<font face="'PrimaSans BT,Verdana,sans-serif'">I do apologize, <br>I made a mistake when I paste the string that I use to compile it.<br>Anyway I tried both the static and the dynamics compilation using either this string <br>ifort -o resp_mat.x resp_mat.f90 \<br>-L/opt/intel/mkl/10.1.2.024/lib/32 -lmkl_lapack -lmkl_ia32 -lguide -lpthread<br>and the output is : <br><br>resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED.<br>resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED.<br>resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED.<br>resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED.<br>/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_fun'<br>/opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so: undefined reference to `mkl_serv_load_dll'<br><br>and <br>ifort -o resp_mat.x resp_mat.f90 \<br>-L/opt/intel/mkl/10.1.2.024/lib/32 <font face="'PrimaSans BT,Verdana,sans-serif'">-lmkl_lapack64 -lmkl -lguide -lpthread</font><br><br>resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED.<br>resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED.<br>resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED.<br>resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED.<br>resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED.<br>ld: cannot find -lmkl_lapack64<br><br>I
cannot manage this problem. If it doesn't matter, can you please send
me, privately, your compiled version? I am struggling a lot.<br>Thank you, Piero</font><br><br>----- Original Message -----<br>From: Gabriele Sclauzero <sclauzer@sissa.it><br>Date: Monday, May 25, 2009 3:18 pm<br>Subject: Re: [Pw_forum] Difficulties compiling comp_resp_mat.j to optimize the U<br>To: PWSCF Forum <pw_forum@pwscf.org><br><br>> I thought Axel already replied to this post...<br>> <br>> Pieremanuele Canepa wrote:<br>> > Dear all,<br>> > I'm trying to compile the comp_resp_mat.j concerning the <br>> DFT+U. Then, I <br>> > tried to use the default settings (ifort-o r.x resp_mat.f90 <br>> > -L/soft/local/intel/mkl72/lib/32 -lmkl_lapack \ -lmkl_ia32 -<br>> lguide <br>> > -lpthread) and what I got, is reported below. Do you have any <br>> suggestions?> Probably I am making some stupid mistake. <br>> have somebody already <br>> > compiled this program on 32 bit machine?<br>> > I don't have any experience compiling with lapack libraries as <br>> you can <br>> > probably realize.<br>> > Cheers, Piero <br>> <br>> I don't understand if you're trying to link against mkl 7.2 or <br>> 10.1. Do you know at least <br>> which version is installed on your machine and where? Try to <br>> find out this first.<br>> <br>> On my 32 bit Intel machine I've got MKL 9.0 installed in <br>> /usr/local/intel_mkl_90/ and this <br>> compilation command works (for static linking):<br>> <br>> ifort -o resp_mat.x resp_mat.f90 \<br>> -L"/usr/local/intel_mkl_90/lib/32" -<br>> lmkl_lapack -lmkl_ia32 -lguide -lpthread<br>> <br>> ldd resp_mat.x<br>> linux-gate.so.1 <br>> => (0x00af3000)<br>> libguide.so => <br>> /usr/local/intel_mkl_90/lib/32/libguide.so (0x00a55000)<br>> libm.so.6 => <br>> /lib/libm.so.6 (0x00aa4000)<br>> libc.so.6 => <br>> /lib/libc.so.6 (0x00111000)<br>> libgcc_s.so.1 <br>> => /lib/libgcc_s.so.1 (0x00d4c000)<br>> libdl.so.2 => <br>> /lib/libdl.so.2 (0x00acd000)<br>> libpthread.so.0 <br>> => /lib/libpthread.so.0 (0x00ad3000)<br>> /lib/ld-<br>> linux.so.2 (0x00941000)<br>> <br>> <br>> If you want to link dynamically use:<br>> <br>> ifort -o resp_mat.x resp_mat.f90 \<br>> -L"/usr/local/intel_mkl_90/lib/32" -<br>> lmkl_lapack64 -lmkl -lguide -lpthread<br>> <br>> ldd resp_mat.x<br>> linux-gate.so.1 <br>> => (0x00741000)<br>> <br>> libmkl_lapack64.so => <br>> /usr/local/intel_mkl_90/lib/32/libmkl_lapack64.so (0x00c32000)<br>> libmkl.so => <br>> /usr/local/intel_mkl_90/lib/32/libmkl.so (0x00eba000)<br>> libguide.so => <br>> /usr/local/intel_mkl_90/lib/32/libguide.so (0x00111000)<br>> libpthread.so.0 <br>> => /lib/libpthread.so.0 (0x00ad3000)<br>> libm.so.6 => <br>> /lib/libm.so.6 (0x00aa4000)<br>> libc.so.6 => <br>> /lib/libc.so.6 (0x0095f000)<br>> libgcc_s.so.1 <br>> => /lib/libgcc_s.so.1 (0x0015e000)<br>> libdl.so.2 => <br>> /lib/libdl.so.2 (0x00acd000)<br>> /lib/ld-<br>> linux.so.2 (0x00941000)<br>> <br>> Obviously you must change the path to the exact location of your <br>> MKL installation<br>> <br>> <br>> HTH<br>> <br>> GS<br>> <br>> <br>> > <br>> > resp_mat.f90(92): (col. 7) remark: LOOP WAS VECTORIZED.<br>> > resp_mat.f90(93): (col. 7) remark: LOOP WAS VECTORIZED.<br>> > resp_mat.f90(119): (col. 7) remark: LOOP WAS VECTORIZED.<br>> > resp_mat.f90(188): (col. 7) remark: LOOP WAS VECTORIZED.<br>> > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>> > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>> > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>> > resp_mat.f90(235): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>> > resp_mat.f90(252): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED.<br>> > resp_mat.f90(265): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED.<br>> > resp_mat.f90(322): (col. 7) remark: LOOP WAS VECTORIZED.<br>> > resp_mat.f90(339): (col. 7) remark: LOOP WAS VECTORIZED.<br>> > resp_mat.f90(365): (col. 16) remark: PARTIAL LOOP WAS VECTORIZED.<br>> > resp_mat.f90(377): (col. 7) remark: LOOP WAS VECTORIZED.<br>> > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>> > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>> > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>> > resp_mat.f90(483): (col. 13) remark: PARTIAL LOOP WAS VECTORIZED.<br>> > resp_mat.f90(500): (col. 10) remark: PERMUTED LOOP WAS VECTORIZED.<br>> > resp_mat.f90(513): (col. 16) remark: PERMUTED LOOP WAS VECTORIZED.<br>> > /tmp/ifortMUANg9.o: In function `invmat_.':<br>> > resp_mat.f90:(.text+0x9c33): undefined reference to `dcopy_'<br>> > resp_mat.f90:(.text+0x9c4a): undefined reference to `dgetrf_'<br>> > resp_mat.f90:(.text+0x9c6a): undefined reference to `dgetri_'<br>> > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so <br>> > <http://10.1.2.20/lib/32/libmkl_lapack.so>: undefined <br>> reference to <br>> > `mkl_pds_c_blklu_unsym_omp_pardiso'<br>> > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so <br>> > <http://10.1.2.20/lib/32/libmkl_lapack.so>: undefined <br>> reference to <br>> > `mkl_lapack_cpptrs'<br>> > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so <br>> > <http://10.1.2.20/lib/32/libmkl_lapack.so>: undefined <br>> reference to <br>> > `mkl_serv_strcat2'<br>> > /opt/intel/mkl/10.1.2.024/lib/32/libmkl_lapack.so <br>> > <http://10.1.2.20/lib/32/libmkl_lapack.so>: undefined <br>> reference to <br>> > `mkl_pds_blklu_omp_pardiso'<br>> > and so on <br>> > -- <br>> > Pieremanuele Canepa<br>> > Room 230<br>> > School of Physical Sciences, Ingram Building,<br>> > University of Kent, Canterbury, Kent,<br>> > CT2 7NH<br>> > United Kingdom<br>> > -----------------------------------------------------------<br>> > <br>> > <br>> > ---------------------------------------------------------------<br>> ---------<br>> > <br>> > _______________________________________________<br>> > Pw_forum mailing list<br>> > Pw_forum@pwscf.org<br>> > http://www.democritos.it/mailman/listinfo/pw_forum<br>> <br>> -- <br>> <br>> <br>> o ------------------------------------------------ o<br>> | Gabriele Sclauzero, PhD <br>> Student |<br>> | c/o: SISSA & CNR-INFM <br>> Democritos, |<br>> | via Beirut 2-4, <br>> 34014 Trieste (Italy) |<br>> | email: <br>> sclauzer@sissa.it |<br>> | phone: +39 040 3787 <br>> 511 |<br>> | skype: <br>> gurlonotturno |<br>> o ------------------------------------------------ o<br>> _______________________________________________<br>> Pw_forum mailing list<br>> Pw_forum@pwscf.org<br>> http://www.democritos.it/mailman/listinfo/pw_forum<br>> <br><br>---<br>Pieremanuele Canepa<br>Room 230<br>School of Physical Sciences, Ingram Building, <br>University of Kent, Canterbury, Kent, <br>CT2 7NH<br>United Kingdom<br><br>