MPI_IGATHERV(3)

📦
version 5.0.11 · 16 versions
, in arch rolling

🌸 MPI_IGATHERV(3) — the Open MPI glow-up edition 💅

ok so bestie, picture this: it's giving everybody sends the root a different amount of stuff and the root has to cope. This is the page for MPI_Gatherv, MPI_Igatherv and MPI_Gatherv_init — the varying-amounts version of gather. No cap, it's iconic. ✨

💖 NAME

MPI_Gatherv, MPI_Igatherv, MPI_Gatherv_init — gathers varying amounts of data from all processes to the root process 🫶

✨ SYNTAX ✨

🎀 C Syntax

#include <mpi.h>

int MPI_Gatherv(const void *sendbuf, int sendcount, MPI_Datatype sendtype,
    void *recvbuf, const int recvcounts[], const int displs[], MPI_Datatype recvtype,
    int root, MPI_Comm comm)

int MPI_Igatherv(const void *sendbuf, int sendcount, MPI_Datatype sendtype,
    void *recvbuf, const int recvcounts[], const int displs[], MPI_Datatype recvtype,
    int root, MPI_Comm comm, MPI_Request *request)

int MPI_Gatherv_init(const void *sendbuf, int sendcount, MPI_Datatype sendtype,
    void *recvbuf, const int recvcounts[], const int displs[], 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_GATHERV(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNTS,
        DISPLS, RECVTYPE, ROOT, COMM, IERROR)
    <type>  SENDBUF(*), RECVBUF(*)
    INTEGER SENDCOUNT, SENDTYPE, RECVCOUNTS(*), DISPLS(*)
    INTEGER RECVTYPE, ROOT, COMM, IERROR

MPI_IGATHERV(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNTS,
        DISPLS, RECVTYPE, ROOT, COMM, REQUEST, IERROR)
    <type>  SENDBUF(*), RECVBUF(*)
    INTEGER SENDCOUNT, SENDTYPE, RECVCOUNTS(*), DISPLS(*)
    INTEGER RECVTYPE, ROOT, COMM, REQUEST, IERROR

MPI_GATHERV_INIT(SENDBUF, SENDCOUNT, SENDTYPE, RECVBUF, RECVCOUNTS,
        DISPLS, RECVTYPE, ROOT, COMM, INFO, REQUEST, IERROR)
    <type>  SENDBUF(*), RECVBUF(*)
    INTEGER SENDCOUNT, SENDTYPE, RECVCOUNTS(*), DISPLS(*)
    INTEGER RECVTYPE, ROOT, COMM, INFO, REQUEST, IERROR

🌈 Fortran 2008 Syntax

USE mpi_f08

MPI_Gatherv(sendbuf, sendcount, sendtype, recvbuf, recvcounts, displs,
        recvtype, root, comm, ierror)
    TYPE(*), DIMENSION(..), INTENT(IN) :: sendbuf
    TYPE(*), DIMENSION(..) :: recvbuf
    INTEGER, INTENT(IN) :: sendcount, recvcounts(*), displs(*), root
    TYPE(MPI_Datatype), INTENT(IN) :: sendtype, recvtype
    TYPE(MPI_Comm), INTENT(IN) :: comm
    INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Igatherv(sendbuf, sendcount, sendtype, recvbuf, recvcounts, displs,
        recvtype, root, comm, request, ierror)
    TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
    TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
    INTEGER, INTENT(IN) :: sendcount, root
    INTEGER, INTENT(IN), ASYNCHRONOUS :: recvcounts(*), displs(*)
    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_Gatherv_init(sendbuf, sendcount, sendtype, recvbuf, recvcounts, displs,
        recvtype, root, comm, info, request, ierror)
    TYPE(*), DIMENSION(..), INTENT(IN), ASYNCHRONOUS :: sendbuf
    TYPE(*), DIMENSION(..), ASYNCHRONOUS :: recvbuf
    INTEGER, INTENT(IN) :: sendcount, root
    INTEGER, INTENT(IN), ASYNCHRONOUS :: recvcounts(*), displs(*)
    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

🧚 INPUT PARAMETERS 🧚

  • sendbuf — starting address of the send buffer (that's a choice, bestie).
  • sendcount — number of elements in the send buffer (an integer).
  • sendtype — datatype of the send buffer elements (a handle).
  • recvcounts — integer array (of length group size) holding the number of elements that are received from each process. Highkey only matters at the root. 👑
  • displs — integer array (of length group size). Entry i says the displacement relative to recvbuf at which to place the incoming data from process i. Again, significant only at root. 📍
  • recvtype — datatype of the recv buffer elements (significant only at root), a handle.
  • root — rank of the receiving process (an integer). Main character energy fr fr.
  • comm — the communicator (a handle).
  • info — info (a handle, persistent only).

🍓 OUTPUT PARAMETERS 🍓

  • recvbuf — address of the receive buffer (a choice, and yes, significant only at root 🙈).
  • request — request (a handle, non-blocking only).
  • ierror — Fortran only: error status (an integer).

💅 DESCRIPTION 💅

okay so bestie, MPI_Gatherv is literally MPI_Gather but with ✨main character energy✨ — it levels up the whole thing by letting each process send a varying count of data, because recvcounts is now an array instead of one sad little number. 🎀 And it's also way more flexible about where all that data lands on the root, thanks to a brand-new argument called displs. iconique behavior only.

The outcome is as if every process (root included, no exceptions, no cap) sends a message to the root:

MPI_Send(sendbuf, sendcount, sendtype, root, ...);

and the root executes n receives, like so:

MPI_Recv(recvbuf + disp[i] * extent(recvtype), recvcounts[i],
         recvtype, i, ...);

Messages get placed in the receive buffer of the root process in rank order — so the data from process j goes into the jth portion of the receive buffer recvbuf on process root. The jth portion of recvbuf starts at offset displs[j] elements (in terms of recvtype) into recvbuf. it's giving organized. it's giving assigned seating. 💅

fr fr: the receive buffer is ignored for all nonroot processes. like, they don't even look at it. 🙈

The type signature implied by sendcount, sendtype on process i must be equal to the type signature implied by recvcounts[i], recvtype at the root. 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 though — see Example 2 below, it slays. 🦋

All arguments to the function are significant on process root, while on other processes, only the arguments sendbuf, sendcount, sendtype, root, comm are significant. And listen — root and comm must have identical values on all processes. no funny business, bestie. 😤

🚨 The specification of counts, types, and displacements should not cause any location on the root to be written more than once. Such a call is erroneous. that's the danger zone, don't go there. 💀

🌸 Example 1 🌸

Now have each process send 100 ints to root, but place each set (of 100) stride ints apart at receiving end. Use MPI_Gatherv and the displs argument to achieve this effect. Assume stride >= 100.

MPI_Comm comm;
int gsize, sendarray[100];
int root, *rbuf, stride;
int *displs, i, rcounts;
...

MPI_Comm_size(comm, &gsize);
rbuf = (int)malloc(gsize * stride * sizeof(int));
displs = (int)malloc(gsize * sizeof(int));
rcounts = (int )malloc(gsize * sizeof(int));

for (i=0; i<gsize; ++i) {
  displs[i] = i * stride;
  rcounts[i] = 100;
}
MPI_Gatherv(sendarray, 100, MPI_INT, rbuf, rcounts, displs, MPI_INT,
            root, comm);
⚠️ Note that the program is erroneous if stride < 100. like literally don't. 😭

🦋 Example 2 🦋

Same as Example 1 on the receiving side, but send the 100 ints from the 0th column of a 100 150 int array, in C.

MPI_Comm comm;
int gsize, sendarray[100][150];
int root, *rbuf, stride;
MPI_Datatype stype;
int displs,i, rcounts;
...

MPI_Comm_size(comm, &gsize);
rbuf = (int )malloc(gsize * stride * sizeof(int));
displs = (int)malloc(gsize * sizeof(int));
rcounts = (int )malloc(gsize * sizeof(int));

for (i=0; i<gsize; ++i) {
  displs[i] = i * stride;
  rcounts[i] = 100;
}

// Create datatype for 1 column of array
MPI_Type_vector(100, 1, 150, MPI_INT, &stype);
MPI_Type_commit( &stype );
MPI_Gatherv(sendarray, 1, stype, rbuf, rcounts, displs, MPI_INT,
            root, comm);

⭐ Example 3 ⭐

Process i sends (100-i) ints from the ith column of a 100 x 150 int array, in C. It is received into a buffer with stride, as in the previous two examples.

MPI_Comm comm;
int gsize, sendarray[100][150], *sptr;
int root, *rbuf, stride, myrank;
MPI_Datatype stype;
int displs, i, rcounts;
...

MPI_Comm_size(comm, &gsize);
MPI_Comm_rank( comm, &myrank );
rbuf = (int)malloc(gsize * stride * sizeof(int));
displs = (int)malloc(gsize * sizeof(int));
rcounts = (int )malloc(gsize * sizeof(int));

for (i=0; i<gsize; ++i) {
  displs[i] = i * stride;
  rcounts[i] = 100-i; // note change from previous example
}

// Create datatype for the column we are sending
MPI_Type_vector(100-myrank, 1, 150, MPI_INT, &stype);
MPI_Type_commit( &stype );
// sptr is the address of start of "myrank" column
sptr = &sendarray[0][myrank];
MPI_Gatherv(sptr, 1, stype, rbuf, rcounts, displs, MPI_INT,
            root, comm);
💖 Note that a different amount of data is received from each process. yes queen, variety. 🍓

💜 Example 4 💜

Same as Example 3, but done in a different way at the sending end. We create a datatype that causes the correct striding at the sending end so that we read a column of a C array. lowkey genius tbh 🧚

MPI_Comm comm;
int gsize, sendarray[100][150], *sptr;
int root, *rbuf, stride, myrank, disp[2], blocklen[2];
MPI_Datatype stype, type[2];
int displs, i, rcounts;
...

MPI_Comm_size(comm, &gsize);
MPI_Comm_rank(comm, &myrank );
rbuf = (int )alloc(gsize * stride * sizeof(int));
displs = (int )malloc(gsize * sizeof(int));
rcounts = (int)malloc(gsize * sizeof(int));

for (i=0; i<gsize; ++i) {
  displs[i] = i* stride;
  rcounts[i] = 100-i;
}
// Create datatype for one int, with extent of entire row
disp[0] = 0;
disp[1] = 150 * sizeof(int);
type[0] = MPI_INT;
type[1] = MPI_UB;
blocklen[0] = 1;
blocklen[1] = 1;

MPI_Type_struct( 2, blocklen, disp, type, &stype );
MPI_Type_commit(&stype );
sptr = &sendarray[0][myrank];
MPI_Gatherv(sptr, 100-myrank, stype, rbuf, rcounts, displs, MPI_INT,
            root, comm);

🌈 Example 5 🌈

Same as Example 3 at sending side, but at receiving side we make the stride between received blocks vary from block to block. chaotic? maybe. iconic? absolutely. 👑

MPI_Comm comm;
int gsize, sendarray[100][150], *sptr;
int root, *rbuf, *stride, myrank, bufsize;
MPI_Datatype stype;
int *displs, i, *rcounts, offset;
...

MPI_Comm_size( comm, &gsize);
MPI_Comm_rank( comm, &myrank );
de = (int )malloc(gsize * sizeof(int));
...
// stride[i] for i = 0 to gsize-1 is set somehow

// set up displs and rcounts vectors first
displs = (int)malloc(gsize * sizeof(int));
rcounts = (int )malloc(gsize * sizeof(int));
offset = 0;

for (i=0; i<gsize; ++i) {
  displs[i] = offset;
  offset += stride[i];
  rcounts[i] = 100-i;
}

// the required buffer size for rbuf is now easily obtained
bufsize = displs[gsize-1]+rcounts[gsize-1];
rbuf = (int )malloc(bufsize * sizeof(int));
// Create datatype for the column we are sending
MPI_Type_vector(100-myrank, 1, 150, MPI_INT, &stype);
MPI_Type_commit( &stype );
sptr = &sendarray[0][myrank];
MPI_Gatherv(sptr, 1, stype, rbuf, rcounts, displs, MPI_INT,
            root, comm);

🎀 Example 6 🎀

Process i sends num ints from the ith column of a 100 x 150 int array, in C. The complicating factor is that the various values of num are not known to root, so a separate gather must first be run to find these out. The data is placed contiguously at the receiving end. two-step slay 💅✨

MPI_Comm comm;
int gsize, sendarray[100][150], *sptr;
int root, *rbuf, stride, myrank, disp[2], blocklen[2];
MPI_Datatype stype,types[2];
int *displs, i, *rcounts, num;
...

MPI_Comm_size( comm, &gsize);
MPI_Comm_rank( comm, &myrank );

// First, gather nums to root
rcounts = (int )malloc(gsize * sizeof(int));
MPI_Gather( &num, 1, MPI_INT, rcounts, 1, MPI_INT, root, comm);
// root now has correct rcounts, using these we set
// displs[] so that data is placed contiguously (or concatenated) at receive end

displs = (int)malloc(gsize * sizeof(int));
displs[0] = 0;
for (i=1; i<gsize; ++i) {
  displs[i] = displs[i-1]+rcounts[i-1];
}

// And, create receive buffer
rbuf = (int *)malloc(gsize * (displs[gsize-1]+rcounts[gsize-1]) * sizeof(int));
// Create datatype for one int, with extent of entire row
disp[0] = 0;
disp[1] = 150 * sizeof(int);
type[0] = MPI_INT;
type[1] = MPI_UB;
blocklen[0] = 1;
blocklen[1] = 1;
MPI_Type_struct(2, blocklen, disp, type, &stype );
MPI_Type_commit( &stype );
sptr = &sendarray[0][myrank];
MPI_Gatherv(sptr, num, stype, rbuf, rcounts, displs, MPI_INT, root, comm);

🫶 USE OF IN-PLACE OPTION 🫶

The in-place option operates in the same way as it does for MPI_Gather. When the communicator is an intracommunicator, you can perform a gather 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 sendbuf. In this case, sendcount and sendtype are ignored, and the contribution of the root process to the gathered vector is assumed to already be in the correct place in the receive buffer. no extra work needed, she's already seated. 💅

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

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. ✨

🦋 WHEN COMMUNICATOR IS AN INTER-COMMUNICATOR 🦋

When the communicator is an inter-communicator, the root process in the first group gathers data from all the 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. And the send buffer argument of the processes in the first group must be consistent with the receive buffer argument of the root process in the second group. it's giving cross-group teamwork, we love to see it 🫶👑

💀 ERRORS 💀 (aka: what happens when the vibes go sideways)

okay bestie, listen — this is the part nobody reads until something breaks, and then suddenly it's the most important page in the world 😭

Almost all MPI routines return an error value, fr fr. How they hand it to you depends on your language era:

  • In C, it's the return result of the function — like, the function's whole personality is "here's your error code."
  • In Fortran, it comes back in the last argument (classic Fortran, always putting things at the end 💅).

🫶 so what actually happens before you get that error value?

Before the error value is returned, the current MPI error handler associated with the communication object is called. And what's a communication object, you ask? Think e.g., communicator, window, file — the whole little family.

But here's the plot twist 🌈: if no communication object is associated with the MPI call, then the call is considered attached to MPI_COMM_SELF and will call the associated MPI error handler. It's giving "no plus-one? guess I'm going with myself" energy.

And when MPI_COMM_SELF is not initialized — and that means:

  • before MPI_Init/MPI_Init_thread
  • after MPI_Finalize
  • or when using the Sessions Model exclusively

…then the error raises the initial error handler. Scary? A lil. 🙈

✨ how to change that initial error handler ✨

You can swap it out, and here's exactly how, no cap:

  • by calling MPI_Comm_set_errhandler on MPI_COMM_SELF when using the World model, or
  • the mpi_initial_errhandler CLI argument to mpiexec, or
  • info key to MPI_Comm_spawn/MPI_Comm_spawn_multiple.

And if no other appropriate error handler has been set, then the defaults kick in like this:

  • MPI_ERRORS_RETURN gets called for MPI I/O functions 📁
  • MPI_ERRORS_ABORT gets called for all other MPI functions 💀

🎀 the three predefined error handlers Open MPI ships with 🎀

Open MPI includes three predefined error handlers that can be used — collect them all like stickers:

  • MPI_ERRORS_ARE_FATAL — Causes the program to abort all connected MPI processes. Like, everyone. Gone. Periodt. 😭
  • MPI_ERRORS_ABORT — An error handler that can be invoked on a communicator, window, file, or session. When called on a communicator, it acts as if MPI_Abort was called on that communicator. If called on a window or file, acts as if MPI_Abort was called on a communicator containing the group of processes in the corresponding window or file. If called on a session, aborts only the local process. iykyk 👑
  • MPI_ERRORS_RETURN — Returns an error code to the application. The chill one. The one that lets you handle it yourself 🫶

💅 and you can build your own, bestie

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

main character energy, honestly — write your own handler, make it your brand ✨

⚠️ real talk warning: note that MPI does not guarantee that an MPI program can continue past an error. So like… don't build your whole coping strategy around surviving it, okay? 🥲

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. (yes, the actual standard, go read it, I believe in you 📚)

🌸 SEE ALSO 🌸

  • MPI_Gather — go give it a read, it's the whole reason you're here 🦋

2003-2026, The Open MPI Community — absolute icons, we stan 👑

and that's the whole page, bestie!! 🍓✨ MPI_IGATHERV(3), dated Sep 11, 2026, brought to you by the Open MPI Community — go forth and gatherv (in a non-blocking way) 💖🧚‍♀️🌈

📦 where this page comes from