<p>Can you change everything to complex*16 ? It makes sense to do everything with double precision, I would be concerned with the accuracy of complex*8 as well as mixing precisions.</p>
<p>---------------------------<br>
Professor Laurence Marks<br>
Department of Materials Science and Engineering<br>
Northwestern University<br>
<a href="http://www.numis.northwestern.edu">www.numis.northwestern.edu</a> 1-847-491-3996<br>
"Research is to see what everybody else has seen, and to think what nobody else has thought"<br>
Albert Szent-Gyorgi<br>
</p>
<div class="gmail_quote">On Aug 23, 2012 6:19 PM, "Gavin Abo" <<a href="mailto:gsabo@crimson.ua.edu">gsabo@crimson.ua.edu</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000">
<div>The situational problem with the fftpack routine might be due to some inconsistency in the array usage. Maybe Prof. Blaha can provide or confirm whether the fix below works properly.<br>
<br>
Lines 392-294 in SRC_lapw0/fft_modules.F:<br>
<br>
real(kind=8) :: DWORK(:)<br>
complex(kind=8) :: CWORK(:)<br>
complex(kind=8) :: C(LDC1,LDC2,N3,2)<br>
<br>
Lines 21-22 in SRC_lapw0/vresp.F:<br>
<br>
DOUBLE PRECISION DWORK(:)<br>
COMPLEX*16 CWORK(:)<br>
<br>
It runs without error with the following changes.<br>
<br>
Lines 392-293 in SRC_lapw0/fft_modules.F:<br>
<br>
real(kind=8) DWORK(*)<br>
complex(kind=8) CWORK(*)<br>
<br>
"complex(kind=8) C(LDC1,LDC2,N3,2)" needed too?? <br>
<br>
Lines 21-22 in SRC_lapw0/vresp.F:<br>
<br>
DOUBLE PRECISION DWORK(*)<br>
COMPLEX*16 CWORK(*)<br>
<br>
as the subroutine in SRC_lapw0/fftpack_helpers.f has "DWORK(*)"<br>
<br>
On 8/23/2012 12:58 PM, Gavin Abo wrote:<br>
</div>
<blockquote type="cite">
<div>I was able to reproduce the error with your files when the fftpack routine is used in Wien2k 12.1. Ran a couple cycles, and the error did not appear when fftw3 was used instead. So a possible solution may be to use the fftw3 library.<br>
<br>
The fftw3 may be faster than the fftpack, so you probably want to use it anyway. It should be easy to use the fftw3 on Debian. fftw3 might already be installed or I believe you can install it with:<br>
<br>
<b>apt-get install libfftw3-dev</b><br>
<br>
Open the Makefile in a text editor:<br>
<br>
<b>vi $WIENROOT/SRC_lapw0/Makefile</b><br>
<br>
Edit and save settings to use for sequential (non-mpi):<br>
<br>
FOPT = -FR -mp1 -w -prec_div -pc80 -pad -ip -DINTEL_VML <b>-DFFTW3</b> -traceback<br>
R_LIBS = -lmkl_lapack95_lp64 -lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core -openmp -lpthread<b> -lfftw3</b><br>
<br>
In $WIENROOT/SRC_lapw0:<br>
<br>
<b>make<br>
cp lapw0 ..<br>
</b><br>
As described in section 11.1.1 of the Wien2k 12.1 userguide, the same can be done with lapw2. Instructions for using the possible faster mkl-fftw3 for sequential (non-mpi) instead of the above described fftw3 from a Debian repository is also given.<br>
<br>
On 8/23/2012 8:06 AM, Luis Carlos Ogando Dacal wrote:<br>
</div>
<blockquote type="cite">
Dear Wien2k users and developers,
<div><br>
</div>
<div> I would like to report the same problem sent to the list by Dr. Eitel Peltzer.</div>
<div> I am running WIEN2k_12.1 on a DELL Precision workstation with two QuadCore Xeon processors and Debian Linux. It was compiled using ifort 2011.3.174, icc and MKL. The compilation options were:</div>
<div><br>
</div>
<div>
<div> O Compiler options: -FR -mp1 -w -prec_div -pc80 -pad -ip -DINTEL_VML -traceback</div>
<div> L Linker Flags: $(FOPT) -L/opt/intel/composerxe-2011.3.174/mkl/lib/intel64 -pthread</div>
<div> P Preprocessor flags '-DParallel'</div>
<div> R R_LIB (LAPACK+BLAS): -lmkl_lapack95_lp64 -lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core -openmp -lpthread</div>
<div><br>
</div>
<div> I have followed section 4.5.9 of the Users Guide and everything is 0K until the change of indxc to 28 in case.in0 and 50 in case.in0_grr. After this, the SCF cycle stops in the second lapw0 run with the following error message: </div>
<div><br>
</div>
<div>
<div>hup: Command not found.</div>
<div> LAPW0 END</div>
<div>forrtl: severe (174): SIGSEGV, segmentation fault occurred</div>
<div>Image PC Routine Line Source </div>
<div>lapw0 000000000040505E c3fft_1_ 119 fftpack_helpers.f</div>
<div>lapw0 0000000000412D5B fftpack_mp_c3fft_ 397 fft_modules.F</div>
<div>lapw0 000000000047F4EC vresp_ 106 vresp.F</div>
<div>lapw0 0000000000495769 xcpot3_ 147 xcpot3.F</div>
<div>lapw0 000000000045C064 MAIN__ 1935 lapw0.F</div>
<div>lapw0 0000000000403D6C Unknown Unknown Unknown</div>
<div>libc.so.6 00002B1096C82C8D Unknown Unknown Unknown</div>
<div>lapw0 0000000000403C69 Unknown Unknown Unknown</div>
<div><br>
</div>
<div>> stop error</div>
</div>
<div><br>
</div>
<div><br>
</div>
<div> I am sending the case.struct and case.in0 files as requested by Prof. Blaha. I have also saved the previous PBE calculation and I can send it if necessary (8.5 MB).</div>
<div> Thanks in advance,</div>
<div> Luis Ogando</div>
<div></div>
</div>
</blockquote>
</blockquote>
<br>
</div>
</blockquote></div>