[mvapich-discuss] SIGSEGV in MPIDI_CH3I_MRAIL_Parse_header

[Johannes Ziegenbalg]

Hi.

First of all, I managed to get my code running, but without the "enable
-hybrid" configure option. With it, the bug still appears.

The system is a Bull HPC-Custer. One Node has 2 Xeon E5-2670 CPUs and 
a Mellanox MT27500 Family [ConnectX-3] FDR-Infiniband Adapter.

For more specific information about the System you can look at:
https://doc.zih.tu-dresden.de/hpc-wiki/bin/view/Compendium/SystemTaurus

I'm using the current 2.1 version of MVAPICH2, together with Slurm
14.11.6.  My configure call is:  ./configure --enable-threads=multiple 
--enable-fortran=all --with-pm=slurm --with-pmi=pmi1 --with-rdma=gen2 -
-enable-hybrid --with-device=ch3:mrail --enable-g=dbg,log,most -q -
-enable-debuginfo --disable-fast
The compiler is a GCC 4.8.2.

Attached are small versions of the server and client. With them I was
able to reproduce the Bug. You can execute them with: 
MV2_SUPPORT_DPM=1 PORT_FILE="Server_port.txt" ./server
MV2_SUPPORT_DPM=1 PORT_FILE="Server_port.txt" ./client

Johannes Ziegenbalg 

On Do, 2015-06-25 at 00:06 -0400, Jian Lin wrote:
> Hi, Johannes,
> 
> Thanks for your notes! However, we didn't observe such issue in our
> tests. Will you please provide more detailed information on your
> environment, including platform, network adapter, MVAPICH2 version,
> configuration options, and the runtime command-line? Also, will you
> please provide a reproducer that can trigger the issue? With those, 
> we can further explore it. Thank you!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: client.c
Type: text/x-csrc
Size: 872 bytes
Desc: not available
URL: <http://mailman.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20150625/2f007a36/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: server.c
Type: text/x-csrc
Size: 1067 bytes
Desc: not available
URL: <http://mailman.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20150625/2f007a36/attachment-0001.bin>