[Wien] MKL FATAL ERROR on load the function mkl_blas_dpotrf_u_small

[Laurence Marks]

Read!

You have mixed linking to "Compiler/11.1" and "composerxe-2011.3.191".
These are different versions, and can easily lead to chaos. Yes, this
will require that you read up and understand a bit about linking,
environments and similar -- and you need to do this as it is not
possible for me or anyone else to guess exactly where things are wrong
by email. You can probably find a sysadmin or someone else at your
institution who is familiar with linux to help you, buy them a beer
(or flowers).

2011/11/10 sufyan <sufeer03 at gmail.com>:
> Thank you for interest in my e-mail,i used this command  and i have
>
>  linux-vdso.so.1 =>  (0x00007fff4c565000)
>         libmkl_lapack.so =>
> /home/sufyan2020/intel/Compiler/11.1/064/mkl/lib/em64t/libmkl_lapack.so
> (0x00002b4cad5bc000)
>         libmkl_intel_lp64.so =>
> /home/sufyan2020/intel/composerxe-2011.4.191/mkl/lib/intel64/libmkl_intel_lp64.so
> (0x00002b4cae211000)
>         libmkl_intel_thread.so =>
> /home/sufyan2020/intel/composerxe-2011.4.191/mkl/lib/intel64/libmkl_intel_thread.so
> (0x00002b4cae9ef000)
>         libmkl_core.so =>
> /home/sufyan2020/intel/composerxe-2011.4.191/mkl/lib/intel64/libmkl_core.so
> (0x00002b4cafc78000)
>         libpthread.so.0 => /lib64/libpthread.so.0 (0x00002b4cb0c9c000)
>         libguide.so =>
> /home/sufyan2020/intel/Compiler/11.1/064/lib/intel64/libguide.so
> (0x00002b4cb0eb9000)
>         libm.so.6 => /lib64/libm.so.6 (0x00002b4cb105e000)
>         libiomp5.so =>
> /home/sufyan2020/intel/composerxe-2011.4.191/compiler/lib/intel64/libiomp5.so
> (0x00002b4cb12b3000)
>         libdl.so.2 => /lib64/libdl.so.2 (0x00002b4cb14a2000)
>         libc.so.6 => /lib64/libc.so.6 (0x00002b4cb16a6000)
>         libgcc_s.so.1 => /lib64/libgcc_s.so.1 (0x00002b4cb1a01000)
>         /lib64/ld-linux-x86-64.so.2 (0x00002b4cad39d000)
>
> but i do not find the problem??
> thanks for your helps
> sufyan naji
> 2011/11/10 Laurence Marks <L-marks at northwestern.edu>
>>
>> For something like this it is ALWAYS a good idea to put the error into
>> a google search, e.g. search on
>>
>> MKL FATAL ERROR on load the function mkl_blas_dpotrf_u_small
>>
>> This does not bring up an exact match, but the link
>>
>> http://software.intel.com/en-us/forums/showthread.php?t=61978
>>
>> looks useful. This suggests that you have something wrong with how the
>> mkl libraries have been linked in. Did you use -istatic which will
>> force them to be static linked, and do you have LD_LIBRARY_PATH setup
>> right and exported? A useful command is ldd, e.g.
>>
>> ldd /home/sufyan2020/WIENROOT/lapw1
>>
>> This will show you the shared libraries that lapw1 is expecting to
>> find, and will say if they are not available. This should get you
>> started on some debugging of the problem.
>>
>> 2011/11/10 sufyan <sufeer03 at gmail.com>:
>> >
>> >
>> > Dears users,
>> > i am new user in wien2k community  and i need your help
>> > i run my case  with the version 11.1 but i have
>> >
>> > using WIEN2k_11.1 (Release 5/4/2011) in /home/sufyan2020/WIENROOT
>> >
>> >
>> >     start     (Thu Nov 10 12:58:20 WET 2011) with lapw0 (40/99 to go)
>> >
>> >     cycle 1     (Thu Nov 10 12:58:20 WET 2011)     (40/99 to go)
>> >
>> >>   lapw0     (12:58:20) 8.292u 0.108s 0:08.40 99.8%    0+0k 0+3720io
>> >> 0pf+0w
>> >>   lapw1  -up        (12:58:29) MKL FATAL ERROR on load the function
>> >> mkl_blas_dpotrf_u_small
>> > 0.820u 0.036s 0:01.01 84.1%    0+0k 0+1216io 0pf+0w
>> > error: command   /home/sufyan2020/WIENROOT/lapw1 uplapw1.def   failed
>> >
>> >>   stop error
>> >
>> > Thank you for helps.
>> >        Sincerely,
>> >    sufyan Naji
>> >
>> > _______________________________________________
>> > 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