[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Fatal: core dump in binutils-2.12/ld
From: |
Nick Clifton |
Subject: |
Re: Fatal: core dump in binutils-2.12/ld |
Date: |
09 Apr 2002 10:59:54 +0100 |
User-agent: |
Gnus/5.0808 (Gnus v5.8.8) Emacs/21.1 |
Hi Jens,
> > > #0 0x10032738 in bfd_xcoff_size_dynamic_sections (output_bfd=0x20017b98,
> > > info=0x20009028, libpath=0x20023998 <Address 0x20023998 out of
> > > bounds>,
> > > entry=0x0, file_align=554186380, maxstack=0, maxdata=804395112,
> > > gc=true,
> > > modtype=12620, textro=true, export_defineds=false,
> > > special_sections=0x2ff21910) at xcofflink.c:3203
> >
> > It's hard to see how 2.12 xcofflink.c line 3203
> >
> > if (! _bfd_coff_free_symbols (sub))
> >
> > can fail with a segfault, unless your compiler is generating bad code.
> > Would you please verify that your sources are 2.12, and that xcofflink.c
> > line 3203 is as I've quoted above?
> Indeed, it's the line. The compiler was either the precompiled GCC 3.0.1
> from Bull or the self-compiled GCC 3.0.4.
Hmm, in which case, please could you start up gdb again, and check
the value of the variable 'sub'. Also what is the value of
'sub->filename' and 'sub->tdata.coff_obj_data' ?
Also - is it possible for you to generate a *small* test case that
reproduces the problem. eg a small object file, plus the command
line given to the linker, plus a description of the environment in
which this command is run.
Cheers
Nick