[Users] OpenSM Simulator

[Parker, Matthew - IS]

Thanks Hal!

I've just started learning about Infiniband and we're having issues with our setup (OFED and multiple OpenSM managers) in production. I wanted a tool that I could test different scenarios to see if they would shed some light on the problems. Would you recommend one over the other?

Regards,

Matt

________________________________
From: Hal Rosenstock [hal.rosenstock at gmail.com]
Sent: Monday, July 22, 2013 9:17 PM
To: Parker, Matthew - IS
Cc: Users at lists.openfabrics.org
Subject: Re: [Users] OpenSM Simulator



On Mon, Jul 22, 2013 at 7:49 PM, Parker, Matthew - IS <Matthew.Parker at exelisinc.com<mailto:Matthew.Parker at exelisinc.com>> wrote:
TWIMC:

I would like to use the OpenSM with IBMgtSim.

I have installed IBMgtSim and create a small topology and net list file. It starts without error.

As I understand it, opensm must be recompiled to work with ibmgtsim.

I've tried recompiling OpenSM (3.3.15-1) on CentOS (x64 arch).

./autogen.sh (works without issue).

Running ./configure --with-osmv=sim, it trys to find libraries in /usr/lib, but all the dependencies are under /usr/lib64. Looking at the [opensm home directory]/configure file, the paths are hardcoded to $with_sim/lib, where $with_sim=/usr as the default. I modified them to point to $with_sim/lib64 where appropriate. ./configure --with-osmv=sim now runs to competition.

Running the command, make, I now get the following compiliation errors while compiling the libvendor module:

[root at localhost libvendor]# make
Making all in .
make[1]: Entering directory `/home/mparker/Programs/opensm-3.3.15/libvendor'
/bin/sh ../libtool  --tag=CC   --mode=compile gcc -DHAVE_CONFIG_H -I. -I../include -I../include/opensm -I/usr/include -I./../include   -Wall -g -g -O2 -MT libosmvendor_la-osm_pkt_randomizer.lo -MD -MP -MF .deps/libosmvendor_la-osm_pkt_randomizer.Tpo -c -o libosmvendor_la-osm_pkt_randomizer.lo `test -f 'osm_pkt_randomizer.c' || echo './'`osm_pkt_randomizer.c
libtool: compile:  gcc -DHAVE_CONFIG_H -I. -I../include -I../include/opensm -I/usr/include -I./../include -Wall -g -g -O2 -MT libosmvendor_la-osm_pkt_randomizer.lo -MD -MP -MF .deps/libosmvendor_la-osm_pkt_randomizer.Tpo -c osm_pkt_randomizer.c  -fPIC -DPIC -o .libs/libosmvendor_la-osm_pkt_randomizer.o
osm_pkt_randomizer.c: In function ‘osm_pkt_randomizer_mad_drop’:
osm_pkt_randomizer.c:238: warning: passing argument 3 of ‘osm_dr_path_init’ makes pointer from integer without a cast
../include/opensm/osm_path.h:147: note: expected ‘const uint8_t *’ but argument is of type ‘uint8_t’
osm_pkt_randomizer.c:238: error: too many arguments to function ‘osm_dr_path_init’
make[1]: *** [libosmvendor_la-osm_pkt_randomizer.lo] Error 1
make[1]: Leaving directory `/home/mparker/Programs/opensm-3.3.15/libvendor'
make: *** [all-recursive] Error 1

I was wonder whether anyone else has had any issues recompiling opensm, and whether they we're able to fix it.

Please try the attached patch.

FWIW, there's another simulator (ibsim) which does not require a different OpenSM "vendor". Is there a particular reason you want to use ibmgtsim ? Just curious...

-- Hal


TIA,

Matt

________________________________

This e-mail and any files transmitted with it may be proprietary and are intended solely for the use of the individual or entity to whom they are addressed. If you have received this e-mail in error please notify the sender. Please note that any views or opinions presented in this e-mail are solely those of the author and do not necessarily represent those of Exelis Inc. The recipient should check this e-mail and any attachments for the presence of viruses. Exelis Inc. accepts no liability for any damage caused by any virus transmitted by this e-mail.

_______________________________________________
Users mailing list
Users at lists.openfabrics.org<mailto:Users at lists.openfabrics.org>
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/users


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/users/attachments/20130723/d59176de/attachment.html>