[SCIP] SCIP JNI - JNIKnapsack example fails to run on an x86 architecture

[Ambros Gleixner]

Dear David,

We have definitely run SCIP successfully on an ARM machine with Debian.
 I am pretty sure we have not tested ARM+JNI.

But the JNI in general is a tricky interface.  If there is any way
around it, you should consider it.

Currently, our JNI maintainer is away from keyboard, so you will have to
wait a bit until we can look at the bug.  Please submit it also at

   http://scip.zib.de/bugs.php

Thanks,
Ambros



Am 23.06.2016 um 18:00 schrieb David Rua:
> Dear all,
> 
> I don't know if this question was already put, so my apologies for that
> beforehand if it was.
> 
> I'm trying to use the SCIP as an alternative solver to GLPK in a
> raspberry PI platform. Due to some previous code development
> restrictions the objective is to use the JNI of SCIP to run a MILP.
> To that end, I have tried to compile the source code, of SCIPSuite in a
> linux distro, only successfully using a x64 architecture.
> 
> Although the compilation in an x86 architecture is possible, when
> running the JniKnapsack example of the SCIP JNI interface I keep getting
> segmentation faults. This does not happen in the x64 counterpart.
> 
> This issue was partly reported by me in SO and the output of the
> SEGFAULT can be checked there:
> http://stackoverflow.com/questions/37964358/scip-jni-jniknapsack-example-fails-to-run
> 
> 
> Since there is an x86 package version for download at the SCIP site I
> guessing that something may not be correct with the SCIPSuite source code.
> 
> The broader question still stands: can we run SCIP in an ARM based
> architecture like a raspberry PI?
> 
> Any thoughts on this are most welcomed!
> Thanks.
> 
> Best,
> David
> 
> _______________________________________________
> Scip mailing list
> Scip at zib.de
> http://listserv.zib.de/mailman/listinfo/scip

-- 
Ambros Gleixner, Zuse Institute Berlin, http://www.zib.de/gleixner