Re: [patch] fix wrong gimple code with ptr to forward vla

[Mark Mitchell <mark at codesourcery dot com>]

Richard Kenner wrote:
> > I think I understand why the heuristic works, in practice.  It's based 
> > on Richard's explanation, which assumes an Ada-like language in which 
> > types are elaborated at a well-defined point.  
>
> I don't follow that comment.  It would seem to me that in *any*
> language where you have a definition of a type that involves
> variables, you have a be able to specify which value of the variable
> is used, meaning that you have to elaborate the type at a well-defined
> point.

Yes -- but that doesn't necessarily correspond to a type-declaration, as 
signified by a TYPE_DECL.

Your approach is imposing TYPE_DECLs on front ends.  Joseph's is 
requiring front ends to make clear when things are evaluated.

Joseph's approach is more explicit, and therefore more flexible; some 
languages might require that array bounds be evaluated in a particular 
order, for example, so that the equivalent of "int x[f()][g()];" had 
well-defined behavior.  A TYPE_DECL for an ARRAY_TYPE with the obvious 
bounds doesn't make that explicit.  So, the front end would need to 
expand things anyhow; Joseph (and I) think we might as well just require 
that the front end always do that.

--
Mark Mitchell
CodeSourcery
mark@codesourcery.com
(650) 331-3385 x713