory_template_warp_compress(code_builder_t *builder, unsigned long dense_block_id);
shared_memory_template_warp_compress_t *init_shared_memory_template_warp_compress(shared_memory_template_t *old_template);
bool is_supported_by_shared_memory_template_warp_compress(code_builder_t *builder, unsigned long dense_block_id);
bool is_supported_by_shared_memory_template_warp_compress(sparse_struct_t *matrix, unsigned long dense_block_id);
void store_template_data(shared_memory_template_warp_compress_t *output_template, string output_dir, bool force_not_share_global_sort_index = false);
string code_of_template_data_struct(shared_memory_template_warp_compress_t *output_template, unsigned long dense_block_id);
string code_of_read_template_data_from_file_func_define(shared_memory_template_warp_compress_t *output_template, unsigned long dense_block_id, bool force_not_share_global_sort_index = false);
string code_of_template_kernal(shared_memory_template_warp_compress_t *output_template, unsigned long dense_block_id);
string code_of_write_template_data_to_gpu(shared_memory_template_warp_compress_t *output_template, unsigned long dense_block_id, bool force_not_share_global_sort_index = false);
string code_of_kernal_function_call(shared_memory_template_warp_compress_t *output_template, unsigned long dense_block_id);
// 一系列压缩
bool compress_block_begin_thread_index_offset(shared_memory_template_warp_compress_t *output_template, bool need_check = true, arr_compress_type type = LINEAR_COMPRESS);
bool compress_thread_block_size_in_block(shared_memory_template_warp_compress_t *output_template, bool need_check = true, arr_compress_type type = CONSTANT_COMPRESS);
bool compress_block_nz_begin_offset(shared_memory_template_warp_compress_t *output_template, bool need_check = true, arr_compress_type type = LINEAR_COMPRESS);
// 每个block首行行号的压缩
bool compress_block_first_row_index(shared_memory_template_warp_compress_t *output_template, bool need_check = true, arr_compress_type type = LINEAR_COMPRESS);
// 每一行中间结果的压缩，可以线性压缩
bool compress_row_offset_in_thread_tmp_result(shared_memory_template_warp_compress_t *output_template, bool need_check = true, arr_compress_type type = LINEAR_COMPRESS);
// 尝试所有的压缩
void try_all_compress(shared_memory_template_warp_compress_t *output_template);
// 设定每一行归约的并行度
bool set_row_reduce_thread_num(shared_memory_template_warp_compress_t *output_template, unsigned long row_reduce_thread_num);
#endif