Since your mail contains the word "MPICH", I assume you are using the
MPICH implementation of MPI, and you should probably be writing to the
MPICH maintainers with questions about their software.
This list is for the LAM implementation of MPI (hence the name
"lam_at_[hidden]") which is wholly unrelated to MPICH except insofar as
they are both MPI implementations. We cannot and will not speak for
the MPICH maintainers.
Thanks.
On Feb 14, 2005, at 7:54 AM, Ru-Zhen Li wrote:
> Dear All,
>
> I am a new user of MPI, i met a problem when I tried to compile some
> package, the makefile is like this
>
>
> $(MAKE) LD="ifort -o" \
>
> LDFLAGS="-L/usr/local/encap/mpich-125-2-intel/lib -lmpich -lpmpich" \
>
> FC=ifort FFLAGS="-c -O -B100" TIMER="" \
>
> CPFLAGS="-D$(STRESS) -DMPI -P -D'POINTER=integer' \
>
> -I/usr/local/encap/mpich-125-2-intel/include/" \
>
> EX=$(EX) BINROOT=$(BINROOT) $(TYPE)
>
> the machine is a Linux cluster, with intel fortran compiler.
>
> I found out that all some of the source files written in fortran, the
> .f files, were changed to .i files by cpp and then copied to .tmp.f
> files, then the .tmp.o was mv to .o and then the .i file was deleted
>
> the problem is, afte the .f file were changed to .tmp.f file, some of
> the statements was moved to the left one bit, so the format was messed
> up. for example
>
> close ....... was now changed to
>
> close .....
>
> does any one know why?
>
> thank you very much in advance
>
>
>
> best wishes,
>
> Ru
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
--
{+} Jeff Squyres
{+} jsquyres_at_[hidden]
{+} http://www.lam-mpi.org/
|
