MPI_RGET_ACCUMULATE(3)

💅 MPI_RGET_ACCUMULATE(3) — Open MPI, bestie edition 🎀

okay so this is from the Open MPI manual and it's literally about combining your origin buffer with a target buffer and handing you back the target's old value. two-for-one slay, no cap ✨

---

🌸 NAME 🌸

MPI_Get_accumulate, MPI_Rget_accumulate — combines the contents of the origin buffer with that of a target buffer and returns the target buffer value. it's giving "read it and write it, periodt" 💅

---

✨ SYNTAX ✨

🧚 C Syntax

#include <mpi.h>

int MPI_Get_accumulate(const void *origin_addr, int origin_count,
    MPI_Datatype origin_datatype, void *result_addr,
    int result_count, MPI_Datatype result_datatype,
    int target_rank, MPI_Aint target_disp, int target_count,
    MPI_Datatype target_datatype, MPI_Op op, MPI_Win win)

int MPI_Rget_accumulate(const void *origin_addr, int origin_count,
    MPI_Datatype origin_datatype, void *result_addr,
    int result_count, MPI_Datatype result_datatype,
    int target_rank, MPI_Aint target_disp, int target_count,
    MPI_Datatype target_datatype, MPI_Op op, MPI_Win win,
    MPI_Request *request)

🦋 Fortran Syntax

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

MPI_GET_ACCUMULATE(ORIGIN_ADDR, ORIGIN_COUNT, ORIGIN_DATATYPE, RESULT_ADDR,
        RESULT_COUNT, RESULT_DATATYPE, TARGET_RANK, TARGET_DISP, TARGET_COUNT,
        TARGET_DATATYPE, OP, WIN, IERROR)
    <type> ORIGIN_ADDR, RESULT_ADDR(*)
    INTEGER(KIND=MPI_ADDRESS_KIND) TARGET_DISP
    INTEGER ORIGIN_COUNT, ORIGIN_DATATYPE, TARGET_COUNT, TARGET_DATATYPE,
        TARGET_RANK, TARGET_COUNT, TARGET_DATATYPE, OP, WIN, IERROR

MPI_RGET_ACCUMULATE(ORIGIN_ADDR, ORIGIN_COUNT, ORIGIN_DATATYPE, RESULT_ADDR,
        RESULT_COUNT, RESULT_DATATYPE, TARGET_RANK, TARGET_DISP, TARGET_COUNT,
        TARGET_DATATYPE, OP, WIN, REQUEST, IERROR)
    <type> ORIGIN_ADDR, RESULT_ADDR(*)
    INTEGER(KIND=MPI_ADDRESS_KIND) TARGET_DISP
    INTEGER ORIGIN_COUNT, ORIGIN_DATATYPE, TARGET_COUNT, TARGET_DATATYPE,
        TARGET_RANK, TARGET_COUNT, TARGET_DATATYPE, OP, WIN, REQUEST, IERROR

🌈 Fortran 2008 Syntax

USE mpi_f08

MPI_Get_accumulate(origin_addr, origin_count, origin_datatype, result_addr,
        result_count, result_datatype, target_rank, target_disp,
        target_count, target_datatype, op, win, ierror)
    TYPE(*), DIMENSION(..), INTENT(IN) :: origin_addr
    TYPE(*), DIMENSION(..) :: result_addr
    INTEGER, INTENT(IN) :: origin_count, result_count, target_rank, target_count
    TYPE(MPI_Datatype), INTENT(IN) :: origin_datatype, target_datatype, result_datatype
    INTEGER(KIND=MPI_ADDRESS_KIND), INTENT(IN) :: target_dist
    TYPE(MPI_Op), INTENT(IN) :: op
    TYPE(MPI_Win), INTENT(IN) :: win
    INTEGER, OPTIONAL, INTENT(OUT) :: ierror

MPI_Rget_accumulate(origin_addr, origin_count, origin_datatype,
        result_addr, result_count, result_datatype, target_rank,
        target_disp, target_count, target_datatype, op, win, request,
        ierror)
    TYPE(*), DIMENSION(..), INTENT(IN) :: origin_addr
    TYPE(*), DIMENSION(..) :: result_addr
    INTEGER, INTENT(IN) :: origin_count, result_count, target_rank, target_count
    TYPE(MPI_Datatype), INTENT(IN) :: origin_datatype, target_datatype, result_datatype
    INTEGER(KIND=MPI_ADDRESS_KIND), INTENT(IN) :: target_dist
    TYPE(MPI_Op), INTENT(IN) :: op
    TYPE(MPI_Win), INTENT(IN) :: win
    TYPE(MPI_Request), INTENT(OUT) :: request
    INTEGER, OPTIONAL, INTENT(OUT) :: ierror

🫶 yes bestie, target_dist in the Fortran 2008 signatures is spelled exactly like that in the page — retype it at your own risk, iykyk 💀

---

💖 INPUT PARAMETERS 💖

• origin_addr — initial address of buffer (choice). the starting point, the main character 💅
• origin_count — number of entries in buffer (nonnegative integer). no negatives, we don't do that here 🙈
• origin_datatype — data type of each buffer entry (handle).
• result_addr — initial address of result buffer (choice). this is where the before pic of the target lands 📸
• result_count — number of entries in result buffer (nonnegative integer).
• result_datatype — data type of each result buffer entry (handle).
• target_rank — rank of target (nonnegative integer). who's getting the update 👑
• target_disp — displacement from start of window to beginning of target buffer (nonnegative integer).
• target_count — number of entries in target buffer (nonnegative integer).
• target_datatype — data type of each entry in target buffer (handle).
• op — reduce operation (handle). the actual vibe of the combine ✨
• win — window object (handle).

---

🎀 OUTPUT PARAMETER 🎀

• MPI_Rget_accumulate — RMA request
• ierror — Fortran only: error status (integer).

---

👑 DESCRIPTION 👑

MPI_Get_accumulate is a one-sided MPI communication function, and it's basically a two-in-one. it takes the contents of the origin buffer — that's origin_addr, origin_count, and origin_datatype — and combines it with the buffer specified by the arguments target_count and target_datatype, at offset target_disp, in the target window specified by target_rank and win, using the operation op. and then it also hands you back, in the result buffer result_addr, the contents of the target buffer before the accumulation. like, it read it first. icon behaviour 💅

any of the predefined operations for MPI_Reduce, plus MPI_NO_OP, can be used here. user-defined functions cannot be used — sorry to everyone with a custom function, not today 😭. so like, if op is MPI_SUM, each element of the origin buffer gets added to the corresponding element in the target, replacing the former value in the target.

every datatype argument has to be a predefined data type or a derived data type where all basic components are of the same predefined data type. both datatype arguments must be constructed from the same predefined data type. the operation op applies to elements of that predefined type. also, the target_datatype argument must not specify overlapping entries, and the target buffer must fit in the target window. no overlap, no overflow, no drama 🚫

there's a new predefined operation, MPI_REPLACE. it corresponds to the associative function f(a, b) =b; that is, the current value in the target memory is replaced by the value supplied by the origin. straight-up swap, slay.

and another new predefined operation, MPI_NO_OP. it corresponds to the associative function f(a, b) = a; that is, the current value in the target memory is returned in the result buffer at the origin and no operation is performed on the target buffer. so it's a read. that's it. that's the function 🫶

MPI_Rget_accumulate is similar to MPI_Get_accumulate, except that it allocates a communication request object and associates it with the request handle (the argument request) that can be used to wait or test for completion. completion of an MPI_Rget_accumulate operation indicates that the data is available in the result buffer and the origin buffer is free to be updated. it does not indicate that the operation has been completed at the target window. read that again bestie, it matters ⚠️

---

🦋 NOTES 🦋

the generic functionality of MPI_Get_accumulate might limit the performance of fetch-and-increment or fetch-and-add calls that might be supported by special hardware operations. lowkey the general version can be slower than the special one 😔. MPI_Fetch_and_op thus allows for a fast implementation of a commonly used subset of the functionality of MPI_Get_accumulate.

MPI_Get is a special case of MPI_Get_accumulate, with the operation MPI_NO_OP. but heads up — note, however, that MPI_Get and MPI_Get_accumulate have different constraints on concurrent updates. don't assume they're interchangeable, fr fr 🙈

it's the user's responsibility to guarantee that, when using the accumulate functions, the target displacement argument is such that accesses to the window are properly aligned according to the data type arguments in the call to the MPI_Get_accumulate function. alignment is your job, not the library's, bestie ✨

---

💀 ERRORS 💀

almost all MPI routines return an error value. C routines return it as the return result of the function, and Fortran routines put it 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 you can use:

• MPI_ERRORS_ARE_FATAL — causes the program to abort all connected MPI processes. going out with a bang, literally 💀
• 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. 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

note that MPI does not guarantee that an MPI program can continue past an error. no promises, 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_Put MPI_Reduce

---

🍓 COPYRIGHT 🍓

2003-2026, The Open MPI Community

---

and that's the whole page, bestie — go forth and accumulate 💜✨

⭐ Open MPI · page dated Sep 11, 2026 · 2003-2026, The Open MPI Community · mwah, stan the manual writers 🎀💅