Support and discussions for Molcas and OpenMolcas users and developers
You are not logged in.
Please note: The forum's URL has changed. The new URL is: https://molcasforum.univie.ac.at. Please update your bookmarks!
You can choose an avatar and change the default style by going to "Profile" → "Personality" or "Display".Pages: 1
Hi everybody,
I'm trying to compile OpenMolcas on a Linux machine. Cmake is done successfully without any error and it could find Fortran and C compilers. But, when I tried to run "make" it stopped in rassi part with this error:
.
.
.
[ 95%] Built target rassi
make: *** [all] Error 2
Does anyone know what is wrong here and how to fix it?
Many Thanks
Offline
That's not of any help. If you did use the -j flag for make, the actual error may be somewhere above. Run "make" again, without any flag, or "make VERBOSE=1"
Offline
Thanks Ignacio. I ran "make" again witjout the -j flage and with VERBOSE=1. I got this error massage:
-----------------------------------------------------------------------------------------
/lustre/hpc/software/kemi/OpenMolcas/v18.09/OpenMolcas/src/single_aniso/torque.f:39.38:
Real(kind=wp), allocatable :: ZT ! ZT(nTempMagn) ! total statis
1
Error: Scalar object 'zt' at (1) may not be ALLOCATABLE
make[2]: *** [CMakeFiles/single_aniso.dir/src/single_aniso/torque.f.o] Error 1
make[2]: Leaving directory `/lustre/hpc/software/kemi/OpenMolcas/v18.09/build'
make[1]: *** [CMakeFiles/single_aniso.dir/all] Error 2
make[1]: Leaving directory `/lustre/hpc/software/kemi/OpenMolcas/v18.09/build'
make: *** [all] Error 2
-----------------------------------------------------------------------------------------
Offline
Dear Ignacio,
The problem is fixed by changing Fortran compiler from GNU to Intel and running "make" without the -j flag. Thanks.
Offline
Thanks, that looks like a bug.
Offline
moabe
yes, it is a known (and fixed) bug, appearing with newest gfortran.
Offline
One thing is the compiler accepting or not such a construct. It should accept it, I believe, because it's valid in the latest Fortran specification. But the bug I refer to is that the ZT variable is never allocated and it looks like it should actually be declared as a plain static scalar.
Offline
Pages: 1