On Sat, 3 May 1997 14:38:14 -0500 (CDT) 
 Dave Huang <khym@bga.com> wrote:

 > What does this mean for those of us cross-compiling (with the stock gcc
 > 2.7.2.x)? Can I do a diff between the previous gcc/arch/m68k/m68k.h and
 > the current one from the NetBSD source and make the same change to GNU's
 > config/m68k/m68k.h? And were there any other gcc files that changed, or is
 > m68k.h the only one?

...here's the diff.  The BIGGEST_ALIGNMENT change was the bug fix.

Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939

Index: m68k.h
===================================================================
RCS file: /cvsroot/src/gnu/usr.bin/gcc/arch/m68k/m68k.h,v
retrieving revision 1.3
retrieving revision 1.4
diff -c -r1.3 -r1.4
*** m68k.h	1996/05/17 01:56:13	1.3
--- m68k.h	1997/05/02 22:16:46	1.4
***************
*** 215,230 ****
  #define PARM_BOUNDARY (TARGET_SHORT ? 16 : 32)
  
  /* Boundary (in *bits*) on which stack pointer should be aligned.  */
! #define STACK_BOUNDARY 16
  
  /* Allocation boundary (in *bits*) for the code of a function.  */
! #define FUNCTION_BOUNDARY 16
  
  /* Alignment of field after `int : 0' in a structure.  */
! #define EMPTY_FIELD_BOUNDARY 16
  
  /* No data type wants to be aligned rounder than this.  */
! #define BIGGEST_ALIGNMENT 16
  
  /* Set this nonzero if move instructions will actually fail to work
     when given unaligned data.  */
--- 215,230 ----
  #define PARM_BOUNDARY (TARGET_SHORT ? 16 : 32)
  
  /* Boundary (in *bits*) on which stack pointer should be aligned.  */
! #define STACK_BOUNDARY (TARGET_SHORT ? 16 : 32)
  
  /* Allocation boundary (in *bits*) for the code of a function.  */
! #define FUNCTION_BOUNDARY (TARGET_SHORT ? 16 : 32)
  
  /* Alignment of field after `int : 0' in a structure.  */
! #define EMPTY_FIELD_BOUNDARY (TARGET_SHORT ? 16 : 32)
  
  /* No data type wants to be aligned rounder than this.  */
! #define BIGGEST_ALIGNMENT (TARGET_SHORT ? 16 : 32)
  
  /* Set this nonzero if move instructions will actually fail to work
     when given unaligned data.  */