Lockless Inc. MPI

Lockless Inc
Purchase »
- Linux
- Windows
- Developers
Benchmarks
Installation
Articles
Technical
Downloads
Documentation
What's New
About Us
Help

NAME

MPI_Irsend - Perform a nonblocking "ready" send of a buffer

SYNOPSIS

#include <mpi.h> 
int MPI_Irsend(void *buf , int count , MPI_Datatype datatype , int dest , int tag , MPI_Comm comm ,
	 MPI_Request *rq ); 
#include <pmpi.h> 
int PMPI_Irsend(void *buf , int count , MPI_Datatype datatype , int dest , int tag , MPI_Comm comm ,
	 MPI_Request *rq ); 
#include <mpi.h> 
MPI::Request MPI::Comm::Irsend(const void *buf , int count , const MPI::Datatype &datatype ,
	 int dest , int tag ) const; 
INCLUDE 'mpif.h' 
MPI_IRSEND(buf , count , datatype , dest , tag , comm , rq ,  ierr ) 
<type>  buf (*) 
INTEGER  count ,  datatype ,  dest ,  tag ,  comm ,  rq ,  ierr

INPUT PARAMETERS

buf - buffer (array)

count - count of elements in buffer (integer)

datatype - type of elements in buffer (handle)

dest - destination rank (integer)

tag - communication tag (integer)

comm - communicator for message (handle)

OUTPUT PARAMETER

rq - request handle for this message (handle)

DESCRIPTION

The MPI_Irsend() function is used to send a non-blocking message to the process with rank dest in the communicator comm , using the tag envelope, with the assumption that a receive at the destination has already been posted. The send consists of count elements of type datatype in an array specified by buf . This buffer may not be altered until the request rq is known to be completed via a function such as MPI_Wait().

If rq is NULL then some other form of synchronization is required for the sender to know that the buffer may be reused. However, this lack of need to directly notify the sender may lead to improved performance.

If the destination is MPI_PROC_NULL then no send is posted, and MPI_SUCCESS is returned.

The C++ version of this function passes the new request as the return value.

The communicator must be a valid one (not MPI_COMM_NULL). PMPI_Irsend() is the profiling version of this function.

ERRORS

All MPI routines except for MPI_Wtime and MPI_Wtick return an error code. The the current MPI error handler is invoked if the return value is not MPI_SUCCESS. The default error handler aborts, but this may be changed with by using the MPI_Errhandler_set() function. The predefined error handler MPI_ERRORS_RETURN may be used to cause error values to be returned instead. Note that MPI does not guarentee that an MPI program can continue past an error. In this implementation, all errors except MPI_ERR_INTERN or MPI_ERR_OTHER should always be recoverable.

In C, the error code is passed as the return value. In FORTRAN, all functions have an parameter ierr which returns the error code. MPI C++ functions do not directly return an error code. However, C++ users may want to use the MPI::ERRORS_THROW_EXCEPTIONS handler. This will throw an MPI::Exception with the corresponding error code. To prevent exceptions from being raised from within C and Fortran code, they will see all error return values as MPI_ERR_PENDING when this handler is chosen. In this implementation, call MPI::throw_exception() to throw the correct exception if this occurs.

MPI_SUCCESS - No error;

MPI_ERR_PENDING - Pending exception;

MPI_ERR_COMM - Invalid communicator;

MPI_ERR_COUNT - Invalid element count;

MPI_ERR_BUFFER - Invalid buffer;

MPI_ERR_TAG - Invalid tag;

MPI_ERR_TYPE - Invalid data type;

MPI_ERR_RANK - Invalid destination;