I don't think it's really a fault -- it's just how we designed and implemented it.
On Sep 6, 2010, at 7:40 AM, lyb wrote: > Thanks for your answer, but I test with MPICH2, it doesn't have this fault. > Why? >> Message: 9 >> Date: Wed, 1 Sep 2010 20:14:44 -0600 >> From: Ralph Castain<r...@open-mpi.org> >> Subject: Re: [OMPI users] MPI_Comm_accept and MPI_Comm_connect both >> use 100% one cpu core. Is it a bug? >> To: Open MPI Users<us...@open-mpi.org> >> Message-ID:<4e4bc153-b4e3-43e2-b980-904dabe78...@open-mpi.org> >> Content-Type: text/plain; charset="us-ascii" >> >> It's not a bug - that is normal behavior. The processes are polling hard to >> establish the connections as quickly as possible. >> >> >> On Sep 1, 2010, at 7:24 PM, lyb wrote: >> >> >>> > Hi, All, >>> > > I tested two sample applications on Windows 2003 Server, one use >>> > MPI_Comm_accept and other use MPI_Comm_connect, >>> > when run into MPI_Comm_accept or MPI_Comm_connect, the application use >>> > 100% one cpu core. Is it a bug or some wrong? >>> > > I tested with three version including Version 1.4 (stable), Version >>> > 1.5 (prerelease) and trunk 23706 version. >>> > > ... >>> > MPI_Open_port(MPI_INFO_NULL, port); >>> > MPI_Comm_accept( port, MPI_INFO_NULL, 0, MPI_COMM_WORLD,&client ); >>> > ... >>> > > ... >>> > MPI_Comm_connect( port, MPI_INFO_NULL, 0, MPI_COMM_WORLD,&server ); >>> > ... >>> > > thanks a lot. >>> > > lyb >>> > > > > > > _______________________________________________ >>> > users mailing list >>> > us...@open-mpi.org >>> > http://www.open-mpi.org/mailman/listinfo.cgi/users >>> >> -------------- next part -------------- >> HTML attachment scrubbed and removed >> > > > > _______________________________________________ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users -- Jeff Squyres jsquy...@cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/
