MPI_Ialltoallv(3)

💖 MPI_Ialltoallv — the manual page (MPI, section 3) 💖

okay bestie, gather round, because this one is a bop. We're in the MPI manual, section 3, and this page is all about MPI_Ialltoallv. Periodt. ✨

---

🌈 NAME 🌈

MPI_Ialltoallv — sends data from all to all processes in a nonblocking way; each process may send a different amount of data and provide displacements for the input and output data.

no cap, that's the whole vibe: everyone talks to everyone, nobody waits around, and every process gets to be a lil different from the others. main character energy fr fr. 👑

---

✨ SYNOPSIS ✨

int MPI_Ialltoallv(const void *sendbuf, const int sendcounts[],
const int sdispls[], MPI_Datatype sendtype, void *recvbuf,
const int recvcounts[], const int rdispls[],
MPI_Datatype recvtype, MPI_Comm comm, MPI_Request *request)

int MPI_Ialltoallv_c(const void *sendbuf, const MPI_Count sendcounts[],
const MPI_Aint sdispls[], MPI_Datatype sendtype,
void *recvbuf, const MPI_Count recvcounts[],
const MPI_Aint rdispls[], MPI_Datatype recvtype,
MPI_Comm comm, MPI_Request *request)

two flavors, both iconic: the classic one with int counts and the _c one that levels up to MPI_Count and MPI_Aint for when your numbers are just built different. 💅

---

🎀 INPUT PARAMETERS 🎀

• sendbuf — starting address of send buffer (choice). it's giving "where my stuff lives".
• sendcounts — non-negative integer array (of length group size) specifying the number of elements to send to each rank (non-negative integer). literally one entry per bestie you're sending to.
• sdispls — integer array (of length group size). Entry j specifies the displacement (relative to sendbuf) from which to take the outgoing data destined for process j (integer). so entry j = where to scoop from when you're packing for rank j. 🎁
• sendtype — data type of send buffer elements (handle).
• recvcounts — non-negative integer array (of length group size) specifying the number of elements that can be received from each rank (non-negative integer). can be received, bestie — it's a capacity, not a promise.
• rdispls — integer array (of length group size). Entry i specifies the displacement (relative to recvbuf) at which to place the incoming data from process i (integer). entry i = the lil spot on the shelf reserved for rank i's package. 📦
• recvtype — data type of receive buffer elements (handle).
• comm — communicator (handle). the group chat itself. 🫶

---

💜 OUTPUT PARAMETERS 💜

• recvbuf — address of receive buffer (choice).
• request — communication request (handle). hang onto this one, because nonblocking means you have to come back for it later. it's your receipt. 🧾

---

🌸 THREAD AND INTERRUPT SAFETY 🌸

This routine is thread-safe. which means multiple threads can use it at the same time and you do not need to bring your own thread locks. she's self-sufficient. slay. 💅

However, the routine is not interrupt safe. Typically, this is due to the use of memory allocation routines such as malloc or other non-MPICH runtime routines that are themselves not interrupt-safe.

so like… thread-safe ✅, interrupt-safe ❌. don't get it twisted bestie, those are two different things and one of them will absolutely humble you. 😭

---

🦋 NOTES FOR FORTRAN 🦋

All MPI routines in Fortran (except for MPI_WTIME and MPI_WTICK) have an additional argument ierr at the end of the argument list. ierr is an integer and has the same meaning as the return value of the routine in C. In Fortran, MPI routines are subroutines, and are invoked with the call statement.

All MPI objects (e.g., MPI_Datatype, MPI_Comm) are of type INTEGER in Fortran.

so if you're in Fortran: extra ierr at the end, everything is a subroutine, call it, and all the fancy objects are just INTEGER. no cap, that's the whole translation. ✨

---

🌈 ERRORS 🌈

All MPI routines (except MPI_Wtime and MPI_Wtick) return an error value; C routines as the value of the function and Fortran routines in the last argument. Before the value is returned, the current MPI error handler is called. By default, this error handler aborts the MPI job. The error handler may be changed with MPI_Comm_set_errhandler (for communicators), MPI_File_set_errhandler (for files), and MPI_Win_set_errhandler (for RMA windows). The MPI-1 routine MPI_Errhandler_set may be used but its use is deprecated. The predefined error handler MPI_ERRORS_RETURN may be used to cause error values to be returned. Note that MPI does not guarantee that an MPI program can continue past an error; however, MPI implementations will attempt to continue whenever possible.

⚠️ real talk: the default handler kills the whole job. if you'd rather get an error code back and deal with it yourself, that's what MPI_ERRORS_RETURN is for. and yeah, MPI_Errhandler_set still exists but it's deprecated — we don't use deprecated things in this house, use the modern setters. 💅

the error codes, one by one, no skips:

• MPI_SUCCESS — No error; MPI routine completed successfully. 🎉
• MPI_ERR_ARG — Invalid argument. Some argument is invalid and is not identified by a specific error class (e.g., MPI_ERR_RANK).
• MPI_ERR_BUFFER — Invalid buffer pointer. Usually a null buffer where one is not valid.
• MPI_ERR_COMM — Invalid communicator. A common error is to use a null communicator in a call (not even allowed in MPI_Comm_rank). 💀
• MPI_ERR_COUNT — Invalid count argument. Count arguments must be non-negative; a count of zero is often valid.
• MPI_ERR_TYPE — Invalid datatype argument. Additionally, this error can occur if an uncommitted MPI_Datatype (see MPI_Type_commit) is used in a communication call. 🚩
• MPI_ERR_OTHER — Other error; use MPI_Error_string to get more information about this error code.

---

and that's the whole page, bestie. every flag, every handle, every error code — nothing left behind. obsessed with her honestly. 💖

🌸✨ signed off on 4/10/2026 — this was the MPI_Ialltoallv(3) page from the MPI manual. go forth and alltoallv, queen. 👑🦋