Barry Smith <[email protected]> writes:

>   It seems some compilers don’t like applying attributes to only some 
> variables being declared. Is this legitimate? Should we always be marking the 
> unused variables on lines of their own? (I have no problem doing that).
>
> /sandbox/petsc/petsc.clone-2/arch-linux-complex-gcov/bin/mpicxx -o ex10.o -c 
> -Wall -Wwrite-strings -Wno-strict-aliasing -Wno-unknown-pragmas 
> -fprofile-arcs -ftest-coverage -g -O0  -fPIC    
> -I/sandbox/petsc/petsc.clone-2/include 
> -I/sandbox/petsc/petsc.clone-2/arch-linux-complex-gcov/include    `pwd`/ex10.c
> /sandbox/petsc/petsc.clone-2/src/ts/examples/tutorials/ex10.c: In function 
> ‘PetscErrorCode RDIJacobian_FE(TS, PetscReal, Vec, Vec, PetscReal, _p_Mat**, 
> _p_Mat**, MatStructure*, void*)’:
> /sandbox/petsc/petsc.clone-2/src/ts/examples/tutorials/ex10.c:710:36: 
> warning: ignoring attributes applied to ‘std::complex<double>’ after 
> definition [-Wattributes]
>        PetscScalar D_R,PETSC_UNUSED rad;
>
> /sandbox/petsc/petsc.clone-2/src/ts/examples/tutorials
>
> request-assigned: petsc-dev PetscScalar D_R,PETSC_UNUSED rad; warning: 
> ignoring attributes applied to ‘std::complex<double>’ after definition

Okay, but what compiler is being this unhelpful?

Attachment: pgpZCBjBQglwj.pgp
Description: PGP signature

Reply via email to