[Wien] SIGBUS occurred in lapw1 on RHEL6.2

Peter Blaha pblaha at theochem.tuwien.ac.at
Wed Jun 20 08:46:36 CEST 2012 

When I tested     composer_xe_2011_sp1.10.319   I got similar problems during runtime
in the libiomp5.so library.

Intel is producing severe bugs in at least every second compiler version they release.

We are using:   composerxe-2011.2.137   (64 bit version)
and this seems to work fine.

I'd recommend to try this compiler version.

Am 20.06.2012 03:54, schrieb Nobuaki Igarashi:
> Thank you for your reply.
>
>> Which version are you using?
>
> We installed Cluster Studio XE as Intel compiler.
> Its version is composer_xe_2011_sp1.6.233.
> $ ifort -v
> ifort version 12.1.0
> $ ./getversion.out
> Intel(R) Math Kernel Library Version 10.3.6 Product Build 20110809 for
> Intel(R) 64 architecture applications
>
> My RHEL5.5 machine, on which I could run wien2k without error, is
> $ ifort -v
> Version 12.0.2
> $ ./getversion.out
> Intel(R) Math Kernel Library Version 10.3.2 Product Build 20110112 for
> Intel(R) 64 architecture applications
> composer's version is composerxe-2011.2.137.
>
> Thanks
> N.Igarashi
>
> Laurence Marks wrote:
>
>> Did you search the mailing list for ifort versions -- several are
>> buggy, see http://zeus.theochem.tuwien.ac.at/pipermail/wien/2011-November/015682.html
>> as well as earlier emails.
>>
>> Which version are you using?
>>
>> On Tue, Jun 19, 2012 at 7:16 PM, Nobuaki Igarashi
>> <igarashi at imr.tohoku.ac.jp> wrote:
>>> Thank you for your quick reply.
>>>
>>> Regarding "Bug 11357 - SIGBUS in getenv(). ", I've already check.
>>> But it seems problem of the programing method.
>>>
>>> I removed " -warn nogeneral ?-g -openmp -static-libgcc".
>>> (I set the option "-static-libgcc" to copy the executables compiled on
>>> RHEL5.5.)
>>> The "-i-static" option has already been deprecated currently, ifort tells us
>>> to use -static-intel instead.
>>>
>>> I compiled all modules with
>>> "-free -w -mp1 -prec-div -pc80 -DINTEL_VML -traceback -static-intel".
>>> However, now, SIGSEGV occurred in openmp library.
>>>
>>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>>> Image ? ? ? ? ? ? ?PC ? ? ? ? ? ? ? ?Routine ? ? ? ? ? ?Line ? ? ? ?Source
>>> libiomp5.so ? ? ? ?00007F77009F4A5A ?Unknown ? ? ? ? ? ? ? Unknown ?Unknown
>>> libiomp5.so ? ? ? ?00007F77009E128C ?Unknown ? ? ? ? ? ? ? Unknown ?Unknown
>>> libiomp5.so ? ? ? ?00007F77009DDBB0 ?Unknown ? ? ? ? ? ? ? Unknown ?Unknown
>>> libiomp5.so ? ? ? ?00007F77009DD238 ?Unknown ? ? ? ? ? ? ? Unknown ?Unknown
>>> libiomp5.so ? ? ? ?00007F77009FAEBB ?Unknown ? ? ? ? ? ? ? Unknown ?Unknown
>>> libpthread.so.0 ? ?000000379F4077F1 ?Unknown ? ? ? ? ? ? ? Unknown ?Unknown
>>> libc.so.6 ? ? ? ? ?000000379F0E5CCD ?Unknown ? ? ? ? ? ? ? Unknown ?Unknown
>>>
>>> Thanks
>>> N.Igarashi
>>>
>>> Laurence Marks wrote:
>>>
>>>> A google search on "sigbus getenv", for instance
>>>> http://sources.redhat.com/ml/glibc-bugs/2010-03/msg00005.html suggests
>>>> that this might be a known bug. I think you will have to try various
>>>> things until they work. Some ideas:
>>>>
>>>> a) Check against the mailing list archive to see if there any other
>>>> reports of bugs with this version of ifort. Of late there have been
>>>> problems.
>>>> b) remove " -warn nogeneral ?-g -openmp -static-libgcc". I don't think
>>>> -g is a good idea except when writing code, and I do not believe
>>>> Wien2k currently uses openmp anywhere, everything is blas.
>>>> c) If this does not work, also remove "-pad"
>>>> d) If this does not work, add "-i-static".
>>>>
>>>> On Sun, Jun 17, 2012 at 6:54 PM, Nobuaki Igarashi
>>>> <igarashi at imr.tohoku.ac.jp> wrote:
>>>>> Dear developers and WIEN2k users,
>>>>>
>>>>> Although I completed compiling WIEN2k along siteconfig procedure, lapw1
>>>>> was abnormally ended as soon as it started by SIGBUS.
>>>>>
>>>>> The following is the debugger's output.
>>>>> (idb) run
>>>>> Starting program: /home007/lacse/my_wien2k/wien2k.org/lapw1
>>>>> [New Thread 8144 (LWP 8144)]
>>>>> Program received signal SIGBUS
>>>>> getenv () in /lib64/libc-2.12.so
>>>>> (idb) where
>>>>> #0 ?0x000000332c4352c0 in getenv () in /lib64/libc-2.12.so
>>>>> #1 ?0x0000000000488288 in for_check_env_name () in
>>>>> /home007/lacse/my_wien2k/wien2k.org/lapw1
>>>>> #2 ?0x00000000004ccfa2 in for__open_proc () in
>>>>> /home007/lacse/my_wien2k/wien2k.org/lapw1
>>>>> #3 ?0x000000000048d908 in for_open () in
>>>>> /home007/lacse/my_wien2k/wien2k.org/lapw1
>>>>> #4 ?0x00000000004184f3 in errclr (fname=(...)) at
>>>>> /home007/lacse/my_wien2k/wien2k.org/SRC_lapw1/errclr.f:64
>>>>> #5 ?0x0000000000443e92 in lapw1 () at
>>>>> /home007/lacse/my_wien2k/wien2k.org/SRC_lapw1/lapw1_tmp_.F:90
>>>>> #6 ?0x0000000000404c5c in main () in
>>>>> /home007/lacse/my_wien2k/wien2k.org/lapw1
>>>>> #7 ?0x000000332c41ecdd in __libc_start_main () in /lib64/libc-2.12.so
>>>>>
>>>>> This issue occurred on only RHEL 6.2, there is no problem on RHEL5.5.
>>>>> Now, I compiled WIEN2k with -static-libgc, which is one of options of Intel
>>>>> compiler, on RHEL5.5,
>>>>> I copied it to new server. ?Then, no error occurred.
>>>>>
>>>>> Although it may come from a problem of glibc of RHEL 6.2,
>>>>> had you had such a information, yet?
>>>>> If you know regarding this, please let me know anything.
>>>>>
>>>>> -----------------------------------------------------------------------
>>>>> I write our system informations :
>>>>>
>>>>> [The new system ]
>>>>> IBM BladeCenter HS22
>>>>> Operating System : Red Hat Enterprise Linux ComputeNode release 6.2
>>>>> (Santiago)
>>>>> CPU model : Intel(R) Xeon(R) CPU X5650 @ 2.67GHz
>>>>> CPU cores : 12
>>>>> Memory : 24594924kB (free 14196424kB)
>>>>> Kernel : Linux 2.6.32-220.7.1.el6
>>>>> GLIBC : GNU C Library stable release version 2.12, by Roland McGrath et al.
>>>>>
>>>>> [RHEL 5.5 Systems]
>>>>> IBM iDataPlex dx360
>>>>> Operating System : Red Hat Enterprise Linux Server release 5.5 (Tikanga)
>>>>> CPU model : Intel(R) Xeon(R) CPU X5650 @ 2.67GHz
>>>>> CPU cores : 12
>>>>> Memory : 24GB
>>>>> Kernel : Linux ?2.6.18-194.26.1.el5
>>>>> GLIBC : GNU C Library stable release version 2.5, by Roland McGrath et al.
>>>>>
>>>>> [Compiler]
>>>>> ifort version 12.1.0
>>>>> Intel(R) Math Kernel Library Version 10.3.6 Product Build 20110809 for
>>>>> Intel(R) 64 architecture applications
>>>>>
>>>>> [WIEN2k Compile options]
>>>>> WIEN2k version : WIEN2k_11.1
>>>>> FOPT:-free -mp1 -warn
>>>>> nogeneral -prec-div -pc80 -pad -ip -DINTEL_VML -traceback -g -openmp -static-libgcc
>>>>> current:LDFLAGS:$(FOPT) -L$(MKLROOT)/lib/$(MKL_TARGET_ARCH) -static-libgcc
>>>>> DPARALLEL:'-DParallel'
>>>>> R_LIBS:$(MKLROOT)/lib/intel64/libmkl_blas95_lp64.a
>>>>> $(MKLROOT)/lib/intel64/libmkl_lapack95_lp64.a \
>>>>> -Wl,--start-group $(MKLROOT)/lib/intel64/libmkl_intel_lp64.a \
>>>>> $(MKLROOT)/lib/intel64/libmkl_intel_thread.a
>>>>> $(MKLROOT)/lib/intel64/libmkl_core.a \
>>>>> -Wl,--end-group -openmp -lpthread -lm
>>>>>
>>>>> * Not MPI version, but MKL + OpenMP version
>>>>>
>>>>> Thanks in advance.
>>>>> N.Igarashi
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Wien mailing list
>>>>> Wien at zeus.theochem.tuwien.ac.at
>>>>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>>>
>>>>
>>>>
>>>> --
>>>> Professor Laurence Marks
>>>> Department of Materials Science and Engineering
>>>> Northwestern University
>>>> www.numis.northwestern.edu 1-847-491-3996
>>>> "Research is to see what everybody else has seen, and to think what
>>>> nobody else has thought"
>>>> Albert Szent-Gyorgi
>>>> _______________________________________________
>>>> Wien mailing list
>>>> Wien at zeus.theochem.tuwien.ac.at
>>>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Wien mailing list
>>> Wien at zeus.theochem.tuwien.ac.at
>>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>>
>>
>> --
>> Professor Laurence Marks
>> Department of Materials Science and Engineering
>> Northwestern University
>> www.numis.northwestern.edu 1-847-491-3996
>> "Research is to see what everybody else has seen, and to think what
>> nobody else has thought"
>> Albert Szent-Gyorgi
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>

-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------

More information about the Wien mailing list