Yoinks!
Yes, this is definitely a problem.
Fixed in SVN -- will be in tonight's nightly tarball (and in 7.1.2 when
it is released).
Thanks for finding this!
On Jul 13, 2005, at 7:49 PM, Roland Krause wrote:
> Hello all,
>
> I tried to use MPI_Bcast with an intercommunicator from a Fortran
> program. I included mpif.h. The call got me an error message saying
> that the root for the communicator was invalid and so I investigated a
> little and found out that there is no MPI_ROOT in mpif.h. I found it in
> mpi.h (set to -4), copied that into my Fortran program (sort of) and
> voila it works now.
>
> I dug a bit into mpif.h.in and didn't see it there either so I believe
> now that this might be an oversight/bug.
>
> Well maybe I am wrong and I misconfigured the whole thing. Here is some
> of my specs.
>
> Lam-7.1.1 recompiled for ifort-8.1 and disabled your malloc
> implementation since it interferes with large memory block allocations
> (different subject).
>
> Opteron x86_64, SuSE-9.2 Linux
>
> Hope this is enough information but a look into your codebase should
> suffice to tell me whether I am full of it....
>
>
> Best regards and many thanks for your help
>
> Roland
>
>
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
--
{+} Jeff Squyres
{+} jsquyres_at_[hidden]
{+} http://www.lam-mpi.org/
|
