------- Comment #6 from law at redhat dot com 2006-01-24 22:50 -------
Subject: Re: [4.2 Regression] ICE on ACATS cxac004 in
Tree-VRP
On Tue, 2006-01-24 at 22:15 +0000, laurent at guerby dot net wrote:
>
> ------- Comment #5 from laurent at guerby dot net 2006-01-24 22:15 -------
> c99004a randomly fails PR21317, unrelated to your fix, so go ahead :).
Ah. Well, I feel better now -- I couldn't see how my change would
be causing an execution failure regression.... It should only have been
avoiding the segfault!
I'm going to ponder the SCEV code in question a little more -- while
it's easy enough to filter out unwanted stuff in VRP, I'm really
more concerned that SCEV gave us junk in the first place.
jeff
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25900
