[mpich-discuss] MPI_Recv problem

[Ken Raffenetti]

He was suggesting using print (cout) statements to verify the parameters of your MPI calls before the error occurs.

----- Original Message -----
> From: "Soheil Hooshdaran" <shooshdaran577 at gmail.com>
> To: "Pavan Balaji" <balaji at mcs.anl.gov>
> Cc: discuss at mpich.org
> Sent: Monday, December 2, 2013 8:06:06 AM
> Subject: Re: [mpich-discuss] MPI_Recv problem
> 
> 
> 
> What should I use?
> 
> 
> Even the statement right after MPI_Sent doesn't execute. My code has
> something like this:
> //////////////////////in the root process
> 
> while(i<iWorldSize)
> {
> cout << "executing send for the " << i << "th time" << endl;
> MPI_Send( A[i], n/iWorldSize * m, MPI_INT, i, itag, MPI_COMM_WORLD );
> sent++;
> cout << "executed send for the " << i << "th time, sending " <<
> n/iWorldSize * m << " bytes to processor " << i << endl;
> 
> i+=n/iWorldSize;
> }//end while
> 
> 
> 
> 
> ////////////////////in the slave process
> 
> while(sent)
> {
> cout << "**********sent= " << sent << endl;
> cout << "\n\n\n**********Executing receive*******" << endl;
> MPI_Recv(&buffer, n/iWorldSize*m, MPI_INT, 0, itag, MPI_COMM_WORLD,
> &status);
> sent=0;
> cout << "**********receive executed***********" << endl;
> }
> 
> 
> 
> On Mon, Dec 2, 2013 at 5:21 PM, Pavan Balaji < balaji at mcs.anl.gov >
> wrote:
> 
> 
> 
> Did you try printing how many bytes you are receiving before calling
> MPI_Recv?
> 
> — Pavan
> 
> 
> 
> On Dec 1, 2013, at 11:43 PM, Soheil Hooshdaran <
> shooshdaran577 at gmail.com > wrote:
> 
> > Hello. Please tell me what is wrong with my program that it doesn't
> > goes no further than executing MPI_Send() for the first time.
> > The file is attached
> > 
> > 
> > On Sat, Nov 30, 2013 at 7:02 PM, Soheil Hooshdaran <
> > shooshdaran577 at gmail.com > wrote:
> > So what do I have to do now, sir? Especially regarding my Bcast()s
> > a,d Recv()s
> > 
> > 
> > On Sat, Nov 30, 2013 at 1:05 AM, Pavan Balaji < balaji at mcs.anl.gov
> > > wrote:
> > 
> > Please keep discuss at mpich.org cc’ed.
> > 
> > On Nov 29, 2013, at 2:19 PM, Soheil Hooshdaran <
> > shooshdaran577 at gmail.com > wrote:
> > > executing send for the 3th time
> > > executed send for the 3th time, sending 40 bytes
> > > Fatal error in MPI_Recv: Message truncated, error stack:
> > > MPI_Recv(184).....................: MPI_Recv(buf=0xbff95c98,
> > > count=0, MPI_INT, src=0, tag=1, MPI_COMM_WORLD,
> > > status=0xbff95c70) failed
> > > MPIDI_CH3U_Receive_data_found(131): Message from rank 0 and tag 1
> > > truncated; 160 bytes received but buffer size is 0
> > 
> > I don’t think I can explain the error any better than what the
> > message above already says.
> > 
> > --
> > Pavan Balaji
> > http://www.mcs.anl.gov/~balaji
> > 
> > 
> > <p1_Hooshdaran_static[debug].cpp>
> 
> 
> 
> --
> Pavan Balaji
> http://www.mcs.anl.gov/~balaji
> 
> 
> 
> _______________________________________________
> discuss mailing list     discuss at mpich.org
> To manage subscription options or unsubscribe:
> https://lists.mpich.org/mailman/listinfo/discuss