Re: [PATCH v2, middle-end]: Introduce TARGET_REJECT_COMBINED_INSN target hook

[Uros Bizjak <ubizjak at gmail dot com>]

On Sun, Aug 26, 2012 at 8:11 AM, Uros Bizjak <ubizjak@gmail.com> wrote:

>> What if we just add a global variable during_combine, and the insn
>> predicates can check that variable?  Then we wouldn't need a new
>> target hook, one that seems slightly obscure to me.
>
> No, in x86 case we need to look at operand constraints of the insn,
> and we need to recognize and analyse the insn. This way, we can avoid
> i.e. "ax_reg", "nonimm_ax_reg" and such register constraints.

Please also note that with "enabled" attribute, operand constraints
are dynamically changed, so predicate would need to track these
changes (i.e. "*umul<mode><dwi>3_1").

BTW: I found the proposed "legitimate_combined_insn" name much more
descriptive, I will update the patch accordingly.

Uros.