dolfin team mailing list archive

[Anders Logg <logg@xxxxxxxxx>]

ok, so then pcimpl.h can be removed from DOLFIN?

I will create a new release within the next hour (will do FFC first,
FIAT is ready), so make the change now if you want to include it in
the release.

/Anders

On Tue, Jul 05, 2005 at 05:41:12PM +0200, Johan Jansson wrote:
> Hi,
> 
> Below is some discussion about PETSc standard installation and
> packages. It seems that a petsc-dev package should include the PETSc
> source, and thus that DOLFIN can assume the PETSc source is available
> when building. This would solve both the pcimpl.h issue as well as the
> issue of extracting PETSc compiler flags.
> 
>   Johan
> 
> ----- Forwarded message from Matthew Knepley <knepley@xxxxxxxxxxx> -----
> 
> X-Original-To: johanjan@xxxxxxxxxxxxxxxx
> X-Medic-Info: 2d27.42caa824.0 wCApK6Z7Wvkurayi 
> To: Johan Jansson <johanjan@xxxxxxxxxxxxxxxx>
> Cc: petsc-maint@xxxxxxxxxxx
> Reply-To: petsc-maint@xxxxxxxxxxx
> Subject: Re: [PETSC #13286] Make pcimpl.h a public API
> From: Matthew Knepley <knepley@xxxxxxxxxxx>
> Date: Tue, 05 Jul 2005 10:32:50 -0500
> X-Spam-Checker-Version: SpamAssassin 2.64 (2004-01-11) on anubis.medic.chalmers.se
> X-Virus-Scanned: by amavisd-new at chalmers.se
> X-Spam-Status: No, hits=0 tagged_above=-100 required=5 tests=
> X-Spam-Level: 
> 
> Johan Jansson <johanjan@xxxxxxxxxxxxxxxx> writes:
> 
> > On Tue, Jul 05, 2005 at 07:20:35AM -0500, Matthew Knepley wrote:
> >> 
> >> Not sure why you want this. The organization of PC mirrors that of
> >> every other package in PETSc. The public headers in /include define
> >> the interface that users need to use the libraries and nothing
> >> else. The implementation headers define the underlying structure of
> >> these objects. This is a very common component programming
> >> structure. The PETSc build system will allow this include using
> >> <src/ksp/pc/pcimpl.h>. To me this makes sense since you are doing
> >> something which depends on PETSc internals rather than just the
> >> interface, but I am willing to listen to the other side.
> >
> > Including <src/ksp/pc/pcimpl.h> is a good solution, but is that
> > accessible from a standard PETSc installation? Can DOLFIN assume that
> > the /src structure will be present?
> 
>   I assume this is a Debian-esque package question. I would liken building
> a package implementation to operations which require a -devel package. So
> I would say that requiring the source is fine. We always install this way
> since debugging is impossible without the source, however I do not use
> any packing system for PETSc currently, so I do not know what they do.
> 
>     Thanks,
> 
>         Matt

-- 
Anders Logg
Research Assistant Professor
Toyota Technological Institute at Chicago
http://www.tti-c.org/logg/