Re: CIE version number vis-a-vis 000330.1

[Michael Eager <eager at eagercon dot com>]

If there is value in retaining the existing version as valid, then we
should document both versions and make it explicit that both are acceptable.

"Burton, Felix" wrote:
> 
> I believe it is necessary to change the version number since there is no way
> for existing DWARF readers to skip the new CIE instructions.
> 
> If somebody would like to generate CIE information with the old version
> number we have two options:
> 
> 1. Refer to the old document
> 2. Describe what they should not use.
> 
> I think option 2) is better but more work...
> 
> Felix
> 
> > -----Original Message-----
> > From: brender@gemevn.zko.dec.com [mailto:brender@gemevn.zko.dec.com]
> > Sent: Thursday, April 20, 2000 1:58 PM
> > To: DWARF2@corp.sgi.com; BRENDER@gemevn.zko.dec.com
> > Subject: CIE version number vis-a-vis 000330.1
> >
> >
> > Yesterday we approved 000330.1, which includes in part a
> > change in the CIE
> > version number to 2 (from 1 in DWARF 2.0) -- but we did not
> > discuss that
> > point specifically.
> >
> > I question whether the version number change is really
> > necessary. Shouldn't
> > it already be the case that a DWARF consumer should recognise and deal
> > delicately with any CIE instruction code it does not
> > understand (whether
> > vendor defined or not)? So long as we don't change the
> > encoding or meaning
> > of any *existing* (V2.0) CIE instruction, it seems like
> > adding a couple new CIE
> > instructions to the list should not require an increase in
> > the CIE version
> > number. Am I missing something?
> >
> > Ron
> >
> > p.s. If that isn't quite right, perhaps an alternative is to
> > specify that the
> > version number is 1 if only codes from Dwarf V2.0 are used,
> > and is 2 only
> > if one of the new codes from V2.1 is actually present.
> >


-- 
Michael Eager	 Eager Consulting     eager@eagercon.com
1960 Park Blvd., Palo Alto, CA 94306  650-325-8077