MPI_WIN_CREATE_DYNAMIC(3)

📦
version 5.0.11 · 16 versions
, in arch rolling

💅 MPI_WIN_CREATE_DYNAMIC(3) 💅

okay bestie, so this is a page from the Open MPI manual and it's literally all about MPI_Win_create_dynamic — the one-sided MPI call that hands you back a window object for RMA operations. no cap, it's a whole vibe. 🦋

✨ 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, bestie (handle). 👑
  • comm — the communicator, also a handle (handle). 🫶

🌸 OUTPUT PARAMETERS 🌸

  • win — the window object that the call returns to you (handle). it's giving main character energy ✨
  • ierror — Fortran only: error status (integer). 💅

🦋 DESCRIPTION 🦋

MPI_Win_create_dynamic is a one-sided MPI communication collective call — meaning every process in the group of comm is in on it together, fr fr. It returns a window object with no memory attached that all these processes can then use to do RMA operations. iconic.

A window created with MPI_Win_create_dynamic requires the target_disp argument for all RMA communication functions to be the actual address at the target. like, the actual address. no shortcuts. 😭

The following info keys are supported:

  • no_locks — if you set this to true, the implementation may assume that the local window is never locked (i.e., nobody's calling MPI_Win_lock or MPI_Win_lock_all on it). Setting this value if only active synchronization is used may let the implementation enable certain optimizations. slay, free performance. 🔥
  • accumulate_ordering — by default, accumulate operations 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 like... no ordering of accumulate operations is guaranteed. 🙈 The key can also be a comma-separated list of required orderings consisting of rar, war, raw, and waw 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 — so yes, chaos can be faster. ⭐
  • accumulate_ops — if set to same_op, the implementation will assume that all concurrent accumulate calls to the same target address will use the same operation. If set to same_op_no_op, then the implementation will assume that all concurrent accumulate calls to the same target address will use the same operation or MPI_NO_OP. The default is same_op_no_op. noted. 🎀

💜 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 some communication. like, you literally have to tell the other side yourself. iykyk. 💀

Dynamic memory does not have any disp_unit associated with it, and it requires correct offset calculations with proper type handling. so be careful, bestie — do the math right. 🧚

🌈 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. classic.

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. big yikes but okay. 😭

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, 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. 🫶
  • MPI_ERRORS_RETURN — returns an error code to the application. polite of it. ✨

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. so like, don't count on it, 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_Win_attach
  • MPI_Win_detach
  • MPI_Get_address

2003-2026, The Open MPI Community

and that's the whole page, bestie — no cap, you now know MPI_Win_create_dynamic better than most people. mwah 💋 this was the Open MPI version, dated Sep 11, 2026 — go forth and create those dynamic windows. 🎀💖✨

📦 where this page comes from