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: [PATCH][RFC] Fix alias-set generation for array types


> in component_uses_parent_alias_set we check
> 
>         case ARRAY_REF:
>         case ARRAY_RANGE_REF:
>           if (TYPE_NONALIASED_COMPONENT (TREE_TYPE (TREE_OPERAND (t, 0))))
>             return true;
> 
> while in alias.c we do
> 
>    switch (TREE_CODE (type))
>      {
> -    case ARRAY_TYPE:
> -      if (!TYPE_NONALIASED_COMPONENT (type))
> -       record_alias_subset (superset, get_alias_set (TREE_TYPE (type)));
> -      break;
> 
> thus, one time we check TYPE_NONALIASED_COMPONENT on the element type
> and one time on the array type.  What is the correct check?

I may be missing something, but both look like on the array type to
me.  Operand 0 of an ARRAY_REF is the array.


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