| To: | Ian Soanes <ians@lineo.com> |
|---|---|
| Subject: | Re: AW: GCC and Modules |
| From: | Ralf Baechle <ralf@oss.sgi.com> |
| Date: | Wed, 25 Jul 2001 19:00:46 +0200 |
| Cc: | Andre.Messerschmidt@infineon.com, linux-mips@oss.sgi.com |
| In-reply-to: | <3B5ED17F.6D50331F@lineo.com>; from ians@lineo.com on Wed, Jul 25, 2001 at 03:02:39PM +0100 |
| References: | <86048F07C015D311864100902760F1DDFF0016@dlfw003a.dus.infineon.com> <3B5ED17F.6D50331F@lineo.com> |
| Sender: | owner-linux-mips@oss.sgi.com |
| User-agent: | Mutt/1.2.5i |
On Wed, Jul 25, 2001 at 03:02:39PM +0100, Ian Soanes wrote: > This may be a little out of date, but compiling with the module with > -mlong-calls worked for me when I had similar problems loading modules a > while back. Relinking the module with 'ld -r -o new_mod.o orig_mod.o' > was useful too ...it worked around some 'exceeds local_symtab_size' > messages. The local_symtab_size symtab is caused by the fine differences between IRIX ELF and ABI ELF. More recent binutils versions switched to ABI ELF as the default. Not using -mlong-calls results in ``Relocation overflow'' messages. That's not the case here, so this option probably was used. Error messages about _gp_disp proof that the module contains PIC code that is has not been built using -mno-abicalls -fno-pic. Ralf |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: AW: GCC and Modules, Ian Soanes |
|---|---|
| Next by Date: | Re: ATI Victoria on Malta, James Simmons |
| Previous by Thread: | Re: AW: GCC and Modules, Ian Soanes |
| Next by Thread: | AW: GCC and Modules, Andre . Messerschmidt |
| Indexes: | [Date] [Thread] [Top] [All Lists] |