bug-gnu-utils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: problems with binutils-2.14 on sparc-sun-solaris


From: Michael van Elst
Subject: Re: problems with binutils-2.14 on sparc-sun-solaris
Date: Thu, 19 Jun 2003 10:04:02 +0200
User-agent: Mutt/1.4i

On Thu, Jun 19, 2003, Alan Modra wrote:

> On Wed, Jun 18, 2003 at 02:47:24PM +0200, Michael van Elst wrote:
> > .../.libs/libstdc++.a(io-inst.o)(.eh_frame+0x2d8): relocation truncated
> > to fit: R_SPARC_DISP32 .gnu.linkonce.t._ZNSdD1Ev
> 
> I thought I fixed this a while ago.  See
> http://mail.gnu.org/archive/html/bug-gnu-utils/2002-09/msg00263.html
> 
> Are you sure you're using the right tools?  Check by adding -v to the
> failing gcc command line.

I am sure as installation of binutils-2.13.2.1 in the same place
fixes the problem.

I had a look at the particular section in the binutils-2.14 source:

elf-eh-frame.c:582ff
        goto free_no_table;
      if ((*reloc_symbol_deleted_p) (buf - ehbuf, cookie))
        {
          /* This is a FDE against a discarded section.  It should
         be deleted.  */
          new_size -= hdr.length + 4;
          sec_info->entry[sec_info->count].removed = 1;
        }

So the cookie->rel = rel; is gone, but the memset didn't appear.

In elflink.h:8692ff
      if (_bfd_elf_discard_section_eh_frame (abfd, info, eh,
                         elf_reloc_symbol_deleted_p,
                         &cookie))
        ret = TRUE;

      if (cookie.rels != NULL
          && elf_section_data (eh)->relocs != cookie.rels)
        free (cookie.rels);
    }

ret is set to true as in your patch but (eh)->relocs is not set.

I will try to reapply your patches to the binutils-2.14 sources.

Greetings,
-- 
    ,eM""=.            a"-.                         Michael van Elst
   dWWMWM" -          :GM==;                        address@hidden
  :WWMWMw=--.          "W='  cable & wireless
   9WWMm==-.
    "-Wmw-"  CABLE & WIRELESS




reply via email to

[Prev in Thread] Current Thread [Next in Thread]