MPI_recv blocks in a small code

Question

0.00/5 (No votes)

See more:

Hi,

If i am using MPI_Send, I am not getting any problem but when I am using both MPI_Send and MPI_Recv my program blocks.

C++

#include <iostream>
#include <fstream>
#include <cmath>
#include <mpi.h>
#include <ctime>
#include <vector>

//using namespace std;

// Sorts the input row into chunks to be scattered two all the processors.
//void sortByProcess(std::vector<double> list1, double* list2, int count);

// Swaps two rows.
//void swap(double** list, int count, int row1, int row2);

//int N=32;
int rank, size;
int tag = 99;
int main(int argc, char * argv[])
{
  double sTime, eTime, rTime;
  std::ifstream inFile;
  int num_rows = 3200;
  int num_cols = 3200;
  int cur_control = 0;
  double * send_buffer = NULL;
  double * recv_buffer = NULL;
  double ** data = NULL;
  double determinant;
  MPI_Status stat;
  std::vector<double> file_buffer;
  
  // Just get the initialization of the program going.
  MPI_Init(&argc, &argv);
  MPI_Comm_rank(MPI_COMM_WORLD, &rank);
  MPI_Comm_size(MPI_COMM_WORLD, &size);

  // If the input file is not given, print message and exit.
  if(argc < 2)
  {
    std::cout << "No input file given." << std::endl;
    MPI_Finalize();
    return 0;
  }
  // If the root node (0), then open the input file and read in the
  // number of rows.
  if(!rank)
  {
    inFile.open(argv[1]);
    inFile >> num_rows;
    file_buffer.resize(num_rows);
  }

  send_buffer = new double[num_rows];
  // Broadcasts p2p the number of rows to each processor.
  //MPI_Bcast (&num_rows, 1, MPI_INT, 0, MPI_COMM_WORLD);
  if(!rank) {
  for(int i=1; i<size; ++i) {
     
     MPI_Send(&num_rows, 1, MPI_INT, i, tag, MPI_COMM_WORLD);
  }
  }
  for(int i=1; i<size; ++i) {
     
     MPI_Recv(&num_rows, 1, MPI_INT, 0, tag, MPI_COMM_WORLD, &stat);
  }
  num_cols = num_rows / size;

  delete [] send_buffer;
  MPI_Finalize();
  return 0;

Some body please guide me how to correct this error?

Zulfi.

What I have tried:

I have checked the syntax of MPI_recv(...). i can't understand any reason for blocking.

MPI_Recv(...) is a blocking call but it must get the data.

Posted 1-May-19 13:15pm

zak100

Updated 2-May-19 7:51am

v3

Add a Solution

2 solutions

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

Rick York · Answer 1 · 2019-05-01T15:35:00

Solution 1

Have you tried inserting a MPI_Barrier(MPI_COMM_WORLD); call in there?

See this Q & A[^]

Posted 1-May-19 15:35pm

Rick York

KarstenK · Answer 2 · 2019-05-02T00:30:00

Solution 2

Sending and Receiving are concurrent tasks and so have always some side effects and they need some time (in milliseconds). You need to syncronize receive and send calls.

The first problem I see in your code is that you dont check the return codes of the functions.

Posted 2-May-19 0:30am

KarstenK

Comments

zak100 2-May-19 13:54pm

I tried the following program. It does not have any Barrier but its working fine:

#include <mpi.h>
#include <stdio.h>
#include <string.h>
#include <math.h>

int main(int argc, char **argv)
{
char idstr[100];
char buff[128];
int numprocs;
int myid;
int i;
MPI_Status stat;
/*********************************************
Begin program
*********************************************/
MPI_Init(&argc,&argv); // Initialize
MPI_Comm_size(MPI_COMM_WORLD, &numprocs); // Get # processors
MPI_Comm_rank(MPI_COMM_WORLD, &myid); // Get my rank (id)
if( myid == 0 )
{ // Master
printf("WE have %d processors\n", numprocs);
for( i = 1; i < numprocs; i++)
{
sprintf(buff, "Hello %d", i);
MPI_Send(buff, 128, MPI_CHAR, i, 0, MPI_COMM_WORLD);
}
for( i = 1; i < numprocs; i++)
{
MPI_Recv(buff, 128, MPI_CHAR, i, 0, MPI_COMM_WORLD, &stat);
printf("%s\n", buff);
}
}
else
{ // Slave
MPI_Recv(buff, 128, MPI_CHAR, 0, 0, MPI_COMM_WORLD, &stat);
sprintf(idstr, "Hello 0, Processor %d is present and accounted for !",myid);
MPI_Send(buff, 128, MPI_CHAR, 0, 0, MPI_COMM_WORLD);
}
MPI_Finalize();
}


/*
$ mpicc -o SR sendRec.c
$ mpirun -np 4 ./SR
WE have 4 processors
Hello 1
Hello 2
Hello 3
$ 
*/

zak100 2-May-19 13:55pm

Is else part necessary?

Zulfi.