[mvapich-discuss] mvapich2 1.4.0 and MPI_Comm_Spawn fatal errors

[Dhabaleswar Panda]

Dr. Kallies - This is to let you know that this fix has been applied to
the bugfix branch version of mvapich2 1.4. This latest version also has a
few more bug fixes. You can check out the latest version from the
following URL:

http://mvapich.cse.ohio-state.edu/nightly/mvapich2/branches/1.4

All these fixes will be reflected in the upcoming 1.4.1 release.

Thanks,

DK


On Tue, 10 Nov 2009, Dhabaleswar Panda wrote:

> Dr. Kallies - Thanks for reporting this error in the mpich2 sources
> distributed with mvapich2. We will examine this and make this change in
> the MVAPICH2 branch version. For our next mvapich2 release, we are working
> on updating to the latest MPICH2 1.2.1 release.
>
> Thanks,
>
> DK
>
> On Tue, 10 Nov 2009, Bernd Kallies wrote:
>
> > Dear mvapich2 dev team,
> >
> > I'm currently playing with dynamic process creation with mvapich2 1.4.0.
> >
> > Note that there is a funny bug in
> > src/mpid/ch3/src/ch3u_comm_spawn_multiple.c line 195:
> >
> > if (mpi_errno != PMI_SUCCESS) {
> >
> > must read
> >
> > if (pmi_errno != PMI_SUCCESS) {
> >
> > otherwise fatal spawn errors cannot be handled correctly, when using
> > pmi.
> >
> > My question: will you fix this in the mpich2 sources distributed with
> > mvapich2, or will you update to a more recent mpich2?
> >
> > Sincerely, BK
> >
> > --
> > Dr. Bernd Kallies
> > Konrad-Zuse-Zentrum für Informationstechnik Berlin
> > Takustr. 7
> > 14195 Berlin
> > Tel: +49-30-84185-270
> > Fax: +49-30-84185-311
> > e-mail: kallies at zib.de
> >
> >
> > _______________________________________________
> > mvapich-discuss mailing list
> > mvapich-discuss at cse.ohio-state.edu
> > http://mail.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss
> >
>
>
> _______________________________________________
> mvapich-discuss mailing list
> mvapich-discuss at cse.ohio-state.edu
> http://mail.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss
>