This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] RHS INDIRECT_REF optimizations (PR tree-optimization/26069)


On Jan 12, 2008 1:55 PM, Andrew Pinski <pinskia@gmail.com> wrote:
> On Jan 12, 2008 3:40 AM, Richard Guenther <richard.guenther@gmail.com> wrote:
> > Technically I think we should use common machinery via
> > folding VIEW_CONVERT_EXPRs for this kind of type-punning and
> > not sticking the simplification in indirect-ref folding (that is, in
> > indirect-ref folding build_fold a VIEW_CONVERT_EXPR).
>
> To do this, you would then need to extend DECL_GIMPLE_REG_P to all
> types and maybe have a pass which also converts some non
> DECL_GIMPLE_REG_P to be true if it is valid for them.

I'm curious, why?

We can simply insert VIEW_CONVERT_EXPRs only if they fold to sth non
VIEW_CONVERT_EXPR if this is a problem.

Richard.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]