Run MPI_Send on MPI1.8.5 without multithread enabled:
it hangs on mca_pml_ob1_send() -> opal_progreses() ->
btl_openib_component_progress() -> poll_device() -> libmlx4-rdmav2.so -> cq
-> phread_spin_unlock
The program can run on TCP with no error.
I thought he already tried a larger -Xmx setting of 2GB, which is more
than enough for the simple HelloWord application
On Wed, Jan 20, 2016 at 5:11 PM, Gilles Gouaillardet <
gilles.gouaillar...@gmail.com> wrote:
> the limit could come from the JVM.
> iirc, you have to add extra flags to the jav
the limit could come from the JVM.
iirc, you have to add extra flags to the java command line so the JVM can
allocate more memory.
java -Xmx=...
or something like that (and that could be JVM dependent)
Cheers,
Gilles
On Thursday, January 21, 2016, Saliya Ekanayake wrote:
> Hi Ibrahim,
>
> This
Hi Ibrahim,
This is most likely due to a ulimit setting. Check with *ulimit -a*
and set *ulimit
-n unlimited* and *ulimit -l unlimited*.
I've experienced the same problem in the past and this was the reason.
Hope this helps
On Wed, Jan 20, 2016 at 9:49 AM, Ibrahim Ikhlawi
wrote:
>
>
> Hi Howa
Hi Howard,
I use 64bit JVM. The problem is not in the code, because I am still getting the
same error when I run the Hello.java from examples folder.
Any other suggestion?
Ibrahim
List-Post: users@lists.open-mpi.org
Date: Tue, 19 Jan 2016 09:48:24 -0700
From: hpprit...@gmail.com
To: us...@op
Hi Gilles,
thank you very much for your help! Your workaround works on my
Linux machine as well.
Best regards
Siegmar
On 01/20/16 03:59, Gilles Gouaillardet wrote:
Siegmar,
that looks like a libtool issue : "-z noexecstack" is not correctly passed to cc
a simple workaround is to append "-
