MPI_WIN_CREATE_DYNAMIC(3)

📦
version 5.0.10 · 7 versions
, in ubuntu stonking

💖 MPI_Win_create_dynamic — the Open MPI glow-up ✨

okay bestie, this is MPI_Win_create_dynamic, and it's a one-sided MPI call that hands you back a window object for RMA operations. no memory attached, all vibes. 🦋

✨ SYNTAX ✨

C Syntax 💅

#include <mpi.h>

MPI_Win_create_dynamic(MPI_Info info, MPI_Comm comm, MPI_Win *win)

Fortran Syntax 🌸

USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_WIN_CREATE_DYNAMIC(INFO, COMM, WIN, IERROR)
     INTEGER INFO, COMM, WIN, IERROR

Fortran 2008 Syntax 🎀

USE mpi_f08
MPI_Win_create_dynamic(info, comm, win, ierror)
     TYPE(MPI_Info), INTENT(IN) :: info
     TYPE(MPI_Comm), INTENT(IN) :: comm
     TYPE(MPI_Win), INTENT(OUT) :: win
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

🌈 INPUT PARAMETERS 🌈

  • info — the Info argument, aka a handle. it's giving "settings container," iykyk.
  • comm — the communicator, a handle. your little friend group. 👑

💜 OUTPUT PARAMETERS 💜

  • win — the window object this call hands back to you (a handle). your new main character artifact. ✨
  • ierror — Fortran only: the error status, as an integer. 😭

🫶 DESCRIPTION 🫶

MPI_Win_create_dynamic is a one-sided MPI communication collective call, and it gets executed by every process in the group of comm. it returns a window object with no memory attached that all those processes can then use to do RMA operations. collective, bestie — everybody's in on it. 💅

here's the tea: a window made with MPI_Win_create_dynamic requires the target_disp argument, for all RMA communication functions, to be the actual address at the target. no offset games, the real address, periodt.

these are the info keys that are supported, bestie, write them down 📝:

  • no_locks — if this is set to true, then the implementation is allowed to assume that the local window is never locked (not by a call to MPI_Win_lock and not by MPI_Win_lock_all). setting this value when only active synchronization is happening may let the implementation turn on certain optimizations. free real estate for the optimizer. 🔥
  • accumulate_ordering — by default, accumulate operations going from one initiator to one target on the same window memory location are strictly ordered. if the info key accumulate_ordering is set to none, then no ordering of accumulate operations is guaranteed. the key can also be a comma-separated list of the required orderings, made up of rar, war, raw, and waw — which stand for read-after-read, write-after-read, read-after-write, and write-after-write, respectively. looser ordering constraints are likely to result in improved performance. speedy but lawless, pick your fighter. 🏎️
  • accumulate_ops — if this is set to same_op, the implementation will assume that all concurrent accumulate calls to the same target address use the same operation. if it's set to same_op_no_op, then the implementation will assume that all concurrent accumulate calls to the same target address use the same operation or MPI_NO_OP. the default is same_op_no_op.

🌟 NOTES 🌟

since dynamically attaching memory to a window is a local operation, you have to communicate the actual address at the target using MPI_Get_address and, like, some communication. yes bestie, actual communication, you have to tell people. 🗣️

dynamic memory doesn't have any disp_unit associated with it, and it requires correct offset calculations with proper type handling. math is math, do it right. 🙈

💀 ERRORS 💀

almost all MPI routines return an error value. C routines give it back as the return result of the function, and Fortran routines put it in the last argument. classic.

before that error value is returned, the current MPI error handler associated with the communication object (e.g., communicator, window, file) gets called. if there is no communication object 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 an 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 you can use, and yes they're iconic:

  • MPI_ERRORS_ARE_FATAL — causes the program to abort all connected MPI processes. everyone goes down together, no cap. 💀
  • 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, it 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, it aborts only the local process. it's giving "choose your own apocalypse." 😭
  • MPI_ERRORS_RETURN — returns an error code to the application. the chill one. 🫶

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
heads up bestie: note that MPI does not guarantee that an MPI program can continue past an error. like, it might just be over. 🙈

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.

also see, since we're being thorough and organized:

  • MPI_Win_attach
  • MPI_Win_detach
  • MPI_Get_address

2003-2026, The Open MPI Community. yeah, the whole era. 💖

and that's the whole page, bestie — MPI_WIN_CREATE_DYNAMIC(3) from the Open MPI manuals, dated Sep 10, 2026. slay, go create a dynamic window, and remember: communicate the address. fr fr. ✨🎀💅

📦 where this page comes from