Re: powerpc64-linux infrastructure 5 of 6

[Geoff Keating <geoffk at geoffk dot org>]

> Date: Thu, 9 Aug 2001 09:14:09 +0930
> From: Alan Modra <amodra@bigpond.net.au>
> Cc: Geoff Keating <geoffk@redhat.com>, binutils@sourceware.cygnus.com
> Mail-Followup-To: Torbjorn Granlund <tege@swox.com>,
> 	Geoff Keating <geoffk@redhat.com>, binutils@sourceware.cygnus.com
> Content-Disposition: inline
> User-Agent: Mutt/1.3.17i
> 
> On Wed, Aug 08, 2001 at 10:55:37PM +0200, Torbjorn Granlund wrote:
> > Geoff Keating <geoffk@geoffk.org> writes:
> > 
> >   > Put relocs in the opcode table.
> >   
> >   This seems like a bad design.
> >   
> > It replaces an ugly chunk of ad-hoc code in tc-ppc.c.
> 
> Boils down to:
> 
> Simplified code + larger data structure
>  vs.
> more complex code + smaller data structure.
> 
> I admit it's arguable in this case whether the simplification warrants
> the extra data size.  Note that at this stage we haven't actually
> introduced any changes in assembler behaviour.

I think it's arguable that there is any simplification at all.  The
code may be shorter, but not simpler, because now there is a blurring
of the boundary between the opcodes library (which is supposed to be
independent of the object file format) and BFD.

-- 
- Geoffrey Keating <geoffk@geoffk.org>