This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: [patch] ivopts fixes
- From: Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>
- To: Giovanni Bajo <rasky at develer dot com>
- Cc: gcc at gcc dot gnu dot org
- Date: Mon, 13 Sep 2004 19:20:02 +0200
- Subject: Re: [patch] ivopts fixes
- References: <20040913154455.GA30018@atrey.karlin.mff.cuni.cz> <000701c499b5$9763b1c0$f503030a@mimas>
Hello,
> > this patch fixes several problems with ivopts:
> >
> > 1) Ada failures due to misshandling of array_ref_element_size and
> > array_ref_low_bound are fixed.
> > 2) missing usharing of expressions is added to rewrite_use_outer.
> > This makes previous fixes of sharing problems in get_computation_at
> > and iv_value redunant and they are removed.
> > 3) signed and unsigned variants of types are obtained through langhooks
> > 4) dumps now show also types of induction variables
>
> Thanks. While you were away, there has been some discussion about
> reimplementing peel_address in terms of get_inner_reference (or better, of a
> new function which is the extract body of the loop of get_inner_reference).
> Does this make sense to you?
yes; however Honza told me that the patch he proposed did not work. I
will have a look at the problem tomorrow.
Zdenek