Hi,
mpirun internally uses ssh to launch a program on multiple nodes.
I would like to see the various parameters that are sent to each of
the nodes. How can I do this?
-chev
Dear Open MPI users list,
From time to time, I experience a mutex deadlock in Open-MPI 1.1.2. The stack
trace is available at the end of the mail. The deadlock seems to be caused by
lines 118 & 119 of the ompi/mca/btl/tcp/btl_tcp.c file, in function
mca_btl_tcp_add_procs:
OBJ_RELEASE(t
On Mon, Jan 29, 2007 at 10:49:10PM -0800, Chevchenkovic Chevchenkovic wrote:
> Hi,
Hi
> mpirun internally uses ssh to launch a program on multiple nodes.
> I would like to see the various parameters that are sent to each of
> the nodes. How can I do this?
You mean adding "pls_rsh_debug=1" to you
The master process uses both MPI_ANY_SOURCE and MPI_ANY_TAG while
waiting for requests from slave processes. The slaves sometimes use
MPI_ANY_TAG but the source is always specified.
We have run the code through valgrid for a number of cases including the
one being used here.
The code is Fortran
On Jan 30, 2007, at 9:35 AM, Fisher, Mark S wrote:
The master process uses both MPI_ANY_SOURCE and MPI_ANY_TAG while
waiting for requests from slave processes. The slaves sometimes use
MPI_ANY_TAG but the source is always specified.
I think you said that you only had corruption issues on the s
The slaves send specific requests to the master and then waits for a
reply to that request. For instance it might send a request to read a
variable from the file. The master will read the variable and send it
back with the same tag in response. Thus there is never more than one
response at a time t
Is there any way that you can share the code?
On Jan 30, 2007, at 9:57 AM, Fisher, Mark S wrote:
The slaves send specific requests to the master and then waits for a
reply to that request. For instance it might send a request to read a
variable from the file. The master will read the variable a
The code can be Freely downloaded for US citizens (it is export
controlled) at http://zephyr.lerc.nasa.gov/wind/. I can also provide you
the test case which is very small. I am a developer of the code and can
help you dig through it if you decide to download it. On the above page
you will need to r
Jeremy,
You're right. Thanks for point it out. I do the change in the trunk.
george.
On Jan 30, 2007, at 3:40 AM, Jeremy Buisson wrote:
Dear Open MPI users list,
From time to time, I experience a mutex deadlock in Open-MPI 1.1.2.
The stack
trace is available at the end of the mail. The d
Hello,
I see the extern definitions in mpi.h for ompi_mpi_2cplex and
ompi_mpi_2dblcplex, but no #define for MPI_2COMPLEX and MPI_2DOUBLE_COMPLEX.
Greetings
Bert Wesarg
Jeff,
Upgrading to 1.1.3 solved both issues - thank you very much!
Avishay
On Mon, 2007-01-29 at 20:59 -0500, Jeff Squyres wrote:
> I'm quite sure that we have since fixed the command line parsing
> problem, and I *think* we fixed the mmap problem.
>
> Is there any way that you can upgrade to
Please note that due to a mixup in the 1.1.3 release, we just
released v1.1.4. :-(
See http://www.open-mpi.org/community/lists/announce/2007/01/0010.php
for the official announcement.
The short version is that the wrong tarball was posted to the OMPI
web site for the 1.1.3 release (doh!)
Hi,
Recently I installed OpenMPI 1.1.4 using the source RPM on Fedora Core
6. then I tried to run some benchmarks from NASA. First I tried is
some I/O benchmarks, It compiles, but when I run it, it generates the
following error:
[abc:25584] *** An error occurred in MPI_Type_create_subarray
[abc
13 matches
Mail list logo
