Re: [Mpi3-ft] MPI_ANY_SOURCE

13 Oct 2011

      yeah you're right. We only need to lock around the wait for the
nonblocking case. One thing that slipped my mind was that MPI_Irecv
will return regardless of the state of other processes - since it is
nonblocking. So eventhough it is ANY_SOURCE, it will still return
MPI_SUCCESS once a request is generated. The request might be marked
as failed, but the call to MPI_Irecv will not return the error, the
subsequent MPI_Wait will return the error.

So it is only really necessary to lock around the Wait.

To tweak your example a bit:
--------------------------
/* Assume the user will cancel the request outside of this function,
 * so we do not need to cancel it for them inside the function.
 */
int My_AS_MPI_Wait(MPI_Request *req, MPI_Status *status)
{
   while(1) {
       reader_lock();
       if (my_cnt != recognize_cnt) {
           /* New failures were detected */
           /* check failed_group and decide if ok to continue */
           if (ok_to_continue(req, failed_group) == FALSE) {
               reader_unlock();
               /* caller responsible for canceling the request */
               return MPI_ERR_PROC_FAIL_STOP;
           }
           my_cnt == recognize_cnt;
       }
       err = MPI_Wait(req, status);
       if (err == MPI_WARN_PROC_FAIL_STOP) {
           /* Failure case */
           reader_unlock();
           writer_lock();
           if (my_cnt != recognize_cnt) {
               /* another thread has already re-enabled wildcards */
               writer_unlock();
               continue;
           }
           MPI_Comm_reenable_any_source(comm, &failed_group);
           ++recognize_cnt;
           writer_unlock();
           continue;
       } else if (MPI_SUCCESS != err) {
           reader_unlock();
           /* caller responsible for canceling the request */
           return MPI_ERR_PROC_FAIL_STOP;
       }

       /* Successful completion of the MPI_Wait */
       reader_unlock();
       return MPI_SUCCESS;
   }
}
--------------------------

On Tue, Oct 11, 2011 at 4:16 PM, Darius Buntinas <[email protected]> wrote:
...
I think you're right Josh, the blocking version wouldn't need to be changed.
For the nonblocking version, wouldn't we only need to lock around the Wait, not between the Recv and Wait? ï¿½If we're worried about hanging in a blocking Wait, I think we just need to check for all-clients-failed before calling Wait. ï¿½If anysources are reenabled by another thread before this thread calls Wait, that's OK, so long as the thread checks first.
Here's a function a user could implement to use whenever waiting on an anysource:
int My_AS_MPI_Wait(MPI_Request *req, MPI_Status *status)
{
ï¿½ ï¿½while(1) {
ï¿½ ï¿½ ï¿½ ï¿½reader_lock();
ï¿½ ï¿½ ï¿½ ï¿½if (my_cnt != recognize_cnt) {
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½/* New failures were detected */
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½/* check failed_group and decide if ok to continue */
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½if (ok_to_continue(req, failed_group) == FALSE) {
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½reader_unlock();
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½return MPI_ERR_PROC_FAIL_STOP;
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½}
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½my_cnt == recognize_cnt;
ï¿½ ï¿½ ï¿½ ï¿½}
ï¿½ ï¿½ ï¿½ ï¿½err = MPI_Wait(req, status);
ï¿½ ï¿½ ï¿½ ï¿½if (err == MPI_WARN_PROC_FAIL_STOP) {
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½/* Failure case */
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½reader_unlock();
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½writer_lock();
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½if (my_cnt != recognize_cnt) {
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½/* another thread has already re-enabled wildcards */
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½writer_unlock();
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½continue;
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½}
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½MPI_Comm_reenable_any_source(comm, &failed_group);
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½++recognize_cnt;
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½writer_unlock();
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½continue;
ï¿½ ï¿½ ï¿½ ï¿½} else {
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½reader_unlock();
ï¿½ ï¿½ ï¿½ ï¿½ ï¿½ ï¿½return MPI_ERR_PROC_FAIL_STOP;
ï¿½ ï¿½ ï¿½ ï¿½}
ï¿½ ï¿½ ï¿½ ï¿½reader_unlock();
ï¿½ ï¿½}
}
-d
_______________________________________________
mpi3-ft mailing list
[email protected]
http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-ft
-- 
Joshua Hursey
Postdoctoral Research Associate
Oak Ridge National Laboratory
http://users.nccs.gov/~jjhursey

Re: [Mpi3-ft] MPI_ANY_SOURCE

Josh Hursey