Re: [Patch, fortran-dev] PR 43896: [fortran-dev Regression] ICE in gfc_conv_variable, at fortran/trans-expr.c:551

[Paul Richard Thomas <paul dot richard dot thomas at gmail dot com>]

Janus,

That's OK for fortran-dev and so implicitly for the trunk-to-be.

Thanks for the rapid fix.

Cheers

Paul

On Mon, Apr 26, 2010 at 11:55 PM, Janus Weil <janus@gcc.gnu.org> wrote:
> Hi all,
>
> just a few hours after we had fixed the last fortran-dev regression
> today, a new bug report was rolling in which contained a new one :(
>
> Fortunately it is easily fixed: I just had to remove the NULL_EXPR
> initializers of the vtabs' PPC components, which anyway are not
> required, since all the vtab-PPCs are initialized explicitly later on.
>
> The attached patch regtests cleanly on x86_64-unknown-linux-gnu and
> fixes the fortran-dev regression part of this PR, but not the
> remaining ICE which also appears on trunk (this can wait until after
> the merge). Ok for fortran-dev?
>
> Btw, I want to thank the gfortran/GCC community for accepting me again
> as a GSoC student this year, and Tobias for being my mentor again. I'm
> looking forward to making some serious progress on OOP this summer :)
>
> Cheers,
> Janus
>
>
>
> 2010-04-26 ?Janus Weil ?<janus@gcc.gnu.org>
>
> ? ? ? ?PR fortran/43896
> ? ? ? ?* symbol.c (add_proc_component,copy_vtab_proc_comps): Remove
> ? ? ? ?initializers for PPC members of the vtabs.
>
>
> 2010-04-26 ?Janus Weil ?<janus@gcc.gnu.org>
>
> ? ? ? ?PR fortran/43896
> ? ? ? ?* gfortran.dg/class_16.f03: New.
>



-- 
The knack of flying is learning how to throw yourself at the ground and miss.
       --Hitchhikers Guide to the Galaxy