MPI_Init_thread - Initialize the MPI execution environment
#include <mpi.h> int MPI_Init_thread(int *pargc, char ***pargv, int requested, int *pprovided)
LAM currently supports MPI_THREAD_SINGLE , MPI_THREAD_SERIALIZED , and MPI_THREAD_FUNNELED . MPI_THREAD_MULTIPLE is not yet supported. The use of MPI_INIT_THREAD is recommended instead of MPI_INIT if an MPI program is going to be multi-threaded. However, the LAM_MPI_THREAD_LEVEL environment variable can be used to the same effect. See MPI_Init(3) for more details.
There is currently no distinction between MPI_THREAD_SINGLE and MPI_THREAD_FUNNELED . An application using MPI_THREAD_FUNNELED promises to never call MPI from a different thread than the one that invoked MPI_INIT (or MPI_INIT_THREAD ). LAM does no checking to ensure that this is true. As such, no locks are placed around MPI calls to ensure that only one thread is making MPI calls at a time (since the application has promised that only one thread will be making MPI calls).
Using MPI_THREAD_SERIALIZED will cause LAM to place locks around all MPI calls such that only one thread will be able to enter the MPI library at a time; beware of this fact for portability with other MPI implementations. Even with multiple threads, deadlock is still possible when using MPI_THREAD_SERIALIZED -- applications still need to be aware of this and code appropriately.
Other than this restriction, this function behaves identically to MPI_INIT (see MPI_Init(3)).
Note that both MPI_INIT and MPI_INIT_THREAD are allowed to changed the requested thread level based on lower-level device and SSI module initializations. For example, if a user application invokes MPI_Init , some SSI modules may require threads and therefore upgrade the thread support level to something higher than MPI_THREAD_SINGLE .
If an error occurs in an MPI function, the current MPI error handler is called to handle it. By default, this error handler aborts the MPI job. The error handler may be changed with MPI_Errhandler_set ; the predefined error handler MPI_ERRORS_RETURN may be used to cause error values to be returned (in C and Fortran; this error handler is less useful in with the C++ MPI bindings. The predefined error handler MPI::ERRORS_THROW_EXCEPTIONS should be used in C++ if the error value needs to be recovered). Note that MPI does not guarantee that an MPI program can continue past an error.
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. The C++ bindings for MPI do not return error values; instead, error values are communicated by throwing exceptions of type MPI::Exception (but not by default). Exceptions are only thrown if the error value is not MPI::SUCCESS .
Note that if the MPI::ERRORS_RETURN handler is set in C++, while MPI functions will return upon an error, there will be no way to recover what the actual error value was.
- - No error; MPI routine completed successfully.
- - Invalid argument. Some argument is invalid and is not identified by a specific error class. This is typically a NULL pointer or other such error.
- - This error class is associated with an error code that indicates that an attempt was made to call MPI_INIT a second time. MPI_INIT may only be called once in a program.
- - Other error; use MPI_Error_string to get more information about this error code.
MPI_Init, MPI_Finalize, lamboot, mpirun, lamhalt, lamssi
For more information, please see the official MPI Forum web site, which contains the text of both the MPI-1 and MPI-2 standards. These documents contain detailed information about each MPI function (most of which is not duplicated in these man pages).
The LAM Team would like the thank the MPICH Team for the handy program to generate man pages ("doctext" from ftp://ftp.mcs.anl.gov/pub/sowing/sowing.tar.gz ), the initial formatting, and some initial text for most of the MPI-1 man pages.