[Insight-developers] Re: Controlling number of lines with CabeSwig (was Re: gcc-3.3 : final_scan_insn, at final.c:2117)

[Mathieu Malaterre]

Ok, actually this is a two lines summary:

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14711#c26
"Fixed in 4.0; will not be fixed in 3.4.x."

Basically this bug affects:
3.0 3.0.1 3.0.2 3.0.3 3.0.4 3.1 3.1.1 3.1.2 3.2 3.2.1 3.2.2 3.2.3 3.3
3.3.1 3.3.2 3.3.3 3.4.0

Known to work:
gcc 2.95.3 and gcc 4.x

-Mathieu

On 9/12/07, Mathieu Malaterre <mathieu.malaterre at gmail.com> wrote:
> Hi Luis,
>
>   The one line summary is:
>
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14711#c17
>
> "This is a long standing regresssion (since 3.0).  It is fixed in
> 3.4.0 and 3.5.0.  It will not be fixed in 3.3.x."
>
> Without that patch you get an ICE on any gcc 3.x compiler (too many
> lines). gcc 2.95 and gcc 4.x are not concerned with this issue.
>
> HTH
> -Mathieu
>
> On 9/12/07, Luis Ibanez <luis.ibanez at kitware.com> wrote:
> >
> >
> > Gaetan,
> >
> >
> > Thanks for this new patch.
> > The patch has been committed.
> >
> >
> > However, I'm afraid I missed the discussion that led to this patch.
> >
> > Just for the record, could you please give us some background ?
> >
> >
> >    Thanks
> >
> >
> >
> >        Luis
> >
> >
> >
> > -----------------------
> > Gaëtan Lehmann wrote:
> > >
> > > Le 12 sept. 07 à 11:58, Mathieu Malaterre a écrit :
> > >
> > >> Hi Luis,
> > >>
> > >>   I believe the following patch is completely backward compatible and
> > >> should be completely invisible to most user (you even save some disk
> > >> space).
> > >
> > >
> > > I do agree. This flag has been tried before, and disabled because
> > > reading the produced code with it is nearly impossible - not really
> > > handy while debugging typemaps.
> > > I believe it can be used now that the typemaps have proved to be  usable
> > > for a long enough time.
> > >
> > > Gaëtan
> > >
> > >>
> > >>   Could you please apply that before the next ITK release.
> > >>
> > >> Thanks to gaetan for the patch,
> > >> -Mathieu
> > >>
> > >> On 9/12/07, Gaëtan Lehmann <gaetan.lehmann at jouy.inra.fr> wrote:
> > >>
> > >>>
> > >>> Hi Mathieu,
> > >>>
> > >>> Le 12 sept. 07 à 11:47, Mathieu Malaterre a écrit :
> > >>>
> > >>>> The trick is to simply add -fcompact. The time to compile the  file is
> > >>>> exactly the same but at least I am now pleasing gcc 3.3, and  manage to
> > >>>> compile the object file.
> > >>>>
> > >>>
> > >>> That's a good news :-)
> > >>>
> > >>>> Now I need to figure out how to add this -fcompact to all target,
> > >>>> geatan any suggestion ?
> > >>>
> > >>>
> > >>> yes, you can add it in CreateWrapperLibrary.cmake, line 416
> > >>>
> > >>>    ADD_CUSTOM_COMMAND(
> > >>>     SOURCE ${input_idx}
> > >>>     COMMAND ${CSWIG}
> > >>>     ARGS ${swig_libs}
> > >>>          -I${CSWIG_DEFAULT_LIB}
> > >>>          ${CSWIG_IGNORE_WARNINGS}
> > >>>          -noruntime
> > >>>         -fcompact <------------ HERE !
> > >>>          ${cindex}
> > >>>          -depend ${input_xml}.depend
> > >>>          -outdir ${outdir}
> > >>>          -o ${output_cxx}
> > >>>          -c++
> > >>>          ${CSWIG_ARGS_${language}}
> > >>>          ${extra_args}
> > >>>          ${input_xml}
> > >>>     TARGET ${library_name}
> > >>>
> > >>> Gaëtan
> > >>>
> > >>>
> > >>>
> > >>> --
> > >>> Gaëtan Lehmann
> > >>> Biologie du Développement et de la Reproduction
> > >>> INRA de Jouy-en-Josas (France)
> > >>> tel: +33 1 34 65 29 66    fax: 01 34 65 29 09
> > >>> http://voxel.jouy.inra.fr
> > >>>
> > >>>
> > >>>
> > >>>
> > >>>
> > >>
> > >>
> > >> --
> > >> Mathieu
> > >
> > >
> > > --
> > > Gaëtan Lehmann
> > > Biologie du Développement et de la Reproduction
> > > INRA de Jouy-en-Josas (France)
> > > tel: +33 1 34 65 29 66    fax: 01 34 65 29 09
> > > http://voxel.jouy.inra.fr
> > >
> > >
> > >
> >
>
>
> --
> Mathieu
>


-- 
Mathieu