Re: Correct logic for "defined by object"

["H.J. Lu" <hjl dot tools at gmail dot com>]

On Tue, Dec 23, 2014 at 4:41 AM, Alan Modra <amodra@gmail.com> wrote:
> The old code missed testing bfd_link_hash_undefweak, and wrongly
> excluded bfd_link_hash_common symbols.  It is also clearer to invert
> the set of enum bfd_link_hash_type values tested.
> bfd_link_hash_indirect and bfd_link_hash_warning will never appear
> here.
>
>         * ldexp.c (update_definedness): Correct logic setting by_object.
>
> diff --git a/ld/ldexp.c b/ld/ldexp.c
> index 2828933..e9e95f9 100644
> --- a/ld/ldexp.c
> +++ b/ld/ldexp.c
> @@ -304,11 +304,11 @@ update_definedness (const char *name, struct bfd_link_hash_entry *h)
>      einfo (_("%P%F: bfd_hash_lookup failed creating symbol %s\n"), name);
>
>    /* If the symbol was already defined, and not by a script, then it
> -     must be defined by an object file.  */
> +     must be defined by an object file or by the linker target code.  */
>    if (!defentry->by_script
> -      && h->type != bfd_link_hash_undefined
> -      && h->type != bfd_link_hash_common
> -      && h->type != bfd_link_hash_new)
> +      && (h->type == bfd_link_hash_defined
> +         || h->type == bfd_link_hash_defweak
> +         || h->type == bfd_link_hash_common))
>      defentry->by_object = 1;
>
>    defentry->by_script = 1;
>

No testcase?


-- 
H.J.