[Bug bootstrap/11777] m68k bootstrap failure on 3.4-branch and HEAD multilib build (-mcpu32)

["bernie at develer dot com" <gcc-bugzilla at gcc dot gnu dot org>]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11777


bernie at develer dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |UNCONFIRMED
         Resolution|DUPLICATE                   |
            Summary|m68k bootstrap failure on   |m68k bootstrap failure on
                   |3.4-branch and HEAD multilib|3.4-branch and HEAD multilib
                   |build                       |build (-mcpu32)


------- Additional Comments From bernie at develer dot com  2003-08-03 19:03 -------
It's most probable that defining __mc68020__ for CPU32 targets was 
done intentionally, since the supported instruction set and addressing 
modes are basically identical. 
 
There's a lot of code relying on __mc68020__, expecially in C 
libraries, so we must keep it for compatiblity. 
 
__mc68000__ is also being defined for all m68k targets, even 
though some ColdFire cores don't implement the full 68000 
instruction set. 
 
My patch cleanly fixes the current bootstrap failure and 
wouldn't break even if __mc68020__ was dropped at a later point.