MPI_TYPE_UB(3)

💅 MPI_TYPE_UB(3) — Open MPI, bestie edition 💅

okay so this is the MPI_TYPE_UB page from Open MPI, and before you even scroll, know this: she's deprecated. Like, still iconic, still works, but the cool kids moved on. Let's get into it 🫶

---

✨ NAME ✨

MPI_Type_ub — Returns the upper bound of a datatype — and heads up, this routine is deprecated in favor of MPI_Type_get_extent. No cap.

---

🌸 SYNTAX 🌸

💖 C Syntax

#include <mpi.h>

int MPI_Type_ub(MPI_Datatype datatype, MPI_Aint *displacement)

💖 Fortran Syntax

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

MPI_TYPE_UB(DATATYPE, DISPLACEMENT, IERROR)
     INTEGER DATATYPE, DISPLACEMENT, IERROR

---

🎀 INPUT PARAMETER 🎀

• datatype — the datatype, as a handle. she's the main character here 💅

---

🦋 OUTPUT PARAMETERS 🦋

• displacement — displacement of the upper bound from the origin, in bytes (integer). yes bestie, bytes.
• ierror — Fortran only: the error status (integer). classic Fortran, always in the back seat with the error code 🙈

---

🌈 DESCRIPTION 🌈

okay so first thing's first: use of this routine is deprecated as of MPI-2. Please use MPI_Type_get_extent instead. i'm not saying it, the standard is saying it 😭

So what does MPI_Type_ub actually do? It returns the upper bound of a data type. That upper bound will differ from zero if the type was constructed using MPI_UB. And she's thorough — the upper bound takes into account any alignment considerations. slay ✨

Now the fun part: the "pseudo-datatypes," MPI_LB and MPI_UB, can be used, respectively, to mark the upper bound (or the lower bound) of a datatype. These pseudo-datatypes occupy no space — literally extent (MPI_LB) = extent (MPI_UB) =0. zero, nada, nothing 🫥 They do not affect the size or count of a datatype, and they do not affect the context of a message created with this datatype. However — and this is the plot twist — they do affect the definition of the extent of a datatype and, therefore, affect the outcome of a replication of this datatype by a datatype constructor. so like, invisible but extremely powerful. iykyk 👑

In general, if

Typemap = {(type(0), disp(0)), ..., (type(n-1), disp(n-1))}

then the lower bound of Typemap is defined to be

                     (min(j) disp(j)                          if no entry has
       lb(Typemap) = (                                        basic type lb
                     (min(j) {disp(j) such that type(j) = lb} otherwise

Similarly, the upper bound of Typemap is defined to be

                     (max(j) disp(j) + sizeof(type(j) = lb}   if no entry has
       ub(Typemap) = (                                        basic type ub
                     (max(j) {disp(j) such that type(j) = ub} otherwise

Then

extent(Typemap) = ub(Typemap) - lb(Typemap)

And finally, if type(i) requires alignment to a byte address that is a multiple of k(i), then e is the least nonnegative increment needed to round extent(Typemap) to the next multiple of max(i) k(i). that's the whole lore, bestie, write it down 📝

---

💀 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. yes, the Fortran girlies get it handed to them at the end, we been knew.

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 — lemme introduce the squad 💅

• MPI_ERRORS_ARE_FATAL — Causes the program to abort all connected MPI processes. dramatic? yes. effective? also yes 💀
• 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. she's versatile 🦋
• MPI_ERRORS_RETURN — Returns an error code to the application. the chill one. the "here's your problem back, handle it yourself" 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

⚠️ Note that MPI does not guarantee that an MPI program can continue past an error. so like, don't assume you can just vibes your way through a failure, 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_Type_get_extent — the glow-up version of this whole page ✨

---

🍓 COPYRIGHT 🍓

2003-2026, The Open MPI Community

---

and that's the whole page, bestie 💖 no notes, she ate. this came straight from the Open MPI girlies — the copyright is 2003-2026, The Open MPI Community, and this edition is dated Sep 11, 2026. go forth and compute responsibly 🧚‍♀️✨