Thanks for the bug report. This has been fixed in r26784 of the trunk and
should be in tonight's tarball.
Brian
On 7/11/12 6:09 AM, "Mikhail Kurnosov" <mkurno...@gmail.com> wrote:
>Hello,
>
>In the case of single process the MPI_Ibarrier call fails (seg. fault).
>Request object does not initialized in this function (LibNBC is not
>called for it).
>
>The following example fails on Open MPI 1.9a1r26779:
>
>$ cat ibarrier_test.c
>#include <mpi.h>
>
>int main(int argc, char **argv)
>{
> MPI_Request req;
> MPI_Status status;
>
> MPI_Init(&argc, &argv);
> MPI_Ibarrier(MPI_COMM_WORLD, &req);
> /* MPI_Wait will crash. */
> MPI_Wait(&req, &status);
> MPI_Finalize();
> return 0;
>}
>
>$ mpiexec -n 1 ./ibarrier_test
>[mkws:26288] *** Process received signal ***
>[mkws:26288] Signal: Segmentation fault (11)
>[mkws:26288] Signal code: Address not mapped (1)
>[mkws:26288] Failing at address: 0x58
>[mkws:26288] [ 0] /lib64/libpthread.so.0() [0x3d6040f500]
>[mkws:26288] [ 1]
>/home/mkurnosov/opt/openmpi-1.9a1r26779/lib/libmpi.so.0(ompi_request_defau
>lt_wait+0x17)
>[0x7f75d824c707]
>[mkws:26288] [ 2]
>/home/mkurnosov/opt/openmpi-1.9a1r26779/lib/libmpi.so.0(MPI_Wait+0x5d)
>[0x7f75d8276e0d]
>[mkws:26288] [ 3] ./ibarrier_test(main+0x46) [0x40089a]
>[mkws:26288] [ 4] /lib64/libc.so.6(__libc_start_main+0xed) [0x3d5f82169d]
>[mkws:26288] [ 5] ./ibarrier_test() [0x400799]
>[mkws:26288] *** End of error message ***
>
>--
>Mikhail Kurnosov
>Computer Systems Department
>Siberian State University of Telecommunications and Information Sciences
>Address: 630102, 86 Kirova str., Novosibirsk, Russia
>Email: mkurno...@gmail.com
>http://cpct.sibsutis.ru/~mkurnosov
>
>_______________________________________________
>devel mailing list
>de...@open-mpi.org
>http://www.open-mpi.org/mailman/listinfo.cgi/devel
>
>
