Re: Update readelf to know about the new ELF constants

[Hans-Peter Nilsson <hans-peter dot nilsson at axis dot com>]

> Date: Mon, 27 Nov 2000 11:12:51 -0800
> From: Nick Clifton <nickc@redhat.com>

> Looking through common.h, I think that we ought to try to register the
> following values:

> Hi Uli,
> : I''m not sure this will be OK.  All the numbers allocates are in
> : sequence (with a few holes left).
> 
> Is there a good reason for this ?  These are just numbers right ?
> 
> The values mentioned above are already in use in real code, so
> changing them would not be a good idea.  Unless there is a real need
> to have consecutive EM_ values, I would prefer to leave them alone.
> 
> : Besides, there is already a EM_ALPHA.
> 
> Except that it is not used, at least not by the binutils code.
> Would you agree to calling it:
> 
>  #define EM_ALPHA_ALTERNATE     0x9026  /* An alternate DEC Alpha machine number.  */

I might be pointing out the obvious (but it doesn't look like
it), please forgive:

In elf-bfd.h, struct elf_backend_data, you'll find
infrastructure to recognize (but not emit, methinks) legacy
(unregistered ad-hoc) e_machine values like the ones you listed:

  /* Alternate EM_xxxx machine codes for this backend.  */
  int elf_machine_alt1;
  int elf_machine_alt2;

Each entity or maintainer responsible for the machines with
unofficial e_machine numbers should ask registry@sco.com for an
officially blessed (in-sequence) number.

Perhaps you knew that.  Now maybe more people know it.

brgds, H-P