GEOM architecture and the (lack of) need for foot-shooting

[Andrey Chernov]

On Thu, Apr 07, 2005 at 08:06:20PM -0700, Marcel Moolenaar wrote:
> Typical problem cases include partial overlaps between on-disk and
> in-core partitions. This is easy to catch and can always result in a

I don't think that in-core and on-disk partition must have influence to 
each other in any way. Better for them be completely independent. Once 
read, in-core partition should not know anything about what happens 
with on-disk one.

> I think that having a single view is probably what's biting. If you

Yes. But who speak about single view? If we have in-core and on-disk 
partition separately, we need _two_ independent views, choosed f.e. by 
some option.

-- 
http://ache.pp.ru/