Re: [patch v3]middle-end: Teach ranger about POLY_INT_CST ranges for VRP.

Richard Sandiford <[email protected]>
Newsgroups gmane.comp.gcc.patches
Message-ID <[email protected]>
Tamar Christina <[email protected]> writes:
>> > +
>> > +	  if (!wi::le_p (bounds[0], bounds[1], sign))
>> > +	    {
>> > +	      r.set_varying (type);
>> > +	      return true;
>> > +	    }
>> 
>> Is this possible after the above?  I would hope that we could either drop
>> this or turn it into an assert.  Either way is ok with me.
>
> You're right, this is mostly a defensive check for a broken target bounds.
> but an assert is 100% more appropriate.
>
>> 
>> OK with those changes, thanks.  I think the patch has been around long
>> enough that more active folks would have commented by now if they wanted
>> to.
>> But please say if you think the above le_p is still needed.
>
> No I think it can be an assert.  OK with you to make it that, apply your
> patch and
> push or do you want to see the final version?

OK like that from my POV, no need for another round.  Thanks for
checking though.

Richard
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.