Re: Fix PR debug/45088

[Jason Merrill <jason at redhat dot com>]

On 11/30/2010 11:06 AM, Dodji Seketeli wrote:
> Jason Merrill<jason@redhat.com>  writes:
>
> > On 11/12/2010 06:47 AM, Dodji Seketeli wrote:
> > > One possible way to do resolve that dilemma is [at the FE level] to
> > > fixup the type of *ai to make it be A instead of the self-reference
> > > type of A. I figure the fact that the self-reference type is different
> > > from A itself should be kept as an implementation detail of the FE
> > > anyway.
> >
> > I think the typedef should be emitted so that name lookup in the
> > debugger can find it.  TYPE_DECL_IS_STUB should not be true for the
> > injected-class-name.
>
> Yes this makes a lot of sense. I initially got somewhat confused by this
> code snippet in is_redundant_typedef:
>
>    if (DECL_ARTIFICIAL (decl)
>        &&  DECL_CONTEXT (decl)
>        &&  is_tagged_type (DECL_CONTEXT (decl))
>        &&  TREE_CODE (TYPE_NAME (DECL_CONTEXT (decl))) == TYPE_DECL
>        &&  DECL_NAME (decl) == DECL_NAME (TYPE_NAME (DECL_CONTEXT (decl))))
>      /* Also ignore the artificial member typedef for the class name.  */
>      return 1;
>
>  From what I understand it basically is tailored to not emit debug info
> precisely for injected-class-names.

Yep.

> But maybe at that time we didn't have the help of cgraph and the unused
> type debug info pruning we have today. In other words I think emitting
> debug info for the injected-class-name won't necessarily increase bloat
> because if the debug info of the injected-class-name typedef will be
> pruned if it is not used.

No, I don't think we can prune those.  Perhaps bloat is a reason not to 
go this way.

What about just having gen_type_die_with_usage check is_redundant_typedef?

Jason