This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: PATCH to ld: embedded relocs for m68k COFF, 2nd take
- To: binutils at sources dot redhat dot com
- Subject: Re: PATCH to ld: embedded relocs for m68k COFF, 2nd take
- From: msokolov at ivan dot Harhan dot ORG (Michael Sokolov)
- Date: Fri, 21 Jul 00 19:09:21 CDT
Nick Clifton <nickc@cygnus.com> wrote:
> I have now looked over them, and they are fine, apart from one
> problem. The change to bfd/coff-m68k.c defines the new global
> function bfd_m68k_coff_create_embedded_relocs unilaterally.
Well, I copied this with no change from Ian's MIPS ECOFF, but...
> This
> is a problem when building with --enable-targets=all, since it is
> not multiply defined for the several different m68k COFF targets.
Whoops! Are you saying that coff-m68k.c is compiled multiple times for
different m68k COFF targets? I didn't think of that... I know of course about
--enable-targets=all, etc., but I thought every .c file in bfd/ is compiled
only once. I knew there was some code that was compiled multiple times, but I
thought it was all made into .h'es, like coffcode.h. Am I right in assuming
that the coff-m68k.c behavior is more of an exception than a rule? What other
backends behave like this? I hope none of the ELF ones do...
But anyway, thanks for catching and fixing this.
> I have added some #define tests which allow the --enable-targets=all
> toolchain to be built, but I will leave it up to you to test to see
> if this change actually prevents you from building your own,
> particularly targeted toolchain. I hope that it will not, but if
> there are problems please let me know.
OK, I'll test it when I get to that stage in building the toolchain, right now
I'm fighting with libiberty. (Just yesterday I started setting up my new
development system for GNU toolchain hacking, having left the old one at my
previous company where I no longer work.)
> Anyway I have now checked this patch in, and I hope we can continue
> development work from here.
Thanks!
--
Michael Sokolov Harhan Engineering Laboratory
Public Service Agent International Free Computing Task Force
International Engineering and Science Task Force
615 N GOOD LATIMER EXPY STE #4
DALLAS TX 75204-5852 USA
Phone: +1-214-824-7693 (Harhan Eng Lab office)
E-mail: msokolov@ivan.Harhan.ORG (ARPA TCP/SMTP) (UUCP coming soon)
P.S. When applying the bfd patch,
> : 2000-06-08 Michael Sokolov <msokolov@ivan.Harhan.ORG>
> :
> : * coff-m68k.c (bfd_m68k_coff_create_embedded_relocs): New function.
> : * bfd-in.h (bfd_m68k_coff_create_embedded_relocs): Add declaration.
> : * bfd-in2.h: Regenerate.
you forgot the last one. You may also want to fix the date on the ChangeLog
entry.