FWIW, we have seen bugs in the intel compiler suite before.  We usually advise 
people to get the latest version of the particular intel compiler suite version 
that they have a license to obtain.

I don't know if that is the exact cause of the issue here, but it's definitely 
something we've seen before.


On Nov 2, 2012, at 5:02 PM, Orion Poplawski wrote:

> On 11/01/2012 09:56 AM, J R Jones wrote:
>> Hi
>> 
>> I am trying to build OpenMPI 1.6.3 with the Intel compilers, but one of the
>> tests in make check fails with a segmentation fault.  I wondered if anyone
>> else has come across this and if so, how they got around it?
>> 
>> $ icc --version
>> icc (ICC) 12.0.4 20110427
>> 
>> I have logged the output of the build process and include it  as a tarball.
>> The interesting part seems to be this, in make.check:
>> 
>> [blue34:26873] Datatype 0x6794570[] size 36000 align 8 id 0 length 6 used 3
>> true_lb 0 true_ub 39592 (true_extent 39592) lb 0 ub 39592 (extent 39592)
>> nbElems 4500 loops 2 flags 104 (commited )-c-----GD--[---][---]
>>    contain OPAL_FLOAT8
>> --C--------[---][---]      OPAL_LOOP 450 times the next 2 elements extent 88
>> --C---P-D--[---][---]    OPAL_FLOAT8 count 10 disp 0x0 (0) extent 8 (size 80)
>> --C--------[---][---]  OPAL_END_LOOP prev 2 elements first elem displacement >> 0
>> size of data 80
>> -------G---[---][---]  OPAL_END_LOOP prev 3 elements first elem displacement >> 0
>> size of data 36000
>> Optimized description
>> --C--------[---][---]      OPAL_LOOP 450 times the next 2 elements extent 88
>> -cC---P-DB-[---][---]     OPAL_UINT1 count 80 disp 0x0 (0) extent 1 (size 80)
>> --C--------[---][---]  OPAL_END_LOOP prev 2 elements first elem displacement >> 0
>> size of data 80
>> -------G---[---][---]  OPAL_END_LOOP prev 3 elements first elem displacement >> 0
>> size of data 36000
>> 
>> /bin/sh: line 5: 26873 Segmentation fault      ${dir}$tst
>> FAIL: opal_datatype_test
>> 
>> Has anyone seen this issue before?
> 
> Seems to be working okay for me with icc (ICC) 13.0.1 20121010 x86_64, though 
> perhaps it's dependent on compile flags...
> 
> -- 
> Orion Poplawski
> Technical Manager                     303-415-9701 x222
> NWRA, Boulder Office                  FAX: 303-415-9702
> 3380 Mitchell Lane                       or...@nwra.com
> Boulder, CO 80301                   http://www.nwra.com
> _______________________________________________
> users mailing list
> us...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/users


-- 
Jeff Squyres
jsquy...@cisco.com
For corporate legal information go to: 
http://www.cisco.com/web/about/doing_business/legal/cri/


Reply via email to