It turns out the values selected for MID_X86_64 and MID_SPARC64 overlap
M_MIPS1 and M_MIPS2 in binutils/bfd/libaout.h.  This is unfortunate so
I'd like to renumber them as to not conflict.  And update binutils
to include any missing values that we have and they don't (such as
MID_POWERPC).

Index: sys/exec_aout.h
===================================================================
RCS file: /cvsroot/src/sys/sys/exec_aout.h,v
retrieving revision 1.32
diff -u -3 -p -r1.32 exec_aout.h
--- sys/exec_aout.h     8 Aug 2003 18:54:16 -0000       1.32
+++ sys/exec_aout.h     9 Feb 2004 08:21:28 -0000
@@ -82,11 +82,11 @@ struct exec {
  #define        MID_SH3         145     /* SH3 */
  #define        MID_POWERPC     149     /* big-endian PowerPC */
  #define        MID_VAX         150     /* vax */
-#define        MID_SPARC64     151     /* LP64 sparc */
-#define        MID_X86_64      152     /* AMD x86-64 */
  #define        MID_M88K        153     /* m88k BSD core dumps */
  #define        MID_SH5_32      154     /* ILP32 SH5 */
  #define        MID_SH5_64      155     /* LP64 SH5 */
+#define        MID_SPARC64     156     /* LP64 sparc */
+#define        MID_X86_64      157     /* AMD x86-64 */
  #define        MID_HP300       300     /* hp300 (68020+68881) BSD 
binary */
  #define        MID_HPUX        0x20C   /* hp200/300 HP-UX binary */
  #define        MID_HPUX800     0x20B   /* hp800 HP-UX binary */

This will mean old a.out core files and kernel dumps will not be handled
by gdb.  The latter can be handled and I don't think any of the former
actually exist.

Comments?
-- 
Matt Thomas                     email: matt@3am-software.com
3am Software Foundry              www: http://3am-software.com/bio/matt/
Cupertino, CA              disclaimer: I avow all knowledge of this 
message.