[Pw_forum] espresso compilation error IBM SP4 arch
aaron at nemo.physics.ncsu.edu
aaron at nemo.physics.ncsu.edu
Sat Oct 2 18:51:46 CEST 2004
Thanks very much.
On Sat, 2 Oct 2004, Axel Kohlmeyer wrote:
>
> On Sat, 2 Oct 2004 aaron at nemo.physics.ncsu.edu wrote:
>
> A> Hi All, Carlo,
> A>
> A> Thanks very much for the help. I was able to move past that point with
> A> your file parser.f90. Now I'm stuck again, but I really want the
> A> functionality I read about.
> A>
> A> Here's the error:
> A>
> A> > > ***********************************************************
> A> > > ***********************************************************
> A>
> A> cc -Wp,-D__AIX,-D__PARA,-D__MPI,-DHAS_ZHEGVX -O2 -c c_mkdir.c
> A> "c_mkdir.c", line 17.10: 1506-296 (S) #include file "c_defs.h" not found.
> A> make: 1254-004 The error code from the last command is 1.
> A>
> A>
> A> Stop.
> A> make: 1254-004 The error code from the last command is 2.
> A>
> A>
> A> Stop.
> A>
> A> > > ***********************************************************
> A> > > ***********************************************************
> A>
> A> Usually I would expect the c-compiler to find these ~.h files on its own.
>
> aaron,
>
> this is only true for (some of) the system headers. c_defs.h is part of
> the espresso package, and thus you have to make sure, that a) the IFLAGS
> definition contains -I../include and b) definition of CCFLAGS contains
> $(IFLAGS) so that the preprocessor knows how to find that file.
>
> regards,
> axel.
>
> A>
> A> I can really make use of the functionality to specify intermediate images
> A> in the NEB.
> A>
> A> Thanks
> A>
> A> Aaron
> A>
> A> On
> A> Fri, 1 Oct 2004, carlo sbraccia wrote:
> A>
> A> > Dear Aaron,
> A> > with some version of the AIX compiler, some combination of flags lead to
> A> > variables being defined as static, hence giving a conflict with PURE
> A> > function. To solve the problem (or compiler bug) we must force the
> A> > variable to be AUTOMATIC. Try with the attached routine.
> A> > carlo
> A> >
> A> > On Fri, 2004-10-01 at 00:02, aaron at nemo.physics.ncsu.edu wrote:
> A> > > Hi,
> A> > >
> A> > > I'm very excited to try the new release.
> A> > > However I've run into a small compilation error as follows:
> A> > >
> A> > > ***********************************************************
> A> > > mpxlf -qalias=noaryovrlp -I../include -O3 -qstrict -qarch=auto
> A> > > -qtune=auto -qsuffix=cpp=f90 -qdpc -Q -qalias=nointptr -qfree=f90
> A> > > -I/tmp/work/amgeorge/pw2.1/install
> A> > > -WF,-D__AIX,-D__PARA,-D__MPI,-DHAS_ZHEGVX
> A> > > -I/tmp/work/amgeorge/pw2.1/Modules -I/tmp/work/amgeorge/pw2.1/PW
> A> > > -I/tmp/work/amgeorge/pw2.1/PH -c parser.f90
> A> > > "parser.f90", 1514-286 (S) The variable, j, has the SAVE or STATIC
> A> > > attribute. A variable declared in a pure subprogram must not have the
> A> > > SAVE or STATIC attribute. You may have specified the SAVE option
> A> > > explicitly or by default.
> A> > > "parser.f90", 1514-286 (S) The variable, sep2, has the SAVE or STATIC
> A> > > attribute. A variable declared in a pure subprogram must not have the
> A> > > SAVE or STATIC attribute. You may have specified the SAVE option
> A> > > explicitly or by default.
> A> > > "parser.f90", 1514-286 (S) The variable, sep1, has the SAVE or STATIC
> A> > > attribute. A variable declared in a pure subprogram must not have the
> A> > > SAVE or STATIC attribute. You may have specified the SAVE option
> A> > > explicitly or by default.
> A> > > ** parser === End of Compilation 1 ===
> A> > > ***********************************************************
> A> > > ***********************************************************
> A> > >
> A> > > Please advise.
> A> > >
> A> > > Thanks in advance,
> A> > >
> A> > > Aaron
> A> > >
> A> > >
> A> > >
> A> > >
> A> > > _______________________________________________
> A> > > Pw_forum mailing list
> A> > > Pw_forum at pwscf.org
> A> > > http://www.democritos.it/mailman/listinfo/pw_forum
> A> >
> A>
> A> _______________________________________________
> A> Pw_forum mailing list
> A> Pw_forum at pwscf.org
> A> http://www.democritos.it/mailman/listinfo/pw_forum
> A>
> A>
>
>
More information about the users
mailing list