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
|
//------------------------------------------------------------------------------
// GB_convert_b2s_template: construct triplets or CSC/CSR from bitmap
//------------------------------------------------------------------------------
// SuiteSparse:GraphBLAS, Timothy A. Davis, (c) 2017-2025, All Rights Reserved.
// SPDX-License-Identifier: Apache-2.0
//------------------------------------------------------------------------------
{
//--------------------------------------------------------------------------
// get inputs
//--------------------------------------------------------------------------
#ifdef GB_JIT_KERNEL
const int64_t avdim = A->vdim ;
const int64_t avlen = A->vlen ;
const int8_t *restrict Ab = A->b ;
const GB_Cp_TYPE *restrict W = W_input ;
const GB_Cp_TYPE *restrict Cp = Cp_input ;
GB_Ci_TYPE *restrict Ci = Ci_input ;
GB_Cj_TYPE *restrict Cj = Cj_input ;
#endif
#ifdef GB_A_TYPE
const GB_A_TYPE *restrict Ax = (GB_A_TYPE *) A->x ;
GB_C_TYPE *restrict Cx = (GB_C_TYPE *) Cx_new ;
#endif
//--------------------------------------------------------------------------
// convert A from bitmap to C sparse (Cp, Ci, Cj, and Cx)
//--------------------------------------------------------------------------
if (W == NULL)
{
//----------------------------------------------------------------------
// construct all vectors in parallel (no workspace)
//----------------------------------------------------------------------
int64_t j ;
#pragma omp parallel for num_threads(nthreads) schedule(static)
for (j = 0 ; j < avdim ; j++)
{
// gather from the bitmap into the new A (:,j)
int64_t pC = GB_IGET (Cp, j) ;
int64_t pA_start = j * avlen ;
for (int64_t i = 0 ; i < avlen ; i++)
{
int64_t pA = i + pA_start ;
if (Ab [pA])
{
// A(i,j) is in the bitmap
if (Ci != NULL) GB_ISET (Ci, pC, i) ; /* Ci [pC] = i */
if (Cj != NULL) GB_ISET (Cj, pC, j) ; /* Cj [pC] = j */
// Cx [pC] = Ax [pA])
GB_COPY (Cx, pC, Ax, pA) ;
pC++ ;
}
}
ASSERT (pC == GB_IGET (Cp, j+1)) ;
}
}
else
{
//----------------------------------------------------------------------
// compute blocks of rows in parallel
//----------------------------------------------------------------------
int tid ;
#pragma omp parallel for num_threads(nthreads) schedule(static)
for (tid = 0 ; tid < nthreads ; tid++)
{
#ifdef GB_JIT_KERNEL
const GB_Cp_TYPE *restrict Wtask = W + tid * avdim ;
#else
uint32_t *restrict Wtask32 = Cp_is_32 ? (W32 + tid * avdim) : NULL ;
uint64_t *restrict Wtask64 = Cp_is_32 ? NULL : (W64 + tid * avdim) ;
#endif
int64_t istart, iend ;
GB_PARTITION (istart, iend, avlen, tid, nthreads) ;
for (int64_t j = 0 ; j < avdim ; j++)
{
// gather from the bitmap into the new A (:,j)
int64_t pC = GB_IGET (Cp, j) + GB_IGET (Wtask, j) ;
int64_t pA_start = j * avlen ;
for (int64_t i = istart ; i < iend ; i++)
{
// see if A(i,j) is present in the bitmap
int64_t pA = i + pA_start ;
if (Ab [pA])
{
// A(i,j) is in the bitmap
if (Ci != NULL) GB_ISET (Ci, pC, i) ; /* Ci [pC] = i */
if (Cj != NULL) GB_ISET (Cj, pC, j) ; /* Cj [pC] = j */
GB_COPY (Cx, pC, Ax, pA) ;
pC++ ;
}
}
}
}
}
}
#undef GB_C_TYPE
#undef GB_A_TYPE
|
