MPI_CART_SHIFT(3)

📦

version 5.0.10 · 7 versions

💫 MPI_CART_SHIFT(3) 💫

okay bestie, this is the Open MPI manual page for MPI_Cart_shift — and no, it's not about shifting your vibes (tho honestly? we could all use a shift). It's about ranks. Sit down, grab a strawberry 🍓, and let me tell you about the most underrated little helper in Cartesian-land.

🎀 NAME 🎀

MPI_Cart_shift — Returns the shifted source and destination ranks, given a shift direction and amount.

Like literally that's it. You give it a direction and a distance, it gives you back "who's behind you" and "who's ahead of you." Matchmaker behavior honestly. 🫶

✨ SYNTAX ✨

🌸 C Syntax

#include <mpi.h>

int MPI_Cart_shift(MPI_Comm comm, int direction, int disp,
    int *rank_source, int *rank_dest)

🌸 Fortran Syntax

USE MPI
! or the older form: INCLUDE 'mpif.h'

MPI_CART_SHIFT(COMM, DIRECTION, DISP, RANK_SOURCE,
        RANK_DEST, IERROR)
    INTEGER COMM, DIRECTION, DISP, RANK_SOURCE
    INTEGER RANK_DEST, IERROR

🌸 Fortran 2008 Syntax

USE mpi_f08

MPI_Cart_shift(comm, direction, disp, rank_source, rank_dest, ierror)
    TYPE(MPI_Comm), INTENT(IN) :: comm
    INTEGER, INTENT(IN) :: direction, disp
    INTEGER, INTENT(OUT) :: rank_source, rank_dest
    INTEGER, OPTIONAL, INTENT(OUT) :: ierror

💖 INPUT PARAMETERS 💖

comm : Communicator with Cartesian structure (handle). i.e. the group chat that actually has a layout. 👑
direction : Coordinate dimension of shift (integer). Which axis are we sliding along, bestie?
disp (Displacement ( > 0: upward shift, < 0: downward shift)) (integer). Positive = upward, negative = downward. That's the whole vibe check right there. ⬆️⬇️

🌈 OUTPUT PARAMETERS 🌈

rank_source : Rank of source process (integer). Your upstream bestie.
rank_dest : Rank of destination process (integer). Your downstream bestie.
ierror : Fortran only: Error status (integer).

🦋 DESCRIPTION 🦋

So here's the tea. 🍵

If the process topology is a Cartesian structure, an MPI_Sendrecv operation is likely to be used along a coordinate direction to perform a shift of data. As input, MPI_Sendrecv takes the rank of a source process for the receive, and the rank of a destination process for the send. If the function MPI_Cart_shift is called for a Cartesian process group, it provides the calling process with the above identifiers, which then can be passed to MPI_Sendrecv. The user specifies the coordinate direction and the size of the step (positive or negative). The function is local.

Local, babes. It doesn't talk to anybody else. It just does math on your own rank. Iconic, low-maintenance, no drama. 💅

The direction argument indicates the dimension of the shift, i.e., the coordinate whose value is modified by the shift. The coordinates are numbered from 0 to ndims-1, where ndims is the number of dimensions.

Note: The direction argument is in the range [0, n-1] for an n-dimensional Cartesian mesh. Stay in your lane literally — go outside that range and it's not gonna be cute.

Depending on the periodicity of the Cartesian group in the specified coordinate direction, MPI_Cart_shift provides the identifiers for a circular or an end-off shift. In the case of an end-off shift, the value MPI_PROC_NULL may be returned in rank_source or rank_dest, indicating that the source or the destination for the shift is out of range.

Translation:</purgle> if your topology wraps around, you get a cute little circle 🔄. If it doesn't wrap, somebody at the end just gets MPI_PROC_NULL — which is MPI-speak for "nobody's there, ghost them, the sendrecv will handle it." 😭

🧚 Example 🧚

The communicator, comm, has a two-dimensional, periodic, Cartesian topology associated with it. A two-dimensional array of REALs is stored one element per process, in variable A. One wishes to skew this array, by shifting column i (vertically, i.e., along the column) by i steps.

! find process rank
    CALL MPI_COMM_RANK(comm, rank, ierr)
! find Cartesian coordinates
    CALL MPI_CART_COORDS(comm, rank, maxdims, coords, ierr)
! compute shift source and destination
    CALL MPI_CART_SHIFT(comm, 0, coords(2), source, dest, ierr)
! skew array
    CALL MPI_SENDRECV_REPLACE(A, 1, MPI_REAL, dest, 0, source, 0, comm, status,
                              ierr)

See how she computed the shift amount straight from coords(2)? Main character energy. Skewing a whole array by asking the topology who lives next door. Obsessed. ✨

⭐ NOTE ⭐

In Fortran, the dimension indicated by DIRECTION = i has DIMS(i+1) nodes, where DIMS is the array that was used to create the grid. In C, the dimension indicated by direction = i is the dimension specified by dims[i].

bestie this is the classic off-by-one trap 🪤 — Fortran arrays start at 1, C arrays start at 0, and the direction argument follows the language you're in. Don't mix them up, fr fr.

💀 ERRORS 💀

Almost all MPI routines return an error value; C routines as the return result of the function and Fortran routines in the last argument.

Before the error value is returned, the current MPI error handler associated with the communication object (e.g., communicator, window, file) is called. 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. When MPI_COMM_SELF is not initialized (i.e., before MPI_Init/MPI_Init_thread, after MPI_Finalize, or when using the Sessions Model exclusively) the error raises the initial error handler. The initial error handler can be changed 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. If no other appropriate error handler has been set, then the MPI_ERRORS_RETURN error handler is called for MPI I/O functions and the MPI_ERRORS_ABORT error handler is called for all other MPI functions.

Open MPI includes three predefined error handlers that can be used:

MPI_ERRORS_ARE_FATAL Causes the program to abort all connected MPI processes.
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.
MPI_ERRORS_RETURN Returns an error code to the application.

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

Note that MPI does not guarantee that an MPI program can continue past an error.

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.

okay real talk: MPI_ERRORS_ARE_FATAL kills everything, and MPI does not promise your program survives an error. So handle your errors, bestie. Self-care is checking ierror. 🩷

👑 Copyright 👑

2003-2026, The Open MPI Community

That's the whole page, bestie!! 🎀 Go forth and shift those ranks, you gorgeous Cartesian icon. Sending you good vibes along direction 0, displacement +1 💖🦋🌸

— with love, the Open MPI crew, page dated Sep 10, 2026, from the MPI_CART_SHIFT(3) chapter of the Open MPI manual 🌈

📦 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