Debian Codegraph @

hpl/www/HPL_plindx10.html

raw
1 <HTML>
2 <HEAD>
3 <TITLE>HPL_plindx10 HPL 2.0 Library Functions September 10, 2008</TITLE>
4 </HEAD>
5
6 <BODY BGCOLOR="WHITE" TEXT = "#000000" LINK = "#0000ff" VLINK = "#000099"
7 ALINK = "#ffff00">
8
9 <H1>Name</H1>
10 <B>HPL_plindx10</B> Compute the logarithmic maps for the spreading.
11
12 <H1>Synopsis</H1>
13 <CODE>#include "hpl.h"</CODE><BR><BR>
14 <CODE>void</CODE>
15 <CODE>HPL_plindx10(</CODE>
16 <CODE>HPL_T_panel *</CODE>
17 <CODE>PANEL</CODE>,
18 <CODE>const int</CODE>
19 <CODE>K</CODE>,
20 <CODE>const int *</CODE>
21 <CODE>IPID</CODE>,
22 <CODE>int *</CODE>
23 <CODE>IPLEN</CODE>,
24 <CODE>int *</CODE>
25 <CODE>IPMAP</CODE>,
26 <CODE>int *</CODE>
27 <CODE>IPMAPM1</CODE>
28 <CODE>);</CODE>
29
30 <H1>Description</H1>
31 <B>HPL_plindx10</B>
32 computes three arrays IPLEN, IPMAP and IPMAPM1 that
33 contain the logarithmic mapping information for the spreading phase.
34
35 <H1>Arguments</H1>
36 <PRE>
37 PANEL (local input/output) HPL_T_panel *
38 On entry, PANEL points to the data structure containing the
39 panel information.
40 </PRE>
41 <PRE>
42 K (global input) const int
43 On entry, K specifies the number of entries in IPID. K is at
44 least 2*N, and at most 4*N.
45 </PRE>
46 <PRE>
47 IPID (global input) const int *
48 On entry, IPID is an array of length K. The first K entries
49 of that array contain the src and final destination resulting
50 from the application of the interchanges.
51 </PRE>
52 <PRE>
53 IPLEN (global output) int *
54 On entry, IPLEN is an array of dimension NPROW + 1. On exit,
55 this array is such that IPLEN[i] is the number of rows of A
56 in the processes before process IMAP[i] after the sort, with
57 the convention that IPLEN[nprow] is the total number of rows.
58 In other words, IPLEN[i+1] - IPLEN[i] is the local number of
59 rows of A that should be moved for each process. IPLEN is
60 such that the number of rows of the source process row can be
61 computed as IPLEN[1] - IPLEN[0], and the remaining entries of
62 this array are sorted so that the quantities IPLEN[i+1] -
63 IPLEN[i] are logarithmically sorted.
64 </PRE>
65 <PRE>
66 IPMAP (global output) int *
67 On entry, IPMAP is an array of dimension NPROW. On exit, this
68 array contains the logarithmic mapping of the processes. In
69 other words, IPMAP[myrow] is the corresponding sorted process
70 coordinate.
71 </PRE>
72 <PRE>
73 IPMAPM1 (global output) int *
74 On entry, IPMAPM1 is an array of dimension NPROW. On exit,
75 this array contains the inverse of the logarithmic mapping
76 contained in IPMAP: IPMAPM1[ IPMAP[i] ] = i, for all i in
77 [0.. NPROW)
78 </PRE>
79
80 <H1>See Also</H1>
81 <A HREF="HPL_pdlaswp00N.html">HPL_pdlaswp00N</A>,
82 <A HREF="HPL_pdlaswp00T.html">HPL_pdlaswp00T</A>,
83 <A HREF="HPL_pdlaswp01N.html">HPL_pdlaswp01N</A>,
84 <A HREF="HPL_pdlaswp01T.html">HPL_pdlaswp01T</A>.
85
86 </BODY>
87 </HTML>