This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Fix glibc builds


> On Wed, Nov 12, 2003 at 08:29:14PM +0100, Jan Hubicka wrote:
> > 	* c-decl.c (finish_decl): Use change_decl_assembler_name.
> > 	* c-pragma.c (handle_pragma_redefine_extname): Likewise.
> > 	* varasm.c (make_decl_rtl): Likewise.
> > 	* cgraph.c (change_decl_assembler_name): New function.
> > 	* tree.h (set_decl_assembler_name): Kill dead declaration.
> > 	(change_decl_assembler_name): Declare.
> > 
> > 	* decl.c (make_rtl_for_nonlocal_decl): Use change_decl_assembler_name.
> > 	* decl2.c (make_rtl_for_nonlocal_decl): Use change_decl_assembler_name.
> 
> Mostly ok.
> 
> > + 	  /* Check that we didn't introduce entry earlier.  */
> > + 	  slot1 = 
> > + 	    htab_find_slot_with_hash (cgraph_hash, name,
> > + 				      IDENTIFIER_HASH_VALUE (name), NO_INSERT);
> > + 	  if (slot1)
> > + 	    abort ();
> 
> This check is redundant with ...
> 
> > +       slot = 
> > + 	htab_find_slot_with_hash (cgraph_hash, name,
> > + 				  IDENTIFIER_HASH_VALUE (name), INSERT);
> > +       if (*slot)
> > + 	abort ();
> 
> ... this one.  Similarly for variables.
> 
> I'll say it would be nice if we had a unified hash table, since
> checking everything twice like this is no fun.

The probem is that I would like to keep datastructures different so we
don't end up with unnecesary memory and for most of use in optimizers
this is OK.
Perhaps I can add some functions around to make this easier.

Honza
> 
> 
> r~


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]