This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Fix PR rtl-optimization/89795
- From: Segher Boessenkool <segher at kernel dot crashing dot org>
- To: Eric Botcazou <ebotcazou at adacore dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Wed, 11 Sep 2019 19:42:25 -0500
- Subject: Re: Fix PR rtl-optimization/89795
- References: <1596660.6ZbOInhRmP@polaris> <20190911120500.GX9749@gate.crashing.org> <10527824.N1McutAL5X@polaris>
On Wed, Sep 11, 2019 at 02:18:36PM +0200, Eric Botcazou wrote:
> > What does it mean for
> > (set (reg:QI) (const_int -128))
> > for example? This is the canonical form of moving positive 128 into a
> > register as well, of course.
>
> Nothing, since word_register_operation_p returns false on CONST_INTs .
>
> > W_R_O needs many more restrictions if it can work at all, but those aren't
> > documented.
>
> The problematic issue is more basic and pertains to mere SUBREGs.
I'd say this is more basic, but heh... It's an example. If it isn't
documented what W_R_O really *does*, how can we ever hope to make it
works stably?
Segher