root / man / man3 / HPL_pdmatgen.3
Historique | Voir | Annoter | Télécharger (1,88 ko)
1 | 1 | equemene | .TH HPL_pdmatgen 3 "September 10, 2008" "HPL 2.0" "HPL Library Functions" |
---|---|---|---|
2 | 1 | equemene | .SH NAME |
3 | 1 | equemene | HPL_pdmatgen \- Parallel random matrix generator. |
4 | 1 | equemene | .SH SYNOPSIS |
5 | 1 | equemene | \fB\&#include "hpl.h"\fR |
6 | 1 | equemene | |
7 | 1 | equemene | \fB\&void\fR |
8 | 1 | equemene | \fB\&HPL_pdmatgen(\fR |
9 | 1 | equemene | \fB\&const HPL_T_grid *\fR |
10 | 1 | equemene | \fI\&GRID\fR, |
11 | 1 | equemene | \fB\&const int\fR |
12 | 1 | equemene | \fI\&M\fR, |
13 | 1 | equemene | \fB\&const int\fR |
14 | 1 | equemene | \fI\&N\fR, |
15 | 1 | equemene | \fB\&const int\fR |
16 | 1 | equemene | \fI\&NB\fR, |
17 | 1 | equemene | \fB\&double *\fR |
18 | 1 | equemene | \fI\&A\fR, |
19 | 1 | equemene | \fB\&const int\fR |
20 | 1 | equemene | \fI\&LDA\fR, |
21 | 1 | equemene | \fB\&const int\fR |
22 | 1 | equemene | \fI\&ISEED\fR |
23 | 1 | equemene | \fB\&);\fR |
24 | 1 | equemene | .SH DESCRIPTION |
25 | 1 | equemene | \fB\&HPL_pdmatgen\fR |
26 | 1 | equemene | generates (or regenerates) a parallel random matrix A. |
27 | 1 | equemene | |
28 | 1 | equemene | The pseudo-random generator uses the linear congruential algorithm: |
29 | 1 | equemene | X(n+1) = (a * X(n) + c) mod m as described in the Art of Computer |
30 | 1 | equemene | Programming, Knuth 1973, Vol. 2. |
31 | 1 | equemene | .SH ARGUMENTS |
32 | 1 | equemene | .TP 8 |
33 | 1 | equemene | GRID (local input) const HPL_T_grid * |
34 | 1 | equemene | On entry, GRID points to the data structure containing the |
35 | 1 | equemene | process grid information. |
36 | 1 | equemene | .TP 8 |
37 | 1 | equemene | M (global input) const int |
38 | 1 | equemene | On entry, M specifies the number of rows of the matrix A. |
39 | 1 | equemene | M must be at least zero. |
40 | 1 | equemene | .TP 8 |
41 | 1 | equemene | N (global input) const int |
42 | 1 | equemene | On entry, N specifies the number of columns of the matrix A. |
43 | 1 | equemene | N must be at least zero. |
44 | 1 | equemene | .TP 8 |
45 | 1 | equemene | NB (global input) const int |
46 | 1 | equemene | On entry, NB specifies the blocking factor used to partition |
47 | 1 | equemene | and distribute the matrix A. NB must be larger than one. |
48 | 1 | equemene | .TP 8 |
49 | 1 | equemene | A (local output) double * |
50 | 1 | equemene | On entry, A points to an array of dimension (LDA,LocQ(N)). |
51 | 1 | equemene | On exit, this array contains the coefficients of the randomly |
52 | 1 | equemene | generated matrix. |
53 | 1 | equemene | .TP 8 |
54 | 1 | equemene | LDA (local input) const int |
55 | 1 | equemene | On entry, LDA specifies the leading dimension of the array A. |
56 | 1 | equemene | LDA must be at least max(1,LocP(M)). |
57 | 1 | equemene | .TP 8 |
58 | 1 | equemene | ISEED (global input) const int |
59 | 1 | equemene | On entry, ISEED specifies the seed number to generate the |
60 | 1 | equemene | matrix A. ISEED must be at least zero. |
61 | 1 | equemene | .SH SEE ALSO |
62 | 1 | equemene | .BR HPL_ladd \ (3), |
63 | 1 | equemene | .BR HPL_lmul \ (3), |
64 | 1 | equemene | .BR HPL_setran \ (3), |
65 | 1 | equemene | .BR HPL_xjumpm \ (3), |
66 | 1 | equemene | .BR HPL_jumpit \ (3), |
67 | 1 | equemene | .BR HPL_drand \ (3). |