[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Release

On 9/22/07, Matthew Knepley <knepley@xxxxxxxxx> wrote:
> What do you think about 2.4 in October? I really want to get the new
> install out there.'

I'm experimenting at home about a way to compile petsc with
'-fstrict-aliasing', in order to follow C99 strict aliasing rules.
However, If I am right, this requiered a new layout for PetscObect's
structure (I basically followed the approach being used for Python
3.0). As a side effect, some very commonly used struct members (like
'comm', 'type_name', and 'prefix') have to be accessed by casting
derived objecs as PetscObject, i.e.:   Vec v;  ((PetscObject)v)->comm.

I have also some problems when calling PetscObjectQueryFunction, as
the compiler complains about strict aliasing rules, and I need to
define an additional function pointer, get the function, and next cast
to the a actual func. pointer to do the call. Not suure yet if this
can be improved, perhaps a macro can help.

Is there any interest in this be completed? It's a nonsense? In case
you want me to continue this work, I will need a bit of help in a few
places related to low-level calls related to sockets and sys calls.

-- 
Lisandro Dalcín
---------------
Centro Internacional de Métodos Computacionales en Ingeniería (CIMEC)
Instituto de Desarrollo Tecnológico para la Industria Química (INTEC)
Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)
PTLC - Güemes 3450, (3000) Santa Fe, Argentina
Tel/Fax: +54-(0)342-451.1594