[Wien] [EXTERNAL] Re: Wien2k with oneAPI on Mac OSX (Monterey)

Gavin Abo gabo13279 at gmail.com
Fri Aug 5 05:44:41 CEST 2022


The sed may need different arguments for macOS based on webpages at the 
following two links:

https://stackoverflow.com/questions/29081799/sed-1-invalid-command-code-f

https://stackoverflow.com/questions/19456518/error-when-using-sed-with-find-command-on-os-x-invalid-command-code

WIEN2k I installed in Linux using oneAPI 2021.6.0 as described at the 
following link:

https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg21943.html

It is not shown in the above link, but the TiC example from the WIEN2k 
userguide did run and the calculation did complete the scf cycles just 
fine.

The SIGSEGV error I think is likely not related to the sed error.  I 
don't have MacOS, such that that I'm unable help troubleshot the error.

There were two links that helped me in the past with SIGSEGV errors.  So 
they might help you as well, they are:

https://www.intel.com/content/www/us/en/developer/articles/troubleshooting/determining-root-cause-of-sigsegv-or-sigbus-errors.html

http://www.democritos.it/pipermail/pw_forum/2005-March/002272.html

Kind Regards,

Gavin
WIEN2k user

On 8/4/2022 8:54 AM, Zhu, Jianxin via Wien wrote:
>
> Dear Laurence,
>
> I would like to respond to your N.B. on the absolute path. –
>
> 1/
>
> After I fire up ./siteconfig_lapw, and choose “O”, and then “S” (for 
> save), it displays
>
>    ***********************************************************************
>
>    *                     Changing compiler options       *
>
>    ***********************************************************************
>
>   Changing Parallel Compiler to
>
> sed: 1: "/Users/jxzhu/wien2k.19. ...": invalid command code j
>
>   changing Makefile in SRC_lib/blas_lapw
>
>   changing Makefile in SRC_lib/lapack_lapw
>
>>
> Since I only want to install the non-MPI version of the code, I don’t 
> understand why there is a header “Changing Parallel Compiler to”.
>
> But maybe this block for parallel compiler is skipped, I am not so 
> concerned.
>
> 2/ oneAPI on Mac has a strange way to set up on tcsh shell that I am 
> using. It is advised that we do the following on an open terminal not 
> placed into the login shell script file
>
> bash-c'source /opt/intel/oneapi/setvars.sh intel64 ; exec csh'
>
> 3/
>
> The more concerning line is:
>
> sed: 1: "/Users/jxzhu/wien2k.19. ...": invalid command code j
>
> Could it be the root cause for the Segment Fault error in “x lapw0”?
>
> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>
> Image             PC                Routine            Line   Source
>
> lapw0             000000010484EE84  Unknown   Unknown  Unknown
>
> libsystem_platfor  00007FF80F4C8DFD  Unknown             Unknown  Unknown
>
> lapw0             00000001047FE935  _xcpot1_   1  xcpot1.f
>
> lapw0             00000001047BF484  _MAIN__   2308  lapw0.F
>
> lapw0             000000010472CC2E  Unknown   Unknown  Unknown
>
> 0.083u 0.110s 0:00.34 55.8% 0+0k 0+0io 949pf+0w
>
> I am now coming back to try to understand this “sed” error.
>
> 4/ I do set a WIENROOT in the PATH of the login shell script file.
>
> Best,
>
> Jianxin
>
> #############################
> Jian-Xin Zhu, Ph.D
> Theoretical Division, MS B262
> Los Alamos National Laboratory
> Los Alamos, New Mexico 87545
> Phone: (505) 667 2363 (T-4);
>
> (505) 667 6602 (CINT)
>
> Fax: (505) 665 4063
> Email: jxzhu.at.lanl.gov <http://jxzhu.at.lanl.gov>
> URL: http://cint.lanl.gov <http://cint.lanl.gov>
> ##############################
>
> *From: *Laurence Marks <laurence.marks at gmail.com>
> *Reply-To: *"laurence.marks at gmail.com" <laurence.marks at gmail.com>
> *Date: *Sunday, July 10, 2022 at 12:35 PM
> *To: *A Mailing list for WIEN2k users <wien at zeus.theochem.tuwien.ac.at>
> *Cc: *Jianxin Zhu <jxzhu at lanl.gov>
> *Subject: *[EXTERNAL] Re: [Wien] Wien2k with oneAPI on Mac OSX (Monterey)
>
> I only compiled on Linux, and I suspect Peter did as well. Maybe 
> others have in OSX?
>
> You should probably check the compiler options for anything specific 
> to OSX, and . The errors entering xcpot1 indicates something major 
> wrong with the compilation.
>
> N.B., it may not matter but I do not understand why you have a full 
> path in the parallel compiler. More normal is to setup your PATH 
> variable properly, then just use the short name. Did you source the 
> Intel script to set the environment (needed)?
>
> --
> Professor Laurence Marks
> Department of Materials Science and Engineering, Northwestern University
> www.numis.northwestern.edu 
> <https://urldefense.com/v3/__http:/www.numis.northwestern.edu__;!!Bt8fGhp8LhKGRg!B8CKTbz6Hi7qXQViiz4xp_4KbpuPzRZjbBq18gwZvinvPYVLLy4CNaqjzOGNoPX6aZqwfA7B7oV33VcmmU9l6Fa7$>
> "Research is to see what everybody else has seen, and to think what 
> nobody else has thought" Albert Szent-Györgyi
>
> On Sun, Jul 10, 2022, 1:01 PM Zhu, Jianxin via Wien 
> <wien at zeus.theochem.tuwien.ac.at> wrote:
>
>     Dear All,
>
>     Recently I upgraded my Mac OSX system to Monterey.
>     I also downloaded and installed  Intel oneAPI 2022.1.0 on the system.
>
>     1/ The compilation of  wien2k.21.1 (and wien2k.19.1) seems to be
>     successful.
>     [I only want to install the k-point coarse-grained wien2k on the
>     system.
>     After I enter "O" (Compiling Options) and then "S" (Save and
>     Quit), the following
>     comes up
>
>      ***********************************************************************
>        *                      Changing compiler options              
>         *
>      ***********************************************************************
>
>         Changing Parallel Compiler to
>
>     sed: 1: "/Users/jxzhu/wien2k.19. ...": invalid command code j
>
>     ...
>
>     I suspect this is due to the missing of $MPR variable value and
>     hope it has no impact to the non-MPI installation.]
>
>     2/ The severe problem comes up during the run-time stage. For
>     example, for TiC,
>
>     ..../TiC] jxzhu% x lapw0
>     forrtl: severe (174): SIGSEGV, segmentation fault occurred
>     Image              PC                Routine   Line        Source
>     lapw0              000000010D2EAE84  Unknown      Unknown  Unknown
>     libsystem_platfor  00007FF801D93DFD  Unknown      Unknown  Unknown
>     lapw0              000000010D29A935  _xcpot1_           1  xcpot1.f
>     lapw0              000000010D25B484  _MAIN__         2308  lapw0.F
>     lapw0              000000010D1C8C2E  Unknown      Unknown  Unknown
>     0.086u 0.037s 0:00.16 68.7%     0+0k 0+0io 186pf+0w
>     error: command   /Users/jxzhu/wien2k.21.1/lapw0 lapw0.def   failed
>
>     This problem occurs regardless of wien2k being 21.1 or 19.1.
>     [Since I am using tcsh shell, I have execute the following to run jobs
>     bash -c 'source /opt/intel/oneapi/setvars.sh
>     <https://urldefense.com/v3/__http:/setvars.sh__;!!Bt8fGhp8LhKGRg!B8CKTbz6Hi7qXQViiz4xp_4KbpuPzRZjbBq18gwZvinvPYVLLy4CNaqjzOGNoPX6aZqwfA7B7oV33VcmmRHWC5x8$>
>     intel64 ; exec csh'
>     ]
>
>
>     I read the group mailing list and see the thread by Peter and
>     Laurence in early 2021 indicating the code should run fine with
>     oneAPI.
>     But I am not sure if it is for Linux or Mac system.
>
>     I wonder if any community member has encountered the same problem.
>     Any thoughts and suggestions are appreciated.
>
>     Best,
>
>     Jianxin
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20220804/8c062e7f/attachment.htm>


More information about the Wien mailing list