Re: QNX binutils targets

["Graeme Peterson" <gp at qnx dot com>]

> 
> On Fri, 25 Oct 2002, H. J. Lu wrote:
> > On Fri, Oct 25, 2002 at 02:17:05PM -0400, Hans-Peter Nilsson wrote:
> > > On Fri, 25 Oct 2002, H. J. Lu wrote:
> > > > On Fri, Oct 25, 2002 at 01:13:07PM -0400, Hans-Peter Nilsson wrote:
> > > > > Consider instead creating a special-named section, as proposed
> > > > > in <URL:http://sources.redhat.com/ml/binutils/2002-10/msg00454.html>.
> > > >
> > > > You have no ideas what you are talking about.
> > >
> > > Yes I do.  (Is this one of those flame wars? :-)
> > >
> > > Please avoid sounding condescending, at least on the subject of
> > > a debated issue.
> > >
> > > I'm considering compatibility with existing tools and objects
> > > too, which I think you're not with the ELFOSABI change.  Having
> >
> > Backward compatibility is possible with the ELFOSABI change.
> 
> Right; I can't see any exclusion on the EI_OSABI field.  So
> indeed it seems either solution would work for
> backward-compatibility, technically, as far as binutils is
> concerned.  Other tools and loaders may hork, but we don't need
> to consider them! ;-)

Aye, there's the rub.  _I_ need to consider them, as we have various
and sundry such tools in house.  But then that is my problem, not
yours.  :-)

I want to leave that support in place, so unless someone says not to, 
I think I will start investigating the special section approach.

Cheers, and thanks as always.
GP

> 
> > > > Please see
> > > > http://sources.redhat.com/ml/binutils/2002-10/msg00436.html
> > >
> > > I did (see below); it indicates that a EI_OSABI/ELFOSABI change
> > > is wrong.
> > >
> >
> > It is wrong because the ABI is the same. In the case of QNX, the ABI
> > is different. Did you get it?
> 
> No new facts.  That's what I meant. by "what system the object
> is for".  Still, the ELF fields (for example SHT_LOOS through
> SHT_HIOS) are interpreted the same, so an EI_OSABI change is
> wrong from the specification viewpoint (if only that).  But hey,
> changing the EI_OSABI field opens up for QNX to have other
> interpretations for the OS-specific ELF fields without
> introducing ambiguity!
> 
> brgds, H-P
> 
>