On Mon, 5 Jun 2000, Lars Marowsky-Bree wrote:
> On 2000-06-05T07:57:36,
> Rik van Riel <riel@conectiva.com.br> said:
>
> > This is all very cool, but if it means that the on-disk format
> > will keep changing every few weeks, I think it may be better to
> > delay reiserfs integration in the kernel until 2.5 ...
>
> I disagree. As long as the version in 2.4 is bugfixed, I see no problem with
> that.
I see no problem with that either - provided the version of ReiserFS
included in 2.4 is properly maintained in addition to the development work
on 2.5's version. I see MAJOR problems if 2.4 ends up with a development
snapshot, and all the work then goes into developing 2.5 to the detriment
of 2.4.
> Obviously, major changes should happen in 2.5 and not 2.4. But that is the way
> it has always been with Linux.
Major changes mustn't happen in 2.4 - and 2.4 must be properly maintained.
> Opposing a merge of a very useful feature into a stable kernel just because
> major improvements are planned for the next development cycle seems
> counterproductive.
Opposing a merge of a very useful feature which will not be properly
maintained into a stable kernel, OTOH, is just common sense. Whatever goes
into 2.4 MUST be production-quality code, with production-quality
maintenance and stability (and I mean no compatability-breaking changes
between 2.4.0 and 2.4.99, not "it doesn't crash").
IF ReiserFS is reliable and tested enough to go into production (and it
appears to be, AFAICS), AND the version going into 2.4 is going to be
properly stabilised and maintained, then I would agree with including it.
The worst case scenario is that ReiserFS goes into 2.4.0 - and then
becomes abandoned, or the on-disk format changes within 2.4. This would be
an absolute disaster. The best case scenario is that ReiserFS is ready for
inclusion now, AND will have the appropriate attention devoted to it,
without attention being focussed on the 2.5 version instead. Leaving
ReiserFS out entirely is not ideal, but it's certainly better than the
first option.
James.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Jun 07 2000 - 21:00:22 EST