1
.TH pvm_send 1 "May 1998" "Scilab // Group" "Scilab function"
4
pvm_send - immediately sends (or multicast) data.
7
[info] = pvm_send(tids,buff,msgtag)
12
: row of integers, contains the task IDs of the tasks to be sent to.
17
msgtag : integer, message tag supplied by the user. msgtag should be
18
>= 0. It allows the user's program to distinguish between different
22
: integer, status code returned by the routine.
23
Values less than zero indicate an error.
27
sends (or multicasts) a message
28
to the PVM process identified in the
30
array. Note that the message is not sent to the caller even if listed
33
is used to label the content of the message.
34
If pvm_send is successful,
36
will be 0. If some error occurs then
40
The pvm_send routine is asynchronous. Computation on the sending
41
processor resumes as soon as the message is safely on its way to the
42
receiving processor. This is in contrast to synchronous
43
communication, during which computation on the sending processor halts
44
until the matching receive is executed by the receiving processor.
46
If a multicast is performed, pvm_send first determines which other
47
pvmds contain the specified tasks. Then passes the message to these
48
pvmds which in turn distribute the message to their local tasks
49
without further network traffic.
51
The PVM model guarantees the following about message order. If task 1
52
sends message A to task 2, then task 1 sends message B to task 2,
53
message A will arrive at task 2 before message B. Moreover, if both
54
messages arrive before task 2 does a receive, then a wildcard receive
55
will always return message A.
57
Terminating a PVM task immediately after sending a message or messages
58
from it may result in those messages being lost. To be sure, always
59
call pvm_exit() before stopping.
64
deff('[x]=f(y)','x = 1/y')
65
info = pvm_send([262150, 262152], A(1:2:5,:), 100)
66
pvm_send(262146,f,200)