This is the mail archive of the
mailing list for the GCC project.
Re: loop optimizer naivety
- From: Richard Guenther <rguenth at tat dot physik dot uni-tuebingen dot de>
- To: gcc at gcc dot gnu dot org
- Cc: Glen Nakamura <glen at imodulo dot com>,Eric Botcazou <ebotcazou at libertysurf dot fr>
- Date: Tue, 23 Sep 2003 12:14:02 +0200 (CEST)
- Subject: Re: loop optimizer naivety
Glen Nakamura writes:
> On Tue, Sep 23, 2003 at 10:38:05AM +0200, Eric Botcazou wrote:
> > I see you fixed PR opt/10087 by tweaking the loop optimizer. However, it
> > still thinks that the biv is incremented twice and that the total increment
> > is 2, which badly fools the unroller.
> The patch for PR opt/10087 was minimal and as non-intrusive as possible
> to fix the testcase. I wasn't very familiar with the loop optimizer and
> didn't want to change variables such as 'biv_count' and 'incremented' without
> fully understanding what effect it would have in other parts of the code.
> I'm willing to look into this, but unfortunately the testcase for PR opt/12340
> is difficult to work with and apparently doesn't segfault on my system even
> with a vanilla 3.3 branch build.
The segfault probably only occours on the ia32 platform, I'm at the moment
checking if the mentioned patch fixes/hides the problem.
But as Eric seems to have found the underlying problem, it might be
possible to construct a C testcase with this information?
Richard Guenther <richard dot guenther at uni-tuebingen dot de>