dolfin team mailing list archive

[Anders Logg <logg@xxxxxxxxx>]

On Tue, Mar 25, 2008 at 02:58:39PM +0100, Ola Skavhaug wrote:
> I'm using the latest slepc (2.3.3), a slightly older petsc (2.3.3-p8),
> bleeding edge dolfin, and I get the following compile error on my amd64 ubuntu
> hardy:
> 
> scons: Building targets ...
> g++ -o dolfin/la/SLEPcEigenvalueSolver.os -c -Wall -pipe -ansi -DDEBUG -g
> -Werror -DNDEBUG -O2 -DPACKAGE_VERSION=\"0.7.2\" -DHAS_PETSC=1 -DHAS_SLEPC=1
> -DHAS_SCOTCH=1 -DHAS_GTS=1 -fPIC -I. -I/usr/include -I/usr/include
> -I/usr/include/libxml2
> -I/home/skavhaug/extsoft/petsc-2.3.3-p8/bmake/linux-gnu-cxx-debug
> -I/home/skavhaug/extsoft/petsc-2.3.3-p8/include -I/usr/local/include
> -I/usr/local/lib -Idolfin -I/home/skavhaug/extsoft/slepc-2.3.3
> -I/home/skavhaug/extsoft/slepc-2.3.3/include
> -I/home/skavhaug/extsoft/petsc-2.3.3-p8/bmake/linux-gnu-cxx-debug
> -I/home/skavhaug/extsoft/petsc-2.3.3-p8/include -I/usr/local/include
> -I/usr/local/lib -Idolfin -I/usr/local/include -I/usr/include
> -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include
> dolfin/la/SLEPcEigenvalueSolver.cpp
> cc1plus: warnings being treated as errors
> dolfin/la/SLEPcEigenvalueSolver.cpp: In member function ‘void
> dolfin::SLEPcEigenvalueSolver::solve(const dolfin::PETScMatrix&, const
> dolfin::PETScMatrix*, dolfin::uint)’:
> dolfin/la/SLEPcEigenvalueSolver.cpp:130: warning: comparison with string
> literal results in unspecified behaviour
> scons: *** [dolfin/la/SLEPcEigenvalueSolver.os] Error 1
> scons: building terminated because of errors.
> 
> 
> The troublesome line is the following:
> 
> EPSType eps_type = getType(type);
> if(eps_type != "default")  // THIS LINE GOES WRONG
>     EPSSetType(eps, eps_type);

Does it help if you just replace this line with

  if(strcmp(eps_type, "default") != 0)

?

-- 
Anders


> As EPSType is a macro for const char*, I have no clue why this fails to
> compile. The error message is rather confusing as well, or at least not
> verbose enough for me. Does anybody know that causes this misbehaviour?
> 
> Thanks,
> Ola
> _______________________________________________
> DOLFIN-dev mailing list
> DOLFIN-dev@xxxxxxxxxx
> http://www.fenics.org/mailman/listinfo/dolfin-dev