more m68k breakage on m68k-linux

[Gunther Nikl]

Hello Roman,

On Thu, Mar 04, 2004 at 02:44:26PM +0100, Roman Zippel wrote:
> 
> On Thu, 4 Mar 2004, Bernardo Innocenti wrote:
> 
> > >   Can you please post the patch that corrects the FIXED_REGISTERS problem?
> >
> > I will.  Please allow some time since I'm on a tight work schedule
> > right now.  I'll regtest the patch on m68k-linux and m68k-netbsd.
> > Additionally, I'll build the m68k-uclinux kernel & userland.
> 
> Does your patch look very different from the one below?
> That's the one I'm testing for m68k-linux right now.

  Thanks for providing something to look at!

> Index: config/m68k/m68k.c
> ===================================================================
> RCS file: /cvsroot/gcc/gcc/gcc/config/m68k/m68k.c,v
> retrieving revision 1.128
> diff -u -r1.128 m68k.c
> --- config/m68k/m68k.c	17 Feb 2004 20:24:44 -0000	1.128
> +++ config/m68k/m68k.c	4 Mar 2004 11:25:54 -0000
> @@ -44,6 +44,18 @@
>  #include "debug.h"
>  #include "flags.h"
> 
> +char regno_reg_class[] =
> +{
> +  DATA_REGS, DATA_REGS, DATA_REGS, DATA_REGS,
> +  DATA_REGS, DATA_REGS, DATA_REGS, DATA_REGS,
> +  ADDR_REGS, ADDR_REGS, ADDR_REGS, ADDR_REGS,
> +  ADDR_REGS, ADDR_REGS, ADDR_REGS, ADDR_REGS,
> +  FP_REGS, FP_REGS, FP_REGS, FP_REGS,
> +  FP_REGS, FP_REGS, FP_REGS, FP_REGS,
> +  ADDR_REGS
> +};

  Wouldn't it be better to make this "const enum reg_class"? Then you don't
  need to cast in REGNO_REG_CLASS. I guess using char[] saves space?

  Gunther