Hi,
I have entered the IP address of the system in the hosts file (twice
the IP for two CPU's) .I don't know about this mca parameter. Can you
please tell me about this or any reference material for this parameter?
Actually the input file is to perform minimization of the protein using
CH
Could you pass along what is in your hosts file? Did you set any mca
params in the default mca parameter file, or in your environ?
I note that you redirected stdin. Which rank is running and which is
stopped? How big is your input file? I am not familiar with your
program - are both ranks e
Hi All,
I am a new user of Open MPI. I have installed open mpi-1.3 on a
RedHat Linux-5 ver system with F77 set as gfortran compiler. I tested the
programs in examples and all ran. When I tried the CHARMM program with
mpirun (2 CPU's) the job runs on single processor and it is stopped in t
On Feb 12, 2009, at 2:00 PM, jody wrote:
In my application i use MPI_PROC_NULL
as an argument in MPI_Sendrecv to simplify the
program (i.e. no special cases for borders)
With 1.3 it works, but under 1.3.1a0r20520
i get the following error:
[jody@localhost 3D]$ mpirun -np 2 ./sr
[localhost.locald
Hi,
please note, that for plain standard valgrind runs (without ompi-internal
checks), You will not need to turn on the v1.3 options --enable-memchecker.
However, if You do, access to buffers, some MPI-standard conformance and (if
You select so) defined-ness checks of ompi-structures will be ena
Hi
In my application i use MPI_PROC_NULL
as an argument in MPI_Sendrecv to simplify the
program (i.e. no special cases for borders)
With 1.3 it works, but under 1.3.1a0r20520
i get the following error:
[jody@localhost 3D]$ mpirun -np 2 ./sr
[localhost.localdomain:29253] *** An error occurred in MP
On Feb 12, 2009, at 1:30 PM, George Bosilca wrote:
Some of the configure options changed their name for the 1.3. Here
is the updated list you should use in order to get valgrind support
for the 1.3:
--enable-memchecker --with-valgrind --with-memory-manager=none
Note that --enable-memcheck
Some of the configure options changed their name for the 1.3. Here is
the updated list you should use in order to get valgrind support for
the 1.3:
--enable-memchecker --with-valgrind --with-memory-manager=none
george.
On Feb 12, 2009, at 16:23 , Reuben D. Budiardja wrote:
On Thursday 1
On Feb 12, 2009, at 1:23 PM, Reuben D. Budiardja wrote:
From memory and reading the above links (i.e. untested advice):
1) Use OpenMPI 1.3 where the default is not to include this.
I tried it with OpenMPI 1.3, I didn't give any configure option
except --prefix, and Valgrind still does not work
On Thursday 12 February 2009 11:50:07 am Ashley Pittman wrote:
>
> > I am using OpenMPI-1.2.8 with all the default configure option. What
> > should I
> > do to be able use Valgrind with program compiled by OpenMPI ?
>
> From memory and reading the above links (i.e. untested advice):
> 1) Use Open
Hello,
Thanks for pointing me the location of the problem.
That came from my /dev, and the rights on the files.
I reinstalled linux, and i can now use openmpi correctly.
Still i don't know how i broke badly the system like this
Yann
Ralph Castain a écrit :
Never seen that one before, though
On 12 Feb 2009, at 15:53, Reuben D. Budiardja wrote:
Hello,
I am having problem that if a program is compiled with OpenMPI,
Valgrind
doesn't work correctly, i.e: it does not show the memory leak like it
supposed too. The same test program compiled with regular "gfortran"
and run
under Valgri
Hello,
I am having problem that if a program is compiled with OpenMPI, Valgrind
doesn't work correctly, i.e: it does not show the memory leak like it
supposed too. The same test program compiled with regular "gfortran" and run
under Valgrind will show the memory leak.
I search the list archive
You don't necessarily have to open all ports. Open MPI can use a range
of ports if specified.
The intervals used can be twitched using [btl_tcp_port_min_v4,
btl_tcp_port_range_v4] for the MPI layer and there should be something
similar for our daemons.
george.
On Feb 11, 2009, at 21:49
Windows do not have natively ssh installed, so we didn't bother making
sure we can start MPI applications using rsh/ssh. There is a special
PLM (process starter) for Windows called process, but I didn't test it
in a long time. Anyway, this only allow you to start local MPI jobs.
george.
The warning is a side-effect of how we're probing for OpenFabrics-
capable hardware (e.g., IB HCAs). While annoying, the warning is
harmless -- it's just noisily indicating that you seem to have no
OpenFabrics-capable hardware.
We made a probe a bit smarter in Open MPI v1.3. If you upgrade
I don't believe we support cygwin at this time...native Windows
support is coming in a later release.
On Feb 12, 2009, at 7:09 AM, Siegmar Gross wrote:
Hi,
I have installed openmpi-1.3 using Cygwin 1.5 on top of Windows
XP Pro with gcc-3.4.4 with the following commands.
At first I added "#d
Hi,
I have installed openmpi-1.3 using Cygwin 1.5 on top of Windows
XP Pro with gcc-4.3.2 with the following commands.
At first I added "#define NOMINMAX" before the line
"#include MCA_timer_IMPLEMENTATION_HEADER" and "#undef NOMINMAX"
after that line in file "ompi/tools/ompi_info/param.cc" as de
Yes, thanks, that seems to have been the problem,
Gary
Ralph Castain wrote:
The ^ applies to everything that follows, so you just turned off all
of the tcp, self, and openib comm paths. :-)
If you just wanted to drop tcp from that list, you should just use
-mca btl self,openib.
Ralph
On Fe
Hi,
I have installed openmpi-1.3 using Cygwin 1.5 on top of Windows
XP Pro with gcc-3.4.4 with the following commands.
At first I added "#define NOMINMAX" before the line
"#include MCA_timer_IMPLEMENTATION_HEADER" and "#undef NOMINMAX"
after that line in file "ompi/tools/ompi_info/param.cc" as de
Hi,
I have installed openmpi-1.2.8 using Cygwin 1.5 on top of Windows
XP Pro with gcc-4.3.2 with the following commands.
configure --prefix=/usr/local/openmpi-1.2.8-gcc-4 \
CFLAGS="" CXXFLAGS="" FFLAGS="" FCFLAGS="" \
CPPFLAGS="" C_INCL_PATH="" C_INCLUDE_PATH="" CPLUS_INCLUDE_PATH="" \
OBJC
Hi,
I have installed openmpi-1.2.8 using Cygwin 1.5 on top of Windows
XP Pro with gcc-3.4.4 with the following commands.
configure --prefix=/usr/local/openmpi-1.2.8 \
CFLAGS="" CXXFLAGS="" FFLAGS="" FCFLAGS="" \
CPPFLAGS="" C_INCL_PATH="" C_INCLUDE_PATH="" CPLUS_INCLUDE_PATH="" \
OBJC_INCLU
Hello,
I use openmpi. I have installed it in the directory :
/usr/lib64/openmpi/1.2.5-gcc/
If I execute a test with openMPI (for exemple helloworld!!!) I have the
next problem:
/*[nmoulin@clusterdell ~/mpi-test]$ mpirun -np 4 mpi-test
libibverbs: Fatal: couldn't read uverbs ABI version.
--
Ralph Castain wrote:
Never seen that one before, though we did put the test in there "just
in case".
Basically, this message indicates that the system returned an error
when we attempted to dup2 the stdout/err file descriptors so we could
setup a pipe by which we can forward that output to yo
On Wed, 2009-02-11 at 17:14 -0700, Ralph Castain wrote:
> Actually, this was also the subject of another email thread on the
> user list earlier today. The user noted that we had lost an important
> line in our Makefile.am for the tm plm module, and that this was the
> root cause of the probl
25 matches
Mail list logo
