linux-mips
[Top] [All Lists]

Re: problems with insmod ...

To: Jun Sun <jsun@mvista.com>
Subject: Re: problems with insmod ...
From: Ralf Baechle <ralf@oss.sgi.com>
Date: Tue, 24 Oct 2000 03:18:28 +0200
Cc: Keith Owens <kaos@melbourne.sgi.com>, linux-mips@oss.sgi.com
In-reply-to: <39F4E113.6BC85A5C@mvista.com>; from jsun@mvista.com on Mon, Oct 23, 2000 at 06:08:35PM -0700
References: <39F48742.933941B8@mvista.com> <7690.972340323@ocs3.ocs-net> <20001024024040.D1009@bacchus.dhis.org> <39F4E113.6BC85A5C@mvista.com>
Sender: owner-linux-mips@oss.sgi.com
On Mon, Oct 23, 2000 at 06:08:35PM -0700, Jun Sun wrote:

> > > Jun Sun <jsun@mvista.com> wrote:
> > > >I tried with 2.3.19, and now I am having problem with out of bound index
> > > >in symbol table.  See the output below.
> > > >
> > > >---------
> > > >sh-2.03# insmod hello.o
> > > >hello.o: local symbol gcc2_compiled. with index 10 exceeds
> > > >local_symtab_size 10
> > > >hello.o: local symbol __gnu_compiled_c with index 11 exceeds
> > > >local_symtab_size 10
> > > >---------
> > >
> > > It is a toolchain bug, I think it is in the assembler.  I have a dim
> > > distant memory from about a month ago that somebody on linux-mips found
> > > the problem.  Ask the toolchain experts.
> > 
> > It's a bug bug in ld, one in BFD and a sillyness in IRIX ELF which the 
> > linker
> > uses.  IRIX ELF uses different sorting rules for the symbol table, see
> > mips_elf_sym_is_global in bfd/elf32-mips.c.
> > 
> >  - Bug one: ld generated output should follow the same rules as assembler
> >    generated output.
> >  - Bug two is more a design flaw - why does Linux/MIPS and most other
> >    MIPS ELF configurations use IRIX and not ABI ELF?
> >  - Bug three is that mips_elf_sym_is_global applies these IRIX ELF sorting
> >    rules even to ABI ELF.
> > 
> >   Ralf
> 
> This sounds like a serious problem to me.  So here are the questions : 
> 
> 1) is it fixed in the latest binutils?

No, all binutils ever are affected.

> 2) is it worth fixed for binutil v2.8.1?

Probably not.  I believe modulo some testing current CVS binutils are ready
for more serious use than just binutils fixing.  In any case fixing should
be easy.

  Ralf

<Prev in Thread] Current Thread [Next in Thread>