Re: [Boost-users] mpi non-blocking communication

On 2010-12-02 18:00:27 +0100, Riccardo Murri said:

Hi Philipp,

On Thu, Dec 2, 2010 at 4:12 PM, Philipp Kraus wrote:

...

There is only one problem at the time:  The CPU 0 does not catch every message. The main method:

MPI::Init_thread( argc, argv, MPI_THREAD_SERIALIZED ); mpi::communicator loMPICom; tl::logger::getInstance()->startListener(loMPICom); tl::logger::getInstance()->write(loMPICom, tl::logger::warn, "test all"); tl::logger::getInstance()->shutdownListener(loMPICom); MPI::Finalize();

So every process shoudl create a message and the CPU 0 should catch them all. The write method sends the data with isend (non-blocking). The CPU 0 runs a thread in background with

while (m_listenerrunnging) {        while (boost::optionalmpi::status l_status = p_mpi.iprobe(mpi::any_source, LOGGER_MPI_TAG)) {                        std::string l_str;                p_mpi.recv(  l_status->source(), l_status->tag(), l_str );        }        boost::this_thread::yield(); }

This code for the background thread will only probe for new messages until m_listenerrunnging is true; as soon as it turns to false (in logger::getInstance()->shutdownListener?), it will stop and receive no more messages.

You probably want to make it run until all messages have arrived. How you do this depends on your application: if you know how many messages will be coming, just count them; if you don't, make ranks !=1 send an "end of transmission" message when they are done and only shutdown the listener when all "end of transmission" messages have been received. There might be other methods that are more suited to your application.

I have created a end-of-transmission code in my shutdown method. That works now very well Thanks Phil