Re: [patch] for PR 17531

[Zdenek Dvorak <rakdver at atrey dot karlin dot mff dot cuni dot cz>]

Hello,

> And all that said, I have trouble believing that iv-opts isn't buggy
> asking for this narrow address in the first place.
> 
> >From what I can determine looking at get_computation_at, you're
> asking what's the cost of expressing USE with CAND.  If CAND is a
> pointer type and USE is a smaller integer type then either
> 
> 2134      if (TYPE_PRECISION (utype) > TYPE_PRECISION (ctype))
> 
> is reversed, or there's some later substitution of utype for ctype.
> In either case, UTYPE doesn't have enough precision to compute CTYPE,
> and thus *any* answer you give is incorrect.

this is because you read it wrong.  You need to express USE using CAND,
therefore this test is correct -- you need type of CAND (ctype) to be
at least as wide as the type of USE (utype), so that you have enough
information.

Zdenek