[Facets-devel] PETSC ERROR

[Barry Smith]

    I agree with Matt here. If we decomposed in then the initial  
initialization would setup certain things and the next initialize  
(depending on options) would have to decompose the exact same things  
before setting another way.

    Lisandro,

       I do not understand from below what problem you have with them  
being together and what advantage you get from them being apart.

    Barry

On Jun 4, 2009, at 5:51 PM, Matthew Knepley wrote:

> I think there is too much setup that can be customized from the  
> command line. I
> don't think you can decompose like that.
>
>   Matt
>
> On Thu, Jun 4, 2009 at 5:33 PM, Lisandro Dalcin <dalcinl at gmail.com>  
> wrote:
> On Thu, Jun 4, 2009 at 6:17 PM, Satish Balay <balay at mcs.anl.gov>  
> wrote:
> > This doesn't work - as some of the options are required during
> > PetscIntialize().
> >
>
> This behaviour had always bothered my, but never invested time to try
> to fix that. Currently, I'm doing a lot of hackery in  petsc4py for
> being able to use command line.
>
> In the bootstrapping script for a simulation, I have to do first:
>
> >>> import sys, petsc4py
> >>> petsc4py.init(sys.argv)
>
> and then next I'm able to do
>
> >>> from petsc4py import PETSc
>
> and being able to make PETSc use my args.
>
> In short, It would be realy nice to split PetscInitialize() in two:
> the fist part initializes everything without looking at command line
> stuff, and then a second part command line options initialization and
> some more initialization stuff. Something like
>
> PetscInitalize(argc,argv, file) {
>    PetscInitializeBasic(argc,argv)
>    PetscSetUp(argc,argv, file)
> }
>
> The argc,argv to PetscInitializeBasic() just in case PETSc still
> supports (does it?) MPI-1 implementations requiring argc,argv.
>
>
>
>
> --
> 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
>
>
>
> -- 
> What most experimenters take for granted before they begin their  
> experiments is infinitely more interesting than any results to which  
> their experiments lead.
> -- Norbert Wiener