[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: package "optim" compilation oddity
From: |
CdeMills |
Subject: |
Re: package "optim" compilation oddity |
Date: |
Fri, 31 Aug 2012 09:43:08 -0700 (PDT) |
John W. Eaton wrote
>
>
> You should be able to get this list of options and libraries by using
> "mkoctfile -p FLIBS" similar to the way you get the value of LFLAGS.
>
> Or are you saying that mkoctfile should automatically add $FLIBS to
> the link command? I don't know how it would know to do that if the
> command used to link has no reference to any Fortran source files.
>
> jwe
>
No, I'm just saying that
1) either I made a mistake somewhere, and linking non-fortran code with blas
and lapack does not, as it should, include the fortran libs. In this case I
would like to know what went wrong.
2) either the makefile of the optim package should explicitelly include the
fortran libs, through mkoctfile FLIBS value, when linking
__disna_optim__.oct, on Mac OS X systems at least.
In my opinion, this issue is very specific. As you point out, introducing
heuristic in mkoctfile to detect wether or not linking with the fortran libs
is required, is not easy.
Regards
Pascal
Regards
Pascal
--
View this message in context:
http://octave.1599824.n4.nabble.com/package-optim-compilation-oddity-tp4643564p4643569.html
Sent from the Octave - Maintainers mailing list archive at Nabble.com.