[Wien] Installation on Intel Mac

[Ashley Harvey]

Here are the options I see from siteconfig_lapw --> compiler options  
(they are exactly those that are recommended):

Current settings:
      O   Compiler options:        -FR -mp1 -w -prec_div -pc80 -pad - 
ip -DINTEL_VML
      L   Linker Flags:            $(FOPT) -L/Library/Frameworks/ 
Intel_MKL.framework/Versions/9.0/lib/32 -lpthread
      P   Preprocessor flags
      R   R_LIB (LAPACK+BLAS):     -lmkl_lapack64 -lmkl_ia32 -lguide - 
lmkl -lvml -pthread

If there is another location/file in which options are stored, please  
let me know and I will look there.

I rather prefer to not reformat.

Thanks,
Ashley

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

On Jul 20, 2007, at 11:21 AM, Maurits Haverkort wrote:

> Dear Ashley
>
> I had the same problem (2) and (3) after installing Wien2K on a Mac
> Pro. The problem came from the optimization options given to the
> compiler. (Can you sent them) I've removed   -ftree-vectorize and
> then everything worked fine.
>
> Best regards,
> Maurits
>
>
>
> On Jul 20, 2007, at 10:14 AM, Ashley Harvey wrote:
>
>> Dear Wien users,
>>
>> I am having some complications with the installation/running of
>> Wien2K version 7.2 on my Intel MacBook Pro, with up-to-date Intel
>> compilers and MKL:
>>
>> (1) I ran the script siteconfig_lapw, kindly modified by our
>> department's computer manager to work with my system (file a bit too
>> big to attach).  There are no errors upon compilation, but the
>> following warning occurs, and I am wondering if it is a problem:
>>
>>   warning suggest use of -bind_at_load, as lazy binding may result in
>> errors or different symbols being used
>>
>>
>> (2) When I start up w2web, with the intention to work the TiC example
>> to make sure everything is OK, in StructGen, after I enter the basic
>> information (e.g. space group, parameters), click on Save and then
>> proceed to set automatically RMT and continue editing, no RMTs are
>> actually generated - the value remains blank.  So, I stopped here
>> with w2web.
>>
>>
>> (3) If I then go to the command line (X11 xterm on the Mac), and
>> start init_lapw with setrmt, I am able to enter a value for the %
>> reduction, but when it proceeds to nn, this message comes up:
>>
>> matl-hci-d185:~/TiC/ asharvey$ init_lapw
>> next is setrmt
>> Automatic determination of RMTs. Please specify the desired RMT
>> reduction
>> compared to almost touching spheres.
>> Typically, for a single calculation just hit enter, for force
>> minimization
>> use 1-5; for volume effects you may need even larger reductions.
>> Enter reduction in %
>> 2
>>
>> Unrecognized option: nn
>> use: X [:<display>] [option]
>>
>> [followed by a long list of options, and then:]
>>
>> atom  Z   RMT-max   RMT
>> 1  22.0  2.11   2.11
>> 2   6.0  1.87   1.87
>> Do you want to accept these radii; discard them; or rerun setRmt  
>> (a/d/
>> r):
>> a
>>>   nn  (08:43:55)   please specify nn-bondlength factor: (usually=2)
>> 2
>> DSTMAX:   20.0000000000000
>> NAMED ATOM: Ti2.15    Z changed to IATNR+999 to determine equivalency
>> NAMED ATOM: C1.91     Z changed to IATNR+999 to determine equivalency
>>
>>      ATOM  1  Ti2.15     ATOM  2  C1.91
>> RMT(  1)=2.11000 AND RMT(  2)=1.87000
>> SUMS TO 3.98000  LT.  NN-DIST= 4.08937
>>
>>      ATOM  2  C1.91      ATOM  1  Ti2.15
>> RMT(  2)=1.87000 AND RMT(  1)=2.11000
>> SUMS TO 3.98000  LT.  NN-DIST= 4.08937
>> NN ENDS
>> 0.005u 0.006s 0:03.16 0.0%      0+0k 1+1io 0pf+0w
>> -----> check in  TiC.outputnn  for overlapping spheres,
>>         coordination and nearest neighbor distances
>> -----> continue with sgroup or edit the TiC.struct file (c/e)
>> c
>>>   sgroup      (08:44:58) Error lat==CUBIC in det_pgrp()
>> Error lat==CUBIC in det_pgrp()
>> diff: TiC.outputsgroup: No such file or directory
>> diff: TiC.outputsgroup1: No such file or directory
>> Error lat==CUBIC in det_pgrp()
>> 0.000u 0.001s 0:00.00 0.0%      0+0k 0+0io 0pf+0w
>> error: command   /Applications/Wien2K/sgroup -wi TiC.struct -wo
>> TiC.struct_sgroup  -set-TOL=0.00001   failed
>>
>> And then the program stops.
>>
>>
>> (4) I then proceeded to make a new directory, containing just the
>> TiC_2.struct file as generated by StructGen, and through the command
>> line run instgen_lapw.  This seems to work fine.  Then, using
>> setrmt_lapw:
>>
>> matl-hci-d185:~/TiC/TiC_2 asharvey$ setrmt_lapw TiC_2 -r 0
>> Unrecognized option: nn
>> use: X [:<display>] [option]
>>
>> [long list of possible options]
>>
>> readline() on closed filehandle NN at /Applications/Wien2K/
>> setrmt_lapw line 25.
>> readline() on closed filehandle NN at /Applications/Wien2K/
>> setrmt_lapw line 36.
>> atom  Z   RMT-max   RMT
>> Use of uninitialized value in sprintf at /Applications/Wien2K/
>> setrmt_lapw line 115, <STRUC> line 7.
>> Use of uninitialized value in sprintf at /Applications/Wien2K/
>> setrmt_lapw line 115, <STRUC> line 13.
>>
>> And it all stops.
>>
>> (5) Finally, in the command line, typing x - h yields:
>>
>> matl-hci-d185:/Applications/Wien2K asharvey$ x - h
>> Unrecognized option: -
>> use: X [:<display>] [option]
>>
>> [followed by a large list of possible options]
>>
>> But, typing x_lapw -h works fine.
>>
>>
>> I cannot figure out what's going on.  Did something go wrong in the
>> installation?
>> Sorry if these are simple questions:  Your help is greatly
>> appreciated.
>>
>> Ashley
>>
>> _______________________________________________
>> 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