[Wien] ifort classic compiler now discontinued in one-api 2025.0 online repositories

Peter Blaha peter.blaha at tuwien.ac.at
Wed Oct 30 09:13:06 CET 2024


I'm not sure about all details, since there were a couple of different 
contradicting messages. I did not have time yet to install the latest 
ifx myself.

However, I can imagine that the latest reported errors are (in part) 
"real" problems.  I'm referring to the problems listed below:
-----------------------------------------------
SRC_dstart/atom_read.F:
This is a real program bug - although it should never cause a problem.

rholm is allocated and used only within the "#ifdef parallel" section, 
except that there is a read statement with error=888.
Here it would jump out of the parallel section and write rholm to some 
file and stop. rholm is NOT defined at this point.

The fix is to move the

#endif

statement a few lines down after the "stop".
-------------------------------------------------
SRC_3ddens:
I guess this is a linker problem and comes because the fftw software was 
not compiled with omp support. The 2 functions are called only when 
FFTW_OMP is set, and if the linker cannot find them, they must be 
missing in the fftw-library, because the fftw was not compiled with the 
corresponding switches.
...
#ifdef FFTW_OMP
      call dfftw_init_threads(omp_error)
      call dfftw_plan_with_nthreads(omp_get_max_threads())
#endif
....
------------------------------------------------
SRC_lapw5/SearchZ.F:  again, a "real programming error":

...
                 call RTBIS(RHOSTM,X1,X2,XACC,CTarget,ISTM,RT)

RHOSTM  is a function. A function name may be passed into a subroutine 
only if defined as "external".

Insert at the beginning of the subroutine a line:

...
         dimension VT(3), X1(3),X2(3),Rt(3),AA(3)
         external rhostm                            ! insert
....
-----------------------------------------------

Best regards
Peter

Am 28.10.2024 um 08:52 schrieb Michael Fechtelkord via Wien:
> These are the detailed error messages from the compile.msg files (I will 
> report the SearchZ.F internal compiler error to Intel):
> 
> 
> 
> SRC_dstart
> /usr/lib64/gcc/x86_64-suse-linux/14/../../../../x86_64-suse-linux/bin/ 
> ld: atom_read.o: in function `rholm_.t94p.t95p':
> ifxYWjPzh.i:(.text+0x27): undefined reference to `rholm_'
> make[1]: *** [Makefile:99: dstart] Fehler 1
> make[1]: Verzeichnis „/usr/local/WIEN2k/SRC_dstart“ wird verlassen
> make: *** [Makefile:90: seq] Fehler 2
> make: *** Keine Regel, um „complex“ zu erstellen.  Schluss.
> 
> SRC_3ddens
> /usr/lib64/gcc/x86_64-suse-linux/14/../../../../x86_64-suse-linux/bin/ 
> ld: 3ddens.o: in function `dfftw_init_threads_.t1225p':
> ifxEONrLm.i:(.text+0xf797): undefined reference to `dfftw_init_threads_'
> /usr/lib64/gcc/x86_64-suse-linux/14/../../../../x86_64-suse-linux/bin/ 
> ld: 3ddens.o: in function `dfftw_plan_with_nthreads_.t1230p':
> ifxEONrLm.i:(.text+0xf7b7): undefined reference to 
> `dfftw_plan_with_nthreads_'
> make: *** [Makefile:65: 3ddens] Fehler 1
> make: *** Keine Regel, um „complex“ zu erstellen.  Schluss.
> 
> SRC_lapw5
> SearchZ.F: REAL version extracted
> ifx  -O0 -xAVX2 -FR -mp1 -w -prec_div -pc80 -ip -DINTEL_VML -traceback - 
> assume buffered_io -I/opt/intel/oneapi/mkl/2025.0/include -c SearchZ_tmp_.F
>            #0 0x000000000310cce1
>            #1 0x00000000031715d7
>    ..
>           #21 0x0000150ba5a40eec
>           #22 0x0000150ba5a40fb5 __libc_start_main + 135
>           #23 0x0000000002e8a34e
> 
> SearchZ_tmp_.F(114): error #5623: **Internal compiler error: internal 
> abort** Please report this error along with the circumstances in which 
> it occurred in a Software Prob
> lem Report.  Note: File and line given may not be explicit cause of this 
> error.
>                  call RTBIS(RHOSTM,X1,X2,XACC,CTarget,ISTM,RT)
> ---------------------------^
> compilation aborted for SearchZ_tmp_.F (code 3)
> make[1]: *** [Makefile:122: SearchZ.o] Fehler 3
> make[1]: Verzeichnis „/usr/local/WIEN2k/SRC_lapw5“ wird verlassen
> make: *** [Makefile:74: real] Fehler 2
> 
> SearchZ.F: COMPLEX version extracted
> ifx  -O0 -xAVX2 -FR -mp1 -w -prec_div -pc80 -ip -DINTEL_VML -traceback - 
> assume buffered_io -I/opt/intel/oneapi/mkl/2025.0/include -c SearchZ_tmp_.F
>            #0 0x000000000310cce1
>            #1 0x00000000031715d7
>           ..
>           #21 0x000015513de40eec
>           #22 0x000015513de40fb5 __libc_start_main + 135
>           #23 0x0000000002e8a34e
> 
> SearchZ_tmp_.F(114): error #5623: **Internal compiler error: internal 
> abort** Please report this error along with the circumstances in which 
> it occurred in a Software Prob
> lem Report.  Note: File and line given may not be explicit cause of this 
> error.
>                  call RTBIS(RHOSTM,X1,X2,XACC,CTarget,ISTM,RT)
> ---------------------------^
> compilation aborted for SearchZ_tmp_.F (code 3)
> make[1]: *** [Makefile:122: SearchZ.o] Fehler 3
> make[1]: Verzeichnis „/usr/local/WIEN2k/SRC_lapw5“ wird verlassen
> make: *** [Makefile:77: complex] Fehler 2
> 
> Best regards,
> 
> Michael
> 

-- 
-----------------------------------------------------------------------
Peter Blaha,  Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-158801165300
Email: peter.blaha at tuwien.ac.at
WWW:   http://www.imc.tuwien.ac.at      WIEN2k: http://www.wien2k.at
-------------------------------------------------------------------------



More information about the Wien mailing list