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 Mon, Jul 21, 2003 at 01:48:35PM -0700, Richard Henderson wrote:
> On Mon, Jul 21, 2003 at 12:28:08PM -0700, H. J. Lu wrote:
> > A weak symbol in executable shouldn't be overridden unless there is a
> > definition in a shared library at link time.
>
> Who says? IIRC this has varied between glibc versions.
>
There were different interpretations of how weak symbols should work
and we discussed it at the ia64 psABI meeting. The conclusion was
glibc was wrong. I added some weak symbol tests into binutils and
marked them XFAIL. This glibc bug has been fixed and I removed those
XFAILs from binutils tests.
H.J.