MPI_ISCATTERV(3)

📦

version 5.0.10 · 4 versions

💖 MPI_ISCATTERV(3) — the Open MPI glow-up 💖

okay bestie, sit down, hydrate, because we're about to talk about three of the most underrated queen bees in the whole Open MPI friend group: MPI_Scatterv, MPI_Iscatterv, and MPI_Scatterv_init. they take one buffer and scatter it in parts to everybody in the group. it's giving "sharing is caring" but make it high-performance. 🎀

✨ NAME ✨

MPI_Scatterv, MPI_Iscatterv, MPI_Scatterv_init — Scatters a buffer in parts to all tasks in a group.

no cap, that's literally the whole job description and they eat every time. 💅

🌈 SYNTAX 🌈

🦋 C Syntax

#include <mpi.h>

int MPI_Scatterv(const void *sendbuf, const int sendcounts[], const int displs[],
     MPI_Datatype sendtype, void *recvbuf, int recvcount,
     MPI_Datatype recvtype, int root, MPI_Comm comm)

int MPI_Iscatterv(const void *sendbuf, const int sendcounts[], const int displs[],
     MPI_Datatype sendtype, void *recvbuf, int recvcount,
     MPI_Datatype recvtype, int root, MPI_Comm comm, MPI_Request *request)

int MPI_Scatterv_init(const void *sendbuf, const int sendcounts[], const int displs[],
     MPI_Datatype sendtype, void *recvbuf, int recvcount,
     MPI_Datatype recvtype, int root, MPI_Comm comm, MPI_Info info, MPI_Request *request)

🦋 Fortran Syntax

USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_SCATTERV(SENDBUF, SENDCOUNTS, DISPLS, SENDTYPE, RECVBUF,
             RECVCOUNT, RECVTYPE, ROOT, COMM, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER SENDCOUNTS(*), DISPLS(*), SENDTYPE
     INTEGER RECVCOUNT, RECVTYPE, ROOT, COMM, IERROR

MPI_ISCATTERV(SENDBUF, SENDCOUNTS, DISPLS, SENDTYPE, RECVBUF,
             RECVCOUNT, RECVTYPE, ROOT, COMM, REQUEST, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER SENDCOUNTS(*), DISPLS(*), SENDTYPE
     INTEGER RECVCOUNT, RECVTYPE, ROOT, COMM, REQUEST, IERROR

MPI_SCATTERV_INIT(SENDBUF, SENDCOUNTS, DISPLS, SENDTYPE, RECVBUF,
             RECVCOUNT, RECVTYPE, ROOT, COMM, INFO, REQUEST, IERROR)
     <type>  SENDBUF(*), RECVBUF(*)
     INTEGER SENDCOUNTS(*), DISPLS(*), SENDTYPE
     INTEGER RECVCOUNT, RECVTYPE, ROOT, COMM, INFO, REQUEST, IERROR

🦋 Fortran 2008 Syntax

USE mpi_f08
MPI_Scatterv(sendbuf, sendcounts, displs, sendtype, recvbuf, recvcount,
             recvtype, root, comm, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN) :: sendbuf
     TYPE(*), DIMENSION(..) :: recvbuf
     INTEGER, INTENT(IN) :: sendcounts(*), displs(*), recvcount, root
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Iscatterv(sendbuf, sendcounts, displs, sendtype, recvbuf, recvcount,
             recvtype, root, comm, request, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN), ASYNCHRONOUS :: sendcounts(*), displs(*)
     INTEGER, INTENT(IN) :: recvcount, root
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     TYPE(MPI_Request), INTENT(OUT) :: request
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_scatterv_init(sendbuf, sendcounts, displs, sendtype, recvbuf, recvcount,
             recvtype, root, comm, info, request, ierror)
     TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
     TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
     INTEGER, INTENT(IN), ASYNCHRONOUS :: sendcounts(*), displs(*)
     INTEGER, INTENT(IN) :: recvcount, root
     TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
     TYPE(MPI_Comm), INTENT(IN) :: comm
     TYPE(MPI_Info), INTENT(IN) :: info
     TYPE(MPI_Request), INTENT(OUT) :: request
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

like… three whole flavors. vanilla, async, and persistent. pick your fighter. 🔥

🌸 INPUT PARAMETERS 🌸

here's the cast of characters, bestie. memorize them, they show up in every single one of these calls. 👑

sendbuf — Address of send buffer (choice, significant only at root). yeah, only root actually cares about this one, everyone else is like "not my business 💅".
sendcounts — Integer array (of length group size) specifying the number of elements to send to each processor. one number per bestie, so nobody gets left out. 🫶
displs — Integer array (of length group size). Entry i specifies the displacement (relative to sendbuf) from which to take the outgoing data to process i. it's the "where do I grab your slice from" map. 🗺️
sendtype — Datatype of send buffer elements (handle).
recvcount — Number of elements in receive buffer (integer).
recvtype — Datatype of receive buffer elements (handle).
root — Rank of sending process (integer). the main character, obviously. 👑
comm — Communicator (handle).
info — Info (handle, persistent only). persistent only, don't try it anywhere else bestie, i'm serious.

💜 OUTPUT PARAMETERS 💜

recvbuf — Address of receive buffer (choice).
request — Request (handle, non-blocking only). non-blocking only, so that's MPI_Iscatterv territory.
ierror — Fortran only: Error status (integer).

🎀 DESCRIPTION 🎀

okay so. MPI_Scatterv is the inverse operation to MPI_Gatherv. it's literally the uno reverse card of the friend group. 🔄

MPI_Scatterv extends the functionality of MPI_Scatter by allowing a varying count of data to be sent to each process, since sendcounts is now an array. so like… nobody's forced to get the same size gift bag anymore. 💝 and it also gives you more flexibility as to where the data is taken from on the root, thanks to the new argument, displs. iconic behavior, honestly.

the outcome is as if the root executed n send operations,

MPI_Send(sendbuf + displs[i] * extent(sendtype),
         sendcounts[i], sendtype, i, ...);

// and each process executed a receive,

MPI_Recv(recvbuf, recvcount, recvtype, root, ...)

The send buffer is ignored for all nonroot processes. not me pointing out that if you're not root, sendbuf is just… decorative. 🙈

The type signature implied by sendcount[i], sendtype at the root must be equal to the type signature implied by recvcount, recvtype at process i (however, the type maps may be different). This implies that the amount of data sent must be equal to the amount of data received, pairwise between each process and the root. Distinct type maps between sender and receiver are still allowed. so like, the vibes of the data have to match, even if the layouts are doing their own little thing. ✨

All arguments to the function are significant on process root, while on other processes, only arguments recvbuf, recvcount, recvtype, root, comm are significant. The arguments root and comm must have identical values on all processes. fr fr, if your root and comm don't match across processes, that's a whole situation. 😭

⚠️ The specification of counts, types, and displacements should not cause any location on the root to be read more than once.

like pls don't double-dip on the root's buffer, it's not a sample table at the mall. 🛑

🌟 Example 1 🌟

The reverse of Example 5 in the MPI_Gatherv manpage. We have a varying stride between blocks at sending (root) side, at the receiving side we receive 100 - i elements into the ith column of a 100 x 150 C array at process i.

MPI_Comm comm;
int gsize,recvarray[100][150],*rptr;
int root, *sendbuf, myrank, bufsize, *stride;
MPI_Datatype rtype;
int i, *displs, *scounts, offset;
...
MPI_Comm_size( comm, &gsize);
MPI_Comm_rank( comm, &myrank );

stride = (int *)malloc(gsize*sizeof(int));
...
/* stride[i] for i = 0 to gsize-1 is set somehow
 * sendbuf comes from elsewhere
 */
...
displs = (int *)malloc(gsize*sizeof(int));
scounts = (int *)malloc(gsize*sizeof(int));
offset = 0;
for (i=0; i<gsize; ++i) {
    displs[i] = offset;
    offset += stride[i];
    scounts[i] = 100 - i;
}
/* Create datatype for the column we are receiving
 */
MPI_Type_vector( 100-myrank, 1, 150, MPI_INT, &rtype);
MPI_Type_commit( &rtype );
rptr = &recvarray[0][myrank];
MPI_Scatterv(sendbuf, scounts, displs, MPI_INT,
             rptr, 1, rtype, root, comm);

the 100-myrank and the 150 stride? chef's kiss. this is the kind of code that makes you feel like a genius at 2am. 🧚

🌟 Example 2 🌟

The reverse of Example 1 in the MPI_Gather manpage. The root process scatters sets of 100 ints to the other processes, but the sets of 100 are stride ints apart in the sending buffer. Requires use of MPI_Scatterv, where stride >= 100.

MPI_Comm comm;
int gsize,*sendbuf;
int root, rbuf[100], i, *displs, *scounts;

...

MPI_Comm_size(comm, &gsize);
sendbuf = (int *)malloc(gsize*stride*sizeof(int));
...
displs = (int *)malloc(gsize*sizeof(int));
scounts = (int *)malloc(gsize*sizeof(int));
for (i=0; i<gsize; ++i) {
   displs[i] = i*stride;
   scounts[i] = 100;
}
MPI_Scatterv(sendbuf, scounts, displs, MPI_INT,
             rbuf, 100, MPI_INT, root, comm);

see how clean that is? displs[i] = i*stride — one line, all the power. 💅

🫶 USE OF IN-PLACE OPTION 🫶

When the communicator is an intracommunicator, you can perform a scatter operation in-place (the output buffer is used as the input buffer). Use the variable MPI_IN_PLACE as the value of the root process recvbuf. In this case, recvcount and recvtype are ignored, and the root process sends no data to itself.

so basically the root is like "i'm keeping my own snacks 🍓" and just… doesn't send to itself. iconic. selfish but in an efficient way.

Note that MPI_IN_PLACE is a special kind of value; it has the same restrictions on its use as MPI_BOTTOM.

🚨 Because the in-place option converts the receive buffer into a send-and-receive buffer, a Fortran binding that includes INTENT must mark these as INOUT, not OUT.

friendly reminder that Fortran is out here doing the most with its INTENT bookkeeping. don't fight it, just mark it INOUT and move on. 💜

🌈 WHEN COMMUNICATOR IS AN INTER-COMMUNICATOR 🌈

When the communicator is an inter-communicator, the root process in the first group sends data to all processes in the second group. The first group defines the root process. That process uses MPI_ROOT as the value of its root argument. The remaining processes use MPI_PROC_NULL as the value of their root argument. All processes in the second group use the rank of that root process in the first group as the value of their root argument. The receive buffer argument of the root process in the first group must be consistent with the receive buffer argument of the processes in the second group.

okay so group 1 has the root, and that root pops off with the data. 🎤 the rest of group 1 is just standing there like 🧍 with MPI_PROC_NULL. group 2 has to reference the rank of that root — in the first group — as their root value. and the recv buffers on both sides have to be consistent. lowkey a lot of bookkeeping but you got this. 🔥

💖 ERRORS 💖

okay bestie so here's the tea on what happens when stuff goes wrong: almost all MPI routines hand you back an error value. C routines? that's the return result of the function. Fortran routines? it's chillin' in the last argument. classic.

Before that error value even makes it back to you, the current MPI error handler that's associated with the communication object (so like, a communicator, a window, or a file) gets called. now — if there's no communication object associated with the MPI call, then the call is considered attached to MPI_COMM_SELF and it will call the associated MPI error handler. got it?

and when MPI_COMM_SELF is not initialized (like: before MPI_Init / MPI_Init_thread, after MPI_Finalize, or when you're out here using the Sessions Model exclusively 😌), the error raises the initial error handler. that initial error handler can be changed by calling MPI_Comm_set_errhandler on MPI_COMM_SELF when you're using the World model, or with the mpi_initial_errhandler CLI argument to mpiexec, or the info key to MPI_Comm_spawn / MPI_Comm_spawn_multiple.

and if no other appropriate error handler has been set? then MPI_ERRORS_RETURN gets called for MPI I/O functions, and MPI_ERRORS_ABORT gets called for all other MPI functions. periodt.

🎀 the three predefined error handlers Open MPI ships with

MPI_ERRORS_ARE_FATAL — causes the program to abort all connected MPI processes. yeah. everyone. gone. 💀
MPI_ERRORS_ABORT — an error handler that can be invoked on a communicator, window, file, or session. when it's called on a communicator, it acts as if MPI_Abort was called on that communicator. if called on a window or file, it acts as if MPI_Abort was called on a communicator containing the group of processes in the corresponding window or file. and if called on a session, it aborts only the local process. (main character energy, honestly.)
MPI_ERRORS_RETURN — returns an error code to the application. ty bestie, very demure, very mindful.

🦋 cooking up your own error handlers

MPI applications can also implement their own error handlers by calling:

MPI_Comm_create_errhandler then MPI_Comm_set_errhandler
MPI_File_create_errhandler then MPI_File_set_errhandler
MPI_Session_create_errhandler then MPI_Session_set_errhandler or at MPI_Session_init
MPI_Win_create_errhandler then MPI_Win_set_errhandler

⚠️ real talk: note that MPI does not guarantee that an MPI program can continue past an error. like. no cap. don't build your whole house on that foundation bestie 🫶

See the MPI man page for a full list of MPI error codes. 👑

See the Error Handling section of the MPI-3.1 standard for more information. 📚

👑 SEE ALSO 👑

MPI_Gather
MPI_Gatherv
MPI_Scatter

📜 Copyright

2003-2026, The Open MPI Community

okay that's the whole page bestie, you ate that 💅 this was served up by Open MPI, dated Sep 10, 2026, and it's giving MPI_ISCATTERV(3) realness. go forth and scatter (the vector kind) 🌸✨

📦 where this page comes from

ubuntu stonking: openmpi-doc 5.0.10-6 · other
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash on 2026-11-03: a language model's rewrite of the original, under its terms