This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: binutils cvs still broken on ppc
- To: Jack Howarth <howarth at nitro dot msbb dot uc dot edu>
- Subject: Re: binutils cvs still broken on ppc
- From: "H . J . Lu" <hjl at lucon dot org>
- Date: Fri, 12 Oct 2001 14:01:57 -0700
- Cc: binutils at sources dot redhat dot com, amodra at bigpond dot net dot au
- References: <200110121139.HAA27758@nitro.msbb.uc.edu>
On Fri, Oct 12, 2001 at 07:39:56AM -0400, Jack Howarth wrote:
> Alan,
> Could you please fix binutils cvs as HJ has requested on ppc
> (and the other arches)? I am still seeing massive make check failures
> in binutils cvs (just as I see in the binutils 2.11.92.0.5 release).
> If I build binutils cvs on debian ppc sid with...
>
This patch seems to fix Linux/PPC. Alan, could you please double check
all ELF targets for your refcount change? It seems to break almost all
ELF targets :-(.
Thanks.
H.J.
-----
2001-10-12 H.J. Lu <hjl@gnu.org>
* elf32-ppc.c (ppc_elf_adjust_dynamic_symbol): Set plt.offset
to -1 if the symbol is not a function.
--- bfd/elf32-ppc.c.refcount Fri Oct 12 11:50:14 2001
+++ bfd/elf32-ppc.c Fri Oct 12 13:12:09 2001
@@ -1797,6 +1797,11 @@ ppc_elf_adjust_dynamic_symbol (info, h)
return true;
}
+ else
+ {
+ h->plt.offset = (bfd_vma) -1;
+ h->elf_link_hash_flags &= ~ELF_LINK_HASH_NEEDS_PLT;
+ }
/* If this is a weak symbol, and there is a real definition, the
processor independent code will have arranged for us to see the