This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: relocation overflow
- From: Camm Maguire <camm at enhanced dot com>
- To: Alan Modra <amodra at bigpond dot net dot au>
- Cc: gcl-devel at gnu dot org,binutils at sources dot redhat dot com
- Date: 20 Nov 2002 12:01:59 -0500
- Subject: Re: relocation overflow
- References: <E18ELVW-0002rZ-00@intech19.enhanced.com> <20021120033243.GB997@bubble.sa.bigpond.net.au>
Greetings, and thanks for your reply.
The relocation type is R_PPC_REL24. I've got the failure up now in
gdb and can backtrace into the bfd library if this would help.
Take care,
Breakpoint 1, mreloc_overflow (link_info=0x10846afc,
name=0x12754c59 "do_init", reloc_name=0x107fd95c "R_PPC_REL24", addend=0,
abfd=0x127493bc, section=0x1275b014, address=28) at sfaslbfd.c:138
138 printf("reloc for %s is overflowing\n",name);
(gdb) hello
Alan Modra <amodra@bigpond.net.au> writes:
> On Tue, Nov 19, 2002 at 10:26:30PM -0500, Camm Maguire wrote:
> > Greetings! On ppc, GCL loads binary object modules at specific
> > address in its lisp core, relocating them via a call to
> > bfd_get_relocated_section_contents. This works in general, but in
> > certain cases of many such loadings and a large base executable, the
> > routine reports that function symbols in the module overflow on
> > attempted relocation. I thought this was a TOC issue, and could be
> > addressed via -mminimal-toc, but no dice. All tests thus far with gcc
> > 2.95, but I'm inclined to think that 3.2 will be the same.
> >
> > Does this problem ring a bell (and hopefully a solution) with anyone?
>
> Relocation type?
>
> --
> Alan Modra
> IBM OzLabs - Linux Technology Centre
>
>
--
Camm Maguire camm@enhanced.com
==========================================================================
"The earth is but one country, and mankind its citizens." -- Baha'u'llah