This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [RFA] Patch PR debug/41065
On Sat, Aug 29, 2009 at 06:35:46PM +0200, Dodji Seketeli wrote:
> Le 29/08/2009 17:29, Jakub Jelinek a écrit :
> > On Sat, Aug 29, 2009 at 05:20:26PM +0200, Dodji Seketeli wrote:
> >> Fix PR debug/41065
> >>
> >> gcc/ChangeLog:
> >> PR debug/41065
> >> * function.h (global_used_types_hash): Declare ...
> >> * function.c (global_used_types_hash): ... new hash table for types
> >> used at global scope.
> >
> > Wouldn't a pointer_set_t rather than htab_t be a better choice?
> > You never delete from the hash table, only insert into it and traverse it.
>
> Yes probably. I chose a htab_t for the sake of consistency with
> function::used_types_hash. That minimises the changes in
> used_types_insert_helper. If I switch global_used_types_hash to a
> pointer_set_t, shouldn't I change function::used_types_hash as well ? There
> never seem to be any deletion on it either.
I'd say so, both could and IMHO should be pointer sets.
Jakub