This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: [RFC] unify dynamic_symbol_p implementations
On Thu, Jul 17, 2003 at 09:28:15AM -0700, Richard Henderson wrote:
> On Thu, Jul 17, 2003 at 10:52:42PM +0930, Alan Modra wrote:
> > Looks reasonable to me, except these hunks.
>
> Um, those hunks are exactly the unification of the
> STV_PROTECTED handling vs function descriptors.
>
You changed
dynamic_symbol_p = elfNN_ia64_dynamic_symbol_p (h, info);
local_symbol_p = (! dynamic_symbol_p
|| ELF_ST_VISIBILITY (h->other) != STV_DEFAULT);
...
if (! local_symbol_p)
...
if (local_symbol_p)
...
to
dynamic_symbol_p = elfNN_ia64_dynamic_symbol_p (h, info, r_type);
...
if (! dynamic_symbol_p)
...
if (dynamic_symbol_p)
...
For those symbols on which the new elfNN_ia64_dynamic_symbol_p returns
the same as the old one, the code logic is reversed. Is that
intentional?
H.J.