Hi,
I am pretty sure that C++ has no restrict keyword (only C99), so IMHO
it's not the compilers fault is not compiling.
You probably want to turn restricts off for C++ by default, unless you
know the compiler supports it.
best regards,
Samuel
smime.p7s
Description: S/MIME Cryptographic Si
Hi,
I am using the OpenMPI version 1.3 downloaded directly from the site.
I want to establish a client/server connection among two local processes (in
the same machine).
I wrote a simple publisher combining MPI_Open_port / MPI_Publish_name /
MPI_Comm_accept and a connector with MPI_Lookup_name /
Raymond Wan wrote:
>
> Actually, when I run the above mpirun command, I don't see "sleep"
> running locally on machine Y, either. However, if I did this:
>
> mpirun --host Y --np 3 sleep 1000
>
> I see 3 instances of "sleep" when I do ps -aedf. Does mpirun try to
> "ssh" all networked machines
Raymond Wan wrote:
>
> Hi Jeff,
>
> Some "good" news (but still some bad news). Y and Z are part of a set
> of 8 machines and I found out that mpirun works for one of them. I
> didn't checked a couple of them before -- sorry! However, I'm no closer
> to the solution since all 8 should be "iden
Hi Marcia
You have to tell the process doing the lookup the connection info for
the server that is hosting the published info. Otherwise, OMPI
defaults to looking for it on its own mpirun, where it won't be found
in this case.
Here is what you need to do (described in "man mpirun"):
1. m
Hi Ralph,
Helps a lot!
But the option "--report-uri" is unable to mpirun. After some searches in
OpenMPI lists I found a way to run:
1) ompi-server -d --report-uri urifile & -> (must be always running)
2) mpirun -ompi-server file:urifile -np 1 publish-test
3) mpirun -ompi-server file:urifile
Ah - you are quite correct. I had forgotten that option is coming in
1.3.1, and wasn't in 1.3.0.
Glad you found the other solution!
Ralph
On Mar 16, 2009, at 9:44 AM, Marcia Cristina Cera wrote:
Hi Ralph,
Helps a lot!
But the option "--report-uri" is unable to mpirun. After some
searche
On Mar 15, 2009, at 1:32 PM, Yury Tarasievich wrote:
Could you please elaborate? I'm not experienced
with MPI and I need to test a heterogenous
installation, to see whether the computational
task is actually broken down and computed in
collaborative manner.
It sounds like you're asking an app
Jeff Squyres wrote:
On Mar 15, 2009, at 1:32 PM, Yury Tarasievich wrote:
Could you please elaborate? I'm not experienced
with MPI and I need to test a heterogenous
installation, to see whether the computational
task is actually broken down and computed in
collaborative manner.
It sounds like
Hi Josh and Jeff,
I found mpicc in /usr/local/lib (where I put it) and when I
tried the command "ldd mpicc" I got the following;
linux-f2f9:/usr/local/bin # ldd mpicc
libopen-pal.so.0 => /usr/local/lib/libopen-pal.so.0 (0x2afcd6bf6000)
libm.so.6 => /lib64/libm.so.6 (0x
You need to have the Intel compilers (both C and Fortran) in your
LD_LIBRARY_PATH so they can resolve the imf library.
On Mar 16, 2009, at 1:05 PM, Amos Leffler wrote:
Hi Josh and Jeff,
I found mpicc in /usr/local/lib (where I put it) and when I
tried the command "ldd mpicc" I got the f
On Mar 16, 2009, at 2:27 PM, Yury Tarasievich wrote:
Well, no. What I would want is, say, MPI variation of "hello world"
that
would output the phrase char by char, one char per the participating
host, and doing it in correct (so, MPI-synchronised) order. Hosts
may be
heterogenous.
Or possi
Amos,
The libraries that ldd isn't finding are intel fortran libraries. If
you put that directory in your load library path things should work.
Doug Reeder
On Mar 16, 2009, at 12:05 PM, Amos Leffler wrote:
Hi Josh and Jeff,
I found mpicc in /usr/local/lib (where I put it) and when I
tr
On Mar 14, 2009, at 12:28 PM, ben rodriguez wrote:
Both machines are x86_64, although one is a single quad core and the
other is a dual quad core, the RH installs are the same. ompi and
the other program were both statically linked. After the compiles I
copied the target dirs to the new mac
There's usually an "icc_vars.sh" script somewhere in your intel
compiler installation (and corresponding "icc_vars.csh" if you're
using C-flavored shells) that will do things like add the intel paths
to LD_LIBRARY_PATH, etc. If you source this file, all should be well.
You may also need to
Jeff Squyres wrote:
On Mar 16, 2009, at 2:27 PM, Yury Tarasievich wrote:
Well, no. What I would want is, say, MPI variation of "hello world" that
Open MPI contains two test programs: "hello world" and "ring" in each
of the 4 languages (C, C++, F77, F90) in the examples/ directory.
Woul
George and I talked more off-list / phone and he convinced me that
it's ok to do this in a top-level header file:
#if defined(c_plusplus) || defined(__cplusplus)
#undef restrict
#define restrict
#endif
I'll post to the Autoconf list about this soon.
On Mar 16, 2009, at 4:43 AM, Samuel Sarhol
Hi Jeff:
I managed to run it just recently... It turns out that some libraries libib*
were missing, as well as some others. I learned this by trying to install an
old version of openmpi that was in the repository of my Suse Linux. The
"software manager" of Suse told me the missing libraries for
Hello everyone,
Could anybody let me know where I can find a manual or an web page that
describes
the specification for specifying the selective file driven decision functions
in Open MPI's mpirun/mpiexec command as explained below ?
> Jelena Pjesivac-Grbovic wrote:
>This is an excellent solut
19 matches
Mail list logo
