MPI_COMM_FREE_KEYVAL(3)

💖 MPI_COMM_FREE_KEYVAL 💖 (the Open MPI manual page, bestie)

Okay so welcome to MPI_COMM_FREE_KEYVAL(3) from the Open MPI manual — this one is short and sweet and honestly kind of iconic once it clicks. Let's get into it ✨

---

🌸 SYNOPSIS 🌸

MPI_Comm_free_keyval — frees an attribute key for a communicator cache attribute. That's the whole vibe. 🎀

---

🦋 SYNTAX 🦋

C Syntax

#include <mpi.h>

int MPI_Comm_free_keyval(int *comm_keyval)

Fortran Syntax

USE MPI
! or the older form: INCLUDE 'mpif.h'
MPI_COMM_FREE_KEYVAL(COMM_KEYVAL, IERROR)
     INTEGER COMM_KEYVAL, IERROR

Fortran 2008 Syntax

USE mpi_f08
MPI_Comm_free_keyval(comm_keyval, ierror)
     INTEGER, INTENT(INOUT) :: comm_keyval
     INTEGER, OPTIONAL, INTENT(OUT) :: ierror

---

🍓 INPUT/OUTPUT PARAMETER 🍓

• comm_keyval — the key you're handing back to the universe 💫 it's in/out because (spoiler, see DESCRIPTION) it gets set to MPI_KEYVAL_INVALID when this call is done with it.

---

🧚 OUTPUT PARAMETER 🧚

• ierror — Fortran only: error status (integer). Yeah, Fortran besties get their error report in the last argument, not as a return value. That's just how it is 💅

---

🌈 DESCRIPTION 🌈

So, MPI_Comm_free_keyval frees an extant attribute key. Like, an attribute key that exists. It sets the value of keyval to MPI_KEYVAL_INVALID. ✨

And here's the part people get confused about, so listen up 🫶 — it is not erroneous to free an attribute key that is in use. Nope. Not a problem. Because the actual free does not transpire until after all references (in other communicators on the process) to the key have been freed. So the key is out there living its life until every last reference is gone, no cap.

Those references need to be explicitly freed by the program, either:

• via calls to MPI_Comm_delete_attr, which frees one attribute instance, or
• via calls to MPI_Comm_free, which frees all attribute instances associated with the freed communicator.

This call is identical to the call MPI_Keyval_free, but it's needed to match the communicator-specific creation function introduced in the MPI-2 standard. And heads up: the use of MPI_Keyval_free is deprecated. 🙈 So use this one, mmkay?

---

🫧 NOTES 🫧

Key values are global — they can be used with any and all communicators. Every single one. Go off, key value 👑

---

💀 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. 😭 dramatic but effective.
• 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. Like, do not assume you can just keep going, bestie. That's a real caveat, fr fr.

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_Comm_create_keyval

---

👑 COPYRIGHT 👑

2003-2026, The Open MPI Community

---

and that's the whole page, bestie 🌈✨ — signed off with love from the Open MPI crew, dated Sep 11, 2026, page MPI_COMM_FREE_KEYVAL(3). go free those keys 💅🎀