Re: Backward and forward compatibility

[Jeff Johnson <n3npq nc rr com>]

Axel Thimm wrote:

> On Thu, Aug 28, 2003 at 03:29:01PM -0400, Jeff Johnson wrote:
>  
>
> > > > > In fact I am already building with 4.2 for all RH platforms, is there
> > > > > anything to fear?
> > > > >          
> > > Would 4.2.1 be a better candidate for consolidating rpm versions?
> > >      
> > Your choice is not as simple as choosing a version of rpm, you need to
> > worry about what kernel and glibc are on the target systems.
> >
> > This is NPTL/TLS/futex/O_DIRECT prolly more in action.
> >    
>
> Well, O_DIRECT is being ripped out by force in the specfiles (BTW when
> you decided to rip it out, you forgot to change the release tag, so
> there are some versions of rpm-4.2-1 floating around with enabled
> O_DIRECT support). BTW2 there is some discussion on the xfs lists on
> the background of db4/O_DIRECT/sparse files bug. If you like I can add
> you to the discussion.
>  

Um, I'm less than happy about O_DIRECT: stupid interface, arcane problem.
The killer is/was that O_DIRECT was unleashed in the kernel with hardly 
any warning,
certainly not enough to do something rational in distributing rpm.

Sure there are are versions around. I have the (ahem) joy of developing 
without
a release plan or schedule, with whatever I just built released through 
Raw Hide
and betas.

Ready or not, here's rpm!

>  
>
> > If not NPTL, you want rpm version compiled w/o --enable-posixmutexes,
> > that means prefer rpm-4.1.1 over rpm-4.2 on non-NPTL.
> >    
>
> Why not use the same code base (be it 4.2 or 4.2.1) with
> --disable-posixmutexes?
>  

That is essentially rpm-4.2 and rpm-4.1.1. Same code base, built and 
configured
slightly differently.

73 de Jeff