This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: A patch for elflink.h
- To: Hans-Peter Nilsson <hans-peter dot nilsson at axis dot com>
- Subject: Re: A patch for elflink.h
- From: "H . J . Lu" <hjl at lucon dot org>
- Date: Sun, 29 Apr 2001 10:43:27 -0700
- Cc: amodra at one dot net dot au, binutils at sourceware dot cygnus dot com
- References: <20010427114620.A8342@lucon.org> <200104291606.SAA16405@ignucius.axis.se>
On Sun, Apr 29, 2001 at 06:06:13PM +0200, Hans-Peter Nilsson wrote:
> > Date: Fri, 27 Apr 2001 11:46:20 -0700
> > From: "H . J . Lu" <hjl@lucon.org>
>
> > The above patch won't work for hppa. Here is a new one. Any comments?
>
> This version looks right, though in exactly the same way as
> Alan's patch in the same area looked Right to me until tested,
> so I guess I don't know. :-) One thing I would check for
> regression if I was you (and will do when I get around to test
> CVS binutils next time) is to compile stuff (libraries) with
> -fPIC but link statically.
Why? The call is protected by eif->info->shared.
>
> Can you add your test-case to ld or glibc as appropriate?
I will find a place in ld for it.
H.J.