This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix tree-opt/35431 ICE with complex integer, invalid gimple
- From: "Richard Guenther" <richard dot guenther at gmail dot com>
- To: "Andrew Pinski" <pinskia at gmail dot com>
- Cc: "GCC Patches" <gcc-patches at gcc dot gnu dot org>
- Date: Mon, 31 Mar 2008 10:07:59 +0100
- Subject: Re: [PATCH] Fix tree-opt/35431 ICE with complex integer, invalid gimple
- References: <de8d50360803302227n21fe7137kdf908998f1ff24c@mail.gmail.com>
On Mon, Mar 31, 2008 at 6:27 AM, Andrew Pinski <pinskia@gmail.com> wrote:
> Hi,
> The issue here is that PHI-opt will produce invalid gimple for
> complex integeral types. I could not figure out an easy way to
> produce valid gimple for them, so I decided that rejecting would be
> the best option, since later on we will split apart the complex type
> and also complex integeral types are rare so we are not losing that
> much optimizations really.
>
> OK? Bootstrapped and tested on i686-linux-gnu with no regressions.
Ok. Another case of integer_zerop returning true for complex is weird...
Richard.
> Thanks,
> Andrew Pinski
>
> ChangeLog:
> * tree-ssa-phiopt.c (conditional_replacement): Return early for complex types.
>
> * gcc.c-torture/compile/pr35431.c: New testcase.
>