Like I said. COMP_LIBS should only contain specific library names. No paths.
Also, you still have redundancies... ;) Delete this line: COMP_LIBS = -L/prghome/Siesta1/Obj Lastly, please always post the exact error everytime you post something new. It makes it much easier for us to see (regardless of whether you see the "same" error). Den ons. 16. dec. 2020 kl. 22.04 skrev Tamas Karpati <tkarp...@gmail.com>: > Hi, > > Sorry for bothering you with such a messy file last time. > I noticed and removed the redundancies (and still have the same result). > Also made it easier to read, I hope. Can you please take a look at it? > I'm afraid that there is an issue with the ordering of flags for each pkg > (MPI, ELPA etc) or something is missing. > > Best regards, > toma > > On Sun, Dec 13, 2020 at 12:59 PM Tamas Karpati <tkarp...@gmail.com> wrote: > > > > Dear Prof. Papior, > > > > Thank you for your quick response. Please find arch.make attached. > > > > For your information: this file was auto-generated by my home-made > > pkg. manager which added all flags for all the dependencies. The code > > is based on my understanding of Obj/DOCUMENTED-TEMPLATE.make. > > > > I appreciate your aid very much, thank you. > > > > With regards, > > toma > > > > On Sat, Dec 12, 2020 at 10:01 PM Nick Papior <nickpap...@gmail.com> > wrote: > > > > > > Please attach your arch.make file > > > > > > On Fri, 11 Dec 2020, 22:03 Tamas Karpati, <tkarp...@gmail.com> wrote: > > >> > > >> Dear fellow developers, > > >> > > >> > > >> Please help my first attempt to build SIESTA as I'm stuck after > > >> several attempts. No similar symptoms I could find in the archives. > > >> > > >> Shortly: make stops without both executables and error messages. > > >> > > >> > > >> Details: after having > > >> * OpenMPI-4.0.5, OpenBLAS-0.3.10, ScaLAPACK-2.1.0, > > >> Elpa-2020.005.001, Metis-5.1.0 and MUMPS-5.3.5 built, > > >> * downloaded from GitLab the Dec. 11, 2020 master version of SIESTA, > > >> * applied the install scripts in Doc for both NetCDF and FLOOK, > > >> * created the arch.make and finally invoked > > >> * make (either without arguments or adding the "-j N" option). > > >> > > >> Without a better guess I followed siesta-4.1-b4.pdf/Section 2. > > >> > > >> > > >> Two observations: > > >> > > >> 1, there neither executables (ie. siesta) > > >> nor error messages appear > > >> > > >> 2, when I run make several times: > > >> - first time libwxml.a and libmpi_f90.a appear, > > >> - second time libfdf.a and MatrixSwitch.a appear, > > >> - third time libSiestaXC.a materialized but > > >> - starting from the fourth invokation of make I got > > >> > > >> "No rule to make target '-L/home/atom/Siesta/Obj', > > >> needed by 'posix_calls.o'. Stop." > > >> > > >> (iteration converged, this method gives me no more fruit) > > >> > > >> Looking in Obj/Makefile did not help either. Still shut a few with > > >> "make lib", "make siesta" even "make dep" but to no avail. > > >> Apparently I need to learn. > > >> > > >> > > >> I kindly ask your help (instructions or written method on the Web). > > >> Thank you in advance. > > >> > > >> Best regards, > > >> toma > > >> > > >> -- > > >> SIESTA is supported by the Spanish Research Agency (AEI) and by the > European H2020 MaX Centre of Excellence (http://www.max-centre.eu/) > > > > > > > > > -- > > > SIESTA is supported by the Spanish Research Agency (AEI) and by the > European H2020 MaX Centre of Excellence (http://www.max-centre.eu/) > > -- > SIESTA is supported by the Spanish Research Agency (AEI) and by the > European H2020 MaX Centre of Excellence (http://www.max-centre.eu/) > -- Kind regards Nick
-- SIESTA is supported by the Spanish Research Agency (AEI) and by the European H2020 MaX Centre of Excellence (http://www.max-centre.eu/)
