1 | .\"Copyright 2006-2008 Sun Microsystems, Inc. |
---|
2 | .\" Copyright (c) 1996 Thinking Machines Corporation |
---|
3 | .TH MPI_Type_struct 3 "Dec 08, 2009" "1.4" "Open MPI" |
---|
4 | .SH NAME |
---|
5 | \fBMPI_Type_struct\fP \- Creates a \fIstruct\fP data type -- use of this routine is deprecated. |
---|
6 | |
---|
7 | .SH SYNTAX |
---|
8 | .ft R |
---|
9 | .SH C Syntax |
---|
10 | .nf |
---|
11 | #include <mpi.h> |
---|
12 | int MPI_Type_struct(int \fIcount\fP, int\fI *array_of_blocklengths\fP, |
---|
13 | MPI_Aint\fI *array_of_displacements\fP, MPI_Datatype\fI *array_of_types\fP, |
---|
14 | MPI_Datatype\fI *newtype\fP) |
---|
15 | |
---|
16 | .SH Fortran Syntax |
---|
17 | .nf |
---|
18 | INCLUDE 'mpif.h' |
---|
19 | MPI_TYPE_STRUCT(\fICOUNT, ARRAY_OF_BLOCKLENGTHS, |
---|
20 | ARRAY_OF_DISPLACEMENTS, ARRAY_OF_TYPES, |
---|
21 | NEWTYPE, IERROR\fP) |
---|
22 | INTEGER \fICOUNT, ARRAY_OF_BLOCKLENGTHS(*)\fP |
---|
23 | INTEGER \fIARRAY_OF_DISPLACEMENTS(*)\fP |
---|
24 | INTEGER \fIARRAY_OF_TYPES(*), NEWTYPE, IERROR\fP |
---|
25 | |
---|
26 | |
---|
27 | .SH INPUT PARAMETERS |
---|
28 | .ft R |
---|
29 | .TP 1i |
---|
30 | count |
---|
31 | Number of blocks (integer) also number of entries in arrays |
---|
32 | array_of_types, array_of_displacements, and array_of_blocklengths. |
---|
33 | .TP 1i |
---|
34 | array_of_blocklengths |
---|
35 | Number of elements in each block (array). |
---|
36 | .TP 1i |
---|
37 | array_of_displacements |
---|
38 | Byte displacement of each block (array). |
---|
39 | .TP 1i |
---|
40 | array_of_types |
---|
41 | Type of elements in each block (array of handles to datatype objects). |
---|
42 | .sp |
---|
43 | |
---|
44 | .SH OUTPUT PARAMETERS |
---|
45 | .ft R |
---|
46 | .TP 1i |
---|
47 | newtype |
---|
48 | New datatype (handle). |
---|
49 | .ft R |
---|
50 | .TP 1i |
---|
51 | IERROR |
---|
52 | Fortran only: Error status (integer). |
---|
53 | |
---|
54 | .SH DESCRIPTION |
---|
55 | .ft R |
---|
56 | Note that use of this routine is \fIdeprecated\fP as of MPI-2. Use MPI_Type_create_struct instead. |
---|
57 | .sp |
---|
58 | This deprecated routine is not available in C++. |
---|
59 | .sp |
---|
60 | MPI_Type_struct is the most general type constructor. It further generalizes MPI_Type_hindexed in that it allows each block to consist of replications of different datatypes. |
---|
61 | .sp |
---|
62 | \fBExample:\fP Let type1 have type map |
---|
63 | .nf |
---|
64 | |
---|
65 | {(double, 0), (char, 8)} |
---|
66 | |
---|
67 | .fi |
---|
68 | with extent 16. Let B = (2, 1, 3), D = (0, 16, 26), and T = (MPI_FLOAT, type1, MPI_CHAR). Then a call to MPI_Type_struct(3, B, D, T, newtype) returns a datatype with type map |
---|
69 | .nf |
---|
70 | |
---|
71 | {(float, 0), (float,4), (double, 16), (char, 24), |
---|
72 | (char, 26), (char, 27), (char, 28)} |
---|
73 | |
---|
74 | .fi |
---|
75 | That is, two copies of MPI_FLOAT starting at 0, followed by one copy of type1 starting at 16, followed by three copies of MPI_CHAR, starting at 26. (We assume that a float occupies 4 bytes.) |
---|
76 | .sp |
---|
77 | For more information, see section 3.12.1 of the MPI-1.1 Standard. |
---|
78 | |
---|
79 | .SH NOTES |
---|
80 | If an upperbound is set explicitly by using the MPI datatype MPI_UB, the corresponding index must be positive. |
---|
81 | .sp |
---|
82 | The MPI-1 Standard originally made vague statements about padding and alignment; this was intended to allow the simple definition of structures that could be sent with a count greater than one. For example, |
---|
83 | .nf |
---|
84 | struct {int a; char b;} foo; |
---|
85 | .fi |
---|
86 | may have |
---|
87 | .nf |
---|
88 | sizeof(foo) = sizeof(int) + sizeof(char); |
---|
89 | .fi |
---|
90 | defining the extent of a datatype as including an epsilon, which would have allowed an implementation to make the extent an MPI datatype for this structure equal to 2*sizeof(int). However, since different systems might define different paddings, a clarification to the standard made epsilon zero. Thus, if you define a structure datatype and wish to send or receive multiple items, you should explicitly include an MPI_UB entry as the last member of the structure. For example, the following code can be used for the structure foo: |
---|
91 | .nf |
---|
92 | |
---|
93 | blen[0] = 1; indices[0] = 0; oldtypes[0] = MPI_INT; |
---|
94 | blen[1] = 1; indices[1] = &foo.b - &foo; oldtypes[1] = MPI_CHAR; |
---|
95 | blen[2] = 1; indices[2] = sizeof(foo); oldtypes[2] = MPI_UB; |
---|
96 | MPI_Type_struct( 3, blen, indices, oldtypes, &newtype ); |
---|
97 | |
---|
98 | .fi |
---|
99 | |
---|
100 | .SH ERRORS |
---|
101 | Almost all MPI routines return an error value; C routines as the value of the function and Fortran routines in the last argument. C++ functions do not return errors. If the default error handler is set to MPI::ERRORS_THROW_EXCEPTIONS, then on error the C++ exception mechanism will be used to throw an MPI:Exception object. |
---|
102 | .sp |
---|
103 | Before the error value is returned, the current MPI error handler is |
---|
104 | called. By default, this error handler aborts the MPI job, except for I/O function errors. The error handler may be changed with MPI_Comm_set_errhandler; the predefined error handler MPI_ERRORS_RETURN may be used to cause error values to be returned. Note that MPI does not guarantee that an MPI program can continue past an error. |
---|
105 | |
---|
106 | .SH SEE ALSO |
---|
107 | .ft R |
---|
108 | .sp |
---|
109 | MPI_Type_create_struct |
---|
110 | .br |
---|
111 | MPI_Type_create_hindexed |
---|
112 | .br |
---|
113 | |
---|