[mpich-discuss] Problem with Log generation in MPE2

Anthony Chan chan at mcs.anl.gov
Fri Nov 25 11:54:23 CST 2011


I am not familiar with GAMESS's build system.
Did you try putting mpefc's full path in GAMESS's
build system or Makefiles ?

A.Chan

----- Original Message -----
> Hi,
> 
> Seems like I was executing GAMESS with sockets and probably thats why
> I
> wasn't getting
> a log. Now I am using the GAMESS for MPI and to compile it, I replaced
> ifort with mpefc in the compilation script
> and now the compilation given an error that mpefc:command not found.
> But
> when I do
> "which mpefc" it gives me the path to mpefc. Could you please help me
> with
> this error.
> 
> 
> On Mon, Oct 10, 2011 at 3:18 PM, Anthony Chan <chan at mcs.anl.gov>
> wrote:
> 
> >
> > Based on the output, your gamess.00.x does NOT make any MPI calls
> > (so no MPI logging from MPE). Does gamess.00.x call any other
> > executables ?
> >
> > A.Chan
> >
> > ----- Original Message -----
> > > Hi,
> > >
> > > Below is the output.
> > >
> > > nm gamess.00.x | grep -Ei "MPI_|MPE_"
> > >
> > > 0000000000eeb2b0 T setfmpi_
> > > 0000000001253260 T zsumpe_
> > >
> > >
> > > On Mon, Oct 10, 2011 at 3:02 PM, Anthony Chan <chan at mcs.anl.gov>
> > > wrote:
> > >
> > > >
> > > > You still misunderstood me, not MPI_IMPE_, it should be
> > > >
> > > > nm gamess.00.x | grep -Ei "MPI_|MPE_"
> > > >
> > > > Copy and paste the above command to a terminal (i.e. a shell)
> > > > and let me know what the output is..
> > > >
> > > > ----- Original Message -----
> > > > > Hi,
> > > > >
> > > > > The executable in GAMESS is gamess.00.x.
> > > > > When I do nm gamess.00.x | grep -Ei MPI_IMPE_ I do not get any
> > > > > output.
> > > > >
> > > > > On Mon, Oct 10, 2011 at 1:41 PM, Anthony Chan
> > > > > <chan at mcs.anl.gov>
> > > > > wrote:
> > > > >
> > > > > >
> > > > > > I don't mean "MPE_IMPE_". In the command
> > > > > >
> > > > > > nm "executable_name" | grep -Ei "MPI_|MPE_"
> > > > > >
> > > > > > "|" is not "I", it's the vertical bar character on your
> > > > > > keyboard.
> > > > > >
> > > > > > A.Chan
> > > > > >
> > > > > > ----- Original Message -----
> > > > > > > Hi,
> > > > > > >
> > > > > > > I used the command "*grep -ri MPE_IMPE_ ** " in GAMESS
> > > > > > > directory
> > > > > > > and
> > > > > > > it did
> > > > > > > not give
> > > > > > > any output.
> > > > > > >
> > > > > > > On Mon, Oct 10, 2011 at 12:28 PM, Anthony Chan
> > > > > > > <chan at mcs.anl.gov>
> > > > > > > wrote:
> > > > > > >
> > > > > > > >
> > > > > > > > Can you do a "nm" on the final executable to see if it
> > > > > > > > has
> > > > > > > > MPI
> > > > > > > > and
> > > > > > > > MPE
> > > > > > > > symbols
> > > > > > > >
> > > > > > > > nm "executable_name" | grep -Ei "MPI_|MPE_"
> > > > > > > >
> > > > > > > > and send us the output.
> > > > > > > >
> > > > > > > > ----- Original Message -----
> > > > > > > > > Hi,
> > > > > > > > >
> > > > > > > > > I downloaded it from the webpage(
> > > > > > > > >
> > > > http://www.mcs.anl.gov/research/projects/perfvis/download/index.htm
> > > > > > ).
> > > > > > > > > Yeah, the executables are linked by a script lked
> > > > > > > > > finally
> > > > > > > > > in
> > > > > > > > > which
> > > > > > > > > I
> > > > > > > > > replace
> > > > > > > > > gfortran by mpefc -mpilog.
> > > > > > > > >
> > > > > > > > > On Mon, Oct 10, 2011 at 11:08 AM, Anthony Chan
> > > > > > > > > <chan at mcs.anl.gov>
> > > > > > > > > wrote:
> > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > Where did you get your version of MPE2 ? Did you get
> > > > > > > > > > it
> > > > > > > > > > from
> > > > > > > > > > MPICH2
> > > > > > > > > > ?
> > > > > > > > > >
> > > > > > > > > > Also, is the final executable linked by "mpefc
> > > > > > > > > > -mpilog"
> > > > > > > > > > ?
> > > > > > > > > >
> > > > > > > > > > A.Chan
> > > > > > > > > >
> > > > > > > > > > ----- Original Message -----
> > > > > > > > > > > Hi ,
> > > > > > > > > > >
> > > > > > > > > > > I am trying to generate the log of the execution
> > > > > > > > > > > of
> > > > > > > > > > > quantum
> > > > > > > > > > > chemistry
> > > > > > > > > > > software GAMESS
> > > > > > > > > > > using MPE2. Although GAMESS compiles by using
> > > > > > > > > > > mpefc
> > > > > > > > > > > -mpilog by
> > > > > > > > > > > replacing
> > > > > > > > > > > instances of
> > > > > > > > > > > gfortran, the execution does not generate a log.
> > > > > > > > > > > Can
> > > > > > > > > > > anyone
> > > > > > > > > > > help
> > > > > > > > > > > me
> > > > > > > > > > > with
> > > > > > > > > > > this?
> > > > > > > > > > >
> > > > > > > > > > > Thanks.
> > > > > > > > > > >
> > > > > > > > > > > _______________________________________________
> > > > > > > > > > > mpich-discuss mailing list
> > > > > > > > > > > mpich-discuss at mcs.anl.gov
> > > > > > > > > > > To manage subscription options or unsubscribe:
> > > > > > > > > > > https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
> > > > > > > > > > _______________________________________________
> > > > > > > > > > mpich-discuss mailing list mpich-discuss at mcs.anl.gov
> > > > > > > > > > To manage subscription options or unsubscribe:
> > > > > > > > > > https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
> > > > > > > > > >
> > > > > > > >
> > > > > >
> > > >
> >


More information about the mpich-discuss mailing list