MPI_Gatherv function
Gathers variable data from all members of a group to one member. The MPI_Gatherv function adds flexibility to the MPI_Gather function by allowing a varying count of data from each process.
Syntax
int MPIAPI MPI_Gatherv(
_In_ void *sendbuf,
int sendcount,
MPI_Datatype sendtype,
_Out_opt_ void *recvbuf,
_In_opt_ int *recvcounts[],
_In_opt_ int *displs[],
MPI_Datatype recvtype,
int root,
MPI_Comm comm
);
Parameters
sendbuf [in]
The handle to a buffer that contains the data to be sent to the root process.If the comm parameter references an intracommunicator, you can specify an in-place option by specifying MPI_IN_PLACE in all processes. The sendcount and sendtype parameters are ignored. Each process enters data in the corresponding receive buffer element. The nth process sends data to the nth element of the receive buffer. The data that is sent by the root process is assumed to be in the correct place in the receive buffer.
sendcount
The number of elements in the send buffer. If sendcount is zero, the data part of the message is empty.sendtype
The data type of each element in the buffer.recvbuf [out, optional]
The handle to a buffer on the root process that contains the data that is received from each process, including data that is sent by the root process. This parameter is significant only at the root process. The recvbuf parameter is ignored for all non-root processes.recvcounts[] [in, optional]
The number of elements that is received from each process. Each element in the array corresponds to the rank of the sending process. If the count is zero, the data part of the message is empty. This parameter is significant only at the root process.displs[] [in, optional]
The location, relative to the recvbuf parameter, of the data from each communicator process. The data that is received from process j is placed into the receive buffer of the root process offset displs[j] elements from the sendbuf pointer.In the recvbuf, recvcounts, and displs parameter arrays, the nth element of each array refers to the data that is received from the nth communicator process.
This parameter is significant only at the root process.
recvtype
The data type of each element in buffer. This parameter is significant only at the root process.root
The rank of the receiving process within the specified communicator.comm
The MPI_Comm communicator handle.
Return value
Returns MPI_SUCCESS on success. Otherwise, the return value is an error code.
In Fortran, the return value is stored in the IERROR parameter.
Fortran
MPI_GATHERV(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNTS, DISPLS, RECVTYPE, ROOT, COMM, IERROR)
<type> SENDBUF(*), RECVBUF(*)
INTEGER SENDCOUNT, SENDTYPE, RECVCOUNTS(*), DISPLS(*), RECVTYPE, ROOT, COMM, IERROR
Remarks
All the function parameters are significant on the root process, only the sendbuf, sendcount, sendtype, root, and comm are significant on the other processes. The root and comm parameters must be identical on all processes.
Generally, derived data types are allowed for both the sendtype and recvtype parameters. The type signature as specified by the sendtype and recvtype parameters on each process must be equal to the type signature of the recvcount and sendcount parameters on the root process. The amount of data that is sent must be equal to the amount of data that is received between the root process and each individual process. Distinct type maps between sender and receiver are still allowed.
The specification of counts and types should not cause any location on the root to be written more than one time. Such a call is erroneous.
If the comm parameter references an intracommunicator, all processes send the contents of its send buffer to the root process. The root process receives the messages and stores them in rank order. The outcome is as if each of the n processes in the group that are executed a call to MPI_Send(sendbuf, sendcount, sendtype, root, …); and the root had executed n calls to MPI_Recv(recvbuf + i*recvcount*extent(recvtype), recvcount, recvtype, i, …). The value of extent(recvtype) is obtained by using the MPI_Type_get_extent function. An alternative description of the function is that the n messages that are sent by the processes in the group are concatenated in rank order, and the resulting message is received by the root as if by a call to MPI_RECV(recvbuf, recvcountn, recvtype, ...). The receive buffer is ignored for all non-root processes.
If the comm parameter references an intercommunicator, then the call involves all processes in the intercommunicator, but with one group, group A, that defines the root process. All processes in the other group, group B, set the same value in root parameter, that is, the rank of the root process in group A. The root process sets the value MPI_ROOT in the root parameter. All other processes in group A set the value MPI_PROC_NULL in the root parameter. Data is broadcast from the root process to all processes in group B. The buffer parameters of the processes in group B must be consistent with the buffer parameter of the root process.
Requirements
Product |
HPC Pack 2012 MS-MPI Redistributable Package, HPC Pack 2008 R2 MS-MPI Redistributable Package, HPC Pack 2008 MS-MPI Redistributable Package or HPC Pack 2008 Client Utilities |
Header |
Mpi.h; Mpif.h |
Library |
Msmpi.lib |
DLL |
Msmpi.dll |