Hi George, Thank you for your prompt reply, but my apologies for missing it and it's been quite a while.
Anyway, to answer your question I was trying to do sendrecv in a chain where "toSend" and "receiveFrom" ranks are not the same. I was using a single buffer, which resulted in cases where the buffer content got replaced with received values before being able to send its original content. I think it was an error on my side to use a single buffer in a pipeline sendrecev. I fixed it by having two buffers. I guess still with MPI_IN_PLACE this could happen. Please correct me here if I am wrong. Thank you, Saliya On Wed, Nov 27, 2013 at 6:00 AM, George Bosilca <bosi...@icl.utk.edu> wrote: > Why do you need an order? If you plan to send and receive on the same > buffer, you should use the MPI constructs for that, namely MPI_IN_PLACE. > > George. > > On Nov 27, 2013, at 07:20 , Saliya Ekanayake <esal...@gmail.com> wrote: > > Hi, > > Just want to verify if sendrecv provides any guarantees as to which > operation (send or receive) happens first. I think it is not, is it? > > Thank you, > Saliya > > -- > Saliya Ekanayake esal...@gmail.com > Cell 812-391-4914 Home 812-961-6383 > http://saliya.org > _______________________________________________ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users > > > > _______________________________________________ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users > -- Saliya Ekanayake esal...@gmail.com Cell 812-391-4914 Home 812-961-6383 http://saliya.org
